Valine-to-Cysteine Mutation Further Increases the Oxygen Tolerance of Escherichia coli NiFe Hydrogenase Hyd-1
Résumé
Some NiFe hydrogenases are particularly resistant to O 2 , as a result of either the natural presence of a particular FeS cluster or the artificial replacement of a conserved valine residue near the Ni site. We show that the two protective eects can be combined in a single enzyme, by constructing and characterizing the V78C variant of the naturally O 2-tolerant E. coli NiFe hydrogenase Hyd-1. We elucidate the eect of the mutation by comparing the kinetics of inhibition by CO and O 2 of a number of wild-type forms and valine-to-cysteine variants of NiFe hydrogenase. This is the authors' version of doi:10.1021/acscatal.9b00543 Hydrogenases are the enzymes that oxidize and produce H 2. They are all inhibited by O 2 , although the extent, ki-netics and nature of the inactivation vary greatly. Here we discuss the hydrogenases that bear a NiFe dinuclear active site. The so-called "O 2-tolerant" NiFe hydrogenases embed an unusual [4Fe-3S] cluster near the active site, as part of a redox chain that mediates intramolecular electron-transfer. The general consensus is that this cluster causes resistance to O 2 1,2 (see, however, ref. 3). On the other hand, the so-called "standard", O 2-sensitive NiFe hydrogenase from Desulfovibrio fructosovorans (Hyn) can be made significantly resistant to O 2 by replacing a conserved valine residue near the active site 4 (figure 1) with a cystein. 5 (Having any hydrophilic residue at that position in Df Hyn enhances the rate of reactivation after oxidative inactivation 6 and this position also determines the intramolecular diffusion rates. 4) Escherichia coli expresses one hydrogenase from each group, Hyd-1 and Hyd-2, respectively. 1,7,8 Figure 1. Overlay of the structures of the active site of E. coli Hyd-1 NiFe hydrogenase (yellow), E. coli Hyd-2 (pink) and D. fructosovo-rans Hyn (blue), together with the conserved V74/V78 residue, and the proximal [4Fe3S] cluster of Hyd-1. pdb accession codes 3UQY, 9 6EHQ 10 and 1YQW. 11 Lukey and coworkers have reported the production of re-combinant Hyd-1 and Hyd-2 enzymes of E. coli, with the small subunit bearing a C-terminal His 6-tag, from engineered oper-ons on the chromosome. 7 On our side, we produced Hyd-1 (wild-type and V78C) and Hyd-2 enzymes from plasmids in the E. coli FTD147 (DE3) strain carrying chromosomal in-frame deletions of the genes encoding the large subunits of hydrogenase-1,-2, and-3. 12-14 Recombinant Hyd-1 hydroge-nases were produced as a dimeric and soluble form, consisting of the large (HyaB) and the small subunits (HyaA) only (figure S1A). The C-terminal membrane-anchoring hydrophobic helix of HyaA was replaced with a Streptag II. Recombinant Hyd-2 hydrogenase consists of HybO (the small subunit), whose C-terminal transmembrane domain was cleaved by a treatment with trypsin, and HybC (the large subunit) bearing a N-terminal His 6-tag (figure S1B). The enzymes Hyd1 and Hyd2 characterized here appear to have greater and lower activity, respectively, that those prepared using the methods in 7. See in SI more information about the production and activity of these enzymes, and a comparison with previously published activity values. We compared the properties of wild-type recombinant (WT) Hyd-1 and the V78C variant using direct electrochemistry, 15,16 whereby the enzyme is adsorbed onto a rotating electrode and any change in current reports on a change in turnover frequency , resulting e.g. from exposure to inhibitors (CO or O 2) and/or redox-driven (in)activation. Figure 2 compares the response of WT Hyd-1 and V78C to transient exposures to O 2. We produced "bursts" of O 2 by injecting in the electrochemical cell small amounts of O 2-saturated solution while simultaneously flushing the solution with H 2. In both cases the inhibition is reversed by removing O 2 : the enzymes reactivate after exposure to O 2 , even under the very oxidizing conditions used here to prevent the direct reduction of O 2 on the electrode (+140 mV). All things being equal, the V78C variant is less inhibited than the WT enzyme. We determined the rate constants of inactivation (k O 2 i) and reactivation (k O 2 a) in this experiment by fitting 18 a model 4 which assumes that the O 2-adduct can also irreversibly inacti-vate (with 1st-order rate constant k 3): active k O 2 i ⇥[O 2 ] *) k a adduct k 3 ! dead end (1) The rate constants in Table 1 show that V78C inactivates about 5-times more slowly than the WT; the mutation has a smaller effect on the reactivation rate. Whereas the inactiva-tion of the WT enzyme is fully reversible (k 3 = 0), we found that the V78C data are best described using a small value of k 3 ⇡ 0.001 s 1. To determine whether the V78C-induced decrease in the rate of reaction with O 2 is due to the mutation hindering the diffusion along the gas channel, we compared the rates of inhibition of the enzymes by CO, which reports on the rate of intramolecular diffusion. 4,16,19 Figure 3 shows the response of 1
Domaines
Chimie
Origine : Fichiers produits par l'(les) auteur(s)
Loading...