CCMB
Research @ BIC

Research Activities

Structural Bioinformatics

1.1. Towards understanding the mechanism of Type III Polyketide Synthases:

Superfamily of type III polyketide synthases(PKSs) has been a field of immense interest with an emerging repertoire of these enzymes having distinct biological functions. Type III PKSs catalyze iterative Claisen-like decarboxylative condensation reactions of CoA-linked thioesters to biosynthesize structurally diverse polyketide scaffolds that are pharmacologically important. Bioinformatic studies suggested conservation of amino acid residues among plants, bacteria and fungal type III PKSs. It is interesting to note that despite the overall conservation of the fold and the functional residues, these enzymes exhibit metabolite diversity among the evolutionarily related proteins.

a) Type III PKS from Mycobacterium tuberculosis( M.tb ) - This study deals with the investigation of structural basis of function of the first bacterial type III PKS. Earlier, type III PKSs were thought to be specific only to plants, such as Chalcone synthase(chs) from Medicago sativa for flavonoid precursor biosynthesis. Due to emerging genome sequencing projects, type III PKSs are becoming evident in several organisms.

Mycobacterial lipids are biosynthesized by the combined efforts of PKSs and fatty acid synthases(FASs) constituting a protective complex virulent lipid envelope. M.tb genome hosts three genes which encode for type III PKS, namely, pks10, pks11 and pks18. PKS18 biosynthesizes long chain secondary metabolites such as tri and tetra-ketide α-pyrones.

Bioinformatic analysis of various type III PKSs suggested an overall sequence identity of 20-25% and a sequence similarity of 40-45% with a conserved αβαβα-fold architecture and active site residues in PKS18. We performed structure-based sequence analyses which allowed us to identify the key catalytic amino acid residues that would account for the observed differences in the catalytic machinery among the superfamily of type III PKSs. These studies helped us to attempt the crystal structure of PKS18 which led us to identify a 20 Å long ‘novel tunnel’ created in the thiolase fold for substrate-binding and a 16 Å CoA-binding pocket providing structural basis for PKS18 function.

Researcher(s) / References :

  1. Sankaranarayanan, R. (2006) Nat. Chem. Biol. 2, 451-452.
  2. Gokhale R.S., Sankaranarayanan, R., Mohanty, D., (2007) Nat. Prod. Rep 24, 267 – 277.
  3. Rukmini, R., Shanmugam, V.M., Saxena, P., Gokhale R.S., Sankaranarayanan, R., (2004) Acta Crystallogr D Biol Crystallogr 60, 749-751.
  4. Sankaranarayanan, R., Saxena, P., Marathe, U.B., Gokhale R.S., Shanmugam, V.M., Rukmini, R.,(2004) Nat. Struct. Biol. 11, 894-900.

b) Type III PKS from Neurospora crassa - Recent discovery of microbial type III PKS systems has revealed remarkable mechanistic as well as functional versatility. With the advent of Fungal Genome Sequencing Projects, several genes homologous to type III PKSs have been identified. Yet another interesting type III PKS protein (NCU04801.1, gi: 85097336) from Neurospora crassa (referred to as NCIII) revealed a sequence identity of 22-30% with a fold-conservation upon comparative sequence analysis with other functionally characterized plant and bacterial proteins. Initial dendrogram-based analyses showed that the fungal proteins form a distinct clade from plant and bacterial type III PKSs allowing us to speculate that these proteins may be involved in the biosynthesis of structurally related metabolites involving a type III PKS mechanism. Results from biochemical and functional studies indicate that NCIII protein catalyzes biosynthesis of resorcinolic lipid metabolites by utilizing long chain fatty acyl-CoAs.

Bioinformatic analysis of NCIII based on our structure revealed a similar location and orientation of the catalytic triad Cys-His-Asn along with the characteristic long CoA-binding tunnel as observed in other type III PKSs. The analysis aided us to perform structure-based mutagenic studies which revealed interesting molecular features of a long hydrophobic tunnel. We could pinpoint a single residue (S186F) for point mutation lining the tunnel which eventually abrogated formation of resorcinolic molecules and resulted in synthesis of triketide pyrones by retaining the ability to accept long chain starters. With careful sequence analysis along with structure-based information we could perform an additional mutation of Ser340 to Leu which blocked the substrate-binding tunnel. These initial bioinformatic studies provided clues towards the possible existence of a functional link between cyclization and acyl chain binding pockets.

The overall conservation of structural features and the presence of identical catalytic triad in all type III PKSs from various organisms suggest that they share a similar enzymatic mechanism and a common evolutionary origin.

Researcher(s) / References :

  1. Goyal, A., Saxena, P., Rahman, A., Singh, P.K., Kasbekar, D.P., Gokhale, R.S., Sankaranarayanan, R., (2008) J Struct Biol. 162, 411-421.

1.2. Bioinformatics of proteins involved in lipid metabolism in Mycobacterium tuberculosis:

Mechanism of fatty acid activation and its functional implications in Mycobacterium tuberculosis - This project deals with the understanding of enzyme mechanistics involved in the fatty acid activation in M.tb. Degradation of fatty acids is universally initiated by fatty acyl-CoA ligases (FACLs) that convert fatty acids to their corresponding coenzyme A (CoA)-thioesters. In contrast, the close homologues of these proteins, fatty acyl-AMP ligases (FAALs), catalyze formation of fatty acyl-adenylates, which are extended by polyketide synthases to produce structurally distinct and complex virulent lipids in Mycobacterium tuberculosis (M.tb). Several genomes have revealed a large number of such carboxyl-activating enzymes. However, it is not clear how these enzymes could have evolved to support biosynthetic processes.

FadD28 is a FAAL-specific protein involved in the biosynthesis of a virulent lipid pthiocerol dimycocerosate. We performed a combined structural bioinformatic and biochemical studies which provided molecular insights into the fatty acid activation mechanism. Sequence alignment analysis showed a sequence identity of 20-25% with a fold similarity of FAAL28 to the members of acyl adenylating enzymes.

Structure-based sequence alignment of various FAAL, FACL and other structurally characterized homologous protein sequences revealed a 22-amino-acid insertion in FAAL28 sequence, spanning residues 351-372 . Conservation of this insertion FAAL homologs and its absence in FACL proteins gave us a clue to generate loss-of-function and gain-of-function mutants. A FAAL28 mutant with deletion in the helical region (354-365) of the insertion motif could interconvert FAAL and FACL activities demonstrating that the insertion motif dictates formation of acyl-adenylate, thereby revealing a mechanism by which M.tb may have evolved FAAL proteins from the omnipresent FACLs. Combined together our analyses highlight a novel mechanism of evolution of FAAL family proteins by subtle alteration in the ubiquitous FACLs in mycobacterium.

Researcher(s) / References :

  1. Gokhale, R.S., Sankaranarayanan, R. & Mohanty, D. ,(2007) Curr. Opin. Struct. Biol. 17, 736–743.
  2. Arora, P., Goyal, A., Natarajan, V.T., Rajakumara, E., Verma, P., Gupta, R., Yousuf, M., Trivedi, O.A., Mohanty, D., Tyagi, A., Sankaranarayanan, R., Gokhale, R. S., et al., (2009) Nat. Chem. Biol. 5, 166-173.

1.3. Bioinformatic Studies on a Unique Cell Wall-Degrading Esterase:

A Cell Wall–Degrading Esterase of Xanthomonas oryzae Requires a Unique Substrate Recognition Module for Pathogenesis on Rice - Xanthomonas oryzae pv oryzae (Xoo) causes bacterial blight of rice (Oryza sativa). LipA, a secretory virulence factor of Xoo and a general indicator of lipase/esterase function, is implicated in rice cell wall degradation with elicitation of innate immune responses like callose deposition and programmed cell death. LipA homologs are present in several gram-negative bacteria, including xanthomonads whose genomes have been sequenced.

Bioinformatic analysis of LipA showed a low sequence identity of 20% with characterized hydrolase fold proteins. From our structural characterization of LipA, we could identify an all-helical 30Å long module forming the ligand-binding tunnel containing a carbohydrate-anchoring pocket. Multiple sequence analysis showed a high conservation of this module across genus Xanthomonas emphasizing its plant cell wall-degrading function for this pathogenic bacteria. Based on phylogenetic dendrogram analysis, LipA sequence homologs cluster together. Interestingly, homology modeling of LipA-like proteins form one group indicating the presence of ligand-binding tunnel with a carbohydrate-binding module. Another group has a similar tunnel but with no carbohydrate-binding pocket.

A comparison with the related structural families illustrates how a typical lipase is recruited to promote virulence. Thus, the study provides a remarkable example of the emergence of novel functions for proficient pathogenesis during pathogen-plant coevolution.

Researcher(s) / References :

  1. Gudlur, A., Chatterjee, A., et al., (2009) Plant Cell 21, 1860–1873

1.4. Bioinformatic analysis on archael tRNA synthetases:

Fidelity during translation of the genetic code - The project aims to elucidate the proofreading mechanisms during translation at atomic resolution and to provide a structural and functional proof for proofreading/editing defects leading to disease conditions like neurodegeneration. Our efforts are currently focussed on understanding the editing module of Pyrococcus abyssi threonyl-tRNA synthetases (ThrRSs) as most of the archaeal ThrRSs possess a unique editing domain when compared to its counterparts in eubacteria and eukaryotes.

The bioinformatic analysis revealed that the unique editing domain from P. abyssi ThrRSs (Pab-NTD) shares no sequence homology with any other protein of known sequence. Upon determination of crystal structure of Pab-NTD the coordinates were used to search for structural homolog using DALI server. Interestingly, it was found out that this protein shares structural homology with another family of editing modules called as D-aminoacyl-tRNA deacylases (DTDs) (Dwivedi et. al. 2005). Structure based sequence alignment showed conservation of residues in proposed the active site in both family of proteins. Mutational as well as biochemical studies led us to propose the editing mechanism in Pab-NTD. This led us to propose the proof-reading mechanism for DTDs based on our knowledge of editing mechanism in Pab-NTD (Hussain et. al. 2006). Structure-based sequence alignment also showed family-specific conservation of residues in Pab-NTD homologs that are replaced by another residue in DTDs. These residues have been taken up for further studies to elucidate their role in each family.

Bioinformatic analysis has also been used to figure out variants of these editing modules in different genomes. One of such module has also been taken up for further study. Studying such variants will enhance our understanding of these modules and may also shed light on the evolution of editing modules.

Moreover, different editing modules have also been taken up for study based on preliminary bioinformatic analysis.

Researcher(s) / References :

  1. Dwivedi S, Kruparani SP, Sankaranarayanan R (2005) Nature Struct. Mol. Biol. 12: 556-557.
  2. Hussain T, Kruparani SP, Pal B, Dock-Bregeon AC, Dwivedi S, Shekar MR, Sureshbabu K, Sankaranarayanan R (2006) EMBO J 25: 4152-4162.
  3. Hussain T, Kamarthapu V, Kruparani SP, Deshmukh MV, Sankaranarayanan R. (2010) Proc Natl Acad Sci U S A. 107, 22117-22121.