Article pubs.acs.org/biochemistry
Importance of the Active Site “Canopy” Residues in an O2‑Tolerant [NiFe]-Hydrogenase Emily J. Brooke,† Rhiannon M. Evans,† Shams T. A. Islam,† Gerri M. Roberts,‡ Sara A. M. Wehlin,† Stephen B. Carr,§,∥ Simon E. V. Phillips,§,∥ and Fraser A. Armstrong*,† †
Department of Chemistry, University of Oxford, Oxford, U.K. Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States § Research Complex at Harwell, Rutherford Appleton Laboratory, Harwell Campus, Didcot, U.K. ∥ Department of Biochemistry, University of Oxford, Oxford, U.K. ‡
S Supporting Information *
ABSTRACT: The active site of Hyd-1, an oxygen-tolerant membrane-bound [NiFe]-hydrogenase from Escherichia coli, contains four highly conserved residues that form a “canopy” above the bimetallic center, closest to the site at which exogenous agents CO and O2 interact, substrate H2 binds, and a hydrido intermediate is stabilized. Genetic modification of the Hyd-1 canopy has allowed the first systematic and detailed kinetic and structural investigation of the influence of the immediate outer coordination shell on H2 activation. The central canopy residue, arginine 509, suspends a guanidine/guanidinium side chain at close range above the open coordination site lying between the Ni and Fe atoms (N−metal distance of 4.4 Å): its replacement with lysine lowers the H2 oxidation rate by nearly 2 orders of magnitude and markedly decreases the H2/D2 kinetic isotope effect. Importantly, this collapse in rate constant can now be ascribed to a very unfavorable activation entropy (easily overriding the more favorable activation enthalpy of the R509K variant). The second most important canopy residue for H2 oxidation is aspartate 118, which forms a salt bridge to the arginine 509 headgroup: its mutation to alanine greatly decreases the H2 oxidation efficiency, observed as a 10-fold increase in the potential-dependent Michaelis constant. Mutations of aspartate 574 (also salt-bridged to R509) to asparagine and proline 508 to alanine have much smaller effects on kinetic properties. None of the mutations significantly increase sensitivity to CO, but neutralizing the expected negative charges from D118 and D574 decreases O2 tolerance by stabilizing the oxidized resting NiIII−OH state (“Ni-B”). An extensive model of the catalytic importance of residues close to the active site now emerges, whereby a conserved gas channel culminates in the arginine headgroup suspended above the Ni and Fe.
H
proton-coupled electron-transfer process, so the chemistry is of fundamental importance. From extensive efforts to synthesize and test structural analogues with hydrogenase-like activity, it is certain that the inner coordination shell alone is insufficient, because relatively forcing conditions are required to obtain any reaction.4,5 In contrast, the most successful small molecule catalysts are equipped with outer-shell functionalities to facilitate H2 activation by the metal:6 significantly, in terms of their inner-sphere coordination, they do not resemble the active sites of hydrogenases beyond the use of Ni. In [NiFe]-hydrogenases, the residues in the outer shell that must be most intimately engaged with H2 binding and with H−H bond formation and cleavage form a “canopy” immediately above the inner shell, i.e., facing the binding site
ydrogenases catalyze the reversible splitting of molecular hydrogen (H2) into protons and electrons and allow a diverse range of microbes to utilize H2 in their energy metabolism.1,2 These organisms include facultative anaerobes such as Escherichia coli, obligate anaerobes such as Clostridium sp. and Desulfovibrio sp., and aerobes such as green algae, e.g., Chlamydomonas reinhardtii. Their minimal active sites contain Ni and/or Fe with biologically unusual CN− and CO ligands coordinated to the Fe, resulting in their classification into the broad classes of [NiFe]- and [FeFe]-hydrogenases (and [Fe]hydrogenases). Despite using only first-row transition metals, their active sites show a high efficiency for H2 oxidation with rates that may exceed those achieved with platinum.3 Although the direct use of isolated hydrogenases in large-scale energy technology is not feasible, a detailed atomic-level description of their mechanisms is important in guiding the design of synthetic catalysts based on abundant elements. Interconversion between H2 and H+ is the simplest example of a complete, © XXXX American Chemical Society
Received: August 28, 2016 Revised: November 21, 2016
A
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry
Figure 1. (A) Schematic and (B) crystal structure of the active site of E. coli Hyd-1 showing the conserved canopy residues. In panel A, atom X represents the variable (state-dependent) metal-bridging ligand. The bridging hydroxide (in the inactive Ni-B state) and the oxygenation of a bridging cysteine11,12 have been omitted from the crystal structure representation for the sake of clarity so that the active site structure represents the catalytically active Ni-SI state (PDB entry 5A4M).
for exogenous CO,7 the apparently vacant site on the Fe, and the variable exchangeable ligand that bridges the Ni and Fe atoms (Figure 1, where X represents an OH− or H− that is present in the oxidized inactive state, Ni-B, or active states Ni-R and Ni-C,5,8,9 respectively). The membrane-bound [NiFe]hydrogenase-1 (Hyd-1) from E. coli belongs to a special group known as O2-tolerant [NiFe]-hydrogenases (Group 1d1) because of its ability to oxidize H2, continuously, in the presence of O2.10 The active site is otherwise typical of the whole class known to date even, apparently, in the oxygenation of a coordinated cysteine that occurs upon reaction with O2 in the absence of H2, resulting in a recalcitrant inactive species known as “Unready” or Ni-A.11,12 Using Hyd-1 numbering, the central feature of the canopy is an arginine (R509) that suspends its potentially mobile side chain directly above the metal ions (average guanidinium Nζ1−Ni/Fe distances of 4.4 Å13). Indeed, R509 is the only residue that has a side chain close enough to interfere directly with an exchangeable ligand of Ni or Fe. The canopy also comprises two aspartates (D118 and D574) and a proline (P508).12 The carboxylate headgroups of D118 and D574 appear to form salt bridges to the R509 guanidinium headgroup, while P508 may be important in conserving the local fold to ensure that the arginine side chain is held in place. These residues are highly conserved, depending on the subgroup, as summarized in Figure S1 and Table S1. The canopy region contains structurally invariant water molecules that could mediate proton transfer,12 but does not include established important residues glutamate 28, valine 78, and leucine 126 (Hyd-1 numbering) that lie to the side of the Ni, as these are too distant to execute H−H bond formation or cleavage. Mutations of E28, thus far the most essential residue for catalysis as determined in studies of the [NiFe]-hydrogenase from Desulfovibrio f ructosovorans (Df [NiFe]-hydrogenase), abolish (E → Q) or retard (E → D) H2 oxidation by methyl viologen, as well as D2/H+ (solvent) exchange, but do not affect ortho/para H2 interconversion;14 consequently, although E28 is not close enough to be implicated directly in H−H bond cleavage, it must be important in longer range
proton transfer to or from the active site. Mutagenesis studies, again in Df [NiFe]-hydrogenase, have suggested that V78 and L126 control the access of gases to and from the active site.15,16 We recently succeeded in modifying the canopy region of Hyd-1 by stepwise mutagenesis of R509, D118, and D574.12 Our study of the fully formed and stable R509K variant, which showed only 1% of the activity of native Hyd-1, suggested the important new possibility that the arginine headgroup plays an essential mechanistic role, analogous to the pendent N base in [FeFe]-hydrogenases17 and the best functional analogues,18 by employing a “frustrated Lewis pair” effect to activate H2.19 On the basis of the pKa of free arginine (12.520), deprotonation of the guanidinium headgroup may seem difficult;21 however, during H2 production, a strongly basic intermediate (a hydrido ligand) is generated by successive electron transfers. Likewise, a transiently deprotonated guanidine would be a strong base, ideal as a proton acceptor during H2 oxidation.22 In addition to primary activation of H2, a mobile arginine side chain might also mediate proton transfer, an obvious possibility being removal of the H entity occupying the bridging position in the Ni-R and Ni-C states (Figure 1A). The high level of conservation of residues D118 and D574 suggested they also have an important role in H2 activation by Hyd-1, for example, in proton transfer; however, substitution of these residues with neutral asparagine (D574N and D118N/D574N) had little effect on the H2 oxidation activity or the active site structure of Hyd-1.12 The ability to re-engineer the canopy region is also significant for establishing structural aspects that are relevant to O2 tolerance and the often quoted “bias” of hydrogenases to favor a particular catalytic direction.23−25 We have now conducted extensive investigations using protein film electrochemistry (PFE), a suite of techniques that has already played an important role in the characterization of hydrogenases,26,27 to establish the importance of each of the canopy residues in determining some key kinetic parameters. We have therefore been able to measure activation parameters (ΔH⧧) and Michaelis constants (KHM2) over a range of potentials, establish quantitative tolerances to O2 and CO, and examine the abilities B
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry
For most variants, enzyme “films” were prepared in a consistent manner. The PGE electrode (geometric surface area of 0.03−0.05 cm2) was abraded using P400 sandpaper (Tufbak Durite) and wiped to remove any excess carbon. Enzyme (∼0.5−2 μL at 10−100 μM) was then applied directly to the surface of the electrode and allowed to soak for ∼30 s before being held under a stream of purified water to remove excess and/or unadsorbed enzyme. All films were reductively “activated” prior to each experiment to remove inactive states generated during purification: the electrode potential was poised at −0.66 V for 300 s, and then H2 oxidation activity was monitored at 0 or −0.1 V depending on the enzyme in question (see below). This procedure was repeated continuously until the H2 oxidation current had stabilized. In most cases, final current amplitudes were restricted to a maximum of 10 μA by lowering the enzyme coverage (by wiping the enzyme film with cotton wool) to remove H2 mass-transport limitations. An improved signal-to-noise ratio, where required for R509K, which has a much lower activity, was achieved by modifying the PGE electrode using a multiwall carbon nanotube/activated 1pyrenebutyric acid system to covalently attach the enzyme and increase the enzyme coverage, as described previously.32,33 Solution Assays. Steady-state H2 oxidation turnover rates under 100% H2 were measured for exhaustively activated native Hyd-1 and variant enzymes, as previously described.12 A methylene blue solution (Aldrich) at a final concentration of 25 μM in 50 mM potassium phosphate buffer (Fisher) at pH 6 was used as an electron acceptor. Solution assays were conducted at room temperature (25 °C) in an anaerobic glovebox (Belle Technologies) using an Ocean Optics S2000 fiber optic spectrometer controlled with OOIBase32 software (Ocean Optics, Inc.). The extinction coefficient adopted for methylene blue was 22400 cm−1 M−1.34 The concentrations of the enzyme solutions (0.1−5 mg mL−1) were determined by the Bradford assay method.35 The initial velocities measured at 600 nm were used to determine the turnover rates and are expressed as micromoles of H2 oxidized per minute per milligram of enzyme or as the apparent turnover rate, k (s−1). X-ray Crystallography. The Hyd-1 variant P508A was crystallized as described previously for all other canopy variants.12 Crystals were transferred to a cryoprotecting solution [100 mM Bis-tris (pH 6.0), 150 mM NaCl, 200 mM LiSO4, 25% (w/v) polyethylene glycol 3350, and 15% (v/v) glycerol] for 30 s before being flash-cooled in liquid nitrogen. X-ray diffraction data were collected at beamline I04-1 (Diamond Light Source) at a wavelength of 0.92 Å using a Pilatus 2M hybrid pixel array detector. To minimize radiation damage by photoreduction, diffraction data were collected using a helical line scan. Data reduction was performed using the DIALS automated data reduction pipeline, 36 and subsequent stages of model building and refinement were performed as previously described.12
to catalyze H2 evolution at low pH. We have also compared the D/H isotope effect on H2 oxidation by the native form and the slowest variant, R509K. In the interpretation of the large decrease in the activity of R509K, it is important to measure nonscaling parameters that complement the rate comparisons: in this respect, the activation parameters and isotope effects are very helpful. In addition to D118A, R509K, D574N, and D118N/D574N, we have included a new variant, P508A, in which the fold of the canopy could, in principle, be destabilized. Our investigations lead to a detailed assessment of the importance of each of these amino acids located in the outer shell closest to the site of H2 activation.
■
MATERIALS AND METHODS Molecular Biology. Cloning steps for the production of the ΔTM, D118A, R509K, and D118N/D574N variants have been described previously,12 and the P508A mutation was made on a pMAK-hyaB plasmid construct (Table S2). Following confirmation by DNA sequencing, the codon change was transferred from the resulting plasmid to the FTH004 strain, as previously described.12,28 Enzyme Production and Purification. All experiments were performed with native and variant Hyd-1 produced from E. coli MC4100-derived K-12 strain FTH004 or the desired variant strain containing an engineered HyaA protein with a His6 affinity tag at its C-terminus.29 Purification steps were followed as published previously.12 Protein Film Electrochemistry. In PFE, the electrode potential applied to the minuscule sample of enzyme adsorbed on a rotating electrode affords direct control of the oxidation state of the redox centers as various gases are introduced, while the catalytic activity is monitored as current. As a consequence, many reactivity characteristics can be measured over a wide H2 ), potential range, for example, Michaelis constants (KM inhibitor binding, O2 tolerance, activation enthalpies (ΔH⧧), and kinetic isotope effects.27 All PFE experiments were performed in an anaerobic glovebox (MBraun, Vacuum Atmospheres, or Belle Technologies) under a N2 atmosphere (80% activity) for each enzyme (see Figure S7 and Table 1)], measured by conventional steady-state assays in an (initially) H2-saturated solution, are listed in Table 1. The values were also converted to turnover frequencies kcat (per enzyme molecule). In view of the data obtained from PFE measurements (see below), we could be confident that in all cases (apart from D118A) 100% H2 provides saturating conditions (assuming Vmax is normally achieved at >5KHM2). The solution data, which now include P508A, supersede but remain in excellent agreement with those published recently.12 Electrocatalytic Windows of the Variants. The characteristic potentials for reductive activation and onset of H2 oxidation (see later) define the potential window for activity of each enzyme (Figure S8). The onset potential for H2 oxidation is, within error, the same for all variants, but ease of reductive activation, as defined by “Eswitch”,26 differs greatly. Regardless of its mechanistic origins,27 when measured at very slow scan rates, such as 0.1 mV s−1, Eswitch signifies the stability of Ni-B, the oxidized inactive “resting” state, and this stability increases in the following order: native < P508A and R509K < D118A < D574N < D118N/D574N. This order is important for O2 tolerance, because negative shifts in Eswitch mean that increasingly strong reducing conditions are required for sustained activity in the presence of O2 (see below). Differing values of Eswitch also had to be taken into consideration in measuring potential-dependent parameters: with D118N/ D574N, for example, Ni-B stabilization is so great that measurements of KHM2 and ΔH⧧ had to be restricted to potentials of D118 > P508 > D574 D118 > P508 ∼ R509 > D574 D574 > D118 > P508 ∼ R509 D118 > P508 ∼ R509 ∼ D574 D118 ∼ D574 > P508 ∼ R509 R509 > D574 > D118 > P508 G
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry
in k1, it follows that the binding of H2 must be slower for D118A than all other enzymes, and this suggests that D118 influences the access of H2 to the active site. The potential independence and low value of KHM2 for R509K are fully consistent with a Michaelis−Menten process whereby kcat is not limited by long range electron transfer and is sufficiently small that KHM2 simplifies to k−1/k1, the dissociation constant for H2. Replacing any of the canopy residues has a limited influence on the susceptibility to CO inhibition when measured electrochemically in 20% H2. In all cases, the variants retain the unusual characteristic, shared with other O2-tolerant membrane-bound [NiFe]-hydrogenases, that CO is a much less effective inhibitor of H2 oxidation than in standard [NiFe]hydrogenases.29,40 The term “O2 tolerance” refers to the ability to sustain catalytic activity in the presence of O2.10,47 A working hypothesis for the mechanism of O2 tolerance in membranebound [NiFe]-hydrogenases such as Hyd-1 relies on the ability (a) to have four electrons immediately ready to transfer when O2 attacks, completely reducing it to water, and (b) to rereduce, rapidly, the Ni(III)-OH product of O2 attack (known as Ni-B or Ready) and thus reactivate the enzyme.10,47 The FeS electron relay is crucial in supplying the electrons, and a feature of O2-tolerant membrane-bound hydrogenases is that the FeS centers have more positive reduction potentials (i.e., increased electron availability) than those of standard [NiFe]-hydrogenases,48,49 including an unusual proximal [4Fe-3S] cluster that can donate two electrons.50 An inability to deal with O2 in this way (Hyd-1 is a “H2-oxidase”51) results in the accumulation of more permanently inactive states, particularly Ni-A (also defined kinetically as an Unready state) in which one of the bridging Cys-S residues is oxygenated.11,12 The necessity of having electrons available means that in the absence of H2 and in the absence of an electron source, even O2-tolerant [NiFe]hydrogenases should form Ni-A: indeed, this is true for Hyd-1 and the canopy variants, because C79 is modified by oxygenation in each structure of the as-isolated enzymes (see ref 12 and Figure S4). Once the safeguards are in place to guarantee that only Ni-B is produced during H2 oxidation under aerobic conditions, the observed fractional activity (f) is given by f = rateact/(rateact + rateinact),10 where rateact is the rate of activation of Ni-B (potential-dependent) and rateinact is the rate of inactivation by O2. Further advantages are therefore afforded if (a) passage of O2 to the active site is restricted, relative to H2,52,53 and (b) NiB is easily reduced to restore activity. We must therefore acknowledge the differences in Ni-B stabilization reflected by the different potentials required to reduce Ni-B for each variant (which is produced anaerobically as well as aerobically26,38), noting that f and therefore O2 tolerance are always improved at lower potentials.10,38 For native Hyd-1 at pH 6, Eswitch is 0.24 V; hence, in O2 exposure experiments at 0.03 V (transient) or 0 V (prolonged), there is approximately 0.21 or 0.24 V, respectively, of favorable reducing potential for reactivating Ni-B. This advantage is diminished for each of the aspartate variants, D118N/D574N being the extreme case with only 0.09 V to favor activation of Ni-B (at 0 V). None of the single-site variants have acquired any significant O2 sensitivity (at a given potential) beyond that attributable to differences in the ease of reactivation of Ni-B, as reflected in Eswitch. Notably, Eswitch becomes more negative as the formal negative charge of the canopy residues is decreased, a trend that might reflect the stabilization afforded to resident OH− ions that are necessary to
position of the D574 carboxylate group nor the H-bond between Nε of R509 (average distance of 2.8 Å12) and the backbone carbonyl of D574 is affected in the P508A variant. Alterations of the canopy residues have only a small impact on the activation enthalpy for H2 oxidation except in the case of R509K and the two D118 variants. The steep decrease in activation enthalpy for D118A as the potential is increased most obviously reflects the increase in KHM2 and the increasing exothermic contribution to ΔH⧧ on H2 binding at the higher potentials (reflecting the progression of control from kcat/KHM2 to kcat). The same potential-dependent trends are evident from the restricted data that can be measured for D118N/D574N. The most meaningful comparison may be made between native Hyd-1 and R509K because they have greatly different rates, small KHM2 values, and a similar flat dependence of ΔH⧧ on potential. For R509K, the potential profile mirrors that of native Hyd-1 but is lowered uniformly by 5−7 kJ mol−1 at each potential. This result, which must predominantly reflect the contribution to kcat, is very important, mechanistically, because the large attenuation of the H2 oxidation rate for R509K compared with that of native Hyd-1 (a 102-fold decrease) must therefore result from a much larger entropy barrier. On the basis of the ratio of steady-state turnover rates (257.4/3.3), the extra barrier equates to a ΔΔS⧧ of −52 J K−1 mol−1 (see the Supporting Information). The small KIEs for both native Hyd-1 and R509K show that H−H bond cleavage is very unlikely to be involved in the ratedetermining step. The differences in the KIE data for native Hyd-1 (1.4) and R509K (1.1) are significant when considering a mechanism for H2 activation in which the role of the immediate base may be fulfilled by the deprotonated guanidine group of R509. Once H2 is split at the active site into a Ni−H− and H+ (on the base), a proton transfer would be required so that the guanidinium headgroup (or amine for R509K) is deprotonated for the next round of catalysis. A comparison of the pKa’s20 of free lysine (pKa = 10.5) and arginine (pKa = 12.5) shows that deprotonation of a lysine NH3+ group should be easier than for the guanidinium group of arginine, and less likely to be rate-limiting. This explanation would be consistent with the lower ΔH⧧ observed for R509K, for which the far less favorable ΔS⧧ suggests suboptimal preorganization, as supported by the increased flexibility of the lysine side chain that is evident from temperature factor analysis of the crystal structure.12 For the sake of simplicity, we define KHM2 as (k−1 + kcat)/k1, where k1 and k−1 are the on and off rates for the H2 binding step, respectively, and kcat is the turnover frequency,45 the latter including proton transfers and potential-dependent electron transfer.46 The KHM2 value for R509K is low and independent of potential. Conversely, D118A differs from all the other variants by having a large potential-dependent KHM2 that results in the large value of 0.3 mM at 0.24 V. Assuming kcat is not unusually large in D118A (because at the high ρH2 that was required, the condition of Vmax in steady-state assays was not met for this variant), the large disparity in values for R509K and D118A must lie in the relative on and off rates, k1 and k−1, respectively. A large value of k−1 would suggest that the interaction between molecular H2 and the active site is weak, and were this to be the case, D118A should show enhanced H2 production activity; however, D118A is the only variant that does not evolve H2 at low pH. An obvious problem with this simple analysis is that H2 evolution and H2 oxidation occur under very different potential and pH conditions. Conversely, if the differences in behavior lie H
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry coordinate to and stabilize Ni(III). It may therefore be significant that D118 and D574 are conserved in all O2tolerant membrane-bound [NiFe]-hydrogenases [Group 1d (Table S1)]. In no case is an Unready state (assumed to be NiA) produced in PFE experiments under H2; given the hypothesis for the mechanism of O2 tolerance, this result may not be surprising because all variants have a fully formed FeS chain capable of supplying all four electrons required to avoid partially reduced O2 species. These results reaffirm the proposition that it is the FeS clusters that ultimately control O2 tolerance and the active site does not influence the response of Hyd-1 to O2 beyond ensuring there is a wide window for O2tolerant H2 oxidation by ensuring that Ni-B is rapidly reduced even at high potentials. Our experiments show clearly that D118 is important in controlling KHM2. The gas channel proposed by Fontecilla-Camps and co-workers54 (based on Xe binding and calculations for D. f ructosovorans [NiFe]-hydrogenase) and by Scheerer and coworkers52 (based on Kr binding and calculations for R. eutropha membrane-bound [NiFe]-hydrogenase) terminates with residues V78 and L126 (E. coli numbering): the carboxylate of D118 also forms part of this bottleneck leading to the [NiFe] site. The role of residue D118 in mediating H2 access is very interesting: this residue is highly conserved in Group 1 hydrogenases, a notable exception being [NiFeSe]-hydrogenases in Group 1a, which have a serine in this position. The [NiFeSe]-hydrogenase from Desulfomicrobium baculatum has a much higher KHM2 (280 μM at −50 mV) than those of other Group 1 hydrogenases,55 suggesting that experiments with a D118S variant may help resolve this question. Mutations of V78 and L126 that decrease the bottleneck radius in Df [NiFe]-hydrogenase resulted in large increases in KHM2, decreases in CO binding and dissociation rates, and a change in O2 sensitivity.15,16 In contrast, although D118A shows a large increase in KHM2 (noting that the PFE experiments reveal a large potential dependence), there is little change in CO inhibition or O2 tolerance, and as with the singular lack of H2 evolution activity of D118A, the residue probably does more than simply regulate H2 access (compared to V78 and/or L126). We can now establish an overall order of importance of canopy residues in Hyd-1 (R509 ≫ D118 > D574 and P508), based in the first instance on the ability to catalyze H2 oxidation. Comparisons of different characteristics are compiled in Table 2. Combining our results with data and conclusions from others, we can now construct a model of the key residues that line the active site (Figure 7). Residues E28, V78, and L126 form a ring with D118, which defines the entrance of the gas channel culminating in the arginine headgroup that is suspended above the Ni and Fe. Like native Hyd-1, all canopy variants exhibit oxygenation of C79 and have well-defined stable structures, apart from P508A, in which there is some conformational heterogeneity and consequently some loss of Ni. All canopy variants except for D118A exhibit H2 production activity at pH 3. Residues R509 and P508 have little influence on O2 tolerance; on the other hand, mutations of D118 and D574 to asparagine (which decreases the negative electrostatic potential in the active site) stabilize the inactive Ni-B state, and more reducing conditions are required to restore activity following O2 exposure. Once O2 tolerance is safeguarded by electron availability, these canopy residues play a role in providing a wide potential region (at high potential) in which Ni-B can be reactivated to ensure a high level of continuous H2 oxidation activity. Previously, E28 was identified as being
Figure 7. Spatial arrangement of outer coordination-shell residues of Hyd-1, signified according to their importance for catalytic activity, including essential amino acids identified in studies using Df [NiFe]hydrogenase. Residues colored red are highly important for H2 oxidation activity and those colored white less so. Residues E28, V78, D118, and L126 encircle the end of two putative gas channels (shown as mesh), with the guanidine headgroup of R509 at their center positioned immediately next to the substrate binding site of the [NiFe] center. Gas channels were calculated using the Caver plugin for pymol56 using a probe radius of 1.4 Å. Iron and nickel ions are shown as orange and green spheres, respectively, and a bridging ligand site is highlighted (red sphere).
critically important for proton transfer away from the active site, but not directly involved in H2 activation in Df [NiFe]hydrogenase.14 Combined with the fact that R509 is the only noncoordinating residue that can interact directly with the coordinated substrate or intermediate, our new evidence provides further support for a direct role for R509 in catalysis, either in H−H bond cleavage and/or in mediating proton transfer, most obviously from the bridging position at which H− is bound. On a more general note, our work has explored systematically and in detail the importance of each of the canopy residues for the various reactions that O2-tolerant [NiFe]-hydrogenases undergo: the quantitative data thus obtained should be valuable in understanding how these enzymes have evolved.
■
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.6b00868. Figures and tables detailing a sequence alignment, the construction of variants via site-directed mutagenesis, denaturing electrophoresis, crystal structures and refinement statistics of P508A, electrochemical profiles, Eyring plots, determination of the difference in activation entropies, pH optima, transient and prolonged O2 exposure experiments for all enzymes, and current-offset adjustment for R509K (PDF) Accession Codes
The P508A structure is deposited at the Protein Data Bank as entry 5JRD.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Simon E. V. Phillips: 0000-0001-8922-0250 I
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry
(9) Ogata, H., Nishikawa, K., and Lubitz, W. (2015) Hydrogens detected by subatomic resolution protein crystallography in a [NiFe] hydrogenase. Nature 520, 571−574. (10) Evans, R. M., Parkin, A., Roessler, M. M., Murphy, B. J., Adamson, H., Lukey, M. J., Sargent, F., Volbeda, A., Fontecilla-Camps, J. C., and Armstrong, F. A. (2013) The principles of sustained enzymatic hydrogen oxidation in the presence of oxygen − the crucial influence of high potential Fe-S clusters in the electron relay of [NiFe]-hydrogenases. J. Am. Chem. Soc. 135, 2694−2707. (11) Volbeda, A., Martin, L., Barbier, E., Gutiérrez-Sanz, O., De Lacey, A. L., Liebgott, P.-P., Dementin, S., Rousset, M., and FontecillaCamps, J. C. (2015) Crystallographic studies of [NiFe]-hydrogenase mutants: towards consensus structures for the elusive unready oxidized states. J. Biol. Inorg. Chem. 20, 11−22. (12) Evans, R. M., Brooke, E. J., Wehlin, S. A. M., Nomerotskaia, E., Sargent, F., Carr, S. B., Phillips, S. E. V., and Armstrong, F. A. (2015) Mechanism of hydrogen activation by [NiFe] hydrogenases. Nat. Chem. Biol. 12, 46−50. (13) Volbeda, A., Amara, P., Darnault, C., Mouesca, J.-M., Parkin, A., Roessler, M. M., Armstrong, F. A., and Fontecilla-Camps, J. C. (2012) X-ray crystallographic and computational studies of the O2-tolerant [NiFe]-hydrogenase 1 from Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 109, 5305−5310. (14) Dementin, S., Burlat, B., De Lacey, A. L., Pardo, A., AdryanczykPerrier, G., Guigliarelli, B., Fernandez, V. M., and Rousset, M. (2004) A Glutamate is the Essential Proton Transfer Gate during the Catalytic Cycle of the [NiFe] Hydrogenase. J. Biol. Chem. 279, 10508−10513. (15) Dementin, S., Leroux, F., Cournac, L., de Lacey, A. L., Volbeda, A., Léger, C., Burlat, B., Martinez, N., Champ, S., Martin, L., Sanganas, O., Haumann, M., Fernández, V. M., Guigliarelli, B., Fontecilla-Camps, J. C., and Rousset, M. (2009) Introduction of Methionines in the Gas Channel Makes [NiFe] Hydrogenase Aero-Tolerant. J. Am. Chem. Soc. 131, 10156−10164. (16) Liebgott, P.-P., De Lacey, L. A., Burlat, B., Cournac, L., Richaud, P., Brugna, M., Fernandez, V. M., Guigliarelli, B., Rousset, M., Léger, C., and Dementin, S. (2011) Original Design of an Oxygen-Tolerant [NiFe] Hydrogenase: Major Effect of a Valine-to-Cysteine Mutation near the Active Site. J. Am. Chem. Soc. 133, 986−997. (17) Siebel, J. F., Adamska-Venkatesh, A., Weber, K., Rumpel, S., Reijerse, E., and Lubitz, W. (2015) Hybrid [FeFe]-Hydrogenases with Modified Active Sites Show Remarkable Residual Enzymatic Activity. Biochemistry 54, 1474−1483. (18) Dutta, A., DuBois, D. L., Roberts, J. A. S., and Shaw, W. J. (2014) Amino acid modified Ni catalyst exhibits reversible H2 oxidation/production over a broad pH range at elevated temperatures. Proc. Natl. Acad. Sci. U. S. A. 111, 16286−16291. (19) Stephan, D. W., and Erker, G. (2010) Frustrated lewis pairs: metal-free hydrogen activation and more. Angew. Chem., Int. Ed. 49, 46−76. (20) Voet, D., and Voet, J. G. (2011) Biochemistry, 4th ed., pp 68−69, John Wiley & Sons, Inc., New York. (21) Harms, M. J., Schlessman, J. L., Sue, G. R., and García-Moreno E, B. (2011) Arginine residues at internal positions in a protein are always charged. Proc. Natl. Acad. Sci. U. S. A. 108, 18954−18959. (22) Morris, R. H. (2014) Estimating the Acidity of Transition Metal Hydride and Dihydrogen Complexes by Adding Ligand Acidity Constants. J. Am. Chem. Soc. 136, 1948−1959. (23) Hexter, S. V., Esterle, T. F., and Armstrong, F. A. (2014) A unified model for surface electrocatalysis based on observations with enzymes. Phys. Chem. Chem. Phys. 16, 11822−11833. (24) Hexter, S. V., Grey, F., Happe, T., Climent, V., and Armstrong, F. A. (2012) Electrocatalytic mechanism of reversible hydrogen cycling by enzymes and distinctions between the major classes of hydrogenases. Proc. Natl. Acad. Sci. U. S. A. 109, 11516−11521. (25) Murphy, B. J., Sargent, F., and Armstrong, F. A. (2014) Transforming an oxygen-tolerant [NiFe] uptake hydrogenase into a proficient, reversible hydrogen producer. Energy Environ. Sci. 7, 1426− 1433.
Fraser A. Armstrong: 0000-0001-8041-2491 Funding
This research was supported by the UK Biological and Biotechnology Sciences Research Council (Grants BB/ I022309-1 and BB/L009722/1 to F.A.A.). A studentship for E.J.B. was supported by grants from Global Innovation Initiative and the UK Engineering and Physical Sciences Research Council. The Islamic Development Bank provided a studentship for S.T.A.I. under the Merit Scholarship Programme for High Technology. F.A.A. is a Royal SocietyWolfson Research Merit Award holder. We are also that the Margaret Leighton SURF Fellowship (California Institute of Technology) provided a visiting studentship to G.M.R. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank Diamond Light Source for beam time (proposal mx12346) and the staff of beamline I04-1 for their assistance. We thank Elena Nomerotskaia for invaluable technical assistance throughout this work.
■
ABBREVIATIONS Hyd-1, hydrogenase-1; PFE, protein film electrochemistry; PGE, pyrolytic graphite edge; RDE, rotating disc electrode; SCE, saturated calomel electrode; SHE, standard hydrogen electrode; KIE, kinetic isotope effect; scc min−1, standard cubic centimeters per minute; PDB, Protein Data Bank; Df , D. f ructosovorans.
■
REFERENCES
(1) Greening, C., Biswas, A., Carere, C. R., Jackson, C. J., Taylor, M. C., Stott, M. B., Cook, G. M., and Morales, S. E. (2016) Genomic and metagenomic surveys of hydrogenase distribution indicate H2 is a widely utilised energy source for microbial growth and survival. ISME J. 10, 761−777. (2) Vignais, P. M., and Billoud, B. (2007) Occurrence, classification, and biological function of hydrogenases: An overview. Chem. Rev. 107, 4206−4272. (3) Jones, A. K., Sillery, E., Albracht, S. P. J., and Armstrong, F. A. (2002) Direct comparison of the electrocatalytic oxidation of hydrogen by an enzyme and a platinum catalyst. Chem. Commun., 866−867. (4) Simmons, T. R., Berggren, G., Bacchi, M., Fontecave, M., and Artero, V. (2014) Mimicking hydrogenases: From biomimetics to artificial enzymes. Coord. Chem. Rev. 270−271, 127−150. (5) Lubitz, W., Ogata, H., Rüdiger, O., and Reijerse, E. (2014) Hydrogenases. Chem. Rev. 114, 4081−4148. (6) Ginovska-Pangovska, B., Dutta, A., Reback, M. L., Linehan, J. C., and Shaw, W. J. (2014) Beyond the Active Site: The Impact of the Outer Coordination Sphere on Electrocatalysts for Hydrogen Production and Oxidation. Acc. Chem. Res. 47, 2621−2630. (7) Ogata, H., Mizoguchi, Y., Mizuno, N., Miki, K., Adachi, S.-i., Yasuoka, N., Yagi, T., Yamauchi, O., Hirota, S., and Higuchi, Y. (2002) Structural Studies of the Carbon Monoxide Complex of [NiFe] hydrogenase from Desulfovibrio vulgaris Miyazaki F: Suggestion for the Initial Activation Site for Dihydrogen. J. Am. Chem. Soc. 124, 11628−11635. (8) Ogata, H., Krämer, T., Wang, H., Schilter, D., Pelmenschikov, V., van Gastel, M., Neese, F., Rauchfuss, T. B., Gee, L. B., Scott, A. D., Yoda, Y., Tanaka, Y., Lubitz, W., and Cramer, S. P. (2015) Hydride bridge in [NiFe]-hydrogenase observed by nuclear resonance vibrational spectroscopy. Nat. Commun. 6, 7890. J
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry
Hydrogenase Reveals Insight into O2-Tolerant H2 Oxidation. Structure 24, 285−292. (45) Fersht, A. (1999) Structure and mechanism in protein science: A guide to enzyme catalysis and protein folding, 5th Print, 2003 ed., p 105, W. H. Freeman and Co., New York. (46) Léger, C., Jones, A. K., Albracht, S. P. J., and Armstrong, F. A. (2002) Effect of a Dispersion of Interfacial Electron Transfer Rates on Steady State Catalytic Electron Transport in [NiFe]-hydrogenase and Other Enzymes. J. Phys. Chem. B 106, 13058−13063. (47) Cracknell, J. A., Wait, A. F., Lenz, O., Friedrich, B., and Armstrong, F. A. (2009) A kinetic and thermodynamic understanding of O2 tolerance in [NiFe]-hydrogenases. Proc. Natl. Acad. Sci. U. S. A. 106, 20681−20686. (48) Parkin, A., and Sargent, F. (2012) The hows and whys of aerobic H2 metabolism. Curr. Opin. Chem. Biol. 16, 26−34. (49) Roessler, M. M., Evans, R. M., Davies, R. A., Harmer, J., and Armstrong, F. A. (2012) EPR spectroscopic studies of the Fe-S clusters in the O2-tolerant [NiFe]-hydrogenase Hyd-1 from E. coli, and characterization of the unique [4Fe-3S] cluster by HYSCORE. J. Am. Chem. Soc. 134, 15581−15594. (50) Pandelia, M.-E., Nitschke, W., Infossi, P., Giudici-Orticoni, M.T., Bill, E., and Lubitz, W. (2011) Characterization of a unique [FeS] cluster in the electron transfer chain of the oxygen tolerant [NiFe] hydrogenase from Aquifex aeolicus. Proc. Natl. Acad. Sci. U. S. A. 108, 6097−6102. (51) Wulff, P., Day, C. C., Sargent, F., and Armstrong, F. A. (2014) How oxygen reacts with oxygen-tolerant respiratory [NiFe]-hydrogenases. Proc. Natl. Acad. Sci. U. S. A. 111, 6606−6611. (52) Kalms, J., Schmidt, A., Frielingsdorf, S., van der Linden, P., von Stetten, D., Lenz, O., Carpentier, P., and Scheerer, P. (2016) Krypton Derivatization of an O2-Tolerant Membrane-Bound [NiFe] Hydrogenase Reveals a Hydrophobic Tunnel Network for Gas Transport. Angew. Chem., Int. Ed. 55, 5586−5590. (53) Buhrke, T., Lenz, O., Krauss, N., and Friedrich, B. (2005) Oxygen Tolerance of the H2-sensing [NiFe] Hydrogenase from Ralstonia eutropha H16 Is Based on Limited Access of Oxygen to the Active Site. J. Biol. Chem. 280, 23791−23796. (54) Montet, Y., Amara, P., Volbeda, A., Vernede, X., Hatchikian, E. C., Field, M. J., Frey, M., and Fontecilla-Camps, J. C. (1997) Gas access to the active site of Ni-Fe hydrogenases probed by X-ray crystallography and molecular dynamics. Nat. Struct. Biol. 4, 523−526. (55) McDowall, J. S., Murphy, B. J., Haumann, M., Palmer, T., Armstrong, F. A., and Sargent, F. (2014) Bacterial formate hydrogenlyase complex. Proc. Natl. Acad. Sci. U. S. A. 111, E3948− E3956. (56) Chovancová, E., Pavelka, A., Beneš, P., Strnad, O., Brezovský, J., Kozlíková, B., Gora, A., Š ustr, V., Klvaň a , M., Medek, P., Biedermannová, L., Sochor, J., and Damborský, J. (2012) CAVER 3.0: A Tool for the Analysis of Transport Pathways in Dynamic Protein Structures. PLoS Comput. Biol. 8, e1002708.
(26) Vincent, K. A., Parkin, A., and Armstrong, F. A. (2007) Investigating and Exploiting the Electrocatalytic Properties of Hydrogenases. Chem. Rev. 107, 4366−4413. (27) Armstrong, F. A., Evans, R. M., Hexter, S. V., Murphy, B. J., Roessler, M. M., and Wulff, P. (2016) Guiding Principles of Hydrogenase Catalysis Instigated and Clarified by Protein Film Electrochemistry. Acc. Chem. Res. 49, 884−892. (28) Hamilton, C. M., Aldea, M., Washburn, B. K., Babitzke, P., and Kushner, S. R. (1989) New method for generating deletions and gene replacements in Escherichia coli. J. Bacteriol. 171, 4617−4622. (29) Lukey, M. J., Parkin, A., Roessler, M. M., Murphy, B. J., Harmer, J., Palmer, T., Sargent, F., and Armstrong, F. A. (2010) How Escherichia coli is equipped to oxidize hydrogen under different redox conditions. J. Biol. Chem. 285, 3928−3938. (30) Evans, R. M., and Armstrong, F. A. (2014) Electrochemistry of Metalloproteins: Protein Film Electrochemistry for the Study of E. coli [NiFe]-Hydrogenase-1. In Metalloproteins: Methods and Protocols (Fontecilla-Camps, J. C., and Nicolet, Y., Eds.) pp 73−94, Humana Press, Totowa, NJ. (31) Bard, A. J., and Faulkner, L. R. (2001) Electrochemical methods, backcover, Wiley, New York. (32) Krishnan, S., and Armstrong, F. A. (2012) Order-of-magnitude enhancement of an enzymatic hydrogen-air fuel cell based on pyrenyl carbon nanostructures. Chem. Sci. 3, 1015−1023. (33) Rüdiger, O., Abad, J. M., Hatchikian, E. C., Fernandez, V. M., and De Lacey, A. L. (2005) Oriented Immobilization of Desulfovibrio gigas Hydrogenase onto Carbon Electrodes by Covalent Bonds for Nonmediated Oxidation of H2. J. Am. Chem. Soc. 127, 16008−16009. (34) Arp, D. J., and Burris, R. H. (1981) Kinetic mechanism of the hydrogen-oxidizing hydrogenase from soybean nodule bacteroids. Biochemistry 20, 2234−2240. (35) Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248−254. (36) Waterman, D. G., Winter, G., Parkhurst, J. M., FuentesMontero, L., Hattne, J., Brewster, A., Sauter, N. K., and Evans, G. (2013) The DIALS framework for integration software. CCP4 Newsletter on Protein Crystallography 49, 16−19. (37) Lamle, S. E., Vincent, K. A., Halliwell, L. M., Albracht, S. P. J., and Armstrong, F. A. (2003) Hydrogenase on an electrode: a remarkable heterogeneous catalyst. Dalton. Trans., 4152−4157. (38) Lukey, M. J., Roessler, M. M., Parkin, A., Evans, R. M., Davies, R. A., Lenz, O., Friedrich, B., Sargent, F., and Armstrong, F. A. (2011) Oxygen-Tolerant [NiFe]-Hydrogenases: The Individual and Collective Importance of Supernumerary Cysteines at the Proximal Fe-S Cluster. J. Am. Chem. Soc. 133, 16881−16892. (39) Léger, C., Dementin, S., Bertrand, P., Rousset, M., and Guigliarelli, B. (2004) Inhibition and aerobic inactivation kinetics of Desulfovibrio f ructosovorans NiFe hydrogenase studied by protein film voltammetry. J. Am. Chem. Soc. 126, 12162−12172. (40) Vincent, K. A., Cracknell, J. A., Lenz, O., Zebger, I., Friedrich, B., and Armstrong, F. A. (2005) Electrocatalytic hydrogen oxidation by an enzyme at high carbon monoxide or oxygen levels. Proc. Natl. Acad. Sci. U. S. A. 102, 16951−16954. (41) Goldet, G., Wait, A. F., Cracknell, J. A., Vincent, K. A., Ludwig, M., Lenz, O., Friedrich, B., and Armstrong, F. A. (2008) Hydrogen production under aerobic conditions by membrane-bound hydrogenases from Ralstonia species. J. Am. Chem. Soc. 130, 11106−11113. (42) Armstrong, F. A., Belsey, N. A., Cracknell, J. A., Goldet, G., Parkin, A., Reisner, E., Vincent, K. A., and Wait, A. F. (2009) Dynamic electrochemical investigations of hydrogen oxidation and production by enzymes and implications for future technology. Chem. Soc. Rev. 38, 36−51. (43) Trevino, S. R., Schaefer, S., Scholtz, J. M., and Pace, C. N. (2007) Increasing protein conformational stability by optimizing betaturn sequence. J. Mol. Biol. 373, 211−218. (44) Schäfer, C., Bommer, M., Hennig, E., Jeoung, J.-H., Dobbek, H., and Lenz, O. (2016) Structure of an Actinobacterial-Type [NiFe]K
DOI: 10.1021/acs.biochem.6b00868 Biochemistry XXXX, XXX, XXX−XXX
