The stabilizing effects of immobilization in D-amino acid oxidase from Trigonopsis variabilis

Abstract

Background: Immobilization of Trigonopsis variabilis D-amino acid oxidase (TvDAO) on solid support is the key to a reasonably stable performance of this enzyme in the industrial process for the conversion of cephalosporin C as well as in other biocatalytic applications.

Results: To provide a mechanistic basis for the stabilization of the carrier-bound oxidase we analyzed the stabilizing effects of immobilization in TvDAO exposed to the stress of elevated temperature and operational conditions. Two different strategies of immobilization were used: multi-point covalent binding to epoxy-activated Sepabeads EC-EP; and non-covalent oriented immobilization of the enzyme through affinity of its N-terminal Strep-tag to Strep-Tactin coated on insoluble particles. At 50 degrees C, the oriented immobilizate was not stabilized as compared to the free enzyme. The structure of TvDAO was stabilized via covalent attachment to Sepabeads EC-EP but concomitantly, binding of the FAD cofactor was weakened. FAD release from the enzyme into solution markedly reduced the positive effect of immobilization on the overall stability of TvDAO. Under conditions of substrate conversion in a bubble-aerated stirred tank reactor, both immobilization techniques as well as the addition of the surfactant Pluronic F-68 stabilized TvDAO by protecting the enzyme from the deleterious effect of gas-liquid interfaces. Immobilization of TvDAO on Sepabeads EC-EP however stabilized the enzyme beyond this effect and led to a biocatalyst that could be re-used in multiple cycles of substrate conversion.

Conclusion: Multi-point covalent attachment of TvDAO on an isoluble porous carrier provides stabilization against the denaturing effects of high temperature and exposure to a gas-liquid interface. Improvement of binding of the FAD cofactor, probably by using methods of protein engineering, would further enhance the stability of the immobilized enzyme.

Figure 1

Effect of enzyme loading on the activity of TvDAO immobilized on Sepabeads EC-EP. Activity was assayed using measurement of oxygen consumption during oxidative deamination of D-methionine. The solid line shows the trend of the data. Inset: change of the apparent enzyme activity as a function of the stirrer speed used in the O₂-dependent assay; the solid line is a straight line fit to the data. Error bars show S.D. from three independent assays.

Figure 2

Thermally induced inactivation of free and immobilized preparations of TvDAO. Time courses of inactivation of TvDAO immobilized on Strep-Tactin MacroPrep material (open squares) and Sepabeads EC-EP (full circles) are compared to the time course of inactivation of the soluble enzyme (full triangles). The inactivation time course of the Sepabeads immobilizate in the presence of 100 μM FAD is shown as open circles. The enzymes were incubated at 50°C on an Eppendorf Thermomixer using orbital shaking at 500 rpm; the protein concentration was 1 μM in 10 mM Tris-HCl buffer, pH 7.5 in all cases. Solid lines represent least squares fits of a single exponential decay function to the data. Note that the suspensions of immobilized biocatalysts were handled like liquids, which explains the observed scattering of the data for the enzyme immobilized on Sepabeads EC-EP.

Figure 3

Release of FAD cofactor during thermally induced inactivation of different preparations of TvDAO. The concentration of free FAD was measured in samples taken during the course of the inactivation experiment at 50°C. After removal of the enzymes by sedimentation (immobilized preparations) or ultrafiltration (soluble TvDAO) the content of FAD in solution was determined fluorometrically. Symbols are used as in Figure 2; exponential rise functions were fitted to the experimental data (solid lines). The inset compares relative inactivation (open symbols) and relative FAD release (full symbols) for TvDAO immobilized on Sepabeads EC-EP.

Figure 4

Interfacial inactivation of free and immobilized TvDAO preparations. Loss of activity induced by bubble aeration in a miniaturized stirred reactor is shown for the different enzyme preparations and compared to a non-aerated control of soluble TvDAO (full down triangles). Experiments were performed at 30°C with stirring at 300 rpm and an aeration rate of 15 L/h; protein concentration was 1 μM in Tris-buffer. Symbols: full triangles, soluble enzyme; open triangles, soluble enzyme in presence of 1% Pluronic F-68; open squares, TvDAO immobilized on Strep-Tactin MacroPrep beads; full circles, TvDAO on Sepabeads EC-EP.

Figure 5

Stability of TvDAO under conditions of substrate conversion. The operational stability under bubble aeration conditions is compared for soluble enzyme in absence (1; and inset) and presence of 1% Pluronic F-68 (2) and TvDAO immobilized on Strep-Tactin MacroPrep particles (3). The initial concentration of D-methionine was 20 mM and it was added in solid form to the reactor in regular intervals to prevent its depletion. A total volume of 20 mL was used; all other conditions were as described in Figure 4. The black lines for series 2 and 3 show least squares fits of a double exponential decay function to the actual stability curves which are shown in grey.

Figure 6

Stability of TvDAO immobilized on Sepabeads EC-EP in repeated batches of D-Met conversion. Oxidative deamination of the substrate (20 mM initial concentration in 20 mL Tris-buffer) was done in a stirred reactor (300 rpm) at 30°C with aeration at 15 L/h. Panel A shows a typical time-course of the concentration of dissolved oxygen during one batch of D-Met conversion. The inset illustrates the determination of enzymatic activity at the beginning of each batch from initial rates of oxygen consumption (dotted line) after addition of the substrate. Aeration was turned off for the measurement and turned on again when the level of O₂ had dropped below 100 μM, which is marked by an increase of the signal. Panel B shows the operational stability of the immobilized enzyme and compares the three methods to calculate the residual activity in every batch: initial rate measurements (triangles); the average oxygen concentration as long as D-Met is not limiting (diamonds); and the reciprocal of the total batch time (open squares). The solid line is a straight-line fit to the averaged data from all three methods.