Generating Single Metalloprotein Crystals in Well-Defined Redox States – Electrochemical Microspectroscopy As a Tool for Mechanistic Studies

The activation and redox chemistry of small molecules by metalloenzymes play important roles in vital processes such as the global carbon and nitrogen cycles, and as sources of energy or low-potential reducing equivalents in cellular environments (for example bacteria such as R. eutropha have evolved the capability to use CO2 as sole carbon source and H2 as sole energy source). Despite the current urgent need for sustainable fuels and greener routes to chemical synthesis, detailed mechanistic understanding of how small molecule activation reactions are carried out in nature is often lacking. In part this is due to the high turnover frequencies achieved by metalloenzymes such as carbon monoxide dehydrogenase (CO2/CO, >2000 s-1), formate dehydrogenase (CO2/HCOO-, >112 s-1), and hydrogenases (H+/H2, >9000 s-1). By combining electrochemical control with microspectroscopic imaging we are able to take advantage of high protein concentrations in the crystalline state, and apparently retarded reactivity, to probe bioelectrocatalytic processes at the sub-single crystal level and gain fresh mechanistic insight. X-ray crystallography is almost ubiquitous in contributing structural insight into the mechanism of redox proteins. At present, significant challenges arise in preparing (and maintaining) protein crystals in all redox states,1 there is often no means of controlling and verifying the state or uniformity of the crystallised sample, and the possibilities for kinetic study are limited. There is a growing trend towards implementing optical spectroscopic techniques which are sensitive to both electronic structure and radiation-induced effects on crystallography beamlines.2 However, these are mainly used to assess the extent of photodamage and photoreduction during X-ray data collection, or confirm that crystallisation maintains the protein in a physiologically relevant state. Even in cases where microspectroscopy is used to gain information to complement the crystal structure, the ability to precisely control and maintain a desired redox state remains largely elusive. Recently we have developed an approach that allows precise electrochemical control over the redox state of a crystallised protein at a carbon electrode, while monitoring the redox state in situ using diffraction-limited synchrotron infrared microspectroscopic imaging (Figure 1).3 Using NiFe hydrogenase 1 from E. coli as a case study we show that the redox state of a crystal can be rapidly and reversibly manipulated, and that the whole crystal sample responds uniformly to the applied potential. The crystals are stable for several hours over multiple potential cycles. Intriguingly, retarded reactivity in the crystalline state gives insight into proton transfer during H+ reduction by NiFe hydrogenases, adding missing chemical detail to the hydrogenase catalytic cycle.4 The addition of electrochemical control will allow us to ‘trap’ NiFe hydrogenase crystals in elusive (or otherwise short-lived) redox states, enabling structural information to be obtained in catalytically-relevant states that have to date been crystallographically inaccessible. The new electrochemical method presented here can be readily coupled with other spectroscopies, thereby addressing a variety of chromophores in crystallo and greatly broadening the range of bioelectrocatalytic processes that can be studied at the sub-single crystal level. References Bowman SEJ et al., Acc. Chem. Res. 2016, 49, 695. von Stetten D et al., Acta Crystallogr. 2015, 71, 15. Ash PA et al., Chem. Commun. 2017, 53, 5858. Ash PA et al., ACS Catal. 2017, 7, 2471. Figure 1. Combined infrared microspectroscopy and electrochemistry of NiFe hydrogenase single crystals. (A) Schematic view of the reflection-absorbtion cell developed for this work (WE, working electrode; CE, counter electrode; RE, reference electrode). (B) Synchrotron IR microspectroscopy allows mapping at the sub-crystal level during transitions between well-defined redox states, as evidenced by the stretching vibration of the intrinsic CO ligand at the hydrogenase active site (inset and structure). Figure 1