The thermostable alcohol dehydrogenase D (AdhD) from Pyrococcus furiosus is an NAD(H)-dependent enzyme that oxidizes a range of sugars and alcohols but is most active with secondary alcohols. It is monomeric, and a member of the aldo-keto reductase (AKR) superfamily of proteins. We have set out to engineer just about every aspect of this enzyme to improve its performance and develop it for use in several new applications.

First, we have added alpha-helical appendages to both termini of the enzyme to create a bifunctional chimeric protein that can self-assemble into hydrogels. The impact of these modifications on the enzymatic activity is negligible. And the resultant hydrogel is robust and active.[1]

Next we rationally altered the cofactor specificity of the enzyme via two site-directed mutations (K249G/H255R) in the cofactor-binding pocket. This significantly increased activity with NAD(H) and improved the activity with NADP(H) by an order of magnitude. A pre-steady state kinetic analysis was also performed and we observed a wide array of conformational behavior indicating that the previously established kinetic mechanism in the AKRs may need to be reevaluated.[2] We have recently expanded this work through the identification of an amino acid location distal to the cofactor binding pocket that can also be mutated to alter cofactor specificity by altering transition state binding energy as opposed to cofactor ground state binding energy.[3]

We demonstrated the utility of an engineered AdhD for use in enzymatic biofuel cells. The performance of systems employing nicotinamide cofactor-dependent enzymes can be limited by the diffusion of the cofactor between the enzyme and the electrode. The cofactor-specificity mutant (K249G/H255R) fortuitously exhibits activity with nicotinamide mononucleotide (NMN(H)) but with activity that is several orders of magnitude lower that what is observed with the native cofactor. When the enzyme was incorporated into an enzymatic biofuel cell, similar power densities were obtained, and the maximum current density for the NMN(H) electrode was higher than that of the NAD(H). Thus, although there was a significant decrease in kinetic activity of the enzyme, the improvement in the diffusion of the truncated cofactor led to improved performance of the enzymatic biofuel cell.[4]

We also modulated the substrate specificity of the enzyme by making chimeric enzymes between AdhD and the mesostable NADP(H)-dependent human aldose reductase (hAR). Active-site loops were swapped to transfer hAR activity to the thermostable AdhD scaffold, including the reversed cofactor-specificity.[5]

One of the loops that was altered did not ablate AdhD activity, and so we chose this site for the introduction of an RTX domain from the adenylate cyclase protein from Bordatella pertussis. This peptide is disordered and forms the beta roll secondary structure motif upon calcium addition. The insertion of the RTX domain into AdhD reduced overall activity. However, the conformational change of the RTX domain affected NAD+-dependent activity more than NADP+-dependent activity such that calcium could be used as a “rheostat switch” to control cofactor selectivity in the enzyme.[6]

Finally, there is a need to develop sensors for explosive molecules such as TNT and RDX and antibody-based systems suffer from high cost, and poor stability. We have used ribosome display to screen randomized AdhD libraries over immobilized RDX. Mutants with affinity for RDX have been identified and characterized by isothermal titration calorimetry (ITC). Thus we have converted an enzyme into a thermostable antibody replacement.[7]


1. Wheeldon IR, Campbell E, & Banta S (2009) A chimeric fusion protein engineered with disparate functionalities-enzymatic activity and self-assembly. J Mol Biol 392(1):129-142.

2. Campbell E, Wheeldon IR, & Banta S (2010) Broadening the cofactor specificity of a thermostable alcohol dehydrogenase using rational protein design introduces novel kinetic transient behavior. Biotechnol Bioeng 107(5):763-774.

3. Solanki K, Abdallah W, & Banta S (2017) Engineering the cofactor specificity of an alcohol dehydrogenase via single mutations or insertions distal to the 2'-phosphate group of NADP(H). Protein Eng Des Sel 30(5):373-380.

4. Campbell E, Meredith M, Minteer SD, & Banta S (2012) Enzymatic biofuel cells utilizing a biomimetic cofactor. Chemical communications 48(13):1898-1900.

5. Campbell E, Chuang S, & Banta S (2013) Modular exchange of substrate-binding loops alters both substrate and cofactor specificity in a member of the aldo-keto reductase superfamily. Protein engineering, design & selection : PEDS 26(3):181-186.

6. Abdallah W, Solanki K, & Banta S (2017) Insertion of a calcium-responsive beta roll domain into a thermostable alcohol dehydrogenase enables tunable control over cofactor selectivity. (Submitted).

7. Bulutoglu B, Haghpanah J, Campbell E, & Banta S (2017) Directed evolution of a thermostable alcohol dehydrogenase to bind RDX for biosensor applications. (Submitted).