A structural model for chorismate synthase from Mycobacterium tuberculosis in complex with coenzyme and substrate

https://doi.org/10.1016/j.compbiomed.2006.01.001Get rights and content

Abstract

The enzymes of the shikimate pathway constitute an excellent target for the design of new antibacterial agents; chorismate synthase (CS) catalyzes the last step of this pathway. The prediction of Mycobacterium tuberculosis (MTB) CS three-dimensional structure and the geometric docking of the coenzyme FMN and the substrate EPSP were performed using the crystal structure of CS from Streptococcus pneumoniae as template. Energy minimization of the whole complex showed, as expected, that most of the template interactions are preserved in the MTB structure, except for HIS11, ARG139 and GLN255. However, novel interactions involving ARG111, GLY113 and SER317 were also observed.

Introduction

The shikimate pathway is the common way for the production of various products including folic acid, vitamin K, ubiquinone and the three aromatic amino acids, tryptophan, phenylalanine and tyrosine. In bacteria, fungi, plants and apicomplexan parasites, chorismate, the final product of the shikimate pathway, is the branch point in the biosynthesis for all these products that are essential for these species. The absence of the shikimate pathway in all other species makes it an attractive target for the development of new antibacterial agents [1], [2].

Chorismate synthase (CS), the seventh and final step of the shikimate pathway, catalyzes the conversion of 5-enolpyruvylshikimate 3-phosphate (EPSP) to chorismate in the presence of a reduced flavin mononucleotide (FMN) as a coenzyme [3]. The reaction mechanism of the shikimate pathway has been studied extensively and revealed that the reaction of CS is unique in nature. The reaction involves a 1,4 elimination of phosphate and the loss of a proton of the C-6 hydrogen. This consists in the formation of the second out of three necessary double bonds to build an aromatic ring (Fig. 1). The enzyme activity requires a reduced FMN molecule which is not consumed during the reaction [4].

The function of the reduced FMN in catalysis was extensively studied. The most accepted mechanism suggests a direct role of reduced FMN in the elimination reaction. FMN transfers the electron transiently to phosphate and the substrate donates an electron to regenerate the FMN. This reaction does not involve an overall change in the redox state [3], [5].

Recently, with the first high-resolution X-ray structure of CS from Streptococcus pneumoniae (SPN) with the substrate and the coenzyme in the oxidized form [6], the structure of CS from Saccharomyces cerevisiae [7], and the structure of CS from Helicobacter pylori with the coenzyme in the reduced form [8], studies on the binding mode of substrate and coenzyme in the active site has been started.

How reduced FMN is obtained divides the CS into two classes, monofunctional and bifunctional. Bifunctional CS has an intrinsic ability to reduce flavin (specifically FMN) using NADPH. In monofunctional CS this catalytic activity is not present. The bifunctional enzyme is present in fungi and the monofunctional form in plants and bacteria [3].

The active site of SPN CS is very hydrophilic and extremely basic, with six arginine and two histidine residues. The two histidines in the active site, HIS10 and HIS110, are present in both classes of CS and across all known species, being HIS10 part of a characteristic CS signature sequence [6]. HIS110 is involved in FMN binding while HIS10 protonates the leaving phosphate group of EPSP. (The numbering of histidine residues corresponds to the alignment shown in Fig. 2; HIS10 and HIS110 match HIS17 and HIS106, respectively, in the SPN CS original sequence [9].) Site-directed mutagenesis studies indicated that mutation of HIS110 to an alanine reduces CS activity to 10%; and, in changing HIS10 to an alanine only 5% of the original activity can be observed, demonstrating the importance of these histidines in the CS reaction. These results are the same for both bifunctional and monofunctional CS enzymes [9].

The FMN coenzyme is deeply buried into the active site with EPSP blocking any possible exit. FMN makes one hydrogen bond to EPSP and a few polar interactions with the protein. On the other hand, EPSP makes several polar interactions and a few hydrophobic contacts with the protein [6].

Mycobacterium tuberculosis (MTB), the etiological agent of tuberculosis, is responsible for widespread human morbidity and mortality. The development of new effective chemotherapy should aid in the treatment and control of the disease. The World Health Organization (WHO) estimates that there were 8.8 million new cases of tuberculosis in 2002 and 4.4 million new cases in 2003. The global incidence rate of tuberculosis is growing at 1.1% per year and the number of cases at 2.4% per year. Through the years 1980–2004, a total of 81 million cases were reported to the WHO [10].

Sequencing of the MTB genome [11] has revealed a large number of individual enzymes potentially useful in drug design, including CS. Understanding the structure of MTB CS, together with its coenzyme and substrate binding modes, should facilitate the search for inhibitors of this enzyme as possible alternative agents to treat tuberculosis.

In this work, we present three-dimensional (3D) structural models for CS from MTB and evaluate their interactions with the substrate EPSP and the coenzyme FMN by geometric docking and energy minimization studies.

Section snippets

Materials and methods

The starting point of homology modeling is the identification of proteins in the protein data bank (PDB) [12] that are related to the target sequence and then select the templates. In this case, the structure prediction of CS from MTB was based on 3D structures for the homologous SPN CS protein (PDB ID: 1QXO), experimentally determined by X-ray diffraction at 2.0 Å resolution [6]. Blastp [13] was used to search for templates.

The next step is the multiple sequence alignment comparison. The

Homology modeling

CS is a protein of about 360–400 amino acid residues, except in apicomplexan parasites, where it has about 500 amino acids. It has a high degree of sequence conservation among species. The protein has three signature patterns (Fig. 2) from conserved regions rich in basic residues (mostly arginines) [26].

In the search for templates we found four candidates for modeling, CS of SPN (PDB ID: 1QX0) [6], Aquifex aeolicus (PDB ID: 1Q1L) [19], S. cerevisiae (PDB ID: 1R53) [7] and H. pylori (PDB ID:

Conclusions

We have obtained a 3D structural model of MTB CS based on the crystal structure of an orthologous enzyme from SPN.

In addition, we modeled the interactions of the coenzyme FMN and the EPSP substrate with the enzyme using a simple, geometric docking approach. After docking studies we performed refinement with energy minimization using two different protocols. The minimized structures showed improvements when compared to the geometrically docked one; amino acids that were overlapping with FMN were

Summary

The enzymes of the shikimate pathway constitute an excellent target for the design of new antibacterial agents. This pathway is found in bacteria, fungi, plants and apicomplexan parasites but is absent in mammals. CS catalyzes the last step of this pathway, the product of which is utilized in other enzymatic transformations like the biosynthesis of aromatic amino acids, folate, vitamin K and ubiquinone. This reaction is the most unusual of the entire pathway and is unique in nature. It converts

Acknowledgments

This project was supported by grants from CAPES and FAPERGS to O.N.S. and FINEP and Millennium Institute (CNPq/MCT) to D.S.S. and L.A.B.; C.L.F. and A.B., we are supported by MSc scholarships from CAPES. We also thank the referees for their valuable suggestions.

References (26)

  • World Health Organization, USA, Global Tuberculosis Control Report 2004,...
  • S.T. Cole et al.

    Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence

    Nature

    (1998)
  • H.M. Berman et al.

    The protein data bank

    Nucleic Acids Res.

    (2000)
  • Cited by (0)

    View full text