Improvement of the optimum temperature of lipase activity for Rhizopus niveus by random mutagenesis and its structural interpretation.

Random mutagenesis was used to improve the optimum temperature for Rhizopus niveus lipase (RNL) activity. The lipase gene was mutated using the error-prone PCR technique. One desirable mutant was isolated, and three amino acids were substituted in this mutant (P18H, A36T and E218V). The wild-type and this randomly mutated lipase were both purified and characterized. The specific activity of the mutant lipase was 80% that of the wild-type. The optimum temperature of the mutant lipase was higher by 15 degrees C than that of the wild-type. To confirm which substitution contributed to enhancing the optimum temperature for enzymic activity, two chimeric lipases from the wild-type and randomly mutated gene were constructed: chimeric lipase 1 (CL-1; P18H and A36T) and chimeric lipase 2 (CL-2; E218V). Each of the chimeric enzymes was purified, and the optimum temperature for lipase activity was measured. CL-1 had a similar optimum temperature to that of the wild-type, and CL-2 had a higher temperature like the randomly mutated lipase. The mutational effect is interpreted in terms of a three-dimensional structure for the wild-type lipase.

High-level expression of Rhizopus niveus lipase in the yeast Saccharomyces cerevisiae and structural properties of the expressed enzyme.

Rhizopus niveus lipase (RNL) has a unique structure consisting of two noncovalently bound polypeptides (A-chain and B-chain). To improve this enzyme's properties by protein engineering, we have developed a new expression system for the production of recombinant lipase in the yeast Saccharomyces cerevisiae. For the present study, we developed a more efficient expression system using the strain ND-12B and the multicopy-type plasmid pJDB219. We purified two types of recombinant lipases, each to a single peak by gel-filtration HPLC, although they were found to be heterogeneous by SDS-PAGE. Analysis of reversed-phase HPLC, N-terminal amino acid sequence, and sugar content showed that the difference between the two types of lipases was due mainly to their sugar content (high or low mannose type). Moreover, there were two species within each type of lipase. One kind was processed to the A-chain and B-chain as in the native lipase, while the other remained unprocessed. Although these yeast-purified lipases contained several posttranslational modifications and different glycosylations, their secondary structures were the same as those of the native lipase as measured by circular dichroism spectra and determination of disulfide bonding. This suggests that protein folding of the recombinant lipase occurred correctly in yeast.