Las propiedades del receptor nicotínico colinérgico muscular (AChR) están reguladas por los lípidos de la membrana. Por tratarse de una proteína integral, desde su síntesis hasta su aparición en la sinapsis y posterior degradación, el AChR esta inmerso en una bicapa lipídica, cuya composición está pre-determinada y finamente controlada por la célula. Por lo tanto, el análisis funcional de la relación entre los lípidos y el AChR requiere de un sistema celular que funcione a modo de tubo de ensayo, para evitar los modelos artificiales puros con membranas reconstituídas. Con este propósito, generamos modelos celulares que expresan heteróloga y onstitutivamente el AChR y en los que se puede modificar parcial y selectivamente el contenido de ciertas especies lipídicas. En una primera etapa a través de la cotransfección del ADNc de las subunidades del AChR adulto de ratón en una línea celular con metabolismo lipídico normal, CHO-K1 (Forsayeth et al., 1990c), establecimos el sistema de referencia. Posteriormente, generamos dos modelos experimentales con líneas celulares mutantes que eran deficientes en el metabolismo lipídico: una línea mutante de CHO-K1, denominada SPB-1deficiente en esfigomielina (SM) (Hanada et al., 1990) y otra línea derivada de CHO-K1, auxotrófica, llamada PSA-3, deficiente en fosfatidilserina (PS). Obtuvimos la estirpe SPB-1/SPH que expresa el AChR y modula, por efecto de la temperatura, la actividad de la enzima serina palmitoil transferasa (SPT), clave en la vía de síntesis de los esfingolípidos. A temperaturas de cultivo entre 33 C y 37 C la actividad de la enzima representa sólo del 4 al 8% de la actividad normal, lo cual disminuye los niveles celulares de SM y gangliósidos como el GM3 a valores menores del 1%. Con este modelo, determinamos que no existen diferencias en los parámetros electrofisiológicos y en las propiedades farmacológicas del receptor, y caracterizamos la participación activa de la SM en los mecanismos que regulan el tráfico exocítico de la proteína a la membrana plasmática. También generamos la línea clonal PSA/AChR, que expresa el AChR y no es capaz de realizar normalmente la síntesis de PS por tener alterada una de las enzimas que catalizan su síntesis. Por este motivo, requiere del agregado exógeno del fosfolípido para su normal crecimiento. Transfectadas las células y seleccionados los clones positivos, analizamos el ARN por RT-PCR para determinar la expresión de las subunidades. Con los clones elegidos realizamos ensayos de unión a [125I]α-bungarotoxina, un radioligando específico para AChR, determinándose los parámetros cinéticos y farmacológicos. Los parámetros electrofisiológicos del canal se caracterizaron por la técnica de patch-clamp y el análisis topológico de la expresión por microscopía de fluorescencia. De esta manera, en cada modelo celular caracterizamos las diferencias estructurales y topológicas del AChR causadas por las variaciones en el contenido lipídico. En el segundo capítulo de esta tesis analizamos el rol que poseen los aminoácidos que están ubicados, de cara a la membrana, en contacto directo con los lípidos. Con este propósito, modificamos la estructura primaria de la proteína por medio de mutaciones simples y dobles, en residuos cargados y muy conservados, situados en los extremos del segmento M4 a la altura de las cabezas polares de los fosfolípidos. Por co-transfección de una subunidad mutada con el resto de las subunidades en la versión salvaje, generamos cinco líneas clonales mutantes. En estos modelos se realizaron ensayos farmacológicos con [125I]-BTX, microscopía de fluorescencia y técnicas bioquímicas que nos permitieron determinar la magnitud de la expresión del receptor mutado y los cambios estructurales y funcionales que se produjeron cuando los diferentes residuos del AChR fueron mutados. / The properties of the muscle nicotinic acetylcholine receptor (AChR) are modulated by membrane lipids. As an integral protein, the AChR - from synthesis to appearance in synapses and further degradation- is immersed in a lipid bilayer. The composition of this bilayer is pre-established and finely controlled by the cell. Therefore, the functional analysis of the lipid-AChR relationship requires a cell system that behaves as test tube, to prevent pure artificial models with reconstituted membranes. For this purpose, the AChR was heterologoulsy and constitutively expressed in cell lines, where the lipid content can be partially and selectively modified. In a first stage, the reference system was established by means of transfection of adult mouse AChR subunits cDNAs into a cell line with normal lipid metabolism, CHO-K1 (Forsayeth et al., 1990c). Then, two experimental models were generated with mutant cell lines which were defective in lipid metabolism: a CHO-K1 mutant line, called SPB-1, deficient in sphyngomyelin (SM) (Hanada et al., 1990) and another auxotrophic line derived from CHO-K1, called PSA-3 and phosphatydilserine deficient (PS). The SPB-1/SPH clone was obtained, which expresses the AChR and modulates, by effect of temperature, the activity of the serine palmitoyl transferase (SPT), which is key to the sphingolipids synthesis pathway. At culture temperatures between 33 C and 37 C, the enzyme activity represents only from 4% to 8% of normal activity, thus reducing cell levels of SM and gangliosides such as GM3 to values lower than 1%. With this model, we established that the electrophysiological parameters and the pharmacological properties of the receptor show no differences. However, we were able to characterize the active participation of SM in the mechanisms that regulate the exocytic protein traffic to the plasma membrane. In addition, a PSA/AChR clone line was generated, which expresses the AChR and is not capable to synthesize PS normally, because one of the enzymes that catalyzes this synthesis is altered. Thus, it requires exogenous PS for normal growth. Once the cells were transfected and the positive clones selected, we analyzed the RNA by RT-PCR to determine the expression of the subunits. With these clones, we performed binding studies with [125I]-bungarotoxin, an AChR specific radioligand, to determine kinetic and pharmacological parameters. The electrophysiological parameters of the channel were characterized by the patch-clamp technique and the topological analysis of the expression by fluorescence microscopy. In this way, on each cell line, we characterized the structural and topological AChR differences caused by lipid content variations. In a second stage of this Thesis, the role of the aminoacids which are located at the membrane facing extremes in direct contact with the lipids was analyzed. For this purpose, we modified the primary structure of the protein by means of simple and double mutations, in highly conserved and charged residues, located at the M4 segment extremes at the level of the phospholipid polar heads. By transfection of a mutated subunit together with the rest of the subunits in their wild-type version, we generated five mutant clone lines. These models were subjected to pharmacological tests with [125I]-BTX, fluorescence microscopy and biochemical techniques that allowed us to establish the expression amount of the mutated receptor and the structural and functional changes that took place when the different residues were mutated.

En esta Tesis doctoral se profundizó en el estudio de la interacción lípidoreceptor de acetilcolina nicotínico (AChR) en dos aspectos: por un lado, el mecanismo de inhibición de ácidos grasos libres (AGLs), antagonistas no competitivos del AChR, y por el otro, la ubicación del AChR en dominios líquidoordenados (Lo) condicionada por dos características particulares de la membrana. Con la finalidad de dilucidar el mecanismo de antagonismo de los AGLs sobre el AChR, se utilizaron AGLs con un doble enlace único en diferentes posiciones de una cadena acílica de 18 átomos de carbono. Estudios funcionales realizados con la técnica de patch-clamp han mostrado que solo el cis-6-18:1 y el cis-9-18:1 reducen la duración del estado de canal abierto del receptor, sugiriendo, por lo tanto, un mecanismo de bloqueo alostérico del canal. Mediante espectroscopía de fluorescencia se comprobó que todos los AGLs se ubican en la interfase lípido-AChR, quedando el cis-6-18:1 restringido a los sitios denominados sitios anulares, mientras que el resto de los AGLs ocupa también sitios no-anulares. Por otro lado, estudios de polarización de fluorescencia mostraron que el AGLs cis-9-18:1 es el que ocasiona el mayor desorden en la membrana. Se comprobó i) que todos los cis-AGLs generan cambios conformacionales del AChR a nivel transmembrana, ii) que los cis-9-18:1, cis-11- 18:1 y cis-13-18:1 perturban al AChR en su estado de reposo e iii) que los cis-6-18:1 y cis-9-18:1 son los que causan una mayor perturbación del estado desensibilizado. De esta manera, la posición e isomería del ángulo de torsión de los AGLs insaturados serían un factor clave en el bloqueo del AChR, sugiriendo entonces que los AGLs con un único doble enlace y ubicados superficialmente en la membrana inhiben en forma directa la función del AChR, posiblemente al perturbar una secuencia aminoacídica transmembrana involucrada en los cambios alostéricos necesarios para la apertura del canal iónico. Se postula que en la membrana plasmática el AChR se encuentra en dominios lipídicos denominados “balsas” (“rafts”). Sin embargo, el AChR no muestra preferencia por dominios Lo en sistemas modelo compuestos por esfingomielina (SM), colesterol (Col) y POPC (1:1:1), pero sí lo hace un segmento transmembrana (γM4) que exhibe mayor contacto con los lípidos. Es decir, su distribución no dependería exclusivamente de sus propiedades intrínsecas sino también de señales extrínsecas a la proteína. En este trabajo de Tesis se estudió la posible partición diferencial del AChR en los dominios Lo en dos sistemas modelo diferentes en función de: a) la presencia de diferentes especies puras de SM en la membrana, y b) la existencia de asimetría transbicapa en el sistema modelo, mediante el agregado de SM de cerebro (bSM) en la hemicapa externa. Tanto la existencia de asimetría como la presencia de 16:0-SM o 18:0-SM, en comparación con las bSM y 24:1-SM, producen una partición preferencial del AChR en los dominios Lo. De este modo, la localización del AChR en estos dominios depende no solo de sus propiedades sino también de las características propias de la membrana en la que se encuentra. Entender la interacción lípido-AChR es de gran importancia para determinar tratamientos que puedan mejorar o inhibir la función del AChR y tratar enfermedades que lo involucren. / In this Ph. D. thesis the understanding of the lipid-niconitic acetylcholine receptor (AChR) interaction was furthered in two aspects, namely the inhibition mechanism of free fatty acids (FFA), non-competitive AChR antagonists, and AChR location in liquid-ordered (Lo) domains conditioned by two membrane characteristics. To elucidate FFA’s antagonism mechanism, FFA with only one double bond in different positions of an 18-carbon acyl chain were tested on AChR. Patch-clamp functional studies showed that only cis-6-18:1 and cis-9-18:1 reduce the duration of the AChR open state, thus suggesting an allosteric blocking mechanism. Fluorescence spectroscopy measurements demonstrated that all FFA locate in the AChR-lipid interfase, with cis-6-18:1 restricted to anular sites, while the rest of the FFA tested also ocupy non-anular sites. Fluorescence polarization studies showed that cis-9-18:1 causes the highest membrane disorder of all FFA tested. It was determined that i) all cis-FFA generate AChR conformational changes at a transmembrane level, ii) only cis-9-18:1, cis-11-18:1 and cis-13-18:1 disturb AChR resting state and iii) cis-6-18:1 and cis-9-18:1 are the ones that cause the highest disturbance of the desensitized state. Thus, the position and isomerism of the torsion angle of unsaturated FFAs are probably a key factor in terms of AChR blockage, possibly by perturbing a transmembrane aminoacidic sequence involved in the allosteric changes necessary for ion channel gating. In the plasma membrane, AChR is postulated to be located in lipid domains known as rafts. However, AChR shows no preference for Lo domains in model systems – composed of sphingomyelin (SM), cholesterol (Chol) and POPC (1:1:1) –, but a transmembrane segment (γM4) that in closest contact with lipids does have preference for them. This means that AChR distribution seems not to exclusively depend on its intrinsic properties but on signals external to the protein. In this Ph. D. thesis a posible differential AChR partitioning in Lo domains was studied in two model systems as a function of a) the presence of different pure SM species in the membrane and b) the existence of transbilayer asymmetry in the model system, by the addition of brain SM (bSM) in the external hemilayer. Both asymmetry and the presence of either 16:0-SM or 18:0-SM, in comparison with bSM or 24:1-SM, lead to an AChR preferential partitioning in Lo domains. AChR location in these domains depends not only on its properties but also on the characteristics of the membrane in which the ion channel is immersed. Understanding lipid-AChR interaction is of great importance to determine treatments that can either improve or inhibit AChR function and, this, in turn, will help determining the treatment of diseases in which AChR is involved.

Introdução: A grande variabilidade individual em resposta a fármacos antitabagismo sugere que tratamentos específicos podem ser mais efetivos em determinados subgrupos de fumantes. No contexto de medicina personalizada, o principal objetivo do presente estudo foi avaliar se polimorfismos nos genes CHRNA4, CHRNB2, CYP2B6 e ANKK1 estão associados com a resposta às terapias de cessação tabágica em pacientes provenientes de um programa de assistência ao fumante. Métodos: Estudo de coorte com 483 pacientes fumantes que receberam tratamento farmacológico (vareniclina, vareniclina e bupropiona, bupropiona em monoterapia ou coadministrada com terapia de reposição nicotínica). O sucesso na cessação tabágica foi considerado para os pacientes que completaram 6 meses de abstinência contínua. O teste de Fagerström para a dependência à nicotina (FTND) e o escore de consumo situacional Issa foram utilizados para avaliar a dependência à nicotina. Os polimorfismos CHRNA4 (rs1044396 e rs2236196), CHRNB2 (rs2072660 e rs2072661) e ANKK1 (rs1800497) foram genotipados pela análise da curva de melting e os polimorfismos CYP2B6 *9 (rs3745274), *4 (rs2279343), *5 (rs3211371) foram genotipados por restrição enzimática. Resultados: Os pacientes com o genótipo CC para o polimorfismo CHRNA4 (rs10443196) obtiveram menor taxa de sucesso no tratamento com vareniclina (29,5%) em comparação com os portadores dos genótipos CT ou TT (50,9%) (P=0,007; n=167). Os genótipos CT ou TT foram associados com maior odds ratio para o sucesso (OR=1,67; IC 95%=1,10-2,53; P=0,02), em um modelo multivariado. Os pacientes com o genótipo AA para o polimorfismo CYP2B6 (rs2279343) obtiveram maior taxa de sucesso no tratamento com bupropiona (48,0%) em comparação com portadores dos genótipos AG ou GG (35,5%) (P=0,05; n=237). O genótipo AA foi associado com maior odds ratio para o sucesso no tratamento (OR=1,92; IC 95%=1,08-3,42; P=0,03), em um modelo multivariado. Não foram observadas diferenças significativas nos escores FTND e Issa com relação aos polimorfismos estudados. Conclusão: Os polimorfismos CHRNA4 (rs1044396) e CYP2B6 (rs2279343) estão associados com a cessação tabágica em indivíduos tratados com vareniclina e bupropiona, respectivamente. Sugere-se que estes polimorfismos influenciam a resposta farmacológica e podem ser importantes para o desenho de uma farmacoterapia individualizada / Background: The large individual variability in response to drugs for smoking cessation suggests that specific treatments can be more effective in particular subgroups of smokers. In the context of personalized medicine, the main aim of the present study was to evaluate whether the CHRNA4, CHRNB2, CYP2B6 and ANKK1 polymorphisms are associated with response to smoking cessation therapies in patients from a smoker assistance program. Methods: This cohort study enrolled 483 smoking patients patients who received pharmacological treatment (varenicline, varenicline plus bupropion, bupropion in monoterapy or plus nicotine replacement therapy). Smoking cessation success was considered for patients who completed 6 months of continuous abstinence. Fagerström test for nicotine dependence (FTND) and Issa situational smoking scores were analyzed for nicotine dependence. The CHRNA4 (rs1044396 and rs2236196), CHRNB2 (rs2072660 and rs2072661) and ANKK1 rs1800497 polymorphisms were genotyped by high resolution melting analysis and the CYP2B6 *9 (rs3745274), *4 (rs2279343) and *5 (rs3211371) were genotyped by restriction fragment lenght polymorphisms. Results: Patients with CHRNA4 rs1044396 CC genotype had lower success rate in treatment with varenicline (29.5%) compared with carriers of CT or TT genotypes (50.9%) (P=0.007, n=167). The CT or TT genotypes were associated with higher odds ratio for success (OR=1.67, 95%CI=1.10-2.53, P=0.02), in a multivariate model. Patients with CYP2B6 rs2279343 AA genotype had higher success rate in treatment with bupropion (48.0%) compared with carriers of AG or GG genotypes (35.5%) (P=0.05, n=237). The AA genotype was associated with higher odds ratio for success (OR=1.92, 95%CI=1.08-3.42, P=0.03), in a multivariate model. We did not observe significant differences in the FTND and Issa scores according to the studied polymorphisms. Conclusion: The CHRNA4 rs1044396 and CYP2B6 rs2279343 are associated with smoking cessation in individuals on varenicline and bupropion terapies, respectively. We suggest that these polymorphisms influence the pharmacological response of these drugs and it might be important in the design of individualized pharmacotherapy

Introdução: A grande variabilidade individual em resposta a fármacos antitabagismo sugere que tratamentos específicos podem ser mais efetivos em determinados subgrupos de fumantes. No contexto de medicina personalizada, o principal objetivo do presente estudo foi avaliar se polimorfismos nos genes CHRNA4, CHRNB2, CYP2B6 e ANKK1 estão associados com a resposta às terapias de cessação tabágica em pacientes provenientes de um programa de assistência ao fumante. Métodos: Estudo de coorte com 483 pacientes fumantes que receberam tratamento farmacológico (vareniclina, vareniclina e bupropiona, bupropiona em monoterapia ou coadministrada com terapia de reposição nicotínica). O sucesso na cessação tabágica foi considerado para os pacientes que completaram 6 meses de abstinência contínua. O teste de Fagerström para a dependência à nicotina (FTND) e o escore de consumo situacional Issa foram utilizados para avaliar a dependência à nicotina. Os polimorfismos CHRNA4 (rs1044396 e rs2236196), CHRNB2 (rs2072660 e rs2072661) e ANKK1 (rs1800497) foram genotipados pela análise da curva de melting e os polimorfismos CYP2B6 *9 (rs3745274), *4 (rs2279343), *5 (rs3211371) foram genotipados por restrição enzimática. Resultados: Os pacientes com o genótipo CC para o polimorfismo CHRNA4 (rs10443196) obtiveram menor taxa de sucesso no tratamento com vareniclina (29,5%) em comparação com os portadores dos genótipos CT ou TT (50,9%) (P=0,007; n=167). Os genótipos CT ou TT foram associados com maior odds ratio para o sucesso (OR=1,67; IC 95%=1,10-2,53; P=0,02), em um modelo multivariado. Os pacientes com o genótipo AA para o polimorfismo CYP2B6 (rs2279343) obtiveram maior taxa de sucesso no tratamento com bupropiona (48,0%) em comparação com portadores dos genótipos AG ou GG (35,5%) (P=0,05; n=237). O genótipo AA foi associado com maior odds ratio para o sucesso no tratamento (OR=1,92; IC 95%=1,08-3,42; P=0,03), em um modelo multivariado. Não foram observadas diferenças significativas nos escores FTND e Issa com relação aos polimorfismos estudados. Conclusão: Os polimorfismos CHRNA4 (rs1044396) e CYP2B6 (rs2279343) estão associados com a cessação tabágica em indivíduos tratados com vareniclina e bupropiona, respectivamente. Sugere-se que estes polimorfismos influenciam a resposta farmacológica e podem ser importantes para o desenho de uma farmacoterapia individualizada / Background: The large individual variability in response to drugs for smoking cessation suggests that specific treatments can be more effective in particular subgroups of smokers. In the context of personalized medicine, the main aim of the present study was to evaluate whether the CHRNA4, CHRNB2, CYP2B6 and ANKK1 polymorphisms are associated with response to smoking cessation therapies in patients from a smoker assistance program. Methods: This cohort study enrolled 483 smoking patients patients who received pharmacological treatment (varenicline, varenicline plus bupropion, bupropion in monoterapy or plus nicotine replacement therapy). Smoking cessation success was considered for patients who completed 6 months of continuous abstinence. Fagerström test for nicotine dependence (FTND) and Issa situational smoking scores were analyzed for nicotine dependence. The CHRNA4 (rs1044396 and rs2236196), CHRNB2 (rs2072660 and rs2072661) and ANKK1 rs1800497 polymorphisms were genotyped by high resolution melting analysis and the CYP2B6 *9 (rs3745274), *4 (rs2279343) and *5 (rs3211371) were genotyped by restriction fragment lenght polymorphisms. Results: Patients with CHRNA4 rs1044396 CC genotype had lower success rate in treatment with varenicline (29.5%) compared with carriers of CT or TT genotypes (50.9%) (P=0.007, n=167). The CT or TT genotypes were associated with higher odds ratio for success (OR=1.67, 95%CI=1.10-2.53, P=0.02), in a multivariate model. Patients with CYP2B6 rs2279343 AA genotype had higher success rate in treatment with bupropion (48.0%) compared with carriers of AG or GG genotypes (35.5%) (P=0.05, n=237). The AA genotype was associated with higher odds ratio for success (OR=1.92, 95%CI=1.08-3.42, P=0.03), in a multivariate model. We did not observe significant differences in the FTND and Issa scores according to the studied polymorphisms. Conclusion: The CHRNA4 rs1044396 and CYP2B6 rs2279343 are associated with smoking cessation in individuals on varenicline and bupropion terapies, respectively. We suggest that these polymorphisms influence the pharmacological response of these drugs and it might be important in the design of individualized pharmacotherapy