Análise funcional das proteínas desacopladoras mitocondriais de plantas utilizando RNA-seq e mutantes de inserção Alessandra Vasconcellos Nunes Laitz Botucatu-SP 2014 UNIVERSIDADE ESTADUAL PAULISTA “Julio de Mesquita Filho” INSTITUTO DE BIOCIÊNCIAS DE BOTUCATU Análise funcional das proteínas desacopladoras mitocondriais de plantas utilizando RNA-seq e mutantes de inserção Alessandra Vasconcellos Nunes Laitz Orientador: Ivan de Godoy Maia Tese apresentada ao Instituto de Biociências, Câmpus de Botucatu, UNESP, para a obtenção do título de Doutor no Programa de Pós-graduação em Ciências Biológicas (Genética) Genética. Botucatu-SP 2014 Agradecimentos - A Deus; - Aos meus pais, pelo amor, carinho, dedicação e exemplo de vida; - A minha avó Maely, meu avô Ramon (in memoriam), minha irmã Larissa e meu sobrinho Miguel pela presença em minha vida; - Ao Fred, pelo amor, apoio, conversas, companheirismo; - A Universidade Estadual Paulista, em especial ao Departamento de Genética pela estrutura e oportunidade; - Ao Prof. Dr. Ivan de Godoy Maia, pelo profissionalismo, competência e orientação dedicada; - Aos queridos amigos e companheiros: Larissa, Juliana, Alessandra, Rodrigo, Márcio e Eder pelo tempo de convivência, ajudas e conversas; - A todos os colegas do departamento de genética, pelas conversas e risadas durante o período de almoço; - Ao Prof. Paulo Ribolla e ao colega Márcio Luis Acencio pela colaboração no trabalho e ensinamentos; - Ao Dr. Alessandro de Mello Varani pelas análises de bioinformática; - A colega Shelly pela ajuda com as fotos; - Ao Claudinei pela colaboração nas análises; - As amigas Larissa, Rita e Natalia pelas conversas e momentos vividos em Botucatu; - A CAPES pelo apoio financeiro; - A todos que contribuíram de alguma forma para a realização desse trabalho; iv Resumo As proteínas desacopladoras (UCPs) são proteínas especializadas no transporte mitocondrial que dissipam o gradiente eletroquímico de prótons gerados na respiração. Essas proteínas desempenham um papel na manutenção da função mitocondrial e sua importância como componente da tolerância celular ao estresse oxidativo tem sido demonstrada em diversos estudos realizados tanto in vitro com em in vivo. No presente estudo, foram realizados dois estudos empregando UCPs de plantas. Numa primeira abordagem foi realizada uma análise do transcriptoma de plantas transgênicas de tabaco que superexpressam o gene AtUCP1 de Arabidopsis thaliana utilizando a técnica de RNA-Seq. Para o RNA-Seq foi gerado de mais de um milhão de reads com 150 pb em média para cada biblioteca testada. A partir desses reads, um conjunto de aproximadamente 34.000 contigs foi obtido. Após as análises foi possível identificar um total de 816 genes diferencialmente expressos entre as linhagens transgênicas e o controle selvagem, sendo 239 genes induzidos (p≤0,001) e 577 reprimidos (p≤0,001). Em paralelo, uma análise de expressão gênica foi empreendida utilizando mutantes de inserção de arabidopsis para os genes AtUCP1-3, dois deles caracterizados no presente estudo (atucp2 e atucp3), com o objetivo de verificar a funcionalidade e redundância entre essas isoformas. Segundo os resultados obtidos, uma possível compensação só foi observada no mutante atucp3 no qual os genes AtUCP1 e AtUCP2 foram induzidos tanto em condições fisiológicas normais como em condições de estresse salino e osmótico. Palavras-chave: Proteína desacopladora, mitocôndria, RNA-seq, mutantes v Abstract Mitochondrial inner membrane uncoupling proteins (UCP) dissipate the proton electrochemical gradient established by the respiratory chain, thus affecting the yield of ATP synthesis. These proteins play a role in maintaining mitochondrial function and their importance as cellular oxidative stress tolerance component has been demonstrated in several studies performed in vitro and in vivo. In this study, the functional role of plant UCPs was investigated. In a first approach, a transcriptomic analysis of tobacco plants overexpressing the AtUCP1 gene of Arabidopsis thaliana was performed using RNA-seq analysis. The RNA-sequencing generated over a million of reads with 150 base pair on average for each library. From these reads, a set of approximately 34,000 contigs was obtained. A total of 816 differentially expressed genes between transgenic lines and wild-type control was identified. Amongst them, 239 were up-regulated (p≤0,001) and 577 were down-regulated (p≤0,001). In parallel, a gene expression analysis was performed using Arabidopsis insertion mutants for the AtUCP1-3 genes, two of them (atucp2 and atucp3) being characterized in this study. The main purpose was to verify the functionality and the existence of redundancy between the target genes. According to the obtained results, a compensatory expression was observed only in the atucp3 background, in which the AtUCP1 and AtUCP2 genes were induced both in normal physiological conditions and under salt and osmotic stresses. Keywords: Uncoupling protein, mitochondrion, RNA-seq, mutants Sumário Resumo ................................................................................................................................... iv Abstract .................................................................................................................................... v 1- Introdução ........................................................................................................................ 8 2- Revisão Bibliográfica ....................................................................................................... 11 2.1 - Mitocôndrias ........................................................................................................................... 11 2.2 - Proteínas Desacopladoras Mitocondriais ............................................................................... 13 2.3-Proteína Desacopladora Mitocondrial em Plantas ................................................................... 18 3- Objetivos ........................................................................................................................ 22 4- Capitulo I – Análise de Transcriptoma e Expressão Diferencial ........................................ 23 I.1- Transcriptome response signatures associated with the overexpression of a mitochondrial uncoupling protein (AtUCP1) in tobacco ............................................................................................................... 23 5- Capítulo II. Análise dos mutantes de inserção de Arabidopsis thaliana ........................... 49 II.1 – Material e Métodos ................................................................................................................ 49 II.1.1 – Crescimento Vegetal ........................................................................................................... 49 II.1.2 –Extração de DNA genômico e genotipagem ........................................................................ 49 II.1.3 – Sequenciamento ................................................................................................................. 50 II.1.4 - Extração de RNA total e análise de RT-qPCR ....................................................................... 51 II.1.5 – Análise de expressão gênica sob estresses abióticos ......................................................... 52 II.1.6 – Análise de expressão sob estresse biótico .......................................................................... 53 II.1.7 - Detecção de superóxido mitocondrial ................................................................................. 54 II.1.8 - Análise do perfil de expressão dos Carreadores de Dicarboxilatos ...................................... 54 II.2 – Resultados e Discussão........................................................................................................... 55 II.2.1 – Genotipagem ...................................................................................................................... 55 II.2.2 – Localização do T-DNA nos mutantes de inserção .............................................................. 57 II.2.3 – Análise de Expressão Gênica e Redundância Funcional ..................................................... 58 II.2.4 – Análise das isoformas em condições de estresses abióticos .............................................. 60 II.2.5 – Expressão Relativa das AtUCPs em resposta a estresse biótico ......................................... 65 II.2.6 – Detecção de superóxido mitocondrial ................................................................................ 66 II.2.7 – Análise do perfil de expressão dos Carreadores de Dicarboxilatos ..................................... 68 6- Referências Bibliográficas ............................................................................................. 76 7- APÊNDICE ........................................................................................................................ 87 8 1- Introdução As mitocôndrias são organelas celulares que desempenham um papel central numa variedade de processos celulares importantes. É de conhecimento geral que o efeito desses processos é transduzidos por todo o metabolismo, mas a integração e a comunicação entre a mitocôndria e os outros compartimentos subcelulares permanecem pouco compreendidos (Duncan et al., 2013). As plantas, por serem imóveis, devem constantemente se adaptar às condições ambientais adversas, fato que resultou na adaptação também de suas mitocôndrias que adquiriram atributos metabólicos únicos quando comparados aos animais, incluindo a biossíntese do ácido ascórbico, do folato e a aquisição de uma via respiratória alternativa (Millar et al., 2011). As mitocôndrias se tornam um local adequado para a detecção de desequilíbrios funcionais e alterações energéticas, sendo agentes fundamentais na maneira como as plantas responderão aos diferentes tipos de estresses ambientais (Schwarzländer e Finkemeier, 2013; Jacoby et al., 2012). As perturbações no status energético das células provocadas pelos estresses têm consequências diretas sobre a homeostase mitocondrial, e a mitocôndria alterada leva a uma regulação retrógada, onde estímulos mitocondriais regulam a expressão de genes nucleares e alteram a atividade fotossintética (Rhoads 2011; Schwarzländer et al., 2012). A regulação retrógrada mitocondrial em plantas pode estar envolvida com a sinalização metabólica e/ou na resposta a condições de estresse. Determinados estresses ambientais podem provocar um aumento rápido e excessivo de espécies reativas de oxigênio (EROS) e seu acúmulo na mitocôndria é sinalizado ao núcleo induzindo alterações significativas na expressão de diferentes genes (Hartl and Finkemeier, 2012). No entanto, Umbach et al (2012) mostrou que essa regulação retrógrada mitocondrial pode ser independente de EROS, sugerindo a existência de mecanismos distintos de sinalização mitocôndria-núcleo. Na mitocôndria, a cadeia transportadora de elétrons gera EROS continuamente (em dois sítios principais: os complexos I e III). Nas plantas, essa cadeia contém enzimas respiratórias alternativas que não estão presentes em mamíferos, sendo elas a oxidase alternativa (AOX) e a NADH desidrogenase (Moller, 2001). A AOX é frequentemente utilizada como um marcador de disfunção mitocondrial e estresse oxidativo, uma vez que em condições de estresse sua http://www.ncbi.nlm.nih.gov/pubmed/?term=Schwarzl%26%23x000e4%3Bnder%20M%5Bauth%5D http://www.ncbi.nlm.nih.gov/pubmed/?term=Finkemeier%20I%5Bauth%5D 9 expressão é induzida e ela atua na homeostase das moléculas sinalizadoras envolvidas na comunicação núcleo-mitocôndria (Vanlerberghe, 2013). Além da AOX, estudos demonstraram a existência de um segundo sistema dissipador de energia nas mitocôndrias de plantas, as chamadas proteínas desacopladoras mitocondriais (UCP). Ambas atuam na regulação da produção de EROs mitocondrial, e têm sua função fisiológica associada com a proteção celular contra o estresse oxidativo (Jarmuszkiewicz et al.,1998; Almeida et al.,1999; Maxwell et al., 1999; Bransalise et al.,2003). As UCPs (mitochondrial uncoupling protein) catalisam o desacoplamento da cadeia transportadora de elétrons da fosforilação oxidativa. De maneira geral, as proteínas desacopladoras dissipam o gradiente eletroquímico de prótons gerados na respiração na forma de calor (Nicholls, 1999), sendo dependentes de ácidos graxos e sensíveis aos nucleotídeos purínicos. As UCPs foram primeiramente caracterizadas em 1976 por Ricquier e Kader quando a sigla UCP (uncoupling protein) passou a ser oficialmente empregada, sendo esta chamada de UCP1 (uncoupling protein 1). A UCP1 está diretamente envolvida com a principal função fisiológica do tecido adiposo marrom que é a termogênese (sem tremor). Em 1995, Vercesi e colaboradores constataram a presença de uma atividade desacopladora em mitocôndrias isoladas de batata muito parecida com a atividade da UCP1 em tecido adiposo marrom, sugerindo a existência de uma proteína desacopladora em plantas. Desde então, vários homólogos passaram a ser identificados em diferentes espécies vegetais. Em 2006, a análise detalhada do genoma de Arabidopsis thaliana permitiu a identificação de uma família gênica de pUCPs composta por seis membros (chamados inicialmente de AtPUMP1-6; Borecký et al., 2006), porém três deles (AtUCP4-6) foram posteriormente classificados funcionalmente como transportadores de dicarboxilatos (DICs; Palmieri et al., 2008). Até o presente momento dados funcionais só foram obtidos para a AtUCP1 (descrita por Maia et al., 1998), o que gera questionamentos sobre a relevância fisiológica e a redundância funcional entre as diferentes isoformas. Sweetlove e colaboradores (2006) demonstraram que plantas nocaute para o gene AtUCP1 apresentaram reduzida taxa de assimilação fotossintética de carbono, e um estresse oxidativo localizado que não chega a afetar a capacidade do mutante em suplantar estresses abióticos como baixa temperatura, exposição a metal pesado 10 e tratamento com herbicida. Neste estudo, porém, nenhuma investigação sobre uma possível complementação funcional e redundância foi realizada. Portanto, o papel desempenhado pelas outras duas UCPs presentes em arabidopsis (AtUCP2 e AtUCP3) ainda permanece pouco estudado. Neste cenário, considerando a disponibilidade de mutantes nocautes de arabidopsis, o presente estudo teve como um de seus objetivos analisar a funcionalidade destas proteínas desacopladoras, e testar experimentalmente a existência de possível complementação e redundância funcional entre elas. Como realçado, diversos estudos sugerem que as UCPs desempenham um papel importante na defesa celular contra o estresse oxidativo mitocondrial, uma vez que o desacoplamento entre a respiração e a fosforilação oxidativa mediado por essas proteínas seria capaz de aumentar a velocidade respiratória, levando a uma significativa redução na geração mitocondrial de EROs (Houron-Cabassa et al., 2002). Tais resultados sugerem que o principal papel fisiológico dessas proteínas seja o de moderar a produção desses radicais durante a ocorrência do estresse (revisto em Pastore et al., 2007). Em conformidade com tal hipótese foi demonstrado que a expressão constitutiva do gene AtUCP1 em plantas transgênicas de tabaco conduz a um aumento de tolerância ao estresse oxidativo gerado exogenamente (Brandalise et al., 2003) bem como a uma maior tolerância aos estresses salino e osmótico (Begcy et al., 2011). Outros resultados demonstraram que a atividade das UCPs de plantas (pUCP) é estimulada por radicais superóxido e por produtos de peroxidação lipídica como o 4-hidroxi-2-nonenal (HNE), que normalmente são gerados em condições de estresse (Smith et al.,2004). Chen et al. (2014) utilizando tomates onde o gene LeUCP foi silenciado mostraram que a UCP é essencial para manter o equilíbrio redox e a atividade de enzimas antioxidantes em resposta ao estresse oxidativo, reduzindo assim a superprodução de EROS. Até o presente momento, a participação efetiva de uma UCP nos mecanismos de comunicação retrógrada só foi evidenciada em um modelo animal, onde a UCP3 com expressão ectópica no fígado de camundongo controlou a expressão de genes nucleares (Camara et al., 2009). Esta observação, juntamente com todos os relatos da participação das UCPs na tolerância ao estresse oxidativo e equilíbrio redox, dá suporte à hipótese de que a expressão constitutiva da pUCP pode elicitar respostas 11 importantes potencialmente relacionadas com a homeostase mitocondrial e sinalização retrógrada. Neste contexto, a disponibilidade de plantas transgênicas capazes de superexpressar a AtUCP1, e seu respectivo controle não transformado (Brandalise et al., 2003), representam um bom ponto de partida para investigar possíveis alterações transcricionais provocadas pela expressão constitutiva de uma proteína desacopladora. Para tal, dentre as várias técnicas disponíveis para avaliar tais alterações, optou-se no presente estudo pelo uso da técnica de RNA-Seq, vislumbrando que o sequenciamento em larga escala de RNA utilizando sequenciadores de nova geração poderá nos fornecer uma visão global dos transcriptomas analisados. Diante dos dados apresentados, o presente estudo tem como objetivo verificar a funcionalidade e o grau de redundância funcional dos genes codificadores de pUCPs em A. thaliana, bem como realizar uma análise global do perfil de expressão de plantas transgênicas de tabaco que superexpressam a AtUCP1 comparadas com plantas selvagens. Buscando um melhor entendimento, os dados obtidos encontram-se apresentados na forma de capítulos. 2- Revisão Bibliográfica 2.1 - Mitocôndrias Toda energia necessária para as funções celulares dependem de um suprimento de energia proveniente do catabolismo de carboidratos, lipídeos e aminoácidos. Em células eucarióticas, o processo de formação de energia inicia-se no compartimento citosólico e tem continuidade nas mitocôndrias. As mitocôndrias são núcleos centrais na conversão de energia em uma forma biologicamente utilizável e estão ligadas a vias metabólicas de diferentes compartimentos subcelulares atuando assim como sensores do estado energético e metabólicos das células vegetais. Essas organelas são formadas por duas membranas que possuem diferentes propriedades de permeabilidade, um espaço intermembranas e uma cavidade 12 denominada de matriz mitocondrial. A membrana externa permite um transporte de íons, metabolitos e proteínas do citosol para o espaço intermembranas, além de conduzir eventos de sinalização entre a mitocôndria e o citosol. Estudos em células de mamíferos revelaram a existência de mais de 120 proteínas na membrana externa da mitocôndria e que maioria delas desempenha algum papel no processo de sinalização que integram a mitocôndria e as funções celulares. Já em A. thaliana, 46 dessas proteínas foram identificadas (Duncan et al., 2013). A proteína mais abundante encontrada na membrana externa é o canal de ânion voltagem dependente (VDAC; do inglês Voltage-Dependent Anion Channel), a qual está envolvida em diversos processos, como o transporte de metabólitos como ATP e NADPH em humanos (Blachy-Dyson and Forte, 2001), além de uma possível função na resposta a patógenos, apoptose e estresse abiótico em plantas (Desai et al., 2006; Jones et al., 2006; Kussano et al., 2009). Já a membrana interna tem uma elevada resistência ao fluxo de prótons e é estruturalmente dividida em três componentes: a membrana interna propriamente dita, a membrana da crista e o espaço intercrista. É uma membrana altamente especializada e rica em proteínas, dentre as quais estão os componentes da cadeia respiratória e as proteínas responsáveis pelo transporte de metabólitos para a matriz. Todas essas proteínas possuem estruturas semelhantes, com um tamanho em torno de 30 kDa, e fazem parte da superfamília de carreadores mitocondriais, os quais exercem uma grande influencia nas funções da mitocôndria (revisto em Logan., 2006; Laloi., 1999). Na matriz mitocondrial encontra-se a maquinaria genética mitocondrial, DNA, RNA e ribossomos, além de diversas enzimas, como piruvato desidrogenase e enzimas do ciclo do ácido citrico. O DNA mitocondrial se apresenta na forma de complexo DNA-proteina chamados de nucleóides, os quais possuem diversas proteínas que podem ou não desempenhar alguma função na manutenção do mtDNA (Kucej and Butow., 2007). O processo mitocondrial de produção de energia ocorre pela ação combinada entre o ciclo do ácido cítrico, a cadeia respiratória e o sistema de fosforilação oxidativa. A primeira etapa se inicia com a oxidação dos combustíveis como lipídios e carboidratos, gerando o acetil- CoA que entra no ciclo do ácido cítrico e tem seus agrupamentos acetil oxidados em CO2, o qual é liberado como subproduto, gerando energia na forma de elétrons que é então conservada nos compostos NADH (Nicotinamida-Adenina-Dinucleotídeo) e FADH2 (Flavina-Adenina-Dinucleotídeo). 13 Estes compostos são transferidos para a membrana mitocondrial interna, onde entram na cadeia transportadora de elétrons, onde ocorre uma doação de elétrons e NADH é oxidado a NAD+ e FADH2 a FADH (Nelson e Cox., 2011). A cadeia transportadora de elétrons é composta por quatro grandes complexos enzimáticos respiratórios: complexo I ou NADH desidrogenase, complexo II ou succinato desidrogenase, complexo III ou citocromo bc1 e complexo IV ou citocromo c oxidase. Esses complexos interagem entre si via ubiquinona (UQ) e citocromo c que são transportadores de elétrons (Jacoby, 2012). O movimento é dirigido pelo potencial redox e o destino final dos elétrons que se movimentam pela cadeia respiratória é o oxigênio molecular, o qual é reduzido a H2O. No processo de transporte dos elétrons ao longo da cadeia acontece a liberação de energia, sendo uma parte desta dissipada na forma de calor, e a outra utilizada para transportar prótons (H+) da matriz mitocondrial (região com baixa concentração de H+) para o espaço intermembrana da mitocôndria (região com alta concentração de H+), estabelecendo assim um gradiente de prótons. Devido à impermeabilidade da membrana interna aos prótons, eles tendem a voltar para a matriz atravessando a ATP-sintetase (complexo V) que sintetiza ATP a partir de ADP mais fosfato. Tal acoplamento, entretanto, não é perfeitamente ligado à síntese de ATP, visto que os prótons podem retornar para a matriz mitocondrial através de outras proteínas também presentes na membrana interna. Dentre essas estão as UCPs que, como já mencionado, catalisam o desacoplamento da cadeia transportadora de elétrons da fosforilação oxidativa. Como consequência, o potencial de membrana é ligeiramente reduzido e a energia derivada da oxidação dos substratos é perdida na forma de calor (Krauss, 2005). 2.2 - Proteínas Desacopladoras Mitocondriais As proteínas desacopladoras mitocondriais (UCPs) são encontradas em diversos organismos, de mamíferos a plantas. As UCPs pertencem à família dos carreadores mitocondriais (MCF; do inglês Mitochondrial Carrier Family), onde todas as proteínas possuem um peso molecular entre 28-34 kDa e são codificadas exclusivamente por genes nucleares (Borecký et al., 2001). As UCPs foram inicialmente denominadas de termogenina ou proteína de ligação ao GDP (guanosina-difosfato) (Nicholls, 2001). Porém, após a sua devida 14 caracterização em tecido adiposo marrom de camundongos por Ricquier e Kader em 1976, a sigla UCP (uncoupling protein) passou a ser oficialmente empregada, sendo esta chamada de UCP1 (uncoupling protein 1). Até o momento foram descobertas cinco isoformas em mamíferos, as quais foram numeradas de acordo com a ordem de suas descobertas i.e. UCP1-5 (Donadelli et al., 2013). Em 2011, Berardi et al. determinaram a estrutura da UCP2 empregando ressonância magnética nuclear e propuseram que a estrutura da UCP é comparável com a do carreador ADP/ATP (ANT) descrito por Pebay-Peyroula et al. (2003). Ambos, UCPs e ANT, possuem uma estrutura tripartida que consiste em três domínios repetidos, cada qual contendo duas regiões hidrofóbicas que formam α- hélices transmembranas, com a diferença que nas UCPs as orientações relativas dos segmentos helicoidais são diferentes, resultando em uma maior abertura no lado da matriz da membrana interna. Os três domínios são ligados através de um “loop” e uma análise do alinhamento das sequencias dessas regiões sugerem que a UCP1-3 pertencem a uma família e a UCP4-5 pertencem a outra família (Berardi et al., 2011; Donadelli et al., 2013). As UCPs catalisam o desacoplamento da cadeia transportadora de elétrons da fosforilação oxidativa (Ricquier e Kader, 1976). Como consequência, o potencial de membrana é ligeiramente reduzido e a energia derivada da oxidação dos substratos é perdida na forma de calor. De maneira geral, as proteínas desacopladoras dissipam o gradiente eletroquímico de prótons gerados na respiração na forma de calor (Nicholls, 2001), sendo que esta atividade é dependente de ácidos graxos e sensível aos nucleotídeos purínicos (PN). A regulação das UCPs pelos ácidos graxos é um assunto controverso e três mecanismos foram propostos até então: (a) modelo tamponante, onde os ácidos graxos se ligam a sítios que estão junto ao canal de prótons da proteína, criando grupos aceptores/doadores de elétrons que facilitariam o transporte de H+, (b) modelo protonoforético, o qual propõe que as proteínas desacopladoras mitocondriais funcionariam como carreadoras de ânions, exportando ácidos graxos livres desprotonados para a monocamada externa da membrana mitocondrial interna, os quais posteriormente penetram na monocamada interna na forma protonada por um mecanismo denominado “flip-flop”, deixando um próton na matriz mitocondrial para cada ciclo de transporte, e (c) modelo de interação alostérica, que postula que os ácidos graxos não são necessários para o transporte de prótons, mas são necessários para induzir uma mudança alostérica para superar a inibição dos 15 nucleotídeos. Como a cinética de ligação dos nucleotídeos não é alterada na presença de ácidos graxos foi sugerido que esta competição funcional é conseguida por uma simples ligação competitiva (Divakaruni and Brand, 2011). Diferentemente dos mecanismos envolvendo os ácidos graxos, fortes evidências experimentais mostram que os PNs inibem a atividade das UCPs. O sítio de ligação da UCP aos PNs está voltado para o espaço intermembrana contendo “solução citosólica”, logo, a concentração dos PNs nessa solução é que irá permitir o desacoplamento (Dlasková et al., 2006). Como citado anteriormente, a UCP1 foi descrita em 1976 nas mitocôndrias de tecido adiposo marrom de camundongos e sua função está diretamente ligada com a termogênese (produção de calor sem contração muscular; Porter, 2006). A termogênese está diretamente ligada à atividade da UCP1 neste tecido, sendo atribuída ao desacoplamento da respiração causado por ela. Em adipócitos, a UCP1 representa cerca de 10% do total de proteínas da membrana mitocondrial interna (Esteves e Brand, 2005). Grande parte dos estudos relacionados com a UCP1 de mamíferos está diretamente ligada ao tecido adiposo marrom e à sua função termogênica, porém relatos mostram a expressão da UCP1 em timo de ratos e camundongos, no músculo liso do aparelho digestivo, do útero, e do aparelho reprodutor masculino, bem como durante o relaxamento das camadas longitudinais de músculo liso (Nibbelink et al., 2001; Carroll et al., 2005). De 1997 a 2000 foram descobertas quatro UCPs homólogas a UCP1 de mamíferos, sendo as mesmas numeradas de acordo com a ordem de sua descoberta. Quando comparadas em relação à identidade, a UCP2 e a UCP3 apresentam 59 e 57% de identidade com a UCP1, respectivamente, e 73% de identidade entre si (Krauss et al., 2005); já as UCP4 e UCP5 apresentam em torno de 30% de identidade com a UCP1. Segundo Hanak et al. (2001), a UCP4 representa o ancestral a partir do qual as demais UCPs divergiram, sendo as UCP1, 2 e 3 evoluíram tardiamente e passaram a executar funções mais especializadas. Em termos de expressão, a UCP1 é altamente expressa em tecido adiposo marrom, enquanto a UCP2 é expressa em vários tecidos, tais como fígado, cérebro, pâncreas, células do sistema imune, baço, rim, e sistema nervoso central, e a UCP3 somente em tecido adiposo marrom e tecido muscular. Os outros dois membros da família, UCP4 e UCP5, são altamente expressos no sistema nervoso central (Donadelli et al., 2013; Sanchis et al.,1998). http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3782081/#R13 16 Inicialmente, devido à similaridade com a UCP1, as UCPs 2 e 3 foram relacionadas com a termogênese, hipótese que foi descartada após experimentos com camundongos nocautes ucp2 e ucp3, onde a resposta a exposição ao frio foi a mesma tanto nos mutantes como nos selvagens (Arsenijevic et al., 2000; Gong et al., 2000; Vidal-Puig et al., 2000). Sabe-se atualmente que estas estão relacionadas com a dissipação do gradiente de prótons, evitando o excesso de força próton- motriz, diminuindo assim as EROS produzidas pela cadeia transportadora de elétrons (revisto em Donadelli et al., 2013). Entretanto, os estudos procurando relacionar a função desacopladora da UCP2 com a supressão de EROS têm apresentado resultados controversos. Em um estudo usando células em cultura, por exemplo, Pecqueur et al. (2008) demonstraram que a perda de função da UCP2 não resultou em aumento significativo da produção de EROS nem em maior propensão para tais células tornarem-se senescentes. Por outro lado, Deng et al. (2012) mostraram que a superexpressão da UCP2 em células A549 inibiu o acúmulo de EROS e a apoptose. Curiosamente, diversos relatos mostram que a UCP2 tem a sua expressão induzida em diversos tipos de cânceres humanos, dado que foi associado ao aumento do estresse oxidativo nas células cancerígenas (revisto em Robbins and Zhao, 2011). Considerando que o aumento da atividade desacopladora da UCP2 pode estar relacionada ao controle do estresse oxidativo, o desacoplamento possivelmente associado a sua superexpressão parece conferir uma vantagem adicional a células cancerígenas. Como para a maioria das UCPs, a UCP3 continua com sua função ainda não totalmente esclarecida, porém algumas hipóteses já foram levantadas. A mais plausível refere-se à sua participação na proteção contra o estresse oxidativo, atuando como um exportador de ácidos graxos, protegendo mitocôndrias contra a peroxidação lipídica e/ou facilitando a oxidação dos ácidos graxos (Costford et al., 2008). Como realçado anteriormente, a UCP3 é expressa no músculo esquelético e sua ausência neste tecido aumenta o dano causado pelo estresse oxidativo e prejudica o metabolismo muscular (Vidal-Puig et al., 2000). Já a sua superexpressão em camundongos aumenta o metabolismo de ácidos graxos e também reduz a produção de EROS durante a oxidação dos ácidos graxos (Bezaire et al., 2005). Interessantemente, Harper et al. (2008) mostraram que a superexpressão da UCP3 oferece uma proteção contra a resistência à insulina, que é um fator importante na diabetes tipo 2, através da conversão da gordura intramiocelular em energia térmica. 17 Ainda neste contexto, Costford et al. (2008), ao compararem o ganho de gordura em três linhagens de ratos (selvagem, sem UCP3 e com UCP3 superexpressa), verificaram que os ratos que tinham a UCP3 superexpressa tiveram uma proteção contra o ganho de gordura. A expressão ectópica da UCP3 de mamíferos em fígado de camundongos exerceu uma ação direta na atividade da mitocôndria, causando uma sinalização retrógrada que culminou na regulação da expressão de genes nucleares envolvidos na inanição e no catabolismo de ácidos graxos no fígado (Camara et al., 2009). As acentuadas diferenças entre as sequências das UCP4 e 5, quando comparadas com as sequências das UCP1-3, bem como a limitada distribuição tecidual destas proteínas, indicam que elas podem ter diferentes funções em comparação com as UCPs 1 a 3. No entanto, ambas participam do transporte de prótons através da membrana interna para a matriz, ou seja, executam a função considerada essencial para uma proteína desacopladora da fosforilação oxidativa. Este processo é acompanhado por uma redução do estresse oxidativo, o que consequentemente, exerce uma influência protetora em células expostas a injurias mitocondriais (Zhang et al. 2006). Análises independentes realizadas utilizando uma linhagem de célula neural superexpressando a UCP4 (Chu et al., 2009) e outra superexpressando a UCP5 (Kwok et al., 2010) mostraram que a superexpressão tem um importante papel neuroprotetivo contra o dano oxidativo quando comparadas com células controle que apresentam um nível basal de expressão das referidas UCPs. A expressão das UCPs 4 e 5 também já foi relatada em tecidos não neurais, especialmente em células epiteliais mamárias de bovino (Yonezawa et al., 2009). Estes autores mostraram que ambas são reprimidas em células submetidas a tratamento com insulina enquanto que a UCP5 é superexpressa em presença de ácidos graxos saturados, sugerindo que ambas participam do metabolismo de lipídios e energético. Cabe ressaltar que embora as UCPs 4 e 5 sejam relativamente desconhecidas e inexploradas, as suas propriedades, em especial da UCP4, as tornam interessantes como potenciais intervenientes no desencadeamento de doenças (Ramsden et al., 2012). 18 2.3-Proteína Desacopladora Mitocondrial em Plantas As proteínas desacopladoras em plantas foram identificadas em 1995 por Vercesi e colaboradores, que demonstram em mitocôndrias isoladas de batata que um estado acoplado só pode ser alcançado na presença de nucleotídeos purínicos e na ausência de ácidos graxos. Ao mesmo tempo, uma proteína de aproximadamente 32 kDa foi isolada utilizando o protocolo de isolamento da UCP1 de mamíferos (Klingenberg e Winkler, 1986), e incorporada em lipossomos, onde uma alcalinização do meio externo pode ser constatada. Esta observação sugeriu que a referida proteína poderia facilitar o transporte de H+, na presença de ácidos graxos livres, sendo fisiologicamente inibida por ATP, GDP e GTP, da mesma forma que a UCP1 de mamíferos, revelando assim a existência de uma proteína desacopladora nos vegetais (Jezek et al., 1997). A proteína evidenciada foi inicialmente chamada de PUMP (plant uncoupling mitochondrial protein), porém após alguns anos a denominação pUCP (de plant UCP) foi empregada e vem sendo utilizada até hoje. Nos anos subsequentes, Laloi et al. (1997) e Maia et al. (1998) clonaram os primeiros cDNAs que codificam pUCP em batata (StUCP) e em A. thaliana (AtUCP1), respectivamente. A sequência de aminoácidos de um peptídeo derivado da StUCP apresentou 44% e 47% de similaridade com a UCP1 e UCP2 de mamíferos enquanto que a sequência deduzida de aminoácidos da AtUCP1 apresentou 81% de identidade com a da StUCP. Após esses dois relatos, outros cDNAs que codificam para pUCPs foram identificados e isolados em diferentes espécies vegetais (revisado em Vercesi et al., 2006), tanto em dicotiledôneas [AtUCP2 em A. thaliana (Watanabe et al., 1999); SfUCP em repolho (Ito et al., 1999) e HmUCPa em Helicodiceros muscivorus (Ito et al., 2003)], como em monocotiledôneas [WhUCP em trigo (Murayama e Handa, 2000), OsUCP em arroz (Watanabe et al., 2002), e ZmPUMP em milho (Brandalise et al., 2003)]. Análises in silico realizadas utilizando o banco de dados do TIGR (The Institute for Genomic Research) permitiram a identificação de seis genes codificando pUCPs no genoma de A. thaliana, que foram então definidos como uma família gênica e denominados de AtUCP1-6. Palmieri e colaboradores (2008), entretanto, relataram que três desses genes, AtUCP4-6, codificam na realidade carreadores de 19 dicarboxilatos (DICs). Neste estudo, análises filogenéticas revelaram que as AtUCP4-6 são mais próximas dos DICs de animais e leveduras do que os outros três membros da família AtUCP1-3. O mesmo estudo que identificou a família gênica em arabidopsis também identificou uma família gênica em cana-de-açúcar, composta por cinco genes não redundantes denominados de SsUCP1-5. Adicionalmente, investigou o perfil expressão dos genes AtUCP1-6 em diferentes órgãos/tecidos de arabidopsis. Dos seis genes analisados, AtUCP1, AtUCP4 e AtUCP5 apresentaram expressão ubíqua, sendo os genes AtUCP4-5 os mais abundantes. AtUCP2, por sua vez, foi expresso exclusivamente em sílica verde e AtUCP3 foi detectado exclusivamente em raiz. Em contrapartida, transcritos do gene AtUCP6 não foram detectados (Borecky et al., 2006). A identificação de todos esses homólogos levou ao aumento de estudos dedicados a esta família de proteínas, porém o papel fisiológico de seus membros ainda permanecem desconhecido, exceção feita para o gene AtUCP1 que está melhor caracterizado (Vercesi et al., 2006). Devido a sua presença generalizada nos eucariotos, diversas funções foram propostas para as pUCPs (Nogueira et al., 2011) como, regulação no potencial de membrana da mitocôndria (Jezek et al., 1997), regulação do metabolismo energético na mitocôndria (Ricquier e Bouillaud, 2000), redução das EROS (Considini et al., 2003; Popov et al., 2011) e manutenção da homeostase redox (Vercesi et al., 2006), além de influenciar no fluxo do ciclo do ácido tricarboxílico (Smith., 2004). As plantas produzem continuamente EROS como subproduto do metabolismo aeróbio, sendo estas geradas na cadeia respiratória, principalmente nos complexos I e III, ou através do metabolismo de produtos endógenos das mitocôndrias. As EROS mais comuns são o superóxido (O-2), o peróxido de hidrogênio (H2O2) e o radical hidroxil (OH) (Mittler et al., 2002). Condições de estresse ambiental provocam um aumento rápido das EROS na mitocôndria, condição que pode, dependendo da concentração, causar dano nessa organela ou ser usado na sinalização. No segundo caso, a organela, por um mecanismo ainda desconhecido, provoca alterações na expressão de genes destinados a manutenção do equilíbrio redox através de mecanismos antioxidantes ou ajustes metabólicos (Moller et al., 2010). Em conformidade com as variadas funções das pUCPs foi demonstrado que plantas que superexpressam o gene AtUCP1 diminuem a produção de EROS aumentando a tolerância ao estresse oxidativo induzido por peróxido de hidrogênio 20 exógeno (Brandalise et al.,2003). Além disso, essas plantas apresentam alta tolerância a diferentes estresses abióticos (salino e osmótico; Begacy et al., 2011), um menor acúmulo de espécies reativas de oxigênio também sob estresses abióticos, um aumento da biogênese mitocondrial e uma diminuição no conteúdo celular de ATP (Barreto et al., 2014) . Em consonância, em mitocôndrias isoladas de trigo, a atividade da WhUCP em resposta a estresse osmótico demonstrou ser modulada por EROs através de um mecanismo de retroalimentação, sugerindo assim que as pUCPs devem atuar de forma indireta como um sistema antioxidante de defesa (Pastore et al., 2007). Porém, Sweetlove et al. (2006) observaram que a falta do gene AtUCP1 em arabidopsis não afetou a capacidade da planta em tolerar alguns estresses abióticos, contudo, esses autores não levaram em consideração a existência e funcionalidade dos outros genes AtUCP. Trabalhos adicionais realizados em tomate também revelaram o envolvimento das UCPs no controle das EROS. O estudo realizado por Popov e colaboradores em 2011 teve como objetivo quantificar a expressão da UCP e da AOX (os dois sistemas dissipadores de energia encontrados nas mitocôndrias de plantas) em condições de baixa temperatura e em presença de peróxido de hidrogênio. Neste caso, os pesquisadores observaram uma indução tanto da pUCP quanto da AOX em decorrência do aumento da produção de EROS. Conforme postulado por Goglia e Skulachev (2003) todos os homólogos de UCP1 desempenham um papel no equilíbrio redox limitando a produção de EROS. Corroborando essa hipótese e também os dados relatados acima, foi observado que em plantas de tomate onde o gene LeUCP foi silenciado, ocorreu uma redução na eficiência quântica do fotossistema II (PSII) bem como no coeficiente de extinção fotoquímico. Além disso, a ativação de determinadas enzimas antioxidantes foi abrandada fato que agravou o acúmulo de peróxido de hidrogênio e superóxido nas folhas de plantas tratadas com metil viologênio. Este fato também foi observado em condições de alta temperatura. Em suma foi demonstrado que a pUCP é essencial para a manutenção do equilíbrio redox e regulação do sistema antioxidante em resposta ao estresse oxidativo, reduzindo assim a superprodução de EROS (Chen et al., 2014). Em contrapartida, a superexpressão da LeUCP provocou um aumento na eficiência quântica do PSII, do coeficiente de extinção fotoquímico, e da taxa de transferência de elétrons, além de reduzir o acumulo de espécies reativas de oxigênio. As plantas transgênicas que superexpressam o gene LeUCP foram mais tolerantes ao estresse térmico e ao fungo Botrytis cinerea, demonstrando que a 21 LeUCP desempenha um papel no controle de estresses abióticos e bióticos através da manutenção do equilíbrio redox e do aumento da capacidade antioxidante (Chen et al., 2013). 22 3- Objetivos Os objetivos gerais do presente trabalho foram:  Sequenciar o transcriptoma das plantas transgênicas de tabaco que superexpressam o gene AtUCP1, utilizando a técnica de RNA-seq;  Determinar e categorizar os genes diferencialmente expressos comparados com a linhagem de tabaco selvagem;  Validar a expressão diferencial de um conjunto de genes alvos por RT-qPCR;  Caracterizar e analisar a funcionalidade e a redundância dos mutantes de inserção atucp1-3;  Analisar a expressão dos diferentes genes que codificam pUCPs e carreadores de dicarboxilatos (DICs) de Arabidopsis thaliana em resposta aos estresses salino e osmótico; 23 4- Capitulo I – Análise de Transcriptoma e Expressão Diferencial Na presente tese foi gerado um artigo contendo os resultados referente a Análise de transcriptoma e expressão diferencial. O artigo intitulado Transcriptome response signatures associated with the overexpression of a mitochondrial uncoupling protein (AtUCP1) in tobacco foi submetido a revista PLOS ONE em agosto de 2014. Nesse estudo, foi realizado uma análise de RNA-seq buscando-se obter uma visão geral do perfil da expressão gênica em duas linhagens de plantas transgênicas de tabaco que superexpressam o gene AtUCP1 (P07 e P32) em comparação com plantas selvagens (WT). I.1- Transcriptome response signatures associated with the overexpression of a mitochondrial uncoupling protein (AtUCP1) in tobacco Alessandra Vasconcellos Nunes Laitz1, Marcio Luis Acencio2, Ney Lemke2, Paulo Eduardo Martins Ribolla3, Ivan G. Maia1* 1UNESP, Instituto de Biociências, Departamento de Genética, Botucatu, SP, Brazil 2UNESP, Instituto de Biociências, Departamento de Física e Biofísica, Botucatu, SP, Brazil 3UNESP, Instituto de Biociências, Departamento de Parasitologia, Botucatu, SP, Brazil * Corresponding author: Instituto de Biociências, Departamento de Genética, UNESP, 18618- 970 - Botucatu, SP, Brazil - Fone: 55 14 3880 0382 and Fax: 55 14 3815 3744 Abstract Mitochondrial inner membrane uncoupling proteins (UCP) dissipate the proton electrochemical gradient established by the respiratory chain, thus affecting the yield of ATP synthesis. UCP overexpression in plants has been correlated with oxidative stress tolerance, improved photosynthetic efficiency and increased mitochondrial biogenesis. This study reports the main transcriptomic responses associated with the overexpression of an UCP (AtUCP1) in tobacco plants. By using RNA-sequencing, a total of 816 differentially expressed genes between the investigated overexpressor lines and parental untransformed plants (WT) were identified. Among them, 239 were 24 up-regulated and 577 were down-regulated. As a general response to AtUCP1 overexpression, noticeable changes in the expression of genes involved in energy metabolism and redox homeostasis were detected. A substantial set of differentially expressed genes code for products targeted to the chloroplast and mainly involved in photosynthesis. The overall results demonstrate that the alterations in mitochondrial function provoked by AtUCP1 overexpression require important transcriptomic adjustments to maintain cell homeostasis. Moreover, the occurrence of an important cross-talk between chloroplast and mitochondria, which culminates in the transcriptional regulation of several genes involved in different pathways, was evidenced. Introduction Mitochondria are important sensors of the metabolic and energetic status of the cell. As the center of the respiratory metabolism, mitochondria play a key role in various processes, especially in plant response to stressful conditions [1]. In this scenario, disturbances in the energy status of the cell caused by stress have direct consequences on mitochondrial function, and mitochondrial adjustment to these conditions, culminates in important changes in the cellular activity as well as in nuclear gene expression. The communication between the mitochondria and nucleus is performed by a mechanism called retrograde signaling, which is already well characterized in yeast, but poorly investigated in plants (reviewed in [2]). Main studies in plants have been performed employing the alternative oxidase (AOX), an important component of the alternative electron transport pathway. It has been demonstrated that the expression of certain genes encoding AOX in Arabidopsis thaliana (AtAOX1a for example) are induced in response to disturbances in mitochondrial function by retrograde signaling [3, 4]. Collected data suggests that AOX is an important marker of mitochondrial dysfunction and cellular oxidative stress, acting in the homeostasis of the signaling molecules in volved in organelle to nucleus communication (reviewed in [5]). Moreover, recent results have shown that retrograde communication may also be involved in the regulation of nuclear genes implicated in the biogenesis of the electron transport chain (reviewed in [5]). The mitochondrial uncoupling protein (UCP) is a component of the mitochondrial alternative pathway in animals and plants (reviewed by [6]). UCPs are located in the inner mitochondrial membrane and catalyze a proton conductance that dissipates the proton electrochemical gradient established by the respiratory chain, 25 thus affecting the yield of ATP synthesis. UCPs are involved in mitochondrial energy flow regulation and reported to play a critical role in the modulation of mitochondrially generated reactive oxygen species (ROS) [6]. Compelling evidences also point to an involvement of UCPs in plant response to stresses, especially those associated with ROS generation [7, 8]. In this context, the positive impact of UCP overexpression in plant tolerance to drought and salt stresses has been demonstrated [9]. The observed tolerance was correlated with reduced ROS accumulation in the transgenic plants as a result of increased UCP- mediated mitochondrial uncoupling. However, the exact mechanisms implicated in improved plant performance under stress are still poorly understood. Recently, UCP overexpression has been correlated with increased mitochondrial biogenesis [10]. Due to their involvement in the regulation of the cell redox state and in tolerance to oxidative stress [7, 11–13], plant UCPs represent an interesting model for understanding the functional implications of mitochondrial uncoupling in retrograde signaling. In this context, the availability of transgenic tobacco plants overexpressing the UCP1 gene from A. thaliana (AtUCP1) [7], and their respective parental untransformed control, represent an interesting starting point to investigate possible transcriptional changes associated with the constitutive expression of an UCP and to elucidate the mechanisms involved. In a similar approach in mice, it has been observed that the ectopic expression of mammalian UCP3 promotes the up- regulation of genes involved in metabolic response to starvation and fatty acid catabolism in the liver through retrograde signaling [14]. This result, obtained using an animal model, supports the hypothesis that the constitutive expression of an UCP can trigger responses potentially involved in mitochondrial function and retrograde signaling. In the present study, we established the global changes in gene expression associated with the overexpression of AtUCP1 in seedlings of transgenic tobacco using RNA-Sequencing. Interestingly, the transcriptomic responses induced by AtUCP1 overexpression resemble, at least in part, those typically triggered by mitochondrial dysfunction and imply the cross-talk between mitochondria and chloroplast. 26 Materials and Methods Plant materials and growth conditions The Nicotiana tabacum SR1 plants overexpressing the AtUCP1 gene used in the present work have been described previously [7]. In this study, two independent and homozygous transgenic lines, termed P32 and P07, which present different levels of AtUCP1 expression [7] were chosen. Both overexpressors lines (OE) were used as replicas and parental untransformed plants were used as control. The seeds from the transgenic and wild type (WT) tobacco were surface-sterilized and sown in Petri dishes containing solid Murashige-Skoog (MS) medium. Dishes were maintained in a growth chamber at 20-22C with a 16/8 h light/dark photoperiod until seedling sampling. For the RNA-Seq and RT-qPCR analyses, entire 3-week-old seedlings from WT and OE lines (20 seedlings per genotype), respectively, were sampled. RNA isolation, quality control and sequencing The sampled seedlings (n=20) were pooled for total RNA extraction using Trizol reagent (Invitrogen). The RNA samples were then quantified spectrophotometrically using NanoDrop (Thermo Scientific). Total RNA integrity (1 g) was further evaluated using the RNA 6000 Nano LabChip Kit and a Bioanalyzer 2100 (Agilent Technologies). RNA samples with RNA integrity number ≥ 7 were used for library construction. The sequencing libraries were prepared using the TruSeq RNA Sample Preparation Kit v2 (Illumina). Sequencing was performed on the Illumina MiSeq platform using a paired-end 300 base pair (bp) run. De novo transcriptome assembly and mapping The raw data generated for each sample were separately trimmed and de novo assembled in a unique file using the CLC Genomics Workbench software (Version 6.5, CLC Bio). The reads were trimmed using the quality score limit of 0.08 and maximum limit of 2 ambiguous nucleotides, and then assembled using the “de novo assembly” algorithm. The trimmed reads were mapped to the de novo assembled contigs with parameters allowing mapping of reads to the genome with up to 3 mismatches. The reads mapped to ribosomal RNA as well as those not uniquely 27 mapped were removed from subsequent analyses. The expression levels were evaluated using the RPKM (reads per kilobase per million) method essentially as previously described [15]. Data accuracy was checked by the calculation of a correlation coefficient using the SigmaStat package based on the log-transformed RPKM values derived from the OE lines dataset and after eliminating genes with zero count. Functional annotation of transcripts Functional annotation was performed by the assignment of Gene Ontology (GO) terms [16] to the transcripts. For this purpose, the Blast2GO suite [17] with default parameters was used. Blast2GO assigns GO terms to sequences through a 3-step process: blasting, mapping and annotation. In the blasting step, sequences are blasted using BLASTX (E-value cutoff = 1e-5) against the sequence database provided by the National Center for Biotechnology Information (NCBI). In the mapping step, GO terms are mapped on the blast results using annotation files provided by the GO Consortium. In the annotation step, the GO terms mapped to the blast results are transferred to the target sequences using the default annotation parameters [17]. Differentially expressed genes were also annotated against metabolic maps gathered from Kyoto Encyclopedia of Genes and Genomes (KEGG) [18] using Blast2GO. Differential gene expression analysis The differential gene expression between OE lines and WT was determined using the CLC Genomics Workbench v6.5 (CLC Bio). For this, RPKM values were submitted to quantile normalization and then log2-transformed for subsequent statistical analysis. Hierarchical clustering of samples and principal component analysis were carried out to examine data quality and comparability. Aiming the identification of genes with significant changes in expression between the OE lines and WT (P07 vs WT and P32 vs WT), a Kal’s Z-test was performed on log2- transformed data [19]. For that, a 2-fold change cutoff plus a p-value cutoff of 0.001 was employed. GO-based functional enrichment analysis Enrichment analysis was carried out for the differentially expressed transcripts. For each type of GO categories, that is, biological process, molecular function and 28 cellular component, we compared the frequencies of GO terms mapped to the differentially expressed transcripts with those of entire set of expressed transcripts. To this end, The Biological Networks Gene Ontology tool (BiNGO) [20], an open- source Java tool to determine which GO terms are significantly overrepresented in a set of genes, was employed. The statistical test (hypergeometric test), the multiple testing corrections (Benjamini and Hochberg correction) and the confidence level (p < 0.05) were default values. The reference set, the ontology and organism/annotation files were all prepared in a customized way. For each type of ontology, the test set consisted in the set of differentially expressed transcripts and the reference set consisted in the complete set of expressed transcripts. Differentially expressed metabolic pathways To determine which metabolic pathways were differentially expressed between control and the investigated OE lines, we first identified the transcripts likely coding for enzymes using Blast2GO. Besides assigning GO terms, this software also assigns Enzyme Commission numbers (EC numbers) to sequences. The retrieved EC numbers were then mapped to the plant metabolic pathways collected from Gramene [21], an integrated data resource for comparative functional genomics in crops and model plant species. In subsequent analysis, only pathways with at least five mapped EC numbers were kept. Next we mapped back the EC numbers to their coding genes and then calculated the fold change in gene expression in OE lines compared with WT control plants. The fold change is given by log2(ExpA/ExpB), where ExpA and ExpB are, respectively, expression values for genes in transgenic and control plants. For each pathway containing k enzymes, the average fold change was calculated and statistically compared with the average fold change obtained by 100 sets of k enzymes randomly taken from the pathways. From this average fold change and its corresponding standard deviation, a Z-score and a p-value were calculated for each metabolic pathway. We considered differentially expressed the pathways with p-value <0.05. Positive and negative Z-scores indicate, respectively, up-regulated and down-regulated pathways. Validation of the RNA-Seq results by RT-qPCR In order to verify the accuracy of the expression profiles determined by RNA- Seq, 17 transcripts were selected for quantitative RT-PCR (RT-qPCR) assay (Table S1). The RNA (2 µg) of each sample was treated with DNaseI (Fermentas) and the 29 High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) was used for cDNA synthesis following the instructions of manufacturer. All cDNA samples were quantified on a NanoDropTMND-1000 spectrophotometer (NanoDrop Technologies). Relative gene expression levels were determined by quantitative real-time PCR (qPCR) using a Step One Plus Real-Time PCR System (Applied Biosystems) under the following parameters: 95◦C for 10 min, 40 cycles at 94◦C for 15 s, 60◦C for 60 s. Amplicon specificity was checked using the dissociation curve at the end of each run. Each PCR reaction (12 µl) consisted of 6µl of Maxima SYBR Green/ROX qPCR Master Mix (Thermo Scientific), 1 µl of cDNA and 0.4 µM of forward and reserve primers. All gene-specific primer pairs (Table S1) were designed using the Primer Express software (Applied Biosystems). In all reactions, the gene encoding an elongation factor (NtEF) from Nicotiana tabacum was used as an internal control to normalize levels of target transcripts (Table S1). The generated relative expression data were further analyzed using the DataAssist software v.3.01 (Applied Biosystems). Additionally, the correlation between the RNA-Seq and RT-qPCR expression values obtained for the selected genes (total of 17) was evaluated by calculating the Pearson correlation coefficient. Results Global analysis and functional annotation In other to get further insights into the broad transcriptional responses associated with the overexpression of an uncoupling protein, global gene expression profiles from 21-day-old seedlings of WT and two AtUCP1 OE lines (P07 and P32) were determined using RNA-Sequencing. At this developmental stage, the OE lines displayed no phenotypic differences when compared to WT plants with the exception of a faster vegetative growth [9]. Most importantly, however, is the fact that at this stage the AtUCP1 OE lines proved to be highly tolerant to salt and drought stresses [9]. It should be also highlighted that the two selected OE lines show different expression levels of the transgene, with P07 being considered a high expression line and P32 a moderate one [7, 10]. De novo assembling was performed by pooling the resulting RNA-Seq data of the investigated samples. The resulting assembling parameters are presented in Table 1. Among the 33.679 assembled contigs, 29.152 (86%) showed positive 30 matches in BLAST searches against the NCBI database and 25.246 (75%) were associated with at least one gene ontology (GO) term when classified using Blast2GO. Moreover, a high correlation (R2=0.947 with p-value < 0.001) between the two OE lines was observed (Figure 1), thus supporting the reliability of the raw dataset. This dataset was considered satisfactory to investigate the main transcriptomic effects of AtUCP1 overexpression in transgenic tobacco. Table 1. Assembling parameters Reads 9,777,662 Reads in pairs 6,333,375 Number of contigs (≤ 200) 33,679 N50 (bp) 632 Maximum Contig length (bp) 5,386 Minimum Contig length (bp) 197 Average (bp) 602 N50 = statistically weighted average such that 50% of the entire assembly is formed by contigs of equal size or larger than this value. 31 Figure 1. R2 linear regression of the RNA-Seq data from both OE lines (P07 and P32). The R2 values were calculated using the Sigma Stat package based on the RPKM values derived from RNA-Seq data and after eliminating genes with zero count. Differential gene expression and GO enrichment analysis By using a 2-fold change cutoff plus a p-value cutoff of 0.001, a total of 816 differentially expressed genes (DEGs) between the AtUCP1 OE lines and WT plants was detected, among which 239 were up-regulated and 577 were down-regulated. It should be emphasized that only DEGs that were in common between both OE lines were considered. As expected, AtUCP1 was the most up-regulated gene in both transgenic lines by more than 700-fold as compared to WT. As a starting point to functional annotation, we first determined the GO terms assigned to both downand up-regulated genes (Figure 2). It should be mentioned that in many cases, different GO terms were assigned to the same transcript. For the up-regulated genes (Figure 2A), “ATP binding" (23 transcripts - 9.62%) and “nucleotide binding" (15 transcripts - 6.27%) were the most predominant GO terms in the \molecular function" category, while “oxidation-reduction process" (28 transcripts - 11.71%) and “pentose-phosphate shunt" (25 transcripts - 10.87%) were prominent in 32 the “biological process" category. In the “cellular compartment" category, the terms “chloroplast envelope" (50 transcripts - 20.92%) and “chloroplast stroma" (42 transcripts - 17.57%) were overrepresented. For the down-regulated genes (Figure 2B), the terms “ATP binding” (66 transcripts - 11.43%), “response to cadmium ion” (53 transcripts - 9.18%), and “plasma membrane” (110 transcripts - 19.06%) were prominent in each of the three main functional categories, respectively. It is also noteworthy the presence of a large number of down-regulated genes associated with the terms “oxidation-reduction process” (52 transcripts - 9%) in the “biological process” category, and “cytosol” (110 transcripts - 17.33%) followed by “mitochondrion” (57 transcripts - 9.87%) in the “cellular compartment” category. A GO enrichment analysis was subsequently employed to identify the most representative ontologies associated to each DEG dataset. By using a Hypergeometric test with Benjamini and Hochberg correction through BiNGO [20], we compared, for each dataset, the frequencies of GO terms mapped to the up and down-regulated genes with those of entire set of expressed genes. By doing so, we determined the GO terms significantly enriched (FDR-corrected for p < 0.05) in the up-regulated and down-regulated datasets as compared to all expressed genes. For the up-regulated dataset, the enrichment analysis indicated a total of 69 and 12 significantly enriched GO terms associated with “biological process" and “cellular component" categories, respectively (Table S3). A graphical illustration of the results of the GO enrichment analysis for the “biological process" category is available in File S1. In this category, the terms “pentose-phosphate shunt" and “rRNA processing" were the most significantly enriched followed by “cysteine biosynthetic process". Besides “pentose-phosphate shunt", other terms related to carbohydrate metabolism (maltose metabolic process, sucrose metabolic process, starch biosynthetic process, starch metabolic process, polysaccharide catabolic process) were also significantly enriched (Figure 3B). Notably, several terms associated with photosynthesis (photosynthesis, light reaction; photosystem II assembly; photosystem II stabilization; photosynthetic electron transport in photosystem I) were overrepresented in the up-regulated gene dataset (detailed view in Figure 3A). This dataset was also significantly enriched for the term “chlorophyll biosynthetic process" (Figure 3A). Consonant with the observed enrichment of photosynthesis related terms, the majority of the GO terms significantly enriched in the “cellular component” category 33 (Table S3) was associated with the chloroplast (chloroplast envelope; chloroplast stroma; chloroplast thylakoid membrane; chloroplast thylakoid lumen). Overall, these results indicate an important effect of AtUCP1 overexpression on chloroplast metabolism and confirm the previously reported relationship between UCP activity and photosynthetic efficiency [13,22,23]. In this respect, it should be highlighted that the net photosynthetic rates of the AtUCP1 OE lines were higher than WT plants under control conditions [9]. Interestingly, we also observed the enrichment of several GO terms associate with cold and biotic stress responses (defense response to bacteria, systemic acquired resistance, and detection of biotic stimulus) within the set of up-regulated genes (Table S3; Figure 3C). These results suggest a connection between UCP- mediated uncoupling and the triggering of specific plant stress responses. In fact, changes in transcript abundance of defense related genes in response to perturbations in the electron transport chain have already been observed (reviewed in [24]. In addition, the observed up-regulation of genes related to cold response is in accordance with previous studies reporting a link between the cold induction of nuclear gene expression and decreased efficiency of oxidative phosphorylation [25, 26]. It is worthy of note in this context that UCP action can lead to a fast electron flux through the respiratory chain. Last, concerning the “molecular function” category, the only enriched term was “delta12-fatty acid dehydrogenase activity”, which is required for the generation of unusual fatty acids. For the down-regulated dataset, a total of 13 and 5 terms were significantly enriched in the “biological process” and “cellular component” categories, respectively (Table S4). Among them, “chlorophyll catabolic process” and “response to cadmium ion” were the most significantly enriched terms followed by “protein folding”. Although unexpected, the overrepresentation of transcripts associated with cadmium response is in line with a proposed role for the alternative respiratory pathway in modulating cadmium-induced stress responses [27]. Considering the terms significantly enriched in the “cellular component” category, the majority was related to “cytosol” followed by “plasmodesma” and “plasma membrane” (Table S4). Interestingly, GO terms linked to certain abiotic stress responses (response to oxidative stress, response to water deprivation, hyperosmotic responses) and also to fatty acid oxidation were significantly enriched in the down-regulated gene dataset. It seems thus that fatty acid oxidation is lowered in the OE lines in order to compensate for the metabolic alterations promoted by AtUCP1 overexpression. It should be emphasized that fatty 34 acids are important regulators of UCP activity (reviewed in [6]). Genes involved in oxidative phosphorylation Due to the described effect of the dissipating activity of overexpressed AtUCP1 on the yield of ATP synthesis [10], we focused attention on the genes encoding products involved in oxidative phosphorylation. Curiously, none of the examined genes figured as up-regulated in our dataset (Table S2). On the contrary, many genes encoding components of the mitochondrial respiratory chain, such the beta subunit of F1F0 ATP synthase, the iron-sulfur subunit 2 of succinate dehydrogenase and different subunits of the NADH dehydrogenase complex (flavoprotein 1, iron-sulfur protein 6, 1 alpha subcomplex subunit 13), respectively, were down-regulated in the OE lines (Table S2). In view of these results, it seems that during seedling development the negative impact of AtUCP1 action in the intracellular ATP level [10] is not compensated by the transcriptional up-regulation of components of the respiratory chain. 35 Figure 2. Gene Ontology (GO) analysis of the differentially expressed genes. Distribution histograms within the three main GO categories (in red “molecular function", in blue “biological process" and in green “cellular component") are shown for the up- (A) and down-regulated (B) genes. Only differentially expressed genes that were in common between the two AtUCP1 OE lines (P07 and P32) were included in the analysis. 36 Figure 3. Enriched network based on GO terms in the up-regulated gene dataset. Fragment of the complete network presented in File S1 showing significantly enriched GO terms in the up-regulated gene dataset associated with photosynthesis (A), carbon metabolism (B) and biotic and cold stress responses (C). Data were analyzed using the Biological Networks Gene Ontology tool (BiNGO). Node color means enrichment significance i.e. darker the color on a color scale ranging from white (no significant) to orange (FDR-corrected p-value = 3 x 10�10), higher is the enrichment significance. White color nodes are not enriched but show the hierarchical relationship among the enriched ontology branches. 37 Altered metabolic pathways in the OE lines The KEGG database was employed to identify the main metabolic pathways altered in the AtUCP1 OE lines. As a result, 52 KEGG pathways (showing 5 or more altered genes) were affected by AtUCP1 overexpression (Table S5). Amongst them, “starch and sucrose metabolism" was the most affected with a great number of allocated up- and down-regulated genes (16 and 12, respectively). Other pathways implicated in carbohydrate metabolism (such as pyruvate metabolism, glycolysis and gluconeogenesis, glyoxylate and dicarboxylate metabolism, pentose phosphate pathway and Citrate cycle) also presented several allocated genes. A substantial number of genes were also mapped to KEGG pathways directly or indirectly implicated in energy metabolism (such as carbon fixation in photosynthetic organisms, nitrogen metabolism, oxidative phosphorylation and porphyrin and chlorophyll metabolism). In this context, “carbon fixation in photosynthetic organisms" was positively affected with 8 up-regulated associated genes, while “oxidative phosphorylation" was negatively affected with 6 down- regulated associated genes. Concerning lipid metabolism, the pathways “fatty acid elongation", “alpha-linolenic acid metabolism" and “fatty acid degradation" were negatively affected with uniquely down-regulated genes associated to them. Alterations in purine metabolism and in different amino acid metabolic pathways were also prominent in the AtUCP1 OE lines. In this respect, \cysteine and methionine metabolism" was the most altered, while “valine, leucine and isoleucine degradation" and “lysine degradation" were negatively affected. Overall, these results indicate that the most important changes in gene expression detected in the AtUCP1 OE lines were mainly associated with pathways involved in carbohydrate metabolism, cell energy supply and amino acid metabolism. Collectively, these data support the proposed role of AtUCP1 in the control of carbon and nitrogen partitioning [28]. As a complementary approach, by using the pathways retrieved from the Gramene database, we identified the pathways considered to be differentially expressed between OE lines and WT. Based on this analysis, the pathways significantly up-regulated in the AtUCP1 OE lines were: Calvin-Benson-Bassham cycle (Z-score 2.1), chlorophyllide a biosynthesis I (Z-score 3.63) and tetrapyrrole biosynthesis I (Z-score 2.21). On the other hand, the pathways significantly down- regulated were: vitamin E biosynthesis (Z-score -2.5), 4-hydroxybenzoate biosynthesis V (Z-score -2) and tyrosine degradation I (Z-score -3.2). 38 RT-qPCR validation of differential expression The accuracy of the expression profiles derived from the RNA-Seq analysis was further verified using RT-qPCR. For that, 17 DEGs (12 up-regulated and 5 down- regulated; Table S1) were chosen based on their involvement in the aforementioned altered pathways, or randomly selected among the most up and down-regulated genes. Overall, a positive correlation between the expression profiles determined in both assays was observed for each transgenic line investigated (Figure 4) (Pearson coefficient of 0.802 for P07 and 0.7212 for P32), and both correlations were significant (p<0.05). According to the relative expression levels determined by RT-qPCR, all the genes considered upregulated by the RNA-Seq analysis also showed induced expression levels in both OE lines as compared to the WT (Figure 5A). The only exception was the GSA gene whose relative expression was not altered. However, for three of them (CYP, MS and PBGD), a statistically significant up-regulation was only detected in the P07 OE line. Concerning the validation of the selected down-regulated genes, the RT-qPCR results revealed that three of the validated genes (AIM, HGO and HPPD) showed low relative expression in both OE lines as compared to WT control, but were significantly down-regulated only in the P32 OE line (Figure 5B). On the contrary, the 4CL and FAH genes were significantly down-regulated only in P07. Despite the observed divergences, the RT-qPCR results confirmed the down-regulation of these 5 genes in at least one of the OE lines investigated. 39 Figure 4. Correlation between RNA-Seq and RT-qPCR data Twelve up-regulated and five down- regulated genes were selected for validation and a Pearson's correlation was calculated based on the expression values generated for each AtUCP1 OE line (P07 and P32). 40 Figure 5. RT-qPCR validation of a set of differentially expressed genes (A) Validation of the selected up-regulated genes. (B) Validation of the selected down-regulated genes. Assays were performed using the same RNA samples used for RNA-sequencing of both OE lines (P07 and P32) and WT expression was set to 1. Bars indicate standard errors of triplicate reactions (*means p < 0.05). 41 Discussion In this study we determined the global expression changes associated with the overexpression of a mitochondrial uncoupling protein (AtUCP1) in tobacco. For that, developing seedlings of two independent OE lines showing high abiotic stress tolerance and a fast growth rate during initial development were employed [9]. Our results demonstrate that AtUCP1 overexpression promotes substantial alterations in the transcriptome of the OE lines resulting in a total of 816 in common DEGs when compared to WT nontransformed plants. As a general response to UCP1 overexpression, noticeable changes in the expression of genes involved in carbon and energy metabolism were observed, indicating that an important transcriptome reprogramming is necessary to maintain energy homeostasis in the transgenic plants. Interestingly, the transcriptomic patterns identified here resemble, at least in part, those triggered by different mitochondrial dysfunctions [29, 30]. Several genes encoding chloroplast-localized proteins were up-regulated in the OE lines, indicating that AtUCP1 overexpression has an important impact outside the mitochondrion as already observed for the alternative oxidase [4, 31]. AtUCP1 overexpression enhanced, for example, the expression of genes encoding important subunits of the reaction centers of photosystem I and II (PSI and PSII) and also of the cytochrome b6f complex. Moreover, genes encoding oxygen-evolving enhancer proteins and the light-harvesting chlorophyll a/b-binding proteins are amongst the most up-regulated genes in both AtUCP1 OE lines. The detection of transcriptional changes in a significant number of photosynthesis-related genes (Figure 3A) is in line with the proposed role of UCP in maintaining rates of photosynthesis. In this respect, previous studies have demonstrated that UCP activity affects photosynthetic efficiency [13, 22, 23], and increase photosynthetic capacity when overexpressed in transgenic plants [8, 9]. Consistently, a connection between mitochondrial function and the transcriptional regulation of nuclear genes encoding photosynthetic proteins was already evidenced in Arabidopsis and Chlamydomonas reinhardtii [30, 32, 33]. The aforementioned results strongly indicate that AtUCP1 overexpression triggers a retrograde signaling that affect photosynthetic gene expression. Particularly interesting in this context was the up-regulation of pathways associated with tetrapyrrole and chlorophyllide a biosynthesis in the AtUCP1 OE lines. Although the requirement for de novo chlorophyll biosynthesis could be associated with the 42 elevated photosynthetic rates of the AtUCP1 OE lines [9], intermediates of such pathways (Mg-protoporphyrin IX for example) as well as changes in the flux through the tetrapyrrole pathway have been shown to play an important role in plastid-to- nucleus retrograde signaling (reviewed in [34, 35]. Notably, porphyrins have been also postulated to participate in interorganelle signaling (reviewed in [34]). Earlier data reported that the photosynthetic performance of an atucp1 insertion mutant was impaired due to the inhibition of the photorespiratory flux and limited regeneration of ribulose-1.5-bisphosphate (RuBP) [22]. In fact, many published data support the view that impaired photorespiration can restrict photosynthesis under situations that promote high oxygenation of the Rubisco enzyme, especially under stressful conditions (reviewed in [36]. Moreover, evidence from UCP-silenced tomato plants suggests that UCP activity affects Rubisco carboxylation and RuBP regeneration rates [23]. In this respect, the most strongly up- regulated gene in our analysis was GLYK (Table S2), which encodes a key regulator of photorespiration. D-glycerate 3-kinase (GLYK) catalyzes the concluding reaction of the photorespiratory cycle and is crucial for the regeneration of RuBP [37]. It seems plausible that the detected up-regulation of GLYK probably represents a way to sustain the elevated photosynthetic rates displayed by the AtUCP1 OE lines [9]. Consistent with this, the Calvin-Benson cycle figured as up-regulated in the AtUCP1 OE lines indicating that RuPB does not limit cycle activity. Another interesting finding concerning the photorespiratory pathway was the detected up-regulation of the gene encoding the mitochondrial malate dehydrogenase (mMDH). This enzyme, classically known for its role in the tricarboxylic acid (TCA) cycle, also operates during the conversion of glycine to serine by reducing oxaloacetate to malate and regenerating NAD+ for glycine decarboxylase [38]. In this context, Sweetlove et al. [22] showed that the photorespiratory flux of glycine to serine was substantially decreased in the atucp1 mutant. It is therefore possible that the observed mMDH up-regulation in the OE lines would avoid excess NADH accumulation and sustain NAD+ regeneration. Of note, an enrichment of terms associated with the pentose phosphate pathway was observed (Figure 3B) and probably reflects the need of reductant availability to maintain cellular redox homeostasis in the OE lines. UCP overexpression has been previously found to increase TCA cycle flux with no apparent alterations in the maximal catalytic activity of specific TCA cycle 43 enzymes [12]. Among the genes involved in the TCA cycle represented in our dataset, those encoding the E1 beta and alpha subunits of the mitochondrial pyruvate dehydrogenase complex, respectively, were found to be significantly up- regulated in the OE lines (Table S2). The PHD complex catalyzes the conversion of pyruvate into acetyl-CoA that is subsequently consumed in the TCA cycle. We also notice, as aforementioned, the up-regulation of mMDH, a classical TCA cycle enzyme that catalyzes the conversion of malate into oxaloacetate. Additionally, AtUCP1 overexpression negatively regulated the expression of a gene encoding the iron sulfur subunit 2 of the mitochondrial succinate dehydrogenase. Succinate dehydrogenase (Complex II) is a multimeric enzyme that has a dual role in mitochondrial metabolism, acting both in the TCA cycle and as a member of the respiration chain. It was previously demonstrated that plants overexpressing this subunit show high rate of photosynthesis and CO2 assimilation but lower rate of TCA cycle [39]. Thus, substantial transcriptional modulations of TCA cycle-related genes are observed in the AtUCP1 OE lines, which seem likely to occur in response to an altered cycle flux. Similarly, a moderate increase in the TCA cycle enzymes was part of the response induced by the ectopic expression of mammalian UCP1 in yeast [40]. Surprisingly, despite the already described reduction in cellular ATP concentration (20-35%) provoked by the dissipating action of AtUCP1 in leaves of the P07 OE line [10], none of the genes encoding the major components of the mitochondrial respiratory chain complexes figured as up-regulated in our datasets. On the contrary, the expression of a number of genes coding for components of complex I, complex II and the beta subunit of ATP synthase was significantly down- regulated in both OE lines (Tables S2 and S5). The reason for this is unclear but Umbach et al. [31] also observed no changes in the transcript abundance of selected electron transport components in the leaves of normally grown transgenic Arabidopsis overexpressing another component of the mitochondrial alternative pathway (AOX1a). Our results are in line with the undetected differences in the respiration (measured as CO2 release) of the AtUCP1 OE lines and WT plants [9], but contrast with those obtained by Barreto et al. [10] showing an up-regulation of both nuclear-encoded (NADH-DeH and NADH-IS7) and mitochondrial-encoded (NADH-IS2) components of complex I in the leaves of 12-week-old P07 plants. These discrepancies could be attributed to differences in the developmental stage and overall physiological status of the sampled material (intact seedlings vs. leaves), and 44 point to the existence of additional posttranscriptional and posttranslational mechanisms [31]. In conclusion, the transcriptional profiling presented here provides novel insights into the molecular responses brought about by the augmented expression of an uncoupling protein in plants. The overall results also highlight the occurrence of an important transcriptional cross-talk between chloroplast and mitochondria, which culminates in the transcriptional regulation of genes mainly involved in photosynthesis, photorespiration and carbon metabolism. Acknowledgments We thank Dr. Alessandro de Mello Varani from the Faculdade de Ciências Agrárias e Veterinárias at UNESP - Univ Estadual Paulista, Jaboticabal, SP, Brazil, for his help with bioinformatics analysis. Alessandra V. N. Laitz and Marcio L. Acencio are recipients of a PhD and PNPD fellowships from CAPES, respectively. Ivan G. Maia and Paulo E. M. Ribolla are recipients of research fellowships from CNPq. References 1. Jacoby RP, Li L, Huang S, Pong Lee C, Millar AH, et al. (2012) Mitochondrial Composition, function and stress response in plants. Journal of Integrative Plant Biology 54: 887-906. 2. Rhoads DM, Subbaiah CC (2007) Mitochondrial retrograde regulation in plants. Mitochondrion 7:177-194. 3. Van Aken O, Giraud E, Clifton R, Whelan J (2009) Alternative oxidase: a target and regulator of stress responses. Physiologia Plantarum 137: 354-361. 4. Clifton R, Millar AH, Whelan J (2006) Alternative oxidases in Arabidopsis: a comparative analysis of differential expression in the gene family provides new insights into function of non-phosphorylating bypasses. Biochimica et Biophysica Acta (BBA)-Bioenergetics 1757: 730-741. 5. Vanlerberghe GC (2013) Alternative oxidase: a mitochondrial respiratory pathway to maintain metabolic and signaling homeostasis during abiotic and biotic stress in plants. International Journal of Molecular Sciences 14: 6805-6847. 6. Vercesi AE, Boreck_y J, Maia IdG, Arruda P, Cuccovia IM, et al. (2006) Plant uncoupling mitochondrial proteins. Annu Rev Plant Biol 57: 383-404. 7. Brandalise M, Maia IG, Boreck_y J, Vercesi AE, Arruda P (2003) Overexpression of plant uncoupling mitochondrial protein in transgenic tobacco 45 increases tolerance to oxidative stress. Journal of Bioenergetics and Biomembranes 35: 203-209. 8. Chen S, Liu A, Zhang S, Li C, Chang R, et al. (2013) Overexpression of mitochondrial uncoupling protein conferred resistance to heat stress and Botrytis cinerea infection in tomato. Plant Physiology and Biochemistry 73: 245-253. 9. Begcy K, Mariano ED, Mattiello L, Nunes AV, Mazzafera P, et al. (2011) An Arabidopsis mitochondrial uncoupling protein confers tolerance to drought and salt stress in transgenic tobacco plants.PloS One 6: e23776. 10. Barreto P, Okura VK, Neshich IA, de Maia I, Arruda P (2014) Overexpression of UCP1 in tobacco induces mitochondrial biogenesis and amplifies a broad stress response. BMC Plant Biology 14: 144. 11. Kowaltowski AJ, Costa AD, Vercesi AE (1998) Activation of the potato plant uncoupling mitochondrial protein inhibits reactive oxygen species generation by the respiratory chain. FEBS letters 425: 213-216. 12. Smith AM, Ratcliffe RG, Sweetlove LJ (2004) Activation and function of mitochondrial uncoupling protein in plants. Journal of Biological Chemistry 279: 51944-51952. 13. Chen S, Liu A, Ji D, Lin X, Liu Z, et al. (2014) Silencing of tomato mitochondrial uncoupling protein disrupts redox poise and antioxidant enzymes activities balance under oxidative stress. Journal of Plant Biology 57: 9-19. 14. Camara Y, Mampel T, Armengol J, Villarroya F, Dejean L (2009) UCP3 expression in liver modulates gene expression and oxidative metabolism in response to fatty acids, and sensitizes mitochondria to permeability transition. Cellular Physiology and Biochemistry 24: 243-252. 15. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B (2008) Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nature Methods 5: 621- 628. 16. Consortium GO, et al. (2013) Gene Ontology annotations and resources. Nucleic Acids Research 41: D530{D535. 17. Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, et al. (2005) Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:674-3676. 18. Kanehisa M, Goto S, Sato Y, Furumichi M, Tanabe M (2011) KEGG for integration and interpretation of large-scale molecular data sets. Nucleic Acids Research : gkr988. 19. Kal AJ, van Zonneveld AJ, Benes V, van den Berg M, Koerkamp MG, et al. (1999) Dynamics of gene expression revealed by comparison of serial analysis of gene expression transcript profiles from yeast grown on two different carbon sources. Molecular Biology of the Cell 10: 1859-1872. 46 20. Maere S, Heymans K, Kuiper M (2005) BiNGO: a Cytoscape plugin to assess overrepresentation of gene ontology categories in biological networks. Bioinformatics 21: 3448-3449. 21. Monaco MK, Stein J, Naithani S, Wei S, Dharmawardhana P, et al. (2014) Gramene 2013: comparative plant genomics resources. Nucleic Acids Research 42: D1193-D1199. 22. Sweetlove LJ, Lytovchenko A, Morgan M, Nunes-Nesi A, Taylor NL, et al. (2006) Mitochondrial uncoupling protein is required for efficient photosynthesis. Proceedings of the National Academy of Sciences 103: 19587-19592. 23. Chen S, Jin W, Liu Ar, Liu Z, Liu D, et al. (2013) Mitochondrial uncoupling protein silence is compromised in photosynthesis and redox poise. Acta Physiologiae Plantarum 35: 2547-2558. 24. Amirsadeghi S, Robson CA, Vanlerberghe GC (2007) The role of the mitochondrion in plant responses to biotic stress. Physiologia Plantarum 129: 253- 266. 25. Lee Bh, Lee H, Xiong L, Zhu JK (2002) A mitochondrial complex I defect impairs cold-regulated nuclear gene expression. The Plant Cell Online 14: 1235- 1251. 26. Meyer EH, Tomaz T, Carroll AJ, Estavillo G, Delannoy E, et al. (2009) Remodeled respiration in ndufs4 with low phosphorylation efficiency suppresses Arabidopsis germination and growth and alters control of metabolism at night. Plant Physiology 151: 603-619. 27. Keunen E, Jozefczak M, Remans T, Vangronsveld J, Cuypers A (2013) Alternative respiration as a primary defence during cadmium-induced mitochondrial oxidative challenge in Arabidopsis thaliana. Environmental and Experimental Botany 91: 63-73. 28. Gandin A, Denysyuk M, Cousins AB (2014) Disruption of the mitochondrial alternative oxidase (AOX) and uncoupling protein (UCP) alters rates of foliar nitrate and carbon assimilation in Arabidopsis thaliana. Journal of Experimental Botany : eru158. 29. Van Aken, O.; Whelan, J. (2012) Comparison of transcriptional changes to chloroplast and mitochondrial perturbations reveals common and specific responses in Arabidopsis. Front. Plant Science: doi:10.3389/fpls.2012.00281. 30. Schwarzländer M, K• onig AC, Sweetlove LJ, Finkemeier I (2012) The impact of impaired mitochondrial function on retrograde signalling: a meta-analysis of transcriptomic responses. Journal of Experimental Botany 63: 1735{1750. 31. Umbach AL, Fiorani F, Siedow JN (2005) Characterization of transformed Arabidopsis with altered alternative oxidase levels and analysis of effects on reactive oxygen species in tissue. Plant Physiology 139: 1806-1820. 47 32. Matsuo M, Obokata J (2006) Remote control of photosynthetic genes by the mitochondrial respiratory chain. The Plant Journal 47: 873-882. 33. Matsuo M, Hachisu R, Tabata S, Fukuzawa H, Obokata J (2011) Transcriptome analysis of respiration-responsive genes in Chlamydomonas reinhardtii: mitochondrial retrograde signaling coordinates the genes for cell proliferation with energy-producing metabolism. Plant and Cell Physiology 52: 333- 343. 34. Nott A, Jung HS, Koussevitzky S, Chory J (2006) Plastid-to-nucleus retrograde signaling. Annu Rev Plant Biol 57: 739-759. 35. Rintam• aki E, Lepist• o A, Kangasj• arvi S (2009) Implication of chlorophyll biosynthesis on chloroplastto-nucleus retrograde signaling. Plant Signal Behav 4: 545-547. 36. Wingler A, Lea PJ, Quick WP, Leegood RC (2000) Photorespiration: metabolic pathways and their role in stress protection. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 355: 1517-1529. 37. Boldt R, Edner C, Kolukisaoglu Ü, Hagemann M, Weckwerth W, et al. (2005) D-Glycerate 3-kinase, the last unknown enzyme in the photorespiratory cycle in Arabidopsis, belongs to a novel kinase family. The Plant Cell Online 17: 2413- 2420. 38. Tomaz T, Bagard M, Pracharoenwattana I, Lind_en P, Lee CP, et al. (2010) Mitochondrial malate dehydrogenase lowers leaf respiration and alters photorespiration and plant growth in Arabidopsis. Plant Physiology 154: 1143-1157. 39. Araújo WL, Nunes-Nesi A, Osorio S, Usadel B, Fuentes D, et al. (2011) Antisense inhibition of the iron-sulphur subunit of succinate dehydrogenase enhances photosynthesis and growth in tomato via an organic acid{mediated effect on stomatal aperture. The Plant Cell Online 23: 600-627. 40. Douette P, Gerkens P, Navet R, Leprince P, De Pauw E, et al. (2006) Uncoupling protein 1 affects the yeast mitoproteome and oxygen free radical production. Free Radical Biology and Medicine 40:303-315. 48 Supporting Information Legends Table S1. Genes selected for expression validation by RT-qPCR with corresponding primers and detected RNA-Seq fold changes. Table S2. List of differentially expressed genes. Fold change indicates the average of up- or down-regulation in the AtUCP1 OE lines compared with WT. Table S3. Significantly enriched GO terms for up-regulated genes. Table S4. Significantly enriched GO terms for down-regulated genes. Table S5. KEGG metabolic pathways altered in the AtUCP1 OE lines. Only pathways with 5 or more affected genes are represented. File S1. Cytoscape file containing the network visualization of part of the hierarchical relationship among GO biological process terms. Colored nodes are significantly overrepresented GO categories (node colors represent enrichment significance as depicted in Figure 3) associated with the up-regulated gene dataset. Data were analyzed using the Biological Networks Gene Ontology tool (BiNGO). 49 5- Capítulo II. Análise dos mutantes de inserção de Arabidopsis thaliana Em paralelo à análise de transcriptoma e expressão diferencial descrita no capítulo I, estudos empregando mutantes de inserção foram realizados com o objetivo de analisar a funcionalidade das proteínas desacopladoras presentes em arabidopsis (AtUCP1, AtUCP2 e AtUCP3), bem como testar experimentalmente uma possível complementação e redundância funcional entre elas. Os dados obtidos com essa etapa do trabalho serão apresentados a seguir. II.1 – Material e Métodos II.1.1 – Crescimento Vegetal As sementes de Arabidopsis thaliana Col-0 correspondentes aos diferentes mutantes de inserção de T-DNA usados neste estudo foram adquiridas junto ao banco Arabidopsis Biological Resource Center (http://www.arabidopsis.org/). Foram utilizadas sementes de A. thaliana Col-0 contendo inserções nos genes AtUCP2 (At5g58970) (SALK_037074 e SALK_040242) e AtUCP3 (At1g14140) (SALK_123501C). As sementes do mutante de inserção atucp1 foram gentilmente doadas pelo Dr. Lee Sweetlove do Department of Plant Sciences da Universidade de Oxford. Adicionalmente, sementes de A. thaliana Col-0 selvagem foram utilizadas como controle. As sementes foram germinadas em uma mistura de solo com vermiculita na proporção 2:1, sendo primeiramente incubadas a 4ºC por 2 dias. Em seguida foram mantidas em câmara climatizada com fotoperíodo de 16 horas/dia de luz artificial e temperatura de 19º-21ºC. Após 30 dias foram coletadas folhas de cada uma delas, separadamente, para extração de DNA genômico. II.1.2 –Extração de DNA genômico e genotipagem O DNA genômico foi extraído utilizando-se o método de CTAB (brometo de cetil-trimetilamônio) como descrito por Doyle e Doyle (1987) com modificações. Cerca de 50 – 100 mg de tecido foliar dos respectivos mutantes foram macerados em nitrogênio líquido com posterior solubilização em 400 μl do tampão de extração 50 CTAB contendo 2% de β-mercaptoetanol. Adicionou-se em seguida 400 μl de CIA (clorofórmio-álcool isoamílico na proporção 24:1) seguido de uma centrifugação a 13.200 rpm durante 3 min. O sobrenadante foi coletado e ao mesmo foi adicionado 500 μl de isopropanol gelado (-20ºC) com subsequente centrifugação a 13.200 rpm por 10 minutos a 4ºC. O precipitado foi lavado com 1 ml de etanol 70% e centrifugado a 13.200 rpm por 5 min. Descartou-se o sobrenadante e o precipitado foi ressuspendido em água. A genotipagem visando detectar indivíduos homozigotos foi realizada por PCR utilizando o GoTaq Master Mix (Promega) contendo 3 μl do DNA genômico extraído (aprox. 200 ng), 0,6 μl dos oligonucleotídeos (10 mM) específicos para os genes AtUCP2 e AtUCP3, respectivamente, e 0,6 μl do oligo SALK LBb1.3 (10 mM; Tabela 1) que é direcionado para as bordas do T-DNA (http://signal.salk.edu/tdnaprimers.2.html). A reação de PCR foi constituída de um ciclo prévio de desnaturação a 95°C por 5 minutos, seguido de um ciclo de desnaturação a 94°C por 30 segundos, anelamento a 51°C por 1 minuto e extensão a 72°C por 1 minuto, repetido por trinta e cinco vezes, e um ciclo final de extensão a 72°C por 7 minutos. Amostras de DNA genômico de A. thaliana selvagem foram empregadas como controle. Os produtos de amplificação foram visualizados em gel de agarose 1% corado com brometo de etídeo. Tabela 1 – Sequência dos oligonucleotídeos empregados para a genotipagem dos mutantes atucp. Primer Sequência UCP2F 5'- AAGAGGGTCTATCTAATGACG- 3' UCP2R 5' - GCCACAAACCAATAAATAGAA - 3' UCP3F 5' - ACTCTTTCTGATTGTAGGA - 3' UCP3R 5'- GACTTTAAAGCCCAATATTAT - 3' Salk Lb1.3 5'- ATTTTGCCGATTTCGGAA - 3' II.1.3 – Sequenciamento Para confirmar a posição de inserção do T-DNA, 250 ng do produto de PCR de cada mutante (item II.1.2) foi sequenciado utilizando-se o kit comercial Big Dye versão 3.1 (Applied Biosystems). As reações de sequenciamento foram realizadas em sequenciador automático de DNA modelo ABI Prism 377 (Applied Biosystems) 51 utilizando oligonucleotídeos reversos específicos para cada gene alvo e o oligo Salk Lb1.3 (Tabela 1). II.1.4 - Extração de RNA total e análise de RT-qPCR Com o objetivo de avaliar o efeito da inserção do T-DNA na expressão de cada gene alvo procedeu-se a extração de RNA total dos respectivos mutantes seguido de uma análise de RT-qPCR. Para a extração de RNA total foi utilizado o reagente Trizol (Invitrogen) segundo protocolo descrito pelo fabricante. Resumidamente, 1 ml de Trizol foi adicionado à cerca de 100-200 mg do material vegetal macerado e procedeu-se a homogeneização em agitador. Após incubação por 5 minutos a temperatura ambiente, as amostras foram centrifugadas a 12000xg por 15 minutos a 4°C e o sobrenadante foi coletado seguido da adição de 200 μl de clorofórmio. A mistura foi homogeneizada em agitador, incubada por 3 minutos em temperatura ambiente e centrifugada a 12000xg por 15 minutos a 4ºC. A fase aquosa foi transferida para um novo tubo e adicionou-se 500 μl de isopropanol. Após incubação por 60 minutos a -20°C e centrifugação por 10 minutos a 4ºC, o sobrenadante foi removido e o sedimento lavado com 1 ml de etanol 70%. Após uma nova centrifugação a 7500xg por 5 minutos a 4°C, o sedimento foi seco e ressuspenso em 30 μl de H2O tratada com diethylpyrocarbonate (DEPC). A integridade do RNA extraído foi determinada por eletroforese em gel de agarose 1% corado com brometo de etídeo. As amostras de RNA extraídas foram ajustadas para uma concentração de 2000 ng com a ajuda de um espectrofotômetro ND-1000 (Fisher Thermo, USA), sendo em seguida tratadas com DNaseI (Fermentas) e submetidas a transcrição reversa utilizando-se o High Capacity cDNA Reverse Transcription Kit conforme