Structure of B. cereus Class Ib Ribonucleotide Reductase NrdI-Fe2-NrdF Complex

Class Ib ribonucleotide reductases (RNRs) use a dimetal-tyrosyl radical (Y·) cofactor in their NrdF (beta2) subunit to initiate ribonucleotide reduction in the NrdE (beta2) subunit. Contrary to the diferric tyrosyl radical (FeIII2-Y·) cofactor, which can self-assemble from FeII2-NrdF and O2, generation of the MnIII2-Y· cofactor requires the reduced form of a flavoprotein, NrdIhq, and O2 for its assembly. Here we report the 1.8 Å resolution crystal structure of Bacillus cereus Fe2-NrdF in complex with NrdI. Compared to the previously solved Escherichia coli NrdI-MnII2-NrdF structure, NrdI and NrdF binds similarly in Bacillus cereus through conserved core interactions. This protein-protein association seems to be unaffected by metal ion type bound in the NrdF subunit. The Bacillus cereus MnII2-NrdF and Fe2-NrdF structures, also presented here, show conformational flexibility of residues surrounding the NrdF metal ion site. The movement of one of the metal-coordinating carboxylates is linked to the metal type present at the di-metal site, and not associated with NrdI-NrdF binding. This carboxylate conformation seems to be vital for the water network connecting the NrdF di-metal site and the flavin in NrdI. From these observations, we suggest that metal-dependent variations in carboxylate coordination geometries are important for active Y· cofactor generation in class Ib RNRs. Additionally, we show that binding of NrdI to NrdF would structurally interfere with the suggested alfa2beta2 (NrdE-NrdF) holoenzyme formation, suggesting the potential requirement for NrdI dissociation before NrdE-NrdF assembly after NrdI-activation. The mode of interactions between the proteins involved in the class Ib RNR system is, however, not fully resolved. [1]

Structural insight into the function and anti-TB pro-drug activation by KatG

Catalase-peroxidase (KatG) is a dual functional enzyme with both catalase and peroxidase activity. KatGs protect aerobic microorganisms from oxidative damage through their high catalase activity, and in Mycobacterium tuberculosis (M.tb.) KatG's peroxidatic activity is central in converting the anti-tuberculosis (TB) pro-drug isoniazid (INH) into an active bactericidal molecule in vivo. There are several antibiotics currently in use to treat TB with INH being one of the first anti-TB agents. A central goal is to understand the catalytic function of M.tb. KatG in drug activation and how mutations in KatG found in INH-resistant strains interfere with this process. One of the most common INH-resistant M.tb. strains has the KatG[Ser315Thr] mutation, previously reported to cause narrowing of a substrate access channel. We have now solved the structure of another mutant [Asp137Ser], which shows enhanced INH-activation ascribed to an enlarged access channel. This study demonstrates that altering the dimensions of the bottleneck in the substrate access channel in KatG can impede or enhance INH peroxidation rates relative to the WT enzyme.

Nitric oxide synthase and possible redox partners in Bacillus cereus

Nitric oxide synthase (NOS), a BH4-dependent heme-enzyme, is the only enzyme that specifically produces NO in mammals. NO is produced by the NOS homodimer in two multistep reaction cycles involving electron transfer from a reducing domain to the heme active site. The importance of NO in mammals is due to its function in signalling, vasodilation and immune response. Some bacterial species also contain NOS-encoding genes, but these bacterial NOSs are differently organized - they contain no reducing domain - and their functions and mechanism are not fully resolved [1]. Bacterial NOSs are potential drug targets, because of their role in protection against antibiotics and oxidative stress in some pathogenic bacterial species (e.g. Bacillus anthracis) [2]. Flavodoxins (Flds) have been shown to be relevant redox partners for bacterial NOSs [3], but the specificity of the interaction between NOS and Flds remains poorly understood. We have investigated the NOS protein system in Bacillus cereus, whose genome encodes NOS and two Flds, by combining crystallographic and spectroscopic methods. So far the structures of the two Flds have been solved to 0.98 Å and 2.75 Å resolution, while NOS has been solved to 2.9 Å resolution. An important part of the study has been to investigate the effect of synchrotron X-ray radiation on the oxidation state and structure of the Flds, due to their radiation sensitive cofactor flavin mononucleotide (FMN). The high-resolution (0.98 Å), oxidized structure of one Fld indicates that X-rays induce structural changes around the FMN cofactor. Another important part of the study has been to gain further insight into the specificity and flexibility of the interactions between ferredoxin/flavodoxin-NADP+ reductases, Flds and NOS in Bacillus cereus, as well as the possible mechanism of bacterial NOSs.