Structural studies on enzymes involved in the formation of salicylate and p-aminobenzoate

A number of different aromatic molecules are made from chorismate. We are interested in the enzymes that use this molecule and convert it into different products. What do the enzymes look like (what is their 3D protein crystal structure)? How does one enzyme differ from another? Can we interpret these differences to explain the different reactions they catalyse? We plan to study two systems: (a) the enzymes that convert chorismate to salicylate, and (b) the enzymes that convert chorismate to ADC, an intermediate on the way to p-aminobenzoate. The conversion of chorismate to salicylate involves two steps. The conversion of chorismate to isochorismate, and the conversion of isochorismate to salicylate. We have recently solved the first structure of a 'salicylate synthase', an enzyme (Irp9) that does both of these steps. We now plan to solve the crystal structures of an enzyme that just catalyses the first step (called EntC), and one that just catalyses the second step (PchB). We also plan to make variants of Irp9 to get a better understanding of how it works. The conversion of chorismate to ADC also involves two steps. The first is catalysed by PabA. It converts glutamine to ammonia and glutamate. The ammonia is then thought to pass through a tunnel to get to the active site of PabB, where it reacts with chorismate to make ADC. We plan to solve the crystal structure of the PabA:PabB complex to see if we can get information about this tunnel. We also hope to get a crystal structure of a novel reaction intermediate attached to PabB that we have previously detected by mass spectrometry. We will learn a great deal from these two projects. We will have a better understanding of why some enzymes catalyse one reaction and then stop, while another apparently similar enzyme, catalyses two reactions. We will also learn something about how the reactions proceed by getting the structure of intermediates and analogues bound at the active sites of the enzymes

This proposal is to carry out structural studies on two sets of related enzymes, those involved the conversion of chorismate to salicylate, and those involved in the conversion of chorismate to 4-amino-4-deoxychorismate (ADC), and intermediate en route to p-aminobenzoate. We have BMS funding for mechanistic studies on these enzymes, but the part of our application relating to structural studies was not funded because we lacked preliminary data. We have recently solved the first structure of a salicylate synthase. This gives us an international lead in the area and makes this application timely. The conversion of chorismate to salicylate involves two separate reactions, the conversion of chorismate to isochorismate (catalysed by isochorismate synthase), and the conversion of isochorismate to salicylate (catalysed by isochorismate pyruvate lyase). We propose to solve the first crystal structures of an isochorismate synthase (EntC) and an isochorismate pyruvate lyase (PchB). We have over-expressed both enzymes and have established crystallisation conditions. We have cloned and over-expressed the irp9 gene and shown that Irp9 catalyses the conversion of chorismate to salicylate, via isochorismate. Very recently we solved the structure of this Irp9, and of the enzyme with products (salicylate and pyruvate) bound. We now wish to solve structures of Irp9 with analogues of chorismate and isochorismate bound, and of mutants of Irp9 where we have blocked the second step. This may enable us to get a structure with the intermediate, isochorismate bound. Comparision of the EntC structure with Irp9 should help identify those aspects of the active site of Irp9 involved in the catalysis of the first step, to form isochorismate. Likewise, comparison of the structures of PchB and Irp9 should inform us about the catalysis of the second step. Intriguingly these enzymes do not share any sequence similarity, and it may be that they use different catalytic mechanisms to promote the isochorismate pyruvate lyase reaction. The second part of the application is to study the enzymes that convert chorismate to ADC. PabA catalyses the conversion of glutamine to glutamate and ammonia. It is then thought that the ammonia passes through a tunnel within the PabA:PabB heterodimer to reach the PabB active site, where is reacts with a covalent adduct formed from chorismate, converting it to ADC. We have previously used electrospray mass spectrometry to detect this covalent adduct, and provided circumstantial evidence that it is attached to Lys274. We propose to solve the structure of the PabA:PabB heterodimer (referred to as ADC synthase) in order to get structural evidence for the putative ammonia tunnel, and to attempt to solve the structure of the covalent adduct attached to PabA:PabB (or PabB alone). These two complementary studies will provide insight on a number of important questions: what determines the regiochemistry of nucleophilic attack on chorismate; what factors govern the choice of the nucleophile (NH3 v H20); how does one enzyme catalyse two sequential reactions while an apparently similar enzyme stops after only the first reaction, and a second unrelated enzyme then catalyses the second reaction.