Jasvinder K. Shergill

Oxidative metabolism of inorganic sulfur compounds by bacteria

The history of the elucidation of the microbiology and biochemistry of the oxidation of inorganic sulfur compounds in chemolithotrophic bacteria is briefly reviewed, and the contribution of Martinus Beijerinck to the study of sulfur-oxidizing bacteria highlighted. Recent developments in the biochemistry, enzymology and molecular biology of sulfur oxidation in obligately and facultatively lithotrophic bacteria are summarized, and the existence of at least two major pathways of thiosulfate (sulfur and sulfide) oxidation confirmed. These are identified as the ‘Paracoccus sulfur oxidation’ (or PSO) pathway and the ‘S4intermediate’ (or S4I) pathway respectively. The former occurs in organisms such as Paracoccus (Thiobacillus) versutus and P. denitrificans, and possibly in Thiobacillus novellus and Xanthobacter spp. The latter pathway is characteristic of the obligate chemolithotrophs (e.g. Thiobacillus tepidarius, T. neapolitanus, T. ferrooxidans, T. thiooxidans) and facultative species such as T. acidophilus and T. aquaesulis, all of which can produce or oxidize tetrathionate when grown on thiosulfate. The central problem, as yet incompletely resolved in all cases, is the enzymology of the conversion of sulfane-sulfur (as in the outer [S-] atom of thiosulfate [-S-SO3-]), or sulfur itself, to sulfate, and whether sulfite is involved as a free intermediate in this process in all, or only some, cases. The study of inorganic sulfur compound oxidation for energetic purposes in bacteria (i.e. chemolithotrophy and sulfur photolithotrophy) poses challenges for comparative biochemistry. It also provides evidence of convergent evolution among diverse bacterial groups to achieve the end of energy-yielding sulfur compound oxidation (to drive autotrophic growth on carbon dioxide) but using a variety of enzymological systems, which share some common features. Some new data are presented on the oxidation of 35S-thiosulfate, and on the effect of other anions (selenate, molybdate, tu ngstate, chromate, vanadate) on sulfur compound oxidation, including observations which relate to the roles of polythionates and elemental sulfur as intermediates.

COORDINATION OF THE 2FE-2S CLUSTER IN WILD TYPE AND MOLECULAR VARIANTS OF CLOSTRIDIUM PASTEURIANUM FERREDOXIN, INVESTIGATED BY ESEEM SPECTROSCOPY

The [2Fe-2S] ferredoxin from Clostridium pasteurianum contains five cysteine residues in positions 11, 14, 24, 56, and 60. This pattern is unique, and a combination of site-directed mutagenesis and spectroscopy is therefore being implemented to identify the ligands of the [2Fe-2S] cluster. The possible involvement of ligands other than cysteine in some molecular variants of this ferredoxin has been considered, histidines being likely candidates. Therefore, the three histidine residues in positions 6, 7, and 90 of the amino acid sequence have been individually and collectively replaced by alanine or valine. The mutated ferredoxins have been purified and were all found to contain [2Fe-2S] clusters of which the UV-visible absorption spectra were identical to that of the wild-type protein. The H6A/H7A/ H90A triply mutated ferredoxin was further characterized by EPR and by ESEEM spectroscopy and was found to differ only marginally from the wild-type protein. The ESEEM spectra of wild-type ferredoxin displayed weak 14N hyperfine interactions at the three principal g-factors of the [2Fe-2S] center. The estimated 14N coupling constants (Aiso = 0.6 MHz; e2qQ approximately 3.3 MHz) indicate that the ESEEM effect is most likely due to 14N from the polypeptide backbone. 2H2O ESEEM spectra showed that the [2Fe-2S] cluster is accessible for exchange with solvent deuterons. ESEEM spectra of the previously characterized C24A and C14A/C24A variants have been recorded and were also found to be very similar to those of the wild-type protein. There was no evidence for coordination of the [2Fe-2S] cluster by [14N]histidine or other 14N nuclei, in either wild-type or mutant forms of the ferredoxin. By these criteria, the environment of the [2Fe-2S] center is not distinguishable from those in plant-type ferredoxins. Non-cysteinyl coordination most probably occurs only in the C14A/C24A variant, which contains no more than three cysteine residues. The data shown here indicate that the fourth ligand of the [2Fe-2S] cluster is neither a histidine residue nor another nitrogenous ligand. The possibility of oxygenic coordination for this molecular variant is discussed.

Electron spin-echo envelope modulation and spin–spin interaction studies of the iron–sulphur clusters in fumarate reductase of Escherichia coli

Electron spin-echo envelope modulation (ESEEM) spectroscopy has been performed in order to obtain structural information about the reduced [2Fe-2S], [4Fe-4S] and oxidized [3Fe-4S] clusters in purified fumarate reductase from Escherichia coli. Spin echoes were detected from the reduced iron–sulphur clusters (centres 1–3) and also from oxidized centre 3. The ESEEM studies showed that the reduced [2Fe-2S] cluster is coordinated by a peptide 14N. In this study, we also observed a weak interaction between the oxidized [3Fe-4S] cluster and a peptide 14N; there was no evidence for coordination to the 14N-atoms of imidazole rings. Spin–lattice and spin–spin relaxation rates for oxidized and reduced protein were measured, and the results confirmed the presence of a spin–spin interaction between reduced centres 1 and 2.

Electron spin-echo envelope modulation studies of the [3Fe–4S] cluster of Escherichia coli fumarate reductase

The dimeric form of purified Escherichia coli fumarate reductase contains three different iron–sulfur clusters, termed centres 1, 2 and 3 ([2Fe–2S], [4Fe–4S] and [3Fe–4S] clusters, respectively). We have performed electron spin-echo envelope modulation (ESEEM) spectroscopy in order to obtain information about the environment of the [3Fe–4S] cluster. Modulations from 14N were detected in the three-pulse ESEEM spectra from the cluster in the oxidised state. The low-frequency lines observed in the three-pulse ESEEM pattern were similar to those observed for the [3Fe–4S] cluster of bovine heart succinate dehydrogenase (B. A. C. Ackrell, E. B. Kearney, W. B. Mims, J. Peisach and J. H. Beinert, J. Biol. Chem., 1984, 259, 4015). Determination of the hyperfine and quadrupolar coupling (A≈ 0.7 MHz, e2qQ≈ 3.4 MHz) gave values that are similar to those determined for the [2Fe–2S] cluster of centre 1 (R. Cammack, A. Chapman, J. McCracken, J. B. Cornelius, J. Peisach and J. H. Weiner, Biochim. Biophys. Acta, 1988, 956, 307; J. K. Shergill, PhD thesis, 1993). These coupling constants indicate that the 14N-nucleus giving rise to the modulation pattern in the ESEEM spectrum is weakly coupled to centre 3, rather than directly coordinated to the iron.

EPR and mutagenesis studies of the Rieske-type iron-sulphur clusters of benzene dioxygenase

enzyme is a three-component system, comprising a flavoprotein and a ferredoxin, which transfer electrons from NADH, to a terminal dioxygenase containing a Rieske-type [2Fe-2S] cluster and a catalytic iron centre [ 11. The terminal dioxygenase consists of two dissimilar subunits arranged in an a& configuration [2]. We have cloned all of the components of the system in E.