Enzyme-electrode systems are unique because enzymes are biological macromolecules capable of catalyzing redox reactions that can be observed electrochemically. Electron transfer (ET) in the enzyme electrode may be direct (DET) or mediated (MET), where the latter relies on a small redox chemical to shuttle electrons between the enzyme and the electrode surface. Driven by our research interests in DET systems, we look to biology as an inspiration. We were inspired to genetically engineer redox enzymes with gold-binding peptides, as reported by Nam et al. (doi: 10.1126/science.1122716), to produce a wire-ready enzyme capable of self-assembling onto the target surface.
In our previous studies (https://doi.org/10.1021/acsami.8b08876 and doi: 10.1016/j.bios.2018.10.013), we showed that by fusing gold-binding peptide (gbp) to FAD-dependent glucose dehydrogenase, the fusion enzymes showed gold-binding activity and improved DET properties, which were measured by monitoring the catalytic anodic current via cyclic voltammetry, in the presence of glucose as a substrate. It was intriguing how much of an impact a tiny gbp (1.3 kDa) could have. Still, we have many questions, including what the gbp-fused enzyme would look like, and whether this approach will work on other enzymes. The enzyme orientation and electron transfer distance are interrelated, which means that if we can orient the enzyme in an ideal manner that would position the final electron donor within 14 Å from the electrode surface, we can theoretically observe the DET current.
In further discussions, it was noted how the L-subunit of carbon monoxide dehydrogenase (CODH-L) resembles a “dome” shape, with both termini at each end. This description led us to the idea that having gbp at both termini may stabilize the enzyme orientation on the electrode surface. To test this hypothesis, we constructed three fusion enzymes with gbp at either the C- or N-terminus, and gbp at both termini, with native CODH-L as a control.
We obtained predicted structures of the fusion enzymes that were previously unknown, due to Zhang's lab open server, I-TASSER (https://zhanggroup.org/I-TASSER/). Next, we wanted to determine how the ET distance changed depending on enzyme orientation. From the enzyme structure, we estimated the ET distance, that is, the distance between the redox site and the gbp-fusion site, under the assumption that the fusion site is the anchoring point of the enzymes on the electrode surface. Because we know the enzyme size, we employed atomic force microscopy (AFM) to see how the enzymes are oriented on an atomically flat gold surface. We observed a wider range of enzyme film heights when gbp was fused at either terminus, but a more consistent enzyme height for gbp at both termini. The fusion enzymes showed significant gold-binding activity, as evidenced by quartz crystal microbalance (QCM) and surface plasmon resonance (SPR).
The high gold binding affinity allows for soft and fast immobilization, where any electrode treatment or modification is eliminated. By immersing the electrode in an enzyme solution, enzyme molecules can self-assemble on the target surface and allow electrochemical measurements. From the CV profile, no catalytic CO oxidation current was observed in any of the three cases of native CODH-L and CODH-L with gbp at the C- or N-terminus. In the case of gbp at both termini, a clear catalytic CO oxidation current starting from −0.62 V vs Ag/Ag+ was shown. Next, we used a small redox mediator, methylene blue, to determine whether the difference in the CV profile was due solely to the ET distance. The mediated current confirmed that the enzymes were successfully immobilized and catalytically active in the case of gbp at either terminus, but they were not within the DET distance.
These findings motivated us to determine whether this approach could also work for other redox enzymes. For this purpose, we analyzed some commonly used redox enzymes, including glucose oxidase, laccase, and bilirubin oxidase. This approach of gbp fusion at both termini works well for CODH-L because of several specific features, that is, C- and N-termini exposed to the solvent have no contact with the catalytic domain, and the enzyme cofactor–surface distance is <14 Å. Thus, our approach would be applicable if redox enzymes could meet these specific requirements.
In this study, we have shown that the orientation of CODH-L controlled by site-specific immobilization enables direct interfacial electron transfer. However, in the enzyme bioelectrocatalytic system, we are still in need of direct visualization of the enzyme on the electrode because seeing is believing. There are still many questions and associated curiosity that we hope to address in the future.