Use of Chemically Crosslinked Polymer Nanofibers for Hyperthermophilic Enzyme Immobilization

Although enzymes are highly efficient and selective biocatalysts, lack of stability often limits their practical application. Thus, enzymes from hyperthermophilic microorganisms may be useful due to their intrinsic thermostability. For application of biocatalysts, immobilization is often desirable since it improves enzyme stability, avoids product contamination, and eases recovery which facilitates reuse. However, immobilization often leads to reduced catalytic activity. The performance of the immobilized enzyme is significantly affected by the structure of the support material and nanofibers are considered promising materials due to their high specific surface area. Therefore, the aim of this work is to evaluate the use of nanofibers for enzyme immobilization.;We immobilized a model hyperthermophilic enzyme, α-galactosidase from Thermotoga Maritima, by encapsulating it within electrospun nanofibers using poly(vinyl alcohol) (PVA) chemically crosslinked with glutaraldehyde. This approach requires: production of uniform enzyme-loaded nanofibers, chemical crosslinking methods that maintain the high specific surface area, and resulting fibers that are insoluble in water at elevated temperatures to avoid enzyme leaching.;We first explored methods of producing electrospun and chemically crosslinked PVA nanofibers including a single-step reactive electrospinning method. In this method, we add glutaraldehyde to PVA so crosslinking occurs during electrospinning and the resulting as-spun fibers are chemically crosslinked. Using this method, the resulting crosslinked fibers maintained their structure when soaked in water, suggesting that this could be a promising method to immobilize enzymes in a single step.;We also examined how the addition of a model globular protein, bovine serum albumin (BSA), comparable to α-galactosidase in molecular weight affected electrospinning and production of uniform fibers. We found that the presence (up to 50 wt.%) of BSA, did not appear to affect the polymer entanglement required to electrospin uniform fibers. Modulation of the protein distribution within the fiber was achieved by adjusting the pH of the electrospinning solution. When the pH was adjusted to the isoelectric point of the protein, a coaxial distribution with BSA in the core was obtained. For pHs above and below the isoelectric point of the protein, the amount of protein on the surface of the fiber was enriched.;We developed a two-step immobilization method in which, we electrospun a PVA/enzyme blend to produce uniform enzyme-loaded nanofibers, then chemically crosslinked the fibers with glutaraldehyde in acetone. The resulting fibers maintained their structure when soaked in water at 75°C and no enzyme leaching was observed indicating that chemical crosslinking effectively immobilized the enzyme within the fibers. The immobilized enzyme showed improved thermal stability but lower apparent catalytic activity. The loss of activity due to deactivation was explored by examining the effect of the reagents used in the crosslinking reaction on the free enzyme. By varying the enzyme-loading and mat thickness, intra-fiber and inter-fiber mass transfer limitations were evaluated, respectively.;Finally, we developed a single-step immobilization method using reactive electrospinning. Using this process, we produced uniform, enzyme-loaded, water-insoluble fibers from which no enzyme leaching was observed. Single-step immobilization resulted in significantly higher apparent activities when compared to two-step methods (e.g. 4 orders of magnitude higher than acetone-based crosslinking). We explored possible reasons (enzyme deactivation and mass transfer) for the lower activity with the aim of elucidating the key differences in crosslinking methods. Enzyme deactivation and inter-fiber diffusion appeared to affect the performance of the immobilized enzyme using both crosslinking methods. However, intra-fiber diffusion limitations were more significant when using acetone-based crosslinking than in situ crosslinking. We attribute the differences in apparent activity to differences in the network structure resulting from the various crosslinking methods. Differences in effective diffusivities within the various network structures could account for the differences in the catalytic performance of the immobilized enzyme.