Metabolic systems biology of Leishmania major

Tese de Doutoramento em Engenharia Química e BiológicaProtozoan parasitic diseases such as leishmaniasis, toxoplasmosis, and sleeping sickness are one of the major causes of death worldwide. The emerging resistance of parasitic species and adaptive mechanisms of infection are a major concern in developing medical treatment. Understanding the genotype/phenotype of parasites during infection can help in developing effective anti-parasitic therapies. In recent years, systems biology approaches, in particular, genome-scale metabolic modelling has been proposed to attain such understanding. This methodology allows incorporating omics data (e.g. transcriptomics, proteomics and metabolomics) to understand stage-specific metabolism of many organisms including protozoan parasites such as Leishmania, Toxoplasma, and Plasmodium. Metabolic behaviour during infection can be characterized, including the metabolic involvement of small or complex carbohydrates that has been poorly studied, so far, at the systems level. This thesis aims to present a comprehensive understanding of protozoan parasites metabolism and specifically the influence of glycans and glycoconjugates during infection. First, a review on different data types and methodologies used to study glycans and glycoconjugates from their structures to functions in protozoan parasites is presented. Glycobiology databases, in particular, glycomic and glycoproteomic data resources, were extensively reviewed, addressing problems in accessing and integrating data due to inconsistencies in the identification and representation of glycans, as well as the poor inter-linkage between databases. The focus is provided on graphic and text-based glycan structural notations, exploiting available tools to interconvert these encoding formats, in order to improve inter-linkage and interoperability among various glycomic databases. Next, metabolic modelling of parasitic cells using omics data (e.g. transcriptomics, proteomics, and glycomics) is used to understand the metabolism and the role of carbohydrates at the systems level. To explore the metabolism of human protozoan parasites, a constraint-based metabolic model of L. major, iAC560, was used as a case-study and extended to include pathways for the metabolism of lipids and larger fatty acids, and biosynthesis of carbohydrates. Flux Balance Analyses (FBA) was used to simulate the metabolism of Leishmania at promastigote (glucoserich environment) and amastigote (amino acid, amino sugar, and lipid-rich environment) conditions. Also, the model helped to assess active and inactive metabolic pathways to synthesize sugar nucleotides, which are essential precursors in the biosynthesis of glycans and glycoconjugates in promastigote and amastigote stages. Furthermore, in order to improve metabolic predictions, Gene Inactivity Moderated by Metabolism and Expression (GIMME) algorithm was used with flux-based simulations to improve the consistency between predicted fluxes and gene expression data of L. major in promastigote stage. The implementation of GIMME improved flux distribution across various pathways and helped to understand metabolism of Leishmania promastigote stage. Gene deletion analysis using the ext-iAC560 model allowed to predict 53 potential drug target genes in L. major. Many of these genes had been already characterized as essential in other protozoan species, while 10 genes (e.g. LmjF35.5330, LmjF36.2540, LmjF32.1960, LmjF33.0680, LmjF28.1280, LmjF21.1430, LmjF09.1040, LmjF06.1070, and LmjF06.0350 in promastigote stage and amastigote stage, and LmjF36.6950 in only amastigote stage) are predicted as novel drug targets. More than 70% of predicted essential genes showed lethal phenotype by preventing biosynthesis of more than two cellular building blocks. Predicted novel essential genes are associated with lipid and fatty acid biosynthesis, but essentiality in human protozoan parasites or closely related species has not been tested. Searches in literature and chemical databases (e.g. DrugBank and TDR Target) found that around 80% of the predicted essential genes have enzyme-based inhibitors in parasitic and non-parasitic species. Most of these enzyme-based inhibitors were not tested in Leishmania species; however, molecules such as Carbamide phenylacetate, CDV, Terbinafine, and 4-(dimethylaminomethyl)-2,6-di(propan-2- yl)ph enol, which are tested against essential genes in other protozoan species are of major interest, as these are more likely to have similar responses in Leishmania.As doenças parasitárias provocadas por protozoários, tais como a leishmaniose, toxoplasmose e doença do sono, são as uma das principais causas de morte em todo o mundo. A resistência destes parasitas a certas drogas usadas nos tratamentos e as suas capacidades de adapatação durante a infecção são dos principais factores a considerar no desenvolvimento de tratamentos médicos.. Compreender o genótipo / fenótipo destes parasitas durante o processo de infecção é importante no desenvolvimento de terapias anti-parasitárias mais eficazes. Mais recentemente, o uso de abordagens de biologia de sistemas, e em particular, a modelação de redes metabólicas à escala genómica tem sido proposta para compreender o metabolismo dos organismo em diferentes condições de crescimento. Este tipo de metodologias permite incorporar dados ómicos (por exemplo, transcriptomica, proteomica e metabolomica) que melhoram a caracterização do metabolismo dos organismos, incluindo parasitas protozoários, tais como Leishmania, Toxoplasma e Plasmodium.. Esta tese tem como objetivo apresentar uma revisão abrangente do metabolismo de parasitas protozoários e em específico o envolvimento de diferente açucares e glicoconjugados durante o processo de infecção. Para tal, diferentes tipos de dados e metodologias utilizadas para estudar estas moléculas e as suas estruturas, tais como as suas funções em parasitas protozoários foram explorados. Bases de dados de Glicobiologia e em particular, dados ómicos relativos a estas moléculas foramanalisados, abordando a problemática do acesso e integração de dados devido às inconsistências na identificação e representação destes dados.A representação gráfica e textual destas estruturas foi cuidadosamente revista e foram exploradas diferentes ferramentas bioinformáticaspara a interconversão dos diefrentes formatos de modo a melhorar a interligação e a interoperabilidade entre as bases de dados. De modo a explorar o metabolismo de parasitas protozoários, foi usado como caso de estudo o modelo metabólico iAC560 de L. major. O modelo foi extendido (ext-iAC560) de modo a compreender o metabolismo de alguns lipídios e ácidos gordos de cadeia longa e a bisossintese de alguns carboidratos. A simulação do metabolismo foi feirta usando essencialemte métodos baseados na análise de balanço de fluxos (FBA), permitindo descrever o metabolismo de Leishmania no estado promastigota (meio rico em glucose) e amastigota (meio rico em lipídios aminoácidos e amino açúcares e). Métodos como GIMME (Gene Inactivity Moderated by Metabolism and Expression) foram também usados para melhorar a consistência entre os fluxos previstos e os dados de expressão genética. Análises adicionais sobre a essencialidade de genes foram aplicadas, tendo sido encontrados 53 potenciais genes alvo de fármacos na L. major. Muitos desses genes já foram sido caracterizados como essenciais noutras espécies de protozoários, enquanto outros 10 genes (por exemplo, LmjF35.5330, LmjF36.2540, LmjF32.1960, LmjF33.0680, LmjF28.1280, LmjF21.1430, LmjF09.1040, LmjF06.1070, and LmjF06.0350 no estado promastigota e amastigota, LmjF36.6950 no estado amastigota) são previstos como novos alvos farmacológicos. Mais de 70% dos genes essenciais previstos mostraram provocar um fenótipo letal ao prevenir a biossíntese de mais de dois componentes celulares essenciais, se eliminados em condições promastigotas ou amastigotas. Pesquisas na literatura e bases de dados químicas (por exemplo, DrugBank e TDR Target) mostraram que cerca de 80% dos genes essenciais previstos têm inibidores enzimáticos em espécies parasitas e não parasitárias. A maioria desses inibidores não foi testada em espécies de Leishmania; no entanto, moléculas como fenilacetato de carbamida, CDV, terbinafina e 4-(dimetilaminometil)-2,6-di(propan-2-il) fenol, que foram testadas contra genes essenciais em outras espécies de protozoários, são de maior interesse, sendo mais propensos a causar respostas semelhantes em Leishmania.I gratefully express my appreciation to SilicoLife Lda and CEB (University of Minho) for providing required infrastructural facilities for doing research. I especially thank to Simao Soares, CEO SilicoLife Lda, for providing me such nice platform and all help during my work. I also thank Bruno Pereira (systems biologist at SilicoLife Lda) and Hugo Giesteira (exprogrammer at SilicoLife Lda) for scientific and technical assistance during various phases of the work. I am very grateful to BiSBII group (Systems Biology group a University of Minho) for technical suggestions on my project and improving my knowledge in this field. My special thanks to Sara Correia (Researcher at University of Minho) for helping me with java programming during my project. I am also thankful for Initial Training Network, GlycoPar, funded by the FP7 Marie Curie Actions of the European Commission (FP7-PEOPLE-2013-ITN-608295) for providing me financial support