Carnitine palmitoyl transferase 1 (CPT1) enzymes catalyze the conversion of long-chain acyl-CoA to acyl-carnitines, thus facilitating the entry of long-chain fatty acids to the mitochondria, where they undergo β-oxidation. There are three isoforms: the liver isoform CPT1A (Esser, V. 1993), the muscle isoform CPT1B (Yamazaki, N. 1995) and the brain-specific isoform CPT1C (Price, N. 2002). CPT1A and CPT1B are localized in the outer mitochondrial membrane and are rate-limiting enzymes in fatty-acid β-oxidation. The CPT1C isoform, was first described in 2002, is expressed exclusively in the central nervous system, with a homogeneous distribution in all areas such as hippocampus, cortex, hypothalamus, cerebellum and others. CPT1C enzyme highly differs from the two other isozymes. Its C-terminal region is longer than that of the other CPTs (Price, N. 2002). It is located in the endoplasmic reticulum (ER) of cells, rather than in mitochondria, and so it does not facilitate fatty acid oxidation (Sierra, A.Y. 2008). Analysis of amino sequence of CPT1C reveals that all important residues for CPT1 activity are conserved in CPT1C enzyme, despite this, no catalytic activity was found (Price, N. 2002; Wolfgang, M.J. 2006), but it binds the CPT1 physiological inhibitor malonyl-CoA with the same affinity as CPT1A (Wolfgang, M.J. 2006). Finally, CPT1C is only present in mammals and appears to stem from a relatively recent CPT1A gene duplication (Price, N. 2002). The other isozymes are expressed in such organisms as fish, reptiles, amphibians or insects. This suggests a specific role for CPT1C in more evolved brains. At the physiological level, CPT1C contributes to the control of food intake and energy homeostasis (Wolfgang, M.J. 2006; Gao, X.F. 2009). Two independent groups developed a CPT1C-KO mouse, and both lines showed decreased food intake respect to wild-type animals (WT). However, when fed a high-fat diet, they were more susceptible to obesity and diabetes, presenting lower rates of peripheral fatty acid oxidation. All these effects were attributed to the hypothalamic function of CPT1C, since ectopic over-expression of CPT1C in hypothalamus protected mice from adverse weight gain caused by high-fat diet (Dai, Y. 2007). Moreover, the involvement of CPT1C in energy homeostasis has also been confirmed in transgenic animals over-expressing CPT1C specifically in brain (Reamy, A.A. 2011). At the molecular level, in collaboration with the group of Dr. Gary Lopaschuk, we showed that CPT1C is involved in the anorectic action of leptin, by modulating ceramide synthesis in the arcuate nucleus (ARC) of the hypothalamus (Gao, S. 2011). Interestingly, recent findings in tumor cells showed a new, unexpected role of CPT1C in the metabolic transformations reported in tumor cell growth (Zaugg, K. 2011). The authors demonstrated that CPT1C is frequently expressed in human lung tumors and protects cancerous cells from death induced by glucose deprivation or hypoxia. The results suggest that CPT1C might provide unidentified fatty-acid derived products that would be beneficial for cell survival under metabolic stress. However, despite these recent findings about CPT1C, little is known about its catalytic activity or its physiological function in other brain areas. We demonstrate that CPT1C has low CPT1 activity although it has similar affinity for its substrates: carnitine and palmitoyl-CoA than CPT1A isoform. The present study also shows that CPT1-KO mice have reduced long-chain acyl-carnitine levels in the hippocampus, hypothalamus or cerebellum. We examined whether CPT1C is expressed in the peripheral nervous system: in the ventral horn of the spinal cord (motor neurons) and in the sensitive ganglions, in addition to the brain. We found that CPT1C is expressed in both regions, albeit at lower levels than in the brain. We also examined CPT1C expression along mouse development, and we found that CPT1C protein expression is present in early stage of embryos at day 15, is increased postnatally and reaches its expression peak in adulthood. Moreover, CPT1C is expressed in pyramidal neurons of hippocampus and is located in ER throughout the neuron, even inside dendritic spines. We used molecular, cellular and behavioral approaches to determine CPT1C function. First, we analyzed the implication of CPT1C in ceramide metabolism. CPT1C over-expression in primary hippocampal cultured neurons increased ceramide levels, an effect that was blocked by treatment with myriocin, an inhibitor of the de novo synthesis of ceramide. Correspondingly, CPT1C knock-out (KO) mice showed reduced ceramide levels in hippocampus, cerebellum, striatum and motor cortex, mainly during fasting. At the cellular level, CPT1C deficiency altered dendritic spine morphology by increasing immature filopodia and reducing mature mushroom and stubby spines. Total protrusion density and spine head area in mature spines were unaffected. Treatment of cultured neurons with exogenous ceramide reverted the KO phenotype, as did ectopic over-expression of CPT1C, indicating that CPT1C regulation of spine maturation is mediated by ceramide. To study the repercussions of the KO phenotype on cognition and motor function, we performed the hippocampus-dependent Morris Water Maze (MWM) test and some motor tests on mice. Results show that CPT1C-KO mice are hypoactive and exhibit clear deficits in motor function, especially in coordination skills and strength. Moreover, CPT1C deficiency strongly impairs spatial learning without affecting memory or cognitive flexibility. So, all these results demonstrate that CPT1C regulates the de novo synthesis of ceramide in ER of hippocampal neurons and this is a relevant mechanism for the correct maturation of dendritic spines and for proper spatial learning. / ESTUDIO DE LA FUNCIÓN FISIOLÓGICA DE LA ENZIMA CPT1C La isoforma carnitina palmitoil transferasa 1C (CPT1C) se expresa únicamente en cerebro y ha sido implicada en la regulación hipotalámica de la ingesta de alimentos y la homeostasis energética. No obstante, su función molecular y su papel en otras áreas del cerebro son desconocidas. Hemos demostrado que CPT1C se expresa en las neuronas piramidales del hipocampo y se localiza en el retículo endoplásmico a lo largo de la neurona, incluso dentro de las espinas dendríticas. Hemos utilizado métodos moleculares, celulares y conductuales para determinar la función de CPT1C. En primer lugar, analizamos la implicación de CPT1C en el metabolismo de la ceramida. La sobre-expresión de CPT1C en neuronas de hipocampo aumentó los niveles de ceramidas, un efecto que fue bloqueado por el tratamiento con miriocina, un inhibidor de la síntesis de novo de la ceramida. En consecuencia, los ratones CPT1C knock-out (CPT1C-KO) demostraron una reducción de los niveles de ceramidas en el hipocampo, cerebelo, estriado y corteza motora principalmente durante el ayuno. A nivel celular, la deficiencia en CPT1C afecta a la morfología de las espinas dendríticas mediante el aumento de filopodios inmaduros y reduciendo el número de espinas maduras. La densidad de protrusiones totales o el área de la cabeza de la espinas dendrítica no se vieron afectadas. El tratamiento de las neuronas en cultivo con ceramida exógena, como la sobre-expresión ectópica de CPT1C, revirtieron el fenotipo de las espinas CPT1C-KO, lo que indica que CPT1C regula la maduración de las espinas dendríticas a través de las ceramidas. Para estudiar las repercusiones del fenotipo CPT1C-KO en la cognición y en la habilidad motora se realizaron diferentes test conductuales. Los resultados del test cognitivo demostraron que la deficiencia de CPT1C perjudica al aprendizaje espacial. Por otra parte, la realización de test motores demostraron que los ratones CPT1C son hipoactivos y tienen disminuida tanto la coordinación motora como la fuerza muscular. Todos estos resultados demuestran que CPT1C regula la síntesis de novo de ceramidas en el retículo endoplásmico de las neuronas y éste es un mecanismo necesario para la correcta maduración de las espinas dendríticas y para el adecuado procedimiento del aprendizaje espacial y la función motora.

Carnitine palmitoyltransferase 1 (CPT1) catalyzes the conversion of long chain fatty acyl-CoAs into acylcarnitines, the first step in the transport of long chain fatty acids from the cytoplasm to the mitochondrial matrix, where they undergo β-oxidation. This reaction is not only central to the control of fatty acid oxidation, but it also determines the availability of long chain acyl-CoA for other processes. There are three different CPT1 isozymes: CPT1A (expressed in liver, pancreas, kidney, brain, blood, and embryonic tissues), CPT1B (expressed only in brown adipose tissue, muscle, and heart) and the recently described CPT1C. CPT1C protein sequence is highly similar to that of the other two isozymes. Expression studies indicate that CPT1C is localized exclusively in the central nervous system, with homogeneous distribution in all areas (hippocampus, cortex, hypothalamus, and others). It has also been reported that CPT1c is localized in neurons but not in astrocytes of adult brain.1. CPT1C strucutral modelA 3-D structural model of the isozyme has been constructed by homology modeling. Residues contacting both substrates have been determined and compared to the same amino acid positions in CPT1A. The results obtained from the analysis show that the residues involved in the catalysis of the reaction in CPT1A and residues contacting both substrates are conserved mainly conserved in CPT1C or show semi-conservative substitutions. 2. CPT1 enzymatic activityExpression of rat CPT1C in Saccharomyces cerevisiae yields no catalytic activity when testing different conditions (longer periods of time, increased temperature, increased substrate concentration, testing of microsomal fraction or chimeric protein CPT1·ACA). Thus, the yeast expression system is not suitable for studying CPT1C enzymatic activity.3. Subcellular localizationEndogenous and overexpressed CPT1C is basically localized in the endoplasmic reticulum of mammalian cells (HEK293T, PC12, SH-SY5Y, primary cultures of fibroblasts and neurons). Some evidences indicated that CPT1C could also be found, in lower amounts, in mitochondrial associated membranes (MAMs).The specific sequence of CPT1C N-terminal domain (first 150 amino acids) drives the protein to the endoplasmic reticulum.4. CPT1C N-terminus processingThe N-terminal end of endogenous CPT1C in wild type mouse brain is processed (at least until Val27) and is not detected in mouse brain cortex lysates.5. CPT1C membrane topologyThe N- and C-terminal domains of CPT1C are facing the cytosolic side of the endoplasmic reticulum membrane, whereas the loop domain is facing the endoplasmic reticulum lumen.6. CPT1C interacting partnersThe data provided by the yeast two-hybrid assay do not indicate a unique binding partner of CPT1C. Instead the assay retrieved proteins involved in different functions: protein degradation, membrane trafficking, cell structure, signal transduction and metabolism.KEYWORDS: Carnitine palmitoyltransferase, Endoplasmatic reticulum, Subcelular localization, CPT1 activity, Structural model, Membrane topology / La carnitina palmitoiltransferasa 1 (CPT1) es una enzima que cataliza la conversión de aciles-CoA de cadena larga en acil-carnitinas, reacción crucial para el control de la oxidación de ácidos grasos. Existen tres isoformas diferentes de CPT1: CPT1A (isoforma más ubicua), CPT1B (expresada en tejido adiposo, músculo y corazón) y CPT1C que es la isoforma más recientemente descrita.La secuencia de la proteína CPT1C es muy parecida a la de las otras dos isoformas. Estudios de expresión indican que CPT1C se localiza exclusivamente en el sistema nervioso central. También se ha descrito que CPT1C se localiza en neuronas de cerebro adulto pero no en astrocitos.Las conclusiones obtenidas de los resultados presentados en esta tesis son (por apartados):1. Modelo structuralA través de técnicas de modelaje por homología se ha construido un modelo tridimensional teórico de la proteína. De su estudio se concluye que los residuos implicados en la catálisis de la reacción y los residuos en contacto con los sustratos están bien conservados en la secuencia de CPT1C.2. Actividad enzimáticaLa expresión de CPT1C en la levadura Saccharomyces cerevisiae no muestra actividad CPT1 aunque se testen diferentes condiciones (tiempos de reacción más largos, incrementos en la concentración de sustratos, pruebas en fracciones microsomales o pruebas con proteínas quiméricas como la CPT1·ACA)3. Localización subcelularLa proteína CPT1C se localiza básicamente en el retículo endoplasmático de células de mamífero (HEK293T, PC12, SH-SY5Y y en cultivos primarios de fibroblastos y de neuronas). La secuencia concreta de los 150 primeros aminoácidos dirige la porteína al retículo endoplasmático.4. Procesamiento del extremo N-terminal de CPT1CEl extremo N-terminal de la proteína CPT1C endógena sufre un procesamiento.5. Topología en la membrana de CPT1CLos dominios N- y C-terminal (centro catalítico) están orientados hacia la cara citosólica de la membrana del retículo endoplasmático.6. Proteínas de unión Los resultados obtenidos del ensayo de dobles híbridos no indican que CPT1C interaccione con una sola proteína de unión. Del ensayo se obtuvieron proteínas implicadas en diferentes funciones: degradación de proteínas, tráfico de membranas, estructura celular, vías de transducción de la señal y metabolismo.

Una de les accepcions de la farmacogenòmica es l'estudi dels efectes d'un tractament farmacològic sobre els nivells d'expressió gènica. Aquesta estratègia inclou la identificació de les variacions de gens individuals, l'avaluació de les interaccions entre els seus productes i la caracterització dels fenotips derivats de la resposta al fàrmac El metotrexat (MTX) és un fàrmac inhibidor de l'enzim dihidrofolat reductasa (DHFR) utilitzat en el tractament del càncer. El treball presentat en aquesta memòria es troba emmarcat dins de l'estudi de l'expressió gènica diferencial derivada de la resistència al MTX i pretén identificar aquells canvis adaptatius que tenen lloc en les cèl•lules sota tractaments perllongats amb el MTX. Alguns d’aquests gens diferencialment expressats poden contribuir a la resistència al fàrmac. Així doncs, s'identifica el gen S100A4 com a diana diferencialment expressada en 5 de les 7 línies cel•lulars resistents a MTX estudiades, i en cèl•lules de càncer de còlon HT-29, la via de Wnt sembla ser la responsable d’aquesta sobrexpressió. També en cèl•lules HT-29, la disminució dels nivells de proteïna S100A4 contribueix a augmentar la sensibilitat cap al MTX, fet que evidencia el paper important de la sobrexpressió de S100A4 en la resistència al MTX. En els últims anys s'ha descobert que les seqüències codificants representen només el 2% de tot el genoma i que la major part és transcrita com a molècules de RNA que no codifiquen per cap proteïna. Això implica que el que abans es considerava com a "junk DNA", en realitat és transcrit donant molècules de RNA reguladores de múltiples processos cel•lulars. Un dels tipus de RNAs no codificants més estudiats són els microRNAs (miRNAs). Actualment hi ha una extensa quantitat d'informació que descriu un clar paper dels miRNAs en tots els passos del desenvolupament del càncer, així com altres malalties. També s'estudia el paper dels microRNAs (miRNAs) en la resistència al MTX en cèl•lules de càncer de còlon. Els resultats presentats identifiquen el miR-224 i els seus gens diana, SLC4A4, CDS2 i HSPC159 com a dianes importants en la resistència al MTX en càncer de còlon. La reducció dels nivells del miR-224, acompanyada de l’augment dels seus gens diana CDS2, HSPC159 i SLC4A4, comporta una insensibilització cap al MTX, de manera que s’afavoreix la resistència al fàrmac. A més, es presenten altres miRNAs identificats com a sobrexpressats en la mateixa línia cel•lular de càncer de còlon (miR-149, miR-193b, miR-210, miR-27b, miR-320, miR-361-5p, miR-365, miR-455-3p i miR-615-3p) i s'identifiquen els gens diana putatius per a cadascun d'aquests miRNAs. La identificació dels gens diana per un determinat miRNA es realitza mitjançant algoritmes bioinformàtics que requereixen validació experimental. La interacció d’un miRNA amb el seu gen diana s’acostuma a estudiar mitjançant la co-transfecció del miRNA amb un vector reporter que contingui la regió 3’-UTR del gen. Tot i que aquest mètode suggereix una interacció física i funcional, no prova la interacció del miRNA amb el seu gen diana de forma directa. Finalment, es presenten resultats que demostren que els assajos de canvi de la mobilitat electroforètica (EMSA) constitueixen un mètode alternatiu per confirmar de manera directa i específica si un miRNA s’uneix o no al seu mRNA diana. / Pharmacogenomics study the effects of a drug on the expression levels of the genes. This strategy includes the identification of changes in individual genes, evaluating the interactions between their products and the characterization of phenotypes resulting from the response to the drug Methotrexate (MTX) is an inhibitor of the enzyme dihydrofolate reductase (DHFR) used in the treatment of cancer. The work presented in this report studies gene expression patterns derived from MTX resistance and seeks to identify the adaptive changes that occur in cells under prolonged treatment with MTX. Some of these differentially expressed genes may contribute to drug resistance. S100A4 was identified as a target gene overexpressed in 5 out of the 7 studied cell lines resistant to MTX. In HT-29 colon cancer cells, the Wnt signaling pathway appears to be responsible this overexpression. Also in HT-29 cells, S100A4 reduced protein levels contribute to increased sensitivity to MTX, which shows the important role of S100A4 overexpression on MTX resistance. MiRNAs are small non-coding RNAs that negatively regulate gene expression. There is an extensive amount of information that clearly describes the role of miRNAs in all stages of development of cancer as well as other diseases. It is also studied the role of microRNAs (miRNAs) in MTX resistance in colon cancer cells. The results presented identify the under expression of miR-224 and, therefore, the overexpression of its target genes, SLC4A4, and HSPC159 CDS2 as important targets for MTX resistance in colon cancer cells. The reduced levels of miR-224, accompanied by the increase of its target genes CDS2, HSPC159 and SLC4A4 lead to a desensitization toward MTX, which favors the drug resistance. Furthermore, the results identify a group of overexpressed miRNAs in the same colon cancer cell line (miR-149, miR-193b, miR-210, miR-27b, miR-320, miR-361-5p, miR -365, miR-455-3p and miR-615-3p) and the putative target genes for each of these miRNAs are predicted. Finally, we present results of tests showing that electrophoretic mobility shift (EMSA) are an alternative method to confirm directly and specifically whether or not miRNA binds to its mRNA target.

Plasmonics nanostructures are becoming remarkably important as tools towards manipulating photons at the nanoscale. They are poised to revolutionize a wide range of applications ranging from integrated optical circuits, photovoltaics, and biosensing. They enable miniaturization of optical components beyond the "diffraction limit'' as they convert optical radiation into highly confined electromagnetic near-fields in the vicinity of subwavelength metallic structures due to excitation of surface plasmons (SPs). These strong electromagnetic fields generated at the plasmonic "hot spots'' raise exciting prospects in terms of driving nonlinear effects in active media. The area of active plasmonics aims at the modulation of SPs supported at the interface of a metal and a nonlinear material by an external control signal. The nonlinear material changes its refractive index under an applied control signal, thereby resulting in an overall altered plasmonic response. Such hybrid nanostructures also allow for the creation of new kinds of hybrid states. This not only provides tools for designing active plasmonic devices, but is also a means of re-examining existing conventional rules of light-matter interactions. Therefore, the need for studying such hybrid plasmonic nanostructures both theoretically and experimentally cannot be understated. The present work seeks to advance and study the control of SPs excited in hybrid systems combining active materials and nanometallics, by an external optical signal or an applied voltage. Different types of plasmonic geometries have been explored via modeling tools such as frequency domain methods, and further investigated experimentally using both near-field and far field techniques such as scanning near field optical microscopy and leakage radiation microscopy respectively. First, passive SP elements were studied, such as the dielectric plasmonic mirrors that demonstrate the ability of gratings made of dielectric ridges placed on top of flat metal layers to open gaps in the dispersion relation of surface plasmon polaritons (SPPs). The results show very good reflecting properties of these mirrors for a propagating SPP whose wavelength is inside the gap. Another passive configuration employed was a plasmonic resonator consisting of dielectric-loaded surface plasmon polariton waveguide ring resonator (WRR). Also, a more robust variant has been proposed by replacing the ring in the WRR with a disk (WDR). The performance in terms of wavelength selectivity and efficiency of the WDRs was evaluated and was shown to be in good agreement with numerical results. Control of SPP signal was demonstrated in the WRR configuration both electro-optically and all-optically. In the case of electro-optical control, the dielectric host matrix was doped with an electro-optical material and combined with an appropriate set of planar electrodes. A 16% relative change of transmission upon application of a controlled electric field was measured. For all-optical control, nonlinearity based on trans-cis isomerization in a polymer material is utilized. More than a 3-fold change between high and low transmission states of the device at milliwatt control powers ( ~100 W/cm^2 intensity) was observed. Beyond the active control of propagating surface plasmons, further advancement can be achieved by means of nanoscale plasmonic structures supporting localized surface plasmons (LSP). Interactions of molecular excitations in a pi-conjugated polymer with plasmonic polarizations are investigated in hybrid plasmonic cavities. Insights into the fundamentals of enhanced light-matter interactions in hybrid subwavelength structures with extreme light concentration are drawn, using ultrafast pump-probe spectroscopy. This thesis also gives an overview of the challenges and opportunities that hybrid plasmonic functionalities provide in the field of plasmon nano optics. / Las nanoestructuras plasmónicas han adquirido una importante relevancia como herramientas capaces de manipular los fotones en la nanoescala, y pueden llegar a revolucionar un amplio abanico de aplicaciones tales como los circuitos ópticos integrados, la fotovoltaica o los dispositivos biosensores. Dichas estructuras hacen posible la miniaturización de los componentes ópticos más allá del “límite de difracción” de la luz, ya que convierten la radiación óptica en campos electromagnéticos fuertemente confinados en la proximidad de estructuras metálicas de tamaño inferior a la longitud de onda mediante la excitación de plasmones de superficie (SPs). Estos campos electromagnéticos tan intensos generados en los llamados “puntos calientes” plasmónicos brindan perspectivas muy interesantes para la generación de efectos no lineales en medios activos. El área de investigación denominado plasmónica activa busca la modulación de los SPs sostenidos por la intercara entre un metal y un material no lineal mediante una señal de control externa. El índice de refracción del material no lineal cambia bajo la aplicación de la señal de control, lo cual da lugar a la modificación de la respuesta plasmónica. Estas nanoestructuras híbridas también hacen posible la aparición de nuevos tipos de estados híbridos, lo cual proporciona tanto herramientas para diseñar dispositivos plasmónicos activos como mecanismos que permiten re-examinar las reglas convencionales de la interacción luz materia. Por lo tanto, es necesario el estudio de dichas nanoestructuras plasmónicas híbridas de manera teórica y experimental. En este trabajo de tesis se analiza el control de los SPs excitados en sistemas híbridos que combinan materiales activos y nanoestructuras metálicas mediante una señal óptica externa o un voltaje aplicado. Se han investigado distintos tipos de geometrías plasmónicas utilizando herramientas de simulación basadas en métodos en el dominio de la frecuencia, y posteriormente se han caracterizado experimentalmente dichas geometrías mediante técnicas de campo cercano y de campo lejano tales como la microscopía óptica de campo cercano y la microscopía basada en pérdidas radiativas, respectivamente. En primer lugar se estudiaron elementos plasmónicos pasivos, en particular espejos plasmónicos dieléctricos que demuestran la capacidad que tienen las redes periódicas de caballones de material dieléctrico colocados sobre una superficie metálica plana para abrir intervalos prohibidos en la relación de dispersión de los plasmones de superficie propagantes o plasmones-polaritones de superficie (SPPs). Los resultados muestran que dichos espejos poseen muy buenas propiedades reflectantes para SPPs cuya energía está en el intervalo prohibido. Otra configuración pasiva analizada fueron los resonadores plasmónicos basados en anillos de guía de onda plasmónica fabricada a partir de estructuras dieléctricas sobre metal (WRR, del inglés waveguide ring resonator ). Asimismo, se propone una versión más robusta de resonador plasmónico, basada en la sustitución del anillo del WRR por un disco (WDR, del inglés waveguide disk resonator). Se ha evaluado el funcionamiento de los WDRs en términos de selectividad en longitud de onda y de eficiencia, y los resultados obtenidos presentan un buen acuerdo con las predicciones numéricas. Pasando a las configuraciones activas, se demuestra el control de la señal plasmónica en configuración WRR por medios tanto electro-ópticos como completamente ópticos. En el caso del control electro-óptico, el material dieléctrico que compone el WRR estaba dopado con un componente electro-óptico y a la estructura pasiva se le añadió un conjunto de electrodos planos. Bajo la aplicación de un campo eléctrico externo, se midió un cambio relativo en la transmisión a través de la guía plasmónica del 16%. En cuanto al control puramente óptico, se utilizó la no linealidad de un material polimérico con origen en una isomerización trans-cis. En este caso se detectó un factor 3 entre los estados de alta y baja transmisión del dispositivo con potencias de control del orden de milivatios (intensidad del haz óptico de control de unos 100W/cm2 aproximadamente). Además del control activo de los plasmones de superficie propagantes, la utilización de nanoestructuras plasmónicas que poseen resonancias plasmónicas localizadas puede dar lugar a nuevos fenómenos. En esta tesis también se han estudiado las interacciones entre las excitaciones moleculares en un polímero pi-congujado con las polarizaciones plasmónicas en nanocavidades plasmónicas híbridas. Utilizando espectroscopia de tipo bombeo-sonda con pulsos ultrarrápidos, se han analizado diversos aspectos del aumento en la interacción luz-materia para estructuras híbridas de dimensiones inferiores a la longitud de onda sometidas a concentraciones de luz muy altas. Por último, esta tesis también proporciona una visión general de los desafíos y posibilidades que las funcionalidades plasmónicas híbridas ofrecen en el campo de la nano-óptica basada en plasmones de superfície.

Els receptors acoblats a proteïna G són una família de proteïnes de membrana que està implicada en moltes malalties d'una gran importancia en la nostra societat actual. Degut a la manca d'estructures tridimensionals d'aquests receptors s'ha desenvolupat un protocol per tal d'estudiar-los in silico. Aquest protocol utilitza les simulacions de dinàmica molecular per a estudiar l'estructura i la funció dels GPCRs tot tenint en compte l'entorn hidrofòbic de la bicapa lipídica en el que es troben i l'efecte dels residus de serina i treonina en la conformació de les hèlixs alfa d'aquests receptors. A partir del protocol s'ha construit i validat a través del disseny racional de fàrmacs un model tridimensional per receptor 5-HT1A que a més a més és vàlid per a la majoria dels receptors de les amines biogèniques. Aquest model s'ha utilitzat per a identificar el mode d'unió dels receptors serotoninèrgics amb els seus lligands. La modelitzaió del mode d'unió dels receptors amb els seus lligans ha permès el disseny de nous lligands amb més afinitat i selectivitat. Tot el treball computacional s'ha validat a través de la síntesi i avaluació farmacològica dels lligands així com amb experiments de mutagènesi dirigida. / G-protein coupled receptors (GPCRs) are membrane proteins that transduce and amplify extracellular chemical signals to the interior of the cell. Due to its prominent role in cell signalling, GPCR malfunction is involved in a wide variety of diseases. Because of the lack of X-ray resolved GPCR structures, this thesis presents a protocol to study the structure and function of these proteins in silico. This protocol is based on molecular dynamics simulations, and takes into account explicitly the hydrophobic environment of the receptors due to the surrounding cellular membrane. In addition, it considers the influence of serine and threonine residues in the global conformation of the hydrophobic transmembrane alpha-helices that constitute the structural feature common to all GPCRs. Based on this protocol, a three dimensional model for 5-HT1A receptor has been constructed, which is valid for most of the biogenic amine receptor family. The computational models of the binding site of serotoninergic receptors have been validated through the synthesis and pharmacological evaluation of new ligands, and through site-directed mutagenesis experiments. This way, these three-dimensional models have allowed to identify and characterize the binding site of the serotoninergic receptors 5-HT1A, 5-HT4 and 5-HT7, and to design new ligands with improved affinity and selectivity for these receptors.

En aquesta tesi s'apliquen eines bioinformàtiques a l'estudi de determinats sistemes biològics. En particular, l'estudi teòric de la influència de determinats aminoàcids sobre l'estructura i la dinàmica dels elements d'estructura secundària de les proteïnes s'aplica a la modelització per homologia dels receptors acoblats a proteïna G (GPCRs) i a l'estudi dels seus mecanismes d'activació.Se sap que determinats residus, com prolina, serina o treonina, provoquen distorsions locals en l'estructura de les hèlices a. L'anàlisi de bases de dades de seqüències de segments transmembrana mostra com certes combinacions d'aquests residus són més comunes que d'altres, i que algunes d'elles estan sobre-representades de manera significativa, mentre que d'altres estan clarament sots-representades. La restricció d'aquesta anàlisi de seqüències a la regió transmembrana dels GPCRs de la Classe A mostra com aquestes combinacions es troben en posicions específiques i, a més, es troben conservades en certes subfamílies de receptors.L'estructura i la dinàmica de les hèlices transmembrana que contenen aquestes combinacions de prolina i serina o treonina s'han estudiat mitjançant simulacions de dinàmica molecular en un entorn hidrofòbic explícit. Els resultats mostren com algunes d'aquestes combinacions indueixen distorsions importants en l'estructura de l'hèlix a, degut al seu efecte desestabilitzador de la xarxa de ponts d'hidrogen que dóna estabilitat a l'hèlix.Aquests resultats s'han aplicat a la construcció d'un model tridimensional del receptor de quimiocines CCR5 , utilitzant tècniques de modelització molecular per homologia. En aquest model es proposa que les hèlices transmembrana (TMH) 2 i 3 del receptor CCR5 són estructuralment diferents del patró de rodopsina. TMH2 està més doblegada degut a la presència d'un motiu Thr-X-Pro, que, a més, fa que aquesta hèlix es doblegui cap a TMH3. Així doncs, es proposa que, en aquest receptor, aquestes dues hèlices interaccionen. Aquesta interacció estaria mediada per la presència de residus hidrofòbics conservats i específics en les dues hèlices. Aquestes hipòtesis han estat posades a prova mitjançant experiments de mutagènesi dirigida, gràcies a la col·laboració amb l'Institut de Recherche Interdisciplinaire en Biologie Humaine et Nucléaire (IRIBHN), Université Libre de Bruxelles. Els resultats experimentals permeten establir la hipòtesi que la interfície TMH2-TMH3 participa en l'activació induïda per quimiocines del receptor CCR5.Com a conclusió, aquesta tesi pretén mostrar com, mitjançant la utilització d'eines bioinformàtiques, és possible traduir les seqüències primàries de proteïnes i les interaccions a nivell atòmic en estructures tridimensionals de proteïnes. A més, aquesta tesi mostra que, encara que l'estructura tridimensional de la rodopsina bovina és un patró útil per la modelització per homologia de GPCRs, s'han de tenir en compte de manera explícita les especificitats de seqüència de cada receptor per tal de construir models de receptors particulars. Aquestes especificitats de seqüència consisteixen en patrons de seqüència conservats en determinades famílies, que es tradueixen en divergències estructurals. Entre aquests patrons de seqüència, es proposa que els residus de serina i treonina, sols o combinats amb residus de prolina propers, poden modular la geometria de les TMHs, degut a la seva capacitat d'interferir amb la xarxa de ponts d'hidrogen que dóna estabilitat a les hèlices a.Finalment, es proposa que la influència dels motius de serina, treonina i prolina en l'estructura de les TMHs pot estar relacionada amb els processos d'activació dels GPCRs de la Classe A i, possiblement, d'altres proteïnes de membrana. En els GPCRs, aquests motius poden haver evolucionat per tal d'adaptar uns mecanismes d'activació conservats als lligands característics de cada família de receptors. / This thesis is framed in the study of particular biological systems through the use of bioinformatics. In particular, the theoretical study of the influence of certain amino acids on the structure and dynamics of the secondary structure elements of proteins has been applied to homology modelling of G protein-coupled receptors (GPCRs) and to the study of their mechanisms of activation.Certain residues, as proline, serine or threonine, are known to induce local distortions in the a-helical structure. Analysis of sequence databases of transmembrane segments evidence that certain combinations of these residues are more common than others, and that some of them are significantly over-represented, while others are clearly under-represented. The focusing this sequence analysis on the transmembrane region of Class A GPCRs illustrates that these combinations are located in some specific locations and conserved within certain subfamilies of receptors.The structure and dynamics of transmembrane a-helices containing these combinations of proline and serine or threonine have been studied using molecular dynamics simulations in an explicit hydrophobic environment. The results show how some of these combinations induce significant distortions in the a-helical structure, due to their effect on the hydrogen bond network that stabilizes the helix.These results have been applied to the building of a three-dimensional model of the chemokine CCR5 receptor, using homology modelling techniques. In this model, transmembrane helices (TMH) 2 and 3 of CCR5 are proposed to be different from the bovine rhodopsin template. TMH2 is more bent due to the presence of a Thr-X-Pro motif, which, in turn, induces this helix to lean towards TMH3. As a consequence, an interaction between these two helices is proposed for this particular receptor. This interaction would be mediated through the presence of specific and conserved hydrophobic and aromatic residues in both helices. These hypothesis have been tested through site-directed mutagenesis experiments, thanks to a collaboration with the Institut de Recherche Interdisciplinaire en Biologie Humaine et Nucléaire (IRIBHN), Université Libre de Bruxelles. The experimental results let us to hypothesize that the TMH2-TMH3 interface is involved in the chemokine-induced activation of the CCR5 receptor.As a conclusion, this thesis aims to show how through the use of bioinformatics tools, primary sequences of proteins and interactions at an atomic level can be translated to three-dimensional protein structures. In addition, this thesis illustrates that, even though the three-dimensional structure of bovine rhodopsin is a very useful template for homology modelling of GPCRs, the sequence specificities of each receptor have to be explicitly taken into account in order to build models. These sequence specificities consist in sequence patterns conserved within certain families, which are translated into structural divergences. Among these sequence patterns, we hypothesize that serine and threonine, alone or combined with nearby proline residues, can modulate the geometry of TMHs, due to its capability to interfere with the hydrogen bond network that stabilize a-helices.Finally, we propose that the influence of serine, threonine and proline motifs in the structure of TMHs may be related to processes of activation in the Class A of GPCRs, and, possibly, other membrane proteins as well. In GPCRs, these motifs may have evolved in order to adapt a conserved mechanism of activation of the G protein to the cognate ligands of each receptor family.

Carbohydrates had historically been associated to two biological functions: energy storage and structural support. However, in the last decades, new complex structures of oligosaccharides have been found to play vital roles in many biological processes, such as signal transduction, immune response, cell differentiation and cancer development, among others. Advances in the functional understanding of carbohydrate-protein interactions represented a breakthrough in the field of glycobiology and glycochemistry, opening a new branch of potential therapeutic targets (carbohydrate acting enzymes), glycomimetic drugs and biomarkers. The bottleneck in the field of glycochemistry is the synthesis of complex saccharides; hence many efforts have been devoted to the development of novel enzymatic strategies for carbohydrate synthesis. Glycoside transferases (GT) and glycoside hydrolases (GH) are the enzymes that catalyze the formation and the cleavage of the glycosidic linkage respectively. They are used in complex oligosaccharides synthesis, and recently they have been engineered to produce enzymes with particular substrate specificities or even activities. In spite of these advances, the understanding of the molecular mechanisms of enzymatic carbohydrate synthesis and degradation is far from complete. Structural studies have shown that the puckering of the sugar ring at the cleavage point must change during catalysis. Knowing the conformational catalytic itinerary has an impact in the design of GHs inhibitors. However, these itineraries are not known for all families of GHs. On the other hand, the saccharide puckering is not an issue in GTs, but the reaction mechanism is not known. In fact, the glycosidic bond formation in GTs remains one of the most intriguing and unanswered questions in the field of glycobiology. The coming of age of powerful theoretical methods such as quantum mechanics / molecular mechanics (QM/MM) and ab initio molecular dynamics (AIMD) has enabled the elucidation of complex reactive processes in proteins and enzymes. In particular, the modeling of the Michaelis complex and the reaction mechanisms of GHs highlighted the interplay between electronic and structural changes that preactivate the substrate for catalysis. Some of these changes can already be anticipated by analyzing the conformational energy landscape of the substrate. Part of the research of this Thesis complements previous studies of our group by analyzing the factors that govern substrate distortion in GHs. In this respect, it extends the use of conformational free energy landscapes of simple sugars to predict the conformation of the substrate in Michalis complexes. Additionally, the molecular mechanism of retaining glycoside transferases is elucidated. This Thesis is organized as follows: Chapter I contains an introduction of the enzymes studied (GHs and GTs) and presents the main objectives of this work. The theoretical methods used are detailed in Chapter II. Chapters III to V are focused on enzyme-substrate interactions affecting the conformation of the substrate in GHs. Concretely; in Chapter III we test how mutation of the acid/base catalytic residue, the use of a substrate-like thio-analogue inhibitor or fluorometric aglycons affects the distortion of the substrate. In Chapter IV we study the influence of the enzyme-substrate interactions through the 2-OH, in particular the effect of the commonly used 2-deoxy-2-fluoro substitution. The conformational itinerary of this inhibitor during catalysis is modeled in Chapter V. In Chapter VI, the conformational flexibility of β-D-mannopyranose and α-L-fucopyranose molecules is investigated. The topologies of their corresponding conformational free energy landscapes are related with the observed crystallographic structures of β-mannosidases and α-fucosidases, and the predictive potential of such calculations is discussed. Chapter VII focuses on trehalose 6-phosphate synthase (a family 20 retaining GT that belongs to fold type B). The mechanism of glycosidic bond formation in this enzyme is elucidated. Finally, in Chapter VI, the main conclusions of this work are summarized.

El cáncer de mama es una enfermedad con elevada incidencia, prevalencia y mortalidad que puede estar influida por factores nutricionales, entre los que destacan los lípidos de la dieta. A su vez, en cáncer se han descrito alteraciones del epigenoma. Los patrones epigenéticos son susceptibles de ser modificadas por factores ambientales. Así, el objetivo del presente trabajo ha sido definir cambios epigenéticos implicados en el desarrollo del cáncer de mama, así como determinar si el estilo de vida, y especialmente los hábitos dietéticos en relación al consumo de lípidos, influyen en este tipo de neoplasia a través de la alteración de patrones epigenéticos. Dicho objetivo se ha desarrollado en dos líneas de investigación, una en cáncer de mama humano y otra en cáncer de mama experimental. El estudio en humanos se ha realizado en una población de mujeres sanas y de enfermas de cáncer de mama, caracterizadas a nivel clínico y de estilo de vida, a partir de muestras de sangre periférica de las sanas y de sangre periférica, glándula mamaria y tumor de las enfermas. En dichas muestras se determinó la metilación del ADN a nivel global y de un panel de 12 genes implicados en los “hallmarks” del cáncer (BRCA1, p16, RARβ2, ESR1, PGR, RASSF1A, NES1, TWIST1, MASPINA, CDH1, CXCL12 y HLA-A). Por su parte, el estudio experimental se ha desarrollado en el modelo de cáncer de mama inducido con DMBA en la rata Sprague-Dawley, a partir de muestras de glándula mamaria y tumor de animales alimentados con diferentes dietas hiperlipídicas (rica en aceite de maíz o rica en aceite de oliva virgen extra). Por lo que respecta al efecto del cáncer sobre los patrones epigenéticos, los resultados obtenidos mostraron que la influencia que ejerce la propia enfermedad sobre el epigenoma alteraría mecanismos epigenéticos comunes, en términos generales, entre el cáncer de mama humano y el experimental. Estas alteraciones se caracterizarían, entre otras, por el incremento de la metilación gen-específica asociado a un aumento de la actividad ADN metiltransferasa, la disminución de la expresión de genes supresores tumorales (RASSF1A y TIMP3) y la alteración de modificaciones postraduccionales de las histonas (H3K4me2, H3K27me3, H4K20me3 y H4K16ac). Estos cambios, además, podrían estar influidos por factores dietéticos y de estilo de vida, como el consumo de alcohol, la realización de actividad física, la ingesta calórica total así como de proteínas y vitaminas B2, B6 y B12 y, especialmente, de lípidos (tanto en cantidad total como en función del tipo). Asimismo, la influencia de estos y otros factores de riesgo relacionados con la reproducción sobre el desarrollo de cáncer de mama, tales como la edad en la primera gestación y de aparición de la menopausia, podría ser, entre otros mecanismos, a través de la alteración de los patrones epigenéticos. El interés de hallar que determinados factores de riesgo, incluido el estilo de vida, alteran dichos patrones se basa en que los mecanismos epigenéticos, a diferencia de los genéticos, son modificables. Así, los resultados del presente trabajo permiten formular opiniones científicas sobre la importancia que tienen los hábitos dietéticos y el estilo de vida en relación a la salud o al riesgo de enfermedad. En este sentido, se podrían definir factores de riesgo y/o protectores a los que está sometida la población en base a sus hábitos alimenticios en relación al consumo de grasas. Por todo ello, este trabajo en su conjunto se enmarcaría en el campo de la prevención primaria y secundaria del cáncer de mama. / Breast cancer is a disease with high incidence, prevalence and mortality, that may be influenced by nutritional factors, and especially dietary lipids. Furthermore, disruption of epigenome is a major change occurring in all types of cancers. Epigenetic patterns are reversible and may be modified by environmental factors. Thus, the aim of this study was to define epigenetic changes involved in breast cancer and determine if lifestyle, particularly dietary habits in relation to fat intake, influence this type of neoplasia through the alteration of epigenetic patterns. This objective has been developed in two lines of research, one in human breast cancer and the other in an experimental model of breast cancer. The human study was conducted in a population of healthy volunteers and breast cancer patients, who were characterized clinically and in which we evaluated their lifestyle. We obtained blood from healthy volunteers and blood, mammary gland and tumor from breast cancer patients. In such samples we determined global DNA methylation and gene-specific methylation of a panel of 12 genes with an important role on the key hallmarks of cancer (BRCA1, p16, RARβ2, ESR1, PGR, RASSF1A, NES1, TWIST1, MASPINA, CDH1, CXCL12 and HLA-A). On the other hand, the experimental study was developed in the rat dimethylbenz(a)anthracene (DMBA)-induced breast cancer model, using samples from mammary gland and tumor from animals fed different high fat diets (rich in corn oil or in extra virgin olive oil). In relation to the effect of cancer on the epigenetic patterns, the results showed that the influence of the disease on the epigenome alter common epigenetic mechanisms, in general, in human and experimental breast cancer. These alterations would be characterized, among others, by the increase of gene-specific DNA methylation associated with an increase in DNA methyltransferase activity, the decrease of tumor-suppressor genes expression and the alteration of post-translational histone modifications (H3K4me2, H3K27me3, H4K20me3 and H4K16ac). These changes could be influenced by dietary and lifestyle factors, such as alcohol consumption, physical activity, calorie intake, protein and vitamins B2, B6 and B12 consumption, and especially lipid intake (including total amount and type of lipid). In addition, the influence on the development of breast cancer of these and other risk factors related to reproduction, such as age at first pregnancy and onset of menopause, could be done, among other mechanisms, through the alteration of epigenetic patterns. Finding that certain risk factors, including lifestyle, alter these patterns is of interest since epigenetic mechanisms, unlike genetic, are modifiable. Thus, the results of this study allow to formulate scientific opinions about the importance of dietary habits and lifestyle in relation to health or disease risk. In this sense, they could be defined risk and/or protective factors to which the population is exposed based on their dietary habits in relation to fat intake. Therefore, this work as a whole would be framed in the field of primary and secondary prevention of breast cancer.

As espécies reativas de oxigênio geradas in vivo têm sido implicadas em vários mecanismos como apoptose, crescimento e diferenciação celular, câncer e inflamação. O estresse oxidativo é o produto do desbalanço entre os agentes prooxidantes e antioxidantes celulares, causando danos por meio da peroxidação lipídica, oxidação de proteínas e de DNA. Neste trabalho avaliou-se o efeito do estresse oxidativo em glândulas submandibulares de rato e em células acinares dispersas tratadas com isoproterenol (ISO), cicloheximida (CHX), carbacol (CA) e propanolol (PROP). Observou-se aumento dos níveis de MDA em homogenados de glândulas de ratos tratados com ISO e CHX e em células acinares dispersas incubadas com ISO e CHX. Tratamentos semelhantes causaram aumento na oxidação protéica em cultura de células, mas o mesmo aumento não ocorreu in vivo. Em relação aos agentes antioxidantes, observou-se um aumento no poder redutor (conteúdo de GSH) em homogenados de glândulas submandibulares tratadas com ISO e em células dispersas incubadas com ISO. A atividade de superóxido dismutase (SOD) também aumentou em tratamentos com ISO in vivo e os \"Western Blots\" mostraram uma variação acentuada nos níveis protéicos da isoforma MnSOD. A atividade de SOD, induzida por ISO em células dispersas, não ocorre quando as células foram pré-incubadas com PROP, o que sugere a ativação da atividade de SOD via receptores β-adrenérgicos. / Reactive oxygen species (ROS) have been implicated in many mechanisms as apoptosis, cell growth and differentiation, cancer and inflammation. Oxidative stress is the product of the imbalance of the prooxidants and antioxidants agents within the cell and it has been shown to cause damage by producing peroxidation of fatty acid and oxidation of proteins and DNA. In this work the oxidative stress was evaluated on rat submandibular glands and on dispersed acinar cells treated with isoproterenol (ISO), cycloheximide (CHX), carbachol (CA) and propanolol (PROP). The level of lipid peroxidation on rats submandibular glands and on dispersed acinar cells were increased by ISO and CHX treatment. Protein oxidation level was increased on cell culture treated with ISO, but the same results was not found in vivo treatments. Regards the antioxidant agents, the GSH content was found to increase on both in vivo and on cell culture by ISO treatment. The superoxide dismutase (SOD) activity in vivo increased and the Western Blots showed a significant MnSOD variation level. The SOD was induced by ISO treatment on cell culture, but the same increasing was not found with previous cells treatment with PROP. This result suggest the involvement of a β-adrenergic receptor on SOD activity.

Orientador: Elso Drigo Filho / Banca: Álvaro de Souza Dutra / Banca: Gerald Weber / Banca: Carla Goldman / Banca: Jorge Chaine / Resumo: Este trabalho apresenta um estudo termodinâmico de modelos de redes unidimensionais, usados para simular o DNA. Primeiramente, o modelo de Peyrard-Bishop (PB) com o potencial "Corcunda" on site foi utilizado para analisar a termodinâmica da molécula. Com o método do operador integral de transferência foram obtidas as propriedades termodinâmicas do sistema e as soluções da equação tipo - Schrödinger que emerge desse formalismo foram determinadas pelo método variacional. Com os parâmetros do potencialnormalmente utilizados na literaturao valor obtido para a temperatura de desnaturação foi extremamente alto. Por isso, são sugeridos diferentes parâmetros para descrever a termodinâmica da molécula de DNA com esse modelo. Além disso, também é estudada uma das extensões do modelo original de PB, na qual a interação de empilhamento puramente harmônica é modificada por um potencial não harmônico. São apresentados os passos e as aproximações necessárias para determinar a equação tipo - Schrödinger com massa dependente da posição que descreve as propriedades termodinâmicas desse modelo. As soluções dessa equação são determinadas a partir de uma adaptação do método variacionalpara o estudo desse problema. Para adquirir confiança na aplicação do método na solução da equação de Schrödinger com massa dependente da posição, alguns exemplos da literatura foram resolvidos.Por fim,a termodinâmica da extensão do modelo de PB foi determinada com a aplicação desse método semi - analítico. Os resultados mostraram que é possível utilizar o método variacional para solucionar esse tipo de problema e quea transição de fase da molécula ocorre em uma temperatura compatível com a apresentada na literatura / Abstract: This work shows a thermodynamical study of one-dimensional lattice models, used to simulate the DNA. The Peyrard-Bishop (PB) model with the "Hump" potential on site was used to analyze the molecule thermodynamics. The thermodynamical properties of the system were obtained with the transfer integral operator method and the solutions of the type-Schrödinger equation that emerges from the formalism were determined with the variational method. With the potential parameters normally used in the literature the obtained value to the denaturation temperature was extremely high. Because of this, it is suggested different parameters to describe the thermodynamics of the DNA molecule. Besides, it is also studied an extension of the original PB model, in which the purely harmonic stacking interaction is modified to a non-harmonic potential. The steps and necessary approximations to determine the type-Schrödinger equation with position-dependent mass that describes the thermodynamical properties of this model are shown. The solutions of this equation are determined from an adaptation of the variational method to the study of this kind of problem. To gain confidence in the application of this method in the solution of the Schrödinger equation with position-dependent mass, some examples of the literature were solved. Finally, the thermodynamic of the extension of the PB model was determined with the application of this semi-analytical method. The results showed that it is possible to use the variational method to solve this kind of problem and that the phase transition of the molecule occurs in a temperature in agreement with the one present in the literature / Doutor