Characterization of genes responsible for the CO-linked hydrogen production pathway in Rubrivivax gelatinosus.

Upon exposure to carbon monoxide, the purple nonsulfur photosynthetic bacterium Rubrivivax gelatinosus produces hydrogen concomitantly with the oxidation of CO according to the equation CO + H(2)O <--> CO(2) + H(2). Yet little is known about the genetic elements encoding this reaction in this organism. In the present study, we use transposon mutagenesis and functional complementation to uncover three clustered genes, cooL, cooX, and cooH, in Rubrivivax gelatinosus putatively encoding part of a membrane-bound, multisubunit NiFe-hydrogenase. We present the complete amino acid sequences for the large catalytic subunit and its electron-relaying small subunit, encoded by cooH and cooL, respectively. Sequence alignment reveals a conserved region in the large subunit coordinating a binuclear [NiFe] center and a conserved region in the small subunit coordinating a [4Fe-4S] cluster. Protein purification experiments show that a protein fraction of 58 kDa molecular mass could function in H(2) evolution mediated by reduced methyl viologen. Western blotting experiments show that the two hydrogenase subunits are detectable and accumulate only when cells are exposed to CO. The cooX gene encodes a putative Fe-S protein mediating electron transfer to the hydrogenase small subunit. We conclude that these three Rubrivivax proteins encompass part of a membrane-bound, multisubunit NiFe-hydrogenase belonging to the energy-converting hydrogenase (Ech) type, which has been found among diverse microbes with a common feature in coupling H(2) production with proton pumping for energy generation.

Activation of brain calcineurin (Cn) by Cu-Zn superoxide dismutase (SOD1) depends on direct SOD1-Cn protein interactions occurring in vitro and in vivo.

Cn (calcineurin) activity is stabilized by SOD1 (Cu-Zn superoxide dismutase), a phenomenon attributed to protection from superoxide (O2*-). The effects of O2*- on Cn are still controversial. We found that O2*-, generated either in vitro or in vivo did not affect Cn activity. Yet native bovine, recombinant human or rat, and two chimaeras of human SOD1-rat SOD1, all activated Cn, but SOD2 (Mn-superoxide dismutase) did not affect Cn activity. There was also a poor correlation between SOD1 dismutase activity and Cn activation. A chimaera of human N-terminal SOD1 and rat C-terminal SOD1 had little detectable dismutase activity, yet stimulated Cn activity the same as full-length human or rat SOD1. Nevertheless, there was evidence that the active site of SOD1 was involved in Cn activation based on the loss of activation following chelation of Cu from the active site of SOD1. Also, SOD1 engaged in the catalysis of O2*- dismutation was ineffective in activating Cn. SOD1 activation of Cn resulted from a 90-fold decrease in phosphatase K(m) without a change in V(max). A possible mechanism for the activation of Cn was identified in our studies as the prevention of Fe and Zn losses from the active site of Cn, suggesting a conformation-dependent SOD1-Cn interaction. In neurons, SOD1 and Cn were co-localized in cytoplasm and membranes, and SOD1 co-immunoprecipitated with Cn from homogenates of brain hippocampus and was present in immunoprecipitates as large multimers. Pre-incubation of pure SOD1 with Cn caused SOD1 multimer formation, an indication of an altered conformational state in SOD1 upon interaction with Cn.

Energy generation from the CO oxidation-hydrogen production pathway in Rubrivivax gelatinosus.

When incubated in the presence of CO gas, Rubrivivax gelatinosus CBS induces a CO oxidation-H2 production pathway according to the stoichiometry CO + H2O --> CO2 + H2. Once induced, this pathway proceeds equally well in both light and darkness. When light is not present, CO can serve as the sole carbon source, supporting cell growth anaerobically with a cell doubling time of nearly 2 days. This observation suggests that the CO oxidation reaction yields energy. Indeed, new ATP synthesis was detected in darkness following CO additions to the gas phase of the culture, in contrast to the case for a control that received an inert gas such as argon. When the CO-to-H2 activity was determined in the presence of the electron transport uncoupler carbonyl-cyanide m-chlorophenylhydrazone (CCCP), the rate of H2 production from CO oxidation was enhanced nearly 40% compared to that of the control. Upon the addition of the ATP synthase inhibitor N,N'-dicyclohexylcarbodiimide (DCCD), we observed an inhibition of H2 production from CO oxidation which could be reversed upon the addition of CCCP. Collectively, these data strongly suggest that the CO-to-H2 reaction yields ATP driven by a transmembrane proton gradient, but the detailed mechanism of this reaction is not yet known. These findings encourage additional research aimed at long-term H2 production from gas streams containing CO.

Conserved amino acid residues within the amino-terminal domain of ClpB are essential for the chaperone activity.

ClpB from Escherichia coli is a member of a protein-disaggregating multi-chaperone system that also includes DnaK, DnaJ, and GrpE. The sequence of ClpB contains two ATP-binding domains that are enclosed between the amino-terminal and carboxyl-terminal regions. The N-terminal sequence region does not contain known functional sequence motifs. Here, we performed site-directed mutagenesis of four polar residues within the N-terminal domain of ClpB (Thr7, Ser84, Asp103 and Glu109). These residues are conserved in several ClpB homologs. We found that the mutations, T7A, S84A, D103A, and E109A did not significantly affect the secondary structure and thermal stability of ClpB, nor did they inhibit the self-association of ClpB, its basal ATPase activity, or the enhanced rate of the ATP hydrolysis by ClpB in the presence of poly-L-lysine. We observed, however, that three mutations, T7A, D103A, and E109A, reduced the casein-induced activation of the ClpB ATPase. The same three mutant ClpB variants also showed low chaperone activity in the luciferase reactivation assay. We found, however, that the four ClpB mutants, as well as the wild-type, bound similar amounts of inactivated luciferase. In summary, we have identified three essential amino acid residues within the N-terminal region of ClpB that participate in the coupling between a protein-binding signal and the ATP hydrolysis, and also support the chaperone activity of ClpB.

Stability and interactions of the amino-terminal domain of ClpB from Escherichia coli.

ClpB is a member of a multichaperone system in Escherichia coli (with DnaK, DnaJ, and GrpE) that reactivates aggregated proteins. The sequence of ClpB contains two ATP-binding regions that are enclosed between the N- and C-terminal extensions. Whereas it has been found that the N-terminal region of ClpB is essential for the chaperone activity, the structure of this region is not known, and its biochemical properties have not been studied. We expressed and purified the N-terminal fragment of ClpB (residues 1-147). Circular dichroism of the isolated N-terminal region showed a high content of alpha-helical structure. Differential scanning calorimetry showed that the N-terminal region of ClpB is thermodynamically stable and contains a single folding domain. The N-terminal domain is monomeric, as determined by gel-filtration chromatography, and the elution profile of the N-terminal domain does not change in the presence of the N-terminally truncated ClpB (ClpBDeltaN). This indicates that the N-terminal domain does not form strong contacts with ClpBDeltaN. Consistently, addition of the separated N-terminal domain does not reverse an inhibition of ATPase activity of ClpBDeltaN in the presence of casein. As shown by ELISA measurements, full-length ClpB and ClpBDeltaN bind protein substrates (casein, inactivated luciferase) with similar affinity. We also found that the isolated N-terminal domain of ClpB interacts with heat-inactivated luciferase. Taken together, our results indicate that the N-terminal fragment of ClpB forms a distinct domain that is not strongly associated with the ClpB core and is not required for ClpB interactions with other proteins, but may be involved in recognition of protein substrates.