Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 106 (41), 17331-6

How Oxygen Attacks [FeFe] Hydrogenases From Photosynthetic Organisms

Affiliations

How Oxygen Attacks [FeFe] Hydrogenases From Photosynthetic Organisms

Sven T Stripp et al. Proc Natl Acad Sci U S A.

Abstract

Green algae such as Chlamydomonas reinhardtii synthesize an [FeFe] hydrogenase that is highly active in hydrogen evolution. However, the extreme sensitivity of [FeFe] hydrogenases to oxygen presents a major challenge for exploiting these organisms to achieve sustainable photosynthetic hydrogen production. In this study, the mechanism of oxygen inactivation of the [FeFe] hydrogenase CrHydA1 from C. reinhardtii has been investigated. X-ray absorption spectroscopy shows that reaction with oxygen results in destruction of the [4Fe-4S] domain of the active site H-cluster while leaving the di-iron domain (2Fe(H)) essentially intact. By protein film electrochemistry we were able to determine the order of events leading up to this destruction. Carbon monoxide, a competitive inhibitor of CrHydA1 which binds to an Fe atom of the 2Fe(H) domain and is otherwise not known to attack FeS clusters in proteins, reacts nearly two orders of magnitude faster than oxygen and protects the enzyme against oxygen damage. These results therefore show that destruction of the [4Fe-4S] cluster is initiated by binding and reduction of oxygen at the di-iron domain-a key step that is blocked by carbon monoxide. The relatively slow attack by oxygen compared to carbon monoxide suggests that a very high level of discrimination can be achieved by subtle factors such as electronic effects (specific orbital overlap requirements) and steric constraints at the active site.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
Catalytic profile of CrHydA1 at pH 6.0 (solid lines) and pH 8.0 (dashed lines) as viewed by cyclic voltammograms of the enzyme adsorbed on a PGE electrode. The thermodynamic 2H+/H2 potentials at pH 6.0 and pH 8.0 are marked by the vertical lines. The dashed oval highlights the inflection point at the zero-current potential at pH 6.0. Black and open circles mark the potential at which anaerobic inactivation begins to occur as the potential is swept to more positive values and the current at each cyclic voltammogram has been normalized at this potential. Black and open arrows indicate the directions of the scans at pH 6 and pH 8, respectively. Experimental conditions: 20 °C, 1 bar H2, electrode rotation rate 3,000 rpm, scan rate 20 mV/s.
Fig. 2.
Fig. 2.
Inactivation of CrHydA1 by O2 by simultaneous gas exchange and injection of O2-saturated buffers. Experiments were carried out under (A) different concentrations of O2 and (B) different concentrations of H2. (A) Gas mixtures in the headspace contain 80% H2 and the remaining 20% are as indicated. For the experiments in which inactivation was induced by 10% and 5% O2, injections of 2 mL and 0.67 mL buffer saturated with 20% O2 and 80% H2 (respectively) were performed into the cell containing 2 mL at the beginning of the experiment. For the experiments in which inactivation was induced by 0.5% O2, an injection of 0.5 mL buffer saturated with 2.5% O2 and 80% H2 was performed into the cell containing 2 mL at the beginning of the experiment. (B) Inactivation was achieved with 5% O2 in the headspace and the solid and dashed lines represent experiments performed with 8% and 80% H2, respectively. For the experiments in which inactivation was induced by 5% O2 in 8% H2, an injection of 0.67 mL buffer saturated with 20% O2 and 8% H2 was performed into the cell initially containing 2 mL. Other conditions: pH 6.0, 20 °C, electrode rotation rate 3,000 rpm, −0.05 V vs. SHE.
Fig. 3.
Fig. 3.
Inactivation of CrHydA1 by O2 as compared to inhibition by CO, and protection by CO against O2 inactivation. (A) Inactivation and inhibition by 10% O2 (i, red trace) and 10% CO (ii, blue trace) by gas exchange. The black trace shows an experiment in which the enzyme was subjected to 10% O2 after being fully inhibited by 10% CO (iii). The dotted line gives a visual guide to the progression of background film loss. Experimental conditions: pH 6.0, 20 °C, electrode rotation rate 3,000 rpm, −0.05 V vs. SHE, gas mixtures in the headspace as indicated, with 80% H2 and balance of N2 making up the remainder of the headspace atmosphere. The timeline shown in the lower panel provides a guide for the sequence of gas changes. (B) Dependence of rate of inactivation on concentration of O2 (diamonds, red) and CO (squares, blue). Note the rates of inactivation by CO were calculated by performing experiments such as those shown in Fig. 2, that is, by simultaneous injection of CO-saturated buffer and gas exchange.
Fig. 4.
Fig. 4.
XAS comparison of CrHydA1 in its as-isolated reduced form (Hred) and after O2 incubation (Hoxair). The reduced distance is the true metal-backscatterer distance minus approximately 0.4 Å because of a phase shift. (A) Fe K-edge spectra. Inset: isolated preedge features due to 1s→3d electronic transitions. The shown preedge features were derived by subtraction of a polynomial spline from the main edge rise by using the program Xanda. (B) FTs of EXAFS spectra (see Fig. S2) of Hred (solid line) and Hoxair (open circles). The FTs were calculated for k values of 1.6–16.5 Å−1 (7.4–19.5 Å−1 in the inset). Numbers on the FT peaks denote specific Fe-ligand interactions as discussed in the text. The Fe-Fe distance of 2FeH of 2.52 Å (FT peak III) is discernable under both conditions; the Fe-Fe distances of approximately 2.7 Å (FT peak IV) from the [4Fe-4S] cluster are largely diminished in Hred.
Fig. 5.
Fig. 5.
Model scheme for reaction of CO and O2 with the H-cluster. Carbon monoxide binds reversibly to the oxidized state Hox, giving an inhibited species Hox-CO. Oxygen reacts by binding to the H-cluster at the same site as CO, that is, Fed. The O2 is converted to a reactive oxygen species (ROS), most likely superoxide (formed by one electron reduction). The ROS can either migrate the very short distance to oxidize the [4Fe-4S]2+ cluster (A) or it can remain bound and exert its destructive effect by causing a through-bond electron transfer from the [4Fe-4S] cluster (B).

Similar articles

See all similar articles

Cited by 60 PubMed Central articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback