Johann Partridge

Biocatalyst behaviour in low-water systems

An improving understanding of the parameters that affect biocatalyst activity, specificity and stability in low-water non-aqueous media make reliable predictions about the behaviour of such systems increasingly feasible. Here, we discuss some of the key factors, such as control of water activity, and the effects of solvent on K m and protein-ionization state, that must be addressed in order to obtain predictable results.

Activity and mobility of subtilisin in low water organic media: hydration is more important than solvent dielectric.

The relationship between hydration, catalytic activity and protein dynamics was investigated for subtilisin Carlsberg in organic solvents with low water content. The organic media were cyclohexane, dichloromethane or acetonitrile, with controlled thermodynamic water activity (aw). Catalytic rate profiles showed the same dependence on aw for the three different solvents. The structural mobility of the enzyme in air and organic media was probed by proton solid-state NMR relaxation measurements. Both spin-lattice relaxation time (T1 ) and line width at half height (apparent spin-spin relaxation time (T2)) were determined for protein which was exchanged and hydrated with D2O. We found NMR relaxation was much more dependent on aw than medium identity (despite very different dielectrics) showing that enzyme hydration is the primary determinant of mobility. Results suggest that initial hydration up to aw 0.22 causes rigidification of part of the protein structure. As aw is increased further, enzyme mobility is found to increase. Above aw 0.44, a large increase in the proportion of more mobile protons coincides with a steep rise in catalytic activity for the enzyme in each of the solvents studied.

Practical route to high activity enzyme preparations for synthesis in organic media

A single pot method to rapidly prepare immobilised subtilisin Carlsberg and α-chymotrypsin gives 1000-fold greater catalytic activities in polar organic solvents than freeze-dried powders.

Selection of salt hydrate pairs for use in water control in enzyme catalysis in organic solvents

The water activities (a(w)) of 13 salt hydrate pairs were determined from vapor pressure measurements; a(w) values for a subset were also estimated from a study of water transfer to isopropylether. The application of salt hydrates as water buffers was investigated in two models: (i) effect of hydration on the initial rate of subtilisincatalyzed transesterification of the nitrophenol ester of CBZ-alanine with butanol; and (ii) effect of hydrates on the equilibrium concentrations of reactants in the esterification of dodecanol and decanoic acid, catalyzed by lipase. Transfer of ions from salt to enzyme particles was also demonstrated. The implications of the results for the successful use of salt hydrates as water buffers are discussed. (c) 1997 John Wiley & Sons, Inc. Biotechnol Bioeng 55: 367-374, 1997.

α-Chymotrypsin stability in aqueous-acetonitrile mixtures: is the native enzyme thermodynamically or kinetically stable under low water conditions?

Abstract Like many proteins, α-chymotrypsin is denatured in 50% volume aqueous-acetonitrile mixtures. However, it also shows high catalytic activity in 70% or more acetonitrile. Good activity in two different aqueous organic composition ranges has been described for several other enzymes. The stability of the native protein under low water conditions is generally believed to be a kinetic phenomenon, though there are also arguments for thermodynamic stability. We have distinguished between these possibilities by studying the effects of changing medium composition at different times. In preliminary experiments, we found catalytic activity could be recovered by adding neat acetonitrile to chymotrypsin in a 50% mixture, suggesting that the enzyme could renature under these conditions. However, in the 50% mixture, the true initial activity at 30°C is not zero, as the literature suggests. Instead, there is an initial burst of product formation over a few minutes, after which the enzyme becomes inactivated. By pre-incubating a 50% aqueous-acetonitrile mixture at 30°C prior to enzyme addition, the product burst could be eliminated. Activity could not then be recovered by slow addition of acetonitrile to the denatured enzyme. In contrast, it was possible to renature by dilution with aqueous buffer so that regeneration of catalytic activity was achieved. Thus, the good practical performance at high acetonitrile concentrations almost certainly results from a high kinetic barrier towards denaturation. The kinetics of enzyme denaturation in 50% and 70% acetonitrile were also investigated both at 30 and 20°C. Loss of catalytic activity was faster at higher temperature and at lower acetonitrile concentrations.

Solid-state proton/sodium buffers: “chemical pH stats” for biocatalysts in organic solvents

The useful application of enzymes in organic synthesis requires reliable and straightforward methods for maximising efficiency and selectivity. Here we describe how to control the protonation state of enzymes using a new class of solid-state buffers, applied for the first time in polar organic solvents. Remarkably these insoluble buffers are able to rapidly exchange H+ and Na+ ions with both the protein and reaction mixture as demonstrated here using propanol rinsed enzyme preparations (PREPs) of subtilisin Carlsberg and chymotrypsin. The buffers tested were generally mixtures of a zwitterionic biological buffer and its Na+ salt with each buffer pair setting a characteristic fixed ratio of H+ activity to Na+ activity (aH+/aNa+) within the system. Dependent upon the solid buffer pair selected a wide range of different enzymatic activities could be observed. The variation in rate showed a fairly good but not exact correlation with the aqueous pKa of the buffers, indicating that crystal lattice energies have less affect on acid–base strength than might be expected. The solid-state buffers were able to prevent detrimental changes to enzyme activity caused by the presence or build up of acids or bases in the organic reaction mixture (often undetected). They could also be used advantageously to tune the enzyme protonation state in solvent if a previous aqueous preparation step needs to be carried out at a pH not optimal for catalysis. Such buffering systems are expected to find wide-spread use as ‘chemical pH stats’ for reactions in non-aqueous media.

Supercritical fluids are superior media for catalysis by cross- linked enzyme microcrystals of subtilisin Carlsberg

We report on the performance of cross‐linked enzyme microcrystals (CLECs) of subtilisin Carlsberg in supercritical fluids (SC‐fluids). The catalytic activity of CLECs in SC‐ethane was found to be 2‐ to 10‐fold greater than in hexane under the same conditions, using CLECs dried by propanol washing. Air‐dried CLECs and lyophilized powders showed much lower activities, reflecting the same hydration hysteresis effects as in organic solvents. Reaction rates were much lower in SC‐CO2, especially at higher water activity, probably as a result of acid‐base effects of carbonic acid on the enzyme.

Tuning of the product spectrum of vanillyl-alcohol oxidase by medium engineering

The flavoenzyme vanillyl‐alcohol oxidase (VAO) catalyzes the conversion of 4‐alkylphenols through the initial formation of p‐quinone methide intermediates. These electrophilic species are stereospecifically attacked by water to yield (R)‐1‐(4′‐hydroxyphenyl)alcohols or rearranged in a competing reaction to 1‐(4′‐hydroxyphenyl)alkenes. Here, we show that the product spectrum of VAO can be controlled by medium engineering. When the enzymatic conversion of 4‐propylphenol was performed in organic solvent, the concentration of the alcohol decreased and the concentration of the cis‐alkene, but not the trans‐alkene, increased. This change in selectivity occurred in both toluene and acetonitrile and was dependent on the water activity of the reaction medium. A similar shift in alcohol/cis‐alkene product ratio was observed when the VAO‐mediated conversion of 4‐propylphenol was performed in the presence of monovalent anions that bind specifically near the enzyme active site.

Peptide synthesis in dimethylformamide-based organic media catalysed by non-covalent chymotrypsin-polyelectrolyte complexes

Chymotrypsin-catalysed kinetically-controlled peptide synthesis was studied in mixtures of dimethylformamide (DMF) and acetonitrile containing 6% water. The free enzyme showed no detectable activity with 60% or more DMF. However, by pre-forming a complex with poly-acrylate (optimum 50 carboxyl groups to 1 enzyme molecule), activity and stability were good at 60% DMF, and reasonable at 70% DMF. In 60% DMF, the peptide Ac-Tyr-Lys-NH2 could be made in 95% yield using this catalyst.

Non-covalent enzyme-polyelectrolyte complexes as self-buffered catalysts for use in low-water organic media

When free chymotrypsin is used to catalyse hydrolysis of N-acetyl tyrosine ethyl ester in 90% acetonitrile, the reaction rate soon falls because of the accumulation of the acidic product. If the enzyme is used in the form of a suspended complex with polyacrylic acid, the polyelectrolyte acts as an acid-base buffer to permit extended reaction.