Research (Academic) - Simon J. George

Hydrogenase Mechanism and Biosynthesis

The hydrogenases catalyze the simplest chemical reaction: the interconversion of H2 and H+, a reaction that also underpins industrial hydrogen energy technologies such as fuel cells and solar hydrogen generation.

Understanding the biological mechanism is important as current industrial processes require rare and expensive metals such as Pt, whereas biology uses cheap and readily available Fe and Ni. Fifteen of my publications investigate the mechanism and biosynthesis of the hydrogenase metal cofactors as well as related biological chemistry.

In work recently published in two papers in the journal Science and others elsewhere, I, together with Jon Kuchenreuther and James Swartz (Stanford) and R. David Britt (U.C. Davis), investigated the biosynthesis of the H-cluster active site of the FeFe hydrogenases The H-cluster (see figure) comprises a conventional [4Fe-4S] cluster coordinated to an unique 2Fe sub-cluster decorated with CO and CN- ligands. The maturase enzymes HydE, HydF and HydG are known to combine to construct the 2Fe sub-cluster, however the role of each maturase is not clear.

In our Science papers, we used EPR and SF-IR to show that the Radical-SAM enzyme HydG uses a [4Fe-4S] cluster to assemble Fe(CO)2(CN) Synthons for the H-cluster. The figure shows SF-IR data illustrating the step-wise chemistry of HydG with its tyrosine and SAM. At very short times, a 4-oxidobenzyl radical (4OB•) is detected by EPR, at longer times SF-IR observes Complex A: a Fe(CO)(CN) species which we assign to a Fe(CO)(CN):[4Fe-4S] cluster. At longer times this is replaced by Complex B: which now has 2 CO groups bound making a Fe(CO)2(CN):[4Fe-4S] species. Experiments using 57Fe substitution monitored by EPR/ENDOR spectroscopy show that the CO/CN- bound Fe in the H-cluster arises from Fe in HydG. Hence we propose that the product of HydG is a Fe(CO)(CN) synthon and note that in this way biology avoids having to release CO or CN- into the cell.

Lanthanide Biological Chemistry and Disease

Gadolinium-based contrast agents (GBCAs) are used in about 30% of all medical MRI diagnostic procedures, and are generally viewed as safe for patients with normal kidney function.  GBCAs contain Gd coordinated to a chelator, which makes the Gd highly soluble while rendering it essentially non-toxic.  In individuals with normal kidney function the GBCA is excreted in a few hours.  In patients with poor kidney function, exposure to GBCAs is associated with the rare, debilitating and sometimes fatal disease, Nephrogenic Systemic Fibrosis (NSF).

In work published in the British Journal of Dermatology, I, in collaboration with Jerrold Abraham (Upstate Medical U.), used micro-XRF imaging and EXAFS spectroscopy to characterize the insoluble Gd-containing deposits found in the tissues of some NSF patients.  The data from skin autopsy tissue samples (see figure) reveal a strong association of Gd with Ca and P.  EXAFS measurements demonstrate that the Gd is no longer bound the GBCA chelator but instead is coordinated to phosphate.  The implication is the liberation of Gd from the GBCA is responsible for the disease.

Nitrogenase is a key enzyme in the nitrogen cycle which reduces nitrogen gas to ammonia in a complex catalytic cycle involving two component proteins, hydrogen evolution and ATP hydrolysis.  Understanding this nitrogen fixation system has long been recognized as important for agriculture, industry and the environment.  Of particular interest is the chemistry and biosynthesis of the active site: a remarkable MoFeS:homocitrate cluster called FeMo-co (see figure). 

I have published many papers on this system, both when working at the Nitrogen Fixation Laboratory in the UK (later part of the John Innes Centre) and in California with Stephen Cramer (U.C. Davis).  My focus has been to use SF-IR to probe substrate and inhibitor binding to FeMo-co, and EXAFS together with NRVS and DFT calculations to probe conformational change in FeMo-co on substrate binding and reduction.

The figure shows a result published in Angewandte Chemie in collaboration with Lance Seefeldt (Utah State) and Dennis Dean (Virginia Tech).  SF-IR shows how the detailed binding of CO to Azotobacter vinelandii MoFe nitrogenase is changed by modifying the Val-70 residue without introducing charge.  This remarkable steric effect strongly suggests that the CO binding occurs on FeMo-co at Fe atoms on the 2-3-6-7 face.