Chemical biology of Mycobacterium tuberculosis: Deciphering the role of polyketide synthases in Mycobacteria

 
Principal Investigator :  Rajesh S Gokhale

Project Associates / Assistants
Priti Saxena
Rashmi Tickoo
Shobhit Saxena
Archana Vats

Ph D Students
Omita Mehta
Yogyata
Pooja Arora
Jatinder Singh

Collaborators
Debasisa Mohanty

Genome-based approaches to identify and exploit the microbial metabolic pathways that are involved in the biosynthesis of various natural products is the major interest of this project. The present focus is to understand the importance of various polyketide synthase gene clusters from Mycobacterium tuberculosis. The genome sequence of M. tuberculosis has revealed a remarkable array of genes that are homologous to polyketide synthases. Polyketide synthases (PKSs) are a class of enzymes that are involved in the biosynthesis of secondary metabolites such as erythromycin, rapamycin, tetracycline, lovastatin, and resveratrol. Attempts are beingh made to understand and exploit the role of polyketide synthases in the biology of M. tuberculosis. The objectives of the project are (i) identification and biochemical analyses of enzymes that are involved in the biosynthesis of metabolites, (ii) isolation and characterization of PKS gene products by heterologous expression of these genes in Streptomyces coelicolor and Escherichia coli, (iii) characterization of molecular mechanisms mediating the crosstalk between various polyketide synthases (PKSs) and fatty acid synthases (FASs) in M. tuberculosis and (iv) genetic and/or Chemical knock-out of PKS genes to synthesize novel polyketides and to study the effects of these changes on mycobacterial pathogenecity.

Sequence comparison and analyses of Mycobacterium polyketide synthase genes

In collaboration with Dr Mohanty’s group at National Institute of Immunology, we have developed an automated web-based software to identify various domains in PKS genes. Apart from elucidating domain arrangement and organization in modular PKS proteins, this software can also predict the specificity of the acyl transferase domain based on homology modeling. We have also analyzed several other PKS-related proteins in M. tuberculosis including the FadD proteins. Our studies indicate that there are at least two different types of FadD proteins that are scattered in the genome. We have investigated several FadD and PKS proteins in order to understand their biological role in M. tuberculosis.

Biochemical analyses of enzymatic crosstalk between FASs and PKSs

In nature, the formation of fatty acids and polyketides is catalyzed by structurally and mechanistically-related enzymes. Both FASs and PKSs use simple acyl coenzyme A (CoA) precursors in a sequential fashion to generate the carbon backbones of these molecules. Traditionally, the biosynthetic machinery for fatty acids and polyketides has been studied independently, but recent evidence suggests that these enzyme systems may coordinate in many organisms to generate hybrid molecules of fatty acids and polyketides. We have examined number of FadD proteins in order to determine their importance to mycobacteria and understand their utility in different metabolic pathways. Our studies show that there are at least two classes of FadD-like proteins, which uses two different chemical modes to activate fatty acids. FadD proteins present adjacent to PKS/NRPS cluster, such as FadD26, FadD28, FadD30, FadD32 and FadD29 synthesizes acyl-adenylates, whereas other FadD proteins makes acyl-CoA. All the products were characterized using mass spectrometry. To test whether FadD proteins can transfer activated substrates onto PKS proteins, we have functionally expressed and characterized number of PKS proteins that are present in proximity of FadD genes. FadD26, FadD28, FadD30, FadD32 proteins and their cognate PKS proteins PpsA, Mas, PKS6 and PKS13 respectively were expressed and purified from the heterologous host. Initial cell-free reconstitution studies of several FadD and PKS proteins have provided direct evidence of mechanism of priming long chain fatty acids onto PKS proteins.

Cloning and characterization of type III PKSs

The analysis of mycobacterial genome sequences has revealed three type III PKS genes in M. tuberculosis. Two of the type III genes, PKS10 and PKS 11 are clustered in an unusual organization with several other type I PKS genes that encompass 18 kb of genomic DNA. PKS 18 protein is not flanked by any other PKS-related genes. In order to probe their functional relevance in mycobacteria, we screened other mycobacterial genomes to check if these are conserved. Genome analysis suggested that all the three type III PKS genes were present as pseudogenes in M. leprae, whereas they were completely conserved in M. bovis BCG. M. bovis was therefore used as a model to test functional expression of these genes. RT-PCR studies with primers specific for PKS 10 and PKS11 yielded single PCR products of 1.05 kb corresponding to the expected size of pks10 and pks11 transcripts. These genes were cloned and expressed in pET expression system such that proteins were expressed with C-terminal hexahistidine tag. The expressed protein was found as a major band at 39 kDa, although the recombinant protein was almost exclusively localized in the insoluble fraction. These proteins were purified from inclusion bodies and were used to prepare polyclonal antiserum in rabbits. The antiserum generated from PKS11 protein specifically cross-reacted with 39 kDa protein and no cross-reacting bands were observed, whereas antiserum from PKS10 cross-reacted with number of non-specific proteins. PKS11 antiserum was also able to specifically recognize PKS10 protein, which may be a result of high sequence homology between these two proteins. Crude extract of M. bovis BCG cells showed major immunologically cross-reacting bands around 40 kDa. Careful analysis suggested that PKS11 antiserum was also cross-reacting with PKS18 (~42 kDa). PKS18 protein was cloned and expressed in E. coli. The protein was purified to homogeneity using Ni-NTA affinity column and ion exchange chromatography. Cell-free assays showed that this protein is able to synthesize a‑pyrones by using a starter acyl-CoA and 2 molecules of malonyl-CoA. PKS18 protein showed a remarkable broad specificity for starter unit from Hexanoyl (C6) CoA to Stearoyl (C20) CoA. Kinetic studies clearly showed that long chain acyl CoAs were the favoured substrate for this protein. We are presently investigating the role of this metabolite in mycobacteria.

Cloning and expression of several other unusual PKS genes

The 18 kbp cluster consisting of six different proteins was partially characterized by establishing an in vitro assay for PKS11 and PKS 9 proteins. PKS9 protein was able to specifically decarboxylate malonyl CoA to acetyl CoA. PKS11 protein was shown to use three acyl CoA units to synthesize a-pyrones. In order to understand the in vivo relevance of this cluster in the life cycle of mycobacteria, this cluster was knocked-out from M. bovis by using homologous recombination. The hygromycin and SacB gene was used as markers for selecting first and second stage of recombinants respectively. The heterologous expression of all six proteins was confirmed in Streptomyces, although we have been unable to isolate any new product from this recombinant strain.

We have cloned and expressed several other PKS genes. PKS13 and PKS6 proteins were expressed as hexa-histidine tagged proteins. These proteins were post-translationally modified using a heterologous coexpression P-pant transferase. Cell-free assays by using SNAC derivatives of hexanoic acid (synthesized in our laboratory) and malonyl CoA for PKS 6 and methyl malonyl CoA for PKS13 suggested synthesis of unknown metabolites by using radio-TLC. We are presently elucidating the structures of these metabolites.

Another unusual FAS protein, mycocerosic acid synthase (mas), has been functionally reconstituted in vitro. MAS is a 464 kDa dimeric protein and contains six catalytic active sites. Our studies indicate that MAS protein specifically uses methyl malonyl CoA to synthesize long chain methylated fatty acids. Biochemical analysis revealed that this protein was unable to release products from the protein and therefore did not show any turnover. A type II thioesterase gene that is present in its neighbour in the genome was shown to accelerate catalysis of mycocerosic acids. TEII was also cloned and expressed in E. coli. Several other proteins present in its vicinity in the genome were also used to test if these proteins would assist in enzymatic turnover from MAS protein.

Publications

Original peer-reviewed articles

1.     Yadav G, Gokhale RS and Mohanty D (2003) Computational approach for prediction of domain organization and substrate specificity of modular polyketide synthases. JMol Biol 328:335‑363.

2.     Yadav G, Gokhale RS and Mohanty D (2003) SEARCHPKS: A program for detection and analysis of polyketide synthase domains. Nucl Acids Res (in press).