Richard R. Bott

Protein engineering of α-amylase for low pH performance

Industrial-scale starch liquefaction is currently constrained to operating at pH 6.0 and above, as the enzyme used in the process, Bacillus licheniformis α-amylase, is unstable at lower pH under the conditions used. There is a need develop an enzyme that can operate at lower pH. Recent progress has been made in engineering the B. licheniformis enzyme for improved industrial performance. The availability of crystal structures and subsequent analysis of improved variants, in a structural context, is revealing common factors and a rationale to make further improvements.

Three-Dimensional Structure of an Intact Glycoside Hydrolase Family 15 Glucoamylase from Hypocrea jecorina

The three-dimensional structure of a complete Hypocrea jecorina glucoamylase has been determined at 1.8 A resolution. The presented structure model includes the catalytic and starch binding domains and traces the course of the 37-residue linker segment. While the structures of other fungal and yeast glucoamylase catalytic and starch binding domains have been determined separately, this is the first intact structure that allows visualization of the juxtaposition of the starch binding domain relative to the catalytic domain. The detailed interactions we see between the catalytic and starch binding domains are confirmed in a second independent structure determination of the enzyme in a second crystal form. This second structure model exhibits an identical conformation compared to the first structure model, which suggests that the H. jecorina glucoamylase structure we report is independent of crystal lattice contact restraints and represents the three-dimensional structure found in solution. The proposed starch binding regions for the starch binding domain are aligned with the catalytic domain in the three-dimensional structure in a manner that supports the hypothesis that the starch binding domain serves to target the glucoamylase at sites where the starch granular matrix is disrupted and where the enzyme might most effectively function.

A Novel Combination of Two Classic Catalytic Schemes

The crystal structure of an alkaline Bacillus cellulase catalytic core, from glucoside hydrolase family 5, reveals a novel combination of the catalytic machinery of two classic textbook enzymes. The enzyme has the expected two glutamate residues in close proximity to one another in the active-site that are typical of retaining cellulases. However, the proton donor, glutamate 139 is also unexpectedly a member of a serine-histidine-glutamate catalytic triad, forming a novel combination of catalytic machineries. Structure and sequence analysis of glucoside hydrolase family 5 reveal that the triad is highly conserved, but with variations at the equivalent of the serine position. We speculate that the purpose of this novel catalytic triad is to control the protonation of the acid/base glutamate, facilitating the first step of the catalytic reaction, protonation of the substrate, by the proton donor glutamate. If correct, this will be a novel use for a catalytic triad.

Probing the specificity of the S1 binding site of M222 mutants of subtilisin B. lentus with boronic acid inhibitors

Abstract Specificity differences between the S1-pockets of subtilisin B. lentus (SBL), and its M222C/S mutants have been explored with boronic acid inhibitors. Similar binding trends were noted, with 2,4-dichlorophenylboronic acid being the best overall inhibitor for each enzyme. In addition, a correlation between inhibitor binding and the electrophilicity of boron was found for both the M222C and M222S enzymes. Specificity differences between the S 1-pockets of subtilisin from B. lentus (SBL), and its M222C/S mutants, have been explored with boronic acid inhibitors. Similar binding trends were noted, with 2,4-dichlorophenyl boronic acid being the best overall inhibitor for each enzyme. In addition, a correlation between inhibitor binding and the electrophilicity of boron was found for both M222C and M222S enzymes.

Altering the specificity of subtilisin B. Lentus by combining site-directed mutagenesis and chemical modification

The thiol side chain of the M222C mutant of the subtilisin from Bacillus lentus (SBL) has been chemically modified by methyl-, aminoethyl-, and sulfonatoethylthiosulfonate reagents. Introduction of charged residues into the active site of the enzyme reduced the catalytic efficiency with Suc-AAPF-pNA as the substrate, but resulted in better binding of sterically demanding boronic acid inhibitors.

Altering the proteolytic activity of subtilisin through protein engineering.

The utility of protein engineering in redesigning the structure of a protein to tailor its functional properties has been firmly established. In particular, the Bacillus serine protease subtilisin has proven to be a useful model protein for examining the use of systematic structural modification to incorporate novel functional properties into an enzyme.1.2 The list of properties that have been altered in subtilisin via such modification includes oxidative ~tability,”~ thermal ~tability,~ alkaline pH stability,h stability in organic ~olvent ,~ substrate specificity in aqueous nucleophile specificity,l2.l3 and pH activity profile.14 In addition to demonstrating the versatility of protein engineering, these studies have also provided valuable insight into the expected consequences of protein structure modification. For example, it is now recognized that while amino acid substitutions generally lead to only slight structural perturbations, these minor changes in structure can cause significant changes in protein function. Furthermore, it is apparent from several studies with subtilisin that multiple amino acid substitutions may additively affect a particular functional property. Provided with this extensive data base of structure-function relationships in subtilisin, thc cngineering of subtilisin for altered proteolytic activity is now being attempted. Increasing the proteolytic activity of subtilisin could boost the enzyme’s effectiveness as an additive to household laundry detergents. Subtilisin sold for use in laundry detergents accounts for the largest share of the worldwide industrial enzyme market with sales estimated for 1991 at

A silicone-based controlled-release device for accelerated proteolytic debridement of wounds.

200 million. Furthermore, the utility of subtilisin for peptide synthesis in aqueous systems can be enhanced by decreasing the enzyme’s proteolytic activity. This would alleviate the problem of low synthesis yields obtained due to proteolysis of the peptide product.

Comparative NMR study on the impact of point mutations on protein stability of Pseudomonas mendocina lipase

A new device for rapid enzymatic debridement of cutaneous wounds has been developed using a controlled‐release, silicone‐based, dried emulsion. A dehydrated serine protease of the subtilisin family, previously untested for wound debridement, was incorporated into the emulsion. This device exhibited excellent storage stability. Moisture from the wound triggered an even, reproducible, and complete release of the enzyme within the first 8 hours. The device maintains a moist wound environment that allows the enzyme to achieve nearly complete digestion of the hardened eschar of full‐thickness burns in a porcine model after an exposure period of 24 hours. Debridement was faster than in untreated wounds or wounds treated with a currently available enzyme ointment. Following rapid enzymatic debridement, healing appeared to progress normally, with no histological evidence of damage to adjacent healthy tissue.

The role of high-resolution structural studies in the development of commercial enzymes.

In this work we compare the dynamics and conformational stability of Pseudomonas mendocina lipase enzyme and its F180P/S205G mutant that shows higher activity and stability for use in washing powders. Our NMR analyses indicate virtually identical structures but reveal remarkable differences in local dynamics, with striking correspondence between experimental data (i.e., 15N relaxation and H/D exchange rates) and data from Molecular Dynamics simulations. While overall the cores of both proteins are very rigid on the pico‐ to nanosecond timescale and are largely protected from H/D exchange, the two point mutations stabilize helices α1, α4, and α5 and locally destabilize the H‐bond network of the β‐sheet (β7–β9). In particular, it emerges that helix α5, undergoing some fast destabilizing motions (on the pico‐ to nanosecond timescale) in wild‐type lipase, is substantially rigidified by the mutation of Phe180 for a proline at its N terminus. This observation could be explained by the release of some penalizing strain, as proline does not require any “N‐capping” hydrogen bond acceptor in the i+3 position. The combined experimental and simulated data thus indicate that reduced molecular flexibility of the F180P/S205G mutant lipase underlies its increased stability, and thus reveals a correlation between microscopic dynamics and macroscopic thermodynamic properties. This could contribute to the observed altered enzyme activity, as may be inferred from recent studies linking enzyme kinetics to their local molecular dynamics.

Selective Protein Degradation by Ligand-Targeted Enzymes: Towards the Creation of Catalytic Antagonists

Recent developments in both NMR and X-ray crystallography allow the analysis of commercial enzymes in unprecedented detail. The novel methods provide detailed insights into protein dynamics, establish the existence of special catalytic hydrogen bonds and define the ionization states at the enzyme active site. A more detailed understanding of how the changes in structure are related to altered function should facilitate the design of future commercial enzymes with improved performance for different environmental conditions.