( PDF ) Estudo estrutural de proteínas alvo de mycobacterium tuberculosis

Download PDF

ads:

______________________________________________________________________

Marcio Vinicius Bertacine Dias

Estudo Estrutural de Proteínas Alvo

de Mycobacterium tuberculosis

Tese apresentada ao Departamento

de Física do Instituto de

Biociências, Letras e Ciências

Exatas (IBILCE) da Universidade

Estadual Paulista “Júlio de

Mesquita Filho” (UNESP) para

obtenção de título de Doutor em

Biofísica Molecular.

Orientador: Prof. Dr. João Rugguiero Neto

Co-orientador: Prof. Dr. Walter Filgueira de Azevedo Jr.

São José do Rio Preto – SP

2007

______________________________________________________________________

ads:

Livros Grátis

http://www.livrosgratis.com.br

Milhares de livros grátis para download.

O presente trabalho foi realizado no Laboratório de Sistemas Biomoleculares (BMSys),

Departamento de Física, Instituto de Biociências, Letras e Ciências Exatas/IBILCE de

São José do Rio Preto, SP, Universidade Estadual Paulista/UNESP, sob orientação do

Prof. Dr. João Ruggiero Neto e co-orientação do Prof. Dr. Walter Filgueira de Azevedo

Júnior, com auxílio da Fundação de Amparo à Pesquisa do Estado de São

Paulo/FAPESP (processo 03/12472-2).

ads:

À minha amada e querida mãe, Catarina,

E em memória ao meu inestimável pai, David

Dedico este trabalho aos amigos

Aos que se tornaram familiares,

Aos que nasceram familiares

E aos que conheci antes de ontem.

Dedico tanto aos que me deixam louco,

Quanto aos que enlouqueço.

Aos que me criticam em tudo,

E a um ou outro que atura

Minha “chatura”.

Aos amigos que correm,

Aos amigos que contemplam.

Aos que me consideram muito,

E aos muitos que, com razão, fazem pouco.

Aos que conhecem o que penso,

E aos que só conhecem o que faço.

Aos que passam o dia todo comigo,

E aos que estão todo tempo em mim.

Este trabalho é a soma de todos vocês.

E se ele não é melhor,

É por falta de memória,

Mas não por falta de amigos.

(autor desconhecido)

Agradecimentos:

Ao final deste trabalho são inúmeras as pessoas que eu devo o meu sincero obrigado, mas

entre estas as que me veio à mente no momento que eu escrevi esta parte do trabalho, sem

serem mais ou menos importantes as que infelizmente, pelo meu egoísmo ou esquecimento,

deixei de colocar neste local, são:

Ao meu orientador Prof. Dr. João Ruggiero Neto, por ter me socorrido no momento que

mais precisei; Pela amizade e confiança depositada em mim;

Ao meu co-orientador, Prof. Dr. Walter F. de Azevedo Jr., pela amizade, orientação,

ensinamentos, oportunidades e confiança depositada em mim;

Ao Prof. Dr. Valmir Fadel, pela amizade, apoio, conselhos e confiança depositada em

mim e pelo auxílio nas coletas realizadas no LNLS;

À Fernanda Canduri, pela amizade, apoio, conselhos, sendo que mesmo distante sempre

esteve perto nos momentos que eu mais precisei;

Às minhas ex-orientadoras, Profas. Dras. Marlene K. H. Kobayashi, Hermione Elly

Melara de Campos Bicudo, Ana Elizabete Silva e Fernanda Canduri, por a cada uma de

sua maneira, por todos os conhecimentos transmitidos e por me despertarem o gosto pela

ciência;

Às minhas queridas “alunas”, Brianna, Lívia, Michele e Samantha, que estiveram

sempre comigo me ensinando a ensinar e sempre trabalhando com dedicação e afinco no

Laboratório. E que sem essas maravilhosas pessoas grande parte deste trabalho talvez

não teria sido efetuada;

Ao meu grande amigo José Henrique Pereira, por toda amizade, conselhos, e ajuda tanto

na vida profissional como pessoal. A sua esposa, Adriana e sua família, que se tornou

uma extensão da minha própria família;

Ao Prof. Dr. Jorge Chahine e a Profa. Dra. Fernanda Canduri, por terem participado da

minha banca de qualificação;

Aos Prof(a)s Dr(a)s. Beatriz Guimarães, João Alexandre Barbosa, Maria Célia

Bertolini e Valmir Fadel por terem aceitado o convite em participar da minha banca de

doutorado; E também aos Prof(a)s Dr(a)s Marcos Fontes, Maria Cristina Nonato e

Jorge Chahine por terem aceitado o convite de serem suplentes da minha banca

examinadora;

À Marisa, minha leal suportadora, que sempre me agüentou na sala de estudos. Pelos

momentos de risadas, tristezas e desabafos; Pela amizade e sinceridade que sempre teve

comigo;

Aos Profs. Drs. Diógenes Santiago Santos e Luiz Augusto Basso, pelo fornecimento de

proteínas;

Ao pessoal da PUC-Rio Grande do Sul, pela amizade e pela expressão das proteínas

fornecidas, em especial, Fernanda Ely, Igor, Jaím, Ana Luíza, Patrícia, e Jordana;

À grande amiga Márcia, que sempre esteve disposta a me ajudar e me escutar. E que fez

também o ambiente de trabalho ser mais gostoso e profissional;

À outra grande amiga Denise Mello, que também sempre fez a minha vida se tornar mais

alegre;

À Daiana e Angélica por todo o carinho, simpatia e confiança;

A todos os alunos do Grupo BMSys tanto os atuais como os que hoje não freqüentam

mais (Pepeu, Diego, Denis, Janaína, Nathália, Nelsinho, Nelsão, Henrique, Marisa,

Galo, Paulo, Alexandre, Hugo, Helen, Alessandra, Bona, Joane, Guilherme, Daiana,

Ana Helena, Karisa, José Renato, Cristiane, Mariane, Lisandra, Marcos Michel,

Brianna, Angélica, Lívia, Michele e Samantha), pelas gargalhadas, risadas, choros,

brigas e tudo mais que faz parte das novas vidas;

Aos grandes amigos da pós-graduação, por tornar a vida cotidiana menos monótona e

solitária (Priscila, Gisele, Magno, Flávio, Sabrina, Jorge, Eduardo, Sídney, Leandro,

Ronaldo, Ricardo, José Ésio, Aperreio, Cristina, Fernanda, André, Renato);

Ao Prof. Dr. Jorge Chahine, pelo empenho com a pós-graduação e pelo exemplo de

pessoa que é;

Aos demais professores do Departamento pelos ensinamentos transmitidos;

Aos funcionários técnico-admistrativos, que sempre estão correndo para nos ajudar com

os problemas menores, mas que fazem toda a diferença;

À Grande Rosemar, por toda a simpatia que é, e sempre correndo para agilizar as coisas

para os alunos na seção de pós-graduação. Aos demais funcionários da Seção de pós-

graduação por sempre terem contribuído aos alunos;

Aos professores da graduação, por ter me feito um biólogo e me dar à visão de mundo que

tenho hoje, em especial a Cláudia Carareto e Eleni, por quem tenho grande admiração;

Aos funcionários do Laboratório Síncrotron, pela ajuda nos equipamentos de raios X, em

especial ao Lucas,;

Aos amigos eternos da graduação, que mesmo distante no momento, eu sei que estarão

torcendo por mim, principalmente a Ana Paula, Marcela, Leandra, Aline, Amanda e

Renata e Gustavo;

A todos os demais amigos conquistados durante todos estes anos de IBILCE;

Aos amigos conquistados em Fortaleza, em especial ao Gustavo, Taianá e Vitor, por me

ajudarem quando precisei;

Aos amigos conquistados em São Paulo, pelos ensinamentos transmitidos em “virtual

screen”, em especial, a profa. Dra Antônia Tavares e ao Dr. Humberto;

À Andréia, pela amizade e carinho que tem por mim;

Aos meus amigos de Fernandópolis, que com certeza estarão torcendo por mim, não

importa onde eu ou eles estejam, principalmente ao Ricardo, Rosa, Vagner e Adriana,

Fernando e Miriam, Rogério e Elisângela;

À minha família, em especial a minha querida e amada mãe, do qual sem ela nada sou.

Que é o suporte para toda a minha vida;

À Fapesp pelo auxílio financeiro e concessão da bolsa de doutorado;

A todas as pessoas que de alguma forma contribuíram para a minha formação ou para

a realização deste trabalho, inclusive aquelas que por força do destino não tinham

simpatia por mim ou eu por elas, mas que contribuíram para o meu crescimento pessoal e

profissional;

E finalmente a Deus por minha existência;

“A maior recompensa do nosso trabalho

não é o que nos pagam por ele,

mas aquilo que ele nos transforma”

Joln Ruskin

“-Podes dizer-me, por favor, que caminho devo seguir para sair daqui?

-Isso depende muito de para onde queres ir – respondeu o gato.

-Preocupa-me pouco aonde ir – disse Alice.

-Nesse caso, pouco importa o caminho que sigas – replicou o gato.”

(Lewis Carrol – Alice no país das Maravilhas)

Estudo estrutural de proteínas alvo de Mycobacterium tuberculosis

Lista de Abreviações

μM

micro molar

ACP Acyl carrier protein - Proteína carreadora de acila

ADP Adenosina difosfato

AIDS Adquired immunodeficiency syndrome - Síndrome da Imunodeficiência Adquirida

ATP Adenosina trifosfato

BCG Bacilo Calmette Ghérin

C Carbono

CQ Chiquimato quinase

DAHP Deoxi-D-arabino-heptulosonato-7-fosfato

DHFR Enzima Dihidrofolato redutase

E4P Eritrose-4-fosfato

EPSP 5-enoil-piruvinil-chiquimato-3-fosfato

EPSPS 5-enoil-piruvinil-chiquimato-3-fosfato sintase

ESB Extended shikimate binding -domínio de ligação do chiquimato estendido

FAS I Sistema de síntese de ácidos graxos tipo I

FAS II Sistema de síntese de ácidos graxos tipo II

FMN

Flavina mononucleotídeo

G3P D-gliceraldeído-3-fosfato

HCl Ácido clorídrico

Hepes Ácido N-(2-hidroxietil)piperazina-N’-(2-etanosulfônico)

HIV Human Immunodeficiency virus (Vírus da Imunodeficiêrncia Humana)

IGP Indol glicerol-3-fosfato

INH Isoniazida

InhA Proteína Enoil ACP redutase

IPP

Indol propanol-3-fosfato

KatG Proteína catalase peroxidase

Constante de inibição

Íon Magnésio

MtCQ Chiquimato quinase de Mycobacterium tuberculosis

N Nitrogênio

Na Sódio

NAD Nicotinamida adenina difosfato

Estudo estrutural de proteínas alvo de Mycobacterium tuberculosis

NADH-INH Complexo de nicotinamida adenina dinucleotídeo e isoniazida

NADP Nicotinamida adenina dinucleotídeo fosfato

NB Nucleotide binding – referente ao domínio de ligação do nucleotídeo

NMP Mononucleotídeo fosfato

OMS Organização Mundial de Saúde

PDB Protein Data Bank – Banco de dados de proteína

PEG Polietileno glicol

PEP Fosfoenol piruvato

Pi Fosfato inorgânico

PLP Piridoxal fosfato

RC Reduced core – referente ao domínio central reduzido

SB Shikimate binding – referente ao domínio de ligação do chiquimato

SDR Short dehydrogenase/reductase – dehidorgenase/redutase de cadeia curta

SpCS Corismato sintase de Streptococcus pneumoniae

TIM Triose fosfato isomerase

Tris Tris(hidroximetil)aminometano

TRPS Triptofano sintase

RESUMO

A tuberculose representa hoje um dos principais problemas de saúde pública

mundial. É estimado que cerca de cem milhões de pessoas sejam infectadas anualmente.

Para este grande número de casos, dois fatores têm contribuído significativamente, a

resistência da Mycobacterium tuberculosis aos antibióticos existentes e o aumento de

casos de co-infecção com HIV. Por isso, é importante o desenvolvimento de novas

drogas contra o seu agente causador. Há duas abordagens principais para o

desenvolvimento de drogas. Primeiramente pode-se identificar e inibir proteínas que

estejam presentes em uma determinada classe de microorganismos, mas não sejam

conservadas em humanos, levando a antibióticos de largo espectro, ou ainda pode-se

identificar e inibir proteínas de um determinado microorganismo, levando a drogas

específicas. Exemplos da primeira abordagem são as enzimas da via do ácido

chiquímico e suas ramificações, como a via de síntese de triptofano. A via do ácido

chiquímico é responsável pela biossíntese de corismato, que é o precursor comum

utilizado na geração de vários compostos aromáticos importantes para sobrevivência da

bactéria, entre eles os aminoácidos e co-enzimas; e exemplos da segunda abordagem

são enzimas relacionadas com a síntese de ácidos micólicos, que são importantes

constituintes da parede celular de micobactérias. O objetivo do presente trabalho foi

realizar estudos estruturais de quatro proteínas que são alvos em potencial para o

desenvolvimento de drogas contra tuberculose, sendo elas: a Chiquimato quinase e

Corisamto sintase, ambas da via do ácido chiquímico; Triptofano sintase, última enzima

da via de síntese de triptofano; e a enzima dependente de NADH enoil ACP redutase

(InhA), relacionada com a síntese de ácidos micólicos. Neste trabalho são apresentadas

as estruturas da Chiquimato quinase de M. tuberculosis em complexo com ADP e

magnésio, na ausência do substrato ácido chiquímico, e da Chiquimato quinase em

complexo com ADP e ácido chiquímico, na ausência do íon magnésio. Nestas estruturas

pôde-se observar o efeito de íons e a influência do ácido chiquímico sobre a estrutura da

chiquimato quinase. Apresentamos também, a primeira estrutura da corismato sintase de

Mycobacterium tuberculosis, evidenciando suas características estruturais. Além disso,

foram realizados estudos estruturais com a proteína InhA na sua forma nativa,

mostrando o seu estado conformacional na ausência de ligantes, e três estruturas para

esta proteína em complexo com isoniazida ativa, sendo a proteína selvagem e dois

mutantes identificados clinicamente como resistentes a este medicamento (S94A e

I21V). Além desses estudos, foi realizada a modelagem molecular da proteína triptofano

sintase em complexo com seis inibidores, e feita análise da especificidade destes com

relação a esta proteína. Com a realização deste trabalho pretendemos estar contribuindo

para a solução do grande problema que a tuberculose representa hoje, por afligir

milhões de pessoas em todo o mundo.

ABSTRACT

Tuberculosis represents today a large problem of world public health. It is

estimated that approximately a hundred million people are infected annually. For this

large number of cases, two factors have contributed significantly, the resistance of

Mycobacterium tuberculosis to existent antibiotics and the increase of the cases of the

co-infection with HIV. Therefore, the development of new drugs against its etiologic

agent is important. There are two main approaches for the development of drugs.

Firstly, proteins of a determined class of microorganism, that are not present in humans,

can be identified and inhibited leading to large-spectra antibiotics; or it can identify and

inhibit proteins of a determined organism leading to specific drugs. Examples of the

first approach are enzymes of the shikimate pathway and its branches, such as the

tryptophan synthesis pathway. The shikimate pathway is responsible for biosynthesis of

chorismate, it is a common precursor utilized in the production of various important

aromatic compounds used in the survival of the bacteria, such as amino acids and co-

enzymes; an example of the second approach are enzymes related to the synthesis of

mycolic acids, they are important constituents of the cellular wall of mycobacteria. The

objective of the present work was to perform structural studies of four enzymes that are

potential targets to the development of drugs against tuberculosis, such as: shikimate

kinase and chorismate synthase, both of the shikimate pathway; tryptophan synthase,

last enzyme of the trytophan pathway; and the dependent of NADH enzyme enoyl ACP

reductase (InhA) related to the synthesis of mycolic acids. Here, the shikimate kinase

structures from M. tuberculosis in complex with ADP and magnesium ion, in the

absence of shikimate and the shikimate kinase in complex with ADP and shikimate, in

the absence of the magnesium ion, are presented. The effects of ions and shikimate on

the structure of shikimate kinase were observed. Here, we also present the first structure

of chorismate synthase from M. tuberculosis showing its structural characteristics.

Furthermore, we perform structural studies of the InhA protein in its native form

showing its conformational state in the absence of ligands and three structures for this

protein in complex with active isoniazid (wild type and two resistant clinical isolated

mutants to this medicine, S94A and I47T). We also performed the molecular modeling

of the tryptophan synthase protein in complex with six inhibitors, and the analysis of the

specificity of these in relation to the protein was performed. With the completion of this

work, we intend to contribute to the solution of the big program that tuberculosis

represents today causing suffering to the millions of people worldwide.

Estudo Estrutural de Enzimas Alvo de Mycobacterium tuberculosis

ÍNDICE DE FIGURAS

Figura 1 Mycobacterium tuberculosis corado pelo método ácido-resistente ..................... 3

Figura 2 Representação esquemática da via metabólica do ácido chiquímico ................. 5

Figura 3 Representação esquemática das ramificações da via do ácido chiquímico ....... 7

Figura 4 Reação catalisada pela chiquimato quinase ......................................................... 9

Figura 5 Representação esquemática da chiquimato quinase ........................................... 10

Figura 6 Domínios estruturais para a chiquimato quinase ................................................ 11

Figura 7 Alterações estruturais e efeito sinérgico observado na chiquimato quinase ..... 13

Figura 8 Reação catalisada pela corismato sintase ............................................................. 15

Figura 9 Representação esquemática da corismato sintase ........................ ...................... 15

Figura 10 Completo mecanismo catalítico para a corismato sintase .................................. 17

Figura 11 Representação esquemática da via de síntese de triptofano ............................... 20

Figura 12 Reação catalisada pela triptofano sintase ............................................................. 21

Figura 13 Estrutura tridimensional da triptofano sintase ................................................... 22

Figura 14 Formula molecular dos ligantes para triptofano sintase .................................... 24

Figura 15 Representação esquemática do envelope celular de micobactérias ................... 25

Figura 16 Reação catalisada pela enzima InhA .................................................................... 26

Figura 17 Representação esquemática da InhA .................................................................... 27

Figura 18 Mecanismo de ativação da isoniazida .................................................................. 28

Figura 19 Contatos moleculares entre o isonicotínico acil-NADH e o sítio ativo da InhA 29

Estudo estrutural de proteínas alvo de Mycobacterium tuberculosis

SUMÁRIO

1. INTRODUÇÃO

1.1 A tuberculose e a necessidade do desenvolvimento de novas drogas

1.2 A via do ácido chiquímico

1.3 A chiquimato quinase

1.4 A corismato sintase

1.5 A triptofano sintase

1.6 InhA e o mecanismo de resistência ao antibiótico isoniazida

2. OBJETIVOS

3. RESULTADOS

3.1 Crystallization and preliminary X-ray crystallographic analysis of

chorismate synthase from Mycobacterium tuberculosis

3.2 Structure of chorismate synthase from Mycobacterium tuberculosis 36

3.3 Molecular models of tryptophan synthase from Mycobacterium tuberculosis

complexed with inhibitors

3.4 Effects of the magnesium and chloride ions and shikimate on the structure

of shikimate kinase from Mycobacterium tuberculosis

Estudo estrutural de proteínas alvo de Mycobacterium tuberculosis

3.5 Crystallographic studies on the binding of isonicotinyl-NAD adduct to wild-

type and isoniazid resistant 2-trans- Enoyl-ACP (CoA) Reductase from

Mycobacterium tuberculosis

4. CONCLUSÕES

118

5. BIBLIOGRAFIA

121

6. ANEXO – Artigo de revisão publicado

126

6.1 Chorismate Synthase: an attractive target for drug development against

neglected disease

127

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

1. INTRODUÇÃO

1.1 A tuberculose e a necessidade do desenvolvimento de novas drogas

As doenças infecciosas é a quarta maior causa de morte da população humana

(KING et al., 2006), sendo responsável pelo sofrimento e morte de centenas de milhões de

pessoas, especialmente em áreas tropicais e subtropicais do mundo, local onde ocorre cerca

de 90% das mortes causadas por esse tipo de enfermidade (TROUILLER et al., 2001).

Segundo a Organização Mundial de Saúde (OMS) cerca de 14 milhões de pessoas morrem

a cada ano devido a desse tipo de doença, destacando-se entre elas AIDS/HIV, infecções

respiratórias, malária e tuberculose, que lideram a causa de morte e de morbidade,

principalmente nos países da África, Ásia e América do Sul – regiões nas quais estão

presentes oitenta por cento de toda a população mundial (TROUILLER et al., 2001).

A organização mundial de saúde identifica três fatores chaves que podem contribuir

coletivamente para a pandemia associada às doenças infecciosas. São eles: falência no uso

de ferramentas existentes no controle de tais doenças, ferramentas inadequadas ou

inexistentes, e conhecimento insuficiente sobre a doença (TROUILLER et al., 2002).

Dentre essas doenças, a tuberculose, hoje, é considerada a causa infecciosa líder de

mortes. É estimado que cem milhões de pessoas são infectadas anualmente. De cem

milhões, aproximadamente dez milhões desenvolvem a doença e três milhões morrem

(MORGAN; HARITAKUL; KELLER, 2003).

A incidência de tuberculose teve um declínio rápido no início do século vinte nos

países em desenvolvimento, devido à melhora nas condições sanitárias e de moradias. Essa

tendência foi, inicialmente, acelerada pela introdução da vacinação BCG (1927) e da

descoberta de antibióticos como a estreptomicina (1944) e, posteriormente, com a

descoberta do ácido p-aminosalisílico (1946), isoniazida (1952) e rifampicina (1965)

(DUNCAN, 2003). Mas está ocorrendo uma ressurgência dos casos de tuberculose nos

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

países em desenvolvimento. Muitos são os fatores envolvidos, porém, o mais notável é a

falta de recursos e o descaso governamental que impedem a implementação adequada de

medidas de controle. Além disso, a incidência aumentou também nos países desenvolvidos,

principalmente devido à imigração e à negligência direta do mundo todo com relação ao

problema (RAVIGLIONE et al., 1997). Todavia, dois recentes eventos têm contribuído

significativamente para a pandemia de tuberculose: a) o aumento de bactérias multi-

resistentes devido a terapias inadequadas e o uso indiscriminado de antibióticos,

principalmente a isoniazida e rifampicina, as duas drogas mais utilizadas no tratamento

quimioterápico (BAPTISTA et al., 2002; DUNCAN, 2003); e b) a AIDS, pois pessoas

portadoras de ambos os agentes causadores da tuberculose e AIDS têm o sistema imune

debilitado, e, portanto, maiores chances de desenvolver a doença (BAPTISTA et al., 2002;

DYE et al., 2002). Em 1993, a Organização Mundial de Saúde declarou a tuberculose

como uma emergência global. Entretanto, a incidência exata desta doença no panorama

mundial não é bem conhecida (RAVIGLIONE et al., 1997).

O principal responsável pela tuberculose em humanos é a bactéria Mycobacterium

tuberculosis. Este organismo é um bastonete fino, com comprimento médio de 4 μm, como

pode ser visto pela coloração por métodos ácido-resistentes (Fiehl-Neelsen) (Figura 1) ou

fluorescentes (COTRAN et al., 1991). Apresenta crescimento lento, dormência e complexo

envelope celular. É um organismo obrigatoriamente aeróbico e é um parasita intracelular

facultativo, usualmente de macrófagos. O tempo de geração da M. tuberculosis em animais

infectados é de aproximadamente vinte e quatro horas. Isto contribui para a característica

crônica da doença (apesar da doença em si não apresentar uma fase crônica),

impossibilitando fortemente o regime de tratamento e representando um formidável

obstáculo para pesquisas (COLE et al., 1998). M. tuberculosis é quase exclusivamente um

parasita humano e é transmitido pelo ar. Usualmente causa tuberculose pulmonar, uma

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

infecção dos pulmões, embora, seja capaz de infectar outros tecidos ou órgãos do corpo

(PELCZAR et al., 1996).

Figura 1. Mycobacterium tuberculosis corado

pelo método ácido-resistente (Fiehl-Neelsen).

Clinicamente a tuberculose pode ser classificada como primária e secundária.

Sendo que a tuberculose primária é caracterizada por uma forma de doença que se

desenvolve previamente numa pessoa não-exposta e, portanto, não-sensibilizada. A

tuberculose primária inicia-se pela inalação do M. tuberculosis e culmina com o

desenvolvimento da imunidade mediada por células no organismo. Na maioria das pessoas,

a tuberculose primária é assintomática, ainda que possa causar febre e derrame pleural. Das

pessoas recém-infectadas somente 5% desenvolvem a doença em uma forma clinicamente

significante. A tuberculose secundária é um padrão de doença que observa em um

hospedeiro previamente sensibilizado. Pode surgir pouco depois da tuberculose primária,

porém mais frequentemente, ela surge da reativação das lesões primárias latentes muitas

décadas após a infecção inicial, particularmente quando a resistência do hospedeiro estiver

enfraquecida. No caso da tuberculose pulmonar secundária, ela é geralmente localizada no

ápice dos lóbulos superiores de um ou de ambos os pulmões. A tuberculose secundária

localizada pode ser assintomática. Quando as manifestações aparecem, elas são,

geralmente, insidiosas no início. Os sintomas sistêmicos aparecem, com freqüência no

início do curso e incluem indisposição, anorexia, perda de peso e febre. Com o

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

envolvimento pulmonar progressivo, surgem quantidades cada vezes maiores de escarro,

primeiramente mucóide e posteriormente purulento. As manifestações extra pulmonares da

tuberculose são em grande número e dependem do sistema orgânico envolvido (KUMAR,

ABBAS, FASTO, 1999)

Com a determinação completa do genoma da M. tuberculosis ocorreu um profundo

avanço em pesquisas sobre tuberculose (COLE et al., 1998). Novas abordagens e técnicas

têm sido utilizadas para avaliar as informações obtidas pelo projeto genoma deste

organismo. Entre elas, inclui a utilização de ferramentas de bioinformática, proteômica,

transcriptoma e genômica estrutural combinadas com ferramentas de manipulação de

genoma. Com a utilização destas técnicas tem-se permitido um melhor entendimento da

complexa biologia deste patógeno (DUNCAN, 2003). Além disso, a determinação do

genoma e o melhor esclarecimento dos mecanismos de virulência deste organismo,

permitem a possibilidade da descoberta de novos alvos para o desenvolvimento de drogas

(COLE et al., 1998; DUNCAN, 2003; KHASNOBIS; ESCUYER; CHATTERJEE, 2002).

Para a identificação de alvos com potencial de gerar novas drogas contra M.

tuberculosis pode se utilizar duas abordagens significativas. A primeira, em que se podem

isolar proteínas que são conservadas entre diferentes espécies de bactérias, mas que não

conservadas em humanos, conduzindo aos antibióticos de largo espectro; e a segunda na

qual se podem identificar alvos que são únicos em M. tuberculosis, levando a obtenção de

drogas anti-tuberculose específicas (KHASNOBIS; ESCUYER; CHATTERJEE, 2002).

Exemplos da primeira abordagem são as enzimas que compõe a via de síntese de

ácido corísmico (via do ácido chiquímico) e da segunda abordagem são as enzimas

responsáveis pela síntese de ácidos micólicos.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

1.2 A Via do Ácido Chiquímico

A via do ácido chiquímico (Figura 2) é essencial para algas, plantas, bactérias,

fungos (BENTLEY, 1990) e parasitas do filo apicomplexa, como Plasmodium,

Toxoplasma e Cryptosporidium (ROBERTS et al., 1998), e não está presente em

mamíferos.

Figura 2. Representação esquemática da via metabólica do ácido chiquímico.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

A via do ácido chiquímico é definida como os sete passos metabólicos iniciais da

condensação de fosfoenol piruvato (PEP) e eritrose-4-fosfato (E4P) finalizando com a

síntese de corismato. Os sete passos desta via foram originalmente descobertos através de

estudos em bactéria, principalmente Escherichia coli e Salmonella typhimurium

(HERRMANN; WEARVER, 1999).

A via do ácido chiquímico foi inicialmente descrita como a rota biossintética para a

produção de aminoácidos aromáticos (Fenilalanina, Tirosina e Triptofano) através dos

estudos clássicos de Bernhard Daves e David Sprinson e seus colaboradores por volta de

1950. A identificação de seus vários intermediários foi completada no início de 1960,

sendo que o primeiro deles a ser identificado foi o ácido chiquímico (PITTARD, 1987;

HERMANN, 1995). Em microorganismos, esta via, além de ser utilizada para síntese de

aminoácidos aromáticos, também faz parte da síntese de folato, quinonas e alguns

metabólitos secundários, como sideróforos e micobactinas. A via do ácido chiquímico

ramifica-se em muitos pontos, porém, corismato é o último ramo comum da via para os

compostos citados acima. Corismato é convertido por cinco enzimas distintas para

prefenato, antranilato, aminodeoxicorismato, isocorismato e p-hidroxibenzoato. Esses

metabólitos compreendem os primeiros intermediários na biossíntese de fenilalanina e

tirosina, triptofano, folato, menaquinonas e sideróforos, e ubiquinonas, respectivamente

(Figura 3) (DOSSELAERE; VANDERLEYDEN, 2001). Sendo que destes intermediários,

folato é um composto que participa de várias formas na síntese de purinas, timidilato,

metionina, glicina e pantotenato; Sideróforos são moléculas responsáveis pela captação de

ferro por bactérias; Quinonas são moléculas lipofílicas, componentes não protéicos da

cadeia transportadora de elétrons ligados à membrana que podem ser divididas em dois

grandes grupos estruturais: benzoquinonas e naftoquinonas. As benzoquinonas são

denominadas de ubiquinonas e as naftoquinonas são chamadas de menaquinonas

(DOSSELAERRE; VANDERLEYDEN, 2001).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figura 3. Representação esquemática das ramificações da via do ácido chiquímico.

(1) corismato piruvato-liase; (2) isocorismato sintase; (3) aminodeoxicorismato

sintase; (4) antranilato sintase; (5) corismato mutase.

A análise completa da seqüência do genoma da M. tuberculosis revelou a presença

dos sete genes aro envolvidos na biossíntese de corismato. São eles, aro D, B, K, F, G, E e

A. Estes genes são responsáveis pela síntese das enzimas 3-dehidroquinato dehidratase, 3-

dehidroquinato sintase, chiquimato quinase, corismato sintase, DAHP sintase, chiquimato-

5-dehidrogenase e EPSP sintase, respectivamente (Figura 2) (COLE et al., 1998; PARISH;

STOKER, 2002).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

A importância desta via, tanto para M. tuberculosis como para outros patógenos foi

justificada por diversos trabalhos: Parish e Stoker (2002) demonstraram que a via do ácido

chiquímico é essencial para M. tuberculosis, pois, eles foram incapazes de isolar cepas

contendo o gene aroK (responsável pela produção da chiquimato quinase) inativado. A

inativação do gene aroD (responsável pela produção de 3-dehidroquinato dehidratase)

também é usada com sucesso para gerar cepas atenuadas para a vacina oral contra

Salmonella typhi e outras bactérias (TACKET et al., 1997). Os estudos de inibição

utilizando glifosato demonstram que este herbicida inibe a EPSP sintase, e apresenta uma

significativa inibição no crescimento de T. gondii, P. falciparum e C. parvum (ROBERTS

et al., 2002).

1.3 A Chiquimato quinase

A Chiquimato quinase, principalmente de Mycobacterium tuberculosis (peso

molecular de 18583,3 Da), é uma enzima que teve sua estrutura intensivamente estudada a

partir de 2002 (GU et al., 2002, DHALIWAL et al., 2004, PEREIRA et al., 2004, GAN et

al., 2006, HARTMANN et al., 2006). Esta proteína, que é, a quinta enzima da via do

ácido chiquímico, catalisa a fosforilação do grupo 3-hidroxil do ácido chiquímico, usando

ATP como um co-substrato (Figura 4) (HERRMANN, 1995) e os íons magnésio e cloreto

como co-fatores. Em Escherichia coli, esta reação é catalisada por duas diferentes

isoformas, sendo elas a CQ I, codificada pelo gene aroK e a CQ II codificada pelo gene

aroL. Estas proteínas são monoméricas, possuem uma massa molecular de

aproximadamente 19.500 Dáltons e apresentam uma identidade de 30% entre si. Ambas as

enzimas possuem atividade in vitro na biossíntese de aminoácidos aromáticos. Mas in

vivo, a isoenzima II não tem uma atividade exclusiva da via do ácido chiquímico, pois

parece apresentar também uma função na divisão celular (HERRMANN; WEAVER,

1999; 1995; KRELL; COGGINS; LAPTHORN, 1998). Diferente de E. coli, o genoma

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

completo de algumas bactérias, como Mycobacterium tuberculosis e Haemophilus

influenzae, tem revelado a presença de apenas um gene para chiquimato quinase. Muitas

destas chiquimato quinases parecem ser codificadas pelo gene aroK, pois sua seqüência de

aminoácidos tem alta identidade com a CQ I de E. coli (GU et al., 2002).

Chiquimato 3-fosfato

Ácido chiquímico

Figura 4. Reação catalisada pela chiquimato quinase.

A primeira estrutura cristalográfica da chiquimato quinase de M. tuberculosis

(MtCQ) foi resolvida na presença de Mg-ADP (GU et al., 2002). A MtCQ é uma proteína

da classe α/β que apresenta na sua estrutura fitas β rodeadas por hélices α. No centro da

proteína estão localizadas cinco fitas β paralelas (β1-β5) formando uma folha β na ordem

23145, que é flanqueada por oito hélices (Figura 5) (GU et al., 2002).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

β5

β4

β3

β2

β1

Figura 5. Represenção de elementos de estrutura secundária da MtCQ em complexo com Magnésio

(Mg

), Adenosina difosfato (ADP) e ácido chiquímico (PEREIRA et al., 2004) gerada pelo programa

MolMol (KURADI; BILLETER; WÜTHRICH, 1996).

A ordem da folha β 23145 observada na estrutura da MtCQ classifica esta proteína

como pertencente à família das monofosfato nucleosídeo (NMP) quinases (YAN; YSAI,

1999). As proteínas desta família são compostas por três domínios: central (local de

ligação do fosfato), domínio da tampa (Lid domain) e o domínio de ligação do NMP ou SB

(shikimate binding). A MtCQ possui uma seqüência de aminoácidos conservada,

GXXXXGKT/S, conhecida como motivo Walker A, que forma o P-loop (SCHULZ,

1992). Esta região liga-se ao β-fosfato presente no ADP e está localizada entre a β1 e α1.

As NMP quinases sofrem grande mudança conformacional durante a sua atividade

catalítica. As regiões responsáveis por este movimento estão localizadas no sítio de ligação

do NMP e no Lid domain (MÜLLER et al., 1996; GERSTEIN; SCHULZ; CHOTHIA,

1993).

Hartmann et al. (2006) baseado na análise de movimentos globais acompanhados

no processo de ligação de substratos no sítio ativo da chiquimato quinase, reclassificou os

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

domínios da chiquimato quinase. Segundo sua classificação, temos agora quatro domínios,

que são: o ESB (extended SB) que inclui o subdomínio SB; o NB (nucleotide-binding) que

inclui o P-loop e o AB-loop ( adenine binding loop) e um curto seguimento da hélice α 6; o

LID; e a parte remanescente que forma o RC (reduced core) (Figura 6). O antigo domínio

central compreende os presentes domínios RC e NB e um curto seguimento do domínio

ESB.

Figura 6. Domínios estruturais para a Chiquimato quinase proposto por

Hartmann et al. (2006). NB: nucleotide-binding, RC: reduced core, ESB:

extended Shikimate binding, LID: Lid domain

Apesar dos estudos cristalográficos feitos com as chiquimato quinases de M.

tuberculosis (GU et al., 2002), E. coli (ROMANOWSKI; BURLEY, 2002) e Erwinia

chrysanthemi (KRELL; COGGINS; LAPTHORN, 1998) a posição precisa do ácido

chiquímico só foi determina em 2004 por Pereira et al. (2004) e Dhaliwal et al. (2004). A

determinação da estrutura da chiquimato quinase em complexo com ácido chiquímico

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

revelou os resíduos do sítio ativo que participam das interações entre este substrato e a

enzima. Pela análise da estrutura deste complexo puderam-se observar também as

mudanças conformacionais ocasionadas pela presença do substrato em seu sítio ativo. As

principais alterações conformacionais ocorrem no domínio ESB e no Lid domain, na qual

ambos movem-se em direção ao centro da enzima causando um fechamento parcial do sítio

ativo da enzima (PEREIRA et al., 2004; DHALIWAL et al., 2004).

Entretanto, muito recentemente Gan et al. (2006) e Hartmann et al. (2006)

mostraram através de dados estruturais que a molécula de nucleotídeo é o principal

responsável pelo fechamento do Lid domain sobre o sítio ativo da MtCQ. Hartmann et al.

(2006) mostrou que ocorre um sinergismo de ligação dos substratos no sítio ativo desta

enzima. Segundo estes autores, ocorrem um efeito sinérgico aleatório de ligação tanto para

o ácido chiquímico como para o nucleotídeo. Esse efeito sinérgico pode ser consistente

com o possível aumento da afinidade para ambos os substratos, independente da ordem de

ligação. A Figura 7 mostra o efeito sinérgico de ambos os substratos e suas influências na

estrutura da chiquimato quinase de Mycobacterium tuberculosis proposto por Hartmann et

al. (2006).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figura 7. Alterações conformacionais observadas na Chiquimato quinase de M.

tuberculosis na presença de seus ligantes e o possível efeito sinérgico desses ligantes

segundo Hartmann et al. (2006). SKM: ácido chiquímico; ATP: adenosina

trifosfato

Até muito recentemente, não se conhecia a conformação da estrutura da chiquimato

quinase na ausência de qualquer ligante. Entretanto Gan et al. (2006) e Hartmann et al.

(2006) determinaram a estrutura desta enzima no seu estado nativo que possibilita intuir os

movimentos causados pela molécula de ADP. Esses autores também determinaram a

estrutura da chiquimato quinase em complexo com análogo de ATP (GAN et al., 2006) e

ATP (HARTMANN et al., 2006) e, desta forma, pode-se acrescentar informações sobre o

mecanismo de transferência do fosfato da molécula de ATP para a molécula de ácido

chiquímico.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Com relação ao íon cloreto que interage com a chiquimato quinase, foi observada a

influência deste sobre a proteína apenas por técnicas de baixa resolução (CERASOLI et al.,

2003). Assim, por meio de dicroísmo circular e de cinética enzimática foi observado que a

presença do íon cloreto enfraquece a interação entre o ácido chiquímico e a chiquimato

quinase de Erwinia Chrysanthemi, mas, por outro lado, fortalece a afinidade da enzima por

ADP ou ATP (CERASOLI et al., 2003). Entretanto, apesar deste estudo, pouco se sabe

sobre o papel deste íon e do íon magnésio sobre a estrutura tridimensional desta proteína.

1.4 A Corismato Sintase

Outra enzima da via do ácido chiquímico, que a partir de 2004 se tornou alvo de

muitas pesquisas na área de biologia estrutural, é a corismato sintase (AHN et al., 2004,

VIOLA; SARIDAKIS; CHRISTENDAT et al., 2004, MACLEAN; ALI et al., 2003,

QUEVILLON-CHERUEL et al., 2004, DIAS et al., 2006). Esta proteína é a última enzima

da via do ácido chiquímico e realiza a conversão de 5-enoilpiruvil-3-chiquimato fosfato

(EPSP) em corismato, através de uma reação de eliminação de um grupo fosfato da

molécula de EPSP (Figura 8) (KITZING; MACHEROUX; AMRHEIN, 2001). A

eliminação desse grupo fosfato é responsável pela adição da segunda dupla ligação na

molécula de corismato (PITTARD, 1987). A corismato sintase requer para sua reação a

presença de flavina mononucleotídeo (FMN), embora a reação não envolva um estado de

mudança redox (HERRMANN, 1995). A eliminação do grupo fosfato é realizada por um

mecanismo não usual, através de uma 1,4-anti-eliminação. Até agora não foi descrita

nenhuma outra enzima que catalisa uma reação deste tipo. Sabe-se também, que na reação

está envolvido um intermediário instável de vida curta, formado depois da perda do fosfato

da molécula de EPSP. Estudos recentes têm dado evidência de que esse intermediário é um

radical, formado pela transferência de um elétron da molécula de FMN para o EPSP

(MACLEAN; ALI, 2004; KITZING, 2004).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figura 8. Reação catalisada pela corismato sintase.

A reação catalisada pela corismato sintase ocorre em duas etapas, em que,

primeiramente ocorre a clivagem da ligação C3-O3 e o lançamento do fosfato, produzindo

o intermediário instável, que colapsa para formar o corismato no segundo passo devido à

clivagem da ligação 6proR-C-H. A orientação da molécula de FMN na estrutura da

corismato sintase sugere que ela pode promover a estabilização de qualquer intermediário

deficiente em elétron da molécula do EPSP (Figura 9) (MACLEAN; ALI, 2004).

Figura 9. Representação de elementos de estrutura secundária da corismato sintase de

Streptococcus pneunoniae complexada com 5-enoilpiruvil-chiquimato 3-fosfato (EPSP)

e flavina mononucleotídeo (FMN) (MACLEAN; ALI 2004), gerada pelo programa

MolMol (KURADI; BILLETER; WÜTHRICH, 1996).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Corismato sintases de diferentes organismos podem ser classificadas de acordo com

sua capacidade em reduzir a molécula de flavina mononucleotídeo. Desta forma, temos as

corismato sintases monofuncionais e as bifuncionais. Corismato sintases de fungos

apresentam uma segunda função, além da conversão de EPSP em corismato. Elas são

também capazes de realizar a redução da molécula de FMN utilizando NADH. Assim,

estas corismato sintases são classificadas como bifuncionais, por apresentarem também

uma função intrínseca de flavina redutase (MACHEROUX et al., 1999). Por outro lado, as

corismato sintases de plantas, bactérias e parasitas do filo apicomplexa não possuem essa

capacidade de redução de flavina mononucleotídeo, tendo que obter esta molécula na sua

forma reduzida do meio e, portanto, são classificadas como monofuncionais

(MACHEROUX et al., 1999).

Apesar da estrutura cristalográfica da corismato sintase de vários organismos terem

sido resolvidas (AHN et al., 2004, VIOLA; SARIDAKIS; CHRISTENDAT et al., 2004,

MACLEAN; ALI et al., 2003, QUEVILLON-CHERUEL et al., 2004), somente a estrutura

de Streptococcus pneumoniae (SpCS) apresenta ambas moléculas de flavina

mononucleotídeo (FMN) e o substrato 5-enoilpiruvil-3-chiquimato fosfato (EPSP) ligadas

no seu sítio ativo (MACLEAN, ALI, 2003). Desta forma, esta estrutura revelou muitas

informações sobre o complexo mecanismo catalítico da corismato sintase. Baseado nesta

estrutura e na análise de proteínas mutantes de Neurospora Crasa, nas quais dois

importantes resíduos de histidina do sítio ativo foram mutados para alanina, pôde-se propor

um mecanismo mais completo para a reação catalisada por esta enzima. Segundo este

mecanismo, após a ligação da molécula de EPSP no sítio ativo da enzima com a molécula

de FMN reduzida já ligada, um elétron é transferido para a dupla ligação do substrato,

iniciando desta maneira, a clivagem e o lançamento do fosfato, com um resíduo de

histidina atuando como um neutralizador da carga incipiente do átomo de oxigênio (Figura

10). O resultante carbono neutro C(4a) da flavina semiquinona tautomeriza-se para uma

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

espécie radial na qual o elétron não pareado reside sobre o átomo N(5), que

concomitantemente causa a abstração do hidrogênio ligado ao C(6). E finalmente, ocorre a

deprotonação da flavina reduzida restaurando o seu estado inicial (Figura 10) (KITZING et

al., 2004).

Figura 10. Mecanismo catalítico completo para a Corismato Sintase proposto

por Kitzing et al. (2004). Rib: Ribose

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Apesar das estruturas cristalográficas da corismato sintase de Aquiflex aeolicus

(VIOLA; SARIDAKIS; CHRISTENDAT et al., 2004) e S. cerevisiae (QUEVILLON-

CHERUEL et al., 2004) estarem na forma nativa, grande parte de suas estruturas não foi

resolvida, devido à ausência de densidade eletrônica, principalmente alguns importantes

resíduos do sítio ativo. Assim, a conformação desses no sítio ativo da corismato sintase na

ausência de qualquer ligante é desconhecida.

A estrutura da corismato sintase apresenta-se como um tetrâmero, onde cada

monômero possui um único domínio central, que é rodeado por loops e trechos discretos

de hélices α e folhas β. O centro do monômero consiste de uma camada de quatro longas

hélices α, sanduíche entre um par de 4 fitas de folhas β antiparalelas (MACLEAN; ALI,

2003) (Figura 9). O bolsão de ligação apresenta uma parte hidrofóbica, na qual está

localizado o sistema de anéis isoaloxazino da molécula de FMN e uma outra parte

hidrofílica carregada positivamente devido a grande quantidade de argininas e histidinas.

Nesta segunda parte do sítio, está localizado a porção remanescente da molécula de FMN e

o EPSP, na qual estas duas moléculas apresentam-se justapostas a uma distância

aproximada de 3,0 Å, favorecendo, assim, a transferência do elétron entre a molécula de

EPSP e FMN (MACLEAN; ALI, 2004).

1.5 A Triptofano Sintase

As ramificações da via do ácido chiquímico também representam alvos

interessantes para o desenvolvimento de drogas de amplo espectro e vacinas para doenças

causadas por microrganismos, pelo fato de também estarem ausentes em mamíferos.

Exemplos clássicos, é a inibição da dihidrofolato redutase (DHFR), terceira enzima da via

de síntese do ácido p-aminobenzóico, por trimetoprim (HAWSER; LOCIURO; ISLAM,

2006) e a inibição da dihidropteroato sintase, quarta enzima da mesma via, pelos

antibióticos da série das sulfonamidas (SKÖLD, 2001). Outros exemplos de alvo são as

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

enzimas da via de síntese de triptofano, como demonstrado por Smith (2001), em que a

inativação do gene trpD, responsável pela síntese da proteína antranilato

fosforibosiltransferase, produz cepas atenuadas de M. tuberculosis. A inativação do gene

trpA (corresponde a subunidade I da antranilato sintase) também foi demonstrado gerar

cepas atenuadas de Bordetella bronchiseptica e Leptospira meyeri (McARTHUR et al.,

2003; BAUBY; GIRONS; PICARDEAU, 2003). A enzima triptofano sintase demonstrou

ser alvo para o desenvolvimento de herbicidas, pois testes in vitro com inibidores análogos

de substrato demonstraram ter atividade inibitória para esta enzima (FINN et al., 1999).

Desta forma a via de síntese de triptofano é bastante estudada no desenvolvimento

de terapias contra doenças infecciosas. Esta via é composta por cinco enzimas, sendo elas,

a antranilato sintase, antranilato fosforibosiltransferase, fosforibosil-antranilato isomerase,

indol-glicerol-3-fosfato sintase e a triptofano sintase (Figura 11).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figura 11. Via de síntese do aminoácido triptofano

Dentre estas enzimas, uma das mais bem estudadas é a triptofano sintase, sendo

muito bem conhecido os detalhes de sua purificação, bioquímica, inibição e modificações

estruturais de sua atividade (PAN; WOEHL; DUNN, 1997).

A triptofano sintase é responsável pela catálise das duas últimas reações da

biossíntese de L-triptofano. Esta proteína é uma enzima bifuncional com duas subunidades,

denominadas α e β. A subunidade α catalisa uma reação retroaldol em que o indol-3-

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

glicerol fosfato (IGP) é clivado em indol e D-gliceraldeído-3-fosfato (G3P). O indol

produzido pela subunidade α reage com L-serina ativada por uma reação dependente de

piridoxal fosfato (PLP) no sítio ativo da subunidade β para formar L-triptofano e água

(Figura 12A, B).

Indol

Figura 12. Reação catalisada pela triptofano sintase. A) Reação catalisada pela

subunidade α. B) Reação catalisada pela subunidade β. IGP: Indol glicerol 3-fosfato;

G3P: Glicerol fosfato; L-Ser: L-Serina; Enz: Enzima; E(Ain): Almidina interna; E(A-

)

: Aminoacrilato

;

: Es

écie

uinóide

;

L-tr

: L-tri

tofano.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

A subunidade α da triptofano sintase apresenta um enovelamento similar a da

proteína triose fosfato isomerase (TIM) (BANNER et al., 1975), apresentando uma

estrutura α/β barril (MILES, 1995). Entretanto, a subunidade β apresenta dois domínios de

aproximadamente mesmo tamanho, sendo eles denominados N-domínio e C-domínio

(HYDE et al, 1988). O sítio ativo da subunidade α é localizado próximo à interface com a

subunidade β em uma depressão que permite interações de hidrogênio com o grupo fosfato

do substrato IGP e um ambiente hidrofóbico para a sua porção indol (WEYAND, M.;

SCHILICHTING, 1999). O sítio ativo da subunidade β apresenta-se enterrado no centro

desta subunidade entre os N- e C- domínios. A molécula de PLP ligada ao sítio ativo desta

subunidade forma uma ligação covalente tipo C=N (base de Schiff) com um grupo amino

de um resíduo de lisina (Figura 12) (HYDE et al, 1988).

Figura 13. Estrutura tridimensional da Triptofano sintase de Salmonella typhimurium

em complexo com Indol glicerol 3-fosfato (IGP) na subunidade α e piridoxal fosfato

(PLP) na subunidade β (HYDE et. al, 1988).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

A estrutura tridimensional da triptofano sintase de Salmonella typhimurium foi a

primeira a ser resolvida e revelou a presença de um longo canal hidrofóbico que conecta os

sítios ativos da subunidade α e da subunidade β. Este canal teria o papel de impedir a

difusão do substrato e do produto da reação catalisada pela proteína, e também de estar

envolvido na regulação alostérica que sincroniza as reações nas duas subunidades

(SACHPATZIDIS et al., 1999). Após a determinação da estrutura nativa e de outras

estruturas com diferentes ligantes para S. typhimurium, pode-se observar que ligantes

induzem mudanças conformacionais que são importantes para o posicionamento do indol

dentro do canal e também para a comunicação alostérica entre as duas subunidades.

Drásticas mudanças conformacionais ocorrem com a adição de ligantes, tanto na

subunidade α como na subunidade β. Estas mudanças incluem ordenação de loops

próximos ao sítio ativo da subunidade α e movimentos de subdomínios da subunidade β

(MILES, RHEE, DAVIES, 1999).

Entre os ligantes estudados que se ligam na triptofano sintase, está o indol-3-

propanol fosfato (IPP) (WEYAND; SCHLICHTING, 1999) e cinco análogos do estado de

transição ariltioalquil fosfonados, que se ligam e inibem a subunidade α da triptofano

sintase (Figura 14) (SACHPATZIDIS et al., 1999). O IPP é um análogo de IGP que inibe a

subunidade α com um K

de 15 μM e foi o ponto inicial para a estratégia de desenho dos

inibidores ariltioalquil fosfonados (FINN et al., 1999). Estes inibidores foram desenhados

de forma a mimetizar o estado de transição formado durante a reação α da enzima e

possuem uma afinidade maior do que o substrato natural IGP e o análogo IPP. Estes

inibidores são derivados ariltioalquilfosfonados orto-substituídos que apresenta um átomo

de enxofre substituindo C2 (FINN et al., 1999; SACHPATZIDIS et al., 1999). Entretanto,

somente foram realizados estudos sobre o efeito inibitório dos inibidores ariltialquil

fosfonados sobre a triptofano sintase de Arabidopsis thaliana e S. typhimurium.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figura 14. Fórmula molecular dos

ligantes de TRPS. A) inibidores

ariltioalquil fosfonados. 1) ácido 4-(2-

hidroxifeniltio)-1-butenilfosfônico, 2)

ácido 4-(2-

hidroxifeniltio)butenilfosfônico, 3) ácido

4-(2-aminofeniltio)-butilfosfônico, 4)

ácido 4-(2-hidroxi-5-fluorofeniltio)-

butilfosfônico e 5) ácido 4-(2-

hidroxifenilsulfinil)butilfosfônico.

B) IGP (Indol-3-glicerol fosfato e IPP

(Indol-3-propanol fosfato).

IPP

IGP

A B

1.6 InhA e o mecanismo de resistência ao antibiótico isoniazida

Apesar das enzimas comentadas anteriormente serem alvos em potencial para o

desenvolvimento de novos compostos que tenham atividade inibitória e, portanto,

possivelmente medicamentosas, nenhum composto foi desenvolvido especificadamente

para estas enzimas de Mycobacterium tuberculosis. Entretanto, uma das enzimas mais

estudadas para o desenvolvimento de drogas em M. tuberculosis, e que é alvo da

isoniazida, é a enzima dependente de NADH, 2-trans enoil-ACP (acyl carrier protein)

redutase (InhA). Esta enzima exibe alta especificidade para ácidos graxos de cadeia longa

> C

) com grupo enoil tioester, que é consistente com o envolvimento desta enzima na

biossíntese de ácidos micólicos (QUÉRMAD et al. 1995). Ácidos micólicos, que por sua

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

vez, são ácidos graxos α-alquil-β-hidroxil, são os principais componentes da parede celular

de micobactérias (BRENNAN; NIKAIDO, 1995). Esta parede celular ou envelope celular

é uma característica única das micobactérias e apresenta um alto conteúdo de lipídios

constituindo uma barreira composta de ácidos micólicos ancorados a moléculas de

arabinogalactano, ligados ao peptídioglicano da membrana celular da bactéria (Figura 15)

(MOLLE et al., 2006). Essa parede é um grande obstáculo para a penetração de drogas na

célula da micobactéria, representando assim um grande desafio para o desenvolvimento de

novas drogas contra tuberculose.

Figura 15. Representação esquemática do envelope celular de micobactérias.

bicamada lipídica, peptídioglicano, arabinogalactano, micolato, acil

lipídeos,

lipoarabinomanano (LAM), porfirina

Dois tipos de sistemas de síntese de ácidos graxos (FAS) são conhecidos. FAS-I,

encontrado em vertebrados e está relacionado com a síntese de novo de ácidos graxos; e

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

FAS-II, que é usualmente encontrado em bactérias, e está relacionado com a elongação de

cadeias de ácidos graxos. Entretanto, M. tuberculosis apresenta ambos os sistemas, uma

cacterística que não é muito usual (DOVER et al., 2004). InhA é uma proteína pertencente

ao sistema de síntese micobacterial de ácidos graxos tipo II (FAS-II), que realiza a

elongação de precursores acil ácidos graxos, provindos do sistema de síntese micobacterial

de ácidos graxos tipo I (FAS-I) rendendo longas cadeias de carbono da parte meromicolato

dos ácidos micólicos (SCHROEDER et al., 2002). A função da InhA dentro do sistema

FAS-II é catalisar a redução dependente de NADH de duplas ligações trans entre a

posição C2 e C3 de substratos acil ácidos graxos (Figura 16) (QUEMARD et al., 1995).

Figura 16. Reação catalisada pela enzima InhA

A primeira estrutura da InhA em complexo com NADH foi determinada por Dessen

et al. (1995). A estrutura desta proteína apresenta uma estrutura terciária α/β com um

domínio apresentando um enovelamento Rossmann, que consiste em fitas β paralelas

ligadas por hélices α em uma ordem topológica β-α-β-α-β-α. A folha β formada apresenta

6 fitas na ordem 321456 (Figura 17). Este tipo de enovelamento classifica a InhA como

pertencente a superfamília das dehidrogenase/redutase de cadeia curta (SDR) (short

dehydrogenase/reductase) (DESSEN et al., 1995). Posteriormente à determinação desta

estrutura, outras estruturas da InhA com diferentes ligantes foram obtidas, entre elas os

complexos da InhA:NADH:isoniazida (ROSWARSKI et al., 1998), InhA:NADH:ácido

graxo C16 (ROSWARSKI et al., 1999) e o complexo da InhA com diversos ligantes

inibidores análogos de substrato ( KUO et al., 2003). Porém a conformação da InhA na

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

ausência de qualquer ligante ainda não foi elucidada, e assim, não se conhece a

conformação desta proteína antes da ligação da molécula de NADH.

Figura 17. Elementos de estrutura secundária da InhA de

Mycobacterium tuberculosis em complexo com NADH (em amarelo) e o

substrato C16 (em vermelho) (DESSEN et al., 1995)

A relação entre a ação da isoniazida e a inibição da InhA é um pouco complexa,

pois na verdade a isoniazida é uma pré-drogra que deve ser primeiramente convertida para

uma forma ativada através da enzima catalase-peroxidase (KatG) para um radical

isonicotínico acil (Figura 18) (ROZWARSKI et al., 1999). Este radical torna-se

covalentemente ligado ao anel nicotidamida do NADH ancorado no sítio ativo da InhA,

criando, desta forma, um NADH aducto que é o potente inibidor da InhA (ROZWARSKI

et al., 1998). A posição em que o radical isonicotínico acil liga-se à molécula de NADH é a

mesma região responsável pela transferência híbrida do átomo de hidrogênio que ocorre

durante a redução do substrato enoil-ACP (ROZWARSKI et al., 1998).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

isoniazida

ADPR

Aducto Isoniazida-NADH

Figura 18. Mecanismo de ativação da isoniazida pela enzima

KatG. ADPR: ADP ribose.

Porém, tem sido observado mutações na estrutura do gene inhA que confere

resistência a isoniazida. Cepas apresentando mutações foram obtidas de isolados clínicos

de pacientes que vivem nos Estados Unidos (BASSO et al., 1998). Os mutantes isolados

apresentam a substituição de um único resíduo de aminoácido da estrutura primária da

InhA. Entre as mutações obtidas podemos destacar as seguintes: S94A, I21V e I47T. Estas

mutações estão bem caracterizadas cineticamente (RAWART; WHITTY; TONGE, 2003,

OLIVEIRA et al., 2006). Recentemente as estruturas cristalográficas dos complexos de

InhA:NADH com essas mutações foram determinadas (OLIVEIRA et al., 2006).

Entretanto, a estrutura da InhA mutante S94A em complexo com NAD-isoniazida foi

determinada por Rozwarski et al. (1998) a 2,8 Å de resolução, mostrando tanto o efeito

desta mutação sobre a estrutura como a influência na conformação dos resíduos do sítio

ativo (figura 19). Porém, os efeitos de outras mutações como a I21V e I47T sobre a

estrutura do complexo da InhA:NAD-isoniazida ainda não foram revelados. A resistência à

isoniazida por esses mutantes parece estar relacionada a uma baixa afinidade da proteína

pela molécula de NADH, que desta maneira, poderia promover a ligação da molécula de

acil substratos antes da molécula de NADH, desta forma dificultando a ligação da

molécula NAD-isoniazida, devido a um impedimento estérico. Lembrando que análise de

dados de cinética enzimática para InhA tem demonstrado que a seqüência de ligação do

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

NADH e ácidos graxos não é estritamente ordenada, mas ocorre uma preferência para o

NADH ligar-se primeiro à enzima (QUEMARD et al., 1995). Entretanto quando ocorre a

ligação do NAD-isoniazida primeiramente a baixa afinidade pela molécula de NADH

poderia permitir a rápida liberação desta molécula, e assim promovendo posterior ligação

da molécula de NADH permitindo a normal catálise da enzima (ROZWARSKI et al.,

1998, OLIVEIRA et al., 2006).

Figura 19. Contatos moleculares entre o complexo isonicotínico acil-NADH e o sítio ativo

da InhA. O grupo acil isonicotínico derivado da isoniazida está em vermelho, a porção

NADH está em azul, a cadeia lateral da InhA está em verde e a Ser94, o resíduo que causa

a resistência a isoniazida quando convertida em Ala está em rosa. Os números

representam a distância (em Å) entre os átomos selecionados (ROZWARSKI et al., 1998).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

2. OBJETIVOS

Os objetivos do presente trabalho foram:

Cristalizar, coletar e processar dados de difração de raios X e resolver as estruturas

de duas enzimas da via do ácido chiquímico de M. tuberculosis, sendo elas a chiquimato

quinase complexada com ADP-ácido chiquímico (MtCQ-ADP-ác.Chiquímico), da

chiquimato quinase complexada com ADP e da corismato sintase na sua forma nativa

(MtCS).

Realizar a modelagem molecular da proteína triptofano sintase de M. tuberculosis

(MtTRPS) em complexo com inibidores indol propanol fosfato e uma série de 5 inibidores

ariltilalquil fosfonados.

Cristalizar, coletar e processar dados de difração de raios X e resolver as estruturas

das proteínas InhA selvagem e dos mutantes I21V e S94A em complexo com isoniazida e

a proteína mutante S94A na sua forma nativa.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3. RESULTADOS

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.1 Crystallization and preliminary X-ray crystallographic analysis of chorismate

synthase from Mycobacterium tuberculosis.

Marcio V. B. Dias; Fernanda Ely; Fernanda Canduri; José H. Pereira; Jeverson Frazzon;

Luiz A. Basso; Mário S. Palma; Walter F. de Azevedo Jr; Diógenes S. Santos. Acta

Crystallographica Section D Biological Crystallography (ISSN 0907-4449), v. D60, p.

2003-2005, 2004.

Neste trabalho é descrita a cristalização e a análise preliminar de dados de difração

de raios X para a corismato sintase de M. tuberculosis (peso molecular de 42014 Da).

A proteína corismato sintase foi clonada, expressa e purificada pelo grupo de

Pesquisa do Prof. Dr. Diógenes Santiago Santos da PUC – Porto Alegre – RS. A proteína

inicialmente precipitada em sulfato de amônio foi dialisada contra tampão Tris HCl, 50

mM, pH 7.8. Para a cristalização foram utilizados os métodos de difusão de vapor, sistema

hanging drop e matriz esparsa. Após a obtenção dos primeiros cristais, a condição inicial

foi otimizada, variando-se a concentração de sal, precipitante e proteína. Foram obtidos

cristais para a corismato sintase solubilizada em Tris-HCl, 50 mM, pH 7,8, com dimensões

adequadas para a coleta de dados de difração de raios X, em uma condição composta por

Hepes-Na, 0,1 M, pH 7,5, Cloreto de Magnésio hexahidratado, 0,6 M e PEG 400, 25%. A

proteína estava em uma concentração de 60 mg.ml

-1

e a razão entre a solução de

cristalização e a solução de proteína era de 2:1. A coleta de dados de difração de raios X

foi realizada no LNLS. Os cristais obtidos são hexagonais, pertencente ao grupo espacial

22 com dimensões aproximadas de 0,3x0,25x0,25 mm. O conjunto de dados obtido

para um cristal foi processado a 2.8 Å de resolução, apresentou uma completeza de 97,9%

e um R

sym

de 5,6%.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

crystallization papers

Acta Cryst. (2004). D60, 2003±2005 DOI: 10.1107/S0907444904019869 2003

Acta Crystallographica Section D

Biological

Crystallography

ISSN 0907-4449

Crystallization and preliminary X-ray

crystallographic analysis of chorismate synthase

from Mycobacterium tuberculosis

Marcio Vinicius Bertacine Dias,

Fernanda Ely,

Fernanda

Canduri,

a,b

Jose

Henrique

Pereira,

Jeverson Frazzon,

Luiz Augusto Basso,

rio Se

rgio Palma,

Walter Filgueira de

Azevedo Jr

a,b

* and

Dio

genes Santiago Santos

Departamento de Fõ

sica, UNESP, Sa

o Jose

Rio Preto, SP 15054-000, Brazil,

Rede

Brasileira de Pesquisas em Tuberculose Grupo

de Microbiologia Molecular e Funcional,

Departamento de Biologia Molecular e

Biotecnologia, UFRGS, Porto Alegre,

RS 91501-970, Brazil,

Laboratory of Structural

Biology and Zoochemistry, CEIS/Department of

Biology, Institute of Biosciences, UNESP, Rio

Claro, SP 13506-900, Brazil, and

Centro de

Pesquisas em Biologia Molecular e Funcional/

Instituto de Pesquisas Biome

dicas, Pontifõ

cia

Universidade Cato

lica do Rio Grande do Sul,

Porto Alegre, RS 90619-900, Brazil

Correspondence e-mail:

[email protected], diogenes@pucrs.br

# 2004 International Union of Crystallography

The enzymes of the shikimate pathway are potential targets for the

development of new therapies because they are essential for bacteria

but absent from mammals. The last step in this pathway is performed

by chorismate synthase (CS), which catalyzes the conversion of

5-enolpyruvylshikimate-3-phosphate to chorismate. Optimization of

crystallization trials allowed the crystallization of homogeneous

recombinant CS from Mycobacterium tuberculosis (MtCS). The

crystals of MtCS belong to space group P6

22 (or P6

22) and diffract

to 2.8 A

resolution, with unit-cell parameters a = b = 129.7, c = 156.8 A

There are two molecules in the asymmetric unit. Molecular-

replacement trials were not sucessful. Heavy-atom derivative

screening is in progress.

Received 15 July 2004

Accepted 10 August 2004

1. Introduction

Tuberculosis is the second leading cause of

deaths worldwide, killing nearly 2 million

people each year. Most cases are in under-

developed countries; over the past decade,

tuberculosis incidence has increased in Africa,

mainly as a result of the burden of HIV

infection, and in the former Soviet Union,

owing to socioeconomic change and decline of

the health-care system (Frieden et al., 2003). In

1993, the gravity of the situation led the World

Health Organization (WHO) to declare

tuberculosis a global emergency in an attempt

to heighten public and political awareness

(Cole et al., 1998). Thus, newer and more ef®-

cient anti-tuberculosis drugs are needed.

Potential targets for the development of new

therapies are the enzymes of the shikimate

pathway, because they are essential for

bacteria, fungi and apicomplexan parasites, but

absent from mammals (Bentley, 1990; Roberts

et al., 1998). In microorganisms, the shikimate

pathway is used to synthesize the three

proteinogenic aromatic amino acids phenyl-

alanine (Phe), tyrosine (Tyr) and tryptophan

(Trp), the folate coenzymes benzoid and

naphtoid quinones and a broad range of mostly

aromatic secondary metabolites, including

siderophores (Dosselaere & Vanderleyden,

2001). This pathway consists of seven enzymes

that catalyse the sequential conversion of

erythrose-4-phosphate and phosphoenol

pyruvate to chorismate, the common precursor

of aromatic compounds (Herrmann & Weaver,

1999). The last step in this pathway is

performed by chorismate synthase, which

catalyzes the conversion of 5-enolpyruvyl-

shikimate 3-phosphate to chorismate via the

1,4-anti-elimination of phosphate and a proton,

a reaction that is unique in nature. The enzyme

has an absolute requirement for reduced FMN

as a cofactor, although the 1,4-anti-elimination

of phosphate and the C(6-pro-R) H atom does

not involve a net redox change. The role of the

reduced FMN in catalysis has long been

elusive. However, recent detailed kinetic and

bioorganic approaches have fundamentally

advanced our understanding of the mechanism

of action of chorismate synthase (Macheroux et

al., 1999).

Structural information is necessary for

structure-based design in drug discovery,

for a better understanding of the catalytic

mechanism and also for an understanding of

protein±ligand interaction (de Azevedo et al.,

1996, 1997, 2002). Although structures of CS

from some microrganisms have been deter-

mined (Ahn et al., 2004; Maclean & Ali, 2003),

the crystal structure of CS from Mycobacterim

tuberculosis has not been described. The

crystal structure of shikimate-pathway

enzymes will help in the development of new

drugs against tuberculosis and other infectious

diseases (de Azevedo et al., 2002; Pereira et al.,

2003). In this work, we initiated the structure

determination of the aroF-encoded MtCS,

which is composed of 401 residues with a

molecular weight of 41 792 Da. We report here

the crystallization and the preliminary X-ray

crystallographic study of MtCS.

2. Materials and methods

2.1. Cloning, protein expression and

purification

Synthetic oligonucleotide primers were

designed based on the aroF structural gene

sequence of M. tuberculosis H37Rv (Cole et al.,

1998), containing 5

NdeI and 3

BamHI

restriction sites. The PCR product was cloned

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

crystallization papers

into pET-23a(+) (Novagen, USA) expres-

sion vector and transformed into Escher-

ichia coli Rosetta(DE3) host cells.

Transformed E. coli cells were grown in LB

medium at 310 K for 18 h in the absence of

IPTG, as used in previously reported

expression protocols (Oliveira et al., 2001;

Grossman et al., 1998), and harvested by

centrifugation. Although it is often argued

that the cost of IPTG limits the usefulness of

the lac promoter to high-added-value

products, it has previously been shown that

high levels of expression could be obtained

with pET vectors as cells entered the

stationary phase without the addition of

inducer in LB medium. It has been demon-

strated that when DE3 hosts are grown to

stationary phase in media lacking glucose,

cyclic AMP mediated derepression of both

the wild type and lacUV5 promoters occurs

(Oliveira et al., 2001). These authors also

proposed that cyclic AMP, acetate and low

pH are required for high-level expression in

the absence of IPTG induction when cells

approach the stationary phase in complex

media and that derepression of the lac

operon in the absence of IPTG may be part

of a general cellular response to nutrient

limitation.

The cell pellet was resuspended in 50 mM

Tris±HCl pH 7.8 (Sigma Co., USA), soni-

cated and centrifuged to remove cell debris.

The protein-puri®cation protocol included

the following steps: streptomycin sulfate

precipitation, ammonium sulfate precipita-

tion, Q Sepharose anion-exchange chroma-

tography, phenyl Sepharose hydrophobic

interaction chromatography and Mono Q

anion-exchange column (Amersham Phar-

macia Biotech, UK). A detailed description

of the cloning and expression of recom-

binant MtCS, the protein-puri®cation

protocol, N-terminal sequencing, mass

spectrometry and determination of the

oligomeric state of homogeneous M. tuber-

culosis chorismate synthase will be given

elsewhere (manuscript in preparation).

2.2. Crystallization

The puri®ed MtCS was concentrated and

dialyzed against 50 mM Tris±HCl buffer

pH 7.8 (Hampton Research, USA). The

®nal protein concentration was about

10 mg ml

À1

. Crystallization was performed

by the hanging-drop vapour-diffusion and

sparse-matrix methods (Jancarik & Kim,

1991) using tissue-culture multiwell plates

with covers (Linbro, ICN Biomedicals, Inc,

USA) at a temperature of 293 K. Each

hanging drop was prepared by mixing 1 ml

each of protein solution and reservoir solu-

tion and was placed over 700 ml reservoir

solution. Initial conditions were screened

using Crystal Screen I and II kits (Hampton

Research, USA).

Crystal optimization was carried out by

altering the concentration of the salt, preci-

pitant and protein and also the ratio

between the protein solution and reservoir

solution.

2.3. X-ray data collection

A data set was collected at a wavelength

of 1.427 A

using a synchrotron-radiation

source (Station PCr, LNLS, Campinas-

Brazil; Polikarpov et al., 1998). This wave-

length was previously set to optimize the

overall data-collection statistics. The data set

was collected from a single MtCS crystal

using a MAR CCD image-plate system. The

crystal was looped out from the drop and

¯ash-cooled. The PEG 400 present in the

crystallization conditions served as a cryo-

protectant. X-ray diffraction data were

collected at a temperature of 100 K under a

cold nitrogen stream generated and main-

tained with an Oxford Cryosystem. The

crystal was rotated through a total of 160



with a 1



oscillation range per frame, a

crystal-to-detector distance of 130 mm and

an exposure time of 60 s. Data were

processed on a Silicon Graphics Octane2

computer using the programs MOSFLM

(Leslie, 1990) and SCALA (Collaborative

Computational Project, Number 4, 1994).

2.4. Molecular-replacement trials

Molecular-replacement trials were carried

out with the program AMoRe (Navaza,

1994). The crystal structures of SpCs (PDB

code 1qxo; Maclean & Ali, 2003) and AaCs

(PDB code 1q1l; Viola et al., 2004) were

used as search models.

3. Results and discussion

The initial crystals were obtained with

reservoir solution comprising 0.2 M mag-

nesium chloride hexahydrate, 0.1 M Na

HEPES pH 7.5 and 30%(w/v) PEG 400. The

ratio of protein solution to well solution in

the drop was 1:1. After optimization of these

conditions, better crystals were obtained.

These diffracting crystals grew from a

reservoir solution containing 0.6 M mag-

nesium chloride hexahydrate, 0.1 M Na

HEPES pH 7.5 and 25%(w/v) PEG 400 and

the protein solution was concentrated to

60 mg ml

À1

in 50 mM Tris±HCl buffer pH

7.8. The ratio of protein solution to well

solution in the drop was 2:1. The crystals

grew reproducibly to approximate dimen-

sions of 0.3 Â 0.25 Â 0.25 mm within 2 d

(Fig. 1). The crystal diffracted to 2.8 A

with

relatively low mosaicity (0.35



). Fig. 2 shows

a typical X-ray diffraction pattern. A total of

92 610 measured re¯ections were merged

2004 Dias et al.



Chorismate synthase Acta Cryst. (2004). D60, 2003±2005

Figure 1

Hexagonal crystals of MtCS. Approximate dimen-

sions are 0.30 Â 0.25 Â 0.25 mm.

Figure 2

A typical diffraction pattern of the MtCS crystal with



oscillation range. The crystal diffracts to 2.8 A

resolution.

Table 1

Summary of data-collection statistics for MtCS.

Values in parentheses are for the highest resolution shell.

X-ray wavelength (A

) 1.427

Unit-cell parameters

a (A

) 129.74

b (A

) 129.74

c (A

) 156.77

Space group P6

22 or P6

No. measurements with I >2'(I) 92610

No. independent re¯ections 19341 (2716)

Completeness in the resolution

range 55.8±2.8 A

(%)

97.9 (97.9)

sym

² (%) 5.6 (16.5)

hI/'(I)i 3.8 (4.0)

Highest resolution shell (A

) 2.94±2.80

² R

sym



jIh

ÀhIhija



Ih

, where I(h)is

the intensity of re¯ection h,



is the sum over all re¯ections

and



is the sum over i measurements of re¯ection h.

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

crystallization papers

into 19 341 unique re¯ections. The overall

sym

was 5.6% and the completeness was

97.9%. The crystal belongs to the hexagonal

space group P6

22 (or P6

22), with unit-cell

parameters a = b = 129.7, c = 156.8 A

Assuming the asymetric unit content to

be two monomers of molecular weight

41 792 Da, the V

value is 2.28 A

À1

Assuming a value of 0.74 cm

À1

for the

protein partial speci®c volume, the calcu-

lated solvent content in the crystal is 45.96%

(Matthews, 1968). Table 1 summarizes the

data-collection statistics.

With the native data of MtCS to 2.8 A

resolution, molecular replacement was used

to attempt to solve the structure using

AMoRe (Navaza, 1994). Various search

models, including complete and modi®ed

structures of SpCS (PDB code 1qxo) and

AaCS (1q1l), did not yield any meaningful

results. This was most likely to be because of

conformational differences between the

search models and the structure under study.

Heavy-atom screening is in progress.

This work was supported by grants from

FAPESP (SMOLBNet, Proc. 01/07532-0, 02/

04383-7, 04/00217-0), CNPq, CAPES and

Instituto do Mile

nio (CNPq-MCT) to DSS

and LAB. WFA (CNPq, 300851/98-7), MSP

(CNPq, 500079/90-0), DSS (CNPq, 304051/

1975-06), LAB (CNPq, 520182/99-5) and JF

(CNPq, 301131/2003-01) are researchers for

the National Research Council.

References

Ahn, H. J., Yoon, H. J., Lee, B. & Suh, S. W. (2004).

J. Mol. Biol. 27, 903±915.

Azevedo, W. F. de Jr, Canduri, F. & da Silveira

N. J. F. (2002). Biochem. Biophys. Res.

Commun. 293, 566±571.

Azevedo, W. F. de Jr, de Oliveira, J. S., Basso, L. A.,

Palma, M. S., Pereira, J. H., Canduri, F. &

Santos, D. S. (2002). Biochem. Biophys. Res.

Commun. 295, 142±148.

Azevedo, W. F. de Jr, Leclerc, S., Meijer, L.,

Havlicek, L., Strnad, M. & Kim, S.-H. (1997).

Eur. J. Biochem. 243, 518±526.

Azevedo, W. F. de Jr, Mueller-Dieckmann, H. J.,

Schulze-Gahmen, U., Worland, P. J., Sausville,

E. & Kim, S.-H. (1996). Proc. Natl Acad. Sci.

USA, 93, 2735±2740.

Bentley, R. (1990). Crit. Rev. Biochem. Mol. Biol.

25, 307±384.

Cole, S. T. et al. (1998). Nature (London), 393,

537±544.

Collaborative Computational Project, Number 4

(1994). Acta Cryst. D50, 760±763.

Dosselaere, F. & Vanderleyden, J. (2001). Crit.

Rev. Microbiol. 27, 75±131.

Frieden, T. R., Sterling, T. R., Munsiff, S. S., Watt,

C. J. & Dye, C. (2003). Lancet, 362, 887±899.

Grossman, T. H., Kawasaki, E. S., Punreddy, S. R.

& Osburne, M. S. (1998). Gene, 209, 95±103.

Herrmann, K. M. & Weaver, L. M. (1999). Annu.

Rev. Plant. Physiol. Plant. Mol. Biol. 50, 473±

503.

Jancarik, J. & Kim, S.-H. (1991). J. Appl. Cryst. 24,

409±411.

Leslie, A. G. W. (1990). Crystallographic

Computing, edited by D. Moras, A. D. Podjarny

& J. C. Thierry, pp. 27±38. Oxford University

Press.

Macheroux, P., Schmid, J., Amrhein, N. & Schaller,

A. (1999). Planta, 207, 325±334.

Maclean, J. & Ali, S. (2003). Structure, 11, 1499±

1511.

Matthews, B. W. (1968). J. Mol. Biol 33, 491±

497.

Navaza, J. (1994). Acta Cryst. A50, 157±163.

Oliveira, J. S., Pinto, C. A., Basso, L. A. & Santos,

D. S. (2001). Protein Expr. Purif. 22 , 430±435.

Pereira, J. H., Canduri, F., de Oliveira, J. S., da

Silveira, N. J., Basso, L. A., Palma, M. S., de

Azevedo, W. F. Jr & Santos, D. S. (2003).

Biochem. Biophys. Res. Commun. 312, 608±

614.

Polikarpov, I., Perles, L. A., de Oliveira, R. T.,

Oliva, G., Castellano, E. E., Garratt, R. C. &

Craievich, A. (1998). J. Synchrotron Rad. 5, 72±

76.

Roberts, F., Roberts, C. W., Johnson, J. J., Kyle,

D. E., Krell, T., Coggins, J. R., Coombs, G. H.,

Milhous, W. K., Tzipori, S., Ferguson, D. J.,

Chakrabarti, D. & McLeod, R. (1998). Nature

(London), 395, 801±805.

Viola, C. M., Saridakis, V. & Christendat, D.

(2004). Proteins, 54, 166±169.

Acta Cryst. (2004). D60, 2003±2005 Dias et al.



Chorismate synthase 2005

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.2 Structure of chorismate synthase from Mycobacterium tuberculosis.

Marcio V.B. Dias; Júlio C. Borges; Fernanda Ely; José H. Pereira; Fernanda Canduri;

Carlos H.I. Ramos; Jeverson Frazzon; Mário S. Palma; Luis A. Basso; Diógenes S. Santos;

Walter F. de Azevedo Jr. Journal of Structure Biology (ISSN 1047-8477), v. 154, p. 130-

143, 2006.

Neste trabalho é descrita a estrutura cristalográfica da proteína corismato sintase de

M. tuberculosis (peso molecular de 42014 Da).

A estrutura da corismato sintase de M. tuberculosis foi resolvida a 2,65 Å de

resolução através da técnica de substituição molecular utilizando como modelo de busca a

estrutura da corismato sintase de S. pneumoniae (MACLEAN; ALI, 2003). Os resíduos

não conservados entre as duas estruturas foram trocados para alanina, com exceção de

glicinas. A função translação obtida na substituição molecular apresentava um coeficiente

de correlação de 37,6% e um R-factor de 53%. Após a confirmação do grupo espacial

correto (P6

22), realizou-se refinamento cristalográfico. O modelo para a corismato sintase

de M. tuberculosis foi iterativamente ajustado com base em mapas de densidade eletrônica

com coeficientes 2Fo – 2Fc e Fo – Fc. A estrutura final obtida apresenta um R-factor de

16,2% e um R-free de 22,1% e uma boa qualidade estereoquímica, apresentando 99,3%

dos resíduos em regiões favoráveis do gráfico de Ramachandran e apenas 0,3% em regiões

não permitidas.

A estrutura da corismato sintase de M. tuberculosis é similar às outras estruturas

determinadas anteriormente para outros microorganismos. A unidade assimétrica do cristal

apresenta um único monômero, que pertence à classe de proteínas α/β. Este monômero

apresenta somente um domínio, cuja topologia estrutural é um sanduíche β-α-β. Este

sanduíche apresenta um centro formado de hélices que se encontra entre duas camadas de

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

folhas β formadas por quatro fitas. Este domínio é rodeado por loops e trechos discretos de

hélices α e fitas β. Entretanto, a estrutura cristalográfica da corismato sintase de M.

tuberculosis pode ser descrita como um tetrâmero formado por dímeros de dímeros, com

simetria 222, sendo que a principal característica da interface de dimerização é uma folha β

antiparalela intermonomérica, formada por oito fitas, sendo quatro de um monômero e

quatro do outro monômero adjacente. A comparação do sítio ativo da corismato sintase de

M. tuberculosis com a de outros organismos mostra que muitos resíduos são conservados

entre estas estruturas. Entretanto a posição da cadeia lateral de vários resíduos apresenta-se

em conformações diferentes, talvez devido à ausência de ligantes em nossa estrutura e a

presença de ligantes nas outras estruturas que foram utilizadas para a análise. Portanto,

devido à ausência de ligantes em nossa estrutura e a determinação da posição de quase

todos os resíduos do sítio ativo, com exceção dos resíduos 49-52, pode-se inferir

importantes movimentos ocorridos pela ligação do substrato EPSP e da co-enzima FMN no

sítio ativo desta proteína.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Structure of chorismate synthase from Mycobacterium tuberculosis

Marcio V.B. Dias

,Ju

lio C. Borges

, Fernanda Ely

, Jose

H. Pereira

Fernanda Canduri

a,c

, Carlos H.I. Ramos

, Jeverson Frazzon

,Ma

rio S. Palma

Luis A. Basso

, Dio

genes S. Santos

, Walter F. de Azevedo Jr.

a,g,

Programa de Po

s-Graduac¸a

o em Biofı

sica Molecular, Departamento de Fı

sica, UNESP, Sa

o Jose

do Rio Preto, SP 15054-000, Brazil

Rede Brasileira de Pesquisa em Tuberculose, Grupo de Microbiologia Molecular e Funcional, Departamento de Biologia Molecular e Biotecnologia,

UFRGS, Porto Alegre, RS 91501-970, Brazil

Departamento de Morfoﬁsiologia, Laborato

rio de Bioquı

´mica,

CCBS, Universidade Federal de Mato Grosso do Sul,

Campo Grande, MS, CEP 79070-900, Brazil

Centro de Biologia Molecular Estrutural, Laborato

rio Nacional de Luz Sı

´nc

rotron, Campinas SP, Brazil

Departamento de Biologia, CEIS/IBRC, UNESP, Rio Claro, SP, CEP 13506-900, Brazil

Pontifı

cia Universidade Cato

lica do Rio Grande do Sul, Centro de Pesquisa em Biologia Molecular e Funcional, Porto Alegre, RS, Brazil

Faculdade de Biocie

ncias—Pontifı

cia Universidade Cato

lica do Rio Grande do Sul, Av. Ipiranga, 6681. Porto Alegre-RS CEP 90619-900, Brazil

Received 17 July 2005; received in revised form 8 November 2005; accepted 9 December 2005

Available online 17 January 2006

Abstract

In bacteria, fungi, plants, and apicomplexan parasites, the aromatics compounds, such as aromatics amino acids, are synthesized

through seven enzymes from the shikimate pathway, which are absent in mammals. The absence of this pathway in mammals make

them potential targets for development of new therapy against infectious diseases, such as tuberculosis, which is the world’s second

commonest cause of death from infectious disease. The last enzyme of shikimate pathway is the chorismate synthase (CS), which is

responsible for conversion of the 5-enolpyruvylshikimate-3-phosphate to chorismate. Here, we report the crystallographic structure

of CS from Mycobacterium tuberculosis (MtCS) at 2.65 A

resolution. The MtCS structure is similar to other CS structures, present-

ing b–a–b sandwich structural topology, in which each monomer of MtCS consists of a central helical core. The MtCS can be

described as a tetramer formed by a dimer of dimers. However, analytical ultracentrifugation studies suggest the MtCS is a dimer

with a more asymmetric shape than observed on the crystallographic dimer and the existence of a low equilibrium between dimer

and tetramer. Our results suggest that the MtCS oligomerization is concentration dependent and some conformational changes must

be involved on that event.

Keywords: Chorismate synthase; Crystallography; Analytical ultracentrifugation; Mycobacterium tuberculosis; Shikimate pathway

1. Introduction

In mammals, diet has to provide the essential amino

acids, such as, phenylalanine, tryptophan, and tyrosine.

In bacteria, fungi, plants, and apicomplexan parasites,

these aromatic amino acids are synthesized through the

complex shikimate pathway (Bentley, 1990; Macheroux

et al., 1999). Furthermore, the shikimate pathway provides

these organisms the basic building blocks for the synthesis

of the other aromatic compounds required for diﬀerent

functions as UV protection, electron transport, signaling,

iron uptake, etc. The shikimate pathway is responsible

for conversion of

D-erythrose-4-phosphate and phospho-

enolpyruvate to shikimate and subsequently to the dihyd-

roaromatic compound chorismate (Bentley, 1990; Dossela-

ere and Vanderleyden, 2001). The absence of shikimate

pathway in mammals has rendered its enzymes as

www.elsevier.com/locate/yjsbi

Journal of Structural Biology 154 (2006) 130–143

Journal of

Structural

Biology

doi:10.1016/j.jsb.2005.12.008

Corresponding authors. Fax: +55 17 51 3220 3629.

E-mail addresses: diogenes@pucrs.br (D.S. Santos), walter.junior@

pucrs.br (W.F. de Azevedo Jr.).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

potentially targets for the development of new therapy

against infectious disease, such as tuberculosis (TB).

TB is considerate one of the most serious global public

health challenge of the 21st century. It is the world’s second

commonest cause of death from infectious disease, after

acquired immune deﬁciency syndrome. According to a

recent report compiled by the World Health Organization,

the total number of new cases of TB worldwide in 2002 ha s

increased to around 9 million (Duncan, 2004; WHO, 2004),

from which 2–3 million people died, despite the chemother-

apy available (WHO, 2004). Thes e deaths are mostly of

young adults but also include about 100 000 children unde r

the age of 5 years (Gandy and Zumla, 2002). The key driver

of the increase of TB is its synergy with the human immu-

nodeﬁciency virus epidemic, which has a devastating

impact in some parts of the world, such as, the African

region (Duncan, 2004). Thus, there is an urgent need for

new anti-mycobacterial inhibitors.

The enzymes of shikimate pathway are good candidates

for development of new therapies against TB. Enzymes

from this metabolic pathway have been submi tted to inten-

sive structural studies (Azevedo et al., 2002; Basso et al.,

2005; Pereira et al., 2003, 2004; Silveira et al., 2005). The

last enzyme from this pathway is the ch orismate synthase

(CS), which catalyzes the conversion of the 5-enolpyruvyls-

hikimate-3-phosphate (EPSP) to chorismate. The CS reac-

tion comprises an anti-1,4-elimination of the 3-phosphate

group and the C(6proR)-hydrogen (Bornemann et al.,

2003; Hill and Newkome, 1969; Macheroux et al., 1998).

It is the only enzymatic reaction known of such transfor-

mation in biological systems, making the CS a unique

enzyme in the nature. The CS requires reduced ﬂavin

mononucleotide (FMN), an essential cofactor typically

found in many biologi cal redox reactions. Surprisingly,

the reaction catalyzed by CS does not involve an overall

change in redox state (Bornemann et al., 1996;Kitzing

et al., 2004; Macheroux et al., 1996; Macheroux et al.,

1999). According Bornemann et al. (1995), the reduced

FMN donates an electron to EPSP to facilitate the loss

of the phosphate and receive it back after the reaction.

So, only ﬂavin in its reduced form is functional and it is

not consumed during the reaction (Macheroux et al.,

1998; Welch et al., 1974). Furthermore, two classes of CS

are distinguished among the microorganisms that possess

shikimate pathway. CS from yeasts ha ve the ability to

use b-nicotinamide adenine dinucleotide phosphate

(NAD(H)P) for the reduction of oxidized FMN, having

therefore, an additional catalytic activity and wher efore

are called bifunctionals, while all the other CS lacking it

are called monofunctional (Macheroux et al., 1999).

There are crystal structures available of CS from three

bacteria: CS from Streptococcus pneumoniae complex ed

with FMN and EPSP (Maclean and Ali, 2003); CS from

Helicobacter pylori co mplexed with FMN (Ahn et al.,

2004; and native CS from Aquiﬂex aelicus (Viola et al.,

2004) (PDB access codes: 1QXO, 1UMO, and 1Q1L,

respectively) and one CS from yeast (native CS of

Saccharomices cerevisae) (PDB access code: 1R53,

Quevillon-Cheruel et al., 2004). In all these structures, the

CS is presented as tetramer composed of two dimers, like

a dimer of dimers. The monomer comprises a single large

core domain, which is surrounded by loops and discrete

stretches of a-helix and b-sheet. Here, we present structural

data of the CS from M. tuberculosis (MtCS) in its native

form. The structure of the MtCS was determined at

2.65 A

resolution and it presents a b–a–b architecture.

The analytical ultracentrifugation data suggest that the

MtCS is predominantly a dimer in solution presenting

equilibrium with a tetrameric form. This structure can help

us understand the action mecha nism of the MtCS.

2. Experimental procedures

2.1. MtCS recombinant production, puriﬁcation, and

crystallization

The MtCS cloning, expression, puriﬁcation, and crystal-

lization were reported elsewhere (Dias et al., 2004). The

protein concentration was determined spectrophotometri-

cally as described by Edelhock (1967), using a calculated

extinction coeﬃcient for denatured proteins (Gill and von

Hippel, 1989).

2.2. Circular dichroism spectroscopy

Circular dichroism (CD) measurements were performed

using a Jasco J-810 spectropolarimeter with the tempera-

ture controlled by a Peltier-type control system PFD

425S. The data were collected at a scanning rate of

100 nm/min with a spectral band width of 1 nm and using

a 1 mm path length cell. The thermal-induced unfolding

experiments followed by CD were performed at a scan rate

of 1 °C/min. The average of three unfolding curves was

used to build the MtCS thermal-unfold proﬁle and the tem-

perature at the midpoint of the unfolding transition was

determined by ﬁtting with Gaussians of the ﬁrst derivative

function.

2.3. Structure determination

The data set of MtCS was collected at a wavelength of

1.427 A

using the Synchrotron Radiation Source (Station

PCr, LNLS, Campinas—Brazil) (Polikarpov et al., 1998)

and a CCD detect or (MARCCD). The crystal was ﬂash-

frozen at 104 K under cold nitrogen stream generated

and maintained with an Oxford Cryosystem. The data set

was processed up to 2.65 A

resolution using the program

Abbreviations used: TB, tuberculosis; CS, chorismate synthase; MtCS,

chorismate synthase from Mycobacterium tuberculosis; EPSP, 5-eno-

lpyruvylshikimate-3-phosphate; FMN, ﬂavin mononucleotide; CD, circu-

lar dichroism; RMSD, root-mean-square deviation; AUC, analytical

ultracentrifugation.

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 131

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

MOSFLM ( Leslie, 1992) and scaled with SCALA (CCP4,

1994).

The crystal structure of the MtCS was determined by

standard molecular replac ement methods implemented in

the program AMoRe (Navaza, 2001), using as search mod-

el the monomer structure of Streptococcus pneumoniae CS

(SpCS) (PDB access code: 1QXO) (Maclean and Ali,

2003), in which nonconserved residues were changed to

alanines. The rotation search was calculated using data

in the range of 12.0–3.0 A

resolution, which gave a solu-

tion with a correlation of 7.3%. The translation search

yielded a unique solution with a correlation of 37.6%

and an R-factor of 53.0% was obtained. At this stage,

the correct space group was conﬁrmed to be P6

22 by visu-

alization of crystal packing using the program O (Jones

et al., 1991). Further crystallographic reﬁnement was car-

ried out using X-PLOR and simulated annealing (Brunger,

1992). The protein model obtained by molecular replace-

ment was reﬁned resulting R-factor and R-free of about

33.6 and 43.1%, respectively. Throughout the reﬁnement,

the composite electron-density maps with coeﬃcients

À F

and F

À F

were calculated with X-PLOR and

visualized using XtalView/Xﬁt (McRee, 1999), and the

model was built and iteratively adjusted. After several

reﬁnement steps the R-factor and the R-free decreased to

25.7 and 32.5%, respectively. Further, the group B factors

were adjusted and it was used the data from 8.0 to 2.65 A

The model was subjected to further reﬁnement using the

maximum-likelihood based program REFMAC5 (Murs-

hudov et al., 1997) using data between 52.93 and 2.65 A

resolution. At this stage, water molecules were added to

the model using the program XtalView and the ARP rou-

tine of REFMAC5. Reﬁnement converged to R-factor and

R-free values of 16.2 and 22.1%, respectively. The correct-

ness of the stereochemi stry of the model was checked using

PROCHECK (Laskowski et al., 1994). Root-mean-square

deviation (RMSD) diﬀerences from ideal geometries for

bond lengths, angles, and dihedrals were calculated with

X-PLOR (Bru

nger, 1992). Atomic models were superposed

using the program LSQKAB from CCP4 (CCP4, 1994).

The molecular surface areas have been calculated using

the program AREAIMOL/RESAREA (CCP4, 1994).

The PARMODEL (Ucho

a, 2004) was used in the analysis

of the ﬁnal model.

2.4. Gel ﬁltration

Gel ﬁltration was performed using a Superdex S-200

column (Amershan Pharmacia Biotech) in FPLC system.

The column was equilibrated in 50 mmol/L of Tris–HCl

buﬀer, pH 7.8, containing 200 mmol/L of NaCl. Protein

standards used were ribonuclease A (13.7 kDa), chymo-

trypsinogen (25 kDa), ovalbumin (43 kDa), albumin

(67 kDa), aldalose (158 kDa), catalase (232 kDa), ferritin

(440 kDa), and thyroglobulin (669 kDa) (Amershan Phar-

macia Biotech). Elution time of MtCS was recorded and

molecular weight was calculated by estimating the elution

volumes of standards of known molecular weight. The

MtCS were loaded on the gel ﬁltration column at a con-

centration of 1 mg/mL.

2.5. Analytical ultracentrifugation

Sedimentation velocity and sedimentation equilibrium

experiments were performed using a Beckman Optima

XL-A analytical ultracentrifuge. The sedimentation veloc-

ity experiments were carried out at 20 °C and 25000 rpm

(AN-60Ti rotor) with the scan data acquisition at 232

and 236 nm. MtCS was tested in concentrations of

300–1000 lg/mL in 20 mmol/L Tris–HCl buﬀer, pH 8.0,

containing 50 mmol/L NaCl and 1 mmol/L b-mercap-

toethanol. The analysis involved ﬁtting a model of absor-

bance versus cell radius data by nonlinear regression. The

analysis was performed with the ORIGIN software

package (MicroCal Software) supplied with the instru-

ment. The second moment (Goldberg, 1953) an d the

sedimentation time derivative (g (s*) integral distribution)

(Staﬀord, 1994) methods were used to analyze the sedi-

mentation velocity experiments. These methods allow

the calculation of the apparent sedimentation coeﬃcient

(s*), the diﬀusion coeﬃcient D, and the molecular mass

MM. The ratio of the sedimentation coeﬃcient to diﬀu-

sion coeﬃcient gives the molecular mass by use of the

follow equation:

MM ¼

sRT

Dð1 À VbarqÞ

; ð1Þ

where R is the gas constant and T is the absolute tempera-

ture. The software Sednterp (www.jphilo.mailway.com/

download.htm) was used to estimate protein partial speciﬁc

volume at 20 °C(Vbar = 0.7329 mL/g), buﬀer density

(q = 1.00087 g/mL), and viscosity (g = 1.0126 · 10

À2

P),

and the s

max

for globular proteins of about 41.8, 83.6,

and 167.2 kDa. This software was also used to estimate

the standard sedimentation coeﬃcients (s

20,w

) for each pro-

tein concentration that were used to estimate the s

20,w

0 mg/mL protein concentration by extrapolation (s

20;w

Þ.

That procedure minimizes interferences caused by temper-

ature, viscosity solution, and molecular crowd (Laue,

2001).

The sedimentation equilibrium experiments were made

at 20 °C at speeds of 8000, 10000, and 12 000 rpm with

the AN-60Ti rotor, scan data acquisition at 236 nm and

at protein concentration of 300, 450, and 600 lg/mL. Brief-

ly, the samples were accelerated to 8000 rpm and scans

were collected until no changes in the scan were observed.

Three scans were obtained at 8000 rpm. Similar procedures

were made at 10000 and 12 000 rpm. The self-association

method was used to analyze the sedimentation equilibrium

experiments using several models of association to ﬁt the

MtCS data. The distribution of the protein along the cell,

obtained in the equilibrium sedimentation experiments,

was ﬁtted with the following equation (Johnson et al.,

1981):

132 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

C ¼ C

M ð1 À VbarqÞx

ðr

À r

2RT



; ð2Þ

where C is the protein concentration at radial position r , C

is the protein concentration at radial position r

, and x is

the centrifugal angular velocity.

The HydroPro software (Garcı

a de la Torre et al., 2000)

was applied to estimate the standard sedimentation coeﬃ-

cient s

20,w

, starting from the high resolution MtCS struc -

ture: as a monomer, dimer an d tetramer. The HydroPro

software was setup with the radius of the atomic elements

of 2.5 A

, with sigma factors from 5 to 8 (as indicated by

supplier) and minibeads radius (SIGMIN and SIGMAX)

from 6 to 2 A

after initial evaluation of the two extremes.

The parameters MtCS Vbar, and q and g (for standard

conditions) were estimated using the software Sednterp as

described above.

3. Results and discussion

3.1. MtCS spectropolarimetry analysis

The observed spectrum of CD data is shown in Fig. 1,

and it is similar to the one obtained for CS from Escherich-

ia coli (EcCS) (Macheroux et al., 1998). The deconvolution

of CD data predicted that the MtCS has nearly 30 and 18%

of a-helices and b-sheet, respectively. This result is in agree-

ment with the crystallographic structure presented here (34

and 17% of a-helices and b-sheet, respectively). the MtCS

thermal stability is presented in Fig. 1 (inset). The data

show that the MtCS has two well-characterized thermal

unfolding transitions: one at 51.5 °C and another at

64 °C, which is disagreement with the data from thermal

unfolding of the EcCS. The data for EcCS suggests only

one unfolding transition at 54 °C(Fitzpatrick et al.,

2001), despite of the experimental conditions were diﬀerent

than that applied for the MtCS. Thus, the data obtained

here, may suggest that the MtCS possesses two well-deﬁned

domains, each one presenting diﬀerent thermal stabilities,

despite MtCS crystal structure does not show the existence

of these domains (see below). Then, other hypotheses is

that the protein may loss its oligomeric status before under-

going the fully unfold. Furthermore, as the protein is a

dimer in solution (see below), the secondary structures on

the ﬂanks of the dimerization interface core can also under-

go unfolded at 51.5 °C and the core lost its secondary

structure at 64 °C. The MtCS unfolding mechanism is cur-

rently under investigation.

3.2. MtCS presents a b–a–

b architecture

The crystal structure of the MtCS was solved at 2.65 A

using the Synchrotron Radiation Source (Polikarpov et al.,

1998), and the structure was determined by molecular

replacement. The ﬁnal statistics for data processing and

reﬁnement are in Table 1. There is one molec ule of MtCS

in the asymm etric unit, containing residues 1–48 and 53–

392 and 243 water molecules (Fig. 2A). The residues 49–

52 and 393–401 did not present electronic density, and they

were omitted in the ﬁnal model. The analysis of stereo-

chemical quality, G-factor and 3D-proﬁle are presented in

Table 1.

The structure of MtCS is similar the other CS structures

deposited in the Protein Data Bank (PDB) (PDB access

code: 1QXO, 1Q1L, 1UMO, and 1R53; Ahn et al., 2004;

Maclean and Ali, 2003; Quevillon-Cheruel et al., 2004; Vio-

la et al., 2004, respectively). The structure of the MtCS

monomer belongs to the a/b class. Each monomer of

MtCS contains 13 a-helices and 17 b-sheets, which fold

into an approximate dimension of 75 A

· 47 A

· 36 A

MtCS presents only one domain and its dominant structur-

al topology is a b–a–b sandwich, in which each monomer

of MtCS contains a central helical core (formed by helices

a1, a5, a11, and a8). This helical core is sandwiched

between two four-stranded antiparallel b-sheets (b1, b2,

b7, b4, and b8, b9, b15, b10), which are surrounded by

loops and discrete stretches of a-helix and b-sheet. The

topology of MtCS is shown in

Fig. 2B and can be observed

that the core of structure of MtCS has pseudo 2-fold sym-

metry, despite the molecule does not present such

symmetry.

The crystal structure obtained for MtCS may be

described as a tetramer formed by a dimer of dimers, pre-

senting approximately 222 symmetry. The intermolecular

interactions among the monomers A and D form the ﬁrst

dimer and those among the monomers B and C forming

the second dimer. But each monomer of CS is in contact

with three others, creating an intricate packing arrange-

ment (Fig. 3). The contact areas between the monomers

are 796.0, 1944.0, and 3977.0 A

for the unit pairs AB/

CD, AC/BD, and AD/BC, respectively, and many residues

involved on the interface of tetramerization are conserved

062052042032022012002

51-

01-

-12

-10

-8

-6

-4

-2

80706050403020

( erutarepmeT

[] (Deg.cm

.dmol

-1

)x10

)mn( htgnelevaW

[θ]

222nm

(Deg.cm

.dmol

-1

)x10

Fig. 1. Circular dichroism analyses. Residual molar ellipticity of the

MtCS was measured from 195 to 260 nm in 20 mmol/L Tris–HCl buﬀer

(pH 8.0) containing 50 mmol/L of NaCl, and 1 mmol/L b-mercap-

toethanol at 20 °C. The amount of secondary structure was estimated by

CDNN Deconvolution software (Bo

hm et al., 1992). Thermal stability of

MtCS was analyzed by CD at 222 nm.

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 133

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

in several sequences from bacteria (Fig. 4). Fig. 5 introduc-

es the interface contact area between the monomers pairs

AB, AC, and AD, emphasizing the molecular surface.

The dimers form ed with monomers A/D or B/C of CS

are charact erized by a large antiparallel eight-stranded b-

sheet that is formed by combination of two sheets, one

from each monomer (Fig. 5A). This large b-sheet is the

major feature of the MtCS dimer interface and may be said

that the two strands b10 are the principal region for the

dimer stabilization. Another secondary structure element

involved in stabilization of the dimer is the helix a8 that

interacts with a8 from another monomer. These interac-

tions contribute highly to dimer stabi lization, presenting

considerable amount of hydrophobic interactions on the

surface at the dimer interface. The dimer also presents

two additional four-strand antiparallel b-sheet, one of each

side of the core formed by association between the b11 and

b12 from one monomer and b16 and b17 of another mono-

mer. Moreover, the dimer interface presents several inter-

actions (44 hydrogen bonds), mainly involv ing loops and

the a3 and a4.

The main residues involved in the MtCS dimerization

are Lys123, Tyr124, Arg135, Gly171, Asp217, Gly220,

Gly233, Gly235, Asp245, Ile25 5, Lys259, Gly274, and

Pro316, which are conserved in almost all CS sequences

from bacteria including the four-solved structures (Fig. 4,

see below). These results suggest that probably all CS are

at least dimeric due the high number of hydrogen bonds

and conservation of the residues involved in this interac-

tion. Fig. 4 shows the alignment of 16 sequences of CS

from bacteria evidencing the conserved residues involved

on the interface of two monomers A and D or B and C,

which form the dimer.

The MtCS crystal structure clearly shows that the inter-

face interactions between the dimer AD and BC, which

form of tetrameric structure of CS involve both hydropho-

bic and polar residues, and some regions present comple-

mentary of charge.

On the other hand, the forces that stabilize the interac-

tion between the monomers A and B are made mainly by

two salt-bridges and six hydrogen bonds localized in the

following regions: b2, a7, a11, a12, and L29, mainly

between a12 from one monomer with the adjacent from

other monomer. The number of the interactions between

the monomers A and C is higher than those observed for

monomers A and B. The interaction between monomer A

and C is made mainly by 18 hydrogen bonds localized prin-

cipally in regions of loops. The protein regions responsible

for these interactions are: L1, a1, L5, b5, L6, b7, L8, L9,

a3, L10, and L29. The predominant region of interaction

between monomer A and C is L8, which interact with the

L8 from the adjacent monomer.

3.3. Comparison between MtCS structure and others CS

structures

The monomeric structure of MtCS was compared with

ﬁve other CS structures from four organisms: three from

bacteria (Aquifex aeolicus—AaCS; Streptococcus pneumo-

niae—SpCS and Helicobacter pylori—HpCs) and one from

yeast (Saccharomyces cerevisiae—ScCS) deposited in the

PDB (PDB access code: 1Q1L, 1QXO, 1UMO, and

1R53, respectively) (Ahn et al., 2004; Maclean and Ali,

2003; Quevillon-Cheruel et al., 2004; Viola et al., 2004,

respectively). Maclean and Ali (2003) reports the SpCS

structure (which shares with MtCS, on their amino acid

Table 1

Crystallographic data, reﬁnement statistics, and analysis of the quality

from MtCS

1° Crystal data

Unit-cell parameters a = b = 129.74, c = 159.77

Space group P6

Number of measurements with I > 2 r (I) 92610

Number of independent reﬂections 19341

Completeness (%) (outermost shell) 97.90 (97.9)

sym

(%) (outermost shell) 5.40 (16.50)

Highest resolution shell (A

) 2.94–2.65

2° Reﬁnement data

Resolution range (A

) 52.93–2.65

Reﬂections used for reﬁnement 20093

Number of atoms 3083

Number of residues 388

Number of O atoms 243

Final R-factor (%)

16.2

Final R-free ﬁnal (%)

22.1

Correlation coeﬃcient F

À F

(%) 95.6

B values (A

)

Main chain 33.81

Side chain 36.05

3° Quality of structure

Observed RMSD from ideal geometryl

Bonds lengths (A

) 0.021

Bonds angles (°) 2.655

G-factor

Torsion angles À0.49

Covalent geometry À0.15

Overall À0.27

3D proﬁle

(S) = 170.39; IS = 177.45;

S/IS = 0.96 IS

Ramachandran plot (%)

Favorable 92.2

Additional allowed 7.5

Generously allowed 0.0

Disallowed 0.3

sym

= 100 PjI (h)_ÆI (h)æj/PI(h) with I (h), observed intensity and

ÆI (h)æ, mean intensity of reﬂection h overall measurement of I (h).

R-factor = 100 ·

(jF

obs

À F

calc

j)/

obs

), the sums being taken over

all reﬂections with F/r(F)>2r ( F).

R-free = R-factor for 10% of the data that were not included during

crystallographic reﬁnement.

B values = average B values for all non-H atoms.

Ideally, scores should be above À0.5. Values below À1.0 may need

investigation.

Total score is the sum of the 3D–1D scores (statistical preferences) of

each residue present in protein. Ideal score S

Ideal

= exp(À0.83 + 1.008 ·

ln(L)); where L is the number of amino acids. S

Ideal

score is compatibility

of the sequence with their 3D structure. It is obtained from total score/

ideal score. S

Ideal

score above 0.45S

Ideal

134 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

sequence, 46 and 52% of identity and similarity, respective-

ly) in two conformational states. In the monomer A, the

loop 22 is opened over the FMN binding site and in the

other monomers (B, C, and D), the loop 22 is closed over

FMN binding site. Fig. 6 (A to D) shows the superposition

of Ca atoms of the MtCS with others solved structures.

Fig. 6 demonstrates that MtCS core is similar to other

structures. However, some dislocated regions can be

observed, mainly the H7 in MtCS. In the superposition

of the MtCS with HpCS, it can be observed that the H13

in MtCS is in opposing direction. Furthermore, superposi-

tion of MtCS against other CS structures indicates highest

RMSD for HpCS and ScCS, possibly due to low identity of

these structures with MtCS (31 and 33% of identity, respec-

tively). On the other hand, the AaCS and SpCS structures

present higher identity with MtCS (45 and 46% of identity,

respectively) and therefore a major structural resemblance.

Furthermore, the MtCS structure presents highest resem-

blance with the monomer B, C, and D from SpCS (which

are in the closed conformation—Fig. 6D). The residues

of FMN binding site of complex HpCS:FMN also present

diﬀerent conformations when compared with MtCS. The

position of residues of active site of HpCS resemble with

the structure of monomer A of SpCS . The FMN binding

site from AaCS and ScCS structures are not deﬁned and

they can not be compared with that region from MtCS.

Fig. 2. MtCS monomer structure. (A) Representation of the monomer MtCS secondary structure elements. (B) MtCS secondary structure representation.

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 135

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

This information lead us conclude that in MtCS, the struc -

ture may be in its closed conformation, and in the presence

of FMN, the loop 29 (in MtCS) moves away opening the

EPSP binding site. In the EPSP bound state, the loop 29

from MtCS molecule must close over the ligand. The SpCS

structure has conﬁrmed that the binding of FMN and

EPSP is ordered (Maclean and Ali, 2003), so as Macheroux

et al. (1996, 1998) observed in experiments for EcCS that

oxidized FMN which has a K

of about 30 lM, and

decreases to around 20 nM in the presence of EPSP. There-

fore, it is clear that the binding of EPSP blocks any possible

exit of FMN from the active site, and the induced structur-

al changes in the active site allow FMN to make further

interactions with SpCS (Maclean and Ali, 2003).

The residues of FMN and EPSP binding active site were

superposed with the same residues in the two conforma-

tional states of SpCS (Figs. 7A and B). The positions of

many residues are conserved, but there are signiﬁcant dif-

ferences in the positions of some residues, mainly the

Arg46 and Arg341 at the EPSP binding site and in

Arg110 and Ile317 at FMN binding site. The His11, an

essential residue to CS catalysis (Kitzing et al., 2004) is con-

served in several sequences (Fig. 4) and presents similar

conformation in MtCS and SpCS (in the both SpCS con-

formational states). The His115 in the MtCS structure,

another conserved and important residue to CS catalyzed

reaction (Kitzing et al., 2004), also keeps its position, when

compared with the SpCS open structure. This same resi-

due, when compared to the closed structure of SpCS pre-

sents dislocated approximately 30° outward from the

EPSP binding site (Fig. 7). In SpCS, the His110 (equivalent

the His115 in MtCS) demonstrated therefore some confor-

mational ﬂexibility and interacts with FMN in the open

form but with EPSP in the closed form (Maclean and

Ali, 2003). The closing of the loop 22 on the closed

structure of SpCS can be the cause of this displacement.

The Arg3 41, located in the loop 29 in MtCS (relative to

loop the 22 in SpCS) present the Ca closer to the closed

Fig. 3. MtCS tetrameric structure. The ﬁgure represents the MtCS tetrameric structure with its secondary structure content.

Fig. 4. Sequence alignment of several sequences of CS. Alignment was performed with MultAlin program (Corpet, 1988). Numbering is relative to MtCS

primary sequence. Residues shown in red have a correlation of at least 90% with the consensus sequence and are therefore highly conserved. Residues

shown in blue have 50% or more correlation with the consensus sequence and are therefore moderately conserved. Residues in (——) boxes are relative the

active site. Residues in (- - -----)boxes are relative the dimerization interface. Residues in (ÁÁÁÁÁÁÁÁÁ) boxes are relative the tetramerization interface.

Myctub (Mycobacterium tuberculosis); Aquaeo (Aquifex aeolicus); Bacsub (Bacillus subtilis); Laclac (Lactococcus lactis); Strpne (Streptococcus

pneumoniae); Lismon (Listeria monocytogenes); Staaur (Staphylococcus aureus); Brumel (Brucella melitensis); Esccol (Escherichia coli); Saltyp (Salmonella

typhimurium); Yerpes (Yersinia pestis); Haeinf (Haemophilus inﬂuenzae); Xilfas (Xylella fastidiosa) Neimen (Neisseria meningitidis); Camjej (Campylobacter

jejuni); Clamur (Chlamydia muridarum).

136 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 137

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

structure of SpCS, but its side chain is coming from other

direction, occupying the EPSP binding site while in closed

structure of SpCS, it presents dislocated outward from the

site. This eﬀect occurs probably due to the EPSP binding at

its site, therefore in HpCS, this arginine residue is moved

away from the EPSP binding site (data not shown). The dif-

ferent orientations of important residues of active site of

MtCS, as Arg341, Ser342, and Asp343, when compared with

the structure of SpCS, likely occurs due to apo state of MtCS.

The arginine residues 112 and 139, which are involved in

the formation of binding pocket of EPSP by the aliphatic

portions from MtCS structure do not present signiﬁcant

modiﬁcations when compared with SpCS. On the other

hand, the Arg40 of the MtCS is more distended upon the

EPSP binding site than the same arginine in the two SpCS

conformational states, showing that the Arg40 presents

some ﬂexibility. This observation can also be justiﬁed by

absence of the EPSP in the MtCS structure. The arginine

residues 46 and 49 that are important to the EPSP binding

(Maclean and Ali, 2003) are localized at the loop 5, which

was not determined with precision probably due to the

absence of interactions of protein with EPSP.

The largest diﬀerences in FMN binding site are

observed in the position of Arg110 and Thr319, where

Fig. 5. MtCS dimeric structure and interface. (A) Figure shows the molecular surface of MtCS dimer formed by monomer A (pink) and D (lime). (B)

Representation of the molecular surface on the interface between monomers A (pink) and B (wheat). (C) Representation of the molecular surface on the

interface between monomers A (pink) and D (slate). The ﬁgures were generated for Pymol program (DeLano, 2002). (For interpretation of the references

to color in this ﬁgure legend, the reader is referred to the web version of this paper.)

138 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

in MtCS structure, their lateral chains are shifted out-

ward from the FMN binding site. The side chains of

Ile37 and Lys315 in MtCS are occupying the FMN bind-

ing site. The Ser342 and Asp343 also present diﬀerent

orientations when compared with SpCS, may be due to

MtCS to be in apo state.

Fig. 6. Superposition of Ca between MtCS and others CS. (A) from A. aeolicus; (B) H. pylori; (C) S. cerevisae, and (D) S. pneumoniae. Structure in light

blue is MtCS, and in light green is monomer B of SpCS. Structure in dark green is monomer A of SpCS; Region in yellow is loop 22 of SpCS of the

monomer B; region in light red is loop 22 of SpCS of the monomer A; Structure in violet is loop 29 of MtCS. H7 and H13 is referent of MtCS. (For

interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this paper.)

Fig. 7. Superposition of residues of the (A) FMN binding site and (B) EPSP binding site. Residues in black are from MtCS, yellow are from monomer B

of the SpCS and in orange is from monomer A of the SpCS. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the

web version of this paper.)

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 139

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.4. Quaternary structure in solution

Gel ﬁltration column was performed to determine the

native molecular weight of MtCS. The results showed that

MtCS is a dimer in solution.

The hydrodynamic properties of the MtCS were estab-

lished by both sedimentation velocity and sedimentation

equilibrium. The samples did not present apparent aggre-

gation and it shows monodispersive behavior. The sedi-

mentation coeﬃcient s normalized to the standard

sedimentation coeﬃcient s

20,w

and extrapolated at 0 mg/

mL (s

20;w

Þ is 4.60 ± 0.02 S which is in accordance with a

particle with molec ular mass (MM) of 80–100 kDa, and

of asymmetrical shape. Furthermore, considering the

diﬀusion coeﬃcient at standard conditions (3.7–3.8 ·

À7

seg

À1

) determined by dynamic light scattering

and by the application of the Eq. (1), a MM of MtCS

was 80 ± 2 kDa, a value closer of the expected for its dimer

(MM of monomer is 41.7 kDa). The MM derived from the

sedimentation equilibrium was 83 ± 1 kDa and it is also in

accordance with the MM expected for the MtCS dimer.

However, the resultant plot (Fig. 8A, inset) suggests that

20,w

increases with the MtCS con centration, instead to

reduce with the viscosity and molecular crowd eﬀect (Laue,

2001). One hypothesis that can explain this behavior of

MtCS is a rapid dimer–tetramer equilibrium ( Rowe,

1977), which should be concentration dependent.

Furthermore, the results of sedimentation equilibrium

also suggest the existence of dimer–tetramer equilibrium

(Fig. 8B). The association constant determined for sedi-

mentation equilibrium is approxim ately 20 mM

À1

, and

therefore, the dissociation constant—K

—about 50 M.

Thus, due to very high K

, probable the interactions

that form the tetramer are few or/and very weak in

these experimental conditions. SpCS, on the other hand,

also shows dimer–tetramer equilibrium with dissociation

constant of 0.8 lM(Maclean and Ali, 2003). So, we can

conclude that in solution, the MtCS is predominantly

dimer while SpCS is tetramer, despite of their structural

resemblance.

The comparison between the predicted s

20,w

for MtCS

crystal structures and the s

max

for a globular protein rela-

tive MM correspondents (Table 2) shows that the MtCS

in solution can have an asymmetric shape. Through of

the sedimentation equilibrium and sedimentation velocity

data of MtCS, we can considerate the predicted s

0,w

for

future analysis. But, the experimental s

20;w

data for MtCS

do not agree with the predicted data for crystallographic

dimer formed for monomers A/D or B/C of this structure.

These ambiguous results suggest that, in solution, the

dimer of MtCS is slightly diﬀerent from the crystallograph-

ic dimers (A/D or B/C) that form the tetramer, presenting

a more asymmetrical shape.

These diﬀerences of the oligomeric states observed for

MtCS when comp ared with other CS structures may not

5.0

0.1

5.1

Absorbance (AU)

Residuals

50.0-

50.0

06.655.605.654.604.653.6

)mc(suidaR

mpr000,8

mpr000,01

mpr000,21

50.0

01.0

51.0

02.0

52.0

03.0

53.0

g(s*)

0.20.010.80.60.40

Lm/gm6.0

Lm/gm8.0

Lm/gm0.1

1.00.80.60.40.20

5.20

5.00

4.80

4.60

4.40

CS )Lm/gm(

CS S)20.0±06.4(=

s (S)

Fig. 8. Sedimentation velocity experiments. (A) Sedimentation velocity experiments: The g (s*) distributions were ﬁtted using the Origin (Microcal

Software) with a Gaussian giving apparent sedimentation coeﬃcients s* for MtCS at diﬀerent protein concentrations (see Section 2 for details). Inset: plots

of s

20,w

versus protein concentration ﬁtted by linear regression to calculate the s

20;w

: 4.60 ± 0.04 Svedberg. Despite the data were well ﬁtted by only one

Gaussian suggesting the existence of only one MtCS specie in solution, the positive slop suggests that the MtCS undergoes an oligomerization process. (B)

Sedimentation equilibrium experiments. The ﬁgure shows the best ﬁts of experimental data for 450 lg/mL of MtCS at 8000, 10000, and 12000 rpm with

the self-association methods (see Section 2). The random distribution of the residuals (bottom panel) indicates that the ﬁt is satisfactory. The ﬁtness of the

data obtained at other protein concentration presented similar results (data not shown). Sedimentation equilibrium data of MtCS agree with a dimer

structure with 83 ± 1 kDa in equilibrium with a tetrameric form with an aﬃnity constant of 20 ± 5 mM

À1

Table 2

Comparison between the predicted s

20,w

for MtCS crystal structure and

the s

max

estimated for a globular protein

Oligomeric state s

20,w

(S) s

max

(S)

Monomer 3.18 ± 0.05 4.20

Dimer 5.42 ± 0.06 6.70

Tetramer 9.02 ± 0.06 10.70

140 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

be justiﬁed by number of saline bridges and hydrogen

bonds in the dimer interface area, which form the tetramer.

Table 3 shows the analysis of the tetramerization interface

for ﬁve microorganisms. It can be observed that the num-

ber of saline bridges varies is between 0 and 6 and the num-

ber of hydrogen bonds is between 6 and 22 in the analyzed

structures. Analysis of contact area between dimer s reveals

that all structures present contact interfaces above of

4000 A

, but SpCS present a tetramerization interface of

12,132.0 A

. Therefore contacts between dimers that form

the tetramer of SpCS are larger than those observed in

the other structures. The most striking diﬀerence between

MtCS and other CS structures is the area percentage of

polar atoms, which participate in the interface regions.

MtCS present 53% of hydrophobic area (referent to the

carbon atoms) while the other CS struc tures present a min-

imum 60%. Therefore, we can aﬃrm that the interface

between the dimers that forms tetramer in MtCS is more

polar an d hydrophilic when compared with other CS struc-

tures. How ever, this result can not justify the predomi-

nance of dimer in solution, but it shows that the

hydrophobic eﬀect has a minor participation in the forma-

tion of tetramer in MtCS structure. On the other hand, in

some way, a minor hydrophobic surface can contribute to

predominance of dimer in solut ion. Furthermore, a

hypothesis to explain the low presence of tetramer in solu-

tion is that MtCS tetramerization may involve small initial

contacts, which lead to some conformational changes

increasing the contact area between dimers, resulting on

MtCS tetramerization. The initial contacts can be concen-

tration-dependent and/or still due to the presence of some

ligand, which may trigger the conformational changes

resulting in the oligomerization (Fig. 9).

4. Conclusion

Here, we report the ﬁrst crystal structure and analysis in

solution of the chorismate synthase from M. tuberculosis.

The CD analyses show that MtCS secondary structures

content agree with the CS structures from other organisms.

Furthermore, MtCS thermal stability experiments show

two well deﬁned transitions, suggesting a complex unfold-

ing mechanism. The monomeric and tetrameric structures

of MtCS are similar to other CS structures. Many of con-

served residues in the primary sequence of the active site of

CS present conserved conformations in all structures

solved, but some residues have large conformational

changes when compared to other structures. All diﬀerences

and resemblances observed in the MtCS structure when

compared to the SpCS and HpCS structures may be due

to absence of any ligand in the active site from MtCS.

The determination of MtCS structure in the presence of

FMN, EPSP, and chorismic acid may conﬁrm these evi-

dences. The intricate association between two monomer s

lanoitamrofnoC

segnahC

hgiH

noitartnecnoc

gnidnibdnagiL

dimer

remarteT

Fig. 9. MtCS oligomerization mechanism. Initial contacts on the MtCS dimer can be induced for protein concentration or ligand binding, would lead to

some conformational changes increasing the dimers contact area and resulting on MtCS tetramerization.

Table 3

Contact analysis of interface area between the dimers that form the CS tetramer

Structure Mycobacterium

tuberculosis

Streptococcus

pneumoniae

Helicobacter

pylori

Aquiﬂex

aelicus

Saccharomyces

cerevisae

Overall contact surface (A

) 5027.0 12132.0 5305.0 4425.0 5745.0

Contact surface apolar (A

) (%) 2661.0 (53) 7303.0 (60) 3369.0 (63) 2658.0 (60) 3653.0 (63)

Contact surface hydrophilic polar (A

) (%) 2338.0 (47) 4825.0 (40) 1936 (37) 1737.0 (40) 2092.0 (37)

Saline bridges (number) 2 6 1 2 0

Hydrogen bonds (number) 24 40 26 27 22

The contact surface was calculated by AREA/AREAIMOL (CCP4, 1994) and saline bridges and hydrogen bonds were identiﬁed for protein–protein

interaction (http://www.biochem.ucl.ac.uk/bsm/PP/server), on the basis of principles described by Jones and Thornton (1996).

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 141

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

of the MtCS leads to dimer organization, connecting to

form a tetramer, which may be described as a dimer of

dimers. In solution, the MtCS must be mainly in dimeric

state that is slightly diﬀerent from crystallographic dimer,

which forms the tetramer. Analysis of all CS led us to con-

clude that all CS must present quaternary structure in solu-

tion in the dimeric or tetr americ forms or in equilibrium of

them.

Despite of the CS structure presented here is in apo

form, the structural data obtained are enough relevant.

This is the ﬁrst structure in the apo form that presents

almost all residues of active site, excepting the residues

49–52. In this form, it is possible to observe several move-

ments of the important residues of active site of CS. Fur-

thermore, this is the ﬁrst CS predominantly dimeric in

structure, which can be due to the amount of interaction

between two dimers, which form the tetrameric structure.

The PDB accession code for the apo crystal structure of

Chorismate synthase from Mycobacterium tuberculosis is

1ZTB.

Acknowledgments

We thank Fundac¸a

o de Amparo a

Pesquisa do Estado de

o Paulo (SMOLBNet, Proc. 01/07532-0 and 03/12472-2),

Conselho Naci onal de Desenvolvimento Cientı

ﬁco e

Tecnolo

gico and Coordenac¸a

o de Aperfeic¸oamento de

Pessoal de Nı

vel Superior for fellowships and ﬁnancial

support. We thank the LNLS technical staﬀ for assistance.

References

Ahn, H.J., Yoon, H.J., Lee, B., Suh, S.W., 2004. Crystal structure of

chorismate synthase: a novel FMN-binding protein fold and function-

al insights. J. Mol. Biol. 336, 903–915.

Azevedo Jr., W.F., Canduri, F., Oliveira, J.S., Basso, L.A., Palma, M.S.,

Pereira, J.H., Santos, D.S., 2002. Molecular model of shikimate kinase

from Mycobacterium tuberculosis. Biochem. Biophys. Res. Commun.

295 (1), 142–148.

Basso, L.A, Silva, L.H.P., Fett-neto, A.G., Azevedo Jr., W.F., Moreira,

I.S., Calixto, J.B., Astolﬁ Filho, S., Santos, R.R., Soares, M.B.P.,

Santos, D.S., 2005. The use of biodiversity as source of new chemical

entities against deﬁned molecular targets for treatment of malaria,

tuberculosis, and T-cell mediated diseases—A review. Memo

rias do

Instituto Oswaldo Cruz 100 (6), 475–506.

Bentley, R., 1990. The shikimate pathway—metabolic tree with many

branches. Crit. Rev. Biochem. Mol. Biol. 25, 307–384.

hm, G., Muhr, R., Jaenicke, R., 1992. Quantitative analysis of protein

far UV circular dichroism spectra by neural networks. Protein Eng. 5,

191–195.

Bornemann, S., Ramjee, M.K., Balasubramanian, S., Abell, C., Coggins,

J.R., Lowe, D.J., Thorneley, R.N., 1995. Escherichia coli chorismate

synthase catalyzes the conversion of (6S)-6-ﬂuoro-5-enolpyruvylshik-

imate-3-phosphate to 6-ﬂuorochorismate. Implications for the enzyme

mechanism and the antimicrobial action of (6S)-6-ﬂuoroshikimate. J.

Biol. Chem. 270, 22811–22815.

Bornemann, S., Lowe, D.J., Thorneley, R.N., 1996. The transient kinetics

of Escherichia coli chorismate synthase: substrate consumption,

product formation, phosphate dissociation, and characterization of a

ﬂavin intermediate. Biochemistry 35 (30), 9907–9916.

Bornemann, S., Lawson, D.M., Thorneley, R.N., 2003. A branch point in

chorismate synthase research. Structure 11, 1463–1465.

Bru

nger, A.T., 1992. X-PLOR, a System for Crystallography and NMR.

Yale University Press, London.

Collaborative Computational Project No. 4., 1994. The CCP4 suite:

programs for proteins crystallography, Acta Crystallogr., D 50, 760–763.

Corpet, T.F., 1988. Multiple sequence alignment with hierarchical

clustering. Nucleic Acid Res. 16 (22), 10881–10890.

Dosselaere, F., Vanderleyden, J.A., 2001. A metabolic node in action:

chorismate-utilizing enzymes in microorganisms. Crit. Rev. Microbiol.

27, 75–131.

DeLano, W.L., 2002. The PyMOL Molecular Graphics System. CA

DeLano Scientiﬁc.

Dias, M.V.B., Ely, F., Canduri, F., Pereira, J.H., Frazzon, J., Basso, L.A.,

Palma, M.S., De Azevedo, W.F., Santos, D.S., 2004. Crystallization

and preliminary X-ray diﬀraction analysis of chorismate synthase

Mycobacterium tuberculosis. Acta Crystallogr., D. Biol. Crystallogr.

60, 2003–2005.

Duncan, K., 2004. Identiﬁcation and validation of novel drug targets in

tuberculosis. Curr. Pharm. Des. 10 (26), 3185–3194.

Edelhock, H., 1967. Spectroscopic determination of tryptophan and

tyrosine in protein. Biochemistry 6, 1948–1954.

Fitzpatrick, T.B., Killer, P., Thomas, R.M., Jelesarov, I., Amrhein, N.,

Macheroux, P., 2001. Chorismate synthase from the hyperthermophile

thermotoga maritima combines thermostability and increased rigidity

with catalytic and spectral properties similar to mesophilic counter-

parts. J. Biol. Chem. 276, 18052–18059.

Gandy, M., Zumla, A., 2002. The resurgence of disease: social and

historical perspectives on the ’new’ tuberculosis. Soc. Sci. Med. 55 (3),

385–396.

Garcı

a de la Torre, J., Huertas, M.L., Carrasco, B., 2000. Calculation of

hydrodynamic properties of globular proteins from their atomic-level

structure. Biophys. J. 78, 719–730.

Gill, S.C., von Hippel, P.H., 1989. Calculation of protein extinction

coeﬃcients from amino acid sequence data. Anal. Biochem. 182,

319–326.

Goldberg, R.J., 1953. Sedimentation in the ultracentrifuge. J. Phys. Chem.

57, 194–202.

Hill, R.K., Newkome, G.R., 1969. Stereochemistry of chorismic acid

biosynthesis. J. Am. Chem. Soc. 91, 5893–5894.

Johnson, M.L., Correia, J.J., Yphantis, D.A., Halvorson, H.R., 1981.

Analysis of data from the analytical ultracentrifuge by nonlinear least-

squares techniques. Biophys. J. 36, 575–588.

Jones, S., Thornton, J.M., 1996. Principles of protein–protein interactions.

Proc. Natl. Acad. Sci. USA 93, 13–20.

Jones, T.A., Zou, J.Y., Cowan, S.W., 1991. Improved methods for

building protein models in electron density maps and the location of

errors in these models. Acta Crystallogr., A 47, 110–119.

Kitzing, K., Auweter, S., Amrhein, N., Macheroux, P., 2004. Mechanism

of chorismate synthase. Role of the two invariant histidine residues in

the active site. J. Biol. Chem. 279 (10), 9451–9461.

Laskowski, R.A., MacArthur, M.W, Moss, D.S., Thornton, J.M., 1994.

PROCHECK v.3.0—Program to Check the Stereochemistry Quality

of Protein structures—Operating Instructions.

Laue, T.M., 2001. Biophysical studies by ultracentrifugation. Curr. Opin.

Struct. Biol. 11, 579–583.

Leslie, A.G.W., 1992. MOSFLM version 6.11 for processing image plate

and CCD data.

Macheroux, P., Bornemann, S., Ghisla, S., Thorneley, R.N., 1996. Studies

with ﬂavin analogs provide evidence that a protonated reduced FMN

is the substrate-induced transient intermediate in the reaction of

Escherichia coli chorismate synthase. J. Biol. Chem. 271 (42), 25850–

25858.

Macheroux, P., Schonbrunn, E., Svergun, D.I., Volkov, V.V., Koch,

M.H., Bornemann, S., Thorneley, R.N., 1998. Evidence for a major

structural change in Escherichia coli chorismate synthase induced by

ﬂavin and substrate binding. Biochem. J. 335, 319–327.

Macheroux, P., Schmid, J., Amrhein, N., Schaller, A., 1999. A unique

reaction in a common pathway: mechanism and function of choris-

mate synthase in the shikimate pathway. Planta 207, 325–334.

142 M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Maclean, J., Ali, S., 2003. The structure of chorismate synthase reveals a

novel ﬂavin binding site fundamental to a unique chemical reaction.

Structure 11, 1499.

McRee, D.E., 1999. XtalView/Xﬁt—a versatile program for manipulating

atomic coordinates and electron density. J. Struct. Biol. 125, 156–165.

Murshudov, G.N., Vagin, A.A., Dodson, E.J., 1997. Reﬁnement of

macromolecular structures by the maximum-likelihood method. Acta

Crystallogr., D 53, 240–255.

Navaza, J., 2001. Implementation of molecular replacement in AMoRe.

Acta Crystallogr. D. Biol. Crystallogr. 57 (Pt. 10), 1367–1372.

Pereira, J.H., Canduri, F., Oliveira, J.S., Silveira, N.J.F., Basso, L.A,

Palma, M.S., Azevedo Jr., W.F., Santos, D.S., 2003. Structural

bioinformatics study of EPSP synthase from Mycobacterium tubercu-

losis. Biochem. Biophys. Res. Commun. 312 (3), 608–614.

Pereira, J.H., Oliveira, J.S., Canduri, F., Dias, M.V.B., Palma, M.S.,

Basso, L.A., Santos, D.S., Azevedo Jr., W.F., 2004. Structure of

shikimate kinase from Mycobacterium tuberculosis reveals the binding

of shikimic acid. Acta Crystallogr. D. Biol. Crystallogr. 60 (Pt. 12, Pt.

2), 2310–2319.

Polikarpov, I., Perles, L.A., de Oliveira, R.T., Oliva, G., Castellano, E.E.,

Garratt, R.C., Craievich, A., 1998. Set-up and experimental parameters

of the protein crystallography beam line at the Brazilian National

Synchrotron Laboratory. Nucl. Instrum. Methods Phys. Res. A 405,

159–164.

Quevillon-Cheruel, S., Leulliot, N., Meyer, P., Graille, M., Bre-

mang, M., Blondeau, K., Sorel, I., Poupon, A., Janin, J., van

Tilbeurgh, 2004. Crystal structure of the bifunctional chorismate

synthase from Saccharomyces cerevisiae. J. Biol. Chem. 279,

619–625.

Rowe, A.J., 1977. The concentration dependence of transport processes: a

general description applicable to the sedimentation, translational

diﬀusion, and viscosity coeﬃcients of macromolecular. Biopolymers

16, 2595–2611.

Silveira, N.J.F., Ucho

a, H.B., Pereira, J.H., Canduri, F., Basso, L.A.,

Palma, M.S., Santos, D.S., Azevedo Jr., W.F., 2005. Molecular models

of protein targets from Mycobacterium tuberculosis. J. Mol. Model. 11

(2), 160–166.

Staﬀord, W.F., 1994. Boundary analysis in sedimentation velocity

experiments. Methods Enzymol. 240, 478–501.

Ucho

a, H.B, Jorge, G.E., Silveira, N.J.F., Ca

mera, J.C., Canduri, F.,

Azevedo Jr., W.F., 2004. Parmodel: a web server for automated

comparative modeling of proteins. Biochem. Biophys. Res. Commun.

325 (4), 1481–1486.

Viola, C.M., Saridakis, V., Christendat, D., 2004. Crystal structure of

chorismate synthase from aquifex aeolicus reveals a novel beta

alpha beta sandwich topology. Proteins: Struct. Funct. Genet. 54,

166–169.

Welch, G.R., Cole, K.W., Gaertner, F.H., 1974. Chorismate synthase of

Neurospora crassa: a ﬂavoprotein. Arch. Biochem. Biophy. 165, 505–

518.

WHO, 2004. World Health Organization. WHO Report, Geneva, Swit-

zerland. WHO/CDS/TB/2004.331.

M.V.B. Dias et al. / Journal of Structural Biology 154 (2006) 130–143 143

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.3 Molecular models of tryptophan synthase from Mycobacterium tuberculosis

complexed with inhibitors

Marcio V. B. Dias; Fernanda Canduri; Nelson J. F. da Silveira; Clarissa M. Czekster; Luiz

A. Basso; Mário S. Palma; Diógenes S. Santos; Walter F. de Azevedo Jr.Cell

Biochemistry and Biophysics (ISSN 1085-9195), v. 44, p. 375-384, 2006.

Neste trabalho é apresentada a modelagem molecular da proteína triptofano sintase

de M. tuberculosis em complexo com 6 inibidores análogos do estado de transição desta

enzima.

Para a modelagem molecular da triptofano sintase de M. tuberculosis, foi utilizado

como ponto de partida (templates), as estruturas cristalográficas da triptofano sintase de S.

typhimurium em complexo com os respectivos inibidores. Para realizar a modelagem

molecular foi utilizado o programa MODELLER (SALI; BLUNDELL, 1993). As duas

subunidades da triptofano sintase foram modeladas separadamente e foram gerados 1000

modelos para cada subunidade e para cada complexo. Os modelos finais foram

selecionados baseando-se em análise estereoquímica. A análise da afinidade de ligação

entre o complexo proteína:ligante foi realizada baseando-se na contribuição individual de

afinidade de ligação de cada átomo do ligante, utilizando o programa SCORE (WANG et

al., 1998). Além disso foi levado em consideração o número de ligações de hidrogênio e a

área de contato entre o ligante e a proteína. Os modelos gerados apresentam boa qualidade

estereoquímica e são bastante similares a seus templates. A comparação dos sítios ativos

dos complexos mostra que muitas interações de hidrogênio são conservadas entre os

modelos obtidos e os templates, assim como muitas interações entre os próprios inibidores,

apesar de estes apresentarem algumas interações únicas. Análise da área de contato entre a

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

proteína e seus inibidores mostra que a triptofano sintase de M. tuberculosis apresenta uma

maior área de contato do que a de S. typhimurium, entretanto, análise das interações

favoráveis para a ligação, mostra que a triptofano sintase S. typhimurium apresenta uma

afinidade de ligação (score) maior do que a triptofano sintase de M. tuberculosis. Desta

forma, algumas alterações nos átomos dos inibidores estudados podem aumentar a

afinidade destes pela triptofano sintase de M. tuberculosis.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

RIGINAL

RTICLE

All rights of any nature whatsoever reserved.

1085-9195/(Online)1559-0283/06/44:375–384/$30.00

Cell Biochemistry and Biophysics

375 Volume 44, 2006

INTRODUTION

Tryptophan synthase (TRPS) is a target for develop-

ment of vaccines and drugs. The interaction between

Mycobacterium tuberculosis TRPS (MtTRPS) and in-

hibitors derived from the arylthioalkyl-phosphonated

inhibitors (e.g., indole 3-propanol phosphate [IPP]) is

not known. In this work, we show that IPP and their

analogs (or derivatives) present strong interaction with

MtTRPS and that they can inhibit this protein efﬁciently.

The development of new therapies as vaccines and

drugs is vital in developing country, where a majority of

the population die because of infectious diseases, such

as tuberculosis, AIDS, cholera, and malaria, among oth-

ers (1). Among these diseases, tuberculosis deserves

special attention because it is the main cause of death in

these countries. M. tuberculosis, the main causative

agent of tuberculosis, infects approx 8 million individu-

als and causes 2 million deaths annually (2).

Bacteria, fungi, apicomplexan parasites, and plants syn-

thesize aromatic compounds, and this synthesis is essential

for their survival (3,4). Therefore, the enzymes responsible

for synthesis of aromatic compounds are considered targets

Molecular Models of Tryptophan Synthase

From Mycobacterium tuberculosis Complexed With Inhibitors

Marcio Vinicius Bertacine Dias,

Fernanda Canduri,

Nelson José Freitas da Silveira,

Clarissa Melo Czekster,

Luis Augusto Basso,

Mário Sérgio Palma,

Diógenes Santiago Santos,

* and Walter Filgueira de Azevedo, Jr.

Programa de Pós-Graduação em Biofísica Molecular-Departamento de Física, UNESP, São José do Rio Preto, Brazil;

Departamento de Morfoﬁsiologia—CCBS—UFMS, Campo Grande-MS, 79070-900, Brazil;

Rede Brasileira de Pesquisas em

Tuberculose, Grupo de Microbiologia Molecular e Funcional, Centro de Biotecnologia, UFRGS, Porto Alegre-RS, 91501-970,

Brazil;

Faculdade de Biociências—PUCRS, Porto Alegre-RS, 90619-900, Brazil;

Laboratory of Structural Biology and

Zoochemistry, CEIS/Department of Biology, Institute of Biosciences, UNESP, Rio Claro, SP 13506-900, Brazil; and

Centro de Pesquisas em Biologia Molecular e Funcional/PUCRS, Avenida Ipiranga, 6681,

Tecnopuc, Partenon 90619-900, Porto Alegre, RS, Brazil

Abstract

The development of new therapies against infectious diseases is vital in developing countries. Among infec-

tious diseases, tuberculosis is considered the leading cause of death. A target for development of new drugs is

the tryptophan pathway. The last enzyme of this pathway, tryptophan synthase (TRPS), is responsible for con-

version of the indole 3-glycerol phosphate into indol and the condensation of this molecule with serine-produc-

ing tryptophan. The present work describes the molecular models of TRPS from Mycobacterium tuberculosis

(MtTRPS) complexed with six inhibitors, the indole 3-propanol phosphate and ﬁve arylthioalkyl-phosphonated

analogs of substrate of the α-subunit. The molecular models of MtTRPS present good stereochemistry, and the

binding of the inhibitors is favorable. Thus, the generated models can be used in the design of more speciﬁc

drugs against tuberculosis and other infectious diseases.

Index Entries: Tryptophan synthase; Mycobacterium tuberculosis; molecular modeling; drug design; structural

bioinformatics.

Author to whom all correspondence and reprint requests should

be addressed. E-mail: walter[email protected] or diogenes@ pucrs.br

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

for development of new therapies, because these com-

pounds are absent in animals. Thus, the inhibition of amino

acid biosynthesis is a demonstrated mode of action for

important classes of inhibitors. Glyphosate and 6-ﬂuoro-

shikimate, for example, inhibit biosynthesis of aromatic

compounds in bacteria, fungi, apicomplexan parasites, and

plants (5,6). Enzymes involved speciﬁcally in tryptophan

biosynthesis have been proposed as potential herbicide

and vaccine targets (7,8). The last enzyme of the pathway of

tryptophan biosynthesis is TRPS, and the following rea-

sons led to selection of this enzyme as a potential target. For

one, all organisms that synthesize tryptophan are known to

do so by a single route, such that inhibition of this pathway

should dramatically reduce tryptophan levels. TRPS is also

one of the most studied enzymes of those involved in the

biosynthesis of amino acids (9). It catalyzes two distinct

reactions by separate polypeptide chains, referred to as α-

and β-subunits. The α-subunit catalyzes the cleavage of

indole 3-glycerol phosphate (IGP) to yield an indole and

glyceraldehyde-3-phosphate, whereas the β-subunit cat-

alyzes the condensation of the indole with serine to pro-

duce tryptophan (Fig. 1) (10). The three-dimensional (3D)

structure of the TRPS from Salmonella typhimurium

(StTRPS) with several inhibitors has been deposited in the

376 Dias et al.

Cell Biochemistry and Biophysics Volume 44, 2006

Fig. 1. Reaction scheme of the TRPS. (A) Reaction of the α-subunit. (B) Reaction of the β-subunit.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Protein Data Bank (PDB). Among them are IPP (11) and

ﬁve arylthioalkyl-phosphonated transition state analogs

that inhibit the α-subunit (12). IPP is an analog of IGP that

inhibits the α-subunit with a K

value of 15 µM, and it was

the starting point for inhibitor design strategy of the

arylthioalkyl-phosphonate (7). The arylthioalkyl-phos-

phonated inhibitors were designed to mimic the transi-

tion state formed during the α-reaction of the enzyme,

and as expected, they have afﬁnities much greater than

that of the natural substrate IGP or its nonhydrolyzable

analog IPP. These inhibitors are ortho-substituted arylth-

ioalkyl-phosphonate derivatives that have an sp

hybridized sulfur atom, designed to mimic the putative

tetrahedral transition state at the C3 atom of the indole,

and they lack the C2 atom to allow for higher conforma-

tional ﬂexibility (7,12). These phosphonated inhibitors

have been synthesized because of efforts to develop her-

bicidal and antimicrobial agents, and they have been

tested for inhibition of enzyme activity in in vitro assay

and for herbicidal activity in a biological assay (7). These

phosphonated compounds possess enzyme inhibitory

and herbicidal activities with micromolar IC

values (7).

The present work describes six molecular models of

MtTRPS, the complex with IPP, and the complexes with

ﬁve arylthioalkyl-phosphonated analogs bound in α-sub-

unit and pyridoxal 5′- phosphate (PLP) in the β-subunit.

MATERIALS AND METHODS

Molecular Modeling

For modeling of the MtTRPS, we used Parmodel, a Web

server for automated comparative modeling of proteins

(13) based on the program MODELLER (14). The atomic

coordinates of the crystallographic structures of StTRPS

were used as starting models. For the complex with IPP,

we used PDB access code 1QOP (11) and for ﬁve com-

plexes with arylthioalkyl-phosphonated derivatives we

used the following PDB access codes: lC29, IC8V, 1CX9,

1C9D, and 1CW2 (12). The arylthioalkyl-phosphonated

inhibitors are 4-(2-hydroxyphenylthio)-1-butenylphos-

phonic acid, 4-(2-hydroxyphenylthio)butylphosphonic

acid, 4-(2-aminophenylthio) butylphosphonic acid, 4-(2-

hydroxy-5-fluorophenylthio)butylphophonic acid and

4-(2-hydroxyphenylsylﬁnyl)butylphosphonic acid. The

complexes are designated by numbers, with 1 for

TRPS:IPP and 2 to 6 for the ﬁve arylthioalkyl-phospho-

nates, respectively. The chemical structures of inhibitors

are shown in Fig. 2. All the cited compounds bind to the α-

subunit. Moreover, the β-subunit was modeled in complex

with PLP for all of the models. However, the models were

generated separately for the α- and β-subunit. The atomic

coordinates of all water were removed from the TRPS tem-

plates. Several slightly different models can be calculated

by varying the initial structure. In total, 1000 models were

generated for each subunit of tryptophan synthase and

each conformational state, and the ﬁnal models were

selected based on stereochemical quality. Subsequently, we

performed the junction of the two subunits, and the com-

plexes were optimized using MODELLER (14). The opti-

mization of the complexes was carried out by the use of

the variable target function method using methods of con-

Molecular Models of TRPS 377

Cell Biochemistry and Biophysics Volume 44, 2006

Fig. 2. Molecular formulae of TRPS inhibitors. (1) indol-

propanol-phosphate; (2) 4-(2-hydroxyphenylthio)-1-butenyl-

phosphonic acid; (3) 4-(2-hydroxyphenylthio)butylphos-

phonic acid; (4) 4-(2-aminophenylthio)butylphosphonic acid;

(5) 4-(2-hydroxy-5-ﬂuorophenylthio)butylphophonic acid;

and (6) 4-(2-hydroxyphenylsylﬁnyl)butylphosphonic acid.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

jugate gradients and molecular dynamics with simulated

annealing. All modeling processes were performed on a

Beowulf cluster, with 16 nodes (B16/AMD Athlon 1800+;

BioComp, São José do Rio Preto, SP, Brazil).

Analysis of the Model

The overall stereochemical quality of the ﬁnal models

for the complexes of MtTRPS were assessed by the pro-

gram PROCHECK (15). Molecular models were super-

posed using the program LSQKAB from CCP4 (15). The

cutoff for hydrogen bonds and salt bridges was 3.5 Å.

The contact surfaces for the binary complexes were cal-

culated using AREAIMOL and RESAREA (15). The root

mean squared deviation (rmsd) differences from ideal

geometries for bond lengths and bond angles were cal-

culated with X-PLOR (16). The G-factor is essentially just

log-odds score based on the observed distributions of

the stereochemical parameters. It was computed for the

following properties: torsion angles (the analyses pro-

vided the observed distributions of ϕ-ψ, χ

–χ

, χ–1, χ–3,

χ–4, and

values for each of the 20 amino acid types)

and covalent geometry (for the main-chain bond lengths

and bond angles). These average values were calculated

using PROCHECK (15). The 3D proﬁle measures the

compatibility of a protein model with its sequence, and

it was calculated using Verify 3D program (17,18).

Analysis of Binding Afﬁnity

of Protein–Ligand Complexes

For an estimative the absolute binding afﬁnity of the

protein–ligand complexes, we used the program SCORE

(19). According to this method, the binding afﬁnity of the

ligand can be decomposed to the contribution of individ-

ual atoms. Each ligand atom obtains a score, called the

atomic binding score, indicating its role in the binding

process. The program reads the structure, assigns atom

types and parameters, performs the calculation, and gives

the dissociation constant of the given protein–ligand com-

plex. The computational results are outputted into a text

ﬁle in which the detailed information of each ligand atom,

including the atomic binding score, is tabulated.

RESULTS AND DISCUSSION

The sequence alignment of StTRPS (template) with

MtTRPS (target) is shown in Fig. 3. The sequence of α-

subunit of MtTRPS shows 27% of identity with the

sequence of α-subunit of StTRPS, whereas the sequence

of β-subunit of MtTRPS shows 53% identity with the

sequence of β-subunit of StTRPS.

Quality of the Model

Ramachandran plots for the six StTRPS structures

solved by crystallography were generated to compare

the overall stereochemical quality of six MtTRPS mod-

els. Analysis of the Ramachandran plots indicated that

the models present a minimum of 93.4% of the residues

in the most favored regions and a maximum of 0.2% of

the residues in the disallowed regions, whereas struc-

tures solved by crystallography present a minimum of

91.6% of the residues in the most favored regions and

also a maximum of 0.2% of the residues in the disal-

lowed regions (Table 1). The Verify 3D values, the aver-

age G-factor, and rmsd values of bond lengths and

bond angles are shown in Table 2. From analysis of the

overall stereochemical quality of the molecular models,

we feel that it is appropriate for structural studies.

Overall Description

The structures of six molecular models were super-

posed with their templates. Only Cα was considered in

the superposition. The values of RMSD for the superpo-

sitions can be seen in Table 2. These results show that

MtTRPS models are most similar to the StTRPS structure,

as expected. The models of MtTRPS contain 270 residues

in the α-subunit (residues 1–8 were omitted) and 422 in

the β-subunit (residues 1–23 were omitted). The model of

MtTRPS complexed with IPP is presented in Fig. 4.

The structure of the α-subunit follows the 8-fold α/β

barrel motif ﬁrst observed in triosephosphate isomerase

(TIM) (20,21). The α/β barrel structure can be described

in terms of a canonical form in which the molecule is

built up from eight repeating supersecondary structural

units each comprised of a β-strand followed by an α-

helix. The eight helices, connected to both ends of the

strands by loops, pack closely in parallel around the

periphery of the barrel.

The β-subunit of the crystal structure of StTRPS and

the molecular models of MtTRPS reveal the presence of

two domains of nearly equal size, denominated N-

domain and C-domain (21). The core of the N-domain is

formed by four strands, with three helices packed on one

side of the sheet, and a fourth helix (helix 6) packed on the

other side. The part of the N-domain constitutes a domain

that has very few interactions with the rest of the protein

and is named in the COMM-domain (21). The C-domain

is comprised of a six-strand β-sheet surrounded by α-

helices. The core of the C-domain contains a central β-

sheet sandwiched between layers of parallel packed

helices, much like the core of the N-terminal domain. The

β-sheet contains ﬁve parallel strands and one antiparallel.

The interface between α- and β-subunits is mostly

hydrophobic in character, and many residues of the

interface between two subunits are conserved in almost

all sequences. Approximately 1295 Å

of the surface

area of the α-subunit and 1285 Å

of the β-subunit are

buried at the α/β interface.

Interaction Between Protein and Its Inhibitors

All inhibitors studied in this work bind to the active

site of the α-subunit. Potential hydrogen bonding inter-

378 Dias et al.

Cell Biochemistry and Biophysics Volume 44, 2006

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

actions and their relative distances from active site

residues for all different inhibitors to MtTRPS can be

seen in Fig. 5. The majority of the interactions are com-

mon to all inhibitors; only some are unique. The

inhibitors form several electrostatic interactions with

different parts of the enzyme. The alkylphosphonate

portion of the phosphonated inhibitors and portions of

the phosphate of IPP extend approximately at a right

angle with the phenyl ring, and the oxygen of phospho-

nate and phosphate groups forms hydrogen bonds with

the hydroxyl group of Ser240 and the main-chain nitro-

gens of Gly190, Gly219, Gly239, and Ser240. The other

residue that interacts with inhibitors is Asp68, which

forms a hydrogen bond with hydroxyl group for com-

Molecular Models of TRPS 379

Cell Biochemistry and Biophysics Volume 44, 2006

Fig. 3. Sequence alignment for TRPS from M. tuberculosis and S. typhimurium. (A) sequence alignment for α-subunit. (B)

Sequence alignment for β-subunit. The alignments were performed with the program ClustalW (26), with the meaning of

the special characters in the last line as follows: (*) positions that have a single, fully conserved; (:) one of the following

“strong” groups is fully conserved; and (.) one of the following “weaker” groups is fully conserved.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

plexes 2, 3, 5, and 6, or the amino group for complex 4

bonded to the phenyl ring of the inhibitors, or with the

indole nitrogen atom of complex 1. Asp68 is thought to

play a crucial role in catalysis of the α-reaction and the

intersubunit communication interface (22). Compounds

1, 4, 5, and 6 present a third interaction with main chain

nitrogens of Ser240. In compound 1, the introduction of

the double bond does not disturb the pattern of the

hydrogen bond between the enzyme and inhibitor,

although rigid conformation is imposed. The atom of

the ﬂuorine introduced in the ring of compound 5 does

not interact by polar interaction with any residue of the

protein. On the other hand, the sulfoxide oxygen of

compound 6 forms one hydrogen bond interaction with

the hydroxyl group of Tyr181. By comparison of the

binding site of the six ligands in the models of MtTRPS,

we may conclude that it superimposes signiﬁcantly well

(figure not shown). The major differences occur in

inhibitor 4 in overall positioning of its alkyl chain that is

associated with an altered orientation of the phosphate

group position relative to the other inhibitors. Owing to

the orientation of the phosphate group in compound 4,

it presents a weaker interaction between the inhibitor

and the backbone amine of Gly190 compared with oth-

ers inhibitors (3.4 Å).

The presence of different ligands in the α-subunit

does not cause any signiﬁcant change in the binding site

of PLP in the β-subunit, despite that TRPS is an

allosteric protein and binding of the compounds of the

α-subunit is able to cause structural change in the β-

subunit (11).

Analysis of the Interaction Between the Molecular

Models and Inhibitors

Table 3 shows the analysis of the interaction between

the molecular models of MtTRPS and the molecular

structure of StTRPS with the six compounds studied in

the present work. Roughly, all ligands have high afﬁnity

for MtTRPS and StTRPS. They present from six to eight

intermolecular hydrogen bonds, contact area above 179

, and positive score value. Therefore, it can be said

that the high afﬁnity between the TRPS and IPP or

arylthioalkyl-phosphonated inhibitors is related to

directional hydrogen bonds and ionic interactions, as

well as shape complementarity of the contact surface of

the partners (23–25).

From analysis (Table 3), compound 5 presents greater

contact area with MtTRPS and seven hydrogen bonds,

but does not present greater amount of favorable bind-

ing (score value). This suggests that compound 5 can be

buried in the protein, but the binding is not completely

favorable. Compound 6 presents eight hydrogen bonds,

but the score value is the lowest. Therefore, this com-

pound must present binding that is not favorable for

interaction with protein. In contrast, compound 2 pre-

sents greater score value and thus a larger amount of

favorable binding with protein and a large contact area

(199 Å). Therefore, this compound presents higher afﬁn-

ity with the protein compared with other compounds.

In biological activity assay, compound 2 presented

greater herbicide activity against Arabidopsis thaliana (7).

Accordingly, based on modeling results obtained and

experimental results, compound 2 can also inhibit

MtTRPS. Inhibition studies of MtTRPS against com-

pound 2 and other inhibitors studied here may conﬁrm

these predictions.

CONCLUSIONS

The molecular models of MtTRPS generated by

MODELLER program possess good stereochemical

quality. Furthermore, they show that the interaction

with phosphonates and IPP is favorable as observed for

StTRPS. Therefore, the molecular models can be used

in the development of more specific drugs against

tuberculosis and other infectious diseases, because the

380 Dias et al.

Cell Biochemistry and Biophysics Volume 44, 2006

Table 1

Analysis of Ramachandram Plots

Most Additional Generously Disallowed

Model favored (%) allowed (%) allowed (%) (%)

1 96.8 (94.6) 5.5 (5.1) 0.5 (0.2) 0.2 (0.2)

2 93.2 (94.7) 6.2 (5.1) 0.5 (0.2) 0.0 (0.0)

3 93.8 (92.7) 5.7 (7.1) 0.4 (0.2) 0.2 (0.0)

4 94.0 (91.6) 5.5 (8.2) 0.4 (0.0) 0.2 (0.2)

5 93.4 (91.6) 6.0 (8.0) 0.5 (0.4) 0.0 (0.0)

6 93.4 (93.6) 6.0 (6.2) 0.5 (0.2) 0.0 (0.0)

Values in parentheses are for templates.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Molecular Models of TRPS 381

Cell Biochemistry and Biophysics Volume 44, 2006

Table 2

Analysis of Quality of Models of MtTRPS and Templates

rmsd from

3D proﬁle

G-factor

ideal geometry Superposition

Torsion Covalent Bond Bond rmsd

Complex Total score Ideal score S

ideal

score angles geometry Global lengths (Å) angles (°) (Å)

1 301.80 (281.35) 303.60 (298.97) 0.99 (0.94) –0.24 (0.12) –0.25 (–0.21) –0.24 (017) 0.021 (0.008) 2.519 (0.02) 0.128

2 298.95 (338.03) 303.60 (298.97) 0.98 (1.13) –0.27 (0.03) –0.30 (0.43) –0.28 (0.20) 0.021 (0.010) 2.515 (1.70) 0.232

3 302.55 (330.26) 303.60 (298.97) 1.00 (1.10) –0.27 (0.05) –0.28 (0.40) –0.28 (0.20) 0.021 (0.010) 2.686 (1.80) 0.256

4 293.33 (324.04) 303.60 (298.97) 0.97 (1.08) –0.28 (0.02) –0.29 (0.45) –0.28 (0.20) 0.021 (0.009) 2.686 (1.71) 0.248

5 298.52 (323.71) 303.60 (298.97) 0.98 (1.08) –0.30 (0.05) –0.33 (0.41) –0.30 (0.20) 0.022 (0.010) 2.745 (1.83) 0.297

6 297.38 (327.72) 303.60 (298.97) 0.98 (1.10) –0.14 (0.01) –0.44 (0.42) –0.25 (0.18) 0.022 (0.010) 2.785 (1.77) 0.249

Values obtained for the templates used for modeling are presented in parentheses.

Total score is the sum of the 3D-one-dimensional scores (statistical prefer-

ences) of each residue present in protein. Ideal score S

Ideal

= exp(–0.83 + 1.008 × ln(L)), where L is the number of amino acids. S

Ideal

score is compatibility of the

sequence with the 3D structure. It is obtained from total score/ideal score. S

Ideal

score above 0.45S

Ideal

Ideally, scores should be above –0.5. Values below –1.0 may need investigation.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

382 Dias et al.

Fig. 4. Ribbon diagram of the α- and β-subunits of MtTRPS complexed with IPP and PLP generated by Molmol (27).

Fig. 5. Hydrogen bond pattern between MtTRPS and inhibitors. (1) indol-propanol-phosphate; (2) 4-(2-hydroxyphenylthio)-

1-butenylphosphonic acid; (3) 4-(2-hydroxyphenylthio)butylphosphonic acid; (4) 4-(2-aminophenylthio)butylphosphonic acid;

(5) 4-(2-hydroxy-5-ﬂuorophenylthio)butylphophonic acid; and (6) 4-(2-hydroxyphenylsylﬁnyl)butylphosphonic acid.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

examination of the binding site for arylthiophos-

phanate inhibitor presents interactions that could be

exploited to improve affinity with the protein. The

replacement of hydrogen atoms along the alkyl chain

with larger atoms or functional group can increase the

afﬁnity or the inhibition of the ligands.

ACKNOWLEDGMENTS

This work was supported by grants from FAPESP

(SMOLBNet 01/07532-0, 02/04383-7, 03/12472-2,

04/00217-0), CNPq, CAPES, and Instituto do Milênio

(CNPq-MCT) PRONEX, FAPERGS. W.F.A., M.S.P.,

L.A.B., and D.S.S. are researchers for the Brazilian

Council for Scientiﬁc and Technological Development

from Brazil (CNPq).

REFERENCES

1. Trouiller, P., Torreele, E., Olliaro, P., et al. (2001) Drugs for

neglected diseases: a failure of the market and a public

health failure? Trop. Med. Int. Health 6, 945–951.

2. World Health Organization. Global Tuberculosis Control.

WHO Report 2001. Geneva, Switzerland, WHO/CDS/

TB/2001.287

3. Dosselaere, F. and Vanderleyden, J. (2001) A metabolic

node in action: chorismate-utilizing enzymes in microor-

ganisms. Crit. Rev. Microbiol. 27, 75–131.

4. Roberts, F., Roberts, C. W., Johnson, J. J., et al. (1998)

Evidence for the shikimate pathway in apicomplexan par-

asites. Nature 393, 801–805.

5. Schonbrunn, E., Eschenburg, S., Shuttleworth, W. A., et al..

(2001) Interaction of the herbicide glyphosate with its tar-

get enzyme 5-enolpyruvylshikimate 3-phosphate synthase

in atomic detail. Proc. Natl. Acad. Sci. U.S.A. 98, 1376–1380.

6. Daves, G. M, Barrett-Bee, K. J. Jude, D. A., et al. (1994) (6S)-

6-Fluoroshikimic acid, an antibacterial agent acting on the

aromatic biosynthetic pathway. Antimicrob. Agents

Chemother. 38, 403–406.

7. Finn, J., Langevine, C., Birk, I., Nicherson, K., and

Rodaway, S. (1999) Rational herbicide design by inhibition

of tryptophan biosynthesis. Bioorg. Med. Chem. Lett. 9,

2297–2302.

8. Smith, D. A., Parish, T., Stoker, N. G., and Bancroft, G. J.

(2001) Characterization of auxotrophic mutants of

Mycobacterium tuberculosis and their potential as vaccine

candidates. Infect. Immun. 69, 1142–1150.

9. Hyde, C. C. and Miles, E. W. (1990) The tryptophan syn-

thase multienzyme complex: exploring structure-function

relationships with X-ray crystallography and mutagene-

sis. Biotechnology 8, 27–32.

10. Pan, P., Woehl, E., and Dunn, M. F. (1997) Protein architec-

ture, dynamics and allostery in trytophan synthase chan-

neling. Trends Biochem. Sci. 22, 22–27.

11. Weyand, M. and Schlichting, I. (1999) Crystal structure of

wild-type tryptophan synthase complexed with the nat-

ural substrate indole-3-glycerol phosphate. Biochemistry

38, 16,469–16,480.

12. Sachpatzidis, A., Dealwis, C., Lubetsky, J. B., Liang, P. H.,

Anderson, K. S., and Lolis, E. (1999) Crystallographic

studies of phosphonate-based alpha-reaction transition-

state analogues complexed to tryptophan synthase.

Biochemistry 38, 12,665–12,674.

13. Uchoa, H. B., Jorge, G. E., da Silveira, N. J. F., Camera, J. C.,

Jr., Canduri, F., and de Azevedo, W. F., Jr. (2004) Parmodel:

a web server for automated comparative modeling of pro-

teins. Biochem. Biophys. Res. Commun. 325, 1481–1486.

14. Sali, A. and Blundell, T. L. (1993) Comparative protein

modelling by satisfaction of spatial restraints. J. Mol. Biol.

234, 779–815.

15. Collaborative Computational Project No. 4 (1994) The

CCP4 suite: program for protein crystallography. Acta

Crystallogr. D 50, 760–763.

16. Brünger, A. T. (1992) X-PLOR version 3.1: a system for

crystallography and NMR. Yale University Press, New

Haven, CT.

17. Bowie, J. U., Luthy, R., and Eisenberg, D. (1991) A method

to identify protein sequences that fold into a known three-

dimensional structure. Science 253, 164–170.

18. Luthy, R., Bowie, J., and Eisenberg D. (1992) Assessment of

protein models with three-dimensional proﬁles. Nature

356, 83–85.

19. Wang, R., Liu, L., Lai, L., and Tang, Y. (1998) SCORE: a

new empirical method for estimating the binding afﬁnity

of a protein-ligand complex. J. Mol. Model. 4, 379–394.

20. Hyde, C. C., Ahmed, S. A., Padlan, E. A., Miles, E. W., and

Davies, D. R. (1988) Three-dimensional structure of the

tryptophan synthase α

multienzyme complex from

Salmonella typhimurium. J. Biol. Chem. 263, 17,857–17,871.

21. Banner, D. W., Bloomer, A. C., Petsko, G. A., et al. (1975)

Structure of chicken muscle triose phosphate isomerase

determined crystallographically at 2.5 Å resolution using

amino acid sequence data. Nature 255, 609–614.

22. Rhee, S., Miles, E. W., Mozzarelli, A., and Davies, D. R.

(1998) Cryocrystallography and microspectrophotometry

of a mutant (alpha D60N) tryptophan synthase alpha 2

beta 2 complex reveals allosteric roles of alpha Asp60.

Biochemistry 37, 10,653–10,659.

Molecular Models of TRPS 383

Cell Biochemistry and Biophysics Volume 44, 2006

Table 3

Interaction of Between Protein and Inhibitors

Number

hydrogen Contact

Model bonds area (Å

) Score

1 7 (7) 200.0 (167.0) 7.39 (7.76)

2 6 (6) 199.0 (176.0) 7.98 (8.29)

3 6 (6) 197.0 (191.0) 7.70 (7.94)

4 7 (7) 192.0 (172.0) 6.61 (6.95)

5 7 (7) 211.0 (185.0) 6.51 (7.60)

6 8 (7) 179.0 (175.0) 5.52 (7.04)

Values in parenthesis are for templates.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

23. De Azevedo, W. F., Jr., Mueller-Dieckmann, H. J., Schulze-

Gahmen, U., Worland, P. J., Sausville, E., and Kim, S. H.

(1996) Structural basis for speciﬁcity and potency of a

ﬂavonoid inhibitor of human CDK2, a cell cycle kinase.

Proc. Natl. Acad. Sci. U.S.A. 93, 2735–2740.

24. De Azevedo, W. F., Jr., Canduri, F., Dos Santos, D. M., et al.

(2003) Structural basis for inhibition of human PNP by

immucillin-H. Biochem. Biophys. Res. Commun. 309,

922–927.

25. Pereira, J. H., Canduri, F., de Oliveira, J. S., et al. (2003)

Structural bioinformatics study of EPSP synthase from

Mycobacterium tuberculosis. Biochem. Biophys. Res. Commun.

312, 608–614.

26. Thompson, J. D., Higgins, D. G., and Gibson, T. J. (1994)

CLUSTAL W: improving the sensitivity of progressive

multiple sequence alignment through sequence weight-

ing, positions-speciﬁc gap penalties and weight matrix

choice. Nucleic Acids Res. 22, 4673–4680.

27. Koradi, R., Billeter, M., and Wüthrich, K. (1996) MOL-

MOL: a program for display and analysis of macromolec-

ular structures. J. Mol. Graph. 14, 51–55.

384 Dias et al.

Cell Biochemistry and Biophysics Volume 44, 2006

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.4 Effects of the magnesium and chloride ions and shikimate on the structure of

shikimate kinase from Mycobacterium tuberculosis.

Marcio V. B. Dias; Lívia M. Faím; Igor B. Vasconcelos; Jaim S. de Oliveira; Luiz A.

Basso; Diógenes S. Santos; Walter F. de Azevedo Jr. Acta Crystallographica Section F

Structural Biology and Crystallization Communications (ISSN 1744-3091), 63, p. 1-6,

2007.

Neste trabalho é apresentado o efeito dos íons magnésio e cloro e do ácido

chiquímico sobre a estrutura da chiquimato quinase de M. tuberculosis (peso molecular de

18583 Da).

A Chiquimato quinase de M. tuberculosis foi clonada, superexpressada e purificada

pelo grupo de Pesquisas do Prof. Dr. Diógenes S. Santos da PUC – Porto Alegre – RS.

Para a obtenção de cristais da chiquimato quinase de M. tuberculosis em complexo com

ácido chiquímico e ADP, na ausência de Mg

, a proteína foi dialisada contra tampão Tris-

HCl, 50 mM, pH 8,0 e concentrada a 14 mg.mL

-1

e utilizada inicialmente a condição de

cristalização estabelecida por Dhaliwal et al. (2004). Os melhores cristais foram obtidos

em uma condição composta for Tris-HCl, 0,1 M, pH 8, 17% de PEG 1500 e de cloreto de

lítio, 0,5-0,7 M. Por outro lado, a condição de cristalização na qual foram obtidos os

melhores cristais para o complexo da chiquimato quinase de M. tuberculosis em complexo

com ADP e magnésio era composta por Tris-HCl, 0,1 M, 20% PEG 3350 e 0,1 M de

MgCl

.6H

O e a concentração da proteína era de 17 mg.mL

-1

. A coleta de dados de

difração de raios X foi realizada no LNLS. O cristal obtido para a chiquimato quinase de

M. tuberculosis em complexo com ADP e ácido chiquímico é similar aos descritos na

literatura e é pertencente ao grupo espacial P3

21. Este cristal difratou a 1,93 Å de

resolução e apresenta um monômero na unidade assimétrica. Os cristais da chiquimato

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

quinase de M. tuberculosis em complexo com Mg

e ADP são pertencentes ao grupo

espacial P2

e difrataram a 2,8 Å de resolução, e apresentam na unidade assimétrica

quatro monômeros, que formam um tetrâmero com simetria 222. As estruturas para os

conjuntos de dados dos cristais obtidos foram determinadas por substituição molecular

utilizando como modelo de busca a estrutura da chiquimato quinase de M. tuberculosis em

complexo com MgADP e ácido chiquímico (PEREIRA et al., 2004). O refinamento

cristalográfico das estruturas foi realizado pelo programa REFMAC 5 (MURSHUDOV;

VAGIN; DODSON, 1997) e a inspeção visual e adição de moléculas de água foi efetuada

pelo programa XtalView/Xfit (McREE et al., 1999). A estrutura da chiquimato quinase de

M. tuberculosis em complexo com ADP e ácido chiquímico, na ausência do íon magnésio,

foi comparada com estrutura da chiquimato quinase em complexo com MgADP e ácido

chiquímico determinada por Pereira et al., 2004, que é a única estrutura que foi observado

o íon magnésio quando o ácido chiquímico está presente. Pela comparação destas duas

estruturas pôde-se observar o efeito que o íon magnésio causa sobre a estrutura da

chiquimato quinase e sobre o ácido chiquímico. Nesta estrutura foram observadas

alterações na posição da cadeia lateral de importantes resíduos do sítio ativo da enzima e

alterações nos grupos hidroxilas da molécula de ácido chiquímico. Com relação à estrutura

da chiquimato quinase de M. tuberculosis em complexo com magnésio e ADP pôde-se

observar as possíveis interferências do ácido chiquímico sobre a conformação da

chiquimato quinase, uma vez que esta molécula não foi adicionada na condição de

cristalização. Além disso, em um dos monômeros do tetrâmero, a molécula de cloro, que

representa um importante papel na catálise da chiquimato quinase, não foi observada

(CERASOLI et al., 2003). Assim, este monômero apresentou alterações estruturais não

observadas nos outros monômeros. Estas alterações conformacionais foram descritas e

acreditamos que elas podem ser causadas pela ausência do íon cloreto na estrutura da

chiquimato quinase de M. tuberculosis.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

protein structure communications

Acta Cryst. (2007). F63, 1–6 doi:10.1107/S1744309106046823 1

Acta Crystallographica Section F

Structural Biology

and Crystallization

Communications

ISSN 1744-3091

Effects of the magnesium and chloride ions and

shikimate on the structure of shikimate kinase from

Mycobacterium tuberculosis

Marcio Vinicius Bertacine Dias,

Lı

via Maria Faı

Igor Bordin

Vasconcelos,

Jaim Simo

es de

Oliveira,

Luiz Augusto Basso,

Dio

genes Santiago Santos

* and

Walter Filgueira de Azevedo Jr

Programa de Po

s-Graduac¸a

o em Biofı

sica

Molecular, Departamento de Fı

sica, UNESP,

o Jose

do Rio Preto, SP 15054-000, Brazil,

Pontifı

cia Universidade Cato

lica do Rio Grande

do Sul, Centro de Pesquisa em Biologia

Molecular e Funcional, Porto Alegre, RS, Brazil,

and

Faculdade de Biocie

ncias, Pontifı

cia

Universidade Cato

lica do Rio Grande do Sul,

Av. Ipiranga, 6681 Porto Alegre-RS,

CEP 90619-900, Brazil

Correspondence e-mail: [email protected],

[email protected]

Received 27 September 2006

Accepted 6 November 2006

PDB References: shikimate kinase–ADP–

shikimate complex, 2dfn, r2dfnsf; shikimate

kinase–MgADP complex, 2dft, r2dftsf.

Bacteria, fungi and plants can convert carbohydrate and phosphoenolpyruvate

into chorismate, which is the precursor of various aro matic compounds. The

seven enzymes of the shikimate pathway are responsible for this conversion.

Shikimate kinase (SK) is the ﬁfth enzyme in this pathway and converts

shikimate to shikimate-3-phosphate. In this work, the conformational changes

that occur on binding of shikimate, magn esium and chloride ions to SK from

Mycobacterium tuberculosis (MtSK) are described. It was observed that both

ions and shik imate inﬂuence the conformation of residues of the active site of

MtSK. Magnesium inﬂuence s the conformation of the shikimate hydroxyl

groups and the position of the side chains of some of the residues of the active

site. Chloride seems to inﬂuence the afﬁnity of ADP and its position in the active

site and the opening length of the LID domain. Shikimate binding causes a

closing of the LID domain and also seems to inﬂuence the crystallographic

packing of SK. The results shown here could be useful for understanding the

catalytic mechanism of SK and the role of ions in the activity of this protein.

1. Introduction

Mycobacterium tuberculosis, the aetiological agent of tuberculosis

(TB), infects one-third of the world’s population. It is estimated that

1.7 million deaths resulted from TB in 2004, 95% of which occurred in

developing countries (World Health Organization, 2006). The emer-

gence of TB as a public health threat, the high susceptibility of HIV/

TB co-infected patients and the proliferation of multi-drug-resistant

strains have created a need for newer and better drugs for the

treatment of TB.

The shikimate pathway is an attractive target for the development

of herbicides (Coggins, 1998) and antibiotic agents (Davies et al.,

1994) because it is essential in algae, higher plants, bacteria, fungi and

apicomplexan parasites but is absent from mammals (Bentley, 1990;

Roberts et al., 1999). The shikimate pathway is a seven-step biosyn-

thetic route that links the metabolism of carbohydrates to the

synthesis of aromatic amino acids. The shikimate pathway leads to the

biosynthesis of chorismate, which is a precursor of aromatic amino

acids and many other aromatic compounds (Ratledge, 1982). Shiki-

mate kinase (SK; EC 2.7.1.71), the ﬁfth enzyme of this pathway,

catalyzes phosphate transfer from ATP to the carbon-3-hydroxyl

group of shikimate, forming shikimate 3-phosphate (S3P).

SK belongs to the nucleoside monophosphate (NMP) kinase

structural family. SK is a / protein consisting of a central sheet of

ﬁve parallel -strands ﬂanked by -helices, with overall topology

similar to that of adenylate kinase (Pereira et al., 2004; Krell et al.,

1998, 2001). A characteristic feature of the NMP kinases is that they

undergo large conformational changes during catalysis (Vonrhein et

al., 1995). The NMP kinases are composed of three domains: the

CORE, which contains a highly conserved phosphate-binding loop

(P-loop), the LID domain, which undergoes substantial conforma-

tional changes upon substrate binding, and the NMP-binding domain,

# 2007 International Union of Crystallography

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

which is responsible for the recognition and binding of a speciﬁc

substrate (Gu et al., 2002).

MgADP induces concerted hinged movements of the shikimate-

binding (SB) and LID domains such that the two domains move

towards each other and towards the active centre of the enzyme in the

presence of this ligand (Gu et al., 2002).

The precise mode of binding of shikimate to MtSK and some

conformational changes upon shikimate binding to MtSK have

recently been proposed (Pereira et al., 2004; Dhaliwal et al., 2004).

The binding of shikimate to MtSK causes a concerted conformational

protein structure communications

2 Dias et al.



Shikimate kinase Acta Cryst. (2007). F63, 1–6

Figure 1

(a) Structure of MtSK in complex with ADP, shikimate and Cl

.(b) Tetrameric structure of MtSK in complex with ADP, Mg

and Cl

. The monomers A, B, C and D are

represented in green, blue, pink and yellow, respectively. The dark blue and yellow spheres represent the magnesium and chloride ions, respectively. The ADP molecules are

represented as sticks. (c) Representation of the hydrogen-bonding interactions that occur between the ribose hydroxyl groups of ADP molecules bound to monomers A and

B. The distances are shown in A

. Figures were generated with the program MolMol (Koradi et al., 1996).

change of the LID and SB domains towards each other and results in

an additional closure of the active site.

Chloride ions have been shown to weaken the interaction between

shikimate and SK from Erwinia chrysanthemi and to strengthen the

afﬁnity of the enzyme for ADP and ATP (Cerasoli et al., 2003). Thus,

a chloride ion seems to occupy a site crucial for the binding of the

nucleotide substrate in the correct orientation for catalysis.

In this work, we report two crystallographic structures: the MtSK–

ADP–shikimate and MtSK–MgADP complexes. The MtSK–ADP–

shikimate complex has been solved at 1.93 A

resolution. The crystal

structure of MtSK in complex with ADP and Mg

has been solved at

2.8 A

resolution. The data presented here provide an evaluation of

the effects of both Mg

and Cl

ions and shikimate on the structure

of MtSK. Thus, these results provide further insight into the roles that

different ions and substrate play in the mechanism of action of MtSK.

2. Materials and methods

2.1. Crystallization

The cloning, expression and puriﬁcation of MtSK have been

reported elsewhere (Oliveira et al., 2001). Initially, MtSK was

concentrated to 14 mg ml

À1

and dialyzed against 50 mM Tris–HCl

buffer pH 8.0. For the crystallization of MtSK in complex with ADP

and shikimate in the absence of Mg

ion, variations of the crystal-

lization condition established by Dhaliwal et al. (2004) were used.

Crystals were obtained by the hanging-drop vapour-diffusion

method. The well solution contained 0.1 M Tris–HCl buffer pH 8.0,

17% PEG 1500 and 0.5–0.7 M LiCl. The drops had a protein solu-

tion:well solution volume ratio of 1.6:1.0. For the crystallization of the

protein in complex with ADP and Mg

in the absence of shikimate, a

crystallization solution composed of 0.1 M Tris–HCl buffer pH 8.0,

20% PEG 3350 and 0.1 M MgCl

.6H

O was used. The concentration

of the protein was 17 mg ml

À1

and the drops again had a protein

solution:well solution volume ratio of 1.6:1.0.

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

2.2. Data collection and processing

All data sets were collected at a wavelength of 1.427 A

using a

synchrotron-radiation source (Station PCr, LNLS, Campinas, Brazil;

Polikarpov et al., 1998) and a CCD detector (MAR CCD). Data

collection was performed under cryogenic conditions at a tempera-

ture of 100 K in a cold nitrogen stream generated and maintained

with an Oxford Cryosystem. Prior to ﬂash-cooling, glycerol was

added to the crystallization drop to 20%(v/v). The data sets were

processed using the program MOSFLM (Leslie, 1992) and scaled

with SCALA (Collaborative Computational Project, Number 4,

1994).

2.3. Structure determination

The crystal structures of both complexes were determined by

standard molecular-replacement methods using the program AMoRe

(Navaza, 2001). For both complexes, we used as a search model the

structure of MtSK–MgADP–shikimate (PDB code 1we2; Pereira et

al., 2004). Reﬁnement of the structures was performed using

REFMAC5 implemented in the CCP 4 package (Murshudov et al.,

1997; Collaborative Computational Project, Number 4, 1994). Xtal-

View/Xﬁt (McRee, 1999) was used for visual inspection and addition

of water molecules. The stereochemical correctness of the models was

checked using PROCHECK (Laskowski et al., 1993). The ﬁnal atomic

models were superposed using LSQKAB from the CCP4 package

(Collaborative Computational Project, Number 4, 1994). PAR-

MODEL (Ucho

a et al., 2004) was used in the ﬁnal analysis of the

model.

3. Results and discussion

MtSK crystallized in two different space groups depending on the

complex. The crystals of the MtSK–ADP–shikimate ternary complex

were trigonal, space group P3

21, and diffracted to 1.93 A

resolution.

The asymmetric unit contains one molecule and the ﬁnal values of R

and R

free

were 20.2 and 27.0%, respectively. In contrast, the crystals of

the MtSK–ADP–Mg

ternary complex were orthorhombic, space

group P2

, and diffracted to 2.8 A

resolution. The asymmetric

unit contains four MtSK monomers that form a tetramer and the ﬁnal

values of R and R

free

were 18.3 and 28.0%, respectively. Table 1

details the data processing, reﬁnement statistics and quality analysis

of the two complexes.

The structures present good geometry, although some residues of

the LID domain are located in disallowed regions of the Rama-

chandran plot. The N-terminal methionine residue is not observed

since it was removed during the MtSK expression in E. coli (Oliveira

et al., 2001). The ten C-terminal residues are disordered in both

structures and have not been included in the ﬁnal structure.

The folding of the MtSK complexes presented here is similar to

that of those reported previously (Pereira et al., 2004; Krell et al.,

1998; Gu et al., 2002; Dhaliwal et al., 2004). SK is a / protein and

consists of ﬁve central parallel -strands ﬂanked by -helices. Fig. 1(a)

shows a ribbon representation of the secondary-structure elements of

MtSK–ADP–shikimate at 1.93 A

resolution.

protein structure communications

Acta Cryst. (2007). F63, 1–6 Dias et al.



Shikimate kinase 3

Figure 3

Superposition of the chains of the structure of the MtSK–MgADP binary complex obtained in the absence of shikimate. (a) Chains A (yellow) and C (light blue), (b) chains B

(green) and C (light blue) and (c) chains B (green) and D (red).

Figure 2

Superposition of the structures of the MtSK–MgADP–shikimate and MtSK–ADP–

shikimate ternary complexes. The C



trace of the MtSK–MgADP–shikimate

complex is presented in green and that of the MtSK–ADP–shikimate trace is

presented in blue. The C atoms of MtSK–MgADP–shikimate and MtSK–ADP–

shikimate are coloured white and yellow, respectively. The Mg

shown in yellow

and the chloride ion shown in turquoise refer to the MtSK–MgADP–shikimate

structure, while the chloride ion in dark blue refers to the MtSK–ADP–shikimate

structure. The ﬁgure was generated with the program MolMol (Koradi et al., 1996).

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.1. Structure of the MtSK–ADP–Mg

ternary complex

The asymmetric unit of the MtSK–ADP–Mg

ternary complex

structure contains four monomers forming a homotetramer with 222

symmetry. However, each subunit presents a different conformation

state. In the A and C subunits the LID domain is disordered and

residues 115–123 and 114–123 were thus not included in the ﬁnal

model.

The structure of the MtSK tetramer is shown in Fig. 1(b). Each

monomer is in contact with the other three, creating an intricate

packing arrangement. The ADP molecules appear to play an

important role in the stabilization of the tetramer, since the hydroxyl

groups of the ribose moiety of ADP of one monomer form hydrogen

bonds with those of a neighbouring monomer (Fig. 1c).

3.2. Influence of Mg

on the structure of MtSK

The absence of Mg

ions seems to have a signiﬁcant effect on the

position of shikimate and some of the active-site residues, mainly

Asp32 and Asp34 (these are conserved residues in all SKs and are

involved in the binding of Mg

; Fig. 2). In addition, the chloride ion

and ADP molecule also undergo changes in position. Fig. 2 shows a

protein structure communications

4 Dias et al.



Shikimate kinase Acta Cryst. (2007). F63, 1–6

superposition of the MtSK–MgADP–shikimate (Pereira et al., 2004)

and MtSK–ADP–shikimate complexes, showing the structural

changes caused by the absence of the Mg

ion.

In the structure of the MtSK–ADP–shikimate ternary complex, the

3-, 4- and 5-hydroxyl groups of shikimate undergo a shift in their

positions, leading to small differences in the hydrogen-bonding

pattern between shikimate and MtSK. Although the Asp34 residue

interacts with shikimate in both structures, a hydrogen bond between

the OD2 atom of Asp34 and the O3 atom of the shikimate is not

observed in the MtSK structure without Mg

(not shown).

The position of the chloride ion is closer to shikimate in the MtSK–

ADP–shikimate ternary complex than in the MtSK–MgADP–

shikimate complex (Pereira et al., 2004). In the structure of the

MtSK–ADP–shikimate ternary complex, the chloride ion is 2.4 A

away from the O1 atom of the shikimate 3-hydroxyl group, while in

the structure of MtSK–MgADP–shikimate the distance between

these two atoms is 3.4 A

. The absence of the Mg

ion can also cause

alterations in the side chains of the hydrophobic residues Phe49 and

Phe57 located in the SB domain and of Pro118, Ala46 and Ile45

located in the LID domain (Fig. 2).

As established previously for other NMP kinases, the Mg

ion is

expected to play an important role in the transfer of the -phosphate

group of ATP to the nucleophilic 3-OH group of shikimate by an

associative reaction mechanism (Schlichting & Reinstein, 1997;

Bellinzoni et al., 2006). Thus, the differences observed in the MtSK–

ADP–shikimate ternary complex in the absence of Mg

may

contribute to understanding of the MtSK chemical mechanism.

3.3. Influence of chloride ion on the structure of MtSK

In our structure of the MtSK–MgADP binary complex, monomer

C of the MtSK tetramer does not contain chloride ion. This ion seems

to inﬂuence the position of the ADP molecule and the position of the

shikimate molecule. This monomer presents a conformation that is

more open than the other monomers (Fig. 3). Furthermore, the LID

domain of monomer C has a mean B factor that is higher than those

of the other monomers (not shown). The ADP molecule bound to

monomer C is shifted by approximately 3.0 A

in the active site (Fig. 3)

compared with the other monomers of the MtSK tetramer or with

other structures of MtSK in complex with ADP. These results are in

Figure 4

Pattern of water bonding in the active site of MtSK. (a) Structure at 1.93 A

resolution (chloride ion and shikimate present, magnesium ion absent), (b) monomer B of the

tetramer (chloride and magnesium ions present, shikimate absent), (c) monomer C of the tetramer (magnesium ion present, shikimate and chloride ion absent). The ﬁgures

were generated using PyMOL (DeLano, 2004).

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

agreement with the kinetic and spectroscopic data previously

obtained by Cerasoli et al. (2003). This work shows that the chloride

ion increases the stability of the E. chrysanthemi SK structure and

that this same ion also inﬂuences the afﬁnity of ADP for SK. Thus,

the increase in the protein stability may be a consequence of chloride

favouring the SK structure in its closed state.

The chloride ion bound to the MtSK active site seems to be part of

an intricate network formed of water molecules, residues of the LID

and SB domains, ADP and shikimate. This network of interactions

appears to cause closure of the structure (Fig. 4). In all structures of

MtSK reported so far and in the three monomers of the MtSK

tetramer, an interaction network involving three water molecules is

formed in the active site (Fig. 4). These water molecules bridge the

interactions between ADP, chloride and magnesium ion and MtSK.

In our MtSK–ADP–shikimate ternary complex, water 306 occupies a

similar position to the Mg

ion. In the absence of chloride ion, this

phenomenon does not occur and furthermore the formation of this

intricate network that can induce the opening of the structure is

avoided.

3.4. Influence of shikimate in the structure of MtSK

The crystals of the MtSK–MgADP binary complex in the absence

of shikimate were obtained in space group P2

, which has not

been previously described for MtSK. The asymmetric unit presents

four MtSK monomers that form a tetramer. The absence of shikimate

from the crystallization conditions inﬂuences the crystal packing and

also the conformation of the SB and LID domains of MtSK, as

observed by Pereira et al. (2004) and Dhaliwal et al. (2004). In

accordance with this, Gan and coworkers recently solved the apo-

MtSK and MtSK–shikimate binary complex structures and suggested

that shikimate binding deﬁnes the conformational change of the

protein that arises when shikimate is bound: the LID domain is

ordered and closes over the bound shikimate (Gan et al., 2006).

4. Conclusion

Here, we report two structures of shikimate kinase from M. tuber-

culosis: the MtSK–ADP–shikimate and MtSK–MgADP complexes.

In the former, we observe the effect of the Mg

ion on the structure

and in the latter we observe the effect of shikimate on the crystal

packing and on the structure of MtSK. The Mg

ion seems to

inﬂuence the position of the hydroxyl groups of the shikimate

molecule and some of the residues of the active site of MtSK. The

crystal structure of the MtSK–MgADP complex was solved in space

group P 2

, which has not previously been described for MtSK. In

this space group, the MtSK presents a tetramer with 222 symmetry, in

which the ribose moiety of the ADP molecule seems to play an

important role in the stabilization of the tetramer and the contacts

between the monomers, which occur mainly in the LID-domain

region. However, one monomer of the tetramer does not contain

chloride ion. The absence of this ion seems to cause large changes in

the position of the ADP molecule and also causes a large opening of

MtSK. This information is accordance with the results obtained by

Cerasoli et al. (2003), which shows the importance of the chloride ion

in the stability and the alignment of the substrates in the active site of

SK.

We hope that the results described here will shed light on the

structural changes of MtSK upon binding of substrate(s) that will be

useful for the understanding of the catalytic mechanism and for

structure-based drug design of novel inhibitors that may be potential

anti-mycobacterial agents.

This work was supported by grants from FAPESP (SMOLBNet,

Proc. 01/07532-0, 03/12472-2, 04/00217-0), CNPq, CAPES and Insti-

tuto do Mile

ncio (CNPq-MCT), DSS, WFA (CNPq, 300851/98-7) and

LAB (CNPq, 520182/99-5) are researchers of the Brazilian Council

for Scientiﬁc and Technological Development.

References

Bellinzoni, M., Haouz, A., Grana, M., Munier-Lehmann, H., Shepard, W. &

Alzari, P. M. (2006). Proteins, 15, 1–5.

Bentley, R. (1990). Crit. Rev. Biochem. Mol. Biol. 25, 307–384.

Cerasoli, E., Kelly, S. M., Coggins, J. R., Lapthorn, A. J., Clarke, D. T. & Price,

N. C. (2003). Biochim. Biophys. Acta, 1648, 43–54.

Coggins, J. R. (1998). In Herbicides and Plant Metabolism, edited by A. Dodge.

Cambridge University Press.

protein structure communications

Acta Cryst. (2007). F63, 1–6 Dias et al.



Shikimate kinase 5

Table 1

Crystallographic data, reﬁnement statistics and analysis of the quality of MtSK

structures.

Values in parentheses are for the outermost shell.

MtSK–ADP–shikimate MtSK–MgADP

Crystallographic data

Unit-cell parameters

a (A

) 63.3 60.2

b (A

) 63.3 62.2

c (A

) 91.6 170.6

Space group P3

21 P2

No. of measurements 76792 97059

No. of independent reﬂections 16017 17057

Completeness (%) 96.9 (91.5) 99.4 (96.9)

sym

† (%) 8.9 (58.8) 12.1 (58.0)

Redundancy 4.8 5.7

Reﬁnement statistics

Resolution range (A

) 35.16–1.93 57.17–2.80

Reﬂections used for reﬁnement 15130 15670

Final R factor‡ (

%) 20.2 18.3

Final R

free

§ (%) 27.0 28.0

Correlation coefﬁcient (%) 95.2 94.3

B values (A

)

Main chain 31 33

Side chain 34 36

ADP 21 27

Shikimate 32 —

Waters 39 30

Quality of structure

Three-dimensional proﬁle} S = 88.07, IS = 74.95,

S/IS = 1.18IS

S = 343.59, IS = 294.34,

S/IS = 1.17IS

Ramachandran plot

Favoured 95.6 84.8

Additionally allowed 2.9 13.3

Generously allowed 0.7 0.8

Disallowed 0.7 1.1

† R

sym

= 100

IðhÞÀhIðhÞij=

IðhÞ, where I (h) is the observed intensity and hI(h)i is

the mean intensity of reﬂection h over all measur ements of I(h). ‡ R factor =

100

obs

À F

calc

obs

, the sums being taken over all reﬂections with F/(F)>

2(F). § R

free

is the R factor for 10% of the data that were not included during

crystallographic reﬁnement. } The ideal score measures the compatibility of a protein

model with its sequence, using a 3D proﬁle. Each residue position in the 3D model is

characterized by its environment and is represented by a row of 20 numbers in the proﬁle.

These numbers are the statistical preferences (called 3D-1D scores) of each of the 20

amino acids for this environment. Environments of residues are deﬁned by three

parameters: the area of the residue that is buried; the fraction of side chain area that is

covered by polar atoms (O and N) and the local secondary structure. The 3D proﬁle score

S for the compatibility of the sequence with the model is the sum, over all residue

positions, of the 3D-1D scores for the amino-acid sequence of the protein. For 3D protein

models known to be correct, the 3D proﬁle score S for the amino-acid sequence of the

model is high, by contrast, the proﬁle score S for the compatibility of a wrong 3D protein

model with its sequence is often low. When this method is used to verify a structure, the

raw compatibility score alone is difﬁcult to interpret. In this case it is necessary to

compare the score to those obtained using structures known to be correct, we use the

Ideal Score (IS), that is calculated from the length of the protein. The IS is determined by

IS ¼ exp½À0:83 þ 1:008 Â lnðLÞ. Where L is the length of the sequence. Severely

misfolded structures typically have scores less than 0.45 IS. A score near or above IS

indicates a reliable structure.

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50,

760–763.

Davies, G. M., Barret-Bee, K. J., Jude, D. A., Lehan, M., Nichols, W. W. &

Pinder, P. E. (1994). Agents Chemother. 38, 403–406.

DeLano, W. L. (2004). The PyMOL Molecular Graphics System. DeLano

Scientiﬁc, San Carlos, CA, USA.

Dhaliwal, B., Nichols, C. E., Ren, J., Lockyer, M., Charles, I., Hawhins, A. R. &

Stammers, D. K. (2004). FEBS Lett. 574, 49–54.

Gan, J., Gu, Y., Li, Y., Yan, H. & Ji, X. (2006). Biochemistry, 45, 8539–8545.

Gu, Y., Reshetnikova, L., Li, Y., Wu, Y., Yan, H., Singh, S. & Ji, S. (2002). J.

Mol. Biol. 319, 779–789.

Koradi, R., Billeter, M. & Wu

thrich, K. (1996). J. Mol. Graph. 14, 51–55.

Krell, T., Coggins, J. R. & Lapthorn, A. J. (1998). J. Mol. Biol. 278, 983–997.

Krell, T., Maclean, J., Boam, D. J., Cooper, A., Resmini, M., Brocklehurst, K.,

Kelly, S. M., Price, N. C., Lapthorn, A. J. & Coggins, J. (2001). Protein Sci. 10,

1137–1149.

Laskowski, R. A., MacArthur, M., Moss, D. S. & Thornton, J. M. (1993). J.

Appl. Cryst. 26, 283–291.

Leslie, A. G. W. (1992). Jnt CCP4/ESF–EACBM Newsl. Protein Crystallogr. 26.

McRee, D. E. (1999). J. Struct. Biol. 125, 156–165.

Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Acta Cryst. D53,

240–255.

Navaza, J. (2001). Acta Cryst. D57, 1367–1372.

Oliveira, J. S., Pinto, C. A., Basso, L. A. & Santos, D. S. (2001). Protein Expr.

Purif. 22, 430–435.

Pereira, J. H., Oliveira, J. S., Canduri, F., Dias, M. V. B., Palma, M. S., Basso,

L. A., Santos, D. S. & Azevedo, W. F. Jr (2004). Acta Cryst. D60, 2310–

2319.

Polikarpov, I., Perles, L. A., de Oliveira, R. T., Oliva, G., Castellano, E. E.,

Garratt, R. C. & Craievich, A. (1998). Nucl. Instrum. Methods Phys. Res. A,

405, 159–164.

Ratledge, C. (1982). The Biology of the Mycobacteria, Vol. 1, edited by C.

Ratledge & J. L. Stanford, pp. 185–271. London: Academic Press.

Roberts, F., Roberts, C. W., Johnson, J. J., Kyle, D. E., Krell, T., Coggins, G. H.,

Milhous, W. K., Tzipoki, S., Ferguson, D. J., Chakrabarti, D. & McLeod, R.

(1999). Nature (London), 397, 219–220.

Schlichting, I. & Reinstein, J. (1997). Biochemistry,

36, 9290–9296.

Ucho

a, H. B., Jorge, G. E., Silveira, N. J. F., Ca

mera, J. C., Canduri, F. &

Azevedo, W. F. Jr (2004). Biochem. Biophys. Res. Commun. 325, 1481–

1486.

Vonrhein, C., Schlauderer, G. J. & Schulz, G. F. (1995). Structure, 3, 483–

490.

World Health Organization (2006). Tuberculosis – Fact Sheet No. 104. http://

www.who.int/mediacentre/factsheets.

protein structure communications

6 Dias et al.



Shikimate kinase Acta Cryst. (2007). F63, 1–6

electronic reprint

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

3.5 Crystallographic studies on the binding of isonicotinyl-NAD adduct to wild-

type and isoniazid resistant 2-trans- Enoyl-ACP (CoA) Reductase from

Mycobacterium tuberculosis

Marcio V. B. Dias; Adriane M. X. Prado; Igor B. Vasconcelos; Valmir Fadel; Luis A.

Basso; Diógenes S. Santos; Walter F. de Azevedo Jr. Artigo aceito no Journal

Structural Biology, 2007.

Neste trabalho são apresentadas as estruturas cristalográficas da enzima Enoil

(ACP) redutase (InhA) (peso molecular de 28528 Da) em complexo com INH-NAD na

forma selvagem e para dois mutantes encontrados em isolados clínicos resistentes a

isoniazida (S94A e I21V). É apresentada também a estrutura do mutante S94A na sua

forma nativa.

As proteínas selvagens e os mutantes S94A e I21V foram expressadas e

purificadas pelo grupo do Prof. Dr. Diógenes S. Santos da PUC – Porto Alegre –RS. Os

cristais da InhA (selvagem e mutantes) em complexo com INH-NAD foram produzidos

seguindo as condições estabelecidas por Dessen et al., 1995 e Rozwarski et al., 1998.

Para a repetição destas condições as proteínas foram dialisadas contra HEPES, 50 mM,

pH 7,5. A condição de cristalização na qual foram obtidos os melhores cristais era

composta por HEPES, 50 mM, pH 7.2, citrato de sódio, 50 mM e 5-10% de MPD. Para

a cristalização do mutante S94A na forma nativa, a proteína foi dialisada contra HEPES,

50 mM, pH 7,5, 10% de glicerol e KCl, 300 mM. A condição na qual foram obtidos os

melhores cristais apresentava citrato de sódio, 100 mM, pH 5,6, acetato de amônio, 200

mM e 20-30% de PEG 4000. Os dados de difração de raios X foram coletados no

LNLS. Os cristais obtidos para a InhA em complexo com INH-NAD são pertencentes

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

ao grupo espacial P6

22 e os cristais obtidos para a proteína na sua forma nativa são

pertencentes ao grupo espacial P1. Todas as estruturas foram determinadas por

substituição molecular e refinadas pelo programa REFMAC 5.2 (MURSHUDOV;

VAGIN; DODSON, 1997).

Os dados obtidos para o complexo da InhA selvagem com INH-NAD apresenta

uma melhor resolução (2.2 Å) do que os dados anteriormente descritos na literatura (2.8

Å) (ROZWARSKI et al., 1998) . Desta maneira, é possível realizar uma melhor análise

do modo de interação desta droga com o sítio ativo da enzima, e evidenciar alterações

ainda não descritas. Além disso, esta estrutura, devido a sua melhor resolução, é mais

confiável para estudos a posteriori no desenvolvimento de novas drogas contra

tuberculose. As estruturas determinadas para os mutantes S94A e I21V contribuem para

a compreensão do mecanismo de resistência destes mutantes à isoniazida, além de

observar a influência que estas mutações causam sobre o processo de ligação do aducto

INH-NAD no sítio ativo da enzima. A estrutura do mutante S94A na sua forma nativa, é

a primeira estrutura apresentada neste estado, e tem sido obtida em uma nova condição

de cristalização e em um novo grupo espacial (P1). A estrutura apresenta um tetrâmero

na unidade assimétrica, que pode corresponder ao tetrâmero encontrado em solução.

Nesta estrutura, foi possível observar o estado da enzima antes da ligação de qualquer

ligante e, desta forma, pôde dar evidências sobre os movimentos de loops e resíduos que

estão envolvidos no processo de ligação de substratos acil ácidos graxos, da molécula de

NADH ou do aducto INH-NAD.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Crystallographic studies on the binding of isonicotinyl-NAD adduct to wild-type

and isoniazid resistant 2-trans- Enoyl-ACP (CoA) Reductase from Mycobacterium

tuberculosis

Marcio Vinicius Bertacine Dias

, Igor Bordin Vasconcelos

, Adriane Michele Xavier

Prado

, Valmir Fadel

, Luiz Augusto Basso

, Walter Filgueira de Azevedo Jr.

Diógenes Santiago Santos

Programa de Pós-Graduação em Biofísica Molecular – Departamento de Física,

UNESP, São José do Rio Preto, SP 15054-000, Brasil;

Centro de Pesquisa em Biologia Molecular e Funcional, Instituto de Pesquisas

Biomédicas, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, RS,

Brasil;

Faculdade de Biociências - Pontifícia Universidade Católica do Rio Grande do Sul,

Av. Ipiranga, 6681. Porto Alegre-RS CEP 90619-900 Brasil.

Corresponding authors: diogenes@pucrs.br (D.S. Santos) and

walter.junior@pucrs.br

(W.F. de Azevedo Jr.).

telephone: +55 51 3320-3500 4529

fax: +55 17 51 3220-3629

* Manuscript

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Abstract

The resumption of tuberculosis led to an increased need to understand the

molecular mechanisms of drug action and drug resistance, which should provide

significant insight into the development of newer compounds. Isoniazid (INH), the most

prescribed drug to treat TB, inhibits an NADH-dependent enoyl-acyl carrier protein

reductase (InhA) that provides precursors of mycolic acids, which are components of

the mycobacterial cell wall. InhA is the major target of the mode of action of isoniazid.

INH is a pro-drug that needs activation to form the inhibitory INH-NAD adduct.

Missense mutations in the inhA structural gene have been identified in clinical isolates

of M. tuberculosis resistant to INH. To understand the mechanism of resistance to INH,

we have solved the structure of two InhA mutants (I21V and S94A), identified in INH-

resistant clinical isolates, and compare them to INH-sensitive WT InhA structure in

complex with the INH-NAD adduct. We also solved the structure of unliganded INH-

resistant S94A protein, which is the first report on apo form of InhA. The salient

features of these structures are discussed and should provide structural information to

should improve our understanding of the mechanism of action of, and resistance to, INH

in M. tuberculosis. The unliganded structure of InhA allows identification of

conformational changes upon ligand binding and should help structure-based drug

design of more potent antimycobacterial agents.

Keywords: M. tuberculosis; InhA; Crystal structure; Isoniazid; drug resistance

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Introduction

Tuberculosis (TB), which is caused mainly by Mycobacterium tuberculosis, is a

global human health emergency that remains the leading cause of mortality among

infectious diseases. It has been estimated that 8.2 million new TB cases occurred

worldwide in the year 2000, with approximately 1.8 million deaths in the same year,

which translates into more than 200 deaths per hour, and more than 95 % of these were

in developing countries (Corbett et al., 2003). In the same year, 3.2 % of the world’s

new cases of TB were multidrug-resistant tuberculosis (MDR-TB), defined as strains

resistant to at least isoniazid and rifampicin (Espinal, 2003; Ormerod, 2005). Treatment

of MDR-TB strains requires the administration of second-line drugs that are more toxic

and less effective, and are given for at least three times as long as, and 100 times as

expensive as basic chemotherapeutic regimens (Pablos-Mendes et al., 2002). More

recently, a survey of the frequency and distribution of extensively drug-resistant (XDR)

TB cases, which are defined as cases in persons with TB whose isolates were resistant

to isoniazid and rifampicin and at least three of the six main classes of second-line

drugs, showed that during 2000-2004, of 17,690 TB isolates, 20% were MDR and 10%

of these were XDR (CDC, 2006). XDR-TB has a wide geographic distribution, poses a

public health threat, is an impediment to TB control, and opens up the possibility that

epidemics of virtually untreatable TB may develop (No authors listed, 2006). New

antimycobacterial agents are thus needed to improve the treatment of MDR- and XDR-

TB, as well as to provide more effective treatment of drug-sensitive TB infection. An

understanding of drug resistance mechanisms in this pathogen should contribute to the

rational design of new chemotherapeutic agents to treat TB.

The modern, standard “short-course” therapy for tuberculosis is based on a four-

drug regimen of isoniazid, rifampicin, pyrazinamide, and ethambutol or streptomycin

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

for two months, followed by treatment with a combination of isoniazid and rifampicin

for four months (Mitchison, 1985). Isoniazid (INH, isonicotinic acid hydrazide) was

first reported to be effective in the treatment of TB in 1952 (Bernstein et al., 1952) and,

soon after, the first INH-resistant M. tuberculosis strains were isolated (Middlebrook,

1953). Genetic and biochemical studies have shown that the inhA-encoded protein is the

primary target for isoniazid (Banerjee et al., 1994; Quémard et al., 1995; Larsen et al.,

2002; Kremer et al., 2003). InhA was identified as an NADH-dependent 2-trans enoyl-

ACP (acyl carrier protein) reductase enzyme that exhibits specificity for long-chain

thioester substrates. InhA is a member of the mycobacterial type II fatty acid synthase

system (FAS-II), which elongates acyl fatty acid precursors yielding the long carbon

chain of the meromycolate branch of mycolic acids, the hallmark of mycobacteria

(Schroeder

et al., 2002). INH is a pro-drug that is activated by the mycobacterial

catalase-peroxidase enzyme KatG in the presence of manganese ions, NAD(H) and

oxygen (Johnsson and Schultz, 1994; Johnsson et al., 1995; Basso et al., 1996; Zabinski

and Blanchard, 1997). The KatG-produced acylpyridine fragment of isoniazid is

covalently attached to the C4 position of NADH forming an INH-NAD adduct, which,

in turn, forms an inhibitory binary complex with the wild-type (WT) enoyl reductase of

M. tuberculosis (Rozwarski et al., 1998) with an equilibrium dissociation constant

value lower than 0.4 nM (Lei et al., 2000). The isonicotinyl-NAD adduct has been

shown to be a slow, tight-binding competitive inhibitor of WT InhA with an overall

inhibition constant value of 0.75 nM (Rawat et al., 2003).

The mechanism of action of isoniazid is complex, as mutations in at least five

different genes (katG, inhA, ahpC, kasA, and ndh) have been found to correlate with

isoniazid resistance (Schroeder

et al., 2002; Blanchard, 1996; Basso and Blanchard,

1998; Glickman and Jacobs, 2001; Basso and Santos, 2005, Oliveira et al., 2007).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Consistent with InhA as the primary target of INH mode of action, INH-resistant

clinical isolates of M. tub erculosis harboring inhA-structural gene missense mu tations,

but lacking mutations in the inhA promoter region, katG gene and oxyR-ahpC region,

were shown to have higher dissociation constant (K

) values for NADH than INH-

sensitive WT InhA, whereas there were only modest differences in the steady-state

parameters (Blanchard, 1996). We have recently reported the crystal structures of binary

complexes formed between NADH and INH-sensitive WT InhA, and INH-resistant

S94A, I47T, and I21V InhA mutant enzymes (Oliveira et al., 2006). Even more

recently, both specialized linkage transduction has been used to introduce S94A single

poin t mu tation within the inhA structural gene and X-ray crystallographic on INH-

resistant S94A InhA protein has been reported (Vilchèze et al., 2006). However, even

though there are several crystal structures of InhA in complex with a variety of ligands,

there has been no report on unliganded InhA structure and, thus, no high resolution

information on the InhA structure before ligand binding. In our efforts to understand the

molecular basis for the reduced inhibition of the INH-NAD adduct to InhA mutants,

here we report co-crystallization of INH-resistant I21V and S94A InhA mutant

enzymes, which were identified in INH-resistant clinical isolates of M. tuberculosis

(Blanchard, 1996; Morlock et al., 2003), with the INH-NAD adduct, and compare them

to the INH-sensitive WT InhA structure. This is the first report on the crystal structure

of the complex formed between INH-resistant I21V InhA and INH-NAD adduct refined

to 2.2 Å. We also report the crystal structure of INH-resistant S94A InhA and INH-

sensitive WT InhA both in complex with INH-NAD adduct to 2.0 and 2.2 Å of

resolu tion, respectively. Moreover, we report, for th e first time, the crystal structur e of

apo INH-resistant S94A InhA refined to 2.15 Å, which shows the protein

conformational changes upon ligand binding. It is hoped that the data presented here

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

will provide structural insight into an understanding of the drug resistance mechanism,

which, in turn, should aid the rational design of chemical compounds to efficiently

inhibit both INH-resistant and –sensitive InhA enzymes with potential

antimycobacterial activity.

Materials and methods

Crysta llizat ion

WT, I21V and S94A InhA enzymes were expressed and purified to homogeneity

as described elsewhere (Quèrmard et al., 1995; Basso et al., 1998). INH-NAD synthesis

was carried out as described elsewhere (Rozwarski et al., 1998) . Crystals of binar y

complex InhA:INH-NAD were obtained by the hanging-drop vapor–diffusion method

under similar conditions as described by Dessen et al., (1995) and Rozwarski et al.,

(1998). InhA enzymes were dialyzed against 50 mM Hepes, pH 7.5 and concentrated to

5-10mg/mL. Enzyme-inhibitor complexes were obtained at room temperature by

incubating 5-10 mg/mL InhA proteins with NADH, MnCl

and isoniazid for 1 hour

using molar ratios of 1:50, 1:10, and 1:100, respectively. The complexes InhA:INH-

NAD were crystallized in hanging droplets containing 1 µL of inhibited-protein solution

and 1 µL of crystallization solution containing 50 mM Hepes, pH 7.2; Sodium citrate

buffer and 5-10% MPD.

The apoenzyme S94A InhA mutant was dialyzed against 50 mM Hepes, pH 7.5,

10% glycerol and 300 mM of KCl. The protein was concentrated to 10 mg/mL and

crystallized in hanging droplets con taining 1 µL of protein solution and 0.5 µL of

crystallization solution containing 100 mM sodium citrate, pH 5.6, 200 mM ammonium

acetate and 20-30% of PEG 4000.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Data collection and processing

All data sets were collected at a wavelength of 1.427 Å using Synchrotron

Radiation Source (Station PCr, LNLS, Campinas, Brazil; and a CCD detector

(MARCCD) (Polikarpov et al., 1998). Data collection was performed in cryogenic

conditions at 100 K in a cold nitrogen stream generated and maintained with an Oxford

Cryosystem. Prior to flash-cooling, glycerol was added to the crystallization drop up to

20 % (v/v). The data sets were processed using the program MOSFLM (CCP4, 1994)

and scaled with SCALA (CCP4, 1994).

Structure determination

The crystal structures were determined by standard molecular replacement

methods using the program AMoRe (Navaza, 1994). Initially, atomic coordinates for

binary complex WTInhA:INH-NAD to 2.7Å resolution were used as search model for

the structure of WT InhA:INH-NAD (PDB access code: 1ZID) (Rozwarski et al., 1998).

Our atomic coordinates for WT InhA:INH-NAD at 2.2Å resolution were then used as

search model to solve the other structures. The structure of apoenzyme S94A InhA

mutant was solved using as search model the structure of binary complex InhA:Genz-

10850 at 2.6Å resolution (PDB access code: 1P44) (Kuo et al., 2003). The atomic

position s, which gen erated the higher correlation coefficien t magnitude obtained from

molecular replacement method, were used for the crystallographic refinement. The

refinements of structures were performed using REFMAC 5.2 program (Murshudov

al., 1997). The XtalView/Xfit (McRee, 1999) was used for visual inspection and

addition of water molecules. The water molecules were also checked based on B factor

values. The stereochemistry correctness of the models was checked using PROCHECK

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

program (Laskowski et al., 1993) and PARMODEL (Uchoa et al., 2004). The final

atomic models were superposed using the program LSQKAB program from CCP4i

package (CCP4, 1994).

Results and discussion

The crystals of apo S94A InhA are triclinic

These crystals diffracted to 2.15 Å

resolution. The asymmetric unit presents four monomers forming the characteristic

tetramer of InhA. This structure presents final R-factor and R-free values of 16.2% and

25.5%, respectively. The crystals of WT, I21V and S94A InhA in complex with the

INH-NAD adduct are hexagonal and crystallized in the space group P6

22, having one

molecule in the asymmetric unit. Table 1 summarizes the data processing,

crystallographic refinement statistics and structural quality for the four structures

presented here. Analysis of the crystallographic refinement for the structures here

described indicates that WtInhA:NADH:INH, I21V InhaA:NADH-INH and

S94Inha:NADH-INH present difference between R-factor and R-free ranging from 3.9

to 5.5 %, which indicates fairly good overall refinement statistics, and the structure

S94A InhA presents a difference of 9.3 %, below 10% , which makes this structure also

acceptable for structural comparisons. These structures present good geometry, although

some residues are located in regions not permitted in Ramachandran plots.

InhA belongs to the short chain dehydrogenase/reductase (SDR) family of

enzymes. Th e main ch aracteristic of th is family is a polyp eptide b ackbone topology in

wh ich each subunit consists of a single domain with a central core that contains a

Rossmann fold supporting an NADH binding site. The structure displays a α/β folding

consisting of a central β-sheet composed of parallel strands and flanked by α-helices

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

(Figure 1A). The structure presented here is in accordance with the homotetrameric

quaternary structure in solution determined by analytical size-exclusion

chromatography, and possesses an internal 222 symmetry (figure 1B).

Influence of INH on the structure of MtInhA

The activated form of INH consists of an isonicotinic-acyl group attached

through its carbonyl group to the C4 of the nicotinamide ring, replacing the 4S hydrogen

of NADH, which is the same position involved in the hydride transfer that occurs during

reduction of enoyl-ACP substrates (Quémard et al., 1995).

We compared the structure of INH-sen sitive WT InhA presented here and the

structure previously solved by Roswarski et al., (1998) at 2.7 Å of resolution. Although

Vilchèze et al., (2006) have solved two structures of InhA in complex with NAD-INH

(WT InhA and S94A mutant at 2.0 Å and 1.9 Å at resolution respectively) it was only

observed one structure deposited in the Protein Data Bank (S94A InhA mutant). This

way, the S94A InhA solved by Vilchèze et al., (2006) is argued in related to effect of

this mutation on the binding process of the INH-NAD at a later stage.

The structure of the INH-sensitive WT InhA presented here shows differences

when compared with the structure determined by Roswarski et al., (1998), mainly in

important residues located in the binding site of INH-NAD (Tyr158, Phe149, Trp222,

Leu218 and Phe41) and in the INH-NAD molecule. There is a difference of

approximately 15º in the position of Tyr158 side chain of WT InhA in comparison with

the structure solved by Roswarski et al., (1998), and the Tyr158 side chain moves closer

to INH. Interestingly, a conformational change involving rotation of Tyr158 side chain

upon binding of the enoyl substrate to InhA has been invoked to account for the

observed inverse solvent isotope effect (Parikh et al., 1999). These authors have

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

proposed that Tyr158 functions as an electrophilic catalyst, stabilizing the transition

state for hydride transfer by hydrogen bonding to the substrate carbonyl. The Phe41 side

chain of WT InhA:INH-NAD structure presented here underwent 30º torsion as

compared to the structure solved by Roswarski et al., (1998). This residue appears to be

important in anchoring the adenine moiety of NADH. The side chain of Phe149 and

Leu218 residues move closer to INH molecule (Figure 2A). Although it could be argued

that these differences are due to the lower resolution of the previously published

structure, these residues appear to play an importance role in the enzyme-ligand

interaction and should be considered in structure-based drug design.

The main effect of INH moiety of INH-NAD adduct on WT InhA is a 90º

rotation of Phe149 side chain (Figure 2B), which provides room to accommodate the

INH moiety. Owing to this rotation, a water molecule, which has been identified in the

InhA structures without the isonicotinic-acyl group (Oliveira et al., 2006), was removed

from the InhA active site (not shown in figure 2). Furthermore, it is observed alterations

in the Met155, Leu218 and Ile215, Met199, probably due to alteration in the side chain

position of Phe149. The Ile215 moves away from the active site due to rotation of the

side chain of the Tyr158 (Figure 2B). The side chain of Met199 underwent a rotation of

approximately 30º away from the active site. Two water molecules are present in the

structure of InhA:INH-NAD complex, but absent from the structures between WT and

S94A, I21V and I47T mutants InhA proteins in complex with NADH (Oliveira et al.,

2006). The isonicotinic group also causes the expulsion of one water molecule next to

C4 of NADH molecule. This causes a rotation of approximately 20º of nicotinamide

group of NADH molecule, when compared with other InhA:NADH structures, but the

phosphates, ribose and adenine are highly superposed among the analyzed structures.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

This change results in a more snugly fit of the INH moiety of INH-NAD in the active

site of InhA.

There is one intermolecular hydrogen bond that is conserved in the present

structures and in the structure reported by Roswarski et al., (1998) involving oxygen

atom from Gly14 and O2 from the pentose of NADH, which strongly indicates the

importance of this hydrogen bond for the interaction of NADH and the InhA.

Influence of mutations I21V and S94A on INH-NAD binding to InhAs

Van der Waals contacts between Val21 side chain and the nicotinamide ring,

nicotinamine ribose, and phosphate oxygen atoms are missing in the INH-resistant I21V

InhA structure as compared to WT-sensitive InhA. These interactions do not occur due

to the absence of the CD1 atom in the mutated residue Val21 (Oliveira et al., 2006),

which appear to play an important role in stabilizing bound NADH or INH-NAD in the

InhA active site (Oliveira et al., 2006; Basso et al., 1998).

The crystal structure of S94A InhA showed that a conserved hydrogen bond to a

water molecule is lost owing to mutation to Ala94 (Oliveira et al., 2006), which has

been proposed to account for the reduction in affinity for NADH observed for the INH-

resistant S94A InhA enzyme (Basso et al., 1998). Vilchèze et al. (2006) have solved the

structure of INH-resistant S94A InhA enzyme (PDB code 2NV6) and INH sensitive

InhA in complex with INH-NAD (data do not deposited in the PDB). These authors

propose the loss of the serine causes a shift in position of water molecule that promotes

a disruption in hydrogen binding network. Analysis of the structure of S94A InhA

enzyme present here does not present the disruption between this water molecule and

the atom O9 of the molecule of INH-NAD. Analysis of other structures of S94A InhA,

such a ones solved by Oliveira et al. (2006), shows a binding between this conserved

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

water molecule and the O9 of the molecule of INH-NAD. Thus, we believed that the

mutation S94A can cause increase flexibility or decrease affinity of this molecule due

the loss binding between this water molecule and the OG atom of the Ser94, which is

observed in the structure wild type. The influence of the mutation on this water

molecule could cause a reduction in the affinity of INH-NAD or NADH by InhA S94A.

Thus, the movement observed in the water molecule by Vilchèze et al. (2006) can be

due an increase of flexibility of this molecule carrying a false impression of an ordered

movement.

The effect of the mutations I21V and S94A on binding process of the NADH on

protein is summarized in the table 2. Analysis of contact area between the InhAs and

INH-NAD reveals that the INH-sensitive WT enzyme presents larger contact area than

the INH-resistant mutants. The value for WT enzyme is 467.4Å

while for S94A and

I21V mutants the values are 464.2 Å

and 461.9 Å

, respectively. The smaller value

observed for I21V mutant can be due the absence of the contact between the CD1 of

Ile21 that is missing in the I21V mutant. The I21V and S94A mutations do not seem to

alter the position of the isonicotinic-acyl group because there are no significant changes

when the mutan t structures are compared with WT type structure (figure 3A-C). These

results are in agreement with the proposal that INH resistance is due to reduction in

NADH affinity for mutant InhAs (S94A and I21V) thereby hampering binding of INH-

NAD adduct.

Effect of INH-NAD on the binding of substrate in the InhA

The isonicotinic-acyl group of INH-NAD adduct replaces the 4S hydrogen of C4

of NADH, which is the same position involved in the hydride transfer catalyzed by

InhA. As a member of the family of SDR enzymes, the catalytic triad formed by

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Phe149, Tyr158 and Lys165 (Rozwarski et al., 1999) is present in InhA (Figure 4). The

Phe149 may play a role in anchoring the nicotinamide moiety of NADH for hydride

transfer to the fatty acyl substrate. Lys165 interacts with the 3’-hydroxyl oxygen of the

nicotinamide ribose of NADH, in agreement with a previous proposal that Lys165 plays

a role in cofactor binding (Parikh et al., 1999).

The proposed mechanism of how InhA catalyzes the reduction of the 2-trans

double bond of the fatty acyl substrate consists of the formation of an enolate

intermediate through the direct transfer of a hydride ion from NADH to position C3 of

the substrate, followed by protonation of position C2 (Rozwarski et al., 1999). It is

noteworthy that the Tyr158 occupies approximately the same position for S94A:INH-

NAD and apo S94A, whereas there is a conformational change upon binding of NADH

to S94A, and NAD

and a C16 enoyl substrate to WT InhA resulting in position of

Tyr158 to similar positions in these latter structures (figure 4). A comparison between

the WT InhA-NADH structure and an inactive ternary complex formed by WT InhA,

NAD

, and a C16 enoyl substrate revealed that upon binding the enoyl substrate there is

a 60° rotation about the Tyr158 Cα-Cβ bond that enables the Tyr158 to hydrogen bond

to the substrate carbonyl group (Parikh et al., 1999). It has thus been suggested that

Tyr158 provides electrophilic stabilization of the transition state(s) for the reaction by

hydrogen bonding to the carbonyl of the substrate. However, the InhA:NAD:C16 enoyl

substrate is a non-productive ternary complex and is likely not part of the reaction

course, which is in agreement with the structural results presented here for apo S94A

InhA. The isonicotyl-acyl group causes a large change in the position of the Phe149 side

chain (figure 4), and appears to be involved in stacking interactions with the INH-NAD

adduct.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

INH-resistant S94A mutant in an uncomplexed form

The structure of unliganded S94A InhA reveals that the protein undergoes three

noticeable conformational changes in its main chain upon NADH binding. The main

alterations occur between the residues 99 and 112, between the residues 197 and 213

(figure 5A), and in the Phe41 (figure 5B). It appears that NADH binding is sufficient to

cause a closure of the active site of InhA. Analysis of B-factor for three different

structures of InhA (figure 6) shows that in the absence of both substrate and cofactor or

in the presence of ones, the protein presents the substrate binding loop (residues 196-

219) more disordered than the stru cture in the presence of cofactor. By analysis of these

three structure, we can observe that in the uncomplexed structure, part of the substrate

binding loop is closed (197-203) and part is opened (205-214) for InhA:NAD

:substrate

structure (figure5A). In the S94A:NADH structure, part of the substrate binding loop is

more open (197-203) whereas part is more closed (205-214). The conformational

change of InhA:NAD

:substrate ternary complex is due to the geometry of the α-helix

of the longer substrate binding loops that could sterically hinder the loop from folding

downward into the substrate binding cavity, leaving a larger opening for the fatty acyl

substrate between residues 205-214 (Rozwarski et al., 1999). The residues 205-214 of

the substrate binding loop moves approximately 6 Å away from the active site in the

unliganded InhA, while the substrate of InhA:NAD

:substrate presents the same loops

away approximately 4Å. The opening of the substrate binding loop for these structures

are consistent with apoenzyme providing access to substrates to enter the enzyme active

site and non-productive ternary complex providing access to solvent for product release.

Analysis of molecular surface shows the effect of the substrate binding loop on the

active site of InhA (figure 7).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

In this manner, the binding of NADH in the active site of the protein can cause

the partial closure of substrate binding loop however arranging a conformation for the

binding of the substrate. In the presence of the co-factor and substrate the protein

presents the active site newly opened for liberation of the product of the reaction. Thus,

in accord ing with th is hypothesis the molecule of NADH is the main responsible by

closure of the active site of InhA.

The van der Waals interactions between the Phe41 and NADH molecule may

play in NADH binding to the enzyme active site, since Phe41 moves closer to the

adenine moiety upon cofactor binding (figure 5B). Asp42 and Arg43 side chains also

undergo conformational changes upon binding of NADH or both NADH and substrate

(figure 5B). However, the absence of substrate does not seem to alter the oligomeric

state of InhA because the tetrameric form in the asymmetric unit and the contact area

among the monomers in the unliganded structure are similar to the other structures

(InhA:NAD

:substrate ternary complex and InhA:NADH binary complex) (data not

show). Thus the movement of substrate binding loop appears not to interfere with the

protein oligomeric state.

Conclusion

Here we present four structures of InhA from M. tuberculosis: three structures

for complexes of InhA:INH-NAD and one in the unliganded form. The structures in

complex with INH-NAD are: INH-sensitive WT and two INH-resistant mutants (I21V

and S94A). Comparison between our WT InhA:INH-NAD structure and the structure

previously determined by Roswarski et al., (1998)

reveals that there are changes in

important residues in the active site that were not previously observed. Moreover, the

comparison of our structures in complex with INH-NAD shows that there is no large

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

influence of mutation in the binding of INH-NAD. The INH-NAD adduct has been

shown to be a slow, tight-binding competitive inhibitor of WT and INH-resistant InhA

mutant enzymes (Rawat et al., 2003). Interestingly, the kinetic and thermodynamic

parameters for the interaction of isonicotinyl-NAD

adduct with INH-resistant I21V,

I47T, and S94A InhA mutant enzymes were found to be similar to those of the WT

enzyme (Rawat et al., 2003). These results prompted the authors to suggest an

alternative hypothesis to explain for INH resistance mechanism in strains harboring

inhA-structural gene mutations, in which Inh A may interact directly with o ther

components of the FAS-II system. Accordingly, the resistance-associated mutations in

the inhA-structural gene would affect the susceptibility of InhA to INH inhibition only

in the context of the multienzyme complex, and not when InhA is tested in isolation as

in in vitro assays. Several protein-protein interactions between FAS-II enzymes have

been detected by yeast two-hybrid and co-immunoprecipitation studies and proposed

that either these complexes might coexist or the quaternary structure of a “unique” FAS-

II might change from one composition to another during the time and according to the

degree of elongation of the substrate (Veyron-Churlet et al., 2004). In particular, M.

tuberculosis InhA was shown to interact with KasA (β-ketoacyl synthase A) and this

protein-protein interaction has been suggested as a probable explanation to occurrence

of INH-resistant mutant in KasA, even if InhA is indeed the only primary target of INH.

However, it remains to be shown whether inhA structural gene mutations identified in

INH-clinical isolates of M. tuberculosis will affect the inhibition of InhA by INH in the

context of, for instance, InhA-KasA multienzyme complex. In agreement with InhA as

the primary target for INH mode of action, recessive mutations in M. smegmatis and M.

bovis BCG ndh gene, which codes for a type II NADH dehydrogenase (NdhII), have

been found to increase intracellular NADH/NAD ratios (Vilchèze et al., 2005).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Increasing NADH levels protected InhA against inhibition by the INH-NAD adduct

formed upon KatG activation of INH. Hence, mutations in mycobacterial ndh gene

resulted in increased intracellular NADH concentrations, which competitively inhibits

the binding of INH-NAD adduct to InhA, in agreement with the higher dissociation

constant values for NADH found for INH-resistant clinical isolates harbouring inh A-

structural gene mutations as compared to WT InhA (Basso et al., 1998). Moreover, it

has been shown that subtle structural changes result in increased values for the limiting

dissociation rate constant for NADH from INH-resistant mutants (Oliveira et al., 2006).

These observations are in agreement with the results described here.

The unliganded structure presented here shows conformational changes that

occur upon binding of substrates (NADH and/or fatty acyl substrate). The structure

presents movements in the substrate binding loop and a striking change in the position

of Phe41, which may play an important role in NADH stabilization in the enzyme active

site. NADH or substrate binding causes no apparent change in the oligomeric state of

InhA.

The results presented here provide structural information that should better our

understanding of the mechanism of action of, and resistance to, isoniazid in M.

tuberculosis. The unliganded structure of InhA allows identification of conformational

changes upon ligand binding that are important for substrate anchoring. We hope that

these results will help structure-based drug design of more potent antimycobacterial

agents.

PDB Accession codes:

Protein Data Bank: Atomic coordinates and structure factors have been deposited with

accession codes: 2idz, 2ie0, 2ieb, 2ied.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Acknowledgments: Financial support for this work was provided by Millennium

Initiative Program MCT-CNPq, Ministry of Health-Department of Science and

Technology (DECIT)-UNESCO (Brazil), and PRONEX/CNPq/FAPERGS (Brazil) to

D.S.S. and L.A.B. This work was also supported by grants from FAPESP (SMOLBNet,

proc. 01/07532-0, 03/12472-2, 04/00217-0) to WFA. D.S.S. (304051/1975-06), L.A.B.

(520182/99-5), and WFA (CNPq, 300851/98-7) are research career awardees from

National Research Council of Brazil (CNPq).

References

Banerjee, A., Dubnau, E., Quémard, A., Balasubramanian, V., Um, K.S., Wilson, T.,

Collins, D., de Lisle, G., Jacobs, W.R. Jr., 1994. inhA, a gene encoding a target for

isoniazid and ethionamide in Mycoba cterium tuberculosis. Science 263, 227-230.

Basso, L.A., Blanchard, J.S., 1998. Resistance to antitubercular drugs. Adv. Exp. Med.

Biol. 456, 115-144.

Basso, L.A., Santos, D.S., 2005. Drugs that inhibit mycolic acid biosynthesis in

Mycobacterium tuberculosis – an update. Med. Chem. Ver. – online 2, 393-413.

Basso, L.A., Zheng, R., Blanchard, J.S., 1996. Kinetics of inactivation of WT and

C243S mutant of Mycobacterium tuberculosis enoyl reductase by activated

isoniazid. J. Am. Chem. Soc. 118, 11301-11302.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Basso, L.A., Zheng, R., Musser, J.M., Jacobs, W.R. Jr., Blanchard. J.S., 1998.

Mechanisms of isoniazid resistan ce in Mycobacterium tuberculosis: enzymatic

characterization of enoyl reductase mutants identified in isoniazid-resistant clinical

isolates. J. Infect. Dis. 178, 769–775.

Bernstein, J.W., Lott, A., Steinberg, B.A., Yale, H.L., 1952 Chemotherapy of

experimental tuberculosis. V. Isonicotinic acid hydrazide (nydrazid) and related

compounds. Am. Rev. Tuberc. 65, 357-374.

Blanchard, J.S., 1996. Mo lecular mechanisms of dru g resistan ce in Mycobacterium

tuberculosis. Ann. Rev. Biochem. 65, 215-239.

CDC (Centers for Disease Control and Prevention), 2006. Emergence of

Mycobacterium tuberculosis with extensive resistance to second-line drugs –

worldwide, 2000-2004. Morb. Mortal Wkly. Rep. 55, 301-305.

Collaborative Computational Project, Number 4., 1994. The CCP4 suite: programs for

protein. Acta Crystallog. Sect. D 50, 760–763.

Corbett, E.L., Watt, C.J., Walker, N., Maher, D., Williams, B.G., Raviglione, M.C.,

Dye, C., 2003. The growing burden of tuberculosis: global trends and interactions

with the HIV epidemic. Arch. Intern. Med. 163, 1009-1021.

DeLano, W.L., 2004. The PyMOL Molecular Graphics System. CA DeLano Scientific.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Dessen, A., Quémard, A., Blanchard, J.S., Jacobs, W.R. Jr., Sacchettini, J.C., 1995

Crystal structure and function of the isoniazid target of Mycobacterium tuberculosis.

Science 267, 1638–1641.

Espinal, M.A., 2003. The global situation of MDR-TB. Tuberculosis: 83, 44-51.

Glickman, M.S., Jacobs, W. R. Jr., 2001 Microbial pathogenesis of Mycobacterium

tuberculosis: dawn of a discipline. Cell, 104, 477-485.

Guex, N., Peitsch, M.C., 1997. SWISS-MODEL and the Swiss-PdbViewer: An

environment for comparative protein modeling. Electrophoresis 18, 2714-2723.

Johnsson, K., King, D.S., Schultz, P.G., 1995. Studies on the mechanism of action of

isoniazid and ethionamide in the chemotherapy of tuberculosis. J. Am. Chem. Soc.

117, 5009-5010.

Johnsson, K., Schultz, P.G., 1994. Mechanistic studies of the oxidation of isoniazid by

the catalase peroxidase from Mycobacterium tuberculosis. J. Am. Chem. Soc. 116,

7425-7426.

Koradi, R., Billeter, M., Wüthrich, K., 1996. MOLMOL: a program for display and

analysis of macromolecular structures. J. Mol. Graph. 14 (1), 51-55.

Kremer, L., Dover, L.G., Morbidoni, H.R., Vilchèze, C., Maughan, W.N., Baulard, A.,

Tu, S.C., Honore, N., Deretic, V., Sacchettini, J.C., Locht, C., Jacobs, W.R. Jr.,

Besra, G.S., 2003. Inhibition of InhA activity, but not KasA activity, induces

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

formation of a KasA-containing complex in mycobacteria. J. Biol. Chem. 278,

20547-20554.

Kuo, M.R., Morbidoni, H.R., Alland, D., Sneddon, S,F., Gourlie, B.B., Staveski, M.M.,

Leonard, M., Gregory, J.S., Janjigian, A.D., Yee, C., Musser, J.M., Kreiswirth, B.,

Iwamoto, H., Perozzo, R., Jacobs. W.R. Jr., Sacchettini. J.C., Fidock, D.A., 2003.

Targeting tuberculosis and malaria through inhibition of enoyl reductase:

Compound activity and structural data. J. Biol. Chem. 278(23), 20851-20859.

Larsen, M.H., Vilchèze, C., Kremer, L., Besra, G.S., Parsons, L., Salfinger, M., Heifets,

L., Hazbon, M.H., Alland, D., Sacchettini, J.C., Jacobs, W.R. Jr., 2002.

Overexpression of inhA, but not kasA, confers resistance to isoniazid and

ethionamide in Mycobacterium smegmatis, M. bovis BCG and M. tuberculosis. Mol.

Microbiol. 46, 453-466.

Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK:

a program to check the stereochemical quality of protein structures. J. Appl.

Crystallog. 26, 283–291.

Lei, B., Wei, C.-J., Tu, S.-C., 2000. Action mechanism of antitubercular isoniazid. J.

Biol. Chem. 275, 2520-2526.

McRee, D.E., 1999. XtalView/Xfit—a versatile program for manipulating atomic

coordinates and electron density. J. Struct. Biol. 125, 156–165.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Middlebrook, G., Cohn, M.L., 1953. Some observations on the pathogenicity of

isoniazid-resistant variants of tubercle bacilli. Science 118, 297-299.

Mitchison, D.A., 1985. The action of antituberculosis drugs in short-course

chemotherapy. Tubercle 66, 219-225.

Morlock, G.P., Metchock, B., Sikes, D., Crawford, J.T., Cooksey, R.C., 2003. ethA,

inhA, and katG loci of ethionamide-resistant clinical Mycobacterium tuberculosis

isolates. Antimicrob. Agents Chemother. 47, 3799–3805.

Murshudov, G.N., Vagin. A.A., Dodson. E.J., 1997. Refinement of macromolecular

structures by the maximum-likelihood method. Acta Crystallog. sect. D 53, 240–

255.

Navaza, J., 1994. AMoRe: an automated package for molecular replacement. Acta

Crystallog. Sect. A 50. 157–163.

No authors listed, 2006 Addressing the threat of tuberculosis caused by extensively

drug-resistant Mycobacterium tuberculosis. Wkly Epidemiol. Rec. 21, 386-390.

Oliveira, J.S., Pereira, J.H., Canduri, F., Rodrigues, N.C., de Souza, O.N., de Azevedo,

W.F. Jr., Basso, L.A., Santos, D.S., 2006. Crystallographic and pre-steady-state

kinetics studies on binding of NADH to wild-type and isoniazid-resistant enoyl-

ACP(CoA) reductase enzymes from Mycobacterium tuberculosis. J. Mol. Biol.

359(3), 646-66.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Oliveira, J. S. Vasconcelos, I. B., Moreira, I. S., Santos, D. S., Basso, L. A., 2007.

Enoyl reductases as targets for the development of anti-tubercular and antimalarial

agents. Curr. Drug Targets 8(3), 399-411.

Ormerod, L.P., 2005. Multidrug-resistant tuberculosis (MDR-TB): epidemiology,

prevention and treatment. British Medical Bulletin 73-74(1), 17-24.

Pablos-Mendez, A., Gowda, D.K., Frieden, T.R., 2002. Controlling multidrug-resistant

tuberculosis and access to expensive drugs: a rational framework. Bull World Health

Organ 80, 489-495.

Parikh, S., Moynihan, D.P., Xiao, G., Tonge, P.J., 1999. Roles of tyrosine 158 and

lysine 165 in th e catalytic mechanism of InhA, the enoyl-ACP reductase from

Mycobacterium tuberculosis. Biochemistry 38, 13623-13634.

Polikarpov, I., Perles, L.A., de Oliveira, R.T., Oliva, G., Castellano, E.E., Garratt, R.C.,

Craievich, A., 1998. Set-up and experimental parameters of the protein

crystallography beam line at the Brazilian National Synchrotron Laboratory. J.

Synchrotron Radiat. 5, 72–76.

Quémard, A., Sacchettini, J.C., Dessen, A., Vilcheze, C., Bittman, R., Jacobs, W.R. Jr.,

Blanchard, J.S., 1995. Enzymatic characterization of the target for isoniazid in

Mycobacterium tuberculosis. Biochemistry 34, 8235-8241.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Rawat, R., Whitty, A., Tonge, P., 2003. The isoniazid–NAD adduct is a slow, tight-

binding inhibitor of InhA, the Mycobacterium tuberculosis enoyl reductase: adduct

affinity and drug resistance. Proc. Natl. Acad. Sci. USA 100, 13881–13886.

Rozwarski, D.A., Grant, G.A., Barton, D.H.R., Jacobs, W.R., Sacchettini, J.C., 1998.

Modification of the NADH of the isoniazid target (InhA) from Mycobacterium

tuberculosis. Science 279, 98-102.

Rozwarski, D.A., Vilchèze, C., Sugantino, M., Bittman, R., Sacchettini, J.C., 1999

Crystal structure of the Mycobacterium tuberculosis enoyl-ACP reductase, InhA, in

complex with NAD

and a C16 fatty acyl substrate. The Journal of Biological

Chemistry 274(22), 15582-15589.

Schroeder, E.K., de Souza, O.N., Santos, D.S., Blanchard, J.S., Basso, L.A., 2002 Drugs

that inhibit mycolic acid biosynthesis in Mycobacterium tuberculosis. Curr. Pharm.

Biotechnol. 3, 197–225.

Uchoa, H.B.; Jorge, G.E.; Freitas Da Silveira, N.J.; Camera, J.C.,Jr.; Canduri, F.; De

Azevedo, W.F.,Jr. Biochem. Biophys. Res. Commun., 2004, 325, 1481-1486.

Veyron-Churlet, R., Guerrini, O., Mourey, L., Daffe, M., Zerbib, D., 2004. Protein-

protein interaction s within the fatty acid synthase-II system of Mycobacterium

tuberculosis are essential for mycobacterial viability. Mol. Microbiol. 54, 1161-

1172.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Vilchèze, C., Morbidoni, H.R., Weisbrod, T.R., Iwamoto, H., Kuo, M., Sacchettini,

J.C., Jacobs, W.R. Jr., 2000. Inactivation of the inhA-encoded fatty acid synthase II

(FASII) enoyl-acyl carrier protein reductase induces accumulation of the FASI end

products and cell lysis of Mycobacterium smegmatis. J. Bacteriol. 182, 4059-4067.

Vilchèze, C., Wang, F., Arai, M., Hazbon, M.H., Colangeli, R., Kremer, L., Weisbrod,

T.R., Alland, D., Sacchettini, J.C., Jacobs, W.R. Jr., 2006. Transfer of a point

mutation in Mycobacterium tuberculosis inhA resolves the target of isoniazid. Nat

Med: 12, 1027-1029.

Vilchèze, C., Weisbrod, T.R., Chen, B., Kremer, L., Hazbon, M.H., Wang, F., Alland,

D., Sacchettini, J.C., Jacobs, W.R. Jr., 2005. Altered NADH/NAD

ratio mediates

coresistance to isoniazid and ethionamide in mycobacteria. Antimicrob. Agents

Chemother. 49, 708-720.

Zabinski, R.F., Blanchard, J.S., 1997. The requirement for manganese and oxygen in the

isoniazid-dependent inactivation of Mycobacterium tuberculosis enoyl reductase. J.

Am. Soc. Chem. 119, 2331-2332.

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figures

Figure 1. Ribbon diagram of InhA. A) Monomeric structure of the wild type InhA:INH-

NAD binary complex, and B) Tetrameric structure of the unliganded InhA. The figures

were drawn using program MolMol (Koradi et al., 1996).

Figure 2. Superposition of active site residues of InhA. A) It is shown the superposition

of the active site structure of wild type InhA: INH-NAD solved by Roswarski et al.,

(1998) at 2.7Å resolution (in green) and wild type InhA: INH-NAD at 2.0 Å resolution

presented here (in grey). B) It is shown the superposition of the structure of wild type

InhA:NADH complex binary solved by Oliveira et al., (2006) (in blue) and the

structure of wild type InhA: INH-NAD presented here (in grey). The figures were

generated with MolMol (Koradi et al., 1996).

Figure 3. Ribbon diagrams for InhA structures, showing the electronic density for INH-

NAD and mutated residue. A) Wild type InhA, B) I21V mutant and C) S94A mutant.

The figures were generated using program SpdbView (Guex and Peitsch, 1997).

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure 4. Catalytic triad for InhA. In green is shown the residues of S94A InhA in

complex with INH-NAD, in grey is shown the residues of S94A InhA unliganded, in

orange is shown the residues of S94A in complex with NADH, and in blue is shown the

residues of WT InhA in complex with C16 enoyl substrate and NAD

(Rozwarski

et al.,

1999. The Figure shows conformational changes in these residues due to different

ligands. The figure was generated using program Molmol (Koradi et al., 1996).

Figure 5. Conformational changes observed in the structure of S94A InhA upon

binding of NADH or both NADH and substrate. A) Cα traces of S94A InhA unliganded

(purple), S94AInhA in complex with NADH (orange), and WT InhA in complex with

both NAD

and substrate (green). B) Conformational changes of residues 41-43

(coloring was the same as in A). The figures were generated using program MolMol

(Koradi et al., 1996).

Figure 6. Residue-averaged B factor for S94A resistant mutant apoenzyme (red line),

WT InhA in complex with NAD

and C16 substrate (black line), and S94A resistant

mutant in complex with NADH (purple line).

Figure 7. Molecular surface for A) Unliganded S94A InhA, B) S94A InhA in complex

with NADH, and C) WT InhA in complex with both NADH and substrate. The figures

were generated using program Pymol (DeLano, 2004).

100

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Table 1. Crystallographic data, refinement statistics and analysis of the quality of InhA structures

Crystallographic data

WtInhA:NADH:INH I21V InhA:NADH-INH S94A InhA:NADH-INH S94A InhA

a (Å) 97.09 96.59 96.83 54.63

b (Å) 97.09 96.59 96.83 63.52

c (Å) 136.94 136.35 136.42 65.18

90 90 90 97.21

90 90 90 85.81

120 120 120 102.87

Space group P6222 P6222 P6222 P1

Number of image 80 60 80 160

Number of measurements with I>2 σ (I)

245823 134890 185407 77775

Number of independent reflections 26409 19563 19782 44054

Completeness (%) (Outermost shell) 99.9 (100) 98.7 (91.6) 99.8 (99.5) 98.6 (85.2)

sym

(%)

(Outermost shell) 9.8 (83.0) 7.2 (59.1) 9.5 (77.4) 5.8 (29)

Redundancy 9.3 6.9 9.4 1.6

Refinement statistics

Resolution range (Å) 40.19-2.0 39.41-2.2 45.64-2.2 42.26-2.14

Reflections used for refinement 25045 18527 18193 41818

Number of atoms 2283 2261 2257 8682

Number of residues 267 267 267 1068

Number of water molecules 239 216 212 701

Final R-factor (¨%)

18.7 18.4 17.9 16.2

Final R-factor free (%)

22.6 23.6 23.4 25.5

Correlation Coefficient (%) 96.0 95.8 95.8 95.8

B values (A

)

Main chain 30.56 36.69 34.08 27.52

Side chain 32.19 37.70 35.33 28.59

NADH-INH 32.19 31.93 29.85

Waters 40.93 43.39 42.13 33.23

Table1_rev

101

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Ramachandran plot

Favorable 86.2 87.1 88.9 88

Additional allowed 9.8 9.8 9.3 10.3

Generously allowed 0.0 0.0 0.0 0.7

Disallowed 4.0 3.1 1.8 1.0

sy m

= 100 P I (h) _ Æ I (h)æ /P I (h) with I (h), observed intensity and ÆI (h)æ, mean intensity of reflection h overall measurement of I (h).

R-factor = 100 × ∑( F

obs

− F

calc

)/∑ (F

obs

), the sums being taken over all reflections with F/σ(F) > 2σ(F).

R -free = R-factor for 10% of the data that were not included during crystallographic refinement.

102

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Table 2. Effect of the mutations I21V and S94A on the binding process of the NADH on

the InhA.

mutation Mutation effect consequence

I21V

Lose of Van der Waals

interaction between NADH

and the CD1 atom present in

the valine residue

Decrease of the stability of

the binding of the NADH in

the active site of the protein

S94A

Alteration in the binding

network involving a

conserved water molecule

and O9 atom of molecule of

NADH

Increase of the flexibility of

the conserved water

molecule and decrease of

the affinity of NADH by

protein

Table2_rev

103

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure1a_rev

Click here to download high resolution image

104

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure1b_rev

Click here to download high resolution image

105

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure2a_rev

Click here to download high resolution image

106

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure2b_rev

Click here to download high resolution image

107

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure3a_rev

Click here to download high resolution image

108

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure3c_rev

Click here to download high resolution image

109

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure3b_rev

Click here to download high resolution image

110

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure4_rev

Click here to download high resolution image

111

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure5a_rev

Click here to download high resolution image

112

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure5b_rev

Click here to download high resolution image

113

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure6_rev

Click here to download high resolution image

114

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure7a_rev

Click here to download high resolution image

115

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure7b_rev

Click here to download high resolution image

116

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Figure7c_rev

Click here to download high resolution image

117

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

4. CONCLUSÃO

Este trabalhou teve como principal objetivo análise estrutural de enzimas que são

importantes para a viabilidade de M. tuberculosis, mas que, entretanto tais enzimas

estivessem ausentes em humanos. Assim essas enzimas são consideradas como alvos para

o desenvolvimento de novas drogas contra a tuberculose e podem apresentar baixa

toxidade ou ausência de efeitos colaterais. Após os estudos realizados com cada uma das

enzimas chegamos as seguintes conclusões para cada uma delas separadamente:

- Foi determinada a primeira condição de cristalização para a proteína corismato

sintase de M. tuberculosis, na qual foram obtidos cristais hexagonais, grupo

espacial P6

22. Estes cristais difrataram em torno de 2,65Å de resolução e os dados

do processamento de difração de raios X foram utilizados para a resolução da

estrutura cristalográfica desta proteína;

- Foi determinada a estrutura cristalográfica da corismato sintase de Mycobacterium

tuberculosis em seu estado nativo a 2,65Å de resolução. A estrutura apresenta um

folding bastante similar às outras estruturas de corismato sintase de outros

microorganismos. Os resíduos do sítio ativo também são bastante conversados,

entretanto a posição de alguns e resíduos apresenta diferente conformações quando

comparado com outras estruturas. Essas diferenças podem ser devido à ausência de

ligantes. A determinação de estruturas da corismato sintase de M. tuberculosis em

complexo com seus substratos podem esclarecer essas evidências;

- Foram construídos seis modelos estruturais para a proteína triptofano sintase de M.

tuberculosis com diferentes inibidores análogos de substratos através da técnica de

modelagem molecular comparativa. Os modelos obtidos apresentam qualidade

estereoquímica satisfatória e mostram as interações entre estes ligantes e a enzima.

118

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Pela análise da ligação podemos concluir que os inibidores estudados apresentam

uma interação satisfatória com a enzima. Testes in vitro ou in vivo poderão

confirmar essa hipótese. Entretanto os modelos obtidos podem ser utilizados na

construção de outros inibidores mais potentes para esta enzima.

- Foram apresentadas também duas estruturas da proteína chiquimato quinase de

Mycobacterium tuberculosis (uma em complexo com ADP e chiquimato e outra em

complexo com MgADP). A estrutura da chiquimato quinase em complexo com

ADP e chiquimato mostra que a ausência do íon magnésio pode influenciar a

posição dos grupos hidroxilas da molécula de chiquimato e também em alguns

importantes resíduos do sítio ativo da enzima, como o Asp34. A estrutura da

chiquimato quinase em complexo com MgADP foi obtida em uma nova condição

de cristalização e apresenta-se como um tetrâmero na unidade assimétrica, com

simetria 222. Nesta estrutura pôde-se observar o possível efeito do chiquimato

sobre o empacotamento cristalino da enzima. Além disso, devido à disposição do

monômero na unidade assimétrica foi possível observar que a molécula ADP tem

uma importante participação na estabilidade da estrutura quaternária da proteína,

pelo menos na rede cristalina. Um dos monômeros do tetrâmero observado para a

chiquimato quinase não apresentava o íon cloreto e assim é possível que ele tenha

uma importante participação da orientação da molécula de ADP no sítio ativo da

enzima, no fechamento do LID domain e também na formação de interações com

moléculas de águas que podem levar o fechamento da enzima.. Desta forma, este

íon pode ter importante influência na estabilidade e no mecanismo de catálise da

enzima. Assim, as informações obtidas neste trabalho podem ser úteis para o

entendimento do mecanismo catalítico desta enzima e também para o desenho de

novos ligantes que tenham atividade inibitória.

119

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

- Foram apresentadas as estruturas cristalográficas da proteína Inha selvagem e de

dois mutantes isolados clínicos resistentes a isoniazida (I21V e S94A) complexada

com o aducto NAD-INH. A partir destas estruturas é possível determinar as

alterações moleculares que a molécula de NAD-INH realiza no sítio ativo da

enzima e também como ocorre o efeito inibitório desta droga. Foi observada nas

estruturas dos mutantes em complexo com NAD-INH que, as mutações I21V e

S94A parecem não influenciar na interação entre a proteína e o aducto. E assim,

estas informações corroboram com os dados existentes na literatura de que a

resistência pode estar relacionada com uma menor afinidade pela molécula de

NADH. Entretanto a resistência a isoniazida poderia estar associada também a

outros fatores nas quais estas mutações poderiam de forma indireta contribuir. É

apresentado também neste trabalho a estrutura do mutante de InhA S94A na sua

forma nativa, em um novo grupo espacial ainda não descrito na literatura para esta

enzima (P1). Nesta estrutura puderam-se concluir quais são os possíveis

movimentos que podem ocorrer durante o processo de ligação do NADH e do

substrato acil C16. Além disso, foi possível evidenciar importantes resíduos que

podem contribuir para a estabilidade da ligação entre o substrato e a enzima e

importantes movimentos que ocorrem em alguns resíduos do sítio ativo durante o

processo de catálise. Esperamos que estas informações contribuam para o

entendimento do mecanismo de inibição e de resistência à isoniazida e que possam

ser úteis na construção de novas drogas contra tuberculose.

120

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

5. BIBLIOGRAFIA

AHN, H. J. et al. Crystal structure of chorismate synthase: a novel FMN-binding protein

fold and functional insights. J. Mol. Biol. v. 336(4), p 903-915, 2004.

BAPTISTA, I. M. et al. Drug resistance and genotypes of strains of Mycobacterium

tuberculosis isolated from human immunodeficiency virus infected and non-infected

tuberculosis patients in Bauru, São Paulo, Brazil. Mem. Inst. Oswaldo Cruz, v. 97(8), p.

1147-1152, 2002.

BASSO, L. A. et al. Mechanisms of isoniazid resistance in Mycobacterium tuberculosis:

enzymatic characterization of enoyl reductase mutants identified in isoniazid-resistant

clinical isolates. J. Infect. Dis., v. 178, p. 769–775, 1998.

BAUBY, H.; GIRONS, I. S.; PICARDEAU, M. Construction and complementation of the

first auxotrophic mutant in the spirochaete Leptospira meyeri. Microbiology, v. 149, p.

689–693, 2003.

BANNER, D. W. et al. Structure of chicken muscle triose phosphate isomerase determined

crystallographically at 2.5 Å resolution using amino acid sequence data. Nature, v. 255, p.

609-614, 1975.

BENTLEY, R. The shikimate pathway - metabolic tree with many branches.Crit. Rev.

Biochem. Mol. Biol., v. 25, p. 307-384, 1990.

BRENNAN, P. J.; NIKAIDO, P. J. The envelope of mycobacteria. Annu. Rev. Biochem.,

v. 64, p. 29-63, 1995.

CERASOLI, E. et al. Effects of salts on the function and conformational stability of

shikimate kinase. Biochem. Biophys. Acta, v. 1648(1-2), p. 43-54, 2003.

COLE, S. T. et al. Deciphering the biology of Mycobacterium tuberculosis from the

complete genome sequence. Nature, v. 393, p.537-544, 1998.

COTRAN, R. S.; KUMAR, V.; ROBBINS, S. L. Robbins: Patologia estrutural e

funcional. Tradução de Amaury José da Cruz Júnior et al. 4.ed. Rio de Janeiro, Guanabara

Koogan, 1991.

DESSEN, A. et al. Crystal structure and function of the isoniazid target of Mycobacterium

tuberculosis. Science, v. 267, p. 1638–1641, 1995.

DIAS, M. V. B. et al. Crystal structure of chorismate synthase from Mycobacterium

tuberculosis. Journal of Structural Biology, v. 154, p. 130-143, 2006.

DHALIWAL, B. et al. Crystallographic studies of shikimate binding and induced

conformational changes in Mycobacterium tuberculosis shikimate kinase. FEBS Lett. V.

574(1-3), p. 49-54, 2004.

121

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

DOSSELAERE, F.; VANDERLEYDEN, J.A. A metabolic node in action: chorismate-

utilizing enzymes in microorganisms. Crit. Rev. Microbiol., v.27, p.75-131, 2001.

DOVER, L. G. et al. Comparative cell wall core biosynthesis in the mycolated pathogens,

Mycobacterium tuberculosis and Corynebacterium diphtheriae. FEMS Microbiology

Reviews, v. 28(2), p. 225-250, 2004.

DUNCAN, K. Progress in TB drug development and what is still needed. Tuberculosis, v.

83, p. 201-207, 2003.

DYE, C. et al. Erasing the world's slow stain: strategies to beat multidrug-resistant

tuberculosis Science, v. 295, p. 2042-2046, 2002.

FINN, J. et al. Rational herbicide design by inhibition of tryptophan biosynthesis. Bioorg.

Med. Chem. Letters, v. 9, p. 2297-2302, 1999.

GERSTEIN, M.; SCHULZ, G.; CHOTHIA C. Domain closure in adenylate kinase. Joints

on either side of two helices close like neighboring fingers. J. Mol. Biol. v. 229(2), p. 494-

501, 1993.

GAN, J. et al. Crystal structure of Mycobacterium tuberculosis shikimate kinase in complex

with skikimic acid and an ATP analogue. Biochemistry, v. 45, p. 8539-8545, 2006.

GU, Y. et al. Crystal structure of shikimate kinase from Mycobacterium tuberculosis reveals

the dynamic role of the LID domain in catalysis. J. Mol. Biol. v. 319, p. 779-789, 2002.

HARTMANN, M. D. et al. Mechanism of phosphoryl transfer catalyzed by shikimate

kinase from Mycobacterium tuberculosis. J. Mol. Biol., 2006 (in press).

HAWSER, S.; LOCIURO, S.; ISLAM, K. Dihydrofolate reductase inhibitors as

antibacterial agents. Biochemical Pharmacology, v. 71(7), p. 941-948, 2006.

HERRMANN, K. M. The shikimate pathway: early steps in the biosynthesis of aromatic

compounds. The Pant Cell, v. 7, p. 907-919, 1995.

HERRMANN, K. M.; WEAVER, L. M. The shikimate pathway. Annu Rev Plant

Physiol Plant Mol Biol. v. 50, p. 473-503, 1999.

HYDE, C. C. et al., Three-dimensional structure of the tryptophan synthase α

multienzyme com plex from Salmonella typhimurium. J. Bio. Chem. v. 263(33), p. 17857-

17871, 1988.

KHASNOBIS, S.; ESCUYER, V.; CHATTERJEE, D. Emerging therapeutic targets in

tuberculosis: post-genomic era. Expert. Opin. Ther. Targets, v. 6, p. 21-40, 2002.

KING, D. A. et al. Epidemiology. Infectious diseases: preparing for the future. Science, v.

313 (5792), p. 1392-1393, 2006.

KITZING, K. et al., Mechanism of chorismate synthase. Role of the two invariant histidine

residues in the active site. J. Biol. Chem. v. 279(10), p. 9451-9461, 2004.

122

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

KITZING, K.; MACHEROUX, P.; AMRHEIN, N. Spectroscopic and kinetic

characterization of the bifunctional chorismate synthase from Neurospora crassa: evidence

for a common binding site for 5-enolpyruvylshikimate 3-phosphate and NADPH. J. Biol.

Chem.,v. 276, p. 42658-42666, 2001.

KUO, M. R. et al. Targeting tuberculosis and malaria through inhibition of enoyl

reductase: Compound activity and structural data. The journal of Biological Chemistry,

v. 278(23), p. 20851-20859, 2003.

KRELL, T.; COGGINS, J. R.; LAPTHORN, A. J. The three-dimensional structure of

shikimate kinase. J. Mol. Biol. v. 278(5), p. 983-997, 1998.

KUMAR, V.; ABBAS, A. K.; FASTO, N. Robbins e Cotran: patologia – bases

patológicas das doenças. Tradução de Maria da Conceição Zacharias et al. Rio de Janeiro:

Elsevier, 2005. 1592p.

KURADI, R.; BILLETER, M.; WÜTHRICH, K. MOLMOL: a program for display and

analysis of macromolecular structures. J. Mol. Graphics v. 14, p. 51-55, 1996.

MACLEAN J.; ALI S. The structure of chorismate synthase reveals a novel flavin binding

site fundamental to a unique chemical reaction. Structure v. 11(12), p. 1499-511, 2003.

MACHEROUX, P. et al. A unique reaction in a common pathway: mechanism and

function of chorismate synthase in the shikimate pathway. Planta, v. 207(3), 325-334,

1999.

McARTHUR J. D. et al. An aromatic amino acid auxotrophic mutant of Bordetella

bronchiseptica is attenuated and immunogenic in a mouse model of infection. FEMS

Microbiology Letters, v. 221, p. 7-16, 2003.

McREE, D. E. XtalView/Xfit--A versatile program for manipulating atomic coordinates

and electron density. J. Struct. Biol., v. 125, p. 156-165, 1999.

MILES, E. W.; RHEE, S.; DAVIES, D. V. The molecular basis of substrate channeling.

The journal of biological chemistry, v. 274 (18), p. 12193-12196, 1999.

MILES, E. W. Tryptophan synthase. Structure, function, and protein engineering. Subcell

Biochem. v. 24, p. 207-254, 1995.

MOLLE, V. et al. The condensing activities of the Mycobacterium tuberculosis type II

fatty acid synthase are differentially regulated by phosphorylation. J. Bio. Chem., v.

281(40), p. 30094-30103, 2006.

MORGAN, J.; HARITAKUL, R.; KELLER, P. A. Anilinopyrimidines as novel

antituberculosis agents. Bioorg. Med. Chem. Lett., v. 13, p. 1755-1757, 2003.

MÜLLER, C. W. et al. Adenylate kinase motions during catalysis: an energetic

counterweight balancing substrate binding. Structure v. 4, p. 147-156, 1996.

MURSHUDOV, G. N.; VAGIN A. A.; DODSON, E. J. Refinement of Macromolecular

Structures by the Maximum-Likelihood Method Acta Cryst. D53, p. 240-255, 1997.

123

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

OLIVEIRA, J. S. et al. Crystallographic and pre-steady-state kinetics studies on binding of

NADH to wild-type and isoniazid-resistant enoyl-ACP(CoA) reductase enzymes from

Mycobacterium tuberculosis. J. Mol. Biol., v. 359(3), p. 646-66 2006.

PAN, P.; WOEHL, E.; DUNN, M. F. Protein architecture, dinamycs and allostery in

tryptophan synthase channeling. TIBS, v. 22, p. 22-27, 1997.

PARISH, T.; STOKER, N. G. The common aromatic amino acid biosynthesis pathway is

essential in Mycobacterium tuberculosis. Microbiology, v. 148, p. 3069-3077, 2002.

PEREIRA, J. H. et al. Structure of shikimate kinase from Mycobacterium tuberculosis

reveals the binding of shikimic acid. Acta Crystallogr. D. Biol., v. 60 (pt 12 pt 2), p.

2310-2319, 2004.

PELCZAR, M. J. Jr.; CHAN, E. C. S; KRIEG, N. R. Microbiologia – conceitos e

aplicações. Tradução: Sueli F. Yamada et al. 2.ed. São Paulo: Makron Books, 1996. 2 v.

PITTARD, A. J. Biosynthesis of the aromatic amino acids. In Escherichia coli and

Salmonella typhimurium: cellular and molecular biology. Neidhardt, F. C., Washington,

DC: American Society for microbiology, pp 368-394,1987

QUÉMARD, A. et al. Enzymatic characterization of the target for isoniazid in

Mycobacterium tuberculosis. Biochemistry, v. 34, p. 8235–8241, 1995.

QUEVILLON-CHERUEL S. et al. Crystal structure of the bifunctional chorismate

synthase from Saccharomyces cerevisiae. J. Biol. Chem. v. 279(1), p. 619-625, 2004.

RAVIGLIONE, M. C. et al., Assessment of worldwide tuberculosis control. WHO global

surveillance and monitoring project. The lancet, v. 350, p. 624-629, 1997.

RAWAT, R.; WHITTYty, A.; TONGE, P. The isoniazid–NAD adduct is a slow, tight-

binding inhibitor of InhA, the Mycobacterium tuberculosis enoyl reductase: adduct affinity

and drug resistance. Proc. Natl Acad. Sci. USA, v. 100, p. 13881–13886, 2003.

ROBERTS, C. W. et al. The shikimate pathway and its branches in apicomplexan

parasites. J. Infec. Dis., v. 185, Suppl 1, p. S25-36, 2002.

ROBERTS, F. et al. Evidence for the shikimate pathway in apicomplexan parasites.

Nature, v. 393(6687), p. 801-805, 1998

ROMANOWSKI, M. J.; BURLEY, S. K. Crystal structure of the Escherichia coli

shikimate kinase I (AroK) that confers sensitivity to mecillinam. Proteins, v. 47(4), p. 558-

62, 2002.

ROSWARSKI, D. A. et al.

Crystal structure of the Mycobacterium tuberculosis enoyl-

ACP reductase, InhA, in complex with NAD

and a C16 fatty acyl substrate. The Journal

of Biological Chemistry, v. 274(22), p. 15582-15589, 1999.

124

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

ROSWARSKI, D. A. et al. Modification of the NADH of the isoniazid target (InhA) from

Mycobacterium tuberculosis. Science, v. 279, p. 98–102, 1998.

SACHPATZIDIS, A. et al. Crystallographic studies of phosphonate-based α-reaction

transition-state analogues complexed to tryptophan synthase. Biochemistry, v. 38, p.

12665-12674, 1999.

SALI, A.; BLUNDELL, T. L. Comparative protein modelling by satisfaction of spatial

restraints. J. Mol. Biol., v. 234, p. 779–815, 1993.

SCHRUEDER, E. K. et al. Drugs that inhibit mycolic acid biosynthesis in Mycobacterium

tuberculosis. Curr. Pharm. Biotechnol., v. 3, p. 197–225, 2002.

SCHULZ, G. E. Binding of nucleotides by proteins. Curr. Opin. Struct. Biol. v. 2, p. 61-

67, 1992.

SKÖLD, O. Resistance to trimethoprim and sulfonamides. Veterinary Research, v. 32, p.

261-273, 2001.

SMITH, D. A. et al. Characterization of auxotrophic mutants of Mycobacterium

tuberculosis and their potential as vaccine candidates. Infection and Immunity, v. 69 (2),

p. 1142-1150, 2001.

TACKET, C. O.; SZTEIN, M. B.; LOSONSKY, G. A.; WASSERMAN, S. S.; NATARO,

J. P.; EDELMAN, R.; PICKARD, D.; DOUGAN, G.; CHATFIELD, S. N.; LEVINE, M.

M. Safety of live oral Salmonella typhi vaccine strains with deletions in htrA and aroC

aroD and immune response in humans. Infect. Immun., v. 65, p. 452-456, 1997.

TROUILLER, P. et al., Drug development for neglected diseases: a deficient market and a

public-health policy failure. Lancet, v. 359(9324), p. 2188-2194, 2002.

TROUILLER, P. et al., Drugs for neglected diseases: a failure of the market and a public

health failure? Tropical Medicine And International Health, v. 6, p.945-951, 2001.

VIOLA, C. M.; SARIDAKIS, V.; CHRISTENDAT, D. Crystal structure of chorismate

synthase from Aquifex aeolicus reveals a novel beta alpha beta sandwich topology.

Proteins, v. 54(1), p. 166-169, 2004.

YAN, H.; YSAI, M. D. Nucleoside monophosphate kinases: structure, mechanism, and

substrat specificity. Advan. Enzymol. Relat. Areas Mol. Biol. v. 73, p. 103-134, 1999.

WANG, R et al. SCORE: a new empirical method for estimating the binding affinity of a

protein-ligand complex. J. Mol. Model. v. 4, p. 379-394, 1998.

WEYAND M.; SCHLICHTING I. Crystal structure of wild-type tryptophan synthase

complexed with the natural substrate indole-3-glycerol phosphate. Biochemistry, v.

38(50), p.16469-16480, 1999.

125

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

4. Anexo

Artigo de Revisão publicado

126

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Chorismate Synthase: an attractive target for drug development against neglected

disease.

Marcio V. B. Dias; Fernanda Ely; Mário S. Palma; Luiz A. Basso; Diógenes S. Santos;

Walter Filgueira de Azevedo Jr. Current Drugs Targets, v. 8, p. 48-55, 2007.

Neste artigo é apresentada uma revisão a respeito das informações sobre a

corismato sintase publicadas em revistas indexadas nos últimos anos.

Nele é descrito a importância desta enzima e da via onde ela se encontra como

alvo para o desenvolvimento de novas drogas. É comentado desde a descoberta do

produto desta enzima, o corismato em 1962 até as informações publicadas em 2004. É

realizado um estudo das características desta enzima em diferentes organismos que a

possui, como fungos, bactérias, plantas e parasitas do filo apicomplexa. Portanto, é

descrita as peculiaridades das corismato sintases monofuncionais encontradas em

bactérias, plantas e parasitas do filo apicomplexa e das bifuncionais encontradas em

fungos. São discutidos também, os dados existentes sobre o possível processo evolutivo

da existência de enzimas monofuncionais e bifuncionais. São mostradas as

peculiaridades das corismato sintases de organismos eurcariotos (Plantas e parasitas de

filo apicomplexa) e sua localização nestas células. É discutido detalhadamente o

mecanismo catalítico desta enzima, com a importância da molécula de FMN para a

transferência de elétrons para a molécula de EPSP e a importância da molécula de

NADPH para as corismato sintases bifuncionais. É realizada uma descrição da estrutura

desta enzima, os aspectos peculiares do seu sítio ativo e a disposição das moléculas de

FMN e EPSP, e como isso pode auxiliar no processo de catálise desta enzima. E

finalmente, são apresentados os estudos já realizados no desenvolvimento de novos

ligantes e as estratégias utilizadas.

127

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Current Drug Targets, 2007, 8, 000-000 1

Chorismate Synthase: An Attractive Target For Drug Development Against Orphan

Diseases

Marcio V. B. Dias

, Fernanda Ely

, Mário S. Palma

, Walter F. de Azevedo Jr.

, Luiz A. Basso

and

Diógenes S. Santos

b**

Programa de Pós-Graduação em Biofísica Molecular – Departamento de Física, UNESP, São José do Rio Preto, SP 15054-000,

Brasil;

Centro de Pesquisas em Biologia Molecular e Funcional / PUCRS. Avenida Ipiranga 6681, Tecnopuc, Partenon 90619-900,

Porto Alegre, RS, Brazil;

Laboratório de Biologia Estrutural e Zooquímica-CEIS / Departamento de Biologia – Instituto de Biociên-

cias, UNESP, Rio Claro, SP 13506-900, Brasil and

Faculdade de Biociências-/ PUCRS. Avenida Ipiranga 6681, Partenon 90619-

900, Porto Alegre, RS, Brazil

Abstract: The increase in incidence of infectious diseases worldwide, particularly in developing countries, is worrying. Each year, 14

million people are killed by infectious diseases, mainly HIV/AIDS, respiratory infections, malaria and tuberculosis. Despite the great

burden in the poor countries, drug discovery to treat tropical diseases has come to a standstill. There is no interest by the pharmaceutical

industry in drug development against the major diseases of the poor countries, since the financial return cannot be guaranteed. This has

created an urgent need for new therapeutics to neglected diseases. A possible approach has been the exploitation of the inhibition of

unique targets, vital to the pathogen such as the shikimate pathway enzymes, which are present in bacteria, fungi and apicomplexan para-

sites but are absent in mammals. The chorismate synthase (CS) catalyses the seventh step in this pathway, the conversion of 5-

enolpyruvylshikimate-3-phosphate to chorismate. The strict requirement for a reduced flavin mononucleotide and the anti 1,4 elimination

are both unusual aspects which make CS reaction unique among flavin-dependent enzymes, representing an important target for the che-

motherapeutic agents development. In this review we present the main biochemical features of CS from bacterial and fungal sources and

their difference from the apicomplexan CS. The CS mechanisms proposed are discussed and compared with structural data. The CS struc-

tures of some organisms are compared and their distinct features analyzed. Some known CS inhibitors are presented and the main charac-

teristics are discussed. The structural and kinetics data reviewed here can be useful for the design of inhibitors.

Key Words: Infectious disease; neglected disease; chorismate synthase; shikimate pathway; flavin-dependent enzymes.

INTRODUCTION

A worrying increase in incidence of infectious diseases was recog-

nized in the late 1980’s, causing the suffering of millions of people,

especially in tropical and subtropical areas as Africa, Asia and

South America – which account for four-fifth of the world’s popu-

lation. Each year, 14 million people are killed by infectious dis-

eases, mainly diseases as HIV/AIDS, respiratory infections, malaria

and tuberculosis [1]. The increase of immunodeficient population,

mainly HIV-positive, the large homeless population and decline in

health care structures are some problems responsible for the in-

creased incidence of many infectious diseases in the world. Among

the neglected infectious diseases are tuberculosis and malaria. The

Mycobacterium tuberculosis, the aetiological agent of tuberculosis

(TB), kills more than 3 million people each year and ninety percent

of TB cases occur in developing countries

[2]. In addition, malaria,

caused by apicomplexan parasites of the genus Plasmodium, infects

millions of people and account for the death of more than 2 million

children annually [3]. In contrast, despite the enormous burden in

the poor countries, drug discovery and development targeted at

tropical diseases are at a standstill [1, 4]. Today, the pharmaceutical

industry is reluctant to invest in drug development to treat the major

diseases of the poor countries, because the financial return cannot

be guaranteed. Thus, an urgent reorientation of priorities in drug

development and health policy is needed, principally by national

and international policies that need to direct the global economy

and address the true health needs of society [1].

With the completion of the genome sequences of several patho-

genic organisms is occurring an enormous impact on our under-

*Address correspondence to this author at the Programa de Pós-Graduação

em Biofísica Molecular – Departamento de Física, UNESP, São José do Rio

Preto, SP 15054-000, Brasil; Tel: ---------------------; Fax: -------------------;

E-mail: diogenes@pucrs.br

standing of the pathogenicity of these organisms. Thus, the infor-

mation obtained by genome projects can be used by national and

international research policy makers and scientific community for

development of drugs against many of these diseases. The genome

allows identification of metabolic pathways present in the patho-

genic microorganisms, which can be target for the development of

new drugs. A possible approach to selective antimicrobial chemo-

therapy has been to exploit the inhibition of unique targets, vital to

the pathogen and absent in mammals [5]. The shikimate pathway is

an attractive example of this kind of targets, since it is present in

bacteria, fungi and apicomplexan parasites but absent from mam-

mals [6]. In this pathway the glycolytic intermediate, phosphoenol

pyruvate, and the pentose phosphate pathway intermediate, D-

erytrose 4-phosphate, are converted to chorismate through seven

metabolic steps [7]. The essentiality of shikimate pathway was ob-

served in some microorganisms such as Plasmodium falciparum

and M. tuberculosis. The disruption of aroK gene, which codes for

the shikimate kinase, showed that this enzyme is essential for M.

tuberculosis viability [8]. In P. falciparum the growth was inhibited

by glyphosate, a well-characterized inhibitor of the shikimate path-

way [9]. These reports provide strong evidence that shikimate

pathway is essential for the survival of these pathogens; therefore,

its enzymes are potential targets for drug development.

The product of shikimate pathway, the chorismate or chorismic

acid, is a dihydroaromatic compound and it was first described by

Frank and Margaret Gibson in 1962 [10]. This compound is the

branch point in the biosynthesis of several important aromatic

molecules. For this reason, it was named chorismate, which means,

in Greek, separation, split, or divorce. The chorismate is the com-

mon precursor for the biosynthesis of a wide range of primary and

secondary metabolites including aromatic amino acids (phenyla-

lanine, tyrosine and tryptophan), folate, naphthoquinones, menaqui-

nones and mycobactins [11]. Chorismate synthase (CS), which is

responsible for the synthesis of chorismate, was first described in

1967 by Morell et al. [12]. This enzyme catalyses the seventh step

128

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

2 Current Drug Targets, 2007, Vol. 8, No. 3 Dias et al.

in the shikimate pathway, the conversion of 5-enolpyruvylshiki-

mate-3-phosphate (EPSP) to chorismate [13].

Brucella suis, a gram-negative bacteria, which is responsible for

animal brucellosis in a variety of mammalian hosts, had the aroC-

encoded CS gene disrupted in an array for oral vaccine against bru-

cellosis. The B. suis mutant is highly attenuated in tissue culture

and murine virulence models [14]. Other pathogenic bacteria as

Salmonella enterica and Yersinia enterocolitica with aroC knock-

out [15-16] also grow slowly in vivo. The in vivo behavior of the

bacterial aroC mutants (low-level residual virulence) suggests the

importance of the CS for these microorganisms and shows that

mutants can be candidate vaccines and the protein can be targeted

for drug development.

The CS enzyme has been described in bacteria, fungi, plants

and apicomplexan parasites. However, the requirement of reduced

flavin cofactor is an intriguing factor, because the CS does not do

net redox change when the substrate EPSP is converted to choris-

mate. The reaction catalyzed by CS involves an 1,4-anti-

elimination of the 3-phosphate group and of the C-(6proR) hydro-

gen from EPSP, which is unusual and unique in nature [17-18]. The

determination of CS structure from Streptococcus pneumoniae

complexed with EPSP and FMN [19] could elucidate or confirm the

role of the flavin in the chemical reaction catalyzed by CS. Kinetics

studies could show that reduced flavin play a pivotal role in the

reaction. In the structure of S. pneumoniae, EPSP molecule is

stacked above the si-face of the isoalloxazine ring in an average

distance of 3.3 Å. This close juxtaposition of FMN and EPSP sup-

ports earlier suggestions for a direct role of reduced FMN in the

elimination reaction, such as a radical mechanism, in which elec-

tron transfer from the reduced cofactor to the substrate initiates C-O

bond breakage of the phosphate group to yield a substrate-derived

neutral radical [18, 20-24]. As there is not net redox change in the

reaction, a reverse transfer of the transiently donated electron is

required to complete the catalytic cycle [18]. This unusual reaction

catalyzed by CS also makes it unique among flavin-dependent en-

zymes.

CS FROM MICROORGANISMS (BACTERIA AND FUNGI)

CS was first described from three different microbial sources:

Escherichia coli [12], Bacillus subtilis [25] and Neurospora crassa

[26] (a gram-negative bacteria, a gram-positive bacteria and a

fungi). Initially, it was believed that these identified proteins were

three different enzymes because they were biochemically different.

However, although they have many differences in their biochemical

proprieties and molecular masses, these enzymes catalyze the same

reaction, the conversion of EPSP to chorismate. Moreover, it had

been observed that there was another feature common among the

three enzymes: the reduced flavin requirement. It was observed that

the E. coli CS is active only in anaerobic conditions in the presence

of either chemically or enzymatically reduced flavin [12], while the

enzyme of N. crassa and B. subtilis appeared to be associated with

a second enzymatic activity, the NAD(P)H dependent-flavin reduc-

tase [25, 27-29]. However, these two organisms use the NAD(P)H

differently. The N. crassa enzyme possesses an intrinsic flavin re-

ductase activity, located in the same polypeptide chain of CS. On

the other hand, the B. subtilis CS appears associated to two other

enzymes: the 3-dehydroquinate synthase, the second enzyme of

shikimate pathway, and a NAD(P)H dependent-flavin reductase,

forming a heterotrimeric complex. Thus, the CS from N. crassa,

owing to this additional capacity to reduce flavin is termed bifunc-

tional CS, while the other two CS, which do not have additional

capacity to reduce flavin, are termed monofunctional CS.

Initially, it was believed that the difference between molecular

weight of monofunctional E. coli

CS enzyme (M

39,138 Da) and

bifunctional N. crassa CS enzyme (M

46,400 Da) was due to a

domain responsible for reduction of flavin [30]. Based on sequence

comparisons with monofunctional CSs, two regions of 18 internal

amino acids residues and 29 C-terminal amino acids residues

unique to N. crassa were deleted. The presence of these two regions

in CS polypeptide chain was found not to be essential by comple-

mentation with an E. coli strain lacking CS [31]. The further char-

acterization of Saccharomyces cerevisiae CS has reported a smaller

(40,800 Da) and yet bifunctional enzyme, which complies with the

hypothesis that is not feasible to predict the mono or bifunctionality

based on the molecular weight [32]. At present, all fungi CSs

known are bifunctional, while all bacterial and plant CSs are mono-

functional. However, it is not known exactly what residues or re-

gion of active site are responsible for the bifunctionality of fungi

CS, despite the experimental efforts trying to determine these prop-

erties.

CSS EVOLUTION

One question that remains without answer is how was the

mono/bifunctional CSs evolution? A possible explanation is the

existence of a common ancestor with flavin reductase activity [32].

It is has been concluded from an earlier phylogenetic analysis that

chorismate synthases are monophyletic [33], it is not know to date

whether the ancestral chorismate synthase is mono – or bifunctional

[34]. However, it has been suggested that the common ancestor was

probably bifunctional given that it is difficult to image the evolution

of the intrinsic reductase activity in a framework of monofunctional

enzymes [33,34]. It was surmised that bifunctionality may have

either been maintained only in organisms in which the availability

of reduced flavin is limiting or perhaps there was positive selection

of monofunctionality. Thermotoga maritima is thought to be one

the oldest eubacterium and appears to have undergone considerable

lateral gene transfer from the archaea. A phylogeneic tree of all CSs

presently known suggests that T. maritima CS diverged with the

archaea and moreover, considerably before any of the CSs for

which bifunctionality is known (fungi CSs). Thus the classification

of its CS as monofunctional could be considered to be cognate to

the ancestral CS and would therefore not lend support to bifunc-

tionality being ascendant [34]. Furthermore, the plant and fungal

CSs emerged from a common ancestral protein after diverging from

the monofunctional bacteria proteins. In this case, if sufficient fla-

vin is available in the cell, there may have been no selective pres-

sure to maintain de flavin reductase activity in some species. By

now, all bifunctional CSs known are from fungal origin, showing

that among these organisms the bifunctionaly persisted under selec-

tive pressure, i.e. in organisms where the availability of reduced

flavin is limiting for growth [33]. The shikimate pathway in fungi

has an interesting feature: the presence of arom complexes - a pen-

tafunctional protein complex that catalyzes the second to sixth reac-

tion of the pathway [35]. Whether there is a causal relationship in

the presence of arom and the bifunctionality of fungal CSs remain

to be solved.

Nowadays several microbial CSs were isolated from different

sources such a M. tuberculosis [36], Helicobacter pylori [37], Sal-

monella typhimurium [38], S. cerevisiae [39] and Staphylococcus

aureus [40] and these could help to solve problem of CS evolution.

APICOMPLEXAN PARASITES

The herbicide glyphosate, a potent inhibitor of the shikimate

pathway enzyme EPSP synthase, has been shown to inhibit the in

vivo growth of the apicomplexan parasites Toxoplasma gondii, P.

falciparum and Cryptosporidium parvum. This effect has provided

evidence for the presence of shikimate pathway in apicomplexan

parasites. The CS was the first shikimate pathway enzyme identi-

fied in apicomplexan parasites by anaerobic assay [9].

The genome projects have identified the ORFs from

aroC-

encoded CS gene by sequence homology in apicomplexan parasites.

Many introns were found in aroC gene from T. gondii, but none

was identified in P. falciparum aroC gene. The lack of introns al-

lows the direct expression of P. falciparum CS in bacterial systems.

129

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Chorismate Synthase: An Attractive Target For Drug Development Current Drug Targets, 2007, Vol. 8, No. 3 3

Moreover, the T. gondii aroC is in a single copy, which allows gene

deletion experiments. Other interesting feature in the T. gondii aroC

gene is a potential GCN4 binding motif (TGACTC) in the 5' non-

coding region [3]. This specific upstream promoter element is pre-

sent in genes that are regulated by the control-activator protein

GCN4. In S. cerevisiae the expression of genes encoding shikimate

pathway enzymes DAHP synthase (3-deoxy-D-arabino-heptulo-

sonate 7-phosphate synthase) and the pentafunctional arom com-

plexes (which catalyzes the steps second to sixth in the pathway)

are regulated by GCN4 protein [41], which indicates a possible

regulation of shikimate pathway by CS expression in apicomplexan

parasites.

In plants, the shikimate pathway enzymes, like other nuclear-

encoded plastid enzymes, are post-translationally targeted to the

plastid by an amino-terminal leader sequence. The apicomplexan

parasites have a plastid-like organelle called apicoplast [42]. Al-

though the apicoplast proteins from apicomplexan parasites also

have leader sequences, there is no evidence that shikimate pathway

enzymes are localized in the apicoplast. Neither the T. gondii CS

gene nor P. falciparium CS gene have an obvious amino-terminal

leader sequence similar to that found on CSs from higher plants,

which strongly suggests that CS apicomplexan enzymes are in the

cytosol, and not in the apicoplast [43]. The shikimate pathway in

fungi also occurs in cytosol and phylogenetic analysis has demon-

strated that CSs from apicomplexan parasites are closer to fungal

CSs than plants CSs, which is consistent with the idea of a non-

apicoplast CS. Moreover, it also indicates a common ancestor to

fungal and apicomplexan enzymes [44].

Studies with CS from P. falciparium have characterized the en-

zyme biochemically and immunologically. Cofocal immunofluo-

rescence and cellular fractions followed by Western blot analysis

has shown that CS is located in the cytosol of this organism in dif-

ferent stages of infection, which is in agreement with the phyloge-

netic analysis. However, despite the close relationship with fungal

proteins, the enzymatic assay has demonstrated that P. falciparum

CS is monofunctional as in plants and bacteria [45]. Therefore, the

CS from P. falciparum has a combination of properties from plant

and fungal enzymes and it seems to possess properties distinct from

previously described CSs. It was found that both T. gondii and P.

falciparum enzymes differ from other known CS in possessing a

number of unique insertions [3]. Although the CSs in apicomplexan

parasites are larger in length, they share all of the amino acids to be

highly conserved in CSs from other species [9]. The implication of

a cytosolic location for the development of new drugs for apicom-

plexan parasites therapy is clear, fewer membranes will have to be

crossed.

The inhibition of P. falciparum growth with a dsRNA encoding

a 900 bp fragment of aroC has shown that CS is required for the

normal growth of this pathogen [46]. It suggests that CS may be a

viable target for chemotherapy. Moreover, two fluorinated analogs

of shikimate: (6R)-6-fluoroshikimate and (6S)-6-fluoroshikimate

have been shown to be inhibitors of the P. falciparum growth in

vitro [47]. The possible function of apicomplexan CS in shikimate

pathway regulation makes this enzyme a promising target for drug

development. Increasing our understanding of the apicomplexan

shikimate pathway enzymes may expedite the discovery and design

of new antiparasitic agents.

CS MECHANISM

The CS-catalyzed chemical reaction was first described in E.

coli extracts. It was reported that CS was inactivated under aerobic

conditions and it could be activated most effectively by a reduced

flavin cofactor [12]. Although many flavoproteins are involved in

redox reactions, there is not an overall redox when EPSP is con-

verted to chorismate and the reduced flavin mononucleotide is not

consumed during the reaction. Another interesting feature in CS

catalysis is an unusual 1,4 anti elimination of 3-phosphate group

and the C(6proR) hydrogen from EPSP, implied in a nonconcerted

reaction, which is interesting because it involved the cleavage of a

nonactivated C-H bond [17]. The strict requirement for a reduced

flavin mononucleotide and the anti 1,4 elimination are both unusual

aspects which make CS mechanism unique among flavin-dependent

enzymes [33]. Although there are two different classes of CS, dis-

tinguished by their ability to use NADPH for the reduction of FMN,

there is no apparent difference in the mechanism of elimination

reaction [18].

A nonconcerted mechanism was demonstrated by studies that

include a secondary tritium kinetic isotopic effect at C(3) [48], tran-

sient kinetic studies [20] and a secondary  deuterium kinetic iso-

topic effect at C(4) [49]. The kinetic isotopic experiments have

revealed that C(6)-H and C(3)-O bonds are cleaved in distinct steps,

and the phosphate group is eliminated before the C-H cleavage.

Two intermediates of nonconcerted mechanism have been proposed

to CS reaction: the cationic and the radical intermediate [50]. Al-

though reported isotopic effects could not clearly distinguish be-

tween the radical and cationic mechanisms, they have provided

evidence against alternative mechanisms. Moreover, the observa-

tion of spectral changes associated with the flavin during the cata-

lytic reaction has provided the strong evidence that the reduced

flavin has been actively involved in the reaction mechanism [51].

The role of flavin mononucleotide in catalysis derives from the lack

of activity of the bifunctional enzyme from N. crassa with 5-deaza-

FMN [52] and the formation of a stable flavin semiquinone radical

with the monofunctional E. coli enzyme and the substrate analog

(6R)-6-fluoro-EPSP, as a ternary complex [23, 52].

The characterization of the flavin intermediate has shown that it

is not simply associated with the conversion of substrate to product

since it must be formed before any chemical step involving EPSP.

The formation of a flavin intermediate is associated with the forma-

tion of the ternary complex between enzyme, EPSP and reduced

flavin, as described above. The flavin intermediate is generated

very rapidly within a few milliseconds and disappears when all

substrate has been consumed [20]. Spectral observations have

shown that the reduced flavin is bound to CS in its monoanionic

(deprotonated) form in the absence of substrate [22]. On the other

hand, studies with flavin analogs have demonstrated that CS binds

preferably the flavin in a neutral form, including the protonated

reduced form, rather than anionic forms in the presence of EPSP or

(6R)-6-F-EPSP [21]. The reported data led to the conclusion that

CS binds flavin in its monoanionic form and the protonation of the

N(1) position of FMN occurs upon binding of either EPSP or (6R)-

6-F-EPSP to the binary complex, which indicates that protonated,

reduced flavin is the observed reaction intermediate. The protona-

tion could lower the reduced flavin redox potential, converting the

reduced flavin into a better reductant, as required to promote the

radical chemistry [20]. Moreover, chorismate was produced from

(6R)-6-F-EPSP with the formation of a flavin semiquinone radical,

indicating the involvement of radical chemistry in this reaction. The

ternary complex is formed and a single electron is transferred from

the low potential flavin to the analogue, which promotes the elimi-

nation of a phosphate group to generate the allylic radical [23]. This

explanation is in agreement with the essential requirement of a fla-

vin co-factor.

Studies with substrate analogs have shown that even though the

(6S)-6-F-EPSP is a competitive inhibitor in N. crassa CS [54], it is

converted to 6-fluorochorismate in E. coli CS at a rate between 270

and 370 times slower than EPSP [50]. Studies with fluorinated sub-

strate analogues have shown that in E. coli CS the C(6)-H bond

breaking occurs after the formation of the flavin intermediate. Al-

though the C(6)-H breaking does not contribute significantly to rate

limitation in E. coli enzyme, it is partially rate-limiting with N.

crassa enzyme [52 54]. On the other hand, this result and deuterium

kinetic isotopic effect studies [55] have provided evidence that

steps after flavin intermediate decay, which may include product

130

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

4 Current Drug Targets, 2007, Vol. 8, No. 3 Dias et al.

release, are not significantly rate-limiting with EPSP or (6S)-6-F-

EPSP.

Spectroscopic and physical studies have shown that E. coli CS

undergoes a major structural change when the ternary complex is

formed. When the EPSP and FMN are bound, the enzyme has a

more compact shape, designated as closed form. On the other hand,

the apoenzyme probably exhibits more conformational flexibility,

designated as opened form [56]. The elucidation of S. pneumoniae

CS structure provides evidence that the binding of EPSP induces

changes in protein conformation, concurrent with a protonation of

FMN to give the neutral reduced form [19]. It is observed mainly

by the clear change in the position of His

110

when the opened and

closed structures are compared. It was proposed that the His

110

was

the probable residue that protonates the FMN and neutralizes the

reduced FMN in the first step of the reaction. The His

also ap-

pears to play a role in the reaction by providing a proton to neutral-

ize the charge on the departing phosphate group [19].

The fundamental role of these two histidine residues in CS ac-

tive site, as observed in S. pneumoniae, was confirmed by site-

directed mutagenesis of two histidine residues in the active site of

N. crassa CS. These histidines, His

106

and His

, were replaced by

alanine, reducing the activity 10- and 20-fold, respectively. The

phosphate release occurs before of the C-(6proR) hydrogen cleav-

age [20] and probably the two histidines residues are involved in

chemical steps leading directly to the initial C-O bond cleavage. It

was proposed that His

106

is involved in the protonation of monoan-

ionic reduced FMN, so it functions as the acid that donates the pro-

ton to N(1). This protonation may serve to direct the electron den-

sity so that charge transfer to the double bond of the substrate is

maximized. The function of His

is protonation of the leaving

phosphate group of EPSP. A detailed mechanism for CS reaction

was proposed based on structure and site-directed mutagenesis.

According to this model the EPSP binds to enzyme-bound reduced

FMN, forming a closed form. Then, an electron is transferred to the

EPSP double bond, initiating C-O bond cleavage and the release of

phosphate. In this step the His

acts as a general acid to neutralize

the incipient charge on the oxygen atom. The resulting C(4a)-

neutral flavin semiquinone tautomerizes to a radical species where

the unpaired electron residues on N(5) with concomitant abstraction

of the (proR-H) from C(6) by the N(5) [18]. Following the cleavage

of phosphate and possibly partial re-opening of the active site, the

intermediate remains bound while the proton is abstracted. The

atom N(5) of FMN is positioned directly below the C(6-proR) hy-

drogen of EPSP and is ideally positioned to remove this proton

from the transient intermediate. In the last step, the deprotonation of

the reduced flavin restores the initial state of the cofactor. The re-

generation of reduced FMN occurs with the reopening of the active

site, involving deprotonation by His

110

in the S. pneumoniae [19]. It

is also proposed that neither of the two histidines residues is essen-

tial for binding of the oxidized cofactor, representing only a role in

the stabilization of the change of the cofactor [18].

The future contributions as studies with substrate analogues,

high-resolution structures information and site-directed mutagenesis

in other invariant amino acids residues in the active site will pro-

vide further insights into this unprecedented flavin-dependent reac-

tion mechanism.

THE ROLE OF NADPH IN THE BIFUNCTIONAL CHO-

RISMATE SYNTHASE

CSs from fungal sources, i.e. S. cerevisiae and N. crassa, have

and additional intrinsic catalytic activity, which utilizes NAD(P)H

to reduce the flavin cofactor. (13, 56). The NAD(P)H is utilized in

the bifunctional CS to generate the essential reduced FMN cofactor.

NAD(P)H is consumed for the reduction of FMN followed by re-

oxidation of the reduced FMN by molecular oxygen. The rate of

flavin reduction probably is independent of the concentration and

presence of EPSP. The most intriguing question about the bifunc-

tional CS is the nature and the location of the NAD(P)H binding

site. All studies carried out to identify the sequence responsible for

binding of NAD(P)H have failed (31), therefore, the ability of bi-

functional CS to bind NAD(P)H is embedded in the protein struc-

ture revealing an unusual mode of interaction between these CS and

NAD(P)H [13]. Furthermore, Kitzing et al. (2001) [13] have ob-

served that EPSP and NAD(P)H compete for a common binding

site because the rate of NAD(P)H oxidation strongly depends on the

concentration of EPSP. It is conceivable because both NAD(P)H

and EPSP possess a phosphate group, which in turn plays an impor-

tant role in binding to the active site of CS. The site-directed

mutagenesis studies of important histidine residues of the active site

of N. crassa CS demonstrated that specific histidines to alanines

replacements affected the intrinsic NAD(P)H:FMN oxidoreductase

activity. It is mentioned that the lack of activity of the two histidine

to alanine mutant proteins under aerobic assay conditions does not

lead to conversion of EPSP to chorismate. Although both amino

acid exchanges affect the intrinsic oxidoreductase activity of CS,

they are apparently not decisive for secondary enzymatic activity,

as they do not abolish the flavin reductase activity [18]. Thus, the

amino acid residues that bring about bifunctionality remain obscure

[18]. Therefore, only the determination of the three-dimensional

structure of a bifunctional CS complexed with NAD(P)H will pro-

vide a description at molecular level of the essential residues re-

sponsible for bifunctionality.

THE STRUCTURE OF CHORISMATE SYNTHASE

Macheroux et al. (1998) [56] have predicted the CS structure in

a - barrel form as the result of secondary structure prediction

efforts. However, the determination of high-resolution structures by

crystallography has shown that they were not correct and the CS

structure is very different from what was expected. The small-angle

X-ray scattering data have indicated that the enzyme presents a

more compact overall shape in the ternary complex when both oxi-

dized FMN and EPSP are bound to the enzyme as compared to the

apoenzyme. Probably, the presented apoenzyme exhibits more con-

formational flexibility, even though these ligands have no effect on

the oligomerization state of the enzyme [5].

All crystal structures of CS solved to date show that the quater-

nary structure is tetrameric with approximately 222 symmetry (Fig.

(1a)). Moreover all structures have high structural similarity repre-

senting other distinct class of a conserved enzyme family, since it

lacks a primary sequence similarity to other classes of enzymes.

Despite the CSs being tretameric in its crystal form, in solution it

can be present as a dimer, tetramer or dimer/tetramer equilibrium

[12, 30, 39-40, 58].

The organisms whose CS crystal structures have been deter-

mined are: S. pneumoniae [19], Aquifex aeolicus [59], H. pylori

[60], S. cerevisaie [61], Campylobacter Jejuni [PDB access code:

1SQ1]. Furthermore preliminary data of X-ray crystallography for

CS from M. tuberculosis have also been reported [36].

The dominant structural topology

of CS monomer is character-

ized by a -- sandwich, in which each monomer consists of a

central helical core formed for four long -helices, sandwiched

between two four-stranded antiparallel beta sheets which is sur-

rounded by loops and discrete stretches of helix  and sheet .

These layers are packed to form a compact structure [19, 59-61]

(Fig. (1b)). The structure could also be described as two intimately

contacted sub-domains, consisting of anti-parallel sheet covered by

helices formed from -2 halves that are related by pseudo 2-fold

symmetry. However, the lack of identity between the sequences,

which form two domains, shows that could not have occurred for

gene duplication events [61-62].

131

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Chorismate Synthase: An Attractive Target For Drug Development Current Drug Targets, 2007, Vol. 8, No. 3 5

Although CS is a tetramer in the crystals, it is better character-

ized as a dimer of dimers in its basic structural unit because two

dimers compose the observed tetramer [19]. The dimer interface is

extensive and includes antiparallel interactions between the ends of

the sheet 2 from each subunit to give an extended eight-stranded 

sheet spanning both subunits (Fig. (1c)) [62].

ACTIVE SITE CHARACTERISTICS

The interesting aspect of active site is the large number of loop

regions that comprise the substrate binding pocket (fig. (1d)). Two

of them are provided by the adjacent subunit across the dimer inter-

face. In the ternary complex, FMN is almost completely buried,

with part of its accessible surface area being occluded by the sub-

strate, EPSP. Thus, the structure indicates an ordered binding of

FMN followed by EPSP, which is consistent with previous rapid

kinetic studies [56].

The FMN-binding site in CS is unique; no other flavoenzyme

presents such characteristic. In the structure of S. pneumoniae, the

FMN molecule makes few specific polar interactions with protein,

most part of them are contacts between hydroxyl and phosphate

oxygens of the ribityl chain. The flexible regions that surround the

FMN site are rich in strictly or highly conserved residues. The

FMN phosphate sits at the dimerization interface and it makes a

number of contacts with residues from the adjacent monomer. Fur-

thermore, there are a considerable number of solvent molecules

close to both the phosphate and ribityl regions of FMN. The water

molecules are discrete and ordered. They are responsible for medi-

ate interactions between FMN and the surrounding residues. The

isoalloxazine ring system of FMN is hydrophobic and makes few

specific interactions with the protein. However, it buries a consider-

able area of hydrophobic surface by packing the re-face against the

complementary surface of the protein [19]. This environment con-

tributes to the remarkable lowering of the flavin´s reduction poten-

tial to a value comparable to the most reducing flavodoxins [62]. A

notable feature of the bound FMN is a significant deviation from

planarity displayed by isoalloxazine ring. The pyrimidine ring and

the dimethylbenzene ring make an angle of approximately 10º and

the re-face of the isolloxazine ring is convex. It appears to assist the

substrate EPSP-binding site. This characteristic is observed in very

few other flavoenzymes [60, 63-65]. Furthermore the FMN binding

site has a number of positive charges in the proximity of the isol-

loxazine ring, which can increase the redox potential of FMN,

whereas a negative charge or a hydrophobic environment is ex-

pected to lower it. The positive charges at this location can possibly

stabilize negatively charged N1 in the reduced flavin, thus keeping

the electron until the substrate EPSP is bound [60].

The EPSP molecule is oriented relative to the N (5) position of

the isoalloxazine ring for electron transfer and hydrogen atom ab-

straction to and from the substrate, respectively. The EPSP mole-

cule is attacked above the si-face of the isoalloxazine ring in an

average distance of 3.3 Å as in the structure of CS from S. pneuno-

miae (Fig. (1d)). Owing to this fact, it can be observed that during

the reaction there is no accumulation of flavin radical intermediates

on the stopped-flow millisecond time scale. Hence, CS catalyzes a

Fig (1). Structure of CS; A)Tetramer structure; B) Monomer structure; C) Dimer; D) structure of active site.

132

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

6 Current Drug Targets, 2007, Vol. 8, No. 3 Dias et al.

unique reaction when occurs only the transfer of one electron,

which is unusual among flavoenzymes. The localization of the

FMN and EPSP in the CS has provided evidence that the flavin is

well placed to abstract a hydrogen atom from the 6-proR position of

the EPSP molecule, rather than just an electron with the proton

being accepted by an amino acid side chain. This form of flavin can

donate an electron and subsequently accept a hydrogen atom, stabi-

lizing again [19, 62].

In the structure of CS from S. pneumoniae, the EPSP molecule

makes a number of polar interactions with the protein, and few

hydrophobic contacts. The site of EPSP is hydrophilic and has an

environment extensively basic with many basic residues, mainly

arginines around of the binding site, which are extremely conserved

in many sequences, thereby showing its importance in the charge

stabilization of EPSP molecule [19].

Maclean and Ali (2003) [19] observed that the quaternary struc-

ture of S. pneumoniae has two conformational states. One of the

four monomers of the tetramer has differences when compared with

the other three. These changes occur principally in two loops of

active site comprised of highly conserved residues. The quaternary

structure shows one monomer with the active site more accessible

having therefore an open conformation, whereas the other three

have closed conformations. These differences provide vital infor-

mation about the mobility of these active site loops, which play an

important mechanistic role [19]. These loops have important basic

residues related with the catalytic mechanism of the protein. Differ-

ences between open and closed active sites demonstrate that the

side chain of one histidine located in one of these loops has some

conformational flexibility, and interacts with the FMN in the open

form but with O12 of EPSP in the closed form. Furthermore other

highly conserved residues in these loops also have different con-

formations in the two states showing that these residues play impor-

tant role in the catalytic mechanism [19]. Ahn et al. (2004) [60]

have observed also that the structure from H. pylori is surrounded

by flexible regions that are also highly conserved. Ahn et al. (2004)

[60] crystallized the apoenzyme and CS complexed with FMN. The

authors have observed that FMN could cause conformational

changes in these loops in CS from H. pylori, furthermore, they have

observed that the movement of these regions results in a more apo-

lar environment for the bound cofactor [60]. This structural infor-

mation confirms the hypothesis, in which the CS undergoes a major

change on binding oxidized flavin and EPSP and that the apoen-

zyme exhibits more conformational flexibility than ternary com-

plex. Furthermore it was observed that alterations occurred in the

visible CD spectrum of the enzyme-bound FMN on binding of

EPSP showing that the environment of the flavin changes consid-

erably [56].

STRATEGY FOR DEVELOPMENT DE DRUG BASED IN

THE CHORISMATE SYNTHASE

Based on the structural knowledge of the enzymes of the shiki-

mate pathway and their mechanisms of catalysis, it is possible to

design potential inhibitors in a rational way against pathogenic

bacteria, fungi and apicomplexan parasites. For amenable manner

and for knowledge about the shikimate, it was chosen for evaluating

of the design potential inhibitors. The strategy used was to incorpo-

rate fluorine at regio- and stereo-specific site of this molecule [66-

68]. With the knowledge of loss of C(6-proR) hydrogen in the reac-

tion catalyzed for CS [17, 66], two shikimate analogs were synthe-

sized, the (6R)-6-fluoro- and (6S)-6-fluoro shikimate acids. The

fluorine in (6R)-6-fluoro shikimate acid occupies the position of the

hydrogen abstracted during conversion of EPSP to chorismate. The

(6R)-6-fluro- and (6S)-6-fluro derivatives from shikimate acid are

metabolized by shikimate kinase and then to EPSP synthase to give

the corresponding diastereoisomeric 6-fluoro-EPSP. The (6R)-6-

EPSP might inhibit the CS, while the (6S)-6-EPSP would be proc-

essed further and might lead to lethal synthesis [66]. The (6R)-6-

fluoro-EPSP binds to E. coli CS in vitro and causes the oxidation of

enzyme-bound flavin, but it is not converted further, and neither

chorismate nor any other product could be detected [53]. On the

other hand, the (6S)-6-fluoro-EPSP is slowly converted to choris-

mate by CS from E. coli to a product identified as 6-fluoro-

chorismate [66] that inhibits the 4-aminobenzoic acid synthesis [55,

66]. Because of its greater potency in vitro, the shikimic acid ana-

logues were tested for antibacterial activity in vivo. The (6S)-6-

fluoro-shikimate was more protective against bacterial intraperito-

neal challenges in a mouse protection test. Mice were infected with

S. aureus and Salmonella dublin, but none of these strains would

grow when mice were treated with (6S)-6-fluro-shikimate on the

minimal chemically defined medium [66, 69].

The 6R- and 6S- isomers of 6-fluoro-shikimates have antibacte-

rial activity displaying minimum inhibitory concentrations of 64

and 0.5 μg mL

-1

against E. coli, respectively [66]. Owing to the lack

of (6proR) hydrogen in (6R)-6-fluoro-EPSP, it cannot be considered

a substrate of CS. Thus, this compound promotes one electron oxi-

dation of the E. coli enzyme [53] and it is a competitive inhibitor of

the N. crassa enzyme. The (6S)-6-fluoro-shikimate is also a com-

petitive inhibitor of the N. crassa enzyme, because there was no

conversion of this analog to 6-fluorochorismate by the enzyme of

this organism [54]. On the other hand, in CS from E. coli, this ana-

logue is not an inhibitor, since the conversion to 6-fluorochorismate

occurs. It can occur due the N. crassa enzyme to be different, once,

it is bifunctional [30, 57], differently of the E. coli enzyme. Despite

of conversion of (6S)-6-fluoro-EPSP for E. coli CS, it is converted

in a rate between 270 and 370 times slower than EPSP [50].

The decreased rate of reaction is consistent with the electron

withdrawing fluoro constituent destabilizing an allylic cationic

intermediate [70] that would be generated by the loss of phosphate

from the substrate [50]. There is a strong evidence for a stepwise,

rather than a concerted condensation reaction, involving an allylic

cationic intermediate with the fluoro substituent adjacent to the

allylic system. An allylic cationic intermediate in the CS reaction

would also be destabilized, in part, by negative hyperconjugative

effect of the adjacent fluoro substituent [71]. The negative hyper-

conjugative effect of the fluorine could destabilize an allylic radical

intermediate that would be formed by a mechanism involving an

additional one-electron reduction [50]. Once a ternary complex is

formed between the reduced holoenzyme and the (6R)-fluoro-EPSP

analogue, a single electron is transferred from the low potential

flavin to the analogue. This promotes the elimination of the phos-

phate group to generate an allylic radical. This intermediate is inca-

pable of transferring an electron and proton to the flavin from the

6R position due to its (6R)-fluoro group [23]. The (6S)-6-fluoro-

shikimate does not inhibit the CS, but the product formed by CS,

the 6-fluorochorismate inhibits the PabA and PabB. These enzymes

are responsible for conversion of chorismate to 4-aminobenzoate

precursor 4-amino-4-deoxychorismate [72].

The analogues of flavin also could be used to the design of in-

hibitors of CS, since it has been demonstrated that occurs lack of

activity of the enzyme with reduced 5-deaza-FMN [23].

Other class of inhibitors of CS was developed by Thomas et al.

(2003) [73]. Among these inhibitors are the benzofuran-3[2H]-one.

This inhibitor is moderately potent, with IC

of 8 μM. Kinetic

studies demonstrated that the benzofuran-3-[2H]-one is a competi-

tive inhibitor with respect to EPSP and noncompetitive with FMN,

suggesting that the compound binds in the active site of the en-

zyme. However, owing to the metabolic vulnerability and potential

toxicity of the 6,7-dihydroxy functional group, and also in an effort

to increase potency, further synthetic efforts have been undertaken

[73].

CONCLUSION

CS is an important enzyme of the shikimate pathway, having a

critical role in metabolism of microorganisms. Furthermore it is

133

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

Chorismate Synthase: An Attractive Target For Drug Development Current Drug Targets, 2007, Vol. 8, No. 3 7

fascinating, mainly for their unusual catalytic mechanism and for

their unique folding in nature. The knowledge about the CSs has

been evaluated day by day. The determination of various structures

of this protein would contribute to an understanding of the details of

the mechanism of catalysis and reveal the role played by important

active site residues, such as the His

and His

110

in S. pneumonie.

Future research could also contribute to the understanding of the

unprecedented flavin-dependence in the reaction mechanism, which

can be through of site-directed mutagenesis and the utilization of

substrate and cofactor analogues. The major challenge today, is to

precisely determine the role of NAD(P)H in bifunctional CSs and

the residues that are involved in the their binding. On the other

hand, today, major new opportunities in structure-based drug design

are possible due the determination of CS structure complexed with

both substrate and cofactor. This knowledge could contribute to the

development of new drugs including the use of combinatorial

chemistry and virtual screening methods that could render inhibi-

tors more specific to compete with substrate and cofactor in binding

to the enzyme. The development of new drugs or medicines against

infectious disease such as tuberculosis, respiratory disease and ma-

laria will help hundreds of millions of people worldwide [74-86],

mainly in developing countries, where poor people will most bene-

fit from these efforts.

ACKNOWLEDGMENTS

This work was supported by grants from FAPESP

(SMOLBNet, Proc. 01/07532-0, 03/12472-2, 04/00217-0) and Mil-

lenium Institute, MCT/CNPq to DSS and LAB. MSP, DSS, WFA

and LAB are Research Awardees from the National Research

Council.

ABBREVIATION LIST

CS = Chorismate synthase

TB = Tuberculosis

EPSP = 5-enolpyruvylshikimate-3-phosphate

FMN = Flavin mononucleotide

FMN

red

= Flavin mononucleotide reduced

FMN

oxi

= Flavin mononucleotide oxidized

FMN

Flavin mononucleotide semiquinone

DAHP = 3-deoxy-D-arabino-heptulosonate 7-

phosphate

6(R)-6-F-EPSP = (6R)-6-fluoro-5-nolpyruvylshikimate-3-

phosphate

6(S)-6-F_EPSP = (6S)-6-fluoro-5-nolpyruvylshikimate-3-

phosphate

His

= Histidine 10

His

110

= Histidine 110

PDB = Protein data bank

REFERENCES

[1] Trouiller, P.; Torreele, E.; Olliaro, P.; White, N.; Foster, S.; Wirth,

D. and Pécoul, B. (2001) Trop. Med. Internat. Health., 6(11), 945-

951.

[2] Corbett, E. L.; Watt, C. J.; Walker, N.; Maher, D.; Williams, B. G.;

Raviglione, M. C. and Dye, C. (2003) Arch. Intern. Med., 1639,

1009-1021.

[3] Roberts, C. W.; Roberts, F.; Lyons, R. E.; Kirisits, M. J.; Mui, E.

J.; Finnerty, J.; Johnson, J. J.; Ferguson, D. J. P.; Coggins, J. R.;

Krell, T.; Coombs, G. H.; Milhous, W. K.; Kyle, D. E.; Tzipori, S.;

Barnwell, J.; Dame, J. B.; Carlton, J. and McLeond, R (2002) J. In-

fect. Dis., 185, S25-36.

[4] Pécoul, B.; Chirac, P.; Trouiller, P. and Pinel, J. (1999) J. of the

American Med. Association, 281, 361-367.

[5] Coates, A.; Hu, Y.; Bax, R. and Page, C. (2002) Nat. Rev. Drug.

Discov., 1, 895 – 910.

[6] Bentley, R. (1990) Crit. Rev. Biochem. Mol. Biol., 1990, 307-384.

[7] Herrmann, K. M. (1995) Plant Cell, 7, 907-919.

[8] Parish, T. and Stoker, N. G. (2002) Microbiol., 148, 3029-3077.

[9] Roberts, F.; Roberts, C. W.; Johnson, J. J.; Kyle, D. E.; Krell, T.;

Coggins, J. R.; Coombs, G. H.; Milhous, W. K.; Tzipori, S.;

Ferguson, D. J. P.; Chakrabarti, D. and McLeond, R. (1998) Na-

ture, 393, 801-805.

[10] Gibson, M.I. and Gibson, F. (1962) Biochim. Biophis. Acta, 65,

160-163.

[11] Ratledge, C. (1982) Nutrition, growth and metabolism in: Ratledge,

C., Stanford, J. L. (Eds.), The biology of the Mycobacteria, vol. 1,

Academic Press, London, pp. 185-271.

[12] Morell, H.; Clark, M.J.; Knowles, P.F. and Sprinson, D.B. (1967) J.

Biol. Chem., 242, 82-90.

[13] Kitzing. K.; Macheroux, P. and Amrhein, N. (2001) J. Biol. Chem.,

276(46), 42658-42666.

[14] Foulongne, V.; Walravens, K.; Bourg, G.; Boschiroli, M.L.; God-

froid, J.; Ramuz, M. and O'Callaghan D. (2001) Infect. Immun.,

69(1), 547-550.

[15] Bowe, F.; O'Gaora, P.; Maskell, D.; Cafferkey, M. and Dougan, G.

(1989), Infect. Immun., 57, 3234-3236.

[16] Lowe, D.C.; Savidge, T.C.; Pickard, D.; Eckmann, L.; Kagnoff,

M.F.; Dougan, G. and Chatfield, S.T. (1999) Infect. Immun., 67,

700-707.

[17] Floss, H. G.; Onderka, D. K. and Carroll, M. (1972) J. Biol. Chem.,

247, 736-744.

[18] Kitzing, K.; Auweter, S.; Amrhein, N. and Macheroux, P. (2004) J.

Biol. Chem., 279, 9451-9461.

[19] Maclean, J. and Ali, S. (2003) Structure, 11(12), 1499-1511.

[20] Bornemann, S.; Lowe, D. J. and Thorneley, R. N. (1996) Biochem-

istry, 35(30), 9907-9916.

[21] Macheroux, P., Bornemann, S.; Ghisla, S. and Thorneley, R. N.

(1996) J. Biol. Chem., 271(42), 25850-25858.

[22] Macheroux, P.; Petersen, J.; Bornemann, S.; Lowe, D. J. and Thor-

neley, R. N. (1996) Biochemistry, 35(5), 1643-1652.

[23] Osborne, A.; Thorneley, R. N.; Abell, C. and Bornemann, S. (2000)

J. Biol. Chem., 275(46), 35825-35830

[24] Machoroux, P.; Bornemann. and Thorneley, R. N. F. (1997) in

Flavins and flavoproteins 1996 (Stevenson, K. J.; Massey, V. and

Williams, C. H. J., eds) 0University of Calgary Press, Calgari,

Canada, pp. 113-122.

[25] Hasan, N. and Nester, E. W. (1978) J. Biol. Chem., 253(14), 4987-

4982.

[26] Gaertner, F. H. and Cole, K. W. (1973) J. Biol. Chem., 248(13),

4602-4609.

[27] Gaertner, F. H. (1987) Methods Enzymol., 142, 362-366.

[28] Hasan, N. and Nester, E. W. (1978) J. Biol. Chem., 253(14), 4993-

4998.

[29] Hasan, N. and Nester, E. W. (1978) J. Biol. Chem., 253(14), 4999-

5004.

[30] White P. J.; Millar, G. and Coggins, J. R. (1988) Biochem. J.,

251(2), 313-322.

[31] Henstrand, J. M.; Amrhein, N. and Schmid, J. (1995) J. Biol.

Chem., 270(35), 20447-20452.

[32] Henstrand, J. M.; Schaller, A.; Braun, M.; Amrhein, N. and

Schmid, J. (1996) Mol. Microbiol., 22(5), 859-866.

[33] Macheroux, P.; Schmid, J.; Amrhein, N. and Schaller, A. (1999)

Planta, 207, 325-334.

[34] Fitzpatrick, T. B.; Killer, P.; Thomas, R.M.; Jelesarov, I.; Amrhein,

N. and Macheroux P. (2001) J. Biol. Chem., 276(21), 18052-18059.

[35] Hawkins, A. R. and Smith, M. (1991) Eur. J. Biochem., 196, 717-

724.

[36] Dias, M. V., Borges, J. C., Ely, F., Pereira, J. H., Canduri, F., Ra-

mos, C. H., Frazzon, J., Palma, M. S., Basso, L. A., Santos, D. S.,

de Azevedo, W. F. Jr. (2006) J. Struct. Biol. 154(2) ,130-143.

[37] Ahn, H. J.; Yang J. K.; Lee, B. I.; Yoon, H. J.; Kim, H. W. and

Suh, S. W. (2003) Acta Crystallogr. D Biol. Crystallog., 59, 569-

571.

[38] Charles, I. G.; Lamb, H. K.; Pickard, D.; Dougan, G. and Hawkins,

A. R. (1990) J. Gen. Microbiol., 136(2), 353-358.

[39] Jones, D. G.; Reusser, U. and Braus, G. H. (1991) Mol. Microbiol.,

5(9), 2143-2152.

[40] Horsburgh, M. J.; Foster, T. J.; Barth, P. T. and Coggins, J. R.

(1996) Microbiology, 142, 2943-2950.

[41] Arndt, K. and Fink, G. R. (1986) Proc. Natl. Acad. Sci. USA, 83,

8516-8620.

134

Estudo Estrutural de Proteínas Alvo de Mycobacterium tuberculosis

8 Current Drug Targets, 2007, Vol. 8, No. 3 Dias et al.

[42] Köhler, S.; Delwiche, C. F.; Denny, P. W.; Tilney, L. G.; Webster,

P. and Wilson, R. J. (1997) Science, 275, 1485-1489.

[43] Mcfadden, G. I.; Reith, M. E.; Munholland, J. and Lang-Unnasch,

N. (1996) Nature, 381, 482.

[44] Keeling, P. J.; Jeffrey, D. P.; Donald, R. G. K.; Roos, D. S.; Waller,

R. F. and McFadden, G. I. (1999) Nature, 397, 219-220.

[45] Fitzpatrick, T.; Ricken, S.; Lanzer, M.; Amrhein, N. Macheroux, P.

and Kappes, B. (2001) Mol. Biol., 40(1), 65-75.

[46] McRobert, L. and McConkey, G. A. (2002) Mol. Biochem. Parasi-

tol., 119, 273-278.

[47] McConkey, G. A. (1999) Antimicrob. Agents Chemotherar., 43,

175-177.

[48] Balasubramanian, S.; Coggins, J. R. and Abell, C. (1995) Biochem-

istry, 34(1), 341-348.

[49] Bornemann, S.; Theoclitou, M. E.; Brune, M.; Webb, M. R.; Thor-

neley, R. N. and Abell, C. (2000) Bioorg. Chem., 28, 191-204.

[50] Bornemann, S.; Ramjee, M. K.; Balasubramanian, S.; Abell, C.;

Coggins, J. R.; Lowe, D. J. and Thorneley, R. N. F. (1995) J. Biol.

Chem., 270, 22811-22815.

[51] Ramjee, M. K.; Coggins, J. R.; Hawkes, T. R.; Lowe, D. J. and

Thorneley, R. N. F. (1991) J. Am. Chem. Soc., 113, 8566-8567.

[52] Lauhon, C. T. and Bartlett, P. A. (1994) Biochemistry, 33, 14100-

14108.

[53] Ramjee M. N.; Balasubramanian, S.; Abell, C.; Coggins, J. R.;

Davies, G. M.; Hawkes, T. R.; Lowe, D. J. and Thorneley, R. N. F.

(1992) J. Am. Chem. Soc., 114, 3151-3153.

[54] Balasubramanian, S. and Abell, C. (1991) Tetrahedron Lett., 32,

963-966.

[55] Bornemann, S.; Balasubramanian, S.; Coggins, J. R.; Abell, C.;

Lowe, D. J. and Thorneley, R. N. (1995) Biochem. J., 305, 707-

710.

[56] Macheroux P.; Schonbrunn, E.; Svergun, D. I.; Volkov, V. V.;

Koch, M. H.; Bornemann, S. and Thorneley, R. N. (1998) Biochem.

J., 335, 319-327.

[57] Welch, G. R.; Cole, K. W.; and Gaertner, F. H. (1974) Arch. Bio-

chem. Biophys., 165, 505-518.

[58] Schaller, A.; Windhofer, V. and Amrhein, N. (1990) Pers. Arch.

Biochem. Biophys., 282, 437-442.

[59] Viola, C.M.; Saridakis, V. and Christendat, D. (2004) Proteins,

54(1), 166-169.

[60] Ahn, H.J.; Yoon, H.J.; Lee, B. 2nd and Suh, S.W. (2004) J. Mol.

Biol., 336(4), 903-915.

[61] Quevillon-Cheruel, S.; Leulliot, N.; Meyer, P.; Graille, M.; Bre-

mang, M.; Blondeau, K.; Sorel, I.; Poupon, ª; Janin, J. and van Til-

beurgh, H. (2004) J. Biol. Chem., 279(1), 619-625.

[62] Bornemann, S.; Lawson, D. M. and Thorneley, R. N. (2003) Struc-

ture, 11(12), 1463-1465.

[63] Binda, C.; Coda, A.; Angelini, R.; Federico, R.; Ascenzi, P. and

Mattevi. A. (1999) Structure, 7, 265-276.

[64] Yue, Q. K.; Kass, I. J.; Sampson, N.S. and Vrielink, A. (1999)

Biochemistry, 38, 4277-4286.

[65] Barber, M. J.; Neame, P. J.; Lim,L. W.; White, S. and Matthews, F.

S. (1992) J. Biol. Chem., 267, 6611–6619.

[66] Davies, G.M.; Barrett-Bee, K.J.; Jude, D.A.; Lehan, M.; Nichols,

W.W.; Pinder, P.E.; Thain, J.L.; Watkins, W.J. and Wilson, R.G.

(1994) Antimicrob. Agents. Chemother., 38(2), 403-406.

[67] Abeles, R.H. and Alston, T.A. (1990) J. Biol. Chem., 265(28),

16705–16708.

[68] Kollonitsch, J. (1978) Isr. J. Chem. 17, 53-59.

[69] Davis, B. D. and Mingioli, E. S. (1950) J. Bacteriol., 60, 17-28.

[70] Hawkes, T. R.; Lewis, T.; Coggins, J. R.; Mousdale, D. M.; Lowe,

D. J. and Thorneley, R. N. F. (1990) Biochem. J., 265, 899-902.

[71] von Ragué Schleyer.; P. and Kos, A. J. (1983) Tetrahedron, 39,

1141-1150.

[72] Green, J. M. and Nichols, B. P. (1991) J. Biol. Chem., 266, 12971-

12975.

[73] Thomas, M.G.; Lawson, C.; Allanson, N.M.; Leslie, B.W.; Bottom-

ley, J.R.; McBride, A. and Olusanya, O.A. (2003) Bioorg. Med.

Chem. Lett., 13(3), 423-426.

[74] Basso, L. A., Pereira da Silva, L. H., Fett-Neto, A. G., de Azevedo

Jr., W. F., Moreira, I. S., Palma, M. S., Calixto, J. B., Astolfi Filho,

S., dos Santos, R. R., Soares, M. B. P., Santos, D. S. (2005) Mem.

Inst. Oswaldo Cruz, 100(6): 475-506.

[75] da Silveira, N. J. F., Uchoa, H. B., Canduri, F.,Pereira, J. H., Cam-

era Jr., J. C., Basso, L. A., Palma, M. S., Santos, D. S., de Azevedo

Jr., W. F. (2004) Biochem Biophys. Res. Commun.,

322(1), 100–

104.

[76] Ducati, R. D., Basso, L. A., Santos, D. S. (2007) Current Drug

Targets, 8(1) In press.

[77] Dias, M. V. B., Faim, L. M., Vasconcelos, I. B., Oliveira, J. S.,

Basso, L. A., Santos, D. S., De Azevedo, W. F. Acta Crystal-

lographica F. (In press).

[78] De Azevedo, W.F.Jr.; Mueller-Dieckmann, H.J.; Schulze- Gahmen,

U.; Worland, P.J.; Sausville, E.; Kim S.-H. (1996) Proc. Natl.

Acad. Sci. USA., 93(7), 2735-2740.

[79] Canduri, F.; Uchoa, H.B.; de Azevedo, W.F.Jr. (2004) Biochem.

Biophys. Res. Commun., 324(2), 661-666.

[80] De Azevedo, W.F.Jr.; Canduri, F.; Silveira, N.J.F. (2002) Biochem

Biophys. Res. Commun., 293(1), 566-571.

[81] De Azevedo, W.F.Jr.; Leclerc, S.; Meijer, L.; Havlicek, L.; Strnad,

M.; Kim, S.-H. (1997) Eur. J. Biochem., 243, 518-526.

[82] De Azevedo, W. F. Jr., Canduri, F. (2005) Current Computer-

Aided Drug Design, 1, 53-64.

[83] Uchoa, H. B., Jorge, G. E., da Silveira, N. J., Camera J. C., Can-

duri, F., De Azevedo, WF. (2004) Biochem. Biophys. Res. Com-

mun. 325(4),1481–1486.

[84] De Azevedo, W.F.Jr., Gaspar, R.T., Canduri, F.; Camera, J.C.Jr.,

Silveira, N.J.F. (2002) Biochem Biophys. Res. Commun., 297(5),

1154-1158.

[85] da Silveira, N. J. F., Bonalumi, C. A., Uchoa, H. B., Pereira, J. H.,

Canduri, F.,Pereira, J. H., de Azevedo Jr., W. F. (2006) Cell Bio-

chem. Biophys. 44(3), 366-374.

[86] Kim, S.-H.; Schulze-Gahmen, U.; Brandsen, J.; De Azevedo, W. F.

Jr. (1996) Prog. in Cell Cycle Res., 2, 137-145.

Received: December 12, 2005 Accepted: May 26, 2006 Updated: September 11, 2006

135

Livros Grátis
( http://www.livrosgratis.com.br )
 
Milhares de Livros para Download:
 
Baixar livros de Administração
Baixar livros de Agronomia
Baixar livros de Arquitetura
Baixar livros de Artes
Baixar livros de Astronomia
Baixar livros de Biologia Geral
Baixar livros de Ciência da Computação
Baixar livros de Ciência da Informação
Baixar livros de Ciência Política
Baixar livros de Ciências da Saúde
Baixar livros de Comunicação
Baixar livros do Conselho Nacional de Educação - CNE
Baixar livros de Defesa civil
Baixar livros de Direito
Baixar livros de Direitos humanos
Baixar livros de Economia
Baixar livros de Economia Doméstica
Baixar livros de Educação
Baixar livros de Educação - Trânsito
Baixar livros de Educação Física
Baixar livros de Engenharia Aeroespacial
Baixar livros de Farmácia
Baixar livros de Filosofia
Baixar livros de Física
Baixar livros de Geociências
Baixar livros de Geografia
Baixar livros de História
Baixar livros de Línguas

Baixar livros de Literatura
Baixar livros de Literatura de Cordel
Baixar livros de Literatura Infantil
Baixar livros de Matemática
Baixar livros de Medicina
Baixar livros de Medicina Veterinária
Baixar livros de Meio Ambiente
Baixar livros de Meteorologia
Baixar Monografias e TCC
Baixar livros Multidisciplinar
Baixar livros de Música
Baixar livros de Psicologia
Baixar livros de Química
Baixar livros de Saúde Coletiva
Baixar livros de Serviço Social
Baixar livros de Sociologia
Baixar livros de Teologia
Baixar livros de Trabalho
Baixar livros de Turismo
 
 