NRPS Prediction Blast Server - Intrduction

We have recently published a article entitled Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domain,Gregory L. Challis, Jacques Ravel and Craig A. Townsend. Chemistry and Biology. 2000. 7:211-224.

Background: Non-ribosomal peptide synthetases (NRPSs) are large modular proteins that selectively bind, activate and condense amino acids in an ordered manner. Substrate recognition and activation occurs by reaction with ATP within the adenylation domain (A-domain) of each module. Recently, the crystal structure of the A-domain from the gramicidin synthetase GrsA with L-phenylalanine and adenosine monophosphate bound has been determined.

Results: Critical residues in all known NRPS A-domains have been identified that align with eight binding pocket residues in the GrsA A-domain and which define sets of remarkably conserved recognition templates. Phylogenetic relationships among these sets and the likely specificity determinants for polar and non-polar amino acids are discussed in light of extensive published biochemical data for these enzymes. The binding specificity of greater than 80% of the known NRPS A-domains has been correlated with more than 30 amino acid substrates.

Conclusions: The analysis presented in this paper allows the specificity of adenylation domains of unknown function (e.g., from PCR amplification or genome sequencing) to be predicted. Furthermore, it provides a rational framework for alteration of A-domain specificity by site-directed mutagenesis, which has significant potential for engineering the biosynthesis of novel natural products.

Purpose of this site

This web site has been designed to help researchers predict the amino acid substrate which is recognized by an unknown nonribosomal peptdie synthetase adenylation domain (A domain) in an automated way. The eight critical residues are first extracted from the unknown sequence, then the use of the Blast program with a database made of all eight amino acids lining the active pocket of adenylation domains of assigned function allows the user to predict the putative amino acid recognized by the unknown A domain. Careful analysis of the output will allow the user to define site directed mutagenesis strategie for alteration of A domain specificity.

BLAST SERVER