Comparative analysis of folding and substrate binding sites between regulated hexameric type II citrate synthases and unregulated dimeric type I enzymes

We describe the first structure determination of a type II citrate synthase , an enzyme uniquely found in Gram-negative bacteria. Such enzymes are hexa meric and are strongly and specifically inhibited by NADH through an allost eric mechanism. This is in contrast to the widespread dimeric type I citrat e synthases found in other organisms, which do not show allosteric properti es. Our structure of the hexameric type II citrate synthase from Escherichi a coli is composed of three identical dimer units arranged about a central 3-fold axis. The interactions that lead to hexamer formation are concentrat ed in a relatively small region composed of helix F, FG and IJ helical turn s, and a seven-residue loop between helices J and K. This latter loop is pr esent only in type II citrate synthase sequences. Running through the middl e of the hexamer complex, and along the 3-fold axis relating dimer units, i s a remarkable pore lined with IS cationic residues and an associated hydro gen-bonded network. Also unexpected was the observation of a novel N-termin al domain, formed by the collective interactions of the first 52 residues f rom the two subunits of each dimer. The domain formed is rich in beta -shee t structure and has no counterpart in previous structural studies of type I citrate synthases. This domain is located well away from the dimer-dimer c ontacts that form the hexamer, and it is not involved in hexamer formation. Another surprising observation from the structure of type II E. coli citra te synthase is the unusual polypeptide chain folding found at the putative acetylcoenzyme A binding site. Key parts of this region, including His264 a nd a portion of polypeptide chain known from type I structures to form an a denine binding loop (residues 299-303), are shifted by as much as 10 Angstr om from where they must be for substrate binding and catalysis to occur. Fu rthermore, the adjacent polypeptide chain composed of residues 267-297 is e xtremely mobile in our structure. Thus, acetylcoenzyme A binding to type II E. coli citrate synthase would require substantial structural shifts and a concerted refolding of the polypeptide chain to form an appropriate bindin g subsite. We propose that this essential rearrangement of the acetylcoenzy me A binding part of the active site is also a major feature of allostery i n type II citrate synthases. Overall, this study suggests that the evolutio nary development of hexameric association, the elaboration of a novel N-ter minal domain, introduction of a NADH binding site, and the need to refold a key substrate binding site are all elements that have been developed to al low for the allosteric control of catalysis in the type II citrate synthase s.