Research Areas

4. Investigation and Enhancement of Protein Stability

Effects of buffer salts on protein stability

Dissolved salts are known to affect properties of proteins in solution including solubility and melting temperature, and the effects of dissolved salts can be ranked qualitatively by the Hofmeister series.  We seek a quantitative model to predict the effects of salts in the Hofmeister series on the deactivation kinetics of enzymes.  Such a model would allow for better prediction of useful biocatalyst lifetimes or improved estimation of protein-based pharmaceutical shelf life.  Here we consider a number of salt properties that are proposed indicators of Hofmeister effects in the literature as a means for predicting salt effects on the deactivation of horse liver alcohol dehydrogenase (HL-ADH), a-chymotrypsin, and monomeric red fluorescent protein (mRFP).  We find that surface tension increments are not accurate predictors of salt effects, but find a common trend between observed deactivation constants and B-viscosity coefficients of the Jones-Dole equation, which are indicative of ion hydration.  This trend suggests that deactivation constants (log kd,obs) vary linearly with chaotropic B-viscosity coefficients but are relatively unchanged in kosmotropic solutions.  The invariance with kosmotropic B-viscosity coefficients suggests the existence of a minimum deactivation constant for proteins.  Differential scanning calorimetry is used to measure protein melting temperatures and thermodynamic parameters which are used to calculate the intrinsic irreversible deactivation constant.  We find that either the protein unfolding rate or the rate of intrinsic irreversible deactivation can control the observed deactivation rates.

Determination long-term biocatalyst process stability through short-term experiments

The deactivation of protein biocatalysts even at relatively low temperatures is one of the principal drawbacks to their use. We have derived an equation for both time- and temperature-dependent activity of the biocatalyst based on known concepts such as transition state theory and the Lumry-Eyring model. We then derived an analytical solution for the total turnover number (ttn), under isothermal operation, as a function of the catalytic constant kcat, the unfolding equilibrium constant K, and the intrinsic first-order deactivation rate constant(s) kd,i. Employing an immobilized glucose isomerase biocatalyst in a CSTR and utilizing a linear temperature ramp beyond the Tm of the enzyme, we demonstrate an accelerated method for extracting the thermodynamic and kinetic constants describing the biocatalyst system. In addition, we demonstrate that the predicted biocatalyst behavior at different temperatures and reaction times is consistent with the experimental observations.