Hybrid multienzyme systems composed of polyketide synthase (PKS) and nonribosomal polypeptide synthetase (NRPS) modules direct the biosynthesis of clinically valuable natural products in bacteria. The fidelity of this process depends on specific recognition between successive polypeptides in each assembly line-interactions that are mediated by terminal 'docking domains'. We have identified a new family of N-terminal docking domains, exemplified by TubCdd from the tubulysin system of Angiococcus disciformis An d48. TubCdd is homodimeric, which suggests that NRPS subunits in mixed systems self-associate to interact with partner PKS homodimers. The NMR structure of TubCdd reveals a new fold featuring an exposed beta-hairpin that serves as the binding site for the C-terminal docking domain of the partner polypeptide. The pattern of charged residues on the contact surface of the beta-hairpin is a key determinant of the interaction and seems to constitute a 'docking code' that can be used to alter binding affinity.

Paracoccus pantotrophus cytochrome cd 1 is an enzyme of bacterial respiration, capable of using nitrite in vivo and also hydroxylamine and oxygen in vitro as electron acceptors. We present a comprehensive analysis of the steady state kinetic properties of the enzyme with each electron acceptor and three electron donors, pseudoazurin and cytochrome c 550 , both physiological, and the non-physiological horse heart cytochrome c. At pH 5.8, optimal for nitrite reduction, the enzyme has a turnover number up to 121 s ؊1 per d 1 heme, significantly higher than previously observed for any cytochrome cd 1 . Pre-activation of the enzyme via reduction is necessary to establish full catalytic competence with any of the electron donor proteins. There is no significant kinetic distinction between the alternative physiological electron donors in any respect, providing support for the concept of pseudospecificity, in which proteins with substantially different tertiary structures can transfer electrons to the same acceptor. A low level hydroxylamine disproportionase activity that may be an intrinsic property of cytochromes c is also reported. Important implications for the enzymology of P. pantotrophus cytochrome cd 1 are discussed and proposals are made about the mechanism of reduction of nitrite, based on new observations placed in the context of recent rapid reaction studies.

The 1.4-Å crystal structure of the oxidized state of a Y25S variant of cytochrome cd 1 nitrite reductase from Paracoccus pantotrophus is described. It shows that loss of Tyr 25 , a ligand via its hydroxy group to the iron of the d 1 heme in the oxidized (as prepared) wild-type enzyme, does not result in a switch at the c heme of the unusual bishistidinyl coordination to the histidine/methionine coordination seen in other conformations of the enzyme. The Ser 25 side chain is seen in two positions in the d 1 heme pocket with relative occupancies of ϳ7:3, but in neither case is the hydroxy group bound to the iron atom; instead, a sulfate ion from the crystallization solution is bound between the Ser 25 side chain and the heme iron. Unlike the wild-type enzyme, the Y25S mutant is active as a reductase toward nitrite, oxygen, and hydroxylamine without a reductive activation step. It is concluded that Tyr 25 is not essential for catalysis of reduction of any substrate, but that the requirement for activation by reduction of the wild-type enzyme is related to a requirement to drive the dissociation of this residue from the active site. The Y25S protein retains the d 1 heme less well than the wild-type protein, suggesting that the tyrosine residue has a role in stabilizing the binding of this cofactor.

Domains within the multienzyme polyketide synthases are linked by noncatalytic sequences of variable length and unknown function. Recently, the crystal structure was reported of a portion of the linker between the acyltransferase (AT) and ketoreductase (KR) domains from module 1 of the erythromycin synthase (6‐deoxyerythronolide B synthase), as a pseudodimer with the adjacent ketoreductase (KR). On the basis of this structure, the homologous linker region between the dehydratase (DH) and enoyl reductase (ER) domains in fully reducing modules has been proposed to occupy a position on the periphery of the polyketide synthases complex, as in porcine fatty acid synthase. We report here the expression and characterization of the same region of the 6‐deoxyerythronolide B synthase module 1 AT‐KR linker, without the adjacent KR domain (termed ΔN AT1‐KR1), as well as the corresponding section of the DH‐ER linker. The linkers fold autonomously and are well structured. However, analytical gel filtration and ultracentrifugation analysis independently show that ΔN AT1‐KR1 is homodimeric in solution; site‐directed mutagenesis further demonstrates that linker self‐association is compatible with the formation of a linker‐KR pseudodimer. Our data also strongly indicate that the DH‐ER linker associates with the upstream DH domain. Both of these findings are incompatible with the proposed model for polyketide synthase architecture, suggesting that it is premature to allocate the linker regions to a position in the multienzymes based on the solved structure of animal fatty acid synthase.