Download PDF
ads:
Estudo estrutural de enzimas da via metabólica do ácido
chiquímico de Mycobacterium tuberculosis
José Henrique Pereira
Orientador: Prof. Dr. Walter Filgueira de Azevedo Júnior
Tese apresentada para obtenção do grau de Doutor
em Biofísica Molecular, área de concentração em
Biofísica Molecular do Programa de Pós-Graduação
do Departamento de Física do Instituto de
Biociências, Letras e Ciências Exatas da
Universidade Estadual Paulista “Julio de Mesquita
Filho” – UNESP.
São José do Rio Preto / Outubro de 2005
ads:
Livros Grátis
http://www.livrosgratis.com.br
Milhares de livros grátis para download.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
O presente trabalho foi desenvolvido entre fevereiro/2002 e outubro/2005, junto ao
Departamento de Física do Instituto de Biociências, Letras e Ciências Exatas da
Universidade Estadual Paulista “Julio de Mesquita Filho” – UNESP, sob a orientação do
Professor Dr. Walter Filgueira de Azevedo Júnior.
José Henrique Pereira
ads:
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Agradecimentos
- Ao Prof. Dr. Walter Filgueira de Azevedo Júnior pela orientação, confiança, incentivo
e oportunidade de trabalho.
- Aos professores Dr. Diógenes Santiago Santos e Dr. Luis Augusto Basso responsáveis
pelo Centro de Pesquisas em Biologia Molecular e Funcional (PUC-RS) que gentilmente
cederam as proteínas envolvidas neste trabalho.
- À Prof. Dra. Fernanda Canduri pela amizade e ensinamentos passados durante todo o
período de iniciação científica até presente momento.
- Ao Dr. Jaim Simões de Oliveira pela fundamental contribuição para o desenvolvimento
deste projeto.
- Aos colegas do Laboratório de Sistemas Biomoleculares (Departamento de Física -
IBILCE): Alessandra, Bona, Briana, Daiane, Danilo, Denise, Diego, Helen, Joane, José
Renato, Júlio, Lisandra, Lívia, Márcio, Marcos, Marisa, Michelli, Nathalia, Nelson,
Pepeu.
- Aos Prof. do Departamento de Física do IBILCE/UNESP.
- À minha Família pelo apoio incondicional.
- À FAPESP, CAPES e CNPq/MCT pela bolsa e apoio aos projetos.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Resumo
A tuberculose ressurgiu na metade dos anos 80 e atualmente, 2 milhões de pessoas
morrem por ano devido a esta doença. O ressurgimento da tuberculose tornou-se uma
ameaça à saúde pública. A alta susceptibilidade dos pacientes infectados com HIV e a
proliferação de cepas resistentes a múltiplas drogas têm criado a necessidade de
desenvolver novas terapias. No entanto, nenhuma nova classe de drogas contra a
tuberculose foi desenvolvida nos últimos 30 anos. Deste modo, existe a necessidade de
desenvolver rapidamente novos agentes anti-tuberculose. As enzimas da via metabólica
do ácido chiquímico são alvos em potencial para o desenvolvimento de agentes
antimicrobianos não-tóxicos e herbicidas, pois esta via é essencial para bactéria e
plantas, enquanto que está ausente em mamíferos.
A via do ácido chiquímico é composta por sete passos metabólicos, catalisados
pelas enzimas: DAHP sintase, 3-desidroquinato sintase, 3-desidroquinato desidratase,
chiquimato desidrogenase, chiquimato quinase, EPSP sintase, e corismato sintase. Este
trabalho apresenta o estudo através da cristalografia de raios X da chiquimato quinase,
assim como a modelagem molecular da EPSP sintase e da corismato sintase.
Os resultados obtidos neste trabalho a partir do estudo estrutural das enzimas da via
do ácido chiquímico presentes em Mycobacterium tuberculosis, usando modelagem
molecular e cristalografia, devem contribuir para o entendimento do mecanismo
catalítico destas enzimas e também para desenvolvimento de novos inibidores que
futuramente poderão ser usados como agentes anti-tuberculose.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Abstract
Tuberculosis resurged in the mid-1980s and now kills approximately 2 million people
a year. The reemergence of tuberculosis as a public health threat, the high susceptibility
of HIV-infected persons, and the proliferation of multi-drug-resistant strains have
created a need to develop new therapies. However, no new classes of drugs for TB have
been developed in the past 30 years. There is an urgent need to developing faster acting
and effective new anti-tubercular agents. The enzymes in the shikimate pathway are
potential target for development of non-toxic antimicrobial agents, and herbicides
because it is essential in bacteria, and plants, whereas it is absent from mammals.
The shikimate pathway is the seven-step biosynthetic route catalyzed by enzymes: 3-
deoxy-D-arabino-heptulosonate 7- phosphate synthase (DAHP synthase), 3-
dehydroquinate synthase, 3-dehydroquinate dehydratase, shikimate dehydrogenase,
shikimate kinase, 5-enolpyruvyl shikimate 3-phosphate synthase (EPSP synthase), and
chorismate synthase. This work presents the X-ray crystallographic studies of shikimate
kinase, as well as the molecular modeling of EPSP synthase and chorismate synthase.
The results obtain with this work through structural studies from enzymes in the
shikimate pathway will provide crucial information for elucidation of the mechanism of
catalyzed reaction and for the development of a new generation of drugs against
tuberculosis.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Lista de Figuras
Figura 1.1 – Microscopia eletrônica de varredura do Mycobacterium tuberculosis. ................………...01
Figura 1.2
– A via do ácido chiquímico. Fosfoenol-piruvato e eritrose 4-fosfato (precursores) são
convertidos em ácido corísmico (corismato). ..................…...............................................……………...06
Figura 1.3
– Estrutura molecular do glifosato (C
3
H
8
NO
5
P). .........................................................….......08
Figura 3.1 -
Representação esquemática da gota pendurada (hanging drop). ………………......………15
Figura 3.2 A e B
- Cristais de MtCQ-MgADP-Chiquimato com tamanho entre 0,35 mm e 0,4 mm. .....17
Figura 3.3
– Gráficos de Ramachandran:
A
) MtCQ-MgADP-Chiquimato,
B
) MtCQ-MgADP (1L4Y),
C
) MtCQ-MgADP (1L4U). ..................………………………………......……......................………….20
Figura 3.4
– Elementos de estrutura secundária da MtCQ. ..........................……………………………21
Figura 3.5
– Alinhamento das chiquimato quinases mostrando que os resíduos envolvidos na ligação do
chiquimato são conservados. .....................................................................................................................22
Figura 3.6
– Coordenação do Mg
+2
.
A)
1L4Y,
B)
1L4U
, C)
MtCQ-MgADP-Chiquimato. …................25
Figura 3.7
– Mapa F
obs
-F
calc
(3σ) gerado para posicionar o ácido chiquímico na estrutura da MtCQ. A
figura mostra também os Cα da MtCQ-mgADP-Ácido chiquímico, juntamente com as moléculas de
ADP e Mg
2+
. ..............................................................................................................................................26
Figura 3.8
– Estrutura molecular do chiquimato mostrando o grupo carboxila e os três grupos
hidroxila. ...................................................................................................................................................27
Figura 3.9
– Interações envolvidas na ligação chiquimato no sítio ativo da MtCQ. ............................... 29
Figura 3.10
– Ligação do íon cloreto no sítio ativo das estruturas da MtCQ-MgADP-Chiquimato e
MtCQ-MgADP (1L4U; Gu et al., 2002). ..................................................................................................31
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.11 –
Superposição da MtCQ-MgADP-Chiquimato (vermelho) e o mutante K15M/P115L da
EcCQ (azul) (Krell et al., 2001) mostrando o grande deslocamento do domínio LID e do domínio SB
devido a ligação do ADP e do ácido chiquímico. ......................................................................................33
Figura 3.12
– Superposição dos C
α
do domínio LID e do domínio SB das estruturas MtCQ-MgADP-
Chiquimato (cinza) e MtCQ-MgADP (verde) (1L4U; Gu et al., 2002). ...................................................35
Figura 3.13
– Superfície molecular das estruturas (A) MtCQ-MgADP (1L4U; Gu et al., 2002) e (B)
MtCQ-MgADP-Chiquimato. .....................................................................................................................36
Figura 3.14 –
Reação catalisada pela EPSPS sintase. ...............................................................................39
Figura 3.15
–
Passos para construção de um modelo utilizando a modelagem molecular
comparativa. ...............................................................................................................................................41
Figura 3.16
– Alinhamento das seqüências de aminoácidos da MtEPSPS e da EcEPSPS. .....................41
Figura 3.17
– (A) MtEPSPS sem ligantes – estrutura aberta e (B) MtEPSPS complexada com
chiquimato-3-fosfato e glifosato – estrutura fechada. ...............................................................................42
Figura 3.18 –
Sobreposição da MtEPSPS aberta (linha grossa) e fechada (linha fina) mostrando a
diferença conformacional entre as estruturas devido a ligação do chiquimato-3-fosfato e do
glifosato. ....................................................................................................................................................44
Figura 3.19
– (A) Sítio de ligação do chiquimato-3-fosfato e do (B) glifosato da MtEPSPS. .................45
Figura 3.20
– Reação catalisada pela corismato sintase. ..........................................................................48
Figura 3.21
– O alinhamento das sequências da MtCS e SpCS. Os elementos de estrutura secundária
para o modelo MtCS são indicados de acordo com a região onde são encontrados. ................................51
Figura 3.22 –
Diagrama de Ribbon do modelo molecular da MtCS complexado com FMN e EPSP. A)
Forma monomérica e B) Forma oligomérica (tetrâmero). ........................................................................53
Figura 3.23A
– Sítio de ligação da FMN. ................................................................................................54
Figura 3.23B
– Sítio de ligação do EPSP. ................................................................................................55
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.24 –
Sobreposição dos espectros de dicroísmo circular da corismato sintase de Mycobacterium
tuberculosis sem ligante e na presença de ligante. ...................................................................................56
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Lista de Tabelas
Tabela 1.1
– Fármacos utilizados no tratamento de tuberculose. .............................................................04
Tabela 3.1
. Estruturas cristalográficas da chiquimato quinase previamente depositadas. .......................13
Tabela 3.2 –
Estatística do processamento de dados de difração de raios X para MtCQ-MgADP-
Chiquimato. ...............................................................................................................................................17
Tabela 3.3
– Estatística do refinamento para MtCQ-MgADP-Chiquimato. ............................................19
Tabela 3.4 – Ligação do Mg
2+
nas MtCQs. ..............................................................................................26
Tabela 3.5
– Ligações de hidrogênio direta ou mediada por água entre o chiquimato e MtCQ. .............29
Tabela 3.6
– Ligação do íon cloreto nas estruturas da MtCQ-MgADP (1L4Y e 1L4U) e MtCQ-MgADP-
Chiquimato. ...............................................................................................................................................32
Tabela 3.7
– Ligações de hidrogênios entre MtEPSP e chiquimato-3-fosfato. ........................................45
Tabela 3.8
– Ligações de hidrogênio entre MtEPSP e glifosato. .............................................................46
Tabela 3.9 –
Resultados das análises estruturais da MtCS e SpCS. .........................................................52
Tabela 3.10 – Porcentagem dos elementos de estrutura secundária calculado para os dados experimentais
de CD e para o modelo molecular da MtCS. ..............................................................................................57
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Glossário
ADP – Adenosina difosfato
AK – Adenilato quinase
aroA – gene codificador da EPSP sintase
aroB – gene codificador da 3-desidroquinato sintase
aroD – gene codificador da 3-desidroquinato desidratase
aroE – gene codificador da chiquimato desidrogenase
aroF
–
gene codificador da corismato sintase
aroG – gene codificador da DAHP sintase
aroK – gene codificador da chiquimato quinase (isoforma I)
aroL – gene codificador da chiquimato quinase (isoforma II)
ATP – Adenosina trifosfato
CQ – Chiquimato quinase
CS – Corismato sintase
3D
–
Tridimensional
DAHP – 3-deoxi-D-arabino-heptulosonato 7-fosfato
DHFR – Disidrofolato redutase
DHQase
– Desidroquinase
E. coli – Escherichia coli
EcEPSPS – EPSPS de Escherichia coli
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
EPSP – 5-enolpiruvilchiquimato-3-fosfato
EPSPS – 5-enolpiruvilchiquimato-3-fosfato sintase
ErcCQ – Chiquimato quinase de Erwinia chrysanthemi
FDA – Food and Drug Administration
FMN – Flavina mononucleotídeo
gaps – intervalos na seqüência de aminoácidos
HIV – Vírus da imunodeficiência adquirida humana
Km – Constante de Michaelis-Menten
MTB – Mycobacterium tuberculosis
MtCQ – Chiquimato quinase de Mycobacterium tuberculosis
MtCS – Corismato sintase de Mycobacterium tuberculosis
MtEPSPS – EPSPS de Mycobacterium tuberculosis
NMP – Nucleosideo monofosfato
PAS – ácido p-aminossalissílico
PDB – Protein Data Bank
PEG – Polietilenoglicol
PEP – Fosfoenol-piruvato
P-loop – Alça de ligação do fosfato
RMSD – Root mean square deviation (Desvio médio quadrático)
SpCS – Corismato sintase de Streptococcus pneumoniae
TB – Tuberculose
TyrR – regulador de expressão da CQ II de E. coli
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Índice
Agradecimentos ........................................................................................................ iii
Resumo ..................................................................................................................... iv
Abstract ..................................................................................................................... v
Lista de Figuras ......................................................................................................... vi
Lista de Tabelas ........................................................................................................ ix
Glossário …………................................................................................................... x
1. Introdução ............................................................................................................. 01
1.1 A tuberculose ................................................................................................ 01
1.2 Via metabólica do ácido chiquímico ............................................................ 05
2. Objetivos ............................................................................................................... 09
3. Resultados ............................................................................................................. 10
3.1 Estrutura cristalográfica da chiquimato quinase ..........................................
10
3.1.1 Materiais e Métodos .................................................................. 14
3.1.2 Resultados e discussão .............................................................. 16
3.1.3 Conclusão ..................................................................................
36
3.2 Modelagem molecular da EPSP sintase ...................................................... 39
3.2.1 Materiais e Métodos .................................................................. 40
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.2.2 Resultados e discussão .............................................................. 41
3.3 Modelagem molecular e análise de CD da corismato sintase ......................
48
3.3.1 Materiais e Métodos .................................................................. 49
3.3.2 Resultados e discussão .............................................................. 50
4. Referências ............................................................................................................ 58
Anexos
1 - Pereira, J.H., Oliveira, J.S., Canduri, F., Dias, M.B.V., Palma, M.S., Basso,
L.A, Santos, D.S., De Azevedo, W. F. Jr. (2004). Structure of shikimate
kinase from Mycobacterium tuberculosis reveals the binding of shikimic acid.
Acta Crystallographica Section D. 60, 2310-2319.
2 - Pereira, J.H., Vasconcelos, I.B., Oliveira, J.S., Basso, L.A, Santos, D.S.
Shikimate Kinase: A potential target for development of novel anti-tubercular
agents (Review). Current Drug Targets. Submetido em Maio de 2005.
3 - De Azevedo, W. F. Jr., 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. Biochemical and Biophysical Research
Communications.
295
, 142-148.
4 - Pereira, J.H., Canduri, F., Oliveira, J.S., da Silveira, N.J.F., Basso, L.A,
Palma, M.S., De Azevedo, W. F. Jr., Santos, D.S. (2003). Structural
bioinformatics study of EPSP synthase from Mycobacterium tuberculosis.
Biochemical and Biophysical Research Communications. 312 (3), 608-614.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
5 - Pereira, J.H., Dias, M.V.B., Canduri, F., Ely, F., Ruggiero, J. N., Frazzon, J.,
Basso, L.A, Palma, M.S., Santos, D.S., De Azevedo, W. F. Jr.
Molecular
modeling and CD analysis of chorismate synthase.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
1. Introdução
1.1 A tuberculose
As doenças infecciosas ainda são responsáveis pelo sofrimento e morte de centenas
de milhões de pessoas e 90% das mortes causadas por esse tipo de enfermidade ocorrem
em áreas tropicais e subtropicais, regiões nas quais estão presentes 80% de toda
população mundial (Trouiller et al., 2001). Entre estas doenças destacam-se a
AIDS/HIV, infecções respiratórias, malária e tuberculose. A infecção pelo
Mycobacterium tuberculosis
1
, agente causador da tuberculose, é a principal causa de
morte em humanos devido a um único agente infeccioso. A tuberculose também pode
ser causada por outras bactérias do gênero Mycobacterium, como por exemplo, o
Mycobacterium africanum, o Mycobacterium microti e o Mycobacterium bovis.
Figura 1.1
–
Microscopia eletrônica de varredura do Mycobacterium tuberculosis. Os bacilos
apresentam 1-4
μ
m de comprimento e 0,3-0,6
μ
m de diâmetro (Pelczar et al., 1996).
1
Mycobacterium tuberculosis também é conhecido como bacilo de Koch, pois em 24 de Março de 1882, Robert
Koch identificou o agente causador da tuberculose.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
De acordo com a Organização Mundial da Saúde, um terço da população mundial
está infectada, e a estimativa é que 10% desta fração desenvolva a doença ao longo de
sua vida (WHO, 2004a; Bloom & Murray, 1992). Anualmente, são mais de 8 milhões de
pessoas que desenvolvem a tuberculose ativa, resultando em uma taxa de 2 milhões de
mortes/ano (WHO, 2004a; 2004b). A cada dia, mais de 25 mil pessoas ficam doentes, e
5 mil pessoas acabam morrendo (WHO, 2003). No Brasil, de acordo com o relatório da
Funasa
2
, entre os anos de 1980 a 2000 a tuberculose apresentou uma ocorrência de 80 a
100 mil casos/ano, resultando em uma taxa de óbitos em torno de 6 mil mortes/ano
3
(FUNASA, 2002).
Apesar da tuberculose ser considerada por muitos como uma doença do século XIX,
na verdade ela nunca foi erradicada. Nos últimos anos tem-se verificado um aumento em
sua incidência: entre 1997 e 2000, o número de novos casos de tuberculose aumentou
1,8% ao ano (Corbett et al., 2003). Estima-se que entre 2002 e 2020, se não houver um
esforço para seu controle, aproximadamente 1 bilhão de pessoas serão infectadas, mais
de 150 milhões ficarão doentes, e a tuberculose levará à morte mais de 38 milhões de
pessoas (WHO, 2001; 2002). Dada a gravidade da situação global da doença, em 1993 a
Organização Mundial da Saúde declarou a tuberculose um sério problema de saúde
pública mundial, como forma de alertar os setores de saúde pública e os governos para a
necessidade de contenção desta doença.
2
FUNASA – Fundação Nacional de Saúde.
3
Oficialmente, a doença mata a cada ano 5 a 6 mil pessoas no Brasil, mas especialistas acreditam que o número de
óbitos deve chegar a 10 mil (Pivetta, 2004).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
A transmissão do bacilo ocorre via aérea, devido à inalação de pequenas partículas de
aerossóis expelidas através da fala, espirro e, principalmente, pela tosse dos pacientes
com a forma pulmonar da doença. Estas partículas são inaladas pelo indivíduo sadio,
atingindo o espaço alveolar, onde iniciam sua multiplicação e são fagocitadas pelos
macrófagos do hospedeiro (Hiriyanna & Ramakrishnan, 1986). Como o Mycobacterium
tuberculosis (MTB) é capaz de resistir ao ataque, sobreviver e multiplicar-se no interior
de células fagocitárias, como os macrófagos, o MTB é considerado um parasita
intracelular (Clemmens, 1996).
À medida que macrófagos, monócitos e células T são recrutados para a área em torno
do bacilo, ele diminui sua replicação e permanece em estado latente até um momento de
deficiência do sistema imunológico. O MTB normalmente sobrevive durante anos neste
estado de dormência dentro dos tecidos (Manabe & Bishai, 2000).
A incidência de tuberculose teve um declínio rápido no início do século vinte nos
países em desenvolvimento, devido à melhora nas condições sanitárias e das moradias.
Essa tendência foi acelerada inicialmente pela introdução da vacinação BCG
4
(1921) e
da descoberta de antibióticos como a estreptomicina (1944) e, posteriormente, com a
descoberta do ácido p-aminossalissílico (PAS) (1946), isoniazida (1952) e rifampicina
(1965) (Drobniewski et al., 2003; Duncan, 2003).
Até o início dos anos 80, o tratamento com estes quimioterápicos promoveu o
controle e o declínio da doença. Entretanto dois principais fatores têm contribuído para
4
A Vacina BCG (Bacilo de Calmette-Guérin) contra a tuberculose foi desenvolvida pelo Dr. Albert Calmette,
juntamente com seu assistente Camille Guérin. Os primeiros indivíduos (120 recém nascidos de mães tuberculosas)
receberam a vacina em Julho de 1921 (Hawgood, 1998).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
aumento no número de casos de tuberculose. O primeiro é a suscetibilidade de pacientes
co-infectados com o vírus HIV, cujo sistema imunológico enfraquecido não pode
controlar o crescimento do bacilo. Estes pacientes apresentam um risco cem vezes maior
de desenvolver a doença (El Syed et al., 2000). O outro fator é a emergência de cepas
resistentes aos antimicrobianos de primeira linha (isoniazida e rifampicina) utilizados no
tratamento, devido a terapias inadequadas e o uso indiscriminado destes antibióticos
(Baptista et al., 2002; Duncan, 2003).
O tratamento contra tuberculose deve incluir uma ação bactericida contra o bacilo
ativo, seguida da esterilização da população de bacilos dormentes (Mitchison, 1985).
Atualmente, existem 10 medicamentos que estão aprovados pela Food and Drug
Administration (FDA) dos E.U.A. (Villar, 2004). A isoniazida, rifampicina,
pirazinamida e o etambutol são considerados como agentes terapêuticos de 1ª linha. A
Cicloserina, Etionamida, o ácido p-aminosalicílico, capreomicina e a estreptomicina são
considerados fármacos de 2ª linha. A estreptomicina, que pertenceu durante muitos anos
aos antibacilares de 1ª linha, atualmente, devido ao cada vez maior número de
resistências, viu muito diminuída a sua utilidade, tendo sido relegada para 2ª linha. A
Rifabutina é aprovada na utilização para tratamento da tuberculose de pacientes
infectados com vírus HIV.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 1.1
– Fármacos utilizados no tratamento da tuberculose.
1ª linha 2ª linha
Isoniazida Cicloserina
Rifampicina Etionamida
Pirazinamida Levofloxacina
#
Etambutol Moxifloxacina
#
Rifabutina
*
Gatifloxacina
#
PAS
Estreptomicina
Amicacina / Canamicina
#
Capreomicina
*
Utilizado em situações particulares
#
Fármacos usados para tratamento de tuberculose multiresistente, apesar de não serem aprovados pelo
FDA.
O tratamento recomendado pela Organização Mundial da Saúde apresenta uma
duração mínima de seis meses. Nos dois primeiros meses, envolve a combinação de
quatro antibióticos: isoniazida, rifampicina, pirazinamida e o etambutol, seguido da
combinação de isoniazida e rifampicina por mais quatro meses (ou de isoniazida
combinada com etionamida ou etambutol por mais seis meses).
Os mecanismos de resistência aos agentes anti-tuberculose são devido às alterações
no DNA cromossomal, portanto estas cepas não estão sujeitas à seleção reversa, e
permanecerão causando a falha de tratamentos padrões à tuberculose. Desta forma, o
desenvolvimento de novas drogas para substituírem àquelas comprometidas pela
resistência torna-se necessário para que se possa estabelecer um tratamento
quimioterápico eficaz.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
O principal objetivo da quimioterapia é atacar alvos peculiares aos microrganismos
como, por exemplo, vias metabólicas ausentes no organismo humano. Sob este aspecto,
as enzimas da via do ácido chiquímico representam bons exemplos de tal abordagem.
1.2 Via metabólica do Ácido Chiquímico
A via metabólica do ácido chiquímico foi descoberta através dos estudos de Bernhard
Davis e David Sprinson (Davis & Mingioli, 1953; Sprinson,1960). Esta rota biossintética
faz a conexão entre o metabolismo de carboidratos e a síntese de compostos aromáticos
através de sete passos metabólicos, onde o fosfoenol-piruvato e eritrose 4-fosfato são
convertidos em ácido corísmico (Figura 1.2) (Pittard, 1987; Haslam, 1993). O ácido
corísmico é um precursor comum para a síntese de aminoácidos aromáticos, folato,
ubiquinonas, menaquinonas e enterobactim.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 1.2
– A via do ácido chiquímico. Fosfoenol-piruvato e eritrose 4-fosfato (precursores) são
convertidos em ácido corísmico (corismato). Corismato é essencial paras síntese de aminoácido
aromático (Phe, Tyr e Trp), folato, ubiquinonas, menaquinonas e entorobactim. Os sete passos são
catalisados pelas enzimas: 3-deoxi-D-arabino-heptulosonato 7- fosfato sintase (DAHP sintase), 3-
desidroquinato sintase, 3-desidroquinato desidratase, chiquimato desidrogenase, chiquimato quinase, 5-
enolpiruvil chiquimato 3-fosfato sintase (EPSP sintase), e corismato sintase. Figura modificada a partir
de Mathews & van Holde. (Mathews & van Holde, 1990).
As enzimas da via do ácido chiquímico são alvos potenciais para desenvolvimento de
anti-microbianos (Davies et al., 1994) e herbicidas (Coggins, 1989), pois esta via é
essencial para algas, plantas superiores, bactérias, fungos, e inexistente em mamíferos
(Bentley, 1990). Recentemente, a via do ácido chiquímico também foi identificada em
parasitas do filo Apicomplexa (Plasmodium falciparum, Toxoplasma gondii e
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Cryptosporidium parvum), fornecendo novos alvos para o desenvolvimento de drogas
anti-parasitas (Roberts et al., 1998).
A organização molecular das enzimas da via metabólica do ácido chiquímico altera-
se de acordo com os grupos taxonômicos (Coggins et al., 1987). As enzimas de bactérias
apresentam-se na forma de sete cadeias polipeptídicas individuais, que são codificadas
por genes separados. As plantas possuem um arranjo similar às bactérias (Butler et al.,
1974), com a exceção da desidroquinase (DHQase, terceira enzima) e a chiquimato
desidrogenase (quarta enzima) que estão presentes como domínios separados em uma
cadeia polipeptídica bifuncional (Mousdale et al., 1987). Em fungos e parasitas do filo
Apicomplexa, a via do ácido chiquímico apresenta a 3-deoxi-D-arabino- heptulosonato
7- fosfato (DAHP) sintase e a corismato sintase (CS) como enzimas monofuncionais e
um polipeptídeo pentafuncional chamado AROM, que é responsável pela cinco reações
restantes da via do ácido chiquímico (Duncan et al., 1987). Especificamente em
Mycobacterium tuberculosis, com a publicação do genoma do bacilo (Cole et al., 1998),
verificou-se a presença dos genes codificadores das enzimas envolvidas na via do ácido
chiquímico: aroG (DAHP sintase), aroB (3-desidroquinato sintase), aroD (3-
desidroquinato desidratase), aroE (chiquimato desidrogenase), aroK (chiquimato
quinase), e aroA (EPSP sintase), aroF (corismato sintase).
A utilização das enzimas da via biossintética do ácido chiquímico como alvos para o
desenvolvimento de inibidores pode ser plenamente justificada e validada pela inibição
da EPSP sintase através do glifosato (N-(fosfometil)glicina) (Kishore & Shah, 1988).
Glifosato é um importante composto guia para a construção de novos inibidores de EPSP
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
sintase. Este composto atualmente é utilizado como herbicida, e está presente nos
defensivos agrícola Round Up e TouchDown, usados no controle de ervas daninhas.
Roberts e seus colaboradores demonstraram também que o glifosato é capaz de inibir in
vitro o crescimento dos parasitas Plasmodium falciparum, Toxoplasma gondii e
Cryptosporidium parvum (Roberts et al., 1998).
Figura 1.3 – Estrutura molecular do glifosato (C
3
H
8
NO
5
P). Este composto exemplifica o grande
potencial de desenvolvimento de novos inibidores baseados nas enzimas da via metabólica do ácido
chiquímico. As coordenadas atômicas do glifosato em complexo com a EPSPS foram obtidas no PDB
(código PDB:1G6S) (Schönbrunn et al., 2001). Figura gerada utilizando o programa MolMol (Koradi et
al., 1996).
Com o objetivo de demonstrar a importância da via do ácido chiquímico em
Mycobacterium tuberculosis
, foi realizado a disrrupção do gene
aroK
, que codifica a
chiquimato quinase. Os resultados obtidos através destes experimentos confirmaram que
a via do ácido chiquímico é essencial para viabilidade do bacilo (Parish & Stoker, 2002).
Uma vez validado a via como um potencial alvo para inibir o desenvolvimento do
Mycobacterium tuberculosis, estudos estruturais das enzimas através de modelagem
molecular e cristalografia foram realizados com o objetivo de fornecer informações que
possam ser utilizadas para desenvolvimento de novos inibidores.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
2. Objetivos
Este trabalho teve como objetivo realizar o estudo estrutural através da cristalografia
de raios X e modelagem molecular das enzimas responsáveis pelos três últimos passos
catalíticos da via metabólica do ácido chiquímico de Mycobacterium tuberculosis. As
enzimas são: chiquimato quinase (MtCQ), EPSP sintase (MtEPSPS) e corismato sintase
(MtCS).
O estudo cristalográfico da chiquimato quinase teve como finalidade determinar o
modo de interação do ácido chiquímico (chiquimato) com a enzima, pois nenhuma
estrutura depositada de CQ apresentava seu substrato complexado (Krell et al., 1998;
Krell et al., 2001; Gu et al., 2002). Uma vez que pretendíamos auxiliar no
desenvolvimento de novos inibidores para CQ, o conhecimento detalhado do sítio ativo
da CQ era de fundamental importância para atingirmos nosso objetivo.
Experimentos de cristalização com MtEPSPS foram realizados, mas até o presente
momento não foi obtido cristais desta enzima. Assim, a modelagem molecular da
MtEPSPS apo enzima e MtEPSPS complexada com chiquimato-3-fosfato e glifosato foi
realizada para extrair informações do modo que esses ligantes interagem com a enzima.
Da mesma maneira, estudo por modelagem molecular da MtCS complexada com seu
substrato (EPSP) e FMN (cofator) foi realizado, assim como experimentos de CD para
reforçar o modelo que estava sendo proposto.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.Resultados
3.1 Estrutura cristalográfica da chiquimato quinase
complexada com MgADP e
ácido chiquímico (chiquimato)
A chiquimato quinase (EC 2.7.1.71), quinta enzima da via, catalisa a fosforilação do
grupo 3-hidroxila do chiquimato usando ATP como co-substrato (Figura 1.2). Em
Escherichia coli, a reação da CQ é catalisada por duas isoformas: CQ I que é codificada
pelo gene aroK (Whipp & Pittard, 1995) e CQ II que é codificada pelo gene aroL (Millar
et al., 1986). A principal diferença entre as isoenzimas são os valores de Km encontrados
para chiquimato, 20mM para a CQ I e 0,2mM para a CQ II. A isoforma CQ II apresenta
um papel dominante na via do ácido chiquímico, onde a sua expressão é controlada pelo
regulador
tyr
R e a sua repressão é realizada pela tirosina e triptofano (Ely & Pittard,
1979; De Feyter et al.,1986). O papel fisiológico da CQ I em E.coli não está claro.
Mutações em CQ I foram associadas com a sensibilidade ao antibiótico mecilinam
(Vinella et al., 1996), sugerindo que CQ I deve apresentar um papel biológico
alternativo e que promove a fosforilação do chiquimato somente casualmente (De Feyter
& Pittard, 1986).
Ao contrário de E.coli, seqüências de genomas completos de bactérias, como por
exemplo Haemophilus influenzae e Mycobacterium tuberculosis, demonstraram a
presença de somente um gene codificador de CQ. A maioria das CQs são codificadas
pelo gene aroK, devido as seqüências de aminoácidos possuir um maior grau de
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
identidade com a CQ I de E. coli. Os parâmetros cinéticos para MtCQ (codificada pelo
gene aroK) são mais similares a àqueles encontrados para CQ II E. coli do que aqueles
de CQ I E. coli. A MtCQ apresenta um valor de Km para chiquimato de 0,41mM
indicando que nem todas as CQs codificadas pelo gene aroK possuem altos valores de
Km para substrato (Gu et al., 2002).
A estrutura cristalográfica de CQ de Erwinia chrysanthemi (ErcCQ) (Krell et al.,
1998; Krell et al., 2001), que é codificada pelo gene aroL, demonstrou que a chiquimato
quinase é uma proteína da classe α/β que apresenta na sua estrutura fitas β rodeadas por
hélices α. No centro da proteína estão localizadas as 5 fitas β paralelas (β1-β5)
formando uma folha β na ordem 23145, que é flanqueada por oito hélices α. A ordem
da folha β 23145 observada na estrutura da ErcCQ, classifica as CQs como pertencente
à família das nucleosideo monofosfato (NMP) quinases (Yan & Ysai, 1999). Esta
família pode ser exemplificada pelas enzimas: hexoquinases (Bennett & Steitz, 1980),
adenilato quinase (AK) (Dreusicke, et al., 1988; Schlauderer & Schulz, 1996), guanilato
quinase (Stehle & Schulz, 1990), uridilate quinase (Müller-Dieckmann & Schulz, 1994)
e timidino quinase (Wild et al., 1995).
As NMP quinases são compostas de três domínios: domínio central (Core domain),
domínio da tampa (LID domain) e o domínio de ligação do NMP (NMP-binding
domain). Uma característica que retrata as NMP quinases é que estas enzimas sofrem
uma grande mudança conformacional durante a catálise (Vonrhein et al., 1995). Existem
duas regiões flexíveis na estrutura que são responsáveis por este movimento: uma é o
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
domínio de ligação do NMP que é formado por uma série de hélices entre as fitas-β 2 e
3, e a outra região é o domínio LID, uma região de tamanho variado localizado após a
quarta fita-β (β4) (Müller et al., 1996; Gerstein et al., 1993). Em CQ, o domínio de
ligação do chiquimato (Domínio SB) corresponde ao domínio de ligação do NMP de
NMP quinases.
Três motivos funcionais presentes em enzimas de ligação de nucleotídeo são
encontrados nas chiquimato quinases: motivo Walker A, motivo Walker B, e um motivo
de ligação da adenina. O motivo Walker A está localizado entre a primeira fita-β (β1) e a
primeira hélice-α (α1), contendo o trecho de seqüência conservada GXXXXGKT/S,
onde X representa qualquer resíduo (Walker et al., 1982). Este motivo forma uma alça
de ligação do fosfato (
P-loop
), onde diferentes amidas da cadeia principal fazem
ligações de hidrogênio com β-fosfato do ADP (Smith & Rayment, 1996). Em MtCQ,
este motivo compreende a região do resíduo 9 até 17, com a seqüência GLPGSGKST.
Em adição ao motivo Walker A, uma segunda seqüência conservada ZZDXXG é
chamada de motivo Walker B, onde Z representa resíduo hidrofóbico (Walker et al.,
1982). Um Asp
→
Ser é trocado em MtCQ. O consenso para o motivo Walker B em
chiquimato quinases é ZZZTGGG , onde a segunda glicina (Gly80 em MtCQ) faz
ligação de hidrogênio com o γ-fosfato do ATP. A alça de ligação da adenina,
caracterizada pela seqüência I/VDAXQ/NXP, foi descrita em Adenilato Quinase e
ErcCQ. Em MtCQ, o motivo de ligação da adenina apresenta a seqüência VDTNRRNP
(resíduos 148 até 155). Portanto, o motivo de ligação da adenina poderia ser descrito
mais precisamente pelo trecho I/VDXXX(X)XP (Gu et al, 2002).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Os estudos cristalográficos realizados com ErcCQ (código de acesso PDB: 1SHK,
2SHK e 1E6C) (Krell et al., 1998) não apresentaram densidade eletrônica suficiente para
incluir o chiquimato na estrutura. As duas estruturas cristalográficas de CQ de
Mycobacterium tuberculosis resolvidas por Gu et al (Gu et al., 2002) (código de acesso
PDB: 1L4Y e 1L4U) revelou o movimento do domínio LID na catálise, mas a posição
precisa do chiquimato e suas interações com a chiquimato quinase não foram
determinadas. Embora as estruturas de ErcCQ e de MtCQ não apresentarem o
chiquimato, o substrato foi adicionado em suas condições de cristalização (Tabela 3.1).
A tabela 3.1 também mostra os resíduos e os ligantes que foram determinados nas
estruturas de ErcCQ-MgADP e MtCQ-MgADP.
Tabela 3.1. Estruturas cristalográficas da chiquimato quinase previamente depositadas.
Aditivos (mM) Estr. cristalográficas das CQ com ligantes
PDB (Å) Ác. Chiq. ADP MgCl
2
Mol Resid.
depositados
Ác.Chiq MgADP
1SHK
a
1,9 5 5 10 A 1-112, 128-173
(1)
b
B 1-112, 126-172
2SHK
a
2,6 5 5 10 A 1-112, 128-173 (1)
b
B 1-112, 123-173 1
1E6C
b
1,8 2,5 2,5 10 A 3-170
B 3-170
1L4U
d
1,8 4 4 4 -- 2-166 1
1L4Y
e
2,0 5 5 5 -- 2-166 1
a
ErcCQ – Chiquimato quinase de Erwinia chrysanthemi (Krell et al., 1998)
b
Mapas de densidade eletrônica Fo-Fc indicam uma possível posição do chiquimato (Krell et al.,
1998)
c
apo-ErcCQ (K15M) (Krell et al., 2001)
d
Derivativo-Pt da MtCQ (Gu et al., 2002)
e
MtCQ nativa (Gu et al., 2002)
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Embora havendo estruturas de chiquimato depositadas no PDB, o modo de ligação
do substrato com a enzima permanecia indeterminado. Deste modo, o estudo estrutural
com a MtCQ teve como objetivo revelar detalhadamente o modo interação dos resíduos
de aminoácidos com o chiquimato e as mudanças conformacionais que a enzima sofre
devido a ligação do substrato. A estrutura da MtCQ-MgADP-Chiquimato fornecerá
informações importantes para o desenvolvimento de inibidores de CQ e também para
entendimento do mecanismo catalítico.
3.1.1 Materiais e métodos
A MtCQ foi obtida a partir de métodos de clonagem, expressão e purificação pelo
grupo do Prof. Dr. Diógenes Santiago Santos, da Rede Brasileira de Pesquisa em
Tuberculose, do Departamento de Biologia Molecular e Biotecnologia (UFRGS), Porto
Alegre, Brasil (Oliveira et al., 2001).
Cristalização
A proteína, precipitada em sulfato de amônio, foi solubilizada e dialisada contra
Hepes 50 mM (pH 7.5), NaCl 0,5M e MgCl
2
5mM. A esta preparação foram
adicionados chiquimato 13mM e ADP 13mM, e a mistura foi centrifugada por 5 minutos
a 9.300 g. Os cristais foram obtidos usando uma variação do método de difusão de
vapor, chamada de gota suspensa (hanging drop).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Para a cristalização do complexo MtCQ-MgADP-Chiquimato foi adicionado nos
reservatórios de uma placa Linbro, 1 mL de uma solução composta de 0,1 M de Hepes
(pH 7,5), 10 % de isopropanol e 35 % de PEG 3350. As gotas foram feitas sobre uma
lamínula siliconizada de 22x22mm, na qual foi utilizado 1 μl da solução do reservatório
e 1,5 μl da solução de proteína a uma concentração de 17 mg.ml
-1
. Portanto, cada gota
apresentava um volume final de 2,5 μl.
Figura 3.1 -
Representação esquemática da gota pendurada (hanging drop)
Coleta e processamento dos Dados
O conjunto de dados (160 imagens) da MtCQ-MgADP-Chiquimato foi coletado a um
comprimento de onda de 1,431 Å, usando-se como fonte radiação síncroton (Estação
PCr, Laboratório Nacional de Luz Síncroton, LNLS, Campinas, Brasil) (Polikarpov et
al., 1998a; Polikarpov et al., 1998b), e um detector CCD (MARCCD) a uma distância de
90 mm do cristal. O protetor criogênico utilizado era composto glicerol 15 %, PEG 3350
12 %, Hepes 35mM, isopropanol 3,5 %, e a coleta foi mantida a uma temperatura de 104
K. O tempo de exposição foi de 50 segundos por imagem, usando um ângulo de
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
oscilação de 1,0º. Os dados foram processados utilizando os programas MOSFLM
(Leslie, 1992) e escalonados com o programa SCALA (Colaborative Computational
Project nº 4, 1994).
Substituição Molecular e Refinamento cristalográfico
A estrutura cristalográfica da MtCQ-MgADP-Chiquimato foi determinada por
métodos padrões de substituição molecular usando o programa AMoRe (Navaza, 1994),
tendo sido usado como modelos de busca as estruturas da chiquimato quinase de
Mycobacterium tuberculosis resolvida anteriormente (código de acesso PDB: 1L4Y e
1L4U) (Gu et al., 2002). As posições atômicas obtidas por substituição molecular foram
usadas para iniciar o refinamento cristalográfico.
Para o refinamento da estrutura foi utilizado o programa X-PLOR (Brünger, 1992).
A qualidade estereoquímica do modelo final da MtCQ-MgADP-Chiquimato foi acessada
pelo programa PROCHECK (Laskowski et al, 1994). Modelos atômicos foram
superpostos usando o programa LSQKAB presente no CCP4 (Colaborative
Computational Project nº 4, 1994). A superfície molecular da MtCQ-MgADP-
Chiquimato foi calculada usando o programa Swiss PDB Viewer v.3.7 (Guex & Peitsch,
1997).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.1.2 Resultados e discussão
Os cristais de MtCQ-MgADP-Chiquimato foram obtidos na presença de 13mM
chiquimato e 13mM ADP e apareceram após 1 dia a temperatura de 20º C (Figura 3.2).
O conjunto de dados da MtCQ-MgADP-Chiquimato foi coletado a 2.3 Å usando fonte
de radiação síncrotron (Tabela 3.2).
Figura 3.2 A e B
- Cristais de MtCQ-MgADP-Chiquimato com tamanho entre 0,35 mm e 0,4 mm .
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 3.2 – Estatística do processamento de dados de difração de raios X para MtCQ-
MgADP-Chiquimato
a (Å) 62,91
b (Å) 62,91
c (Å) 90,92
α(º)
90,00
β(º)
90,00
γ(º)
120,00
Grupo espacial P3
2
21
Número de medidas com I>2
σ
(I)
34.274
Número de reflexões independentes 9.563
Completeza (%) 98,7
R
*
sym
(%) 3,0
Camada de mais alta resolução (Å) 2,41 – 2,30
Completeza à mais alta resolução (%) 98,7
R
*
sym
à mais alta resolução (%) 7,20
*R
sym
= 100 Σ ⎪I(h) - <I(h)>⎪/ Σ I(h) com I(h), intensidade observada e <I(h)>, intensidade de
reflexões h sobre todas as medidas de I(h).
O cristal apresentava uma molécula de MtCQ-MgADP-Chiquimato por unidade
assimétrica, contendo os resíduos 2-166, Mg
2+
, ADP, ác. chiquímico, dois íons Cl- e 144
moléculas de água (Tabela 3.3). O metionina do N-terminal e os dez resíduos do C-
terminal (NQIIHMLESN) não foram observados na MtCQ-MgADP-Chiquimato.
Informações sobre o refinamento e estatísticas do modelo final são mostradas na tabela
3.3. A média do fator-B para os átomos da cadeia principal é 34,64 Å
2
, enquanto para
átomos da cadeia lateral é de 35,68 Å
2
(Tabela 3.3). Para obter a posição de ligação do
chiquimato na MtCQ, uma maior concentração de substrato foi utilizada (13 mM)
comparado com a usada por Gu et al. (Gu et al., 2002) para obter os cristais (5mM). A
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
média do fator-B de 26,15 Å
2
obtida para os átomos do chiquimato indica que este
ligante está altamente ordenado na estrutura
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 3.3 – Estatística do refinamento para MtCQ-MgADP-Chiquimato
MtCQ-MgADP-Chiquimato
Faixa de resol. usada (Å) 6,00 – 2,30
Reflexões usadas no ref. 8.885
Número de resíduos 165
Número de mol. de águas 144
Número de mol de ADP 1
Número de íons metal Mg
2+
1
Número de Cl
-
2
Número de moléculas de
chiquimato
1
Final R
factor
*
(%) 20,7
Final R
free
**
(%) 28,7
Fator B
+
(Å
2
)
Cadeia principal 34,64
Cadeias laterais 35,68
Águas 39,63
ADP 21,81
Mg
+2
34,70
Cl
-
37,65
Chiquimato 26,15
R.M.S.D observado da
geometria ideal
Comprimento de ligação (Å) 0,017
Ângulos de ligação (graus) 1,905
Diédros (graus) 22,125
Gráfico de Ramachandran
Φ/Ψ mais favoráveis (%) 93,4
Φ/Ψ não permitidos (%) 0
*
R
factor
= 100 x
Σ
|F
obs
- F
calc
|/
Σ
(F
obs
), somas obtidas sobre todas as reflexões com corte de F/
σ
(F)>2.
**
R
free
= R
factor
para 10% dos dados, os quais não foram incluídos no refinamento.
+
Fator B = valores médios do fator de vibração térmica para átomos de não-hidrogênios.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
A qualidade estereoquímica da estrutura MtCQ-MgADP-Chiquimato foi analisada
através do gráfico de Ramachandran (ângulos phi (φ) e psi (ψ)) e apresentou 93,4% dos
resíduos nas regiões mais favoráveis, e nenhum resíduo nas regiões não permitidas. As
estruturas 1L4Y e 1L4U apresentaram, respectivamente, 90,5% e 94,2% dos resíduos
nas regiões mais favoráveis e nenhum nas regiões não permitidas (Figura 3.3).
A B
Figura 3.3
– Gráficos de Ramachandran:
A
) MtCQ-
MgADP-Chiquimato,
B
) MtCQ-MgADP (1L4Y),
C
)
MtCQ-MgADP (1L4U). Em vermelho, regiões mais
favoráveis; em amarelo mais forte, regiões
adicionalmente permitidas; em amarelo claro, regiões
generosamente permitidas; e em branco, regiões não
permitidas. Resíduos de glicina são mostrados como
triângulos.
C
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Como citado anteriormente, uma característica marcante da CQ é que esta enzima
sofre uma grande mudança conformacional durante a catálise. As duas regiões
responsáveis são o domínio de ligação do chiquimato (Domínio SB) e o domínio LID
(Figura 3.4), que correspondem em MtCQ aos resíduos 33-61 e 112-124,
respectivamente.
Figura 3.4
– Elementos de estrutura secundária da MtCQ, juntamente com Mg
2+
, ADP, e Chiquimato.
As regiões LID e SB correspondem aos domínios de ligação do chiquimato e domínio da tampa,
respectivamente. Figura gerada utilizando o programa MolMol (Koradi et al., 1996).
Interação com ADP/Mg
2+
As estruturas da MtCQ complexadas com MgADP (Gu et al., 2002) são altamente
similares, apresentando um RMSD de 0,19 Å para todos os pares de átomos C
α
. A
porção adenina do ADP está localizada entre Arg110 e Pro155, como observado para as
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
estruturas MtCQ-MgADP. A Arg110 em MtCQ representa o primeiro resíduo de um
motivo conservado do domínio LID observado para P-loop quinases (Leipe et al., 2003).
O segundo resíduo básico conservado deste motivo (Arg117) interage com os grupos
α
-
e β-fosfato do ADP. Assim, este motivo conservado em CQ encontrado no domínio LID
pode ser descrito como R(X)
6-9
R (Figura 3.5). A Arg 117 deve estabilizar o estado de
transição neutralizando a carga negativa formada entre os átomos de oxigênio dos
fosfatos β-γ do ATP. A Pro 155 é o último resíduo do domínio de ligação da adenina
(resíduos 148-155 em MtCQ), que pode ser descrito como I/VDXXX(X)XP (Gu et al.,
2002).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.5
– Alinhamento das chiquimato quinases mostrando que os resíduos envolvidos na ligação do
chiquimato são conservados. Os elementos de estrutura secundária para MtCQ são mostrados acima da
seqüência. Motivos identificados em CQ são apresentados na figura, assim como o domínio SB e
domínio LID em azul e verde, respectivamente. Regiões preenchidas em amarelo indicam os resíduos
envolvidos na ligação com chiquimato, e a região preenchida em vermelho mostra a conservação do
Glu61.AK - AroK; AL - AroL; MYCTU – M.tuberculosis; CAMJE – C.jejuni; ECOLI – E.coli; SALTY –
S.typhimurium; ERWCH – E.chrysanthemi.
Apesar das estruturas da MtCQ-MgADP serem altamente similares, existem
diferenças na posição do Mg
2+
. Na MtCQ-MgADP (PDB 1L4Y) uma típica hexa-
coordenação é observada para o Mg
2+
(Figura 3.6A). Para a MtCQ-MgADP derivativo
Pt (PDB 1L4U) hexa-coordenação (na verdade sete interações são observadas)
apresenta-se desordenada (Figura 3.6B). A MtCQ-MgADP-Chiquimato (este trabalho)
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
apresentou uma hexa-coordenação desordenada do Mg
2+
(Figura 3.6C). Distâncias de
ligação para a ligação do Mg
2+
são mostradas na Tabela 3.4.
O Mg
2+
da MtCQ-MgADP-Chiquimato interage com oxigênio β-fosfato do ADP,
átomo OG da Ser16 do motivo Walker A, e quatro moléculas de água. Em ErcCQ, o
Mg
2+
é tetra-coordenado, interagindo com oxigênio β-fosfato do ADP, átomo OG1 da
Thr16, átomo OD2 do Asp32, e uma molécula água (Krell et al., 1998). A ligação do
Mg
2+
nas estruturas MtCQ-MgADP e MtCQ-MgADP-Chiquimato são de certa forma
similares, onde a Água2 coordenada com Mg
2+
faz ligação de hidrogênio com Asp32,
que é conservado em todas as CQs (Figura 3.5). Superposição das estruturas da MtCQ-
MgADP e MtCQ-MgADP-Chiquimato demonstrou que na última estrutura a Água1 e
Água4 estão em posições equivalentes, mas um deslocamento de 1,32 Å e 2,75 Å são
observados para Água2 e Água3, respectivamente (Figura 3.6). Em MtCQ-MgADP-
Chiquimato o Asp32 forma ligação de hidrogênio com Ser16, enquanto que a interação
entre Asp32 e Ser16 ocorre via ligação com uma molécula de água (Água6). A rotação
dos ângulos diedros χ1 e χ2 do Asp32 promove expulsão da Água6 do sítio de ligação
do Mg
2+
e uma interação direta com átomo OG da Ser16 é observada para MtCQ-
MgADP-Chiquimato (Figura 3.6C).
A Água1 coordenada pelo Mg
2+
interage com o íon cloreto ao invés de fazer uma
ligação com a Água5 que apresenta uma interação com Asp34 como observado para
MtCQ-MgADP. Conseqüentemente, o modo diferente de interação observado para o
Asp34 deve ser devido à presença do chiquimato que promove a expulsão da Água5 do
sítio ativo.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
O motivo Walker B para CQs (resíduos 75-81 em MtCQ) teve a seqüência proposta
como ZZZTGGG (Leipe et al., 2003). A segunda glicina (Gly80 em MtCQ) faz ligação
de hidrogênio com o
γ
-fosfato do ATP. O ataque nucleofílico ao
γ
-fosfato do ATP deve
ser facilitado pela ligação do íon metálico (Mg
2+
) aos grupos β- e γ-fosfato (Jencks,
1975a). Embora o mecanismo enzimático da MtCQ permaneça desconhecido, a
interação entre a Gly80 e o íon cloreto, que está ocupando o local de ligação do
γ
-
fosfato, indica que este resíduo deve participar do mecanismo catalítico das CQs.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.6
–
Coordenação do Mg
+2
.
A)
MTCQ-
MgADP (1L4Y),
B)
MTC-MgADP (1L4U)
, C)
MtCQ-MgADP-Chiquimato. Distância das ligações
entre o Mg
+2
e os oxigênios presente no β-fosfato
do ADP, a Ser16 e as moléculas de água (1-5)
estão descritas na tabela 3.4. Figura gerada
utilizando o programa MolMol (Koradi et al.,
1996).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 3.4 – Ligação do Mg
2+
nas MtCQs.
Átomos MtCQ-Chiq. 1L4Y 1L4U
Mg
2+
Ser16-OG 2,36 2,15 2,57
ADP-O2B 2,23 2,15 2,39
Água1-O 2,39 2,24 2,52
Água2-O 2,23 2,04 2,77
Água3-O 2,75 2,25 2,37
Água4-O 2,97 2,05 2,94
Água5-O n.o. n.o. 3,06
n.o.– Não observada
Sítio de ligação do chiquimato
O domínio de ligação do chiquimato, que localiza-se após a fita β2, consiste nas
hélices
α
2,
α
3 e a região N-terminal da hélice
α
4 (resíduos 33-61). Um pico maior que
3σ no mapa de diferença F
obs
-F
calc
revelou com clareza a posição de ligação do substrato
(Figura 3.7).
Figura 3.7 –
Mapa F
obs
-F
calc
(3σ) gerado para posicionar o chiquimato na estrutura da MtCQ. A figura
mostra também os Cα da MtCQ-mgADP-Ác.chiq., juntamente com as moléculas de ADP e Mg
2+
. A
figura foi gerada utilizando o programa XtalView (McRee, 1999).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.8
– Estrutura molecular do chiquimato evidenciando o grupo carboxila e os três grupos
hidroxila. Os átomos de carbono estão numerados em azul e os átomos de oxigênio em vermelho.
A estrutura química do chiquimato [3R-(3
α
,4
α
,5
β
)]3,4,5-trihidroxi-1-ciclohexeno-1-
ácido carboxílico] é mostrada na Figura 3.8. O grupo guanidina da Arg58 e Arg136, e o
grupo NH-cadeia principal da Gly81 interagem com o grupo carboxila do chiquimato. O
grupo 3-hidroxila do chiquimato forma ligações de hidrogênio com o grupo carboxila do
Asp34, grupo NH da Gly80 e uma molécula de água (Figura 3.9). Esta molécula de água
interage com os resíduos do domínio de ligação do chiquimato (Arg58 e Glu61), com os
resíduos do motivo Walker B (Gly79, Gly80 e Gly81) e com Ala82. O grupo 2-hidroxila
do chiquimato faz ligações de hidrogênio com a cadeia lateral do Asp34. As distancias
das ligações de hidrogênio direta ou mediadas por água entre o chiquimato e a MtCQ são
mostradas na Tabela 3.5.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
O resíduo Glu61 é conservado em ambas chiquimato quinases (tanto as codificadas
pelo gene aroK ou pelo gene aroL) e acreditava-se que este resíduo estava envolvido na
ligação do chiquimato (Krell et al., 1998; Gu et al., 2002; Romanowski & Burley, 2002).
Krell e co-autores propuseram que o Glu61 é favoravelmente posicionado para a ligação
com o grupo 5-hidroxila do chiquimato em ErcCQ. Entretanto, a posição do chiquimato
na estrutura da ErcCQ não foi claramente demonstrado devido a densidade eletrônica
insuficiente na região do sítio ativo. Na estrutura da MtCQ-MgADP-Chiquimato, a
cadeia lateral do Glu61 forma uma interação mediada por água com o grupo 3-hidroxila
do chiquimato. Além disso, o Glu61 forma ligações de hidrogênio e uma ponte salina
com a Arg58, auxiliando no posicionamento do grupo guanidina deste resíduo, que é
importante para ligação do substrato através de interações com o grupo carboxila.
Portanto, o resíduo Glu61 demonstra um papel importante no posicionamento da
molécula do chiquimato, embora não esteja diretamente envolvido na ligação. O Glu61 é
conservado em todas as CQs seqüenciadas até o momento, o que reforça o papel de
posicionar a Arg58 no sítio ativo. O Glu54 também apresenta o papel de posicionar a
Arg58 para interação do chiquimato, como sugerido na estrutura da MtCQ-MgADP (Gu
et al., 2002). Entretanto, o Glu54 não é conservado em todas as CQs, com exceção das
codificadas pelo gene aroK (Figura 3.5).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.9
– Interações envolvidas na ligação chiquimato no sítio ativo da MtCQ. Ligações de
hidrogênio entre chiquimato e grupos protéicos estão representadas por linhas tracejadas. A figura foi
gerada utilizando o programa MolMol (Koradi et al., 1996).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 3.5 – Ligações de hidrogênio direta ou mediada por água entre o chiquimato e MtCQ
Ácido
chiquímico
MtCQ Distâncias
(Å)
Ligações de hidrogênio
mediadas por água
Distâncias (Å)
Grupos hidroxila
O1 Gly80-N 3,10
Asp34-OD1 2,82
Asp34-OD2 2,57
Água320-O 2,46 Gly79-O 2,96
Gly80-N 3,13
Gly81-N 2,51
Ala82-N 3,54
Arg58-NH1 3,12
Glu61-OE2 2,99
O2 Asp34-OD1 2,66
Asp34-OD2 2,85
O3 n.o
Grupo carboxila
O4 Gly81-N 3,22
Arg136-NH2 2,68
O5 Arg58-NH2 2,67
Gly 81-N 3,49
Arg136-NH2 2,34
Todas as distâncias < 3,6 Å são mostradas.
n.o., Não observada
Ligação do íon cloreto
O íon cloreto demonstra um papel importante na determinação da afinidade da ErcCQ
pelo ATP e chiquimato (Cerasoli et al., 2003). Entretanto, nenhum íon cloreto foi
encontrado no sítio ativo em posições equivalentes a observada em MtCQ-MgADP-
Chiquimato quando superposta em ambas ErcCQ (Krell et al., 1998) e o mutante K15M
ErcCQ (Krell et al., 2001). Dois íons cloreto foi encontrado em MtCQ-MgADP-
Chiquimato em posições equivalentes aos observados em MtCQ-MgADP. Um dos íons
cloreto está localizado no sítio ativo da enzima fazendo ligações de com o grupo 3-
hidroxila do chiquimato, grupo NH da Gly80 e Água1 (Figura 3.10), enquanto que o
outro está localizado na superfície da proteína, distante do sítio ativo. O resíduo
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
conservado Arg117 do domínio LID está envolvido na ligação do ADP, promovendo o
deslocamento observado na região do domínio LID. A Lys15 forma ligações de
hidrogênio com um átomo de oxigênio do
β
-fosfato e com íon cloreto na estrutura
MtCQ-MgADP (Figura 3.10). O NH da cadeia principal da Gly80 faz ligação de
hidrogênio com íon cloreto em MtCQ-MgADP e MtCQ-MgADP-Chiquimato (Tabela
3.10). Estes resíduos (Lys15, Gly80 e Arg117) estão localizados próximos do local onde
ocorre a reação química e devem apresentar um papel crucial na estabilização do estado
de transição. Consistente com esta proposta, o mutante K15M da ErcCQ não possui
atividade enzimática (Krell et al., 2001). O resíduo Gly80 do motivo Walker-B está
envolvido através de uma ligação de hidrogênio com
γ
-fosfato of ATP (Krell et al.,
1998). A distância entre o íon cloreto e o grupo 3-hidroxila do chiquimato (local onde o
foforil é transferido) é de 3,36 Å em MtCQ-MgADP-Chiquimato. Os resíduos Lys15 e
Gly80 estão em posições significativamente diferente em MtCQ-MgADP-Chiquimato
comparado com MtCQ-MgADP.
A distância entre a Lys15 e o íon cloreto é maior em MtCQ-MgADP-Chiquimato
(Tabela 3.6), enquanto que a distância entre Cl
-
e NH da Gly80 (3,49 – 3,69 Å) em
MtCQ-MgADP é diminuída para 3,24 Å em MtCQ-MgADP-Chiquimato. Estas
mudanças indicam ser um movimento em conjunto dos resíduos Lys15, Gly80 e íon Cl
-
.
Este íon cloreto aparece deslocado em 1,76 Å e 1,52 Å na MtCQ-MgADP-Chiquimato
quando comparado com MtCQ-MgADP 1L4U e 1L4Y, respectivamente. A posição do
Cl
-
e as diferenças observadas em MtCQ-MgADP-Chiquimato sugerem que o íon cloreto
ocupa a posição do fosforil na reação para formação do chiquimato 3-fosfato.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.10
– Ligação do íon cloreto no sítio ativo das estruturas da MtCQ-MgADP-Chiquimato e
MtCQ-MgADP (1L4U; Gu et al., 2002).
O íon Cl
-
, Lys 15 e Gly80 da estrutura MtCQ-MgADP são
coloridos em verde. A figura foi gerada utilizando o programa MolMol (Koradi et al., 1996).
Tabela 3.6 – Ligação do íon cloreto nas estruturas da MtCQ-MgADP (1L4Y e 1L4U) e MtCQ-
MgADP-Chiquimato
Distâncias (Å)
Íon
cloreto
Átomos
1L4Y 1L4U
MtCQ-MgADP-
Chiquimato
Grupo 3-hidroxila do
chiquimato - O1
4,48
a
4,66
a
3,36
Gly80-N 3,49 3,69 3,24
Lys15-NZ 3,17 3,25 3,94
Cl180
Mg
2+
-Água1-O 3,87 3,75 2,84
a
As distâncias foram medidas coma estrutura da MtCQ-MgADP superpostas na estrutura da
MtCQ-MgADP-Chiquimato.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Mudanças conformacionais devido à ligação do chiquimato
Como citado anteriormente, MtCQ pertence a família das nucleosídeo monofosfato
(NMP) quinases, que são compostas por três domínios: CORE, LID, e domínio de
ligação NMP. As quinases apresentam um grande movimento durante a catálise para
defender o sítio ativo da água evitando a hidrólise do ATP (Jencks, 1975b).
A MtCQ-MgADP-Chiquimato representa a estrutura mais fechada da MtCQ, mas não
totalmente, desde que o total fechamento deve ser obtido com a ligação do ATP. O
mutante K15M ErcCQ (Krell et al., 2001) foi cristalizado em uma conformação aberta
que presumidamente deve ser equivalente a conformação de uma estrutura apo-enzima
(sem ADP e chiquimato) (Leipe et al., 2003). Este mutante foi produzido para avaliar o
papel da Lys15 do motivo Walker-A na catálise. Entretanto, uma mutação inesperada no
domínio LID (Pro115Leu) foi detectada durante o refinamento do modelo. Além disso, a
enzima foi cristalizada com duas moléculas na unidade assimétrica com contatos entre
os domínios LID, o que não é observado na estrutura cristalográfica nativa (Krell et al.,
1998; Krell et al., 2001). Apesar do duplo mutante K15M/P115L ErcCQ como modelo
para apo-enzima apresentarem estes “problemas”, consideramos como sendo a
conformação da estrutura apo, uma que vez que não tínhamos um melhor modelo
disponível no momento. A superposição da MtCQ-MgADP-Chiquimato e apo ErcCQ
(K15M+P115L) mostra a grande mudança conformacional nos domínios LID e SB
devido a ligação do ADP e chiquimato na MtCQ (Figura 3.11).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.11 –
Superposição da MtCQ-MgADP-Chiquimato (vermelho) e o mutante K15M/P115L da
ErcCQ (azul) (Krell et al., 2001) mostrando o grande deslocamento do domínio LID e do domínio SB
devido a ligação do ADP e do chiquimato. A figura foi gerada utilizando o programa MolMol (Koradi et
al., 1996).
Mudanças nos elementos de estrutura secundária da ErcCQ foram observadas através
dos espectros obtidos por dicroísmo circular da enzima livre (ErcCQ), do complexo
binário (ErcCQ-Chiquimato), e do complexo ternário (ErcCQ- Chiquimato-Adenil-
imidodifosfato) (Krell et al., 1998). A superposição dos C
α
da MtCQ-MgADP-
Chiquimato e MtCQ-MgADP demonstrou que os domínio LID e SB sofreram um
notável deslocamento em direção um ao outro na estrutura MtCQ-MgADP-Chiquimato
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
(Figura 3.12). A superposição incluindo todos os resíduos apresentou valores de RMSD
de 0,56 Å e 0,54 Å para 1L4U e 1L4Y, respectivamente. Quando esta análise é feita
somente para os resíduos dos domínios LID e SB são observados RMSD de 1,33 Å
(resíduos 112-124 que formam o LID) e 0,74 Å (resíduos 33-61 que formam o SB). O
local de ligação do chiquimato é composto pelos resíduos dos domínios LID e SB,
motivo Walker-B, e Arg136 da hélice-α7. Na estrutura MtCQ-MgADP-Chiquimato
ocorre deslocamento dos resíduos Val116, Pro118 e Leu119 do domínio LID, e Ile45,
Ala46, Glu54, Phe57 e Arg58 do domínio SB em direção ao chiquimato (Figura 3.12).
Figura 3.12
– Superposição dos C
α
do domínio LID e do domínio SB das estruturas MtCQ-MgADP-
Chiquimato (cinza) e MtCQ-MgADP (verde) (1L4U; Gu et al., 2002). A figura mostra deslocamento em
direção ao substrato dos resíduos Val116, Pro118 e Leu119 do domínio LID e Ile45, Ala46,
Glu54,Phe57 e Arg58 do domínio SB devido a ligação do chiquimato. A figura foi gerada utilizando o
programa MolMol (Koradi et al., 1996).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
As mudanças conformacionais descritas acima podem ser demonstradas pela redução
da superfície molecular da MtCQ-MgADP-Chiquimato quando comparada com MtCQ-
MgADP (Figura 3.13). O ADP, Mg
2+
, chiquimato, Cl
-
foram removidos para o cálculo.
Os valores calculados são aproximadamente 7246 Å
2
para MtCQ complexada com
MgADP (1L4U) e 6915 Å
2
para estrutura complexada com MgADP e chiquimato. Desta
forma, aproximadamente 330 Å
2
na superfície molecular é reduzido devido à ligação do
chiquimato. Uma vez que MtCQ-MgADP-Chiquimato e MtCQ-MgADP foram
cristalizadas no mesmo grupo espacial com valores similares para os parâmetros de cela
(Gu et al., 2001), as mudanças conformacionais observadas não estão associadas ao
arranjo cristalino.
Figura 3.13
– Superfície molecular das estruturas (A) MtCQ-MgADP (1L4U; Gu et al., 2002) e (B)
MtCQ-MgADP-Chiquimato. As moléculas de água são coloridas em vermelho. A figura foi gerada
utilizando o programa MolMol (Koradi et al., 1996).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.1.3 Conclusão
Os resíduos identificados na ligação do chiquimato diretamente ou indiretamente
(Asp34, Arg58, Glu61, Gly79, Gly80, Gly81 e Arg136) são conservados em todas as
CQs (codificadas pelos genes aroK e aroL) (Figura 3.5). Estruturas de CQs depositadas
no PDB foram superpostas e as posições dos resíduos envolvidos na ligação entre o
chiquimato e a CQ são altamente conservadas para as proteínas codificadas pelos genes
aroK (Chiquimato quinase de Mycobacterium tuberculosis e Campylobacter jejuni) e
aroL (Chiquimato quinase de Erwinia chrysanthemi). A conservação do sítio ativo pode
ser responsável pelos valores similares de Km encontrados para MtCQ (aroK) e EcSK
(aroL), entretanto esta informação não explica a diferença observada para valores Km da
CQ I de E.coli. Romanowski e co-autor (Romanowski & Burley, 2002) propuseram que
a substituição da Leu83 (ErcCQ) pela Lys86 em CQ I de E.coli atrapalha a ligação do
chiquimato. A perda de um resíduo hidrofóbico nesta posição provocaria uma
diminuição na afinidade pelo chiquimato pela CQ I de E.coli comparada à CQ II. No
entanto, superposição da CQ I de E.coli com MtCQ-MgADP-Chiquimato demonstrou
que a Lys86 localiza-se distante do sítio de ligação do chiquimato, não suportando assim
a proposição de Romanowski & Burley. Além disso, embora a ocorra uma substituição
da Leu83 (ErcCQ) por uma Thr84 (MtCQ), isto é, mudando um resíduo hidrofóbico por
um resíduo polar, os valores de Km para a MtCQ e ErcCQ são similares.
A superposição da CQ I de E.coli na estrutura da MtCQ-MgADP-Chiquimato
mostrou o resíduo Leu123 (E.coli) na posição ocupada pelo chiquimato, porém não
podemos afirmar que os sítios ativos são diferentes, uma vez que a estrutura da CQ I de
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
E.coli foi cristalizada na ausência de chiquimato, e a ligação deste composto pode
promover futuras mudanças de conformação incluindo o deslocamento da Leu123.
Neste trabalho foram descritos os resíduos envolvidos na ligação do chiquimato e as
mudanças conformacionais devido a ligação desta molécula. O fechamento completo do
sítio ativo provavelmente será alcançado através do complexo MtCQ com chiquimato,
Mg
2+
, e um análogo de ATP não hidrolisável 5'-(β,γ-metileno) trifosfato (AMP-PCP). A
disponibilidade da estrutura da chiquimato quinase de M. tuberculosis complexada com
chiquimato irá permitir o desenho molecular de inibidores específicos para CQ baseados
no sítio de ligação do substrato. Além disso, as informações aqui apresentadas
contribuíram para entendimento dos fatores responsáveis pelo fechamento do sítio ativo,
que poderão ser utilizadas o desenvolvimento de inibidores que forcem a estrutura em
uma conformação fechada que seja incapaz de catalisar a transferência do fosfato para o
chiquimato.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.2 Modelagem molecular da 5-enolpiruvilchiquimato-3-fosfato sintase (EPSP
sintase)
A EPSP sintase (EC 2.5.1.19) catalisa o sexto passo da via do ácido chiquímico, que
corresponde à transferência do enolpiruvil do fosfoenol piruvato (PEP) para o
chiquimato-3-fosfato (produto da catálise da chiquimato quinase), dando origem ao 5-
enolpiruvilchiquimato-3-fosfato e um fosfato inorgânico (Figura 3.14) (Bentley, 1990;
Levin & Sprinson, 1964).
Figura 3.14 –
Reação catalisada pela EPSPS sintase. Figura adaptada de Mathews & van Holde.
(Mathews & van Holde, 1990).
Como citado anteriormente, um importante composto guia para a construção de
novos inibidores de EPSPS é o glifosato. Deste modo, dois modelos para EPSPS de M.
tuberculosis (MtEPSPS) são propostos, sendo um na ausência de ligantes (modelo
aberto) e o outro na presença do chiquimato-3-fosfato e do glifosato (modelo fechado),
uma vez que a literatura reporta uma mudança conformacional entre estes dois estados
(Schönbrunn et al., 2001)
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.2.1 Materiais e Métodos
Para construção do modelo da MtEPSPS foi usada modelagem molecular por
homologia implementada no programa MODELLER (Šali & Blundell, 1993). Este
programa aborda modelagem comparativa satisfazendo restrições espaciais. O
procedimento para modelagem inicia-se com o alinhamento da seqüência de
aminoácidos a ser modelada (alvo), com uma ou várias seqüências primárias de
proteínas com estruturas tridimensionais conhecidas (templates), obedecendo ao critério
de maior identidade para a seleção. Este alinhamento é utilizado como informação de
entrada (input) para o programa. O arquivo de saída (output) é um modelo tri
dimensional da seqüência alvo contendo todos os átomos da cadeia principal e das
cadeias laterais, com exceção dos hidrogênios.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.15 – Representação esquematizando as etapas de construção de um modelo utilizando a
modelagem molecular comparativa.
As coordenadas atômicas da estrutura cristalográfica da EPSPS Escherichia coli
(EcEPSPS), que apresenta duas estruturas cristalográficas, uma no estado aberto (sem
ligante – PDB:1EPS) (Stallings et al., 1991) e a outra no estado fechado (complexada
com chiquimato-3-fosfato e glifosato – PDB:1G6S) (Schönbrunn et al., 2001) foram
utilizada como modelo inicial para modelagem da MtEPSPS apo-enzima e MtEPSPS
complexada com chiquimato-3-fosfato e glifosato. As coordenadas atômicas de todas as
águas foram removidas das estruturas da EcEPSPS. Um total de 1000 modelos foram
gerados para cada estado, e o modelo final foi selecionado baseando-se na sua qualidade
esterioquímica.
3.2.2 Resultados e discussão
Alinhamento das seqüências primárias e qualidade dos modelos
O alinhamento entre as seqüências primárias da MtEPSPS e da EcEPSPS (Figura
3.16) apresenta 31% de identidade, indicando que EcEPSPS é um template que pode ser
utilizado para modelagem. A figura 3.16 mostra também que os resíduos de aminoácidos
Lys-23, Arg-124, e Lys-410 da MtEPSPS (correspondente aos resíduos Lys-22, Arg-
124, e Lys-411 em EcEPSPS) são conservados, e estes estão localizados no sítio ativo da
enzima (Shuttleworth et al., 1999).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.16 – Alinhamento das seqüências de aminoácidos da MtEPSPS e da EcEPSPS. Em azul e preto
são mostrados resíduos que não são idênticos ou não foram alinhados, e em vermelho os resíduos
idênticos ou semelhantes entre as seqüências.
A qualidade esterioquímica do modelo da MtEPSPS aberto e fechado foram
analisados pelo diagrama
φ
-
ψ
. O modelo fechado apresentou 96% dos resíduos nas
regiões permitidas, e 4% nas regiões não permitidas. O modelo aberto (sem ligante),
apresentou 91,1% resíduos nas regiões permitidas, e 8,9% nas regiões não permitidas. A
quantidade de resíduos na região não permitida do diagrama para o modelo aberto é
devido ao template utilizado da EcEPSPS ter somente os carbonos
α
depositado no PDB
(código:1EPS), não fornecendo assim, informações suficientes para obtenção de um bom
modelo. Não foi utilizado outro template para a modelagem da MtEPSPS, pois a única
EPSPS depositada no PDB sem ligante é de E. coli.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Caracterização das estruturas aberta e fechada da MtEPSPS
A MtEPSPS é uma proteína da classe
α
/
β
e consiste de fitas
β
rodeadas de hélices
α
.
As figuras 3.17A e B mostram as estruturas secundárias de ambos os modelos de
MtEPSPS. O enovelamento das estruturas apresentou dois domínios similares, que
possuem uma conformação hemisférica, com raios de aproximadamente 21 Å.
Figura 3.17
– (A) MtEPSPS sem ligantes – estrutura aberta e (B) MtEPSPS complexada com
chiquimato-3-fosfato e glifosato – estrutura fechada.
Os dois domínios são ligados por dois segmentos da cadeia com ambos, amino e
carboxi terminais da proteína localizados no mesmo domínio. Os efeitos macrodipolares
observados nas hélices criam um potencial que facilita a ligação do chiquimato-3-fosfato
e do glifosato (Stallings et al., 1991). Os domínios são formados por três cópias de
βαβαββ - unidade de enovelamento. As quatro fitas-β de cada unidade de enovelamento
A
B
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
formam folhas-β paralela e antiparalela, e as duas hélices são paralelas a estas folhas. Na
estrutura fechada, as moléculas do chiquimato-3-fosfato e do glifosato estão ligadas na
região interdomínio, promovendo o fechamento da estrutura.
“Dobradiça” molecular
As estruturas cristalográficas de EcEPSPS e dos modelos moleculares de MtEPSPS
revelam a presença de dois domínios, designados A e B, que sofrem uma translação e
rotação da conformação aberta para a fechada. O centro de massa dos dois domínios é
deslocado 4,6 Å da estrutura aberta para a fechada em MtEPSPS, e ocorre uma diferença
angular na interface de 69° quando os dois domínios estão em conformações distintas
(Figura 3.18).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Figura 3.18 –
Sobreposição da MtEPSPS aberta (linha grossa) e fechada (linha fina) mostrando a
diferença conformacional entre as estruturas devido a ligação do chiquimato-3-fosfato e do glifosato.
Figura gerada utilizando o programa XtalView (McRee, 1999).
Ligação do chiquimato-3-fosfato e do glifosato
Os sítios de ligação para o glifosato e o chiquimato-3-fosfato ficam próximos e o
grupo 5-hidroxila do chiquimato-3-fosfato faz ligação de hidrogênio com o átomo de
nitrogênio do glifosato a uma distância de 2,82 Å. A figura 3.19 mostra o sítio de ligação
do chiquimato-3-fosfato e do glifosato da MtEPSPS.
Figura 3.19
– (A) Sítio de ligação do chiquimato-3-fosfato e do (B) glifosato da MtEPSPS. Figura
gerada utilizando o programa MolMol (Koradi et al., 1996).
A B
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
A análise do complexo da MtEPSPS-Chiquimato-3-fosfato-Glifosato indicou um total
de 15 e 22 ligações de hidrogênio da enzima com o chiquimato-3-fosfato e o glifosato,
respectivamente (Tabelas 3.7 e 3.8).
Tabela 3.7 – Ligações de hidrogênios entre MtEPSP e chiquimato-3-fosfato
Ligações de hidrogênio entre MtEPSP e o substrato Distâncias (Å)
Chiq-3-Fosfato MtEPSP
O3 Lys23 NZ 2,81
O5 Ser24 OG 2,70
O5 Arg28
N
H1 2,86
O4 Arg28
N
H2 2,78
O5 Arg28
N
H2 3,60
O6 Ser167 OG 3,40
O8 Ser167 OG 2,70
O6 Ser168 OG 2,69
O6 Ser168
N
2,84
O1 Gln169
N
E2 3,49
O6 Ser196 OG 3,59
O7 Ser196 OG 2,69
O2 Glu311 OE2 2,65
O2 His340
N
E2 3,14
O1 His340
N
E2 3,57
Todas as distâncias < 3,6 Å são mostradas.
Tabela 3.8 – Ligações de hidrogênio entre MtEPSP e glifosato.
Ligações de hidrogênio entre MtEPSP e o inibidor Distâncias (Å)
Glifosato EPSP
O4 Lys23
N
Z 2,98
O1 Lys23
N
Z 2,84
N1 Lys23
N
Z 3,46
O3 Leu94 O 3,10
O2 Gly96
N
3,53
O3 Gly96
N
2,87
O3 Arg124
N
H1 2,87
O2 Arg124
N
H2 2,81
O3 Arg124
N
H2 3,60
O2 Gln169
N
E1 3,58
O2 Gln169
N
E2 2,88
O5 Glu311 OE1 2,96
N1 Glu341 OE1 3,35
N1 Glu341 OE2 2,88
O4 Glu341 OE2 3,59
O5 Arg344
N
H1 2,81
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
O5 Arg344
N
H2 3,02
O4 His384
N
E2 3,49
O4 Arg385
N
E 3,59
O5 Arg385
N
E 2,75
O4 Arg385
N
H2 3,12
O3 Lys410
N
Z 2,95
Todas as distâncias < 3,6 Å são mostradas.
A maioria das ligações de hidrogênio entre glifosato e a MtEPSPS são formadas
pelos resíduos Lys23, Arg124, Glu341, e Arg385, os mesmos resíduos identificados na
ligação com a EcEPSPS. Além disso, o alinhamento de 69 seqüências primárias de
EPSPS indicou que estes resíduos são conservados em todas as seqüências. Esta
observação sugere que o glifosato deve ser capaz de inibir a maioria ou quase todas
EPSPS, uma vez que a afinidade e especificidade de um inibidor estão relacionadas com
as ligações de hidrogênio, interações iônicas, assim como complementariedade de cargas
e superfície de contato entre enzima-inibidor (de Azevedo et al., 1996; Kim et al., 1996).
Dentre as sete enzimas presentes na via metabólica do ácido chiquímico, acredita-se
que a EPSPS seja um dos principais alvos para inibir o M. tuberculosis, uma vez que a
inibição desta proteína provocou a morte do organismo alvo. Portanto, o conhecimento
das estruturas (na presença e na ausência de ligantes) da MtEPSPS e do modo de
interação com chiquimato-3-fosfato (proteína-substrato) e com glifosato (proteína-
inibidor) se tornam imprescindíveis para desenvolvimento de possíveis inibidores.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.3 Modelagem molecular e análise de CD da corismato sintase
A corismato sintase (EC 4.2.3.5) catalisa a última reação desta via, a conversão do 5-
enolpiruvil-3-chiquimato fosfato (EPSP) para corismato (Bornemann et al., 2002;
Macheroux et al., 1998). A enzima utiliza flavina mononucleotídeo (FMN) reduzida
como cofator, embora a reação catalisada não envolve uma mudança no estado de oxi-
redução. O papel da FMN reduzida na catálise ainda não está totalmente elucidado.
Entretanto, recentes detalhes cinéticos contribuíram para avançar no entendimento do
mecanismo de ação da corismato sintase (Macheroux et al., 1999).
Figura 3.20
– Reação catalisada pela corismato sintase. Figura modificada a partir de Mathews & van
Holde. (Mathews & van Holde, 1990).
O produto da corismato sintase é o precursor para cinco distintas vias metabólicas. O
corismato é necessário para a produção de aminoácidos aromáticos, ácido para-
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
aminobenzóico (PABA), folato, e outros metabólitos cíclicos como ubiquinonas e
menaquinonas (Ratledge, 1982). A via do folato também é ausente em mamíferos, e suas
enzimas estão sendo exploradas com sucesso como alvos antibacterial, como por
exemplo, a inibição da disidrofolato redutase (DHFR) pelo trimetoprim.
A modelagem molecular da corismato sintase de M. tuberculosis (MtCS) complexada
com FMN e EPSP foi realizada, assim como análise de dicroísmo circular da proteína na
presença e ausência de seu cofator.
3.3.1 Materiais e Métodos
Modelagem Molecular
Para construção do modelo da MtCS foi utilizada a mesma abordagem do modelo
MtEPSPS, como descrito anteriormente. As coordenadas atômicas da estrutura
cristalográfica da corismato sintase de Streptococcus pneumoniae (SpCS) (código acesso
PDB: 1QX0) (Maclean & Ali, 2003) resolvida a 2,0 Å resolução foi utilizada como
modelo inicial para modelagem da MtCS complexada com FMN e EPSP. As
coordenadas atômicas de todas as águas foram removidas da estrutura da SpCS. Um
total de 1000 modelos foram gerados para MtCS, e o modelo final foi selecionado
baseando-se na sua qualidade estereoquímica.
Análise do modelo
A qualidade estereoquímica do modelo final do complexo MtCS-FMN-EPSP foi
analisada utilizando o programa PROCHECK (Laskowski, 1994). Os modelos atômicos
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
foram superpostos usando o programa LSQKAB (Colaborative Computational Project nº
4, 1994). As ligações de hidrogênio e as pontes salinas foram consideradas até 3,6 Å. As
diferenças do desvio da raiz média quadrática (r.m.s.d.) da geometria ideal para
comprimento de ligação e ângulos de ligação foram calculadas usando o programa X-
PLOR (Brünger, 1992). O G-fator é essencialmente uma análise obtida através de
parâmetros estereoquímicos. Esta análise é calculada seguindo as propriedades: ângulos
de torção (analisa a distribuição dos ângulos
φ
-
ψ
, χ1– χ2, χ-1, χ-3, χ-4, e ω para cada um
dos 20 tipos de aminoácidos) e geometria covalente (comprimentos e ângulos de ligação
para cadeia principal). A compatibilidade do modelo tri dimensional da proteína com a
sua seqüência primária foi analisada utilizando o programa VERIFY-3D (Bowie et al.,
1991; Luthy
et al.
, 1992).
Dicroísmo Circular
A clonagem, expressão e purificação da MtCS foi realizada como descrito em Dias et
al. (Dias et al., 2004). Espectros de dicroísmo circular foram obtidos entre os
comprimentos de ondas de 195-260 nm, usando um espectropolarímetro Jasco-710
(Jasco, Tókio-Japão) acoplado a um banho térmico Neslab RTE111. Os espectros foram
obtidos a 25º C usando celas com caminho óptico de 0,2 cm. Os dados foram adquiridos
com uma velocidade de leitura de 20 nm/min, largura de banda de 1,0 nm, 1,0s como
tempo de resposta e 0,1 nm de resolução. Um total de cinco acumulações foram
realizadas para obtenção de um espectro médio final. A deconvolução dos dados foi feita
utilizando o programa DICROPROT (Deléage & Geourjon, 1993).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
3.3.2 Resultados e discussão
Qualidade do modelo
O alinhamento das seqüências primárias da MtCS (alvo) e SpCS (template) apresenta
44% de identidade (Figura 3.21). Este grau de identidade entre as seqüências demonstra
que a estrutura cristalográfica da SpCS representa um bom “molde” para ser usado na
modelagem da MtCS.
Figura 3.21
– O alinhamento das sequências da MtCS e SpCS. Os elementos de estrutura secundária
para o modelo MtCS são indicados de acordo com a região onde são encontrados. O alinhamento foi
realizado com o programa CLUSTAL W (Thompson et al., 1994).
O diagrama phi (φ) e psi (ψ) (Diagrama de Ramachandran) foi utilizado para
comparar a qualidade estereoquímica do modelo MtCS e da estrutura da SpCS resolvida
por cristalografia. Análise do modelo da MtCS apresentou 94,6 % dos resíduos nas
regiões mais favoráveis e 5,4% dos resíduos nas regiões permitidas. A mesma análise
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
para a estrutura cristalográfica da SpCS foi realizada e verificou-se 93,1% dos resíduos
nas regiões mais favoráveis e 6,9% dos resíduos nas regiões permitidas. A excelente
qualidade do template permitiu que o modelo gerado também possuísse uma excelente
qualidade esterioquímica. Os valores de r.m.s.d. para comprimento e ângulo de ligação
com relação à geometria ideal, e valores correspondentes a compatibilidade da seqüência
primária do modelo da MtCS com sua estrutura tri dimensional são mostrados na Tabela
3.9.
Tabela 3.9 –
Resultados das análises estruturais da MtCS e SpCS.
Análise do modelo da MtCS e da estrutura da SpCS
Proteínas Verify-3D
a
G-factor
b
rmsd da geometria ideal
Escore
Total
Escore
Ideal
Escore
S
Ideal
Ângulos
Torsão
Geometria
Covalente
Global Comprimentos
de ligação (Ǻ)
Ângulos
de ligação (˚)
MtCS 157,49 179,29 0,88 S -0,07 -0,34 -0,16 0,022 3,172
SpCS 163,85 171,92 0,95 S 0,00 -0,16 -0,03 0,027 2,558
a
Escore Total: é a soma dos escores 3D-1D (preferência estatística) de cada resíduo presente na proteína.
Escore Ideal: S
ideal
= exp(-0,83+1,008xln(L)); onde L é o número de aminoácidos. Escore S
ideal
: é a
compatibilidade da seqüência com sua estrutura 3D, e é obtido através do Escore Total / Escore Ideal.
Escore S
ideal
deve ser acima de 0,45S
ideal
.
b
Idealmente, escores devem ser acima de –0,5. Valores abaixo de –1,0, necessitam ser investigados.
A estrutura da MtCS é pertencente a família α/β. A topologia estrutural dominante
das CS é um sanduíche beta-alfa-beta, onde cada monômero de CS é formado por
hélices na região central, na ordem α1, α6, α12 e α9, entre duas folhas betas compostas
por quatro fitas antiparalelas, na ordem β1, β2, β6 e β3 para folha “um” e na ordem β7,
β
8,
β
14 e
β
9 para folha “dois” (Figura 3.21A). A topologia central da MtCS apresenta
um pseudo eixo de ordem 2, embora esta simetria não seja observada para toda a
molécula. O sítio ativo da MtCS está localizado entre as folhas 1 e 2 e este é formado
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
pelas extremidades das hélices α centrais, e também pelos loops L1, L4, L10, L16, L25 e
L27.
Análise dos alinhamentos das seqüências da corismato sintase de Streptococcus
pneumoniae, Helicobacter pylori, Aquifex aeolicus, e Sacchoromyces cerevisiae indicam
que os resíduos envolvidos para a estabilização do tetrâmero são conservados na
seqüência do Mycobacterium tuberculosis. Baseados nestas evidências, foi proposto que
a estrutura quaternária da MtCS se apresente como um tetrâmero, como é mostrado na
Figura
3.21B
.
Figura 3.21 –
Diagrama de Ribbon do modelo molecular da MtCS complexado com FMN e EPSP. A)
Forma monomérica e B) Forma oligomérica (tetrâmero). A figura foi gerada utilizando o programa
MolMol (Koradi et al., 1996).
Sítio de ligação da FMN
O potencial eletrostático da superfície molecular da MtCS obtido pelo programa
GRASP (Nicholls et al., 1991), demonstrou uma grande concentração de resíduos
A
B
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
positivos ao redor do sítio de ligação da FMN. Análise da superfície molecular do
modelo da MtCS e da estrutura da SpCS indicaram que a FMN possui uma área de
contato de 261 Å
2
e 236 Å
2
, respectivamente.
A molécula de FMN apresenta poucas interações polares com a proteína, sendo a
maioria contatos entre os grupos hidroxila e os oxigênios do fosfato com os resíduos
presentes nos loops L10, L16, e L25. A figura 3.23A mostra as ligações de hidrogênio
entre a molécula de FMN e MtCS. Foram observadas 6 ligações de hidrogênio,
envolvendo os resíduos Ala257, Lys315, Thr319 e Ser342. Estes resíduos são altamente
conservados nas seqüências das corismato sintases, indicando que inibidores
competitivos com a FMN poderão inibir grande parte destas enzimas.
Figura 3.23A – Sítio de ligação da FMN. A
molécula de FMN apresenta ligações de
hidrogênio entre os oxigênios do fosfato e os
resíduos Ala257 e Lys315. A região do sistema de
anéis isoaloxazine da FMN também possui
ligações de hidrogênio com a proteína,
envolvendo os resíduos Thr319 e Ser342. Figura
gerada utilizando o programa MolMol (Koradi et
al., 1996).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Sítio de ligação do EPSP
A molécula do EPSP faz muitas interações polares com a MtCS. O sítio de ligação
do EPSP é extremamente básico, com seis resíduos de arginina (R40, R41, R46, R49,
R112, R139) e três histidinas (H11, H115, H339). Análise das ligações de hidrogênio
entre a enzima e o EPSP indicaram um total de 14 interações. A molécula do EPSP pode
ser dividida em três regiões: carboxila, enol-piruvil e do grupo fosfato. A região do enol-
piruvil interage com três resíduos de argininas (R40, R46 e R139). O grupo fosfato faz
ligações com a H11, R49, E81 e R341. O grupo carboxila forma ligações de hidrogênio
com a H115 e A138 (Figura 3.23B).
Estudos espectroscópicos em corismato sintase de Escherichia coli demostraram que
molécula do EPSP pode ser responsável por induzir mudanças conformacionais na
estrutura da MtCS, diferentemente da FMN que pode ser acomodada no sítio ativo sem
mudanças estruturais (Macheroux et al., 1996a; Macheroux et al., 1996b).
Figura 3.23B
– Sítio de ligação do EPSP. A
molécula de e H115 também faz ligação de
hidrogênio com EPSP apresenta ligações de
hidrogênio com as argininas R40, R46, R49, R139
e R341. Além das arginas, as histidinas H11 o
EPSP, caracterizando o sítio como um local
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
extremamente básico. Figura gerada utilizando o programa MolMol (Koradi et al., 1996).
Dicroísmo Circular
Os espectros de CD da MtCS na presença e na ausência de FMN são mostrados na
figura 3.24. As porcentagens obtidas dos elementos de estruturas secundárias para as
duas medidas independentes e do modelo da MtCS pode ser observado na tabela 3.10.
Figura 3.24 –
Sobreposição dos espectros de dicroísmo circular da corismato sintase de
Mycobacterium tuberculosis sem ligante e na presença de ligante (
_ _ _
) Enzima sem ligante; (
........
)
enzima na presença de 22,5
μ
M flavina mononucleotídeo (FMN). Solução de proteína utilizada nos
experimentos estava a uma concentração de 4,5
μ
M em tampão Fosfato de Sódio 5 mM (pH 7,8). Para
o espectro da enzima com FMN, uma alíquota da solução do ligante a 10mM em Fosfato de Sódio 5
mM (pH 7,8) foi adicionado a solução de proteína, e no espectro resultante foi corrigido a diluição. O
espectro de CD para enzima complexada com FMN também foi corrigido a contribuição do ligante.
200 220 240 260
-10000
-5000
0
5000
10000
15000
[Θ] deg.cm
2
.dmol
-1
Comprimento de onda (nm)
MtCS apo-enzima
MtCS + FMN
200 220 240 260
-10000
-5000
0
5000
10000
15000
[Θ] deg.cm
2
.dmol
-1
Comprimento de onda (nm)
MtCS apo-enzima
MtCS + FMN
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Tabela 3.10 – Porcentagem dos elementos de estrutura secundária calculado para os dados
experimentais de CD e para o modelo molecular da MtCS.
Modelo Molecular da MtCS
a
MtCS com FMN
b
MtCS sem FMN
b
Folhas
β
18 18 16
Hélices
α
40 44 45
Estr. aleatória 42 38 39
a
Elementos de estr. secundária foram calculados usando o programa MOLMOL (Koradi et al., 1996).
b
Elementos de estr. secundária foram calculados usando o programa de análise de CD - DICROPROT
(Deléage & Geourjon, 1993).
A porcentagem das estruturas secundárias obtidas através dos dados de CD da MtCS
apresentam uma grande concordância com os resultados obtidos através da modelagem
molecular, reforçando deste modo o modelo da MtCS. A presença de FMN oxidada
causa pequenos efeitos no espectro do CD, como observado na figura 3.24. Estes
resultados estão de acordo com os obtidos em espectros de CD para CS de E.coli
(Macheroux et al., 1998).
As moléculas de EPSP e de FMN fornecem interessantes alvos para o
desenvolvimento de drogas baseado em estrutura, não somente contra tuberculose, mas
também para outros patógenos. Inibidores podem ser competitivos com substrato e co-
fatores, levando a inativação da enzima. A molécula de 5-deazaflavina (Lauhon &
Bartlett, 1994), um análogo de FMN onde o nitrogênio da posição 5 é substituído por um
carbono, e (6R)-6-fluoro-EPSP, onde o hidrogênio-6R do EPSP é substituído por um
flúor, resulta na ausência de atividade da corismato sintase (Macheroux, 1998;
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Bornemann et al., 2003). Estes são exemplos de moléculas que foram racionalmente
modificadas e possuem a capacidade de inibição da enzima.
4.Referências Bibliográficas
Baptista, I. M., Oelemann, M. C, Opromolla, D. V., Suffys, P. N. (2002). 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,
97,
1147-1152.
Bennett, W.S., Steitz, T.A. (1980). Structure of a complex between yeast hexokinase A
and glucose. II. Detailed comparisons of conformation and active site configuration with
the native hexokinase B monomer and dimer. J. Mol.Biol. 140, 211-230.
Bentley, R. (1990). The shikimate pathway - A metabolic tree with many branches. Crit.
Rev. Biochem. Mol.Biol.
25
, 307-384.
Bloom, B.R., and Murray, C.J.L. (1992). Tuberculosis: Commentary on a reemergent
killer. Science, 257, 1055-1064.
Bornemann, S., Lawson, D.M., Thorneley, R.N. (2003). A branch point in chorismate
synthase research. Structure. 11, 1463-1465.
Bowie, J.U., Luthy, R., Eisenberg, D. (1991). A method to identify protein sequences
that fold into a known three-dimensional structure. Science. 253, 164–170.
Brünger, A. T. (1992). X-PLOR Version 3.1: A system for crystallography and NMR.
Yale University Press, New Haven.
Butler, J.R., Alworth, W.L., Nugent, M.J. (1974)
. J. Am. Chem. Soc.
96
, 1617–1618.
Cerasoli, E., Kelly, S. M., Coggins, J. R., Lapthorn, A. J., Clarke, D. T., Price, N. C.
(2003). Effects of salts on the function and conformational stability of shikimate kinase.
Biochim. Biophys. Acta 1648, 43-54.
Clemmens, D.L. (1996). Characterization of the Mycobacterium tuberculosis
phagossome. Trends Microbiol.,
4
, 13-118.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Coggins, J.R. (1989). The shikimate pathway as a target for herbicides. In Herbicides
and Plant Metabolism (Dodge, A., ed.), Cambridge University Press, Cambridge, UK,
pp. 97-112.
Coggins, J.R., Duncan, K., Anton, I.A., Boocock, M.R., Chaudhuri, S., Lambert,
J.M., Lewendon, A., Millar, G., Mousdale, D.M., Smith, D.D., (1987). The anatomy
of a multifunctional enzyme. Biochem. Soc. Trans. 15, 754–759.
Cole, S.T.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher, C.; Harris, D.; Gordon,
S.V.; Eiglmeier, K.; Gas, S.; Barry III, C.E.; Tekaia, F.; Badcock, K.; Basham, D.;
Brown, D.; Chillingworth, T.; Connor, R.; Davies, R.; Devlin, K.; Feltwell, T.;
Gentles, S.; Hamlin, N.; Holroyd, S.; Hornsby, T.; Jagels, K.; Krogh, A.; McLean,
J.; Moule, S.; Murphy, L.; Oliver, K.; Osborne, J.; Quail, M.A.; Rajandream, M.-
A.; Rogers, J.; Rutter, S.; Seeger, K.; Skelton, J.; Squares, R.; Squares, S.; Sulston,
J.E.; Taylor, K.; Whitehead, S., Barrell, B.G. (1998).
Deciphering the biology of
Mycobacterium tuberculosis from the complete genome sequence. Nature, 393, 537-544.
Collaborative computational project, No. 4, The CCP4 suite: programs for protein
crystallography. (1994) Acta Crystallog. sect. D. 50, 760–763.
Corbett, E.L., Watt, C.J., Walker, N., Mahier, D., Williams, B.G., Raviglione, M.C.,
Dye, C. (2003). The growing burden of tuberculosis. Arch. Intern. Med., 163, 1009-
1021.
Corpet, F. (1988). Multiple sequence alignment with hierarchical clustering. Nucl. Acids
Res., 16, 10881-10890.
Davies, G.M., Barret-Bee, K.J., Jude, D.A., Lehan, M., Nichols, W.W., Pinder, P.E.
(1994). (6S)-6-fluoroshikimic acid, an antibacterial agent acting on the aromatic
biosynthetic pathway. Antimicrob. Agents Chemother., 38, 403-406.
Davis, B.D., Mingioli, E. S. (1953). Aromatic biosynthesis. VII. Accumulation of two
derivatives of shikimic acid by bacterial mutants. J. Bacteriol. 66, 129–136.
de Azevedo, W.F. Jr., Mueller-Dieckmann, H.J., Schulze-Gahmen, U., Worland,
P.J., Sausville, E., 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. USA.
93(7):2735-2740.
De Feyter, R. C., Pittard, J. (1986). Purification and properties of shikimate kinase II
from Escherichia coli K-12. J. Bacteriol. 165, 331-333.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
De Feyter, R.C., Davidson, B.E., Pittard, J. (1986). Nucleotide sequences of the
transcription unit containing the aroL and aroM genes from Escherichia coli K-12. J.
Bacteriol. 165: 233–239.
Deléage, G., Geourjon, C. (1993). An interactive graphic program for calculating the
secondary structures content of proteins from circular dichroism spectrum. Comp. Appl.
Biosc. 9, 197-199.
Dias, M. V. B., Ely, F., Canduri, F., Pereira, J. H., Frazzon, J., Basso, L. A., Palma,
M. S., De Azevedo Jr., W. F., Santos, D. S. (2004). Crystallization and preliminary X-
ray crystallographic analysis of chorismate synthase from Mycobacterium tuberculosis.
Acta Crystal. Section D. 60, 2003-2005.
Dreusicke, D., Karplus, A., Schulz, G. E. (1988). Refined structure of porcine
cytosolic adenylate kinase at 2.1 Å resolution. J. Mol. Biol.
199
, 359-371.
Drobniewski, F. A., Caws, M., Gibson, A., Young, D.(2003). Modern laboratory
diagnosis of tuberculosis. Lancet Infect. Dis., v. 3, p.141-147.
Duncan, K. (2003). Progress in TB drug development and what is still needed.
Tuberculosis, v. 83, p. 201-207.
Duncan, K., Edwards, R.M., Coggins, J.R. (1987). The pentafunctional arom enzyme
of Saccharomyces cerevisiae is a mosaic of monofunctional domains. Biochem. J. 246,
375–386.
El Sayed, K. A., Bartyzel, P., Shen, X. Y., Perry, T. L., Kjawiony, J. K., Hamann,
M. T. (2000)
. Marine natural products as antituberculosis agents. Tetrahedron,
56
, 949-
953.
Ely, B., Pittard, J. (1979). Aromatic amino acid biosynthesis: Regulation of shikimate
kinase in Escherichia coli K-12. J. Bacteriol. 138: 933–943.
Funasa, 2002. Situação da Prevenção e Controle das Doenças Transmissíveis no Brasil.
Setembro/2002 [http://www.funasa.gov.Br/ guia_epi/]
Gerstein, M., Schulz, G. E., Chothia, C. (1993). Domain closure in adenylate kinase:
Joints on either side of two helices close like neighbouring fingers. J. Mol. Biol. 229,
494-501.
Gu, Y., Reshetnikova, L., Li, Y., Wu, Y., Yan, H., Singh, S., Ji, X. (2002). Crystal
Structure of Shikimate Kinase from
Mycobacterium tuberculosis
Reveals the Dynamic
Role of the LID Domain in Catalysis. J. Mol. Biol. 319, 779-789.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Gu, Y., Reshetnikova, L., Li, Y., Yan, H., Singh, S. V., Ji, X.(2001). Crystallization
and preliminary X-ray diffraction analysis of shikimate kinase fom Mycobacterium
tuberculosis in complex with MgADP. Acta crystallographica Section D, D57, 1870-
1871.
Guex, N., Peitsch, M.C. (1997). SWISS-MODEL and the Swiss-PdbViewer: an
environment for comparative protein modeling. Electrophoresis.
18
, 2714-2723.
Haslam, E. (1993). Shikimic acid: metabolism and metabolites. John Wiley & Sons,
Chichester, United Kingdom.
Hawgood, B.J. (1999). Doctor Albert Calmette 1863-1933: founder of antivenomous
serotherapy and of antituberculous BCG vaccination. Toxicon, 37, 1241-1258.
Hiriyanna, K.T., Ramakrishnan, T. (1986). Deoxiribonucleic-acid replication time in
Mycobacterium tuberculosis. Arch. Microbiol., 144, 105-109.
Jencks, W. P. (1975a). In Catalysis in Chemistry and Enzymology. Dover Publications,
Inc., New York, USA.
Jencks, W. P. (1975b). Binding energy, specificity and enzymatic catalysis: the circe
effect. Adv. Enzymol. 43, 219-410.
Kim, S.-H., Schulze-Gahmen, U., Brandsen, J., de Azevedo, W. F. Jr. (1996).
Structural basis for chemical inhibitor of CDK2. Prog. in Cell Cycle Res. 2, 137-145.
Kishore, G.M., Shah, D.M. (1988)
. Amino acid biosynthesis inhibitors as herbicides.
Annu. Rev. Biochem. 57, 627–663.
Koradi, R., Billeter, M., Wüthrich, K. (1996). MOLMOL: A program for display and
analysis of macromolecular structures. J. Molecular Graphics,
14(1)
, 51-55.
Krell, T., Coggins, J. R., Lapthorn, A. J. (1998). The three-dimensional structure of
shikimate kinase. J. Mol. Biol. 278, 983–997.
Krell, T., Maclean, J., Boam, D. J., Cooper, A., Resmini, M., Brocklehurst, K.
(2001). Biochemical and X-ray crystallographic studies on shikimate kinase: the
important structural role of the P-loop lysine. Protein Sci. 10, 1137–1149.
Laskowski, R. A., MacArthur, M. W., Smith, D. K., Jones, D. T., Hutchinson, E. G.,
Morris, A, L., Naylor, D., Moss, D. S., Thorton, J. M. (1994). PROCHECK v.3.0 -
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Program to check the stereochemistry quality of Protein structures - Operating
instructions.
Lauhon, C.T., Bartlett, P.A. (1994). Substrate analogs as mechanistic probes for the
bifunctional chorismate synthase from Neurospora crassa. Biochemistry. 33, 14100-
14108.
Leipe, D. D., Koonin, E. V., Aravind, L. (2003)
. Evolution and classification of P-loop
kinases and related proteins. J. Mol. Biol. 333, 781-815.
Lennon, B.W., Williams Jr., C.H., Ludwig, M.L. (1999). Crystal structure of reduced
thioredoxin reductase from Escherichia coli: structural flexibility in the isoalloxazine
ring of the flavin adenine dinucleotide cofactor. Protein Sci.. 8, 2366–2379.
Leslei, A.G.W. (1992)
. Jnt CCP4/ESF-EACBM Newsl. Protein Crystallogr.
26
.
Levin, J.G., Sprinson, D.B. (1964). The enzymatic formation and isolation os 3-
EnolPyruylshikimate 5-Phosphate. J.Biol.Chem., 239, 1142-1150.
Luthy, R., Bowie, J., Eisenberg, D. (1992). Assessment of protein models with three-
dimensional profiles. Nature, 356, 83–85.
Maclean, J., Ali, S. (2003). The structure of chorismate synthase reveals a novel flavin
binding site fundamental to a unique chemical reaction. Structure. 11, 1499-1511.
Macheroux, P., Bornemann, S., Ghisla, S., Thorneley, R. N. F. (1996b). 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, 25850-25858.
Macheroux, P., Petersen, J., Bornemann, S., Lowe, D., Thorneley, R. N. F. (1996a)
.
Binding of the oxidized, reduced, and radical flavin species to chorismate synthase. An
investigation by spectrophotometry, fluorimetry, and electron paramagnetic resonance
and electron nuclear double resonance spectroscopy. Biochemistry. 35(5), 1643–1652.
Macheroux, P., Schmid, J., Amrhein, N., Schaller, A. (1999). A unique reaction in a
common pathway: mechanism and function of chorismate synthase in the shikimate
pathway. Planta. 207, 325–334.
Macheroux, P., Schönbrunn, E., Svergun, D.I., Volkov, V.V., Koch, M.H.J.,
Bornemann, S., Thorneley, R.N.F. (1998). Evidence for a major structural change in
Escherichia coli chorismate synthase induced by flavin and substrate binding. Biochem.
J. 335, 319–327.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Manabe, Y.C., Bishai, W.R. (2000). Latent Mycobacterium tuberculosis – persistence,
patience, and winning by waiting. Nat. Med., 6, 1327-1329.
Mathews, C.K., van Holde, K.E. (1990). Biochemistry, Benjamin Cummings, 1130p.,
California, USA.
McRee, D. E. (1999).
XtalView/Xfit-A versatile program for manipulating atomic
coordinates and electron density. J. Struct. Biol., 125, 156-165.
Millar, G., Lewendon, A., Hunter, M. G., Coggins, J. R. (1986). The cloning and
expression of the aroL gene from Escherichia coli K-12. Biochem. J. 237, 427-437.
Mitchison, D.A. (1985). The action of antitubercular drugs in short-course
chemotherapy. Tubercle,
66
, 219-225.
Mousdale, D.M., Campbell, M.S., Coggins, J.R. (1987). Purification and
characterization of bifunctional dehydroquinase-shikimate: NADP oxidoreductase from
pea seedlings. Phytochemistry, 26, 2665–2670.
Müller, C. W., Schlauderer, G. J., Reinstein, J., Schulz, G. E. (1996). Adenylate
kinase motions during catalysis: an energetic counterweight balancing substrate binding.
Structure, 4, 147-156.
Müller-Dieckmann, H.-J., Schulz, G. E. (1994). Thestructure of uridylate kinase with
its substrates, showing the transition state geometry. J. Mol. Biol., 236, 361-367.
Navaza, J. (1994)
. AMoRe: an automated package for molecular replacement. Acta
Crystallog. sect. A., 50, 157-163.
Nicholls, A., Sharp, K.A., Honig, B. (1991). Protein folding and association: insights
from the interfacial and thermodynamic properties of hydrocarbons. Proteins, 11(4),
281–296.
Oliveira, J.S., Pinto, C.A., Basso, L.A., Santos, D.S. (2001)
. Cloning and
overexpression in soluble form of functional shikimate kinase and 5-
enolpyruvylshikimate 3-phosphate synthase enzymes from Mycobacterium tuberculosis.
Protein Expression and Purification, 22, 430-435.
Parish, T., Stoker, N. G. (2002). The common aromatic amino acid biosynthesis
pathway is essential in Mycobacterium tuberculosis. Microbiology, 148, 3069-3077.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Pelczar M. J. Jr., Chan, E.C.S., Krieg, N. R. (1996). Microbiologia – Conceitos e
Aplicações,Makron Books, vol. 2, 2ed., 517p., São Paulo.
Pittard, A.J. (1987) In F.C. Neidhardt, J.L. Ingraham, K.B. Low, B. Magasanik, M.
Schaechter, and H.E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium:
cellular and molecular biology. vol.1, pp. 368–394. American Society for Microbiology,
Washington, D.C.
Pivetta, M. (2004). Pesquisa Fapesp 97, 32-37.
Polikarpov, I., Oliva, G., Castellano, E.E., Garratt, R., Arruda, P., Leite, A.,
Craievich, A. (1998a). The protein crystallography beamline at LNLS, the Brazilian
National Synchrotron Light Source. Nuclear Instruments & Methods In Physics
Research A, 405, 159-164.
Polikarpov, I., Perles, L. A., de Oliveira, R. T., Oliva, G., Castellano, E.E., Garratt, R.,
Craievich, A. (1998b). Set-up and experimental parameters of the protein
crystallography beamline at the Brazilian National Synchrotron Laboratory. J.
Synchrotron Rad, 5, 72-76.
Ratledge, C. (1982). Nutrition, growth and metabolism, in: Ratledge, C., Stanford, J. L.
(Eds.), The biology of the mycobacteria, Academic Press, London, Vol 1, 185-271.
Roberts, F., Roberts, C.W., Johnson, J.J., Kyle, D.E., Krell, T., Coggins, J.R.,
Coombs, G.H., Milhous, W.K., Tzipori, S., Fergunson, D.J.P., Chakrebarti, D.,
McLeod, R. (1998). Evidence for the shikimate pathway in apicomplexan parasites.
Nature, 393, 801-805.
Romanowski, M.J., Burley, S.K. (2002). Crystal structure of the Escherichia coli
Shikimate Kinase I (AroK) That confers sensitivity to Mecillinam. Proteins: Structure,
Function, and Genetics, 47, 558-562.
Sali, A., Blundell, T. L. (1993). Comparative protein modelling by satisfaction of
spatial restraints. J. Mol. Biol
., 234, 779-815.
Schlauderer, G. J., Schulz, G. E. (1996). The structure of bovine mitochondrial
adenylate kinase: Comparison with isoenzymes in other compartments. Protein Sci. 5,
434-441.
Schönbrunn, E., Eschenburg, S., Shuttleworth, W. A., Schloss, J. V., Amrhein, N.,
Evans, J. N. S., Kabsch, W. (2001). Interaction of the herbicide glyphosate with its target
enzyme 5-enolpyruvylshikimate 3-phosphate synthase in atomic detail.
Proc. Natl. Acad.
Sci. USA, 98, 1376-1380.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Shuttleworth, W. A., Pohl, M. E., Helms, G. L., Jakeman, D. L., Evans, J. N. S.
(1999). Site-directed mutagenesis of putative active site residues of 5-
enolpyruvylshikimate-3-phosphate synthase. Biochemistry, 38, 296-302.
Smith, C. A., Rayment, I. (1996). Active site comparisons highlight structural
similarities between myosin and other P-loop proteins. Biophys. J.,70, 1590-1602.
Sprinson, D.B. (1960). The biosynthesis of aromatic compounds from D-glucose. Adv.
Carbohydrate Chem. 15, 235–270.
Stallings, W. C., Abdel-Meguid, S. S., Lim, L. W., Shieh, H.-S., Dayringer, H. E.,
Leimgruber, N. K., Stegeman, R. A., Anderson, K. S., Sikorski, J. A., Padgette, S.
R., Kishore, G. M. (1991). Structure and topological symmetry of the glyphosate target
5-enol-pyruvylshikimate-3-phosphate synthase: a distinctive protein fold. Proc. Natl.
Acad. Sci. USA 88, 5046-5050.
Stehle, T., Schulz, G. E. (1990). Three-dimensional structure of the complex of
guanylate kinase from yeast with its substrate GMP. J. Mol. Biol. 211, 249-254.
Thompson, J.D., Higgins, D.G., Gibson, T.J. (1994). CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting,
positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–
4680.
Trouiller, P., Torreele, E., Olliaro, P., White, N., Foster, S., Wirth D., Pecoul, B.
(2001). Drugs for neglected diseases: a failure of the market and a public health failure?
Tropical Medicine And International Health,
v. 6
, p.945-951.
Villar, M. (2004).Toxicidade aos antibacilares.
[http://www.sppneumologia.pt/docspdfs/Toxicidade.pdf]. Acessado em Agosto/2005.
Vinella, D., Gagny, B., Joseleau-Petit, D., D’Ardi, R., Cashel, M. (1996). Mecillinam
resistance in Escherichia coli is conferred by loss of a second activity of the aroK
protein. J. Bacteriol
.
178
, 3818-3828.
Vonrhein, C., Schlauderer, G. J., Schulz, G. E. (1995). Movie of the structural
changes during a catalytic cycle of nucleoside monophosphate kinases. Structure, 3,
483–490.
Walker, J. E., Saraste, M., Runswick, M. J., Gay, N. J. (1982). Distantly related
sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other
ATP-requiring enzymes and a common nucleotide binding fold. EMBO J. 1, 945–951.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Whipp, M. J., Pittard, A. J. (1995). A reassessment of the relationship between aroK-
and aroL-encoded shikimate kinase enzymes of Escherichia coli. J. Bacterial. 177,
1627-1629.
WHO (2001). Stop TB Annual Report 2001. World Health Organization, Geneva,
Switzerland, WHO/CDS/STB/2002.17. Reprint:
[http://www.who.int/gtb/publications/TBCatalogue.htm#2002]. Acessado em
Fevereiro/2004.
WHO (2002)
. Tuberculosis. Fact Sheet N° 104. World Health Organization, Geneva,
Switzerland.August/2002.
[http:www.who.int/mediacentre/factsheets/who104/en/print.html]. Acessado em
Março/2004.
WHO (2003). The global plan to stop tuberculosis. World Health Organization, Geneva,
Switzerland, WHO/CDS/STB/2003.23. Reprint
[http://www.who.int/gtb/publications/TBCatalogue.htm#2003]. Acessado em
Fevereiro/2004.
WHO (2004a). Tuberculosis. Fact Sheet N
°
104. World Health Organization,
Geneva,Switzerland.March/2004.
[http://www.who.int/mediacentre/factssheetts/fs104/en/print.html]. Acessado em
Março/2004.
WHO (2004b). Global tuberculosis control. In WHO Report 2004. World Health
Organization, Geneva, Switzerland, WHO/HTM/TB/2004.331. Reprint
[http://www.who.int/gtb/publications/TBCatalogue.htm#2004]. Acessado em
Março/2004.
Wild, K., Bohner, T., Aubry, A., Folkers, G., Schulz, G. E. (1995). The three-
dimensional structure of thymidine kinase from Herpes simplex virus type 1. FEBS
Letters, 368, 289-292.
Yan, H., Ysai, M.-D. (1999). Nucleoside monophosphate kinases: structure,
mechanism, and substrat specificity. Advan. Enzymol. Relat. Areas Mol. Biol. 73, 103-
134.
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexos
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexos:
Anexo 1 – “Structure of shikimate kinase from
Mycobacterium
tuberculosis
reveals the binding of shikimic acid”
Anexo 2 – “Shikimate Kinase: A potential target for development of
novel anti-tubercular agents” (Review)
Anexo 3 – “Molecular model of shikimate kinase from
Mycobacterium
tuberculosis
”
Anexo 4 – “Structural bioinformatics study of EPSP synthase from
Mycobacterium tuberculosis
”
Anexo 5 – “Molecular modeling and CD analysis of chorismate
synthase”
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexo 1
electronic reprint
Acta Crystallographica Section D
Biological
Crystallography
ISSN 0907-4449
Editors: E. N. Baker and Z. Dauter
Structure of shikimate kinase from
Mycobacterium tuberculosis
reveals
the binding of shikimic acid
Jos
´
e Henrique Pereira, Jaim Sim
˜
oes de Oliveira, Fernanda Canduri, Marcio Vinicius
Bertacine Dias, M
´
ario S
´
ergio Palma, Luiz Augusto Basso, Di
´
ogenes Santiago Santos
and Walter Filgueira de Azevedo Jr
Copyright © International Union of Crystallography
Author(s) of this paper may load this reprint on their own web site provided that this cover page is retained. Republication of this article or its
storage in electronic databases or the like is not permitted without prior permission in writing from the IUCr.
Acta Cryst.
(2004). D60, 2310–2319 Pereira
et al.
Shikimate kinase
research papers
2310 doi:10.1107/S090744490402517X Acta Cryst. (2004). D60, 2310±2319
Acta Crystallographica Section D
Biological
Crystallography
ISSN 0907-4449
Structure of shikimate kinase from Mycobacterium
tuberculosis reveals the binding of shikimic acid
Jose
Â
Henrique Pereira,
a
Jaim Simo
Ä
es de Oliveira,
b
Fernanda Canduri,
a,c
Marcio Vinicius Bertacine Dias,
a
Ma
Â
rio Se
Â
rgio Palma,
c,d
Luiz Augusto Basso,
b,e
Dio
Â
genes Santiago Santos
e
* and
Walter Filgueira de Azevedo
Jr
a,d
*
a
Programa de Po
Â
s-GraduacËa
Ä
o em Biofõ
Â
sica
Molecular, Departamento de Fõ
Â
sica, UNESP,
Sa
Ä
o Jose
Â
do Rio Preto, SP 15054-000, Brazil,
b
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,
c
Center for Applied
Toxinology, Institute Butantan, Sa
Ä
o Paulo,
SP 05503-900, Brazil,
d
Laboratory of Structural
Biology and Zoochemistry, CEIS/Department of
Biology, Institute of Biosciences, UNESP, Rio
Claro, SP 13506-900, Brazil, and
e
Centro de
Pesquisa e Desenvolvimento em Biologia
Molecular e Funcional, Pontifõ
Â
cia Universidade
Cato
Â
lica do Rio Grande do Sul, Porto Alegre,
RS 90619-900, Brazil
Correspondence e-mail: [email protected],
[email protected]
# 2004 International Union of Crystallography
Printed in Denmark ± all rights reserved
Tuberculosis made a resurgence in the mid-1980s and now kills
approximately 3 million people a year. The re-emergence of
tuberculosis as a public health threat, the high susceptibility of
HIV-infected persons and the proliferation of multi-drug-
resistant strains have created a need to develop new drugs.
Shikimate kinase and other enzymes in the shikimate pathway
are attractive targets for development of non-toxic antimicro-
bial agents, herbicides and anti-parasitic drugs, because the
pathway is essential in these species whereas it is absent from
mammals. The crystal structure of shikimate kinase from
Mycobacterium tuberculosis (MtSK) complexed with MgADP
and shikimic acid (shikimate) has been determined at 2.3 A
Ê
resolution, clearly revealing the amino-acid residues involved
in shikimate binding. This is the ®rst three-dimensional
structure of shikimate kinase complexed with shikimate. In
MtSK, the Glu61 residue that is strictly conserved in shikimate
kinases forms a hydrogen bond and salt bridge with Arg58 and
assists in positioning the guanidinium group of Arg58 for
shikimate binding. The carboxyl group of shikimate interacts
with Arg58, Gly81 and Arg136 and the hydroxyl groups
interact with Asp34 and Gly80. The crystal structure of MtSK±
MgADP±shikimate will provide crucial information for the
elucidation of the mechanism of the shikimate kinase-
catalyzed reaction and for the development of a new
generation of drugs against tuberculosis.
Received 24 August 2004
Accepted 5 October 2004
PDB Reference: shikimate
kinase±MgADP±shikimate
complex, 1we2, r1we2sf.
1. Introduction
The shikimate pathway is a seven-step biosynthetic route that
generates chorismic acid from phosphoenol pyruvate and
erythrose 4-phosphate. The shikimate pathway is an attractive
target for the development of antimicrobial agents (Davies et
al., 1994) and herbicides (Coggins, 1989) because it is essential
in algae, higher plants, bacteria and fungi, whilst being absent
from mammals (Bentley, 1990). Several enzymes of this
pathway have been submitted to structural study (Arcuri et al.,
2004; Pereira et al., 2003) in order to propose inhibitors for
these enzymes. More recently, by disruption of the aroK gene,
which codes for the shikimate kinase enzyme, the shikimate
pathway has been shown to be essential for the viability of
Mycobacterium tuberculosis (Parish & Stoker, 2002). Shiki-
mate kinase (SK; EC 2.7.1.71), the ®fth enzyme of the
pathway, catalyses the speci®c phosphorylation of the
3-hydroxyl group of shikimic acid using ATP as a co-substrate.
Three previously solved structures of SK from Erwinia
chrysanthemi (EcSK; Krell et al., 1998, 2001) show that SK
belongs to the same structural family as nucleoside mono-
phosphate (NMP) kinases. The NMP kinases are composed of
three domains: the CORE, LID and NMP-binding (NMPB)
domains. A characteristic feature of the NMP kinases is that
electronic reprint
they undergo large conformational changes during catalysis
(Vonrhein et al., 1995).
Three functional motifs of nucleotide-binding enzymes are
recognizable in M. tuberculosis SK, including a Walker A
motif, a Walker B motif and an adenine-binding loop. The
Walker A motif is located between the ®rst -strand (1) and
®rst -helix (1), containing the conserved sequence
GXXXXGKT/S (Walker et al., 1982), where X represents any
residue. This motif forms the phosphate-binding loop (P-loop;
Smith & Rayment, 1996). In addition to the Walker A motif, a
second conserved sequence ZZDXXG called the Walker B
motif (Walker et al., 1982) is observed, where Z represents a
hydrophobic residue. An Asp3Ser replacement exists in
MtSK. The Walker B motif consensus in SKs is ZZZTGGG
and the second glycine (Gly80 in MtSK) has been implicated
in hydrogen bonding to the -phosphate of ATP. The adenine-
binding loop motif may be described as an I/VDXXX(X)XP
sequence stretch (Gu et al., 2002).
The shikimate-binding domain has previously been
assigned on the basis of the difference Fourier map of EcSK
and structural comparison with NMP kinases. Shikimate was
included in the crystallization conditions (co-crystallization);
however, the electron density was not clear enough to include
shikimate in the previously solved crystallographic structure
(Krell et al., 1998).
Two crystal structures of SK from M. tuberculosis (MtSK;
Gu et al., 2002; PDB codes 1l4y and 1l4u) have revealed the
dynamic role of the LID domain in catalysis, but the precise
position and interactions between shikimate and MtSK were
not unequivocally demonstrated as the shikimate-binding site
was not occupied by the substrate. A previously reported
molecular-modelling study of the complex between MtSK and
shikimate also failed to predict all the intermolecular
hydrogen bonds (de Azevedo et al., 2002). Here, we describe
the crystal structure of the MtSK±MgADP±shikimate ternary
complex at 2.3 A
Ê
resolution, unequivocally revealing in detail
the interactions of amino-acid residues with bound shikimate
and the conformational changes upon substrate binding. The
crystal structure of MtSK±MgADP±shikimate will provide
crucial information for the design of non-promiscuous SK
inhibitors that target both the shikimate- and ATP-binding
pockets or uniquely the shikimate-binding site.
2. Materials and methods
2.1. Crystallization
Cloning, expression and puri®cation have been reported
elsewhere (Oliveira et al., 2001). MtSK was concentrated and
dialyzed against 50 mM Na HEPES buffer pH 7.5 containing
0.5 M NaCl and 5.0 mM MgCl
2
. This protein solution was
brought to 13.0 mM in shikimate and ADP by addition of the
pure solids and centrifuged prior to crystallization. The
protein concentration was about 17.0 mg ml
À1
. Crystals were
obtained by the hanging-drop vapour-diffusion method. The
well solution contained 0.1 M Na HEPES buffer pH 7.5, 10%
2-propanol, 35% PEG 3350 and the drops consisted of a
mixture of 1.0 ml well solution and 1.5 ml protein solution.
2.2. Data collection and processing
The data set for MtSK±MgADP±shikimate was collected at
a wavelength of 1.431 A
Ê
using the Synchrotron Radiation
Source (Station PCr, LNLS, Campinas, Brazil; Polikarpov et
al., 1998) and a CCD detector (MAR CCD). The cryo-
protectant contained 15% glycerol, 12% PEG 3350 and 3.5%
propanol. The crystal was ¯ash-frozen at 104 K in a cold
nitrogen stream generated and maintained with an Oxford
Cryosystem. The oscillation range used was 1.0
, the crystal-
to-detector distance was 90 mm and the exposure time was
50 s. A data set containing 160 frames was collected and
processed to 2.3 A
Ê
resolution using the program MOSFLM
(Leslie, 1992) and scaled with SCALA (Collaborative
Computational Project, Number 4, 1994).
2.3. Molecular replacement and crystallographic refinement
The crystal structure of MtSK±MgADP±shikimate was
determined by standard molecular-replacement methods
using the program AMoRe (Navaza, 1994), using as a search
model the structure of MtSK±MgADP (PDB code 1l4y; Gu et
al., 2002). After translation-function computation the corre-
lation was 65% and the R factor was 38.3%. The highest
magnitude of the correlation coef®cient function was obtained
for the Euler angles = 54.85, = 85.32, = 91.90
.The
fractional coordinates are T
x
= 0.7336, T
y
= 0.5326, T
z
= 0.2768.
The atomic positions obtained from molecular replacement
were used to initiate the crystallographic re®nement. Structure
re®nement was performed using X-PLOR (Bru
È
nger, 1992).
During rigid-body re®nement, the R factor decreased from
38.3 to 35.8%. Further re®nement continued with simulated
annealing using the slow-cooling protocol, followed by alter-
nate cycles of positional re®nement and manual rebuilding
using XtalView (McRee, 1999). Finally, the positions of waters,
MgADP and shikimate were checked and corrected in
F
obs
À F
calc
maps. The ®nal model has an R factor of 20.7%
and an R
free
of 28.7%.
Root-mean-square deviation differences from ideal
geometries for bond lengths, angles and dihedrals were
calculated with X-PLOR (Bru
È
nger, 1992). The overall
stereochemical quality of the ®nal model for MtSK±MgADP±
shikimate was assessed by the program PROCHECK
(Laskowski et al., 1994). Atomic models were superposed
using the program LSQKAB from CCP4 (Collaborative
Computational Project, Number 4, 1994).
The molecular-surface areas were calculated using the
program Swiss PDB Viewer v.3.7 (http://www.expasy.org/
spdbv), a probe radius of 1.4 A
Ê
and a ®xed radius for all atoms.
3. Results and discussion
The crystals of MtSK±MgADP±shikimate were grown in the
presence of 5.0 mM MgCl
2,
13.0 mM ADP and 13.0 mM
shikimate. The data set of MtSK±MgADP±shikimate was
research papers
Acta Cryst. (2004). D60, 2310±2319 Pereira et al.
Shikimate kinase 2311
electronic reprint
collected at 2.3 A
Ê
(Table 1) using the Synchrotron Radiation
Source (Polikarpov et al., 1998) and the structure was solved
by molecular replacement. There is one molecule of MtSK±
MgADP±shikimate in the asymmetric unit, containing resi-
dues 2±166, Mg
2+
, ADP, shikimate, two Cl
À
ions and 144 water
molecules (Fig. 1). Therefore, the enzyme crystallized as a
MtSK±MgADP±shikimate dead-end ternary complex. The
N-terminal methionine residue is not observed and the ten
C-terminal residues (NQIIHMLESN) are disordered. Details
of the re®nement and the ®nal model statistics are presented
in Table 2. Analysis of the Ramachandran diagram 9±2 plot
shows that 93.4% of non-glycine residues lie in most favoured
regions and there are no residues in the disallowed region
(Fig. 2). The average B factor for main-chain atoms is
34.64 A
Ê
2
, whereas that for side-chain atoms is 35.68 A
Ê
2
(Table 2). In order to accurately determine the position of
shikimate binding in the active site of MtSK, larger ®nal
concentrations of ADP and shikimate in the drop (8 mM)
were used than those used by Gu et al. (2002) to obtain the
crystals (4 mM). The average B-factor value of 26.15 A
Ê
2
obtained for shikimate indicates a higher order and occupancy
of this substrate in the present structure than in the previously
solved SK structures, where either shikimate was not
research papers
2312 Pereira et al.
Shikimate kinase Acta Cryst. (2004). D60, 2310±2319
Figure 1
MtSK complexed with Mg
2+
, ADP and shikimate. The LID (residues 112±
124) and SB (residues 33±61) domains are responsible for large
conformational changes during catalysis and are shown. The ®gure was
prepared with MOLMOL (Koradi et al., 1996).
Table 1
Summary of data-collection statistics for MtSK±MgADP±shikimate.
Values in parentheses are for the highest resolution shell.
Unit-cell parameters
a = b (A
Ê
) 62.91
c (A
Ê
) 90.92
Resolution (A
Ê
) 29.74±2.30 (2.41±2.30)
Space group P3
2
21
No. of measurements with I >2'(I) 34274
No. of independent re¯ections 9563
Overall redundancy 2.4
Completeness (%) 98.7 (98.7)
R
sym
² (%) 3.0 (7.2)
Table 2
Summary of re®nement statistics for MtSK±MgADP±shikimate.
Values in parentheses are for the highest resolution shell.
Resolution range (A
Ê
) 6.00±2.30
Re¯ections used for re®nement 8885
No. of residues 165
No. of water O atoms 144
No. of ADP molecules 1
No. of metal ions (Mg
2+
)1
No. of Cl
À
2
No. of shikimate molecules 1
Final R factor² (%) 20.7 (31.8)
Final R
free
³ (%) 28.7 (36.9)
B factors§ (A
Ê
2
)
Main chain 34.64
Side chains 35.68
Waters 39.63
ADP 21.81
Mg
2+
34.70
Cl
À
37.65
Shikimate 26.15
Observed r.m.s.d. from ideal geometry
Bond lengths (A
Ê
) 0.017
Bond angles (
) 1.905
Dihedrals (
) 22.125
Ramachandran plot
Most favoured 9/2 angles (%) 93.4
Disallowed 9/2 angles (%) 0
² R factor = 100 Â
jF
obs
À F
calc
ja
F
obs
, the sums being taken over all re¯ections with
F/'(F )>2'(F ). ³ R
free
= R factor for 10% of the data that were not included during
crystallographic re®nement. § B values = average B values for all non-H atoms.
Figure 2
Ramachandran diagram for MtSK±MgADP±shikimate. 93.4% of non-
glycine residues lie in the most favoured regions and no residues are in
the disallowed region.
electronic reprint
complexed or its electron density was too poor to accurately
locate the substrate (Table 2) (Gu et al., 2002; Krell et al.,
1998).
SK displays an /-fold and consists of ®ve central parallel
-strands ¯anked by -helices. As pointed out above, a char-
acteristic feature of SKs is that they undergo large confor-
mational changes during catalysis. There are two ¯exible
research papers
Acta Cryst. (2004). D60, 2310±2319 Pereira et al.
Shikimate kinase 2313
Figure 3
Mg
2+
coordination. (a) 1l4y, (b) 1l4u (Gu et al., 2002), (c) MtSK±MgADP±shikimate. For simplicity, only protein residues, atoms involved in binding of
Mg
2+
water and ADP are shown. Broken black lines represent hydrogen bonds or Mg
2+
coordination. The distances are in A
Ê
. The following colour
scheme was adopted: grey for carbon, red for oxygen, blue for nitrogen, yellow for chlorine, purple for magnesium and orange for phosphorus.
regions of the structure that are responsible for movement: the
SB and LID domains (Fig. 1), which in MtSK correspond to
residues 33±61 and 112±124, respectively.
3.1. Interaction with ADP/Mg
2+
The structures of MtSK complexed with MgADP (Gu et al.,
2002) are highly similar, with a root-mean-square deviation of
0.19 A
Ê
for all pairs of C
atoms. The adenine moiety of ADP is
sandwiched between Arg110 and Pro155 as observed for the
MtSK±MgADP structure (Gu et al., 2002). The Arg110 in
MtSK represents the ®rst residue in a conserved motif, typi-
cally RXX(X)R, of the LID domain observed for P-loop
kinases (Leipe et al., 2003). The second conserved basic
residue of this motif interacts with the -phosphate of ATP. In
MtSK, this residue would be Arg117, which interacts with the
- and -phosphate groups (Fig. 3), and thus the conserved
motif of the LID domain for P-loop shikimate kinases would
be R(X)
6±9
R (Fig. 4). The Arg117 residue may stabilize the
transition state by neutralizing the developing negative charge
on the ± bridge O atom (Hasemann et al., 1996). The Pro155
is the last residue of the adenine-binding loop motif (residues
148±155 in MtSK), which was ®rst recognized in AK and
EcSK (Krell et al., 1998) and has been described as an
I/VDXXX(X)XP sequence stretch (Gu et al., 2002). This motif
forms a loop that wraps around the adenine moiety of ATP,
connecting the 5-strand with the C-terminal 8-helix. The
second (aspartate) residue and the last (proline) residue are
not conserved in the adenine-binding loops of aroK-encoded
SKs from Escherichia coli (Romanowski & Burley, 2002) and
Campylobacter jejuni (PDB code 1via). However, they are
electronic reprint
recognizable from structural alignment analysis using MtSK±
MgADP as a reference structure. Accordingly, the I/VD-
(X)XP primary sequence cannot be used to identify the
adenine-binding loop sequences of all shikimate kinases. The
carbonyl group of Arg153 residue forms hydrogen bonds to
both the N6 atom of adenine and to a water molecule, which in
turn hydrogen bonds to the N1 atom of adenine.
There are minor differences between the MtSK±MgADP
and MtSK±shikimate±MgADP structures in the position of
Mg
2+
. In the MtSK±MgADP structure (PDB code 1l4y) a
typical six-coordination is observed for Mg
2+
(Fig. 3a),
whereas magnesium six-coordination (or, more accurately,
seven-coordination) is distorted in the structure of the
Pt-derivative of MtSK±MgADP (PDB code 1l4u) (Fig. 3b).
The structure of MtSK±MgADP±shikimate shows a distorted
six-coordinated Mg
2+
(Fig. 3c). Binding distances to Mg
2+
are
shown in Table 3.
The Mg
2+
of MtSK±MgADP±shikimate interacts with a
-phosphate oxygen of ADP, Ser16 OG of the Walker A motif
and four water molecules. In EcSK, the Mg
2+
is four-coordi-
nated, interacting with a -phosphate O atom of ADP,
Thr16 OG1, Asp32 OD2 and a water molecule held in position
by Asp34 (Krell et al., 1998). The Mg
2+
binding in MtSK±
MgADP (Gu et al., 2002) and MtSK±MgADP±shikimate
structures are somewhat similar: the six-coordinated magne-
sium ion structural water2 is held in place by a hydrogen bond
to Asp32, which is conserved in all SKs (Fig. 4). Super-
imposition of MtSK±MgADP on MtSK±MgADP±shikimate
showed that in the latter structure water1 and water4 are in
equivalent positions but that shifts of 1.32 and 2.75 A
Ê
are
observed for water2 and water3, respectively (Fig. 3). Water3
has moved to back the plane formed between the magnesium
ion, the O2B atom of ADP and the side chain of Asp34.
Water3 is held in place by Asp32 and Asp34 (Fig. 3). Asp32
also forms a hydrogen bond to Ser16 of the Walker A motif,
whereas the interaction between Asp32 and Ser16 is via a
bridging water molecule (water6) in the MtSK±MgADP
structure (Gu et al., 2002). A rotation of the Asp32 side-chain
dihedral angles 1
1
and 1
2
leads to a direct interaction with
Ser16 OG, which accounts for the exclusion of water6 from the
magnesium-binding site in the ternary structure. The Mg
2+
-
coordinated water1 interacts with a chloride ion instead of
interacting with Asp34 via a bridging water molecule (Fig. 3,
water5) as observed in the MtSK±MgADP structure (Gu et al.,
2002). This chloride ion also interacts with the 3-hydroxyl
group of shikimate and the backbone amide of Gly80.
Moreover, Asp34 makes hydrogen bonds with the 2- and
3-hydroxyl groups of shikimate. Hence, the different mode of
interaction observed for residue Asp34 arises from the
presence of shikimate, which leads to the exclusion of water5
from the MtSK active site. In addition, the distance between
ADP -phosphate O2 and Mg
2+
changes from 2.23 A
Ê
in our
structure (Table 3) to 2.15 A
Ê
for the
non-distorted six-coordinated magne-
sium ion in the MtSK±ADP structure
(Gu et al., 2002), thereby accounting for
the slightly distorted six-coordination of
Mg
2+
. Interestingly, the two functions of
the aspartate of the ZZDXXG Walker
B motif (hydrogen bonding to Mg
2+
water and to the hydroxy group of the
serine or threonine of the Walker A
motif) are taken over by the two
conserved aspartate residues located at
the last position of strand 2 (Asp32 in
MtSK) and the ®rst position of helix 2
(Asp34 in MtSK) of shikimate kinase
and gluconate kinase (Kraft et al., 2002).
The Walker B motif consensus sequence
in shikimate kinases has been proposed
to be ZZZTGGG (residues 75±81 in
MtSK; Leipe et al., 2003) and the second
glycine (Gly80 in MtSK) has been
implicated in hydrogen bonding the
-phosphate of ATP. Nucleophilic
attack on the -phosphate of ATP will
be most facilitated by metal-ion binding
research papers
2314 Pereira et al.
Shikimate kinase Acta Cryst. (2004). D60, 2310±2319
Figure 4
Alignment of shikimate kinases showing that the residues identi®ed in binding of shikimate are
conserved. The secondary-structure assignment for MtSK±shikimate is shown above the sequence
and the locations of linear motifs are labelled. Shaded regions in yellow are the residues involved in
binding of shikimate and the shaded region in red is the residue Glu61 (MtSK). Stretches in blue
and green are the shikimate-binding domain and the LID domain, respectively. AK_MYCTU,
M. tuberculosis SK I; AK_CAMJE, C. jejuni SK I; AK_ECOLI, E. coli SK I; AL_ECOLI, E. coli SK
II; AL_SALTY, S. typhimurium SK II; AL_ERWCH, E. chrysanthemi.
Table 3
Binding of Mg
2+
in MtSKs.
n.o., not observed.
Atom MtSK±MgADP±shikimate (A
Ê
) 1l4y (A
Ê
) 1l4u (A
Ê
)
Ser16 OG 2.36 2.15 2.57
ADP O2B 2.23 2.15 2.39
Water1 O 2.39 2.24 2.52
Water2 O 2.23 2.04 2.77
Water3 O 2.75 2.25 2.37
Water4 O 2.97 2.05 2.94
Water5 O n.o. n.o. 3.06
electronic reprint
(Mg
2+
) to the - and -phosphate groups, whereas departure
of the leaving group will be most favoured in a structure with
metal binding to the - and -phosphate groups (Jencks,
1975a). In enzyme reactions, the enzyme may facilitate the
reaction and contribute to its speci®city simply by favouring
the binding of metal in the most favourable possible structure
for the particular reaction that is being catalyzed. Although
the mechanism of action of MtSK is still unknown, the inter-
action between Gly80 and the chloride ion may indicate that
the latter occupies the -phosphate position as the chemical
reaction proceeds, thereby suggesting that the interaction
between Gly80 and -phosphate of ATP may play a role in the
catalytic mechanism of MtSK, as discussed below.
3.2. Shikimate binding
The shikimate-binding domain, which follows strand 2,
consists of helices 2 and 3 and the N-terminal region of
helix 4 (residues 33±61). A peak of more than 3' in the ®nal
F
obs
À F
calc
difference Fourier map clearly indicates the
position of bound shikimate in the electron density (Fig. 5).
The chemical structure of shikimic acid [3R-(3,4,5)]-
3,4,5-trihydroxy-1-cyclohexene-1-carboxylic acid] is shown in
Fig. 6. The guanidinium groups of Arg58 and Arg136 and the
NH backbone group of Gly81 interact with the carboxyl group
of shikimate. The 3-hydroxyl group of shikimate forms
hydrogen bonds with the carboxyl group of Asp34, the main-
chain NH group of Gly80 and a water molecule (Fig. 7). This
water molecule in turn mediates interactions with the side
chains of SB residues Arg58 and Glu61, Walker B residues
Gly79, Gly80 and Gly81, and Ala82. The 2-hydroxyl group of
shikimate hydrogen bonds to the side chain of Asp34. The
distances of direct and water-mediated hydrogen bonds in the
shikimate binding of MtSK are shown in Table 4.
research papers
Acta Cryst. (2004). D60, 2310±2319 Pereira et al.
Shikimate kinase 2315
Figure 5
F
obs
À F
calc
electron density contoured at 3.0' showing the binding of
shikimate. The ®gure was prepared with XtalView (McRee, 1999).
Figure 6
The molecular structure of shikimate shows the carboxyl group and three
hydroxyl groups. C atoms are numbered in blue and O atoms in red.
Figure 7
Polar interactions involved in the binding of shikimate to the MtSK active
site. Hydrogen bonds between shikimate and protein groups and
interactions between protein groups are represented as broken black
lines. The interaction distances are in A
Ê
.
Table 4
Direct and water-mediated hydrogen bonding of shikimate in MtSK.
All distances <3.6 A
Ê
are shown. n.o., not observed.
Shikimate Atom
Distances
(A
Ê
)
MtSK atom
hydrogen bonded
to water molecule
Distances
(A
Ê
)
Hydroxyl groups
O1 Gly80 N 3.10
Asp34 OD1 2.82
Asp34 OD2 2.57
Water320 O 2.46 Gly79 O 2.96
Gly80 N 3.13
Gly81 N 2.51
Ala82 N 3.54
Arg58 NH1 3.12
Glu61 OE2 2.99
O2 Asp34 OD1 2.66
Asp34 OD2 2.85
O3 n.o
Carboxyl group
O4 Gly81 N 3.22
Arg136 NH2 2.68
O5 Arg58 NH2 2.67
Gly81 N 3.49
Arg136 NH2 2.34
electronic reprint
Glu61 is conserved in both aroK- and aroL-encoded
shikimate kinase enzymes and has been implicated in shiki-
mate binding (Gu et al., 2002; Krell et al., 1998; Romanowski &
Burley, 2002). Krell and coworkers proposed that Glu61 is
suitably positioned to bind the 5-hydroxyl group of shikimate
in EcSK (Krell et al., 1998). However, the amino-acid residues
comprising the shikimate-binding domain were not clearly
demonstrated because the electron density was not suf®cient
to position the shikimate molecule in the structure. In our
structure, the Glu61 side chain forms a water-mediated
interaction with the 3-hydroxyl group of shikimate. In addi-
tion, Glu61 forms a hydrogen bond and a salt bridge with the
conserved Arg58 and assists in positioning the guanidinium
group of Arg58 for substrate binding via interactions with the
carboxylate group of shikimate. Glu61 OE2 makes a hydrogen
bond with the amide N atom of Ala82. Moreover, Arg58 NH1
forms a hydrogen bond with Gly80 O. Therefore, Glu61 plays
an important role in positioning a shikimate molecule, even
though it is not directly involved in substrate binding.
Furthermore, Glu61 is conserved in all SKs sequenced so far,
which corroborates its role in anchoring Arg58 in the shiki-
mate-binding site. The Glu54 carboxylate group also appears
to anchor the guanidinium group of Arg58 for interaction with
the shikimate carboxylate group, as suggested in the MtSK±
MgADP structure (Gu et al., 2002). However, Glu54 is not
conserved (Fig. 4), except for aroK-encoded shikimate
kinases.
3.3. Chloride ion binding
Chloride ions have been proposed to play an important role
in determining the af®nity of E. chrysanthemi SK for ATP and
shikimate (Cerasoli et al., 2003). However, no chloride ions in
the active-site cavity are found in equivalent positions when
the MtSK structures are superimposed on both E. chry-
santhemi SK (Krell et al., 1998) and the K15M mutant (Krell et
al., 2001) SKs. We have found only two chloride ions in
MtSK±MgADP±shikimate that have equivalents in the
MtSK±MgADP complex structures (1l4u and 1l4y; Gu et al.,
2002). One of them is bound to the active-site cavity and
makes hydrogen bonds with 3-hydroxyl group of shikimate,
Gly80 NH and water1 (Fig. 8), whereas the other is on the
enzyme surface at a large distance from the active site. The
conserved residue Arg117 in the LID domain is involved in
ADP binding by forming two hydrogen bonds with the -and
-phosphate O atoms (Fig. 7). Lys15 forms a hydrogen bond
with a -phosphate O atom (Fig. 3) and the chloride ion in the
MtSK±MgADP structures (Fig. 8). The main-chain NH of
Gly80 is hydrogen bonded to the chloride ion in the binary
(1l4y) and ternary complex structures (Table 5). These resi-
dues are located in the vicinity of where the chemical reaction
occurs and may thus play a critical role in transition-state
stabilization. Consistent with this proposal, the K15M mutant
of EcSK showed no detectable enzyme activity (Krell et al.,
2001), although it was in fact a K15M/P115L double mutant.
Gly80 of the Walker B motif has been implicated in hydrogen
bonding to the -phosphate of ATP (Krell et al., 1998). The
distance between the chloride ion and the hydroxyl group
bound to carbon 3 of shikimate (to which a phosphoryl
group would be transferred) is 3.36 A
Ê
in the MtSK±MgADP±
shikimate structure (Fig. 8). Lys15 and Gly80 are in slightly
different positions in the ternary complex compared with the
MtSK±MgADP binary complex. The distance between O1B of
the -phosphate of ATP (2.81 A
Ê
) and Lys15 is 2.96 A
Ê
in the
ternary complex (Fig. 3). However, the distance between
Lys15 and the chloride ion is greater in the ternary complex
(3.94 A
Ê
, Table 5). The distance between Cl
À
and the main-
chain NH of Gly80 (3.49±3.69 A
Ê
) in the binary complex is
shortened to 3.24 A
Ê
in the ternary complex. There appears to
be a concerted movement of Lys15, Gly80 and the Cl
À
ion
(Fig. 8). This chloride ion appears to be shifted by 1.76 and
research papers
2316 Pereira et al.
Shikimate kinase Acta Cryst. (2004). D60, 2310±2319
Figure 8
Chloride ions bound to the active sites of the MtSK±MgADP±shikimate
and MtSK±MgADP (1l4u; Gu et al., 2002) structures. For clarity, only
residues Lys15, Ser16, Asp32, Asp34 and Gly80, shikimate, the - and -
phosphates of ADP, Mg
2+
and Cl
À
ions and Mg
2+
-coordinated waters are
shown. The chloride ion hydrogen bonds are represented as broken black
lines and distances are in A
Ê
. The Cl
À
ion and the Lys15 and Gly80
residues of the superimposed MtSK±MgADP structure 1l4u are coloured
cyan.
Table 5
Binding of chloride ion in MtSK±MgADP and MtSK±MgADP±shikimate
structures.
Distance (A
Ê
)
Chloride ion Atom 1l4y 1l4u MtSK±MgADP
±shikimate
Cl180 Shikimate hydroxyl
group O1
4.48² 4.66² 3.36
Gly80 N 3.49 3.69 3.24
Lys15 NZ 3.17 3.25 3.94
Mg
2+
Water1 O 3.87 3.75 2.84
² The distance was measured with the MtSK±MgADP structures (1l4y and 1l4u)
superimposed on the MtSK±MgADP±shikimate structure.
electronic reprint
1.52 A
Ê
in MtSK±MgADP±shikimate compared with the
MtSK±MgADP 1l4u and 1l4y complexes, respectively. It is
thus tempting to suggest that the chloride ion in the ternary
complex occupies the phosphoryl position in the reaction
coordinate to shikimate 3-phosphate formation.
3.4. Conformational changes upon substrate binding
As pointed out above, MtSK belongs to the family of
nucleoside monophosphate (NMP) kinases, which are
composed of three domains: the CORE, LID and NMP-
binding domains. Kinases should undergo large movements
during catalysis to shield their active centre from water in
order to avoid ATP hydrolysis (Jencks, 1975b). The NMP and
LID domains of NMP kinases have been shown to undergo
large motions that are independent, in agreement with the
observed random bi-bi kinetics (Vonrhein et al., 1995).
Ligand-induced changes in the secondary structure of EcSK
have been detected by comparing the circular-dichroism
spectra of free enzyme, EcSK±shikimate binary complex and a
ternary complex of EcSK, shikimate and adenylyl imido-
diphosphate, an ATP analogue (Krell et al., 1998). Alignment
of C
positions of the MtSK±MgADP±shikimate dead-end
ternary complex and the MtSK±MgADP binary complex
structures shows that the LID and SB domains undergo
noticeable concerted movements towards each other (Fig. 9).
The structural alignment included all residues and yielded
r.m.s. deviation values of 0.56 and 0.54 A
Ê
for 1l4u and 1l4y,
respectively. There is a shift of the LID domain, with an r.m.s.
deviation of 1.33 A
Ê
for residues 112±124. The SB domain shift
is somewhat smaller, with a calculated r.m.s. deviation of
0.74 A
Ê
for residues 33±61. The shikimate-binding cavity is
delineated mainly by residues from the LID and SB domains,
the Walker B motif and Arg136 from the 7 helix. A close
inspection of the residues involved in these movements shows
that the side chains of Val116, Pro118 and Leu119 from the
LID domain and Ile45, Ala46, Glu54, Phe57 and Arg58 from
the SB domain shifted upon shikimate binding to the MtSK±
MgADP binary complex (Fig. 9). In MtSK with bound
MgADP and shikimate, a cluster of hydrophobic contacts is
formed between LID residues Val116, Pro118 and Leu119 and
SB residues Ala46 and Phe49, which account for the stabili-
zation of the partially closed shikimate-binding site cavity.
The conformational changes described above result in
closure of the MtSK binding site, as shown by the reduction
in molecular-surface area of MtSK±shikimate±MgADP
compared with MtSK±MgADP (Fig. 10). The ADP, Mg
2+
,
shikimate, Cl
À
ions and water molecules were removed prior
to calculation. The calculated values are approximately
7246 A
Ê
2
for MtSK complexed with MgADP (1l4u) and
6915 A
Ê
2
for MtSK complexed with MgADP and shikimate.
Thus, approximately 330 A
Ê
2
of molecular surface is buried on
shikimate binding. Since both MtSK±MgADP and MtSK±
MgADP±shikimate have been crystallized in the same space
group with similar values for the unit-cell parameters (Gu et
al., 2001) (Table 1) and residues 114±124 from the LID domain
and residues 33±61 from the SB domain form no crystal
contacts with symmetry-related MtSK molecules, the confor-
mational changes observed cannot
merely be a re¯ection of the different
crystal-packing arrangements.
The MtSK±MgADP±shikimate
structure should represent a partially,
but not totally, closed structure, since
total active-site closure upon dead-end
ternary complex formation would result
in locking the enzyme active site in an
inactive form in which shikimate
substrate binding to MtSK enzyme prior
to MgADP dissociation from its active
site would result in an inactive abortive
complex. Consistent with these struc-
tural results, measurements of EcSK
intrinsic tryptophan ¯uorescence
(Trp54) on shikimate binding to either
EcSK or the EcSK±MgADP binary
complex showed a modest synergism of
binding between these substrates, since
the dissociation constant value for
shikimate (K
d
= 0.72 mM) decreased to
0.3 mM in the presence of 1.5 mM ADP
(Idziak et al., 1997). Measurements of
the quenching of protein ¯uorescence
of the aroL-encoded EcSK upon
nucleotide binding demonstrated the
dissociation-constant values for ADP
research papers
Acta Cryst. (2004). D60, 2310±2319 Pereira et al.
Shikimate kinase 2317
Figure 9
Shikimate binding in the MtSK±MgADP±shikimate structure. The C
traces of MtSK±MgADP±
shikimate (grey) and MtSK±MgADP (cyan) (1l4u; Gu et al., 2002) were superimposed. The side
chains with large shifts owing to shikimate binding are shown for both structures: Val116, Pro118
and Leu119 from LID domain and Ile45, Ala46, Glu54, Phe57 and Arg58 from the SB domain. For
clarity, only residues from the LID and SB domains of 1l4u (cyan) are shown.
electronic reprint
and ATP to be 1.7 and 2.6 mM, respectively (Krell et al., 2001).
The K
m
for ATP (620 mM) was found to be approximately four
times lower than the dissociation constant in the absence of
shikimate (Krell et al., 2001). These results prompted the
proposal that the conformational changes in the enzyme
associated with the binding of the ®rst substrate lead to an
increase in the af®nity for the second substrate. However,
even if this holds for EcSK, it does not appear to hold for
MtSK since the apparent dissociation-constant values for ATP
(89 mM) and shikimate (44 mM) are similar to their K
m
values,
83 mM for ATP and 41 mM for shikimate considering a
random-order bi-bi enzyme mechanism (Gu et al., 2002).
Moreover, no evidence for synergism between shikimate and
ATP could be observed in substrate binding to EcSK in a
chloride-free buffer system (Cerasoli et al., 2003). The K15M
EcSK mutant has been crystallized in an open conformation
that is proposed to presumably be equivalent to an apo-
enzyme structure in which neither shikimate nor ADP (or
ATP) would be bound (Krell et al., 2001). This mutant was
produced to evaluate the role of the conserved Lys15 of the
Walker A motif (Krell et al., 2001). However, an unwanted
point mutation in the LID domain (Pro115Leu) was detected
during re®nement of the model and extensive contacts
between LID domains of neighbouring EcSK enzymes were
observed. It therefore appears unwarranted to consider the
double K15M/P115L EcSK mutant a model for the apo
enzyme. The incomplete LID-domain closure observed in the
crystal structure presented here may suggest that the
-phosphate of ATP plays a crucial role in the completion of
the domain movement, as has also been proposed by others
(Krell et al., 1998). The equilibrium constant for the intra-
molecular hydrolysis of bound ATP to bound ADP and
phosphate at enzyme active sites is considerably larger than
the equilibrium constant for ATP hydrolysis in solution
(Jencks, 1975a). Accordingly, the loss of two water molecules
(water5 and water6) described in x3.1 is consistent with the
exclusion of water molecules from the active site owing to the
partial closure of MtSK upon shikimate binding in order to
minimize ATP hydrolysis.
4. Conclusions
The residues identi®ed in the binding of shikimate whether
directly or indirectly (Asp34, Arg58, Glu61, Gly79, Gly80,
Gly81 and Arg136) are conserved in all SKs encoded by aroK
and aroL genes (Fig. 4). The structures of SKs deposited in the
PDB were superimposed and the positions of the residues
involved in binding between shikimate and SK are highly
conserved in aroK-encoded proteins (shikimate kinase from
M. tuberculosis and C. jejuni)andaroL-encoded proteins
(shikimate kinase from E. chrysanthemi). The conserved
active site may account for the similar K
m
values found for
MtSK (aroK-encoded) and EcSK (aroL-encoded); however, it
does not support the difference observed in the K
m
value of
SKI from E. coli. Romanowski & Burley (2002) proposed that
the substitution of Leu83 (EcSK) for Lys86 in E. coli SKI
disrupts the binding to shikimate. The loss of a stabilizing
hydrophobic residue at this position may explain the signi®-
cantly lower af®nity of SKI from E. coli for shikimate
compared with that of SKII. However, superimposition of SKI
from E. coli with MtSK±shikimate indicates that Lys86 is
distant from shikimate, which does not support Romanoswski
and Burley's proposition. Moreover, even though a substitu-
tion of Leu83 (EcSK) for Thr84 occurs in MtSK, which would
result in the loss of a stabilizing hydrophobic residue, the K
m
values for MtSK and EcSK are similar.
research papers
2318 Pereira et al.
Shikimate kinase Acta Cryst. (2004). D60, 2310±2319
Figure 10
Molecular surface of (a) MtSK±MgADP (1l4u; Gu et al., 2002) and (b)
MtSK±MgADP±shikimate structures. Chloride ion is coloured yellow in
MtSK±MgADP and MtSK±MgADP±shikimate. For water molecules only
O atoms are shown.
electronic reprint
Superimposition of SKI from E. coli on the MtSK±
MgADP±shikimate structure shows the overlap of Leu123
onto the shikimate structure; however, we cannot af®rm that
the binding sites are different, since the structure of SKI from
E. coli was crystallized without shikimate and shikimate
binding in the SKI structure may promote further conforma-
tion changes that may include a displacement of the Leu123
side chain.
Here, we describe the residues involved in shikimate
binding and conformational changes upon substrate binding to
the MtSK±MgADP complex, resulting in a partially closed
structure. A complete active-site closure could be achieved in
a complex of MtSK with shikimate, Mg
2+
and the non-
hydrolysable ATP analogue adenosine 5
H
-(,-methylene)
triphosphate (AMP-PCP). A likely drawback of ATP-binding-
site-based SK inhibitors would be their lack of speci®city,
owing to the common fold and similar ATP-binding site
shared by many P-loop kinases (Leipe et al., 2003). The
availability of the M. tuberculosis shikimate kinase structure
complexed with shikimate should allow the molecular design
of speci®c SK inhibitors that target either the shikimate- and
ATP-binding sites or the shikimate-binding site only. More-
over, the knowledge of functional factors that lead to active-
site closure could be used to design inhibitors that force MtSK
into a closed conformation that would be unable to catalyze
the phosphoryl transfer to shikimate.
This work was supported by grants from FAPESP
(SMOLBNet, Proc. 01/07532-0, 02/05347-4, 04/00217-0),
CNPq, CAPES and Instituto do Mile
Ã
nio (CNPq-MCT). WFA
(CNPq, 300851/98-7), MSP (CNPq, 500079/90-0) and LAB
(CNPq, 520182/99-5) are researchers of the Brazilian Council
for Scienti®c and Technological Development.
References
Arcuri, H. A., Canduri, F., Pereira, J. H., Silveira, N. J. F., Camara,
J. C. Jr, Oliveira, J. S., Basso, L. A., de Palma, M. S., Santos, D. S. &
de Azevedo, W. F. (2004). Biochem. Biophys. Res. Commun. 320,
979±991.
Azevedo, W. F. de, Canduri, F., de Oliveira, J. S., Basso, L. A., Palma,
M. S., Pereira, J. H. & Santos, D. S. (2002). Biochem. Biophys. Res.
Commun. 295, 142±148.
Bentley, R. (1990). Crit. Rev. Biochem. Mol. Biol. 25, 307±384.
Bru
È
nger, A. T. (1992). X-PLOR Version 3.1: A System for Crystal-
lography and NMR. New Haven: Yale University Press.
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. (1989). Herbicides and Plant Metabolism, edited by A.
Dodge, pp. 97±112. Cambridge University Press.
Collaborative Computational Project, Number 4 (1994). Acta Cryst.
D50, 760±763.
Davies, G. M., Barrett-Bee, K. J., Jude, D. A., Lehan, M., Nichols,
W. W., Pinder, P. E., Thain, J. L., Watkins, W. J. & Wilson, R. G.
(1994). Antimicrob. Agents Chemother. 38, 403±406.
Gu, Y., Reshetnikova, L., Li, Y., Wu, Y., Yan, H., Singh, S. & Ji, X.
(2002). J. Mol. Biol. 319, 779±789.
Gu, Y., Reshetnikova, L., Li, Y., Yan, H., Singh, S. V. & Ji, X. (2001).
Acta Cryst. D57, 1870±1871.
Hasemann, C. A., Istvan, E. S., Uyeda, K. & Deisenhofer, J. (1996).
Structure, 4, 1017±1029.
Idziak, C., Price, N. C., Kelly, S. M., Krell, T., Boam, D. J., Lapthorn,
A. J. & Coggins, J. R. (1997). Biochem. Soc. Trans. 25, S627.
Jencks, W. P. (1975a). In Catalysis in Chemistry and Enzymology. New
York: Dover.
Jencks, W. P. (1975b). Adv. Enzymol. 43, 219±410.
Koradi, R., Billeter, M. & Wu
È
thrich, K. (1996). J. Mol. Graph. 14, 51±
55.
Kraft, L., Sprenger, G. A. & Lindqvist, Y. (2002). J. Mol. Biol. 318,
1057±1069.
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. R. (2001). Protein Sci. 10, 1137±1149.
Laskowski, R. A., MacArthur, M. W., Smith, D. K., Jones, D. T.,
Hutchinson, E. G., Morris, A. L., Naylor, D., Moss, D. S. &
Thornton, J. M. (1994). PROCHECK v.3.0. Program to Check the
Stereochemistry Quality of Protein Structures. Operating Instruc-
tions. http://www.ccp14.ac.uk/ccp/ccp14/ftp-mirror/llnlrupp/
Procheck_NT/manual.txt.
Leipe, D. D., Koonin, E. V. & Aravind, L. (2003). J. Mol. Biol. 333 ,
781±815.
Leslie, A. G. W. (1992). Jnt CCP4/ESF±EACBM Newsl. Protein
Crystallogr. 26.
McRee, D. E. (1999). J. Struct. Biol. 125, 156±165.
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.
Parish, T. & Stoker, N. G. (2002). Microbiology, 148 , 3069±3077.
Pereira, J. H., Canduri, F., de Oliveira, J. S., da Silveira, N. J. F., Basso,
L. A., Palma, M. S., de Azevedo, W. F. & 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.
Romanowski, M. J. & Burley, S. K. (2002). Proteins Struct. Funct.
Genet. 47, 558±562.
Smith, C. A. & Rayment, I. (1996). Biophys. J. 70, 1590±1602.
Vonrhein, C., Schlauderer, G. J. & Schulz, G. E. (1995). Structure, 3,
483±490.
Walker, J. E., Saraste, M., Runswick, M. J. & Gay, N. J. (1982). EMBO
J. 1, 945±951.
research papers
Acta Cryst. (2004). D60, 2310±2319 Pereira et al.
Shikimate kinase 2319
electronic reprint
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexo 2
1
Shikimate Kinase: A potential target for development
of novel anti-tubercular agents.
J.H Pereira
†
, I.B Vasconcelos
§
, J.S.Oliveira
§
, L.A. Basso
‡*
and D.S. Santos
§*
†
Department of Physics, University of São Paulo State, São José do Rio Preto, SP 15054-
000, Brazil.
§
Centro de Pesquisas em Biologia Molecular e Funcional, Faculdade de Farmácia, Instituto
de Pesquisas Biomédicas, Pontifícia Universidade Católica do Rio Grande do Sul, Porto
Alegre, RS, Brasil.
‡
Centro de Biotecnologia, Instituto de Biociências, Universidade Federal do Rio Grande do
Sul, Porto Alegre, RS, Brasil.
* to whom correspondence may be addressed
LAB: phone, +55-51-33166234; fax, +55-51-3316-7309; e-mail: labasso@cbiot.ufrgs.br
DSS: phone, +55-51-33203629; fax, +55-51-3320-3629; e-mail: [email protected]
Key Words: tuberculosis, malaria, shikimate kinase, shikimate pathway, drug design, drug
target, Mycobacterium tuberculosis, Plasmodium sp
2
ABSTRACT
Tuberculosis (TB) remains the leading cause of mortality due to a bacterial pathogen,
Mycobacterium tuberculosis. However, no new classes of drugs for TB have been developed
in the past 30 years. There is an urgent need to developing faster acting and effective new
anti-tubercular agents, preferably belonging to new structural classes, to better combat TB,
including MDR-TB, to shorten the duration of current treatment to improve patient
compliance, and to provide effective treatment of latent tuberculosis infection. The enzymes
in the shikimate pathway are potential targets for development of a new generation of anti-
tubercular drugs. The shikimate pathway has been shown by disruption of aroK gene, which
codes for the shikimate kinase enzyme, to be essential for the Mycobacterium tuberculosis.
The shikimate kinase (SK) catalyses the phosphorylation of the 3-hydroxyl group of
shikimic acid (shikimate) using ATP as a co-substrate. SK belongs to family of nucleoside
monophosphate (NMP) kinases. The enzyme is an α/β protein consisting of a central sheet of
five parallel β-strands flanked by α-helices. The shikimate kinases are composed of three
domains: Core domain, Lid domain and Shikimate-binding domain. The Lid and Shikimate-
binding domains are responsible for large conformational changes during catalysis. More
recently, the precise interactions between SK and substrate have been elucidated, showing
the binding of shikimate with three charged residues conserved among the SK sequences.
The elucidation of interactions between MtSK and their substrates is crucial for the
development of a new generation of drugs against tuberculosis through rational drug design.
3
ABBREVIATIONS LIST
TB, tuberculosis; Mt, Mycobacterium tuberculosis; SK, shikimate kinase; DAHPS, 3-deoxy-
D-arabino-heptulosonate-7-phosphate synthase; CS, chorismate synthase; EPSPS, 5-
enolpyruvylshikimate-3-phosphate synthase; PABA, p-aminobenzoate; GTP, guanosine
triphosphate; ATP, adenosine triphosphate; ADP, adenosine diphosphate; Km, Michaellis-
Menten constant; NMP, nucleoside monophosphate; AK, adenylate kinase; WHO, World
Health Organization; HIV, human immunodeficiency vírus; MDR, multidrug-resistant;
DHQ, dehydroquinase; PCR, polymerase chain reaction; IPTG, isopropyl-beta-D-
thiogalactoside; Mg, magnesium; r.m.s., root mean square; Erc, Erwinia chrysantemi; AMP-
PCP, adenosine -β,γ-methyleneadenosine 5'-triphosphate; AMP-PCP, β,γ-imidoadenosine 5'-
triphosphate.
4
Fig. 1
1. THE SHIKIMATE PATHWAY
The shikimate pathway was discovered as a biosynthetic route through the studies of
Bernhard Davis and David Sprinson and their collaborators [1,2]. The shikimate pathway
links metabolism of carbohydrates to biosynthesis of aromatic compounds through seven
metabolic steps, where phosphoenolpyruvate and erythrose 4-phosphate are converted to
chorismic acid (Fig. 1) [3,4]. Chorismic acid is a common precursor for the synthesis of
aromatic compounds, such as aromatic amino acids, folate, ubiquinone and menaquinones.
Amongst them, the only ones that can be synthesized by humans are tyrosine, which is
synthesized from phenylalanine through a reaction catalyzed by phenylalanine hidroxylase
enzyme, and ubiquinone from tyrosine through a cascade of eight aromatic precursors [5].
The molecular organization of the shikimate pathway enzymes varies between
taxonomic groups [6]. Bacteria have seven individual polypeptides, which are encoded by
separate genes. Plants have a molecular arrangement similar to bacteria [7], with the
exception of dehydroquinase (DHQase, third enzyme) and shikimate dehydrogenase (fourth
enzyme) which have been shown to be present as separate domains on a bifunctional
polypeptide [8]. In fungi and apicomplexan parasites (Toxoplasma gondii) the shikimate
pathway has been shown to include monofunctional 3-deoxy-D-arabino- heptulosonate 7-
phosphate (DAHP) synthase and chorismate synthase (CS) enzymes and a pentafunctional
polypeptide termed AROM, which accounts for the remaining five shikimate pathway
reactions [9].
The shikimate pathway enzymes are attractive targets for development of non-toxic
anti-bacterial [10] and herbicides [11], because this pathway is essential for algae, higher
plants, bacteria, fungi, whereas it is absent from mammals [12]. Thus, in the case of bacterial
diseases, inhibition of any of shikimate pathway enzymes is unlikely to cause toxic side
5
effects on the host. In addition, the importance of shikimate pathway can be indicated by the
finding that deletion of the aroA gene, which codes EPSPS, causes Streptomyces
pneumoniae and Bordetella bronchiseptica strains to be attenuated for virulence [13,14].
The shikimate pathway has also been discovered in apicomplexan parasites providing
several targets for the development of new anti-parasite drugs [15]. In vitro growth of
apicomplexan parasites such as Plasmodium falciparum (malaria), Toxoplasma gondii
(toxoplasmosis), and Cryptosporidium parvum (cryptosporidiosis) was inhibited by the
herbicide glyphosate, a well-characterized inhibitor [16] of the shikimate pathway enzyme 5-
enolpyruvyl shikimate 3-phosphate synthase (EPSPS), at concentration of 1-6 mM [15]. In
P. falciparum and T. gondii, this inhibitory effect was reversed by co-addition o p-
aminobenzoate (PABA) or folate suggesting that this pathway is essential for the
biosynthesis of folate precursors. Moreover, several shikimate pathway enzymes has been
detected in T. gondii and P. falciparum extracts [15,17] and a gene encoding a
pentafunctional polypeptide AROM has been described in T. goondii indicating that a
complete shikimate pathway is present in this parasite [18]. Whereas the shikimate pathway
gives an array of compounds in bacteria, fungi and plants, only its folate biosynthesis
function has been established is in aplicomplexa parasites. All folate pathway enzymes,
which convert GTP to derivatives of tetrahydrofolate, were found in Apicomplexa when the
complete genome was sequenced [19].
Two shikimic acid analogs, 6-S-fluorshikimate and 6-R-fluorshikimate, have been
shown to inhibit P. falciparum growth and inhibition shown to be specific to the shikimate
pathway [20]. Despite the completion of P.falciparum genome sequence [21], only a single
gene encoding the chorismate synthase enzyme and a potential bifunctional SK/EPSP
synthase protein has been identified in the genome annotation [22]. The coding DNA
sequence of P. falciparum chorismate synthase has been cloned and the protein has been
6
shown to be located on cytosol by immunological studies [23]. Moreover, chorismate
synthase, which catalyzes the last step of shikimate pathway, has been shown to be required
for Plasmodium falciparum growth, as disruption of expression by RNA interference
decreased parasite growth [19]. It has been proposed that the missing shikimate pathway
enzymes are either substituted by non-homologous enzymes that catalyze the same reaction
or that the enzymes are homologous but too divergent to be identified readly [22].
In mycobacteria, the chorismic acid intermediate is a precursor for the synthesis of
naphthoquinones, menaquinones, and mycobactins, besides aromatic amino acids [24]. The
salycilate-derived mycobactins siderophores have been shown to be essential for M.
tuberculosis growth in macrophages [25]. Particularly, in Mycobacterium tuberculosis, the
shikimate pathway has been shown to be essential for the bacterial viability. The disruption
of aroK gene, which codes for the shikimate kinase enzyme (SK), was only possible when
the second functional copy of aroK was integrated into the chromosome. Moreover, excision
of the second integrated copy of aroK by the L5 excisionase could not be achieved in a M.
tuberculosis strain carrying the disrupted copy of aroK gene, but was possible in a strain
carrying a wild-type copy [26].
2. SHIKIMATE KINASE
The shikimate kinase (SK; EC 2.7.1.71), the fifth enzyme of the pathway, catalyses
the specific phosphorylation of the 3-hydroxyl group of shikimic acid (shikimate) using ATP
as a co-substrate (Fig. 1). In Escherichia coli, the SK reaction is catalyzed by two isoforms:
SK I encoded by the aroK gene [27] and SK II encoded by the aroL gene [28]. The major
difference between the isoenzymes is their K
m
for shikimate, 20 mM for the SK I and 0.2
mM for the SK II enzyme [29]. The SK II isoform appears to play a dominant role in the
shikimate pathway, its expression is controlled by the tyrR regulator, and it is repressed by
7
tyrosine and tryptophan [30,31]. The physiological role of SK I in E.coli is not clear. Since
mutations in SK I are associated with sensitivity to the antibiotic mecillinam [32] it has been
suggested that SK I may have an alternative biological role that is distinct and unrelated to
its shikimate kinase activity [29]. As pointed out by Parish and Stoker [26], if M.
tuberculosis aroK-encoded SK I possess a similar activity it is possible that disruption of this
activity can account for the observed inability of M. tuberculosis to grow in the absence of a
functional copy of aroK gene. However, the actual nature of second activity of aroK gene
product remains to be established. Contrary to the presence of isoenzymes in E. coli,
complete genome sequences of a number of bacteria, for example, Haemophilus influenzae
and Mycobacterium tuberculosis, have revealed the presence of only one SK-coding gene.
Most of these SKs appear to be encoded by aroK rather than aroL because their amino acid
sequences have higher degree of identity with E.coli SK I. The kinetic parameters for aroK-
encoded M.tuberculosis SK (MtSK) are more similar to those of aroL-encoded E.coli SK II
than to those of aroK-encoded E.coli SK I. Thus, the MtSK K
m
value (0.41mM) for
shikimate suggests that not all aroK-encoded SKs have high K
m
values for shikimate [33].
2.1 THE ENZYME FOLD
Shikimate kinase displays an
α
/β-fold and consists of five central parallel β-sheet
with the strand order 23145, flanked by α-helices [34](Fig. 2). The three crystal structures of
SK from Erwinia chrysanthemi (ErcSK) [34,35] showed that SK belongs to the same
structural family of nucleoside monophosphate (NMP) kinases for which structures are
known for adenylate kinase (AK) [36,37], guanylate kinase [38], uridylate kinase [39] and
thymidine kinase [40].
The NMP kinases are composed of three domains: CORE, LID and NMP-binding
(NMPB) domains [41]. A characteristic feature of the NMP kinases is that they undergo
8
Fig.2
large conformational changes during catalysis, for which AK is the most extensively studied
[41]. There are two flexible regions of the structures that are responsible for movement: one
is the NMP-binding site which is formed by a series helices between strands 2 and 3 of
parallel β-sheet and the other is the LID domain, a region of varied size and structure
following the fourth β-strand of the sheet [42,43]. In SK, the shikimate-binding (SB) domain
corresponds to the NMPB domain of NMP kinases.
2.2 FUNCTIONAL MOTIFS
Three functional motifs of nucleotide-binding enzymes are recognizable in shikimate
kinase, including a Walker A-motif (A-motif), a Walker B-motif (B-motif), and an adenine-
binding loop. The Walker A-motif is located between the first β-strand (β1) and first α-helix
(α1), containing the GXXXXGKT/S conserved sequence [44], where X represents any
residue. This motif forms the phosphate-binding loop (P-loop), a giant anion hole which
accommodates the β-phosphate of the ADP by donating hydrogen bonds from several
backbone amides [45]. The side chain of P-loop lysine have a catalytic role of stabilizing the
pentavalent transition state of the γ-phosphoryl group as has been shown for adenylate
kinase [46] and p21
ras
[47].
In addition to the Walker A-motif, it is observed a second conserved sequence ZZDXXG
called Walker B-motif [44], where Z represents a hydrophobic residue. More recently,
however, sequence and structural comparisons for all P-loop-fold proteins classified
Shikimate Kinase in the DxD group of enzymes [48], which has a conserved DxD motif in
strand 2. The Walker B motif consensus in shikimate kinases is ZZZTGGG and the second
glycine has been implicated in hydrogen bonding to the γ phosphate of ATP. This motif is
located on the C-terminal segment of the third strand (β3) of the central β-sheet The adenine-
9
binding loop motif may be described as a sequence stretch of I/VDXXX(X)XP [33]. This
motif forms a loop that wraps around the adenine moiety of ATP, connecting the β5-strand
with the C-terminal α-helix.
3. TUBERCULOSIS AND Mycobacterium tuberculosis SHIKIMATE KINASE
Tuberculosis (TB) remains the leading cause of mortality due to a bacterial pathogen,
Mycobacterium tuberculosis. The interruption of centuries of decline in case rates of TB
occurred, in most cases, in the late 1980s and involved the USA and some European
countries due to increased poverty in urban settings and the immigration from TB high-
burden countries [49]. Thus, no sustainable control of TB epidemics can be reached in any
country without properly addressing the global epidemic. It is estimated that 8.2 million new
TB cases occurred worldwide in the year 2000, with approximately 1.8 million deaths in the
same year, and more than 95 % of those were in developing countries [50]. Approximately 2
billion individuals are believed to harbor latent TB based on tuberculin skin test surveys
[51], which represents a considerable reservoir of bacilli. According to a recent report
compiled by the World Health Organization (WHO), the total number of new cases of
tuberculosis worldwide in 2002 had risen to approximately 9 million [52]. A key driver of
the increase is the synergy with the HIV epidemic, which is having a devastating impact on
some parts of the world mostly in the African Region, where 31% of new TB cases were
attributable to HIV co-infection [50]. Another problem is the proliferation of multi-drug
resistant (MDR) strains, defined as resistants to at least isoniazid and rifampicin, which are
the most effective first-line drugs [53]. According to the 2004 Global TB Control Report of
the World Health Organization, there are 300,000 new cases per year of MDR-TB
worldwide, and 79 % of MDR-TB cases are now “super strains”, resistant to at least three of
the four main drugs used to treat TB [52]. The factors that most influence the emergence of
10
drug-resistant strains include inappropriate treatment regimens, and patient noncompliance
in completing the prescribed courses of therapy due to the lengthy standard “short-course”
treatment or when the side effects become unbearable [54]. No new classes of drugs for TB
have been developed in the past 30 years, reflecting the inherent difficulties in discovery and
clinical testing of new agents and the lack of pharmaceutical industry in investing money
and manpower for research in the area [55]. Hence, there is an urgent need to developing
faster acting and effective new anti-tubercular agents, preferably belonging to new structural
classes, to better combat TB, including MDR-TB, to shorten the duration of current
treatment to improve patient compliance, and to provide effective treatment of latent
tuberculosis infection [53].
In M.tuberculosis, the presence of the genes involved in the shikimate pathway began
to be elucidated in the early 90s, with the cloning and characterization of aroA gene product,
which codifies the EPSPS synthase [56]. Nevertheless, the complete genome sequence from
M.tuberculosis, strain H37Rv reported by Cole et al. in 1998 [57] allowed the identification
by sequence homology of all genes coding for shikimate pathway enzymes.
Four homologues to the shikimate pathway enzymes were located in a cluster
containing the aroD-encoded type II DHQ dehydratase (Rv2537c), aroB-encoded DHQ
synthase (Rv2538c), aroK-encoded type I shikimate kinase (Rv2539c), and aroF-encoded
chorismate synthase (Rv2540c). The remaining homologues to shikimate pathway enzymes
were annotated as follows: aroG-encoded class II phenylalanine-regulated DAHPS
(Rv2178c), aroE-encoded shikimate dehydrogenase (Rv2552c), and aroA-encoded EPSP
synthase (Rv3227).
The aroK structural gene is composed by 531 pb and codes a protein of 176 amino
acids. The theoretical molecular mass of MtSK enzyme subunit is 18.58 KDa. The first
report of cloning and overexpression in soluble and functional form of MtSK occurred in
11
2001 [58], where has been reported the PCR amplification aroK gene from genomic DNA of
M. tuberculosis H37Rv strain, cloning in the plasmid pET-23a(+), and overexpression of
aroK-encoded MtSK protein in Escherichia coli BL21(DE3) host cells, without IPTG
induction.
The crystal structure of a protein complexed with its substrates is of crucial
importance for the rational design of inhibitors that target the enzyme. Although two crystal
structures of SK from M.tuberculosis [33] have revealed the dynamic role of LID Domain in
catalysis and position in the structure of ligands Mg
2+
and ADP, the precise positions and
interactions between shikimate and MtSK was not demonstrated because the shikimate-
binding site was not occupied by the substrate or the electron density was not sufficient to
position the shikimate molecule. In case of SK enzymes, a likely drawback of ATP-binding-
site-based SK inhibitors would be their lack of specificity, owing to the common fold and
similar ATP-binding site shared by many P-loop kinases [48]. Hence, trying to obtain a
description of molecular interaction between MtSK and shikimate, molecular-modeling and
docking studies has been carried out and its results reported [59]. Despite these studies failed
to predict all molecular interactions between MtSK and shikimate [60], Asp34 and Arg136
residues and its hydrogen bonds implicated on shikimate binding are in agreement with the
crystal structure of MtSK-MgADP-Shikimate ternary complex reported afterwards [61]. The
crystal structure of MtSK-MgADP-Shikimate was the first crystallographic structure of SK
with bound shikimate deposited in Protein Data Bank (PDB access code: 1WE2) [61].
Crystals were obtained by the hanging-drop vapour-diffusion method, and larger final
concentration of ADP and shikimate in the drop (8.0mM) were used than those used by Gu
et al. (2002) to obtain the crystals (4.0mM), in order to accurately determine the position of
shikimate binding in the active site of MtSK. The structure has been determined at 2.3 Å
resolution, clearly reveling the amino-acids residues involved in shikimate binding. The
12
molecular replacement method was used, using as a search model the structure of MtSK-
MgADP [33]. Almost at the same time of publishing of our article describing the structure of
MtSK-MgADP-Shikimate ternary complex, another article describing a similar structure of
the ternary complex has been published (PDB access code: 1U8A) [60]. The crystal
structures of MtSK-MgADP-Shikimate [61] and MtSK-ADP-Shikimate [60] ternary
complexes have unequivocally revealed in detail the interactions of amino acid residues with
bound shikimate and conformational changes upon substrate binding.
3.1 ADP/Mg
2+
INTERACTION
Essentially all kinases require a Mg
2+
-nucleotide complex as one of the enzyme
substrates [62], an exception being the first partial reaction catalyzed by nucleoside
diphosphate kinase, which can proceed independently of Mg
2+
[63]. Nucleophilic attack on
the γ-phosphate group of ATP will be most facilitated by meta-ion binding (Mg
2+
) to the β-
and γ-phosphoryl groups, whereas departure of the leaving group will be most favoured in a
structure with metal binding to the α- and β-phosphoryl groups [64]. Presumably Mg
2+
also
assists in orienting the γ-phosphoryl group of ‘inline’ with respect to the second substrate,
creating the correct geometry to complete phosphoryl transfer [62]. The binding of this
cation in many P-loop proteins such as myosin [65], elongation factor EF-Tu [66], p21-Ras
[67] and the heterotrimeric G-proteins [68] involves hexa-coordination of the Mg
2+
by two
oxygen atoms (from the β and γ-phosphates of the bound nucleotide), two water molecules,
and two protein ligands. As in the MtSK-MgADP structure (PDB code 1L4Y) [33], a typical
six-coordination has been observed for Mg
2+
in the MtSK-MgADP-Shikimate structure, with
some minor differences between the position of Mg
2+
in the structures. In
the MtSK-
MgADP-Shikimate structure Mg
2+
interacts with a β-phosphate oxygen of ADP, Ser16 OG
13
of the Walker A motif and four water molecules [61]. Superimposition of shikimate-free
MtSK-MgADP structure on shikimate-complexed MtSK-MgADP structure showed that in
the ternary complex water1 and water4 are in equivalent positions but that shifts of 1.32 and
2.75 Å are observed for water2 and water3, respectively. In the MtSK-MgADP structure, the
interaction between Asp32 and Ser16 of the Walker A motif is via a bridging water molecule
(water6), whereas in the ternary complex structure this interaction occurs directly via a
hydrogen bond between the two residues [61], which accounts for the exclusion of water6
from the magnesium-binding site in the ternary structure. In the ternary complex, the water1
molecule coordinated to the Mg
2+
interacts directly with the chloride ion instead of
interacting with Asp34 via a bridging water molecule (water5) as observed in the MtSK-
MgADP structure [33]. This chloride ion also interacts with the 3-hydroxyl group of
shikimate and the backbone amide of Gly80 [61]. Thus, the different mode of interaction
observed for residue Asp34 arises from the presence of shikimate, which leads to the
exclusion of water5 from MtSK active site [61]. The Mg
2+
cation was not included in the
final structure of MtSK-ADP-Shikimate ternary complex reported by Dhaliwal et. al.(2004),
since the best diffracting crystals grew in the absence of MgCl
2
[60]. Accordingly, the
chloride ion observed in the active sites of both MtSK binary complex [33] and MtSK-
MgADP-Shikimate ternary complex [61is absent in the Mg
2+
-free structure of MtSK-ADP-
Shikimate ternary complex [60]. Thus, the absence of MgCl
2
in crystallization mixture
probably have accounted for the small differences observed in conformation, position and
molecular interactions of shikimate when the structure reported by Dhaliwal et al. is
compared with MtSK-MgADP-Shikimate-Cl structure [61]. In Mg
2+
-free MtSK structure the
3-hydroxyl group of shikimate is closer to β-phosphoryl group of ADP than it is in the
MtSK-MgADP-Shikimate structure. Furthermore, two additional water-mediated
interactions between protein residues and shikimate has been shown. The NZ atom of Lys15
14
residue forms a 2.6 Å hydrogen-bond with oxygen atom of another water molecule, which in
turn interacts with the 3-hydroxyl group of shikimate, and the main-chain nitrogen of
Leu119 residue forms a 2.9 Å hydrogen bond with the oxygen atom of a water molecule,
which in turn interacts with 5-hydroxyl group of substrate [60]. In the MtSK-MgADP-
Shikimate structure, Lys15 cannot form a interaction with 3-hydroxyl group of shikimate,
since the chloride ion is bound to the active site cavity between them. The distance between
NZ atom of Lys15 and the chloride ion is 3.94 Å, which in turn forms a 3.36-Å hydrogen
bond with 3-hydroxyl group of shikimate [61].
In the MtSK-MgADP structures [33], Lys15 forms a hydrogen bond with a β-
phosphate O1B atom and the chloride ion. The main-chain NH of Gly80 is hydrogen bonded
to the chloride ion the binary (1L4Y) [33] and ternary complex structures [60,61]. These
residues are located in vicinity of where the chemical reaction occur a may thus play a
critical role in the transition state stabilization.
The molecular interactions that describe the ADP binding mode on enzyme are very
similar in all available structures of MtSK. The adenine moiety of ADP is sandwiched
between Arg110 and Pro155 [33] and this interaction has also been observed in ErcSK [34],
and in Adenilate kinase [69] and isoenzyme II [70]. Arg110 is located at the C-terminus of
α6 where the LID domain starts. In MtSK, Arg110 and Arg117 residues represent,
respectively, the first and the last residue of a conserved motif of LID domain observed for
P-loop kinases (typically RXX(X)R) [48]. In P-loop shikimate kinases the conserved motif
of the LID domain has been proposed to be R(X)
6-9
R [61]. The Arg117 has been shown to
interact with the α− and β-phosphate groups of ADP [33,60,61], it generally interacts with
the γ-phosphate of ATP bound to enzymes, and it may stabilize the transition state by
neutralizing the developing negative charge on the β-γ bridge O atom [71]. The Pro155 is
the last residue of the adenine-binding loop motif (residues 148-155 in MtSK), which was
15
first recognized in Adenilate kinase and ErcSK [34] and has been described as an
I/VDXXX(X)XP sequence stretch [33]. However, the second (aspartate) residue and the last
(proline) rsidue are not conserved in the adenine-binding loops of aroK-encoded SKs from
Escherichia coli [72] and Campylobacter jejuni (PDB acess code: 1VIA). However, they are
recognizable from structural alignment analysis using MtSK-MgADP as a reference
structure [61]. In fact, it has been pointed out that the main-chain contacts with the adenine
base and the presence of a structural motif independent of sequence may be more important
for adenine binding in kinases [62]. The adenine-binding loop motif (residues 148-155 in
MtSK) forms a loop that wraps around the adenine moiety of ATP, connecting the β5-strand
with the C-terminal α8-helix. In addition, Lys15 forms a hydrogen bond with a β-phosphate
oxygen atom [33].
For catalysis, the three protein interacting residues Lys15, Arg117, and Arg136 has
been proposed to be the most important [33]. Consistent with this proposal, the K15M
mutant of ErcSK showed no detectable enzyme activity [35], although it was in fact a
K15M/P115L double mutant. Contrary to the previously proposed, analysis of the MtSK
ternary complexes have revealed that Lys15 and Arg117 are the only positively charged
residues located in the vicinity of where the reaction may occur and may therefore play
critical roles in the stabilization of the transition state [33]. The Arg136 residue, instead,
appears to interact with the carboxyl group of shikimate and probably, it is not involved in
catalysis [60,61].
3.2 INTERACTION WITH SHIKIMATE
The shikimate-binding domain, which follows strand β2, consists of helices α2 and
α3 and the N-terminal region of helix α4 (Fig. 2). In particular for MtSK, the precise
interactions between shikimate and SK have been elucidated, showing that the guanidinium
16
Fig. 3
groups of Arg58 and Arg136 and the NH backbone group of Gly81 interact with the
carboxyl group of shikimate. The 3-hydroxyl group of shikimate forms hydrogen bonds with
the carboxyl group of Asp34 and the main-chain NH group of Gly80 and a water molecule.
The 2-hydroxyl group of shikimate hydrogen bonds to the side chain of Asp34 (Fig. 3B).
The Glu61 is conserved in both aroK and aroL-encoded shikimate kinase enzymes. This
residue is not directly involved in substrate binding, but it forms a hydrogen bond and a salt
bridge with the conserved Arg58 and assists in positioning the guanidinium group of Arg58
for shikimate binding. In the ternary structure, Glu61 side chain also forms a water-mediated
interaction with the 3-hydroxyl group of shikimate. Therefore, Glu61 plays an important role
in the substrate-binding site [61]. The Glu54 carboxylate group also appears to anchor the
guanidinium group of Arg58 for interaction with the shikimate carboxylate group; however,
Glu54 is not conserved, except for aroK-encoded shikimate kinases [61]. In the MtSK-ADP-
shikimate structure [60], the main-chain nitrogen of Leu119 forms a 2.9 A H-bond with the
water3 oxygen atom, which in turns interacts with the 1-hydroxyl group of shikimate, and
the NZ atom of Lys15 forms a 2.6 A H-bond with the oxygen atom of water2, which also
interacts with the 3-hydroxyl group of the substrate.
3.3 CONFORMATIONAL CHANGES UPON SUBSTRATE BINDING
As pointed out above, NMP Kinases undergoes large conformational changes during
catalysis, because their LID domain and the NMP binding-site are very flexible and can
make large movements upon substrate binding. These structural changes act to position
enzyme sidechains appropriately around the substrates and to sequester the substrates so as
to prevent the hydrolysis of bound ATP or other phosphoryl-containing substrates prior to
catalysis [41,62,64]. In addition, it has also been shown that these two domains are capable
17
to make independently moves towards each other [41]. Previous studies have shown that
hexokinases [73] and adenylate kinases [42,74], which are classified as NMP kinases,
undergo a large conformational change during catalysis.
Several structural and spectroscopic studies have demonstrated that SKs undergo
conformational changes on ligand binding. Circular dichroism spectra of unliganded and
liganded ErcSK enzyme in the presence of 2mM shikimate or 2mM γN-ATP (an non-
hydrolisable ATP analogue) have shown that SK undergoes conformational changes upon
ligand binding [34]. Moreover, fluorescence studies were performed using a single
tryptophan residue (W54) as a report group, which is positioned close to the shikimate
binding site. The addition of shikimate to protein solution caused a quenching in ErcSK
protein fluorescence and a blue shift of 3 nm in the emission maximum, consistent with the
loop containing the tryptophan residue becoming more deeply buried within the protein
following ligand-binding [75]. In addition, the averaged B-factor for all residues in crystal
structure of ErcSK showed clear evidence of the flexibility of the molecule, where the
temperature factors for both ErcSK molecules in the asymmetric unit (one with bound
shikimate and other with unbound shikimate) indicate two regions of high mobility,
corresponding to the shikimate binding-site and the LID domain and its flanking regions
[34]. A comparison of the residue-averaged B factors between the crystal structures of
ErcSK-MgADP, MtSK-MgADP (1L4Y), and MtSK-MgADP Pt-derivative (1L4U) binary
complexes [33] shows that the enzyme from M. tuberculosis follows the same pattern of
flexibility in the LID and SB domains as previously observed for ErcSK.
Another method used to evaluate conformational changes in proteins is the superposition of
structures in order to compare different complexes of the same protein. To demonstrate
conformational changes upon shikimate binding in MtSK, alignment of C
α
positions of the
MtSK-MgADP-Shikimate dead-end ternary complex and the MtSK-MgADP binary complex
18
structures were made [61], showing that the LID and SB domains undergoes noticeable
concerted movements towards each other. In this alignment, which included all residues,
were verified r.m.s. deviation values of 0.56 and 0.54 Å for 1L4U and 1L4Y, respectively.
Residues 112-124, that comprises the LID domain, showed a r.m.s. deviation of 1.33 Å, and
the SB doamin shift was somewhat smaller, with an r.m.s. deviation value of 0.74 Å for
residues 33-61. Similar values were found when the same procedure was applied to the
binary complex and the MtSK-ADP-Shikimate structure [60], with a value of 0.7 Å for the
overall structure (excluding residues 114-115), 1.5 Å for the LID domain and 0.4 Å for the
SB domain. The residues directly involved in these movements are Val116, Pro118 and
Leu119 (LID domain), Ile45, Ala46, Glu54, Phe57 and Arg58 (SB domain), where its side
chains shifted upon shikimate binding [61]. A comparison of MtSK-ADP-Shikimate ternary
complex to MtSK-MgADP binary complex showed, within of SB region, a shift of 0.9 Å of
Arg58 residue towards the carboxylate group of shikimate. Phe49 residue moves
approximately ~1.7 Å away from Phe57 and closer to the substrate, translating towards
shikimate [60]. In addition, it has been proposed by Dhaliwal et al. (2004) that the observed
changes in the orientation and position of Phe49 and Phe57 residues disrupt their strong ring
stacking interactions is probably a result from the Van Der Waals contacts made between
shikimate with both phenylalanines [60]. However, in MtSK-MgADP-Shikimate the strong
ring stacking interaction between the phenylalanines is not disrupted in spite of the shift of
Phe49 and Phe57 residues towards shikimate of, 0.49 Å and 0.79 Å, respectively. Moreover,
there is a reduction in molecular surface area value the ternary complex compared to binary
complex. The calculated value for MtSK complexed with MgADP (1L4U) was
approximately 7246 A
2
and for MtSK complexed with MgADP and shikimate a value of
6915 A
2
was found. Based on this results, the difference between the complexes molecular
surface areas is 330 A
2
, which are buried on shikimate binding (Fig. 4) [61]. An reduction in
19
Fig. 4
molecular surface area value has also been observed in the Mg
2+
-free crystal structure of
MtSK-ADP-Shikimate ternary complex (Fig. 4). Nevertheless, in this complex a molecular
surface of only 206 A
2
is buried on shikimate binding, indicating that Mg
2+
binding has a
role in the closure of MtSK active site. It is important to point out that both the MtSK-
MgADP bynary complexes [33] and MtSK-MgADP-Shikimate [61] and MtSK-ADP-
Shikimate [60] ternary complexes have been crystallized in the same space group with
similar unit-cell parameters and residues from LID and SB domains form no crystal contacts
with symmetry-related MtSK molecules. Thus, the conformational changes that has been
described for MtSK cannot merely be a reflection of the different crystal-packing
arrangements.
The incomplete LID-domain closure observed in crystal structures of both MtSK-
MgADP-Shikimate [61] and MtSK-ADP-Shikimate [60] ternary complexes suggests that γ-
phosphate of ATP is necessary for the completion of the domain movement [34]. This find is
not unexpected, since total active-site closure upon dead-end ternary complex formation
would result in locking the enzyme active site in an inactive form in which shikimate
substrate binding to MtSK enzyme prior to MgADP dissociation from its active site would
result in an inactive abortive complex. Several previously reported results are consistent with
these structural. Measurements of ErcSK intrinsic tryptophan fluorescence (Trp54) on
shikimate binding to either ErcSK or the ErcSK-MgADP binary complex showed a modest
synergism of binding between these substrates, since the dissociation constant value for
shikimate (K
d
= 0.72 mM) decreased to 0.3 mM in the presence of 1.5 mM ADP [75]. In
addition, measurements of the quenching of protein fluorescence of the aroL-encoded
ErcSK upon nucleotide binding demonstrated the dissociation-constant values for ADP and
ATP to be 1.7 and 2.6 mM, respectively [35]. The Km for ATP (620 µM) was found to be
20
approximately four times lower than the dissociation constant in the absence of shikimate
[35]. These results prompted the proposal that the conformational changes in the ErcSK
enzyme associated with the binding of the first substrate led to an increase in the affinity for
the second substrate. However, this synergism on substrate binding does not appear to hold
for MtSK since the apparent dissociation-constant values for ATP (89µM) and shikimate
(440 µM) are similar to their K
m
values, 83 µM for ATP and 410 µM for shikimate
considering either a rapid equilibrium random-order bi-bi enzyme mechanism or a steady-
state-ordered bi-bi enzyme mechanism [33]. Moreover, no evidence for synergism between
shikimate and ATP could be observed in substrate binding to ErcSK in a chloride buffer
system [76].
The K15M ErcSK mutant has been crystallized in an open conformation that is
proposed to presumably be equivalent to an apo-enzyme structure in which neither ADP (or
ATP) neither shikimate would be bound [35]. This mutant was produced to evaluate the role
of the conserved Lys15 of the Walker A motif. However, an unwanted mutation point
mutation in the LID domain (Pro115Leu) was detected during the refinement of the model.
Furthermore, the enzyme was crystallized with two independent molecules in the
asymmetric unit and extensive contacts of neighboring LID domains lead to a stabilization
of this part of the molecule that is not visible in the native crystal structure [34,35]. It
therefore appears unwarranted to consider the double K15M/P115L ErcSK mutant a model
for the apo enzyme. Since a better model for the apo-enzyme is not currently available, here
we considered the double K15M/P115L ErcSK mutant as a model for the apo-SK. The
superposition of MtSK-MgADP-Shikimate and apo ErcSK (K15M+P115L) show the large
conformational changes in the LID and SB domains associated with the binding of ADP and
shikimate on MtSK (Fig. 5).
21
Fig. 5
3. CONCLUDING REMARKS
The disruption of gene arok, which codes for the shikimate kinase, has been shown to
be essential for the viability of M. tuberculosis. The evidence that shikimate pathway is
essential for M. tuberculosis even in the presence of exogenous supplements as p-
aminobenzoate, p-hydroxibenzoate and aromatic amino acids reinforces its attractiveness as
a drug target [26]. It is interesting to note that shikimic acid can be used to supply the
aromatic amino acid and aromatic vitamin requirements of E. coli blocked in any of the first
three steps of the shikimate pathway [77]. This ability has been related to the presence of the
shikimic acid transport system encoded by the gene shiA [78]. Thus, there is the possibility
that shikimic acid could be used as a supplement to allow the growth in vitro of mutant M.
tuberculosis strains disrupted on aroG, aroB, aroD or aroE genes, which code for first four
shikimate pathway enzymes in M. tuberculosis. These aro-disrupted M. tuberculosis strains
could be used in experiments to determine if shikimate pathway is essential to in vivo growth
of M. tuberculosis.
A comprehensive structural picture of the interactions between M. tuberculosis SK
enzyme and Mg
2+
, ADP and shikimate substrates have been obtained from crystallographic
studies [33,60,61]. In addition, the LID and SB domains conformational changes upon ADP
and shikimate substrates binding, which result in a partial closure of and solvent expulsion
from MtSK active site, has been described. Probably, a complete active-site closure would
be achieved in a complex of MtSK with shikimate, Mg
2+
and a non-hydrolysable ATP
analogue such as adenosine -β,γ-methyleneadenosine 5'-triphosphate (AMP-PCP) or β,γ-
imidoadenosine 5'-triphosphate (AMP-PCP). These ATP analogues have been used in
22
crystallographic studies for obtaining closed complexes with E. coli Gluconate Kinase [79]
and 3-Phosphoglycerate Kinase enzymes [80].
There has been considerable debate within the literature as to whether enzyme-
catalyzed phosphoryl transfer reactions operate primarily with an associative (S
N
2-like) or
dissociative (S
N
1-like) transition state [62]. A criterion that has been used for distinguishing
between these mechanisms is based on the geometry and reaction coordinate distances
between the terminal phosphoryl group of ATP and the acceptor substrate in ground state
enzyme-substrate or enzyme inhibitor complexes [81]. For the transition state to have some
degree of associative character, these distances are expected to be less than the sum of a P-O
van der Waals contacts and a P-O single bond (i.e in the order of approximately 4.9 Å) [62].
In dissociative mechanisms these distances would be expected to be longer than this to allow
space for the intermediate monomeric metaphosphate to exist between the leaving group
(ADP) and the entering group [81]. A crystal structure of MtSK in complex with MgADP,
Shikimate, and AlF
3
(a structural analog of γ-phosphoryl group transfered in the reaction)
could permit a snapshot of the transition-state like structure of MtSK. As it has been shown
to other kinases [82,83,84], this structural study could distinguish between associative and
dissociative transition states in the reaction catalyzed by MtSK.
Structures of SKs deposited in the protein data bank (PDB) were superimposed and
positions of residues involved in binding (Lys15, Asp34, Arg58, Glu61, Gly79, Gly80,
Gly81, Arg136) between shikimate and SK are highly conserved [61]. The precise
interactions between MtSK and shikimate have been elucidated, showing the residues
involved in the bind of substrate and the conformational changes upon substrate binding.
Accordingly, the availability of the M. tuberculosis shikimate kinase structures complexed
with shikimate provide crucial information for the design of non-promiscuous SK inhibitors
that target both the shikimate- and ATP-binding pockets or uniquely, the shikimate-binding
23
site. Moreover, the knowledge of functional factors that lead to active site closure could be
used for designing inhibitors that force MtSK to a closed conformation that would be unable
to catalyze the phosphoryl transfer to shikimate. These inhibitors could block the
biosynthesis of aromatic aminoacids and other compounds (folate, micobactins, etc), which
are essential for the growth and viability of the microorganism.
24
ACKNOWLEDGMENTS
Financial support for this work was provided by FAPESP (Proc. 02/05347-4) and
Millennium Initiative Program MCT-CNPq, Ministry of Health - Secretary of Health Policy
(Brazil) to DSS and LAB. DSS (304051/1975-06) and LAB (520182/99-5) are research
career awardeed from the National Research Council of Brazil (CNPq).
25
4. REFERENCES
[1] Davis, B.D. and Mingioli, E. S. (1953) J. Bacteriol. 66, 129–136.
[2] Sprinson, D.B. (1960) Adv. Carbohydrate Chem. 15, 235–270.
[3] Pittard, A.J. (1987) In F.C. Neidhardt, J.L. Ingraham, K.B. Low, B. Magasanik, M.
Schaechter, and H.E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular
and molecular biology, vol.1, pp. 368–394. American Society for Microbiology,
Washington, D.C.
[4] Haslam, E. (1993) Shikimic acid: metabolism and metabolites. John Wiley & Sons,
Chichester, United Kingdom.
[5] Folkers, K.(1996). Biochem. Biophys. Res. Commun. 224, 358-361.
[6] Coggins, J.R., Duncan, K., Anton, I.A., Boocock, M.R., Chaudhuri, S., Lambert, J.M.,
Lewendon, A., Millar, G., Mousdale, D.M., Smith, D.D., (1987) Biochem. Soc. Trans. 15,
754–759.
[7] Butler, J.R., Alworth, W.L., Nugent, M.J., (1974) J. Am. Chem. Soc. 96, 1617–1618.
[8] Mousdale, D.M., Campbell, M.S., Coggins, J.R. (1987) Phytochemistry 26, 2665–2670.
[9] Duncan, K., Edwards, R.M., Coggins, J.R. (1987) Biochem. J. 246, 375–386.
[10] Davies, G.M., Barret-Bee, K.J., Jude, D.A., Lehan, M., Nichols, W.W., Pinder, P.E. et
al. (1994). Antimicrob. Agents Chemother., 38, 403-406.
[11] Coggins, J.R. (1989). The shikimate pathway as a target for herbicides. In Herbicides
and Plant Metabolism (Dodge, A., ed.), Cambridge University Press, Cambridge, UK, pp.
97-112.
[12] Bentley, R. (1990). Crit. Rev. Biochem. Mol. Biol., 25, 307-384.
[13] McDevitt, D., Payne, D.J., Holmes, D.J., and Rosenberg, M. (2002) J. Appl. Microbiol.
92, 28S-34S.
26
[14] McArthur, J.D., West, N.P., Cole, J.N., Jungnitz, H., Guzman,C.A., Chin, J., Lehrbach,
P.R., Djordjevic, S.P., and Walker, M.J. (2003) FEMS Microbiol. Lett. 221, 7-16.
[15] Roberts, F., Roberts, C.W., Johnson, J.J., Kyle, D.E., Krell, T., Coggins, J.R., Coombs,
G.H., Milhous, W.K., Tzipori, S., Fergunson, D.J.P., Chakrebarti, D., and McLeod, R.
(1998). Nature, 393, 801-805.
[16] Kishore, G.M. & Shah, D.M. (1988) Annu. Rev. Biochem. 57, 627–663.
[17] Dieckmann, A., and Jung, A. Mol. Biochem.Parasitol., 19, 143-147.
[18] Campbell, S.A., Richards, T.A., Mui, E.J., Samuel, B.U., Coggins, J.R., McLeod, R.,
Roberts, C.W. (2004) Int. J. Parasitol., 34, 5-13.
[19] McRobert, L., McConkey, G.A. (2002) Mol. Biochem. Parasitol., 119, 273-278.
[20] McConkey G.A. (1999) Antimicrob. Agents Chemother., 43, 175-177.
[21] Gardner, M. J., Shallom, S.J., Carlton, J.M., et al. (2002) Nature, 419, 498-511.
[22] McConkey, G.A., Pinney, J.W., Westhead, D.R. (2004) Trends Parasitol., 20, 60-65.
[23] Fitzpatrick T, Ricken S, Lanzer M, Amrhein N, Macheroux P, Kappes B. (2001) Mol.
Microbiol., 4, 65-75.
[24] Ratledge, C. (1982). The Biology of the Mycobacteria. Academic Press, London. 1,
185-271.
[25] Voos, J.J., Rutter, K., Schroder, B.G., Su, H., Zhu, Y., and Barry, C.B. III. (2000) Proc.
Natl. Acad. Sci. USA. 97, 1252-1257.
[26] Parish, T., and Stoker, N. G. (2002). Microbiology, 148, 3069-3077.
[27] Whipp, M.J. & Pittard, A.J. (1995) J. Bacterial. 177, 1627-1629.
[28] Millar, G., Lewendon, A., Hunter, M.G. & Coggins, J.R. (1986). Biochem. J. 237, 427-
437.
[29] De Feyter, R.C. & Pittard, J. (1986) J. Bacteriol. 165, 331-333.
[30] Ely, B. and Pittard, J. (1979) J. Bacteriol. 138: 933–943.
27
[31] De Feyter, R.C., Davidson, B.E., and Pittard, J. (1986) J. Bacteriol. 165, 233–239.
[32] Vinella, D., Gagny, B., Joseleau-Petit, D., D’Ardi, R. & Cashel, M. (1996) J. Bacteriol.
178, 3818-3828.
[33] Gu, Y., Reshetnikova, L., Li, Y., Wu, Y., Yan, H., Singh, S., Ji, X. (2002) J. Mol. Biol.
319, 779-789.
[34] Krell, T., Coggins, J.R. & Lapthorn, A.J. (1998) J. Mol. Biol. 278, 983–997.
[35] Krell, T., Maclean, J., Boam, D.J., Cooper, A., Resmini, M., Brocklehurst, K. et al.
(2001) Protein Sci. 10, 1137–1149.
[36] Dreusicke, D., Karplus, A. & Schulz, G.E. (1988) J. Mol. Biol. 199, 359-371.
[37] Schlauderer, G.J. & Schulz, G.E. (1996) Protein Sci. 5, 434-441.
[38] Stehle, T. & Schulz, G.E. (1990) J. Mol. Biol. 211, 249-254.
[39] Müller-Dieckmann, H.-J. & Schulz, G.E. (1994) J. Mol. Biol. 236, 361-367.
[40] Wild, K., Bohner, T., Aubry, A., Folkers, G. & Schulz, G.E. (1995) FEBS Letters, 368,
289-292.
[41] Vonrhein, C., Schlauderer, G. J. & Schulz, G. E. (1995) Structure, 3, 483–490.
[42] Müller, C.W., Schlauderer, G.J., Reinstein, J. & Schulz, G.E. (1996) Structure, 4, 147-
156.
[43] Gerstein, M., Schulz, G.E. & Chothia, C. (1993) J. Mol. Biol. 229, 494-501.
[44] Walker, J.E., Saraste, M., Runswick, M.J. & Gay, N.J. (1982). EMBO J. 1, 945–951.
[45] Smith, C.A. & Rayment, I. (1996) Biophys. J. 70, 1590-1602.
[46] Reinstein, J., Schlichting, I., and Wittinghofer, A. (1990) Biochemistry 29, 7451–7459.
[47] Sigal, I.S., Gibbs, J.B., D’Alonzo, J.S., Temeles, G.L., Wolanski, B.S., Socher,
S.H., and Scolnick, E.M. (1986) Proc. Natl. Acad. Sci. USA 83, 952–956.
[48] Leipe, D.D., Koonin, E.V., and Aravind, L. (2003) J. Mol. Biol. 333, 781-815.
[49] Raviglione, M.C. (2003) Tuberculosis, 83, 4-14.
28
[50] Corbett, E.L., Watt, C.J., Walker, N., Maher, D., Willians, B.G., Raviglione, M.C., Dye,
C. (2003). Arch. Intern. Med., 163, 1009-1021.
[51] Dye C, Scheele S, Dolin P, Pathania V, Raviglione, M.C. (1999) JAMA, 282, 677-686.
[52] World Health Organization: Global Tuberculosis Control. WHO Report 2004. ISBN 92
4 156264 1.
[53] Basso, L.A., and Santos, D.S. (2005) Med. Chem. Ver.- Online. (accepted for
publication; Ms. MCRO/03/INV).
[54] Duncan, K. (2003) Tuberculosis, 83, 201-207.
[55] O’Brien, R.J., Nunn, P.P. (2001) Am. J. Resp. Crit. Care Med., 163, 1055-1058.
[56] Garbe, T., Joens, C., Charles, I., Dougan, G., Young, D. Journal of Bacteriology 172,
6774-6782.
[57] Cole, S.T.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher, C.; Harris, D.; Gordon, S.V.;
Eiglmeier, K.; Gas, S.; Barry III, C.E.; Tekaia, F.; Badcock, K.; Basham, D.; Brown, D.;
Chillingworth, T.; Connor, R.; Davies, R.; Devlin, K.; Feltwell, T.; Gentles, S.; Hamlin, N.;
Holroyd, S.; Hornsby, T.; Jagels, K.; Krogh, A.; McLean, J.; Moule, S.; Murphy, L.; Oliver,
K.; Osborne, J.; Quail, M.A.; Rajandream, M.-A.; Rogers, J.; Rutter, S.; Seeger, K.; Skelton,
J.; Squares, R.; Squares, S.; Sulston, J.E.; Taylor, K.; Whitehead, S. and Barrell, B.G. (1998)
Nature, 393, 537-544.
[58] Oliveira, J.S., Pinto, C.A., Basso, L.A., Santos, D.S. (2001) Protein Expr. and Purif.,
22, 430-435.
[59] Azevedo, W. F. de, Canduri, F., de Oliveira, J. S., Basso, L. A., Palma, M.S., Pereira, J.
H. & Santos, D. S. (2002). Biochem. Biophys. Res. Commun., 295, 142-148.
29
[60] Dhaliwal, B., Nichols, C.E., Ren, J., Lockyer, M., Charles, I., Hawkins, A.R.,
Stammers, D.K. FEBS Letter 2004, 574, 49-54.
[61] Pereira, J.H., Oliveira, J.S, Canduri, F., Dias, M.V.B., Palma, M.S., Basso, L.A.,
Santos, D.S., and Azevedo Jr., W.F. (2004) Acta Cryst. Section D 60, 2310-2319.
[62] Matte, A., Tari, L.W., and Delbaere, T.J. (1998) Structure, 6, 413-419.
[63] Williams, R.L., Oren, D.A., Muñoz-Dorado, J., Inouye, S., Inouye, M., and Arnold, M.
(1993) J. Mol. Biol., 234, 1230-1247.
[64] Jencks, W.P. (1975) Adv. Enzymol. 43, 219-410.
[65] Smith, C.A., Reyment, I. (1995) Biochemistry, 34, 8973-8981.
[66] Berchthold, H., Reshetnikovca, L., Reiser, C.O.A., Schrimer, N.K., Sprinzl, M.,
Hilgenfeld, R. Nature 1993, 365, 126-132.
[67] Pai, E.F., Krengel, U., Petski, G.A., Goody, R.S., Kabsch, W., Wittinghofer, A. EMBO
1990, 9, 2351-2359.
[68] Coleman, D.E., Berghuis A.M., Lee, E., Linder, M.E., Gilman, A.G., Sprang, S.R.
Science (1994), 265, 1405-1412.
[69] Abele, U. and Schulz, G. E. (1995) Protein Sci., 4, 1262-1271.
[70] Schlauderer, G.J. & Schulz, G.E. (1996) Protein Sci. 5, 434-441.
[71] Hasemann, C. A., Istvan, E. S., Uyeda, K. & Deisenhofer, J. (1996).
Structure, 4, 1017-1029.
[72] Romanowski, M.J., Burley, S.K. (2002) Proteins: Structure, Function, and Genetics.
47, 558-562.
[73] Bennett, W.S. and Steitz, T.A. (1980) J. Mol.Biol. 140, 211-230.
[74] Schulz, G.E., Müller, C.W. and Diederichs, K. (1990) J. Mol. Biol. 213, 627-630.
[75] Idziak, C., Price, N.C., Kelly, S.M., Krell, T., Boam, D.J., Lapthorn, A.J., and Coggins,
J.R. (1997) Biochem. Soc. Trans., 25, S627.
30
[76] Cerasoli, E., Kelly, S.M., Coggins, J.R., Lapthorn, A.J., Clarke D.T. (2003) Biochim.
Biophys.Acta, 1648, 43-54.
[77] Pittard, A.J., and Wallace, B.J. (1966) J. Bacteriol., 92, 1070-1075.
[78] Whipp, M.J., Camakaris, H., and Pittard, A.J. (1998) Gene, 209, 185-192.
[79] Kraft, L., Sprenger, G. A., and Lindqvist, Y. (2002) J. Mol. Biol., 318, 1057-1069.
[80] Flachner, B., Kovári, Z, Varga, A., Gugolya, Z., Vonderviszt, F., Náray-Szabó, G., and
Vas, M. (2004) Biochemistry, 43, 3436-3449.
[81] Mildvan, A.S. (1979) Adv. Enzymol. Relat. Areas Mol. Biol., 49, 103-126.
[82] Schlichting, I., and Reinstein, J. (1997) Biochemistry, 36, 9290-9296.
[83] Xu, Y.-W., Moréra, S., Janin, J., and Cherfils, J. (1997) Proc. Natl. Acad. Sci. USA, 94,
3579-3583.
[84] Scheffzek, K., and Wittinghofer, A. (1997) Science, 277, 333-338.
31
Figures
Fig. (1).
Fig. (1). The Shikimate pathway. Erythrose 4-phosphate and phosphoenolpyruvate
(precursors) are converted to chorismic acid (chorismate). Chorismate is a essential
intermediate for the synthesis of aromatic amino acid (Phe,Tyr and Trp), folate, ubiquinones,
menaquinones, and enterobactin. The seven steps are catalyzed by enzymes: 3-deoxy-D-
arabino-heptulosonate 7- phosphate synthase (DAHP synthase), 3-dehydroquinate synthase,
3-dehydroquinate dehydratase, shikimate dehydrogenase, shikimate kinase, 5-enolpyruvyl
shikimate 3-phosphate synthase (EPSP synthase), and chorismate synthase. The reaction
catalyzed by shikimate kinase is circled by a drawn line.
32
Fig. (2).
Fig. (2). Overall structure of Shikimate kinase from M. tuberculosis [61]. MtSK displays an
α
/β-fold and the precise ordering of the strands 23145 in the parallel β-sheet classifies MtSK
as belonging to the family as the NMP kinases. The SKs are composed of three domains:
CORE, LID and Shikimate-binding (SB) domains. The positions of the ADP, Mg
2+
and
shikimate are shown in the MtSK structure.
33
Fig. (3).
Fig. (3). A) The molecular structure of shikimate [3R-(3α,4α,5β)]3,4,5-trihydroxy-1-
cyclohexene-1-carboxylic acid] shows the carboxyl group and three hydroxyl groups B)
A
B
34
Interactions involved in the binding of shikimate to the MtSK active site (PDB acess code:
1WE2) [61]. Hydrogen bonds are represented as broken lines. For clarity, the the water320-
mediated hydrogens bonds between Gly79, Gly80, Gly81, Ala82, Arg58, and Glu61 protein
residues and 3-hydroxyl group of shikimate are not shown.
35
Fig. (4).
Fig. (4). Molecular surface of (A) MtSK-MgADP (1L4U) [33], (B) MtSK-MgADP-
Shikimate (1WE2) [61], and (C) MtSK-ADP-Shikimate (1U8A) [60] structures The
molecular surface areas have been calculated using the program Swiss-PDBViewer v3.7
(www.expasy.org/spdbv), probe radius of 1.4 Å, and a fixed radius for all atoms. Calculated
values are approximately 7246 Å
2
for MtSK complexed with only MgADP (A), 6915 Å
2
for
structure complexed with MgADP and Shikimate (B), and 7040 Å
2
for the Mg
2+
-free
structure complexed with ADP and Shikimate. Thus, approximately 330 Å
2
and 206 Å
2
of
solvent-acessible surface are buried on shikimate binding to, respectively, MtSK-MgADP-
Shikimate and MtSK-ADP-Shikimate ternary complexes. The shikimate binding leads to a
counter movement of LID and SB domains and, consequently, to a partial closure of
shikimate-binding pocket. Probably, the Mg
2+
cation has a role in the closure of MtSK active
site. All atoms of shikimate and ADP, Mg
+2
cation, and oxygen atoms of waters are colored
grey.
36
Fig. (5).
Fig. (5). Superposition of SK-MgADP-Shikimate ternary complex from Mycobacterium
tuberculosis [61] (red line) and apo-SK (K15M/P115L) from Erwinia chrysanthemi [34]
(blue line).
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexo 3
Molecular model of shikimate kinase from Mycobacterium tuberculosis
Walter Filgueira de Azevedo Jr.,
a,b,
*
Fernanda Canduri,
a,b
Jaim Sim
~
ooes de Oliveira,
c
Luiz Augusto Basso,
c
M
aario S
eergio Palma,
b,d
Jos
ee Henrique Pereira,
a
and Di
oogenes Santiago Santos
c
a
Departamento de F
iisica, UNESP, S
~
aao Jos
ee do Rio Preto, SP 15054-000, Brazil
b
Center for Applied Toxicology, Instituto Butantan, S~aao Paulo, SP 05503-900, Brazil
c
Rede Brasileira de Pesquisa de Pesquisas em Tuberculose, Departamento de Biologia Molecular e Biotecnologia, UFRGS, Porto Alegre,
RS 91501-970, Brazil
d
Laboratory of Structural Biology and Zoochemistry/Department of Biology, Institute of Biosciences, UNESP, Rio Claro, SP 13506-900, Brazil
Received 26 May 2002
Abstract
Tuberculosis (TB) resurged in the late 1980s and now kills approximately 3 million people a year. The reemergence of tuber-
culosis as a public health threat has created a need to develop new anti-mycobacterial agents. The shikimate pathway is an attractive
target for herbicides and anti-microbial agents development because it is essential in algae, higher plants, bacteria, and fungi, but
absent from mammals. Homologs to enzymes in the shikimate pathway have been identified in the genome sequence of Myco-
bacterium tuberculosis. Among them, the shikimate kinase I encoding gene (aroK) was proposed to be present by sequence ho-
mology. Accordingly, to pave the way for structural and functional efforts towards anti-mycobacterial agents development, here we
describe the molecular modeling of M. tuberculosis shikimate kinase that should provide a structural framework on which the design
of specific inhibitors may be based. Ó 2002 Elsevier Science (USA). All rights reserved.
Keywords: Shikimate kinase; Bioinformatics; Structure; Drug design; Mycobacterium tuberculosis
The fifth annual report on global tuberculosis (TB)
control of the World Health Organization found that
there were an estimated 8.4 million new cases in 1999, up
from 8.0 million in 1997 [1]. It is expected that there will
be 10.2 million new cases in 2005 if the present trend
continues. Approximately 3 million persons die from the
disease each year [2]. Ninety percent of tuberculosis
cases occur in developing countries, where few resources
are available to ensure proper treatment and where
human immunodeficiency virus (HIV) infection may be
common. The concentration of deaths due to tuber-
culosis in demographically developing nations and
mortality rate in the range from 25 to 54 years, the most
economically fruitful years of life, causes substantial
losses in productivity and contributes to the impover-
ishment of third-world countries [3]. The reemergence of
TB as a public health threat, the high susceptibility of
HIV-infected persons to the disease, and the prolifera-
tion of multi-drug (MDR) strains have created much
scientific interest in developing new anti-mycobacterial
agents to both treat Mycobacterium tuberculosis strains
resistant to existing drugs and shorten the duration of
short-course treatment to improve patient compliance
[4].
The shikimate pathway is an attractive target for the
development of herbicides and anti-microbial agents
because it is essential in algae, higher plants, bacteria,
and fungi, but absent from mammals [5]. In mycobac-
teria, the shikimate pathway leads to the biosynthesis of
precursors for the synthesis of aromatic amino acids,
naphthoquinones, menaquinones, and mycobactin [6].
Homologs to enzymes in the shikimate pathway have
been identified in the complete genome sequence of
Mycobacterium tuberculosis H37Rv strain [7]. Among
them, the shikimate kinase I (mtSK, EC 2.7.1.71) en-
coding gene (aroK, Rv2539c) was proposed to be present
Biochemical and Biophysical Research Communications 295 (2002) 142–148
www.academicpress.com
BBRC
*
Corresponding author. Fax: +55-17-221-2247.
E-mail address: [email protected] (W. Filgueira de
Azevedo Jr.).
0006-291X/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved.
PII: S 0 0 0 6 - 2 9 1 X ( 0 2 ) 0 0 6 3 2 - 0
by sequence homology. Shikimate kinase catalyzes a
phosphate transfer from ATP to the carbon-3 hydroxyl
group of shikimate resulting in the formation of shiki-
mate-3-phosphate (S3P) and ADP.
The present paper describes the molecular model of
M. tuberculosis shikimate kinase (mtSK) and analysis of
SK and shikimate complex obtained by docking simu-
lations. The homology modeling was performed using
three crystallographic structures of SK from Erwinia
chrysanthemi, solved to resolution better than 2.6
AA, as
templates [8]. The mtSK has been cloned, sequenced,
overexpressed in soluble and functional forms [9], thus
allowing enzymological studies to be performed. The
results presented here should provide a three-dimen-
sional model of mtSK to both guide enzymological
studies and aid the design of specific inhibitors.
Methods
Molecular modeling. For modeling of the mtSK we used restrained-
based modeling implemented in the program MODELLER [10]. This
program is an automated approach to comparative modeling by sat-
isfaction of spatial restraints [11–13]. The modeling procedure begins
with an alignment of the sequence to be modeled (target) with related
known three-dimensional structures (templates). This alignment is
usually the input to the program. The output is a three-dimensional
model for the target sequence containing all main-chain and side-chain
non-hydrogen atoms.
The degree of primary sequence identity between mtSK and Er-
winia chrysanthemi shikiamte SK (ecSK) indicates that the crystallo-
graphic structures of ecSK are good models to be used as templates for
mtSK. The atomic coordinates of three crystallographic ecSK struc-
tures (PDB access code: 1SHK, 2SHK, and 1E6C) [8,14], with two
independent structures in each asymmetric unit, solved to resolution
better than 2.6
AA were used to build up an ensemble of SK structures to
be used as starting models for modeling of the mtSK. The atomic
coordinates of all waters and ligands were removed from the ecSK
structures. Next, the spatial restraints and CHARMM energy terms
enforcing proper stereochemistry [15] were combined into an objective
function. Finally, the model is obtained by optimizing the objective
function in Cartesian space. The optimization is carried out by the use
of the variable target function method [16] employing methods of
conjugate gradients and molecular dynamics with simulated annealing.
Several slightly different models can be calculated by varying the initial
structure. A total of 500 models were generated for mtSK, and the final
model was selected based on stereochemical quality.
Docking simulations. To obtain information about the docking of
shikimate to ecSK and mtSK, several rigid docking simulations were
performed using the geometric recognition algorithm, which was de-
veloped to identify molecular surface complementarity. The geometric
recognition algorithm was implemented in the program GRAMM [17].
The atomic coordinates of shikimate, used in the docking simula-
tions, were obtained from structure of 5-enolpyruvylshikimate-3-
phosphate synthase liganded with shikimate-3-phosphate and
glyphosate (PDB access code: 1G6S) [18]. To generate the ternary
complex mtSK–shikimate–ADP/Mg
2þ
we superposed the atomic co-
ordinates of the ADP/Mg
2þ
to the binary complex of mtSK–shikimate.
The optimization of the complexes was carried out by the use of the
variable target function method [10] employing methods of conjugate
gradients and molecular dynamics with simulated annealing. All
docking simulations and optimization process were performed on SGI
Octane, R12000.
Analysis of the model. The overall stereochemical quality of the final
model for mtSK complex was assessed by the program PROCHECK
[19]. The cutoff for hydrogen bonds and salt bridges was 3.6
AA.
Results and discussion
Primary sequence comparison
The sequence alignment of ecSK (template) and
mtSK (target) is shown in Fig. 1. The secondary struc-
tural elements are indicated in the figure. The sequence
mtSK shows 34% of identity with the sequence of ecSK.
Quality of the model
Figs. 2A and B show the Ramachandran diagram /–
w plots for the mtSK structure and for three crystallo-
graphic SK structures solved to resolution better than
2.6
AA. The Ramachandran plot for the three ecSK
structures was generated to compare the overall stereo-
chemical quality of mtSK model against SK structures
solved by biocrystallography. Analysis of the Rama-
chandran plot of the mtSK model shows that 91.1% of
the residues lie in the most favorable regions and the
Fig. 1. The sequence alignment of ecSK and mtSK indicating the secondary structural elements. The sequence mtSK shows 34% of identity with the
sequence of ecSK. The alignment was performed with the program CLUSTAL V [31].
W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148 143
remaining 8.9% in the additional allowed regions. The
same analysis for three crystallographic ecSK structures
(six chains) present 93.7% of residues in the most fa-
vorable, 6.1% additional allowed regions, and 0.7%
generously allowed regions. The overall rating for the
mtSK model is slightly poorer than the one obtained for
the three structures of SK. However, it has over 90% of
the residues in the most favorable regions.
Overall description
MtSK is an a=b protein consisting of a mixed b sheet
surrounded by a helices. A central five stranded parallel
b-sheet (b1–b5) presents the strand order 23145. The
b-strands are flanked on either side by a helices (a1 and
a8 on one side, a4, a5, and a7 on the other). Fig. 3
shows a schematic diagram of the mtSK structure, with
shikimate and ADP/Mg
2þ
bound to the structure.
The ordering of the strands 23145 observed the mtSK
structure classifies it as belonging to the same structural
family as the nucleoside monophosphate (NMP) ki-
nases. The mtSK structure exhibits the Walker A-motif
located between b1anda1 forming a canonical phos-
phate-binding loop (P-loop). The core of the mtSK
structure forms a classical mononucleoside-binding fold
[20].
It has been reported that NMP kinases undergo large
confomation changes during catalysis. The regions re-
sponsible for this movement are NMP-binding site and
the lid domain. The NMP-binding site is formed by a
series of helices between strands 1 and 2 of the parallel
b-sheet. The lid domain is a region of variable size and
structure following the forth b-strand of the sheet
[21,22]. The residues from 112 to 123 form the lid do-
main in the mtSK which has been reported to be highly
dynamic and possibly flexible in solution in the ecSK
structure [8].
ADP =Mg
2þ
-binding site
The molecular model for ternary complex mtSK–
shikimate–ADP/Mg
2þ
indicates that ADP/Mg
2þ
is
Fig. 2. (A) Ramachandran diagram /–w plots for the mtSK structure and (B) for three crystallographic SK structures solved to resolution better than
2.6
AA.
Fig. 3. Ribbon diagram of the mtSK structure with shikimate and
ADP/Mg
2þ
bound to the structure generated by MOLSCRIPT [32].
144 W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148
tightly bound to the mtSK structure. The intermolecular
hydrogen bonds are described in Table 1. Most of the
intermolecular hydrogen bonds observed in the ecSK
structure is conserved in the ternary complex mtSK–
shikimate–ADP/Mg
2þ
. Fig. 4 shows the superposition of
the ATP-binding site of mtSK and ecSK. As previously
described a phosphate-binding loop (P-loop) accom-
modates the b-phosphate of ADP by donating hydrogen
bonds from several backbone amides [8,14]. SKs contain
a conserved stretch of sequence GXXXXGKT/S known
as the Walker A-motif [23]. This motif forms the P-loop
in the ecSK and mtSK structures. In addition to the
Walker A-motif, the mtSK structure presents a modified
Walker B-motif. The Walker B-motif is present in the
majority of purine–nucleotide binding proteins. This
motif, Z–Z–Asp–X–X–Gly (where Z is a hydrophobic
residue and X is any residue) forms a loop around the
c-phosphate of the nucleotide. The B-motif present in
the mtSK has a different conformation from that ob-
served in proteins with full Walker B-motif [8], the Asp
is replaced by Ser77. However, the conserved Gly80 of
mtSK (Fig. 1) is in an almost identical position to the
conserved Gly found in proteins with the full B-motif
with its amide nitrogen hydrogen-bonded to the
c-phosphate of a bound ATP.
Shikimate-binding site
The crystallographic structure of ecSK indicated the
presence of a strong electron density peak attributed to
shikimate. However, the electron density was not clear
enough to include shikimate in the molecular structure
[8]. The docking simulations of shikimate to ecSK and
mtSK identified that shikimate binds in a position
analogous to nucleotide monophosphate in NMP
kinases [8]. The shikimate binding domain, which fol-
lows strand b2, consists of helices a2 and a3 and the
N-terminal region of helix a4. A total of four hydrogen
bonds between ecSK and shikimate, in the model,
involving the residues Lys15, Asp34, and Arg 136 was
observed. For the mtSK model the same pattern was
observed. Fig. 5A and B show the main residues
involved in contact with shikimate in the complexes.
Tables 2 and 3 show the intermolecular hydrogen bonds
for both structures. The residues involved hydrogen
Table 1
Intermolecular hydrogen bonds between mtSK and ADP
Hydrogen bonds between active site and inhibitor
Distance (
AA)
ADP mtSK
O1B Leu10 O 3.49
O3B Gly12 N 3.15
O1B Ser13 N 3.08
O1B Ser13 OG 2.66
O3A Gly14 N 2.75
O3A Lys15 N 3.35
O2B Lys15 NZ 2.80
O3B Lys15 NZ 2.54
O1B Lys15 NZ 2.82
O2A Ser16 OG 2.65
O2A Ser16 N 3.40
O1A Ser16 OG 2.98
O2A Thr17 OG1 3.02
O2A Thr17 N 3.08
N1 Arg110 NH1 3.05
O4 Arg110 NE 3.41
N3 Arg110 NH1 2.88
N7 Arg110 NH1 3.29
N6 Arg152 O 2.52
N1 Arg152 O 3.23
N1 Arg153 O 3.37
N6 Arg153 O 2.66
N3 Arg153 NE 2.80
N3 Arg153 NH2 2.68
N1 Arg153 N 3.20
Fig. 4. Superimposed binding pockets of the ATP-binding site of mtSK (thick line) and ecSK (thin line).
W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148 145
bond identified from the docking simulation of shiki-
mate to ecSK are in good agreement with that of crys-
tallographic structure.
The electrostatic potential surface of the ecSK and
mtSK complexed with shikimate calculated with
GRASP [24] indicates the presence of some charge
complementarity between shikimate and enzyme,
nevertheless most of the residues in the binding pocket is
hydrophobic in all structures. Figs. 6A and B show the
molecular surfaces for mtSK and ecSK complexed with
shikimate. The electrostatic potential surfaces of mtSK
and ecSK show some striking differences. The main
Table 2
Intermolecular hydrogen bonds between mtSK and shikimate
Hydrogen bonds between active site and inhibitor
Distance (
AA)
Shikimate mtSK
O1 Asp34 OD2 2.58
O2 Asp34 OD2 3.55
O3 Lys15 NZ 3.36
O5 Arg136 NH1 3.47
Table 3
Intermolecular hydrogen bonds between ecSK and shikimate
Hydrogen bonds between active site and inhibitor
Distance (
AA)
Shikimate ecSK
O1 Asp34 OD2 2.64
O2 Asp34 OD2 3.17
O3 Lys15 NZ 3.46
O5 Arg11 NH1 2.77
O5 Arg136 NH1 3.51
A
B
Fig. 5. Main residues involved in contact with shikimate in the complexes (A) mtSK:shikimate and (B) ecSK:shikimate.
146 W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148
difference is the presence of a positive potential patch on
the surface of mtSK, which is not observed in the ecSK
surface. This positive patch indicates a concentration
of positive charged residues in the mtSK structure,
involving residues Arg21, Arg125, Lys128, Arg130,
Lys135, Arg142, Arg147, Arg153, Arg160, His161, and
Arg165. These residues are changed to polar or hydro-
phobic or negative charged residues in the ecSK
sequence. Particularly interesting is the intermolecular
hydrogen bond pattern between SK and shikimate. The
residues Lys15, Asp34, and Arg136, involved in inter-
molecular hydrogen bonds, are conserved in both
structures. The model strongly indicates that the shiki-
mate binding domain is a well-conserved motif in SK
structures. Furthermore, the alignment of 37 SK
sequences, figure not shown, indicates that the main
residues involved in intermolecular hydrogen bonds are
conserved in all sequences. Such observation suggests
that competitive inhibitors with shikimate will be able to
inhibit most or even all SKs, since specificity and affinity
between enzyme and its inhibitor depend on directional
hydrogen bonds and ionic interactions, as well as on
shape complementarity of the contact surfaces of both
partners [25–30]. Further inhibition experiments may
confirm this prediction.
Acknowledgments
This work was supported by grants from FAPESP (SMOLBNet),
CNPq, CAPES, and Instituto do Mil
^
eenio (CNPq-MCT). WFA
(CNPq, 300851/98-7) and MSP (CNPq, 500079/90-0) are researchers
for the Brazilian Council for Scientific and Technological Develop-
ment.
References
[1] World Health Organization, Global Tuberculosis Control, WHO
Report 2001, Geneva, Switzerland, WHO/CDS/TB/2001.287.
[2] N.E. Dunlap, J. Bass, P. Fujiwara, P. Hopewell, C.R. Horsburgh
Jr., M. Salfinger, P.M. Simone, Diagnostic standards and classi-
fication of tuberculosis in adults and children, Am. J. Respir. Crit.
Care Med. 161 (2000) 1376–1395.
[3] C.J.L. Murray, in: B.R. Bloom (Ed.), Tuberculosis: Pathogenesis,
Protection, and Control, ASM, Washington, 1994, pp. 583–622.
[4] L.A. Basso, J.S. Blanchard, Resistance to antitubercular drugs,
Adv. Exp. Med. Biol. 456 (1998) 115–144.
[5] R. Bentley, The shikimate pathway
—a metabolic tree with many
branches, Crit. Rev. Biochem. Mol. Biol. 25 (1990) 307–384.
[6] C. Ratledge, Nutrition, growth and metabolism, in: C. Ratledge,
J.L. Stanford (Eds.), The Biology of the Mycobacteria, vol. 1,
Academic Press, London, 1982, pp. 185–271.
[7] S.T. Cole, R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D.
Harris, S.V. Gordon, K. Eiglmeier, S. Gas, C.E. Barry III, F.
Tekaia, K. Badcock, D. Basham, D. Brown, T. Chillingworth, R.
Connor, R. Davies, K. Devlin, T. Feltwell, S. Gentles, N. Hamlin,
S. Holroyd, T. Hornsby, K. Jagels, B.G. Barrel, Deciphering the
biology of Mycobacterium tuberculosis from the complete genome
sequence, Nature 393 (1998) 537–544.
[8] T. Krell, J.R. Coggins, A.J. Lapthorn, The three-dimensional
structure of shikimate kinase, J. Mol. Biol. 278 (1998) 983–997.
[9] J.S. Oliveira, C.A. Pinto, L.A. Basso, D.S. Santos, Cloning and
overexpression in soluble form of functional shikimate kinase and
5-enolpyruvylshikimate-3-phosphate synthase enzymes from My-
cobacterium tuberculosis, Protein Express. Purif. 22 (2001) 430–
435.
[10] A. Sali, T.L. Blundell, Comparative protein modelling by satis-
faction of spatial restraints, J. Mol. Biol. 234 (1993) 779–815.
[11] A. Sali, J.P. Overington, Derivation of rules for comparative
protein modeling from a database of protein structure alignments,
Protein Sci. 3 (9) (1994) 1582–1596.
[12] A. Sali, L. Potterton, F. Yuan, H. van Vlijmen, M. Karplus,
Evaluation of comparative protein modeling by MODELLER,
Proteins 23 (3) (1995) 318–326.
Fig. 6. (A) Molecular surfaces for mtSK and (B) ecSK complexed with shikimate generated with GRASP [24].
W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148 147
[13] A. Sali, Modeling mutations and homologous proteins, Curr.
Opin. Biotechnol. 6 (4) (1995) 437–451.
[14] T. Krell, J. Maclean, D.J. Boam, A. Cooper, M. Resmini, K.
Brocklehurst, S.M. Kelly, N.C. Price, A.J. Lapthorn, J.R.
Coggins, Biochemical and X-ray crystallographic studies on
shikimate kinase: the important structural role of the P-loop
lysine, Protein Sci. 10 (2001) 1137–1149.
[15] B.R. Brooks, R.E. Bruccoleri, B.D. Olafson, D.J. States, S.
Swaminathan, M. Karplus, CHARMM: a program for macro-
molecular energy minimization and dynamics calculations, J.
Comp. Chem. 4 (1983) 187–217.
[16] W. Braun, N. Go, Calculation of protein conformations by
proton–proton distance constraints. A new efficient algorithm, J.
Mol. Biol. 186 (3) (1985) 611–626.
[17] E. Katchalski-Katzir, I. Shariv, M. Eisenstein, A.A. Friesem, C.
Aflalo, I.A. Vakser, Molecular surface recognition: determination
of geometric fit between proteins and their ligands by correlation
techniques, Proc. Natl. Acad. Sci. USA 89 (1992) 2195–2199.
[18] E. Sch
€
oonbrunn, S. Eschenburg, W.A. Shuttleworth, J.V. Schloss,
N. Amrhein, J.N.S. Evans, W. Kabsch, Interaction of the
herbicide glyphosate with its target enzyme 5-enolpyruvylshiki-
mate-3-phosphate synthase in atomic detail, Proc. Natl. Acad. Sci.
USA 98 (2001) 1376–1380.
[19] R.A. Laskowski, M.W. MacArthur, D.K. Smith, D.T. Jones,
E.G. Hutchinson, A.L. Morris, D. Naylor, D.S. Moss, J.M.
Thorton, PROCHECK v.3.0
—program to check the stereochem-
istry quality of protein structures—operating instructions, 1994.
[20] G.E. Schulz, Binding of nucleotides by proteins, Curr. Opin.
Struct. Biol. 2 (1992) 61–67.
[21] C.W. M
€
uuller, G.J. Schlauderer, J. Reinstein, G.E. Schulz, Ade-
nylate kinase motions during catalysis: an energetic counterweight
balancing substrate binding, Structure 4 (1996) 147–156.
[22] M. Gerstein, G. Schulz, C. Chothia, Domain closure in adenylate
kinase Joints on either side of two helices close like neighboring
fingers, J. Mol. Biol. 229 (2) (1993) 494–501.
[23] J.E. Walker, M. Saraste, M.J. Runswick, N.J. Gay, Distantly
related sequences in the a and b subunits of ATP synthase,
myosin, kinases and other ATP-requiring enzymes and a common
nucleotide binding fold, EMBO J. 1 (1982) 945–951.
[24] A. Nicholls, K.A. Sharp, B. Honig, Protein folding and associ-
ation: insights from the interfacial and thermodynamic properties
of hydrocarbons, Proteins 11 (4) (1991) 281–296.
[25] F. Canduri, L.G.V.L. Teodoro, C.C.B. Lorenzi, V. Hial, R.A.S.
Gomes, J. Ruggiero Neto, W.F. de Azevedo Jr., Crystal structure
of human uropepsin at 2.45
AA resolution, Acta Crystallogr. D 57
(2001) 1560–1570.
[26] W.F. De Azevedo Jr., H.J. Mueller-Dieckmann, U. Schulze-
Gahmen, P.J. Worland, E. Sausville, S.-H. Kim, Structural basis
for specificity and potency of a flavonoid inhibitor of human
CDK2, a cell cycle kinase, Proc. Natl. Acad. Sci. USA 93 (7)
(1996) 2735–2740.
[27] W.F. De Azevedo Jr., S. Leclerc, L. Meijer, L. Havlicek, M.
Strnad, S.-H. Kim, Inhibition of cyclin-dependent kinases by
purine analogues: crystal structure of human CDK2 complexed
with roscovitine, Eur. J. Biochem. 243 (1997) 518–526.
[28] W.F. De Azevedo Jr., F. Canduri, V. Fadel, L.G.V.L. Teodoro,
V. Hial, R.A.S. Gomes, Molecular model for the binary complex
of uropepsin and pepstatin, Biochem. Biophys. Res. Commun.
287 (1) (2001) 277–281.
[29] W.F. De Azevedo Jr., F. Canduri, N.J.F. da Silveira, Structural
basis for inhibition of cyclin-dependent kinase 9 by flavopiridol,
Biochem. Biophys. Res. Commun. 293 (1) (2002) 566–571.
[30] S.-H. Kim, U. Schulze-Gahmen, J. Brandsen, W.F. de Azevedo
Jr., Structural basis for chemical inhibitor of CDK2, Prog. Cell
Cycle Res. 2 (1996) 137–145.
[31] D.G. Higgins, A.J. Bleasby, R. Fuchs, CLUSTAL V: improved
software for multiple sequence alignment, Comput. Appl. Biosci.
8 (2) (1992) 189–191.
[32] P. Kraulis, MOLSCRIPT: a program to produce both detailed
and schematic plots of proteins, J. Appl. Cryst. 24 (1991) 946–950.
148 W. Filgueira de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 295 (2002) 142–148
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexo 4
Structural bioinformatics study of EPSP synthase
from Mycobacterium tuberculosis
Jos
ee Henrique Pereira,
a
Fernanda Canduri,
a,b,
*
Jaim Sim
~
ooes de Oliveira,
c
Nelson Jos
ee Freitas da Silveira,
a
Luiz Augusto Basso,
c
M
aario S
eergio Palma,
b,d
Walter Filgueira de Azevedo Jr.,
a,b,
*
and Di
oogenes Santiago Santos
e,
*
a
Departamento de F
ıısica, UNESP, S
~
aao Jos
ee do Rio Preto, SP 15054-000, Brazil
b
Center for Applied Toxinology, Instituto Butantan, S
~
aao Paulo, SP 05503-900, Brazil
c
Rede Brasileira de Pesquisa de Pesquisas em Tuberculose Grupo de Microbiologia Molecular e Funcional,
Departamento de Biologia Molecular e Biotecnologia, UFRGS, Porto Alegre, RS 91501-970, Brazil
d
Laboratory of Structural Biology and Zoochemistry, CEIS/Department of Biology, Institute of Biosciences, UNESP, Rio Claro, SP 13506-900, Brazil
e
Faculdade de Farm
aacia/Instituto de Pesquisas Biom
eedicas, Pontif
ııcia Universidade Cat
oolica do Rio Grande do Sul, Porto Alegre, RS, Brazil
Received 23 October 2003
Abstract
The shikimate pathway is an attractive target for herbicides and antimicrobial agent development because it is essential in algae,
higher plants, bacteria, and fungi, but absent from mammals. Homologues to enzymes in the shikimate pathway have been identified
in the genome sequence of Mycobacterium tuberculosis. Among them, the EPSP synthase was proposed to be present by sequence
homology. Accordingly, in order to pave the way for structural and functional efforts towards anti-mycobacterial agent develop-
ment, here we describe the molecular modeling of 5-enolpyruvylshikimate-3-phosphate (EPSP) synthase isolated from M. tuber-
culosis that should provide a structural framework on which the design of specific inhibitors may be based on. Significant differences
in the relative orientation of the domains in the two models result in “open” and “closed” conformations. The possible relevance of
this structural transition in the ligand biding is discussed.
Ó 2003 Elsevier Inc. All rights reserved.
Keywords: EPSP synthase; Bioinformatics; Molecular modeling; Mycobacterium tuberculosis
Tuberculosis (TB) remains one of the most deadly
infectious diseases in the world. It is estimated that ap-
proximately 1 billion individuals are infected with latent
TB. Tuberculosis has made a steady global comeback in
the late 1980s and now kills more than 2 million people
each year worldwide, according to the World Health
Organization (WHO) which in 1993 declared tubercu-
losis to be a global emergence [1].
Mycobacterium tuberculosis is by far the biggest killer
among infectious agents, accounting for 7% of all deaths
and 26% of all avoidable deaths. In impove rished third
world countries where 95% of cases and 98% of deaths
occur, 70–80% of TB cases are in the economically most
productive 15–50 year age bracket [1,2].
The treatment of tuberculosis is a special problem in
the field of chemotherapy. Many of the drugs employed
to treat the disease are used only for treating infections
caused by mycobacteria. Treatment of the active case of
TB always includes simultaneous therapy with two or
more of the frontline drugs: isoniazid, ethambutol, rif-
ampicin, and streptomycin which are used to decrease
the rate of emerg ence of resistant strains as well to
increase the antibacterial effect [3–5].
Recent outbreaks of tuberculosis caused by multi-
drug-resistant (MDR) strains, mainly in individuals in-
fected with HIV, have created a scaring element to the
scenario and also created a worldwide interest in ex-
panding current programs of development of new drugs
*
Corresponding authors. Fax: +55172212247.
E-mail addresses: [email protected] (F. Canduri), wal-
[email protected] (W.F. Azevedo Jr.), [email protected] (D.S. San-
tos).
0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2003.10.175
Biochemical and Biophysical Research Communications 312 (2003) 608–614
BBRC
www.elsevier.com/locate/ybbrc
different in kinds from existing ones and based on the
principle of selective toxicity on enzymes or structure of
the bacillus to both treat M. tuberculosis strains resistant
to existing drugs and shorte n the duration of short-
course treatment to improve patient compliance [4,6].
The shikimate pathway is an attractive target for the
development of herbicides and antimicrobial agents be-
cause it is essential in algae, higher plants, bacter ia, and
fungi, but absent from mammals [7]. In mycobacteria, the
shikimate pathway leads to the biosynthesis of precursors
for the synthesis of aromatic amino acids, naphthoqui-
nones, menaquinones, and mycobactin [8]. Homologu es
to enzymes in the shikimate pathway have been identified
in the complete genome sequence of M. tuberculosis
H37Rv strain [9]. Among them, 5-enolpyruvylshikimate-
3-phosphate (EPSP) synthase encoded by aroA gene,
which catalyzes the transfer of the enolpyruvyl moiety
from phosphoenolpyruvate (PEP) and inorganic phos-
phate. A valuable lead compound in the search of new
inhibitors is glyphosate, which has proven to be potent
and specific inhibitor of EPSP synthase [10]. Glyphosate is
successfully used as a herbicide, being the active ingredi-
ent of the widely used weed control agent Roundup
Ready, and was recently shown to inhibit the growth of
the pathogenic parasiti es Plasmodium falciparum, Toxo-
plasma gondii,andCryptosporidium parvum [11,12].
The present paper describes the two molecular
models of M. tuberculosis EPSP synthase, one without
any ligand and another in complex with 3-phosphos-
hikimate (S3P) and glyphosate. The homology model-
ing was performed using the crystallographic structures
of EPSP synthase from Escherichia coli [12,13], as
templates. The EPSP synthase has been cloned, se-
quenced, and overexpressed in soluble and functional
form [14], thus allowing enzymological studies to be
performed. The results presented here should provide a
three-dimensional model of EPSP synthase to both
guide enzymological studies and aid in the design of
specific inhibitors.
Methods
Molecular modeling. For modeling of the EPSP synthase we used
restrained-based modeling implemented in the program MODELLER
[15]. This program is an automated approach to comparative modeling
by satisfaction of spatial restraints [16–18]. The modeling procedure
begins with an alignment of the sequence to be modeled (target) with
related known three-dimensional structures (templates). This align-
ment is usually the input to the program. The output is a three-
dimensional model for the target sequence containing all main-chain
and side-chain non-hydrogen atoms.
The degree of primary sequence identity between MtEPSP synthase
and EPSP synthase from E. coli indicates that the crystallographic
structures of EcEPSP are good models to be used as templates for
MtESPS synthase. The EPSP synthase isolated from E. coli showed
two conformation states, open (unliganded structure) and closed
(liganded structure) [12]. Models in the two conformations were gen-
erated, using the a-carbon atomic coordinates from 1EPSP [13] to
generate the open structure and the atomic coordinates from 1G6S [12]
to generate the closed structure. The atomic coordinates of all waters
were removed from the EPSP structure. The 3-phosphoshikimate and
glyphosate of the template were kept in the structure for the closed
Fig. 1. The sequence alignment of EcEPSP and MtEPSP indicating the secondary structural elements. The sequence MtEPSP shows 31% of identity
with the sequence of EcEPSP. The alignment was performed with the program CLUSTAL V [37].
J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614 609
conformation. Next, the spatial restraints and CHARMM energy
terms enforcing proper stereochemistry [19] were combined into an
objective function. Finally, the model is obtained by optimizing the
objective function in Cartesian space. The optimization is carried out
by the use of the variable target function method [20] employing
methods of conjugate gradients. Several slightly different models can
be calculated by varying the initial structure. A total of 1000 models
were generated for each EPSP synthase conformation state, and the
final models were selected based on stereochemical quality. The opti-
mization of the complex was carried out by the use of the variable
target function method [15] employing methods of conjugate gradients
and molecular dynamics with simulated annealing. All modeling pro-
cess was performed on a Beowulf cluster, with 16 nodes (B16/AMD
Athlon 1800+; BioComp, S
~
aao Jos
ee do Rio Preto, SP, Brazil).
Analysis of the model. The overall stereochemical quality of the
final model for EPSP complex was assessed by the program
PROCHECK [21]. The cutoff for hydrogen bonds and salt bridges
was 3.6
AA.
Results and discussion
Primary sequence comparison
The sequence alignment of EcEPSP (template) and
MtEPSP (target) is shown in Fig. 1. The secondary
structural elements are indicated in the figure. The se-
quence MtEPSP shows 31% identity with the sequence
of EcEPSP.
Quality of the model
The Ramachandran plot for the 2 EcEPSP structures
was generated in order to compare the overal l stereo-
chemical quality of MtEPSP model against EPSP
Fig. 2. Ribbon diagram of the MtEPSP structure in the open (A) and closed (B) conformations generated by Molscript [38].
610 J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614
structures solved by biocrystallography. Analysis of the
Ramachandran plot of the MtEPSP models shows that
84.8% of the residues lie in the most favorable regions and
the remaining 10.9% in the additional allowed regions for
the closed structure, and 91.1% lie in the allowed regions
for the open structure. The same analysis for two crys-
tallographic EcEPSP structures present 91.2% of residues
in the most favorable, 8.5% additional allowed regions,
and 0.3% generously allowed regions. The overall rating
for the MtEPSP model is slightly poorer than the one
obtained for the two structures of EcEPSP. However, it
has over 90% of the residues in the most favorable regions.
Overall description
Figs. 2A an d B show the secondary structure of both
models of MtEPSP synthase. EPSP synthase is an a=b
protein consisting of a mixed b sheet surrounded by a
helices. The structure folds into two similar domains,
which are approximately hemis pheric, each with a
radius of about 21
AA. The domains are linked by two
crossover chain segments with both the amino and
carboxyl termini of the protein in the lower domain. The
two flat surfa ces of the hemispheres, which in projection
form a “V,” are almost normal and accommodate the
amino termini of the six helices in each domain. The
helical macrodipolar effects create a potential well that
facilitates the binding of glyphosate and S3P [13]. The
domains are composed of three copies of a bababb-
folding unit. The four-stranded b-sheet structures con-
tain both parallel and antiparallel strands and the heli-
ces are parallel. In the closed structure the glyphosate
and S3P molecules are bound in the interdomain cleft,
promoting the closure of the structure.
Molecular hinge
Both crystal structures of EcEPSP synthase and the
molecular models of MtEPSP synthase reveal the pres-
ence of two domains in which the domains, designated
as domains A and B, undergo a translation and rotation
from the open to the closed co nformation. The center of
mass of the two domains moves 4.6
AA from the open to
the closed structure, for MtEPSP synthase. In the open
and closed structures of MtEPSP synthase an angular
difference at the interface of approximately 69° between
the two domains results in different conformations, as
illustrated in Fig. 3. Superposition of Ca atoms of the
constituent domains of each structure results in rms
differences of 0.20 and 0.39
AA for EcEPSP and MtEPSP
synthases, respectively. The individual domains from
both structures were superposed using the Ca atoms in
the four major a-helices (residues 240–252, 270–280,
343–355, and 385–395). The low rmsd values, observed
in the superpositions, indicate that the alternative
conformation of the two-domain structures is due to
relative motions of structurally conserved domains.
However, the difference in conformation suggests a
Fig. 3. Molecular hinge. a-Carbon traces of the superimposed structures of MtEPSP synthase from the open conformation (dark line) and closed
conformation (light line).
J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614 611
degree of flexibility in the interdomain contact at the
interface region, which acts as a hinge between the two
monomers resulting in “open” and “close d” forms of the
EPSP synthase structure. Similar molecular hinge is
observed in PLA
2
structures [22–24].
3-Phosphoshikimate and glyphosate binding
Fig. 4 shows the active site of MtEPSP synthase. The
glyphosate and S3P are close to each other and the
5-hydroxyl group of S3P is hydrogen bonded to the ni-
trogen atom of glyphosate with a distance of 2.82
AA.
Analysis of the molecular surface of the MtEPSP syn-
thase indicates that both ligands are buried in between
the two domains, with a contact area of 29 and 134
AA
2
for
glyphosate and 3-phosphoshikimate, respectively. The
electrostatic potential surface of the MtEPSP synthase
complexed with glyphosate and S3P calculated with
GRASP [25] indicates the presence of some charge
complementarity between ligands and enzyme (figure not
shown), nevertheless most of the residues in the binding
pocket are hydrophobic in all structures (see Fig. 4).
The analysis of the MtEPSP synthase complex indi-
cates a total of 22 and 15 hydrogen bonds between the
enzyme and glyphosate and S3P, respectively. Tables 1
and 2 show the intermolecular hydrogen bonds between
Fig. 4. Stereo view of the interdomain interface region illustrating the binding of glyphosate and S3P in the MtEPSP synthase structure.
Table 1
Hydrogen bonds between EPSP:glyphosate
Hydrogen bonds between active site and inhibitor
Distance (
AA)
Glyphosate EPSP
O4 Lys23 NZ 2.98
O1 Lys23 NZ 2.84
N1 Lys23 NZ 3.46
O3 Leu94 O 3.10
O2 Gly96 N 3.53
O3 Gly96 N 2.87
O3 Arg124 NH1 2.87
O2 Arg124 NH2 2.81
O3 Arg124 NH2 3.60
O2 Gln169 NE1 3.58
O2 Gln169 NE2 2.88
O5 Glu311 OE1 2.96
N1 Glu341 OE1 3.35
N1 Glu341 OE2 2.88
O4 Glu341 OE2 3.59
O5 Arg344 NH1 2.81
O5 Arg344 NH2 3.02
O4 His384 NE2 3.49
O4 Arg385 NE 3.59
O5 Arg385 NE 2.75
O4 Arg385 NH2 3.12
O3 Lys410 NZ 2.95
Table 2
Hydrogen bonds between EPSP:shikimate 3-phosphate
Hydrogen bonds between active site and ligand
Distance (
AA)
Shikimate
3-phosphate
EPSP
O3 Lys23 NZ 2.81
O5 Ser24 OG 2.70
O5 Arg28 NH1 2.86
O4 Arg28 NH2 2.78
O5 Arg28 NH2 3.60
O6 Ser167 OG 3.40
O8 Ser167 OG 2.70
O6 Ser168 OG 2.69
O6 Ser168 N 2.84
O1 Gln169 NE2 3.49
O6 Ser196 OG 3.59
O7 Ser196 OG 2.69
O2 Glu311 OE2 2.65
O2 His340 NE2 3.14
O1 His340 NE2 3.57
612 J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614
the enzyme and the ligands. Most of the hydrogen bonds
between glyphosate and EPSP synthase involve residues
Lys23, Arg124, Glu341, and Arg385, the same residues
involved in hydrogen bonds between EcEPSP and
glyphosate. Furthermore, the alignment of 69 EPSP
synthase sequences indica tes that the main residues in-
volved in intermolecular hydrogen bonds, between the
glyphosate and the enzyme, are conserved in all se-
quences (figure not shown). Such observation suggests
that inhibitors derived from glyphosate will be able to
inhibit most or even all EPSP synthases, since specificity
and affinity between enzyme and its inhibitor depend
on directional hydrogen bonds and ionic interactions,
as well as on shape complementarity of the contact
surfaces of both partners [26–36]. Further inhibition
experiments may confirm this prediction.
Acknowledgments
This work was supported by Grants from FAPESP (SMOLBNet,
Proc. 01/07532-0), CNPq, CAPES and Instituto do Mil
^
eenio (CNPq-
MCT) to DSS and LAB. WFA (CNPq, 300851/98-7) and MSP (CNPq,
500079/90-0) are researchers for the National Research Council.
References
[1] World Health Organization. Global Tuberculosis Control. WHO
Report 2001. Geneva, Switzerland, WHO/CDS/TB/2001.287.
[2] N.E. Dunlap, J. Bass, P. Fujiwara, P. Hopewell, C.R. Horsburgh
Jr., M. Salfinger, P.M. Simione, Diagnostic standards and
classification of tuberculosis in adults and children, Am. J. Respir.
Crit. Care Med. 161 (2000) 1376–1395.
[3] J.D. McKinney, W.R. Jacobs Jr., B.R. Bloom, Persisting prob-
lems in tuberculosis, in: R.M. Krause (Ed.), Emerging Infections,
Academic Press, New York, 1998, pp. 51–146.
[4] C.J.L. Murray, Issues in operational, social, and economic
research on tuberculosis, in: B.R. Bloom (Ed.), Tuberculosis:
Pathogenesis, Protection, and Control, ASM Press, Washington,
DC, 1994, pp. 583–622.
[5] B.R. Bloom, C.J.L. Murray, Tuberculosis: commentary on a
reemergent killer, Science 257 (1992) 1055–1064.
[6] L.A. Basso, J.S. Blanchard, Resistance to antitubercular drugs,
Adv. Exp. Med. Biol. 456 (1998) 115–144.
[7] R. Bentley, The shikimate pathway
—A metabolic tree with many
branches, Crit. Rev. Biochem. Mol. Biol. 25 (1990) 307–384.
[8] C. Ratledge, Nutrition, growth and metabolism, in: C. Ratledge,
J.L. Stanford (Eds.), The Biology of the Mycobacteria, vol. 1,
Academic Press, London, 1982, pp. 185–271.
[9] S.T. Cole, R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D.
Harris, S.V. Gordon, K. Eiglmeier, S. Gas, C.E. Barry III, F.
Tekaia, K. Badcock, D. Basham, D. Brown, T. Chillingworth, R.
Connor, R. Davies, K. Devlin, T. Feltwell, S. Gentles, N. Hamlin,
S. Holroyd, T. Hornsby, K. Jagels, B.G. Barrel, Deciphering the
biology of Mycobacterium tuberculosis from the complete genome
sequence, Nature 393 (1998) 537–544.
[10] H.C. Steinr
€
uucken, N. Amrhein, The herbicide glyphosate is a
potent inhibitor of 5-enolpyruvyl-shikimic acid-3-phosphate syn-
thase, Biochem. Biophys. Res. Commun. 94 (4) (1980) 1207–1212.
[11] F. Roberts, C.W. Roberts, J.J. Johnson, D.E. Kyle, T. Krell, J.R.
Coggins, G.H. Coombs, W.K. Milhous, S. Tzipori, D.J. Fergu-
son, D. Chakrabarti, R. McLeod, Evidence for the shikimate
pathway in apicomplexan parasites, Nature (London) 395 (6699)
(1998) 801–805.
[12] E. Sch
€
oonbrunn, S. Eschenburg, W.A. Shuttleworth, J.V. Schloss,
N. Amrhein, J.N.S. Evans, W. Kabsch, Interaction of the
herbicide glyphosate with its target enzyme 5-enolpyruvylshiki-
mate 3-phosphate synthase in atomic detail, Proc. Natl. Acad. Sci.
USA 98 (2001) 1376–1380.
[13] W.C. Stallings, S.S. Abdel-Meguid, L.W. Lim, H.-S. Shieh, H.E.
Dayringer, N.K. Leimgruber, R.A. Stegeman, K.S. Anderson,
J.A. Sikorski, S.R. Padgette, G.M. Kishore, Structure and
topological symmetry of the glyphosate target 5-enol-pyruvyl-
shikimate-3-phosphate synthase: a distinctive protein fold, Proc.
Natl. Acad. Sci. USA 88 (1991) 5046–5050.
[14] J.S. Oliveira, C.A. Pinto, L.A. Basso, D.S. Santos, Cloning and
overexpression in soluble form of functional shikimate kinase
and 5-enolpyruvylshikimate 3-phosphate synthase enzymes from
Mycobacterium tuberculosis, Protein Express. Purif. 22 (2001)
430–435.
[15] A. Sali, T.L. Blundell, Comparative protein modelling by satis-
faction of spatial restraints, J. Mol. Biol. 234 (1993) 779–815.
[16] A. Sali, J.P. Overington, Derivation of rules for comparative
protein modeling from a database of protein structure alignments,
Protein Sci. 3 (9) (1994) 1582–1596.
[17] A. Sali, L. Potterton, F. Yuan, H. van Vlijmen, M. Karplus,
Evaluation of comparative protein modeling by MODELLER,
Proteins 23 (3) (1995) 318–326.
[18] A. Sali, Modeling mutations and homologous proteins, Curr.
Opin. Biotechnol. 6 (4) (1995) 437–451.
[19] B.R. Brooks, R.E. Bruccoleri, B.D. Olafson, D.J. States, S.
Swaminathan, M. Karplus, CHARMM: a program for macro-
molecular energy minimization and dynamics calculations, J.
Comp. Chem. 4 (1983) 187–217.
[20] W. Braun, N. Go, Calculation of protein conformations by
proton–proton distance constraints. A new efficient algorithm, J.
Mol. Biol. 186 (3) (1985) 611–626.
[21] R.A. Laskowski, M.W. MacArthur, D.K. Smith, D.T. Jones,
E.G. Hutchinson, A.L. Morris, D. Naylor, D.S. Moss, J.M.
Thorton, PROCHECK v.3.0
—Program to Check the Stereo-
chemistry Quality of Protein structures—Operating Instructions,
1994.
[22] W.F. De Azevedo Jr., R.J. Ward, F.R. Lombardi, J.R. Giglio, A.M.
Soares, M.R.M. Fontes, R.K. Arni, Crystal structure of myotoxin-
II: a myotoxic phospholipase A2-homologue from Bothrops mooj-
eni venom, Protein Peptide Lett. 4 (5) (1997) 329–334.
[23] W.F. De Azevedo Jr., R.J. Ward, J.M. Gutierrez, R.K. Arni,
Structure of a Lys49 phospholipase A2 homologue isolated from
the venom of Bothrops nummifer (jumping viper), Toxicon 37 (2)
(1999) 371–384.
[24] W.F. De Azevedo Jr., R.J. Ward, F. Canduri, A. Soares, J.R.
Giglio, R.K. Arni, Crystal structure of piratoxin-I: a calcium-
independent, myotoxic phospholipase A
2
-homologue from Bothr-
ops pirajai venom, Toxicon 36 (10) (1998) 1395–1406.
[25] A. Nicholls, K.A. Sharp, B. Honig, Protein folding and associ-
ation: insights from the interfacial and thermodynamic properties
of hydrocarbons, Proteins 11 (4) (1991) 281–296.
[26] F. Canduri, L.G.V.L. Teodoro, C.C.B. Lorenzi, V. Hial, R.A.S.
Gomes, J. Ruggiero Neto, W.F. de Azevedo Jr., Crystal structure
of human uropepsin at 2.45
AA resolution, Acta Crystallogr. D 57
(2001) 1560–1570.
[27] W.F. De Azevedo Jr., H.J. Mueller-Dieckmann, U. Schulze-
Gahmen, P.J. Worland, E. Sausville, S.-H. Kim, Structural basis
for specificity and potency of a flavonoid inhibitor of human CDK2,
a cell cycle kinase, Proc. Natl. Acad. Sci. USA 93 (7) (1996) 2735–
2740.
[28] W.F. De Azevedo Jr., S. Leclerc, L. Meijer, L. Havlicek, M.
Strnad, S.-H. Kim, Inhibition of cyclin-dependent kinases by
J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614 613
purine analogues: crystal structure of human CDK2 complexed
with roscovitine, Eur. J. Biochem. 243 (1997) 518–526.
[29] W.F. De Azevedo Jr., F. Canduri, V. Fadel, L.G.V.L. Teodoro,
V. Hial, R.A.S. Gomes, Molecular model for the binary complex
of uropepsin and pepstatin, Biochem. Biophys. Res. Commun.
287 (1) (2001) 277–281.
[30] W.F. De Azevedo Jr., F. Canduri, N.J.F. da Silveira, Structural
basis for inhibition of cyclin-dependent kinase9 by Flavopiridol,
Biochem. Biophys. Res. Commun. 293 (1) (2002) 566–571.
[31] S.-H. Kim, U. Schulze-Gahmen, J. Brandsen, W.F de Azevedo Jr.,
Structural basis for chemical inhibitor of CDK2, Prog. Cell Cycle
Res. 2 (1996) 137–145.
[32] W.F. De Azevedo, R.T. Gaspar, F. Canduri, J.C. Camera, N.J.F.
da Silveira, Molecular model of cyclin-dependent kinase 5
complexed with roscovitine, Biochem. Biophys. Res. Commun.
297 (2002) 1154–1158.
[33] W.F. De Azevedo Jr., J.S. de Oliveira, L.A. Basso, M.S. Palma,
J.H. Pereira, F. Canduri, D.S. Santos, Molecular model of
shikimate kinase from Mycobacterium tuberculosis, Biochem.
Biophys. Res. Commun. 295 (1) (2002) 142–148.
[34] W.F. De Azevedo, F. Canduri, D.M. Santos, R.G. Silva, J.S.
Oliveira, L.P.S. Carvalho, L.A. Basso, M.A. Mendes, M.S. Palma,
D.S. Santos, Crystal structure of human purine nucleoside
phosphorylase at 2.3
AA resolution, Biochem. Biophys. Res.
Commun. 308 (3) (2003) 545–552.
[35] D.M. Santos, F. Canduri, J.H. Pereira, M.V.B. Dias, R.G. Silva,
M.A. Mendes, M.S. Palma, L.A. Basso, W.F. de Azevedo, D.S.
Santos, Crystal structure of human purine nucleoside phosphor-
ylase complexed with acyclovir, Biochem. Biophys. Res. Com-
mun. 308 (3) (2003) 553–559.
[36] W.F. De Azevedo, F. Canduri, D.M. Dos Santos, J.H. Pereira,
M.V.B. Dias, R.G. Silva, M.A. Mendes, L.A. Basso, M.S. Palma,
D.S. Santos, Structural basis for inhibition of human PNP by
immucillin-H, Biochem. Biophys. Res. Commun. 309 (2003) 922–
927.
[37] D.G. Higgins, A.J. Bleasby, R. Fuchs, CLUSTAL V: improved
software for multiple sequence alignment, Comput. Appl. Biosci.
8 (2) (1992) 189–191.
[38] P. Kraulis, MOLSCRIPT: a program to produce both detailed
and schematic plots of proteins, J. Appl. Cryst. 24 (1991) 946–950.
614 J.H. Pereira et al. / Biochemical and Biophysical Research Communications 312 (2003) 608–614
Estudo Estrutural das Enzimas da Via Metabólica do Ácido Chiquímico
Anexo 5
1
Molecular modeling and CD analysis of chorismate synthase
José Henrique Pereira
a,1
, Marcio Vinicius Bertacine Dias
a,1
, Fernanda Canduri
a,b
, Fernanda
Ely
c
, João Ruggiero Neto
a
, Jeverson Frazzon
c
, Luiz Augusto Basso
c
, Mário Sérgio Palma
b,d
,
Diógenes Santiago Santos
e*
, Walter Filgueira de Azevedo Jr
a,b*
a
Programa de Pós-Graduaçãoem Biofísica Molecular - Departamento de Física, UNESP,
São José do Rio Preto, SP 15054-000, Brasil
b
Centro de Toxicologia Aplicada, Instituto Butantan, São Paulo, SP 05503-900, Brasil
c
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
d
Laboratório de Biologia Estrutural e Zooquímica, CEIS/Departamento de Biologia,
Instituto de Biociências, UNESP, Rio Claro, SP 13506-900, Brasil
e
Faculdade de Farmácia/Instituto de Pesquisas Biomédicas, Pontifícia Universidade
Católica do Rio Grande do Sul, Porto Alegre, RS, Brasil
1
J. H. Pereira and M. V. B. Dias contributed equally to this work
*
Corresponding author
E-mail adresses of the corresponding authors: walterfa@df.unesp.br
(W.F.de Azevedo Jr.)
and diogenes@pucrs.br
(D.S. Santos).
Keywords: Chorismate synthase, Circular dichroism, Molecular modeling, Mycobacterium
tuberculosis, Drug design.
2
Abstract
Tuberculosis is considerate a worldwide health problem mainly due to co-infection
with HIV and proliferation of multi-drug-resistant strains. 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 (EPSP) to chorismate. Here, it is described molecular modeling of chorismate
synthase from Mycobacterium tuberculosis complexed with EPSP and FMN (Flavin-
mononucleotide), and analysis of circular dichrosim (CD). The atomic coordinates of the
structure of CS from Streptococcus pneumoniae were used as template for starting the
modeling of MtCS. The molecular model presents good stereochemistry quality and CD
analysis confirms the prediction of the secondary structure. The molecular modeling of
MtCS should provide a structural framework on which the design of specific inhibitors may
be based.
3
Introduction
Tuberculosis resurged in the mid-1980s and now kills approximately 3 million
people a year. The reemergence of tuberculosis as a public health threat, the high
susceptibility of HIV-infected persons, and the proliferation of multi-drug-resistant strains
have created a need to develop new drugs. Potential targets for the development of new
therapies are the enzymes of shikimate pathway, because they are essential for bacteria,
fungi and apicomplexan parasites, but absent from mammals [1,2]. This pathway consists
of seven enzymes that catalyze the sequential conversion of erythrose-4-phosphate and
phosphoenol pyruvate to chorismic acid (chorismate), the common precursor of aromatic
compounds [3]. Chorismate synthase catalyses the most interesting and unusual reaction of
the entire pathway, the conversion of 5-enolpyruvyl-3-shikimate phosphate (EPSP) to
chorismate, via the 1,4-antielimination of phosphate and a proton, a reaction that is unique
in nature [4,5]. The enzyme has an absolute requirement for reduced FMN as a cofactor,
although the catalyzed reaction involves no net redox change. The role of the reduced FMN
in catalysis still not is all clear. However, recent detailed kinetic and bioorganic approaches
have fundamentally advanced our understanding of the mechanism of action of chorismate
synthase [6].
Chorismate synthase is particularly attractive as an anti-infective target as
chorismate lies at a metabolic node, being the precursor for five distinct pathways. It is
necessary for the production of aromatic amino acids, para-aminobenzoic acid (PABA),
folate, and for other cyclic metabolites such as ubiquinone and menaquinone [7]. Much of
the folate pathway is also absent in mammals, and enzymes within it have therefore been
successfully exploited as targets for antibacterial chemotherapy, as exemplified in the
4
inhibition of dihydrofolate reductase (DHFR) by trimethoprim. Despite of the
crystallization of Chorismate Synthase from Mycobacterium tuberculosis has been reported
[8], the structure was not determined, yet.
The present paper describes the molecular model of M. tuberculosis chorismate
synthase (MtCS) complexed with FMN and EPSP, and analysis of circular dichroism
spectra of protein in the presence and absence of co-factor.
Methods
Molecular modeling
For modeling of the MtCS was used restrained-based modeling implemented in the
program MODELLER [9]. The modeling procedure begins with an alignment of the
sequence to be modeled (target) with related known three-dimensional structures
(templates). This alignment is usually the input to the program. The output is a three-
dimensional model for the target sequence containing all main-chain and side-chain non-
hydrogen atoms. The atomic coordinates of the crystallographic structure of Streptococcus
pneumoniae chorismate synthase (SpCS) (PDB access code: 1QXO) [10] solved at 2.0 Å
resolution were used as starting model for modeling of the CS from M. tuberculosis
complexed with FMN and EPSP. The atomic coordinates of all waters were removed from
the SpCS structure. Several slightly different models can be calculated by varying the initial
structure. A total of 1000 models were generated for MtCS, and the final model was
selected based on stereochemical quality.
5
Analysis of the model
The overall stereochemical quality of the final model for the complex MtCS-FMN-
EPSP was assessed by the program PROCHECK [11]. Molecular models were superposed
using the program LSQKAB from CCP4 [12]. The cutoff for hydrogen bonds and salt
bridges were 3.5Å. The contact surfaces for the binary complexes were calculated using
AREAIMOL and RESAREA [12]. The root mean square deviation (rmsd) differences from
ideal geometries for bond lengths and bond angles were calculated with X-PLOR [13]. The
G-factor is essentially just log-odds score based on the observed distributions of the
stereochemical parameters. It is computed for the following properties: torsion angles (the
analyses provided the observed distributions of φ-ψ, χ1– χ2, χ-1, χ-3, χ-4, and ω values for
each of the 20 amino acid types) and covalent geometry (for the main-chain bond lengths
and bond angles). These values’ averages were calculated using PROCHECK [14]. The
Verify-3D measures the compatibility of a protein model with its sequence, and it was
calculated using 3D profile program [14,15].
Circular dichroism
The cloning, protein expression and purification of the MtCS was performed as
described in Dias et al. [8]. Circular dichroism (CD) was measured over the range 195-260
nm, using a Jasco-710 spectropolarimeter (Jasco, Tokyo, Japan) coupled to a Neslab
RTE111 circulating water bath. Spectra were obtained at 25
°
C using cells with a path
length of 0.2 cm. Data points were recorded at a scan speed of 20 nm/min, bandwidth 1.0
nm, 1 s response and 0.1 nm resolution. Five repeat scans were accumulated to obtain the
6
final averaged spectra. The deconvolution was performed using program of CD analysis by
DICROPROT [16].
Results and discussion
Quality of the model
The sequence alignment of MtCS (target) and SpCS (template) shows 44.1% of
identity (Figure 1). This degree of primary sequence identity indicates that the
crystallographic structure of SpCS is good model to be used as template for modeling of
MtCS.
The Ramachandran plot was used to compare the overall stereochemical quality of
MtCS model against SpCS structure solved by crystallography. Analysis of the
Ramachandran plot of the MtCS model shows that 94.6% of the residues lies in the most
favorable regions and the remaining 5.4% in the allowed regions. The same analysis for
crystallographic SpCS structure present 93.1% of residues in the most favorable and 6.9%
allowed regions. The overall rating for the MtCS model is slightly better than the obtained
for the structure of SpCS, however both structures shows excellent stereochemical quality.
The r.m.s.d. of molecular modeling MtCS on crystallography SpCS, in the coordinates of
Cα is 0.13 Å. The r.m.s.d. value of bond lengths and bond angles, the average G-factor
and Verify 3D values are shown in Table 1.
Overall description
The structure of MtCS belongs to the
α
/
β
family. The dominant structural topology
of CS is a beta-alpha-beta sandwich, in which each monomer of CS consists of a central
helical core, helix order
α
1,
α
6,
α
12 and
α
9, sandwiched between two four-stranded
7
antiparallel beta sheets, strand order β1, β2, β6 and β3 for sheet “one” and strand order β7,
β
8,
β
14 and
β
9 for sheet “two” (Figure 2). These layers are packed to form a compact
structure. It is clear from the topology of MtCS that the protein core has a pseudo 2-fold
symmetry, although the molecule as a whole does not.
The MtCS active site is at the interface between
β
sheet 2 and 1 end of the internal
layer of α helices. At this point, close to the internal pseudo 2-fold axis, the two centrals
helix diverges to leave a small hydrophobic pocket. The others secondary structure
elements that participate of active site are the loops L1, L4, L10, L16, L25 and L27. FMN
and EPSP are closely associated with each other and a considerable degree of the binding
surface of each ligand is in contact with the other.
FMN Binding
The electrostatic potential at the molecular surface generated for program GRASP
[17] shows that FMN binding site is highly positive. The six flexible loops are clustered
around or near bond cofactor. These flexible regions are rich in strictly or highly conserved
residues. Analysis of the molecular surface of the model of MtCS and the structure of SpCS
indicates that FMN presents a contact area of the 261.0 Å
2
and 236.0 Å
2
, respectively.
The molecule of FMN makes few specific polar interactions with the protein, the
majority of which are contacts between the hydroxyl and phosphate oxygens of the ribityl
chain with loops L10, L16, and L25. The ribityl-phosphate portion of FMN is buried deeply
into the enzyme, while the entire re face and much of the ortho-xylyx ring of FMN contact
the surface of the FMN binding site. These few residues that form the binding surface can
provide additional flexibility necessary for the protein to accommodate reduced FMN, in
which the isoalloxazine ring system may be bent into a “butterfly”. Furthermore, the lack of
8
specific contacts between CS and the FMN ribityl chain would also allow some flexibility
of the FMN molecule [10]. The figure 3a shows the hydrogen bonds between the molecule
de FMN and MtCS.
The FMN molecule seems not cause structural change in the MtCS as well as of the
planar oxidized and highly puckered reduced flavins can be accommodated in the same
active site without the need of changes in protein conformation [18].
The role of FMN in structure of chorismate synthase seems to be the stabilization
of any electron-deficient intermediate of EPSP, indeed, the N5 of FMN is positioned
directly below the C6-pro-R hydrogen of EPSP and is ideally positioned to remove this
proton from the transient intermediate. Therefore, the flavin could be able to donate an
electron and subsequently accept a hydrogen atom.
EPSP Binding
There are several polar interactions between EPSP and the MtCS. The EPSP site is
an environment extremely basic, with the six arginine (R40, R41, R46, R49, R112, R139)
and three histidine (H11, H115, H339) residues concentrated in a small, tightly enclosed
binding site in the three corners of the site. The contact area for a molecule of EPSP is
119.0 Å
2
and 163.0 Å
2
, for molecular model of MtCS and structure of SpMS, respectively.
Recently, it was demonstrated by site-directed mutagenesis the role of the H11 and H115
[19]. The H115 serves to protonate the reduced FMN while the H11 protonates the leaving
phosphate group of the substrate.
The analysis of the MtCS indicates a total of 11 interactions between the enzyme
and EPSP (Figure 3b). The EPSP molecule can be divided in three moietys: carboxyl, enol-
pyruvyl, phosphate. The enol-pyruvyl moiety interacts with three arginines residues (R40,
9
R46 and R139). The phosphate group forms interactions between H11, R49, E81 and
R341. The carboxyl group forms hydrogen bonds between H115 and A138.
The molecule of EPSP can be responsible to induce changes in the structure of
MtCS, unlike the FMN can be accommodated in the active site without change in protein
conformation. The physical studies and spectroscopic in Escherichia coli showed that CS
undergoes a major structural change when both FMN and EPSP, are bound [20, 21]. The
binding of EPSP must cause protein conformational changes involving movements in some
of the flexible loops regions resulting in a more apolar environment for the bound co-factor.
The CD spectrum of MtCS in the presence and absence of the FMN is shown in
figure 5. The percentage of secondary structure for the two independent measurements and
for model can be observed of table 2. The percentage of the secondary structure elements
determined by CD is in good agreement with the molecular model of MtCS. These results
show that the prediction of secondary structure by homology molecular modeling is
reliable. The presence of oxidized FMN seems to cause minor effects on the CD spectrum.
These results are like those obtained for chorismate synthase from Escherichia coli, where
it was used oxidized FMN and EPSP. In the presence of substrate and co-factor the
chorismate synthase from E. coli was not observed significant variation in the CD spectrum
[5].
The EPSP molecule and FMN provide interesting targets for structure-based drug
design, not alone against M. tuberculosis, but also, for other pathogen, once that specificity
and affinity between enzyme and its inhibitors depend on directional hydrogen bonds and
ionic interactions, as well as on shape complementarily of the contact surfaces of both
partners [22-25]. Inhibitors could compete with the substrate and co-factor in the binding to
10
the enzyme, leaving the enzyme in unproductive conformations. Furthermore, the 5-
deazaflavin [26] an flavin analog in which the nitrogen at the 5-position is replaced by
carbon, and (6R)-6-fluoro-EPSP [27] in which the (6R)-hydrogen of EPSP is replaced by
fluorine, results a lack of activity of the chorismate synthase [6, 28]. These are examples of
molecules that can be rationally modified to optimize the inhibition of the enzyme.
Acknowledgements
This work was supported by grants from FAPESP (SMOLBNet 01/07532-0,
02/04383-7, 04/00217-0), CNPq, CAPES and Instituto do Milênio (CNPq-MCT). WFA
(CNPq, 300851/98-7), and MSP (CNPq, 300337/2003-5) are researchers for the Brazilian
Council for Scientific and Technological Development.
References
[1] R. Bentley, The shikimate pathway—A metabolic tree with many branches, Crit. Rev.
Biochem. Mol. Biol. 25 (1990), 307–384.
[2] F. Roberts, C.W. Roberts, J.J. Johnson, D.E. Kyle, T. Krell, J.R. Coggins, G.H.
Coombs, W.K. Milhous, S. Tzipori, D.J. Ferguson, D. Chakrabarti and R. McLeod,
Evidence for the shikimate pathway in apicomplexan parasites, Nature (London)
395
6699
(1998), 801–805.
[3] K.M. Herrmann, L. M. Weaver, The shikimate pathway, Annu. Rev. Plant. Physiol.
Plant. Mol. Biol.
50
(1999), 473-503.
[4] S. Bornemann, Flavoenzymes that catalyse reactions with no net redox change. Nat.
Prod. Rep. 19, (2002), 761–772.
11
[5] P. Macheroux, E. Schönbrunn, D.I. Svergun, V.V. Volkov, M.H.J. Koch, S.
Bornemann, R.N.F. Thorneley,. Evidence for a major structural change in Escherichia coli
chorismate synthase induced by flavin and substrate binding. Biochem. J. 335 (1998) 319–
327.
[6] P. Macheroux, J. Schmid, N. Amrhein, A. Schaller, A unique reaction in a common
pathway: mechanism and function of chorismate synthase in the shikimate pathway. Planta
207, (1999), 325–334.
[7] C. Ratledge, Nutrition, growth and metabolism, in: C. Ratledge, J. L. Stanford, (Eds.),
The biology of the mycobacteria, Academic Press, London, 1982, Vol 1, 185-271.
[8] M. V. B. Dias, F. Ely, F. Canduri, J. H. Pereira, J. Frazzon, L. A. Basso, M. S. Palma,
W. F. De Azevedo Jr., D. S. Santos, Crystallization and preliminary X-ray crystallographic
analysis of chorismate synthase from Mycobacterium tuberculosis. Acta Crystal. Section
D60, (2004), 2003-2005.
[9] A. Sali, T.L. Blundell, Comparative protein modelling by satisfaction of spatial
restraints, J. Mol. Biol. 234 (1993), 779–815
[10] J. Maclean, S. Ali, The structure of chorismate synthase reveals a novel flavin binding
site fundamental to a unique chemical reaction, Structure 11 (2003), 1499-1511.
[11] R.A. Laskowski, M.W. MacArthur, D.K. Smith, D.T. Jones, E.G. Hutchinson, A.L.
Morris, D. Naylor, D.S. Moss, J.M. Thorton, PROCHECK V.3.0 program to check the
stereo-chemistry quality of protein structures operating instructions, 1994.
[12] Collaborative Computational Project No. 4, The CCP4 suite: program for protein
crystallography, Acta Crystallogr. D 50 (1994) 760–763.
12
[13] A.T. Brünger, X-PLOR Version 3.1: A System for Crystallography and NMR, Yale
University Press, New Haven (1992).
[14] J.U. Bowie, R. Luthy, D. Eisenberg, A method to identify protein sequences that fold
into a known three-dimensional structure, Science 253 (1991) 164–170
[15] R. Luthy, J. Bowie, D. Eisenberg, Assessment of protein models with three-
dimensional profiles, Nature 356 (1992), 83–85.
[16] G. Deléage, C. Geourjon, An interactive graphic program for calculating the
secondary structures content of proteins from circular dichroism spectrum. Comp. Appl.
Biosc. 9 (1993), 197-199.
[17] A. Nicholls, K.A. Sharp, B. Honig, Protein folding and association: insights from the
interfacial and thermodynamic properties of hydrocarbons, Proteins 11 (1991) (4) 281–296.
[18] B.W. Lennon, C.H. Williams Jr., M.L. Ludwig, Crystal structure of reduced
thioredoxin reductase from Escherichia coli: structural flexibility in the isoalloxazine ring
of the flavin adenine dinucleotide cofactor. Protein Sci. 8 (1999) 2366–2379.
[19] K. Kitzing, S. Auweter, N. Amrhein, P. Macheroux, Mechanism of chorismate
synthase. role of the two invariant histidine residues in the active site. J. Biol. Chem.
279
(10), 9451-9461.
[20] P. Macheroux, J. Petersen, S. Bornemann, D. Lowe, R. N. F. Thorneley, Binding of
the oxidized, reduced, and radical flavin species to chorismate synthase. An investigation
by spectrophotometry, fluorimetry, and electron paramagnetic resonance and electron
nuclear double resonance spectroscopy. Biochemistry 35(5) (1996), 1643 – 1652.
13
[21] P. Macheroux
,
S. Bornemann
,
S. Ghisla,
R. N. F. Thorneley, 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
(1996)
,
25850-25858.
[22] W.F. De Azevedo Jr., S. Leclerc, L. Meijer, L. Havlicek, M. Strrnad, S.-H. Kim,
Inhibition of cyclin-dependent kinases by purine analogues: crystal structure of human
CDK2 complexed with roscovitine, Eur. J. Biochem. 243 (1997) 518–526.
[23] J.H. Pereira, F. Canduri, J.S. de Oliveira, N.J. da Silveira, L.A. Basso, M.S. Palma,
W.F. deAzevedo, Jr., D.S. Santos, Structural bioinformatics study of EPSP synthase from
Mycobacterium tuberculosis. Biochem. Biophys. Res. Commun. 312 3 (2003), pp. 608–614.
[24] W.F. De Azevedo, F. Canduri, D.M. dos Santos, J.H. Pereira, M.V.B. Dias, R.G.
Silva, M.A. Mendes, L.A. Basso, M.S. Palma, D.S. Santos, Structural basis for inhibition of
human PNP by immucillin-H. Biochem. Biophys. Res. Commun. 309 (2003), pp. 922–927.
[25] W.F. De Azevedo, Jr., H.J. Mueller-Dieckmann, U. Schulze-Gahmen, P.J. Worland, E.
Sausville and S.-H. Kim , Structural basis for specificity and potency of a flavonoid
inhibitor of human CDK2, a cell cycle kinase. Proc. Natl. Acad. Sci. USA 93 7 (1996),
2735–2740.
[26] C. T Lauhon, P. A. Bartlett
,
Substrate analogs as mechanistic probes for the
bifunctional chorismate synthase from Neurospora crassa. Biochemistry 33(47) (1994)
14100-14108.
[27] J. K. Sutherland, W. J. Watkins, J. P. Bailey, A. K. Chapman, G. M. Davies, The
synthesis of 6a- and 6b-fluoroshikimic acids. J. Chem. Soc.Chem. Commun. (1989) 1386–
1387.
14
[28] S. Bornemann, D.M. Lawson, R.N. Thorneley, A branch point in chorismate synthase
research. Structure 11(12) (2003) 1463-1465.
[29] R. Koradi, M. Billeter and K. Wüthrich, MOLMOL: A program for display and
analysis of macromolecular structures. J. Molecular Graphics, 14(1) (1996), 51-55.
[30] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity
of progressive multiple sequence alignment through sequence weighting, positions-specific
gap penalties and weight matrix choice. Nucleic Acids Res. 22 (1994), 4673–4680.
15
Table 1
.
Summary of structural results for MtCS and SpCS.
Analysis of the model of MtCS and structure of SpCS
Proteins 3D Profile
a
G-factor
b
rmsd from ideal
geometry
Total
Score
Ideal
Score
S
Ideal
Score
Torsion
Angles
Covalent
Geometry
Global Bond
Lengths (Ǻ)
Bond Angles
(˚)
MtCS 157.49 179.29 0.88 S -0.07 -0.34 -0.16 0.022 3.172
SpCS 163.85 171.92 0.95 S 0.00 -0.16 -0.03 0.027 2.558
a
Total Score: is the sum of the 3D-1D scores (statistical preferences) of each residue present in protein. Ideal
Score: S
ideal
= exp(-0.83+1.008xln(L)); where L is number of amino acids. S
ideal
Score: is compatibility of the
sequence with their 3D structure. It is obtained Total Score / Ideal Score. S
ideal
Score above 0.45S
ideal
.
b
Ideally, scores should be above – 0.5. Values below –1.0 may need investigation.
16
Table 2 – Percentage of the secondary structure elements for experimental data of CD and
for MtCS molecular model.
Molecular Model of MtCS
a
MtCS with FMN
b
MtCS without FMN
b
β-sheet
18 18 16
α-helix
40 44 45
Coil 42 38 39
a
Secondary structure elements were calculated using the program MOLMOL [29]
b
Secondary structure elements were calculated using the program of CD analysis
DICROPROT [16]
17
Figure 1.
The sequence alignment of MtCS and SpCS indicating the secondary structural
elements. The sequence MtCS shows 44.1% of identity with the sequence of SpCS. The
alignment was performed with the program CLUSTAL W [30].
18
Figure 2
. Ribbon diagram of the MtCS model complexed with FMN and EPSP generated
by MolMol [29].
19
Figure 3.
Diagrams showing the active sites of the MtCS; (A) active site of the FMN and
(B) active site of the EPSP.
A
B
20
Figure 4
– Superimposed circular dichroism spectra of unliganded and liganded chorismate
synthase from Mycobacterium tuberculosis.( ) Unliganded enzyme; (----) enzyme in the
presence of 22.5
μ
M flavin mononucleotide (FMN). Protein solutions of 4.5
μ
M were in 5
mM Sodium Phosphate (pH 7.8). For the spectra of liganded enzyme an aliquot of a 10 mM
ligand solution in 5 mM Sodium Phosphate (pH 7.8) was added to the protein solution and
resulting spectra were corrected for dilution. The CD spectra were also corrected for the
contribution of ligand.
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