Anchimeric Assistance in the Intramolecular Reaction of Glucose-dehydrogenase-Polyethylene Glycol NAD Conjugate*

Polyethylene glycol-bound derivatives of NAD(P) (PEG-NAD(P)) are water-soluble macromolecular coenzymes used in continuous enzyme reactors. These NAD(P) derivatives have good coenzyme activity for many dehydrogenases, but some enzymes such as glu- cose dehydrogenase (EC 1.1.1.47) show very low activity with these derivatives (less than 0.1% of that for native NAD(P)). In this work, we prepared a covalently linked glucose-dehydrogenase-polyethylene glycol-NAD conjugate (GlcDH-PEG-NAD) and found that the conjugate shows a much higher reaction rate than that of the native enzyme plus PEG-NAD: the ratio of the reaction rates of GlcDH-PEG-NAD and the native enzyme plus PEG-NAD is calculated to be 10,000-fold at the concentrations of the enzyme subunit and NAD moiety of 0.3 1 and 0.65 PM, respectively; the rate of the conjugate is even higher than that of the native enzyme plus native NAD. This rate acceleration is due to the increase in the effective concentration of NAD moiety ("anchimeric assistance") and demon- strates the potential of covalent linking for improving the interaction between an enzyme and a coenzyme derivative. Enzyme Assay-Enzyme reactions were measured at 30 “C with a Hitachi 220A spectrophotometer with a thermostatted cell compart-ment and a magnetic stirrer. The activity in the presence of exogenous NAD was assayed in 75 mM Tris/HC1 buffer, pH 8.0, containing 0.1 M D-gluCOSe and 2.0 mM NAD, and the reactions were recorded as the increase in absorbance at 340 nm. The activity was also assayed by the tetrazolium salt method (12) in 20 mM Tris/HCl buffer, pH 8.0, containing 1 M NaC1, 0.1 M D-glucose, 20 mM PES, and 1.7 mM MTT, and the reactions were recorded as the increase in absorbance at 570 nm due to formazan formation from MTT. The concentration of the formazan was measured using a molar absorption coefficient of 13,000 M” cm” at 570 nm (13).

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Glucose dehydrogenase (EC 1.1.1.47) from Bacillus megaterium (7) is a tetrameric enzyme that catalyzes the oxidation of P-D-glucose to ~1-glucono-1,5-lactone, which is spontaneously hydrolyzed to gluconic acid, using NAD or NADP as a coenzyme. As the equilibrium of the overall reaction lies much in favor of NAD(P)H formation, this enzyme is useful as an NAD(P)H regenerator in enzyme reactors (8). Glucose dehydrogenase, however, shows very low activity for PEG-NAD(P); the reduction rate of PEG-NAD is only 0.08% of that of NAD (5). In the present work, we demonstrate that the activity of this enzyme for PEG-NAD is revived by preparing a covalently linked glucose-dehydrogenase-PEG-NAD conjugate.
Protein ana' Nucleotide Concentrations-The concentrations of NAD(H), their derivatives, and glucose dehydrogenase were measured using the following molar absorption coefficients: NAD, 18,000 M-' cm" at 260 nm; PEG-NAD, 25,000 M" cm" at 266 nm (9); NADH and PEG-NADH, 6,300 M" cm" at 340 nm (10); and the subunit of glucose dehydrogenase, 35,000 M" cm" at 280 nm. The value for the subunit was calculated from the results of absorption measurements, the protein content obtained from amino acid analysis, and the relative molecular mass of 30,000 for the subunit (see "Results" and Ref. 7. The concentration of PEG-NAD and the enzyme subunit in GlcDH-PEG-NAD were calculated as described by Minsson et al. (11) from the ultraviolet absorption spectrum of the conjugate on the basis of the following molar absorption coefficients: 35,000 M" cm" at 280 nm and 24,000 M" cm" at 266 nm for the enzyme subunit; 14,000 M" cm" at 280 nm and 25,000 M" cm" at 266 nm for PEG-NAD. The fraction of reducible NAD in the preparation of GlcDH-PEG-NAD was measured at 30 "C in 20 mM phosphate buffer, pH 6.5 (1 ml), containing about 0.3 M NaCl and GlcDH-PEG-NAD measured by the increase in absorbance at 340 nm due to the addition Preparation of GlcDH-PEG-NAD-PEG-NAD (15.0 rmol) was dissolved in CH2C12 (2.0 ml) containing 54 mM triethylamine, and 3,3'-(1,6-dioxo-1,6-hexanediyl)bis-2-thiazolidinethione (50 pmol, a bifunctional reagent) dissolved in CH2C1, (1.0 ml) was added to the solution. After 1 h at room temperature, CH,Cl, was removed by evaporation. The residue was dissolved in phosphate buffer, pH 6.5, containing 0.1 M NaCl (1.0 ml) and the solution was filtered to remove the remaining precipitate. The filtrate was mixed with glucose dehydrogenase (0.37 rmol as subunit) dissolved in the same buffer containing 0.1 M NaCl described above (1.0 ml). After 10 h at room temperature, the reaction mixture was applied to a DEAE-Sephadex A-50 column (2 X 15 cm) equilibrated with 20 mM phosphate buffer, pH 6.5, containing 0.1 M NaCI. The column was washed with this buffer, and GlcDH-PEG-NAD was eluted with an NaCl gradient of 0.1-1.0 M (total 240 ml). Fractions 67-76 (3.5 ml each) were combined, concentrated by ultrafiltration, and used as GlcDH-PEG-NAD.

Glucose-dehydrogen
Enzyme Assay-Enzyme reactions were measured at 30 "C with a Hitachi 220A spectrophotometer with a thermostatted cell compartment and a magnetic stirrer. The activity in the presence of exogenous NAD was assayed in 75 mM Tris/HC1 buffer, pH 8.0, containing 0.1 M D-gluCOSe and 2.0 m M NAD, and the reactions were recorded as the increase in absorbance at 340 nm. The activity was also assayed by the tetrazolium salt method (12) in 20 mM Tris/HCl buffer, pH 8.0, containing 1 M NaC1, 0.1 M D-glucose, 20 mM PES, and 1.7 mM MTT, and the reactions were recorded as the increase in absorbance at 570 nm due to formazan formation from MTT. The concentration of the formazan was measured using a molar absorption coefficient of 13,000 M" cm" at 570 nm (13). RESULTS

AND DISCUSSION
PEG-NAD was covalently linked to glucose dehydrogenase ( Fig. 1) by the procedure described for the preparation of malate-dehydrogenase-PEG-NAD conjugate (9). Fig. 2 shows the DEAE-Sephadex column chromatography after the reaction of glucose dehydrogenase with the activated PEG-NAD. The first and the second peaks had no enzyme activity, and their ultraviolet spectra showed that unbound NAD derivative and 2-thiazolidinethione produced by the hydrolysis of the activated PEG-NAD were included. GlcDH-PEG-NAD was obtained from the third peak. Under the same experimental conditions, native glucose dehydrogenase was eluted a t a higher ionic strength (32 ms).
GlcDH-PEG-NAD, thus prepared, has the following characteristics. The average number of NAD moieties bound per molecule of enzyme subunit (NAD content (9)) is 2.1. High performance gel filtration chromatography (TSK-Gel G3000SW equilibrated with 50 mM phospahte buffer, pH 6.5, containing 10% glycerol) shows a single peak with M, of 205,000, which is larger than the value of the native enzyme of 140,000. The value of the native enzyme is a little larger  ,ase-NAD Conjugate 16793 than the value of 118,000 obtained in 50 mM phosphate buffer, pH 6.5, containing 0.1 M NaC1; the latter is in good agreement with those reported in Ref. 7. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis as in Ref. 9 shows at least four bands with the apparent M, of 30,000, 39,100, 46,400 and 54,000; these bands seem to correspond to the subunits with 0, 1, 2, and 3 molecules of PEG-NAD, respectively. About 70% of the bound NAD moieties of GlcDH-PEG-NAD is reduced by its enzyme moiety in the presence of D-glucose. The specific activity of GlcDH-PEG-NAD in the presence of exogenous NAD is about 92% of that of the native enzyme.
These results indicate that a tetramer of GlcDH-PEG-NAD has about six molecules of active and covalently linked NAD moieties, and that the covalent linking causes only a slight loss in the activity of its enzyme moiety; similar results were obtained for malate-dehydrogenase-PEG-NAD conjugate (9). The K,,, value of GlcDH-PEG-NAD for exogenous NAD is 0.8 mM and the value is similar to that of the native enzyme (0.9 mM). These results indicate that the NAD(H) moiety of the conjugate does not compete with exogenous NAD for the coenzyme-binding site of the enzyme moiety. This is probably due to the fact that glucose dehydrogenase shows very low activity for NAD(P) derivatives alkylated at the 6-amino group (5,14). The activity of the enzyme moiety of GlcDH-PEG-NAD toward the bound NAD moiety can be measured by the coupled redox system of PES and MTT in the presence of glucose (internal activity (9)). In this reaction system, the NAD moiety is recycled by the two reactions of the enzyme moiety with glucose and PES with MTT, and the concentrations of PES and MTT are made high enough to keep more than 95% of the coenzyme moiety in the oxidized form at steady state. The concentration of glucose (0.1 M ) is much higher than the K,,, for glucose in a similar recycling assay system: GlcDH-PEG-NAD, 31 mM; glucose dehydrogenase + NAD, 18 mM. The concentrations of the coenzymes (<0.1 mM) are far below the K, for NAD (0.9 mM), and the K , for PEG-NAD is supposed to be much larger than that for native NAD (5,14). The internal activity of GlcDH-PEG-NAD at different concentrations of the conjugate is shown in Fig. 3 together with the activities of control systems containing native glucose dehydrogenase (GlcDH) plus NAD and GlcDH plus PEG-NAD; in the system of GlcDH + NAD, the concentrations of the enzyme and NAD are varied with a fixed ratio (1.6) of [NAD]/[GlcDH], whereas in the system of GlcDH + PEG-NAD, concentration of PEG-NAD is varied and that of glucose dehydrogenase is fixed at 0.31 HM.
The slope of the logarithmic plot shown in Fig. 3 gives the order of a reaction. The plot for GlcDH-PEG-NAD fits a straight line with a slope of 1 and a first order rate constant GlcDH + NAD at lower concentrations of the enzyme in Fig.   3 may be due to dissociation of tetrameric glucose dehydrogenase (15).
As the order of these reactions is different, the reaction rate of GlcDH-PEG-NAD can not be compared directly with those of GlcDH + NAD and GlcDH + PEG-NAD; the ratio of the reaction rates of GlcDH-PEG-NAD and GlcDH + NAD (or PEG-NAD) increases with the decrease in the concentration used for the assay. However, it is apparent that the rate of GlcDH-PEG-NAD is much higher than that of GlcDH + PEG-NAD, and is even higher than that of GlcDH + NAD under the conditions shown in Fig. 3. Namely, the low reaction rate of GlcDH + PEG-NAD is greatly enhanced just by covalently linking the enzyme and PEG-NAD. This effect of the covalent linking is a kind of "anchimeric assistance" (16), and the magnitude of the effect can be estimated using the ratio of kl.comp/kp,PEG-NAD, which is known as the effective concentration (16). The effective concentration of the NAD moiety of GlcDH-PEG-NAD is 4.2 mM; this means that the enzyme moiety of GlcDH-PEG-NAD acts as if it were in a solution containing 4.2 mM PEG-NAD irrespective of the actual concentration of the conjugate, assuming the proportionality of the reaction rate of GlcDH + PEG-NAD to the concentration of PEG-NAD up to 4.2 mM (in other words, K,,, for PEG-NAD is much higher than 4.2 mM). The ratio of the effective and the actual concentrations corresponds to the ratio of the reaction rates of GlcDH-PEG-NAD and GlcDH + PEG-NAD at the same actual concentration. For example, the rate of GlcDH-PEG-NAD at its concentration of 0.31 p M is estimated to be 6,500-fold the rate of GlcDH (0.31 p~) + PEG-NAD (0.65 p~) ; this value increases to 10,000-fold if account is taken of the fact that the specific activity of the enzyme moiety of the conjugate is 92% of that of the native enzyme, and that of the enzymically reducible NAD moiety of the conjugate is 70%.
The dramatic improvement of the activity of glucose dehydrogenase for PEG-NAD by this simple method of covalent linking has significant implications for enzyme technology, for it indicates that a much wider range of enzymes and coenzyme derivatives can be made applicable in the analytical and industrial fields.