Incorporation of an Artificial Receptor into a Native Protein - American

Nov 1, 1997 - binds a 1,2- or 1,3-diol unit of sugar to form a negatively charged boronate ... 2, yield ≈ 50%] were also prepared as control protein...
3 downloads 0 Views 369KB Size
Download PDF
Bioconjugate Chem. 1997, 8, 862−868

862

Incorporation of an Artificial Receptor into a Native Protein: New Strategy for the Design of Semisynthetic Enzymes with Allosteric Properties Itaru Hamachi,* Tsuyoshi Nagase, Yusuke Tajiri, and Seiji Shinkai Department of Chemistry and Biochemistry, Graduate School of Engineering, Kyushu University, Fukuoka 812, Japan. Received April 25, 1997X

The sugar-facilitated structure and enzymatic activity change of engineered myoglobins bearing a phenylboronic acid moiety, which were semisynthesized by a cofactor reconstitution method, were studied by the denaturation experiment, spectrophotometric titration of the pKa shift of the axial H2O, circular dichloism (CD), and the kinetics of the myoglobin-catalyzed-aniline hydroxylation reaction. Both boronophenylalanine-appended myoglobin [Mb(m-Bphe)2] and phenylboronic acidappended myoglobin [Mb(PhBOH)2] were stabilized by approximately 2 kcal/mol upon complexation with D-fructose. CD spectral changes and the sugar-induced pKa shift suggested that the microenvironment of the active site of these myoglobins was re-formed from a partially disturbed state to that comparable to the native state upon D-fructose binding. The correlation of pKa with kcat (for the aniline hydroxylase activity) and the ∆GDH2O-kcat profile showed that these structural changes of Mb(m-Bphe)2 and Mb(PhBOH)2 were closely related to their sugar-enhanced aniline hydroxylase activity. Thus, the results established that an incorporation of the artificial receptor molecule can be a valid methodology for the design of stimuli-responsive semiartificial enzymes.

INTRODUCTION

The introduction of unnatural atoms or molecules into naturally occurring enzymes and proteins is one of the most promising approaches for the development of new tailored proteins, which then have potential applications in chemistry and biology (Bell and Hilvert, 1994). Following Kaiser’s proposal on chemical mutation (Kaiser and Lawrence, 1984; Kaiser, 1988; Wu and Hilvert; 1989; Petersen and Hilvert, 1995), a more general biosynthetic method that utilizes aminoacylated suppresser t-RNA has been recently advanced by Schultz and co-workers (Cornish et al., 1995). In addition to developing other methodologies for the incorporation of unnatural molecules, the class of unnatural atoms and molecules that can positively act on native enzymes also needs to be expanded. As one example of such an interaction, we preliminarily reported that the sugar-binding ability of phenylboronic acid that has been incorporated into myoglobin can enhance the dioxygen storage capability of the myoglobin molecule (Hamachi et al., 1994a,b). On the basis of the previous findings, the present study demonstrates here that the aniline hydroxylase activity of these engineered myoglobins is facilitated by sugar molecules. The interesting relationship between the structure and the activity change induced by sugars is also discussed in great detail. RESULTS

Design of Phenylboronic Acid-Appended Myoglobins. Such researchers as Czarnik, Smith, and Shinkai have recently and independently demonstrated that phenylboronic acid is one of the most potentially useful artificial receptors for saccharide molecules (Yoon and Czarnik, 1992; Paugam et al., 1994; James et al., * Author to whom correspondence should be addressed (email [email protected]). X Abstract published in Advance ACS Abstracts, November 1, 1997.

S1043-1802(97)00055-4 CCC: $14.00

1996). A neutral boronic acid (sp2 configuration of boron) binds a 1,2- or 1,3-diol unit of sugar to form a negatively charged boronate ester (sp3 configuration) in aqueous solution. Site-specific alteration in the charge and the hydrophilicity can be induced when boronic acid binds with sugar molecules. Thus, phenylboronic acid is expected to be useful for the modulation of structure and/ or activity of native proteins.

Initially, in this study, heme 1 was synthesized and incorporated into apomyoglobin (apo-Mb) using a conventional cofactor reconstitution method. The resultant Mb(PhBOH)2 displayed a sugar responsive dioxygen storage capability. However, the low yield (30% for the reconstitution) and low stability of Mb(PhBOH)2 and its derivative rendered them unsuitable for detailed analysis. In an attempt to solve these problems, a more sophisticated design for the phenylboronic acid insertion into myoglobin was used. To retain two carboxylic acid groups, which are usually present in the side chains of the native heme cofactor, an unnatural amino acid, boronophenylalanine, was employed for pendant units. m-Boronophenylalanine-appended heme 3 was successfully reconstituted with apo-Mb in an almost quantitative yield [Mb(m-Bphe)2, g 95%]. Phenylalanine-appended Mb [Mb(Phe)2, reconstitution of heme 4, yield > 90%] and © 1997 American Chemical Society

Bioconjugate Chem., Vol. 8, No. 6, 1997 863

Semisynthetic Enzymes with Allosteric Properties

aniline-appended Mb [Mb(PhH)2, reconstitution of heme 2, yield ≈ 50%] were also prepared as control proteins.

After purification using previously reported procedures (Asakura and Yonetani, 1969; Asakura, 1978), the purity of the proteins was assayed using the ratio of Soret absorbance (408 nm) to protein absorbance (280 nm) as a known purity index (Tamura et al., 1973). The ratios were 4.5 for Mb(PhBOH)2, 4.4 for Mb(PhH)2, 4.8 for Mb(m-Bphe)2, and 4.7 for Mb(Phe)2. These are comparable to the value of 4.8 for native myoglobin, indicating that these semisynthetic myoglobins are pure enough to use in the subsequent studies. Sugar-Facilitated Stabilization of Phenylboronic Acid-Appended Myoglobins. The net stability of the present myoglobins was evaluated by urea-induced denaturation experiments. Increasing the concentration of urea (a denaturant) destroys the three-dimensional structure of myoglobin, resulting in the release of a heme cofactor. This denaturation process can be spectrophotometrically observed by the broadening of the Soret band (Puett, 1973). Figure 1a shows the denaturation curve of Mb(m-Bphe)2 plotting the Soret absorbance changes against urea concentration in the absence and presence of D-fructose. Clearly, Mb(m-Bphe)2 with D-fructose is more resistant to denaturation, relative to that without D-fructose. The conventional linear extrapolation of these denaturation curves gives denaturation free energies (∆GDH2O) (Pace, 1986). Table 1 summarizes ∆GDH2O values for all of the myoglobin derivatives. Interestingly, both Mb(m-Bphe)2 and Mb(PhBOH)2 were stabilized by approximately 2 kcal/mol as a result of D-fructose. On the other hand, the stability of Mb(Phe)2 (Figure 1b) and Mb(PhH)2 was not affected by D-fructose. In addition to the sugar-facilitated stabilization of phenylboronic acid-appended myoglobins, there are other important findings related to the design of engineered myoglobins (Table 1): (i) the modification of protoheme at both propionic acid functional groups destabilized the net 3D structure of holo-Mb, (ii) the retention of carboxylic acid groups at heme side chains was effective for the stabilization of reconstituted myoglobins by 2-4 kcal/mol [please compare the ∆GDH2O values for Mb(PhH)2 and Mb(Phe)2 or for Mb(PhBOH)2 to Mb(m-Bphe)2]. The intensified heme-apoprotein interactions by fructose were observed for phenylboronic acid-appended myoglobins using circular dichroism (CD) spectroscopy. Figure 2 compares the CD spectra of Mb(PhBOH)2 in the absence and presence of fructose. A positive peak at 408 nm is due to an induced CD of the heme which is sensitive to the microenvironment of the heme crevice (i.e., myoglobin active site). This peak is clearly intensified by 1.2-fold upon the addition of fructose, whereas two negative (225 and 208 nm) CD peaks and one positive (190 nm) CD peak, which are characteristic of the R-helix structures of myoglobin, scarcely change. Similar CD spectral changes were observed for Mb(m-Bphe)2, but no

Figure 1. Denaturation curve of the myoglobins derivatives by the addition of urea: (a) Mb(m-Bphe)2 in the absence (O) and the presence (b) of D-fructose (0.1 M); (b) Mb(Phe)2 in the absence (O) and presence (b) of D-fructose (0.1 M). The slope m is a parameter showing the cooperativity factor during the protein denaturation. The values are within 2.2 ( 0.5, so the cooperativity factors were practically identical in all cases. Table 1. Parameters for the Urea Denaturation: Free Energy Changes (∆GDH2O) and the Slope (m) none

D-fructose

m ∆GDH2O m ∆GDH2O (kcal/mol) (kcal/mol‚M-1) (kcal/mol) (kcal/mol‚M-1) Mb(PhBOH)2 Mb(PhH)2 Mb(m-Bphe)2 Mb(Phe)2 native Mb

6.3 4.5 8.0 8.3 14.3

2.3 1.9 2.3 2.4 2.7

8.3 4.5 10.1 8.4 13.6

2.3 1.8 2.6 2.3 2.5

a ∆G H2O was determined by the following equation: ∆G ) D D ∆GDH2O - m[urea] (Pace, 1986).

fructose-induced changes occurred for Mb(PhH)2, Mb(Phe)2, and native Mb by fructose (data not shown). It is clear that the heme-apoprotein interactions are considerably reinforced by the fructose molecules bound to phenylboronic acid sites and not by the fructose dissolved in a solution. Sugar-Induced pKa Shift of the Coordinated H2O. Sugar-induced UV-visible spectral changes in Mb(PhBOH)2, such as sharpening of the Soret band (408 nm) and intensification of Q-bands (503 and 630 nm) due to aqua-met-Mb with simultaneous lessening of Q-bands (540 and 580 nm) due to hydroxide-met Mb, were evident.

These changes appear to be due to the pKa shift of a water molecule coordinated to the iron(III) heme center (Brunori

864 Bioconjugate Chem., Vol. 8, No. 6, 1997

Hamachi et al.

Figure 2. CD spectral changes of the CN-coordinated Mb(PhBOH)2 in the absence (s) and presence (- - -) of D,L-fructose (0.1 M). Buffer was Mb(PhBOH)2 7 µM, KCN 10 mM, and 50 mM phosphate (pH 7.5) at 25 °C. The left side (180-250 nm) shows the secondary structure of the protein, and the right side (250-600 nm) shows the induced CD of the heme site. In these experiments, we used low-spin met-Mbs in which a cyanide anion was bound to iron(III) heme because the sugar-induced ligand exchange from hydroxide to H2O (due to the pKa shift) can be neglected in this form. Table 2. pKa of the Axial H2O in the Active Center Mb(PhBOH)2 Mb(PhH)2 Mb(m-Bphe)2 Mb(Phe)2 native Mb

none

D-fructose

8.0 8.3 8.5 8.1 9.0

8.5 8.3 9.0 8.1 9.0

et al., 1968). This suggests that the microenvironment of the active site may be modulated by sugar binding. Table 2 shows the pKa values of the reconstituted myoglobins in the presence and absence of D-fructose. The modification of the heme propionates clearly appears to cause the acidic pKa shift seen in all of the myoglobin derivatives. However, the pKa values for the phenylboronic acid-appended myoglobins were shifted to the basic side by fructose compared to the pKa value without fructose. In particular, the pKa of Mb(m-Bphe)2 in the presence of D-fructose became almost identical to that of native myoglobin. This suggests that the microenvironment of the active site of Mb(m-Bphe)2 is re-formed from a partially disturbed state to that comparable to the native state by D-fructose binding. It is highly likely that the bound sugar causes an increase in both the hydrophilicity and the density of negative charge at boronic acid moieties. On the basis of the pKa values, it is conceivable that D-fructose causes no significant change in the structure of the active site of the myoglobins bearing no sugar-binding sites. From these sugar-induced spectral changes of Mb(PhBOH)2, which show a typical saturation behavior with respect to the D-fructose concentration, we can estimate the binding constant of D-fructose to Mb(PhBOH)2. The Benesi-Hildebrand plot (Benesi and Hildebrand, 1949) gives a good linear relationship against the reciprocal square of the D-fructose concentration, indicating that a 2:1 complex of D-fructose/Mb predominantly formed with an association constant of 8 × 104 M-2. This value proves that 99% of Mb(PhBOH)2 binds two molecules of Dfructose in our normal conditions. Sugar-Enhanced Aniline Hydroxylase Activity of Phenylboronic Acid-Appended Myoglobins. The sugar-induced modulation that was evident not only in the net structure but also at the active site promoted us to investigate whether the enzymatic activity of phenylboronic acid-appended myoglobins can respond to sugar

Figure 3. Dependence of the initial rates on aniline concentration in the aniline hydroxylation reaction catalyzed by Mb ([Mb] ) 8 µM): (a) Mb(m-Bphe)2 in the presence of 0.1 M D-fructose; (b) Mb(m-Bphe)2 in the absence of D-fructose; (c) Mb(Phe)2 in the presence of 0.1 M D-fructose; (d) Mb(Phe)2 in the absence of D-fructose.

molecules. Therefore, an attempt was made to catalyze the aniline hydroxylation reaction using the myoglobin derivatives according to previously reported methodologies (Mieyal et al., 1976; Kokubo et al., 1987; Hamachi et al., 1995). As shown in Figure 3, the initial rates of generation of p-aminophenol were dependent on the aniline concentration and then gradually saturated at >10 mM of aniline. As shown, the initial rates of Mb(m-Bphe)2 were enhanced 3.5-fold by the addition of D-fructose (Figure 3a,b), whereas Mb(Phe)2 was less reactive and not responsive to D-fructose (Figure 3c,d). Double-reciprocal plots of the initial reaction rates against aniline concentrations (the Lineweaver-Burk plot) gave good linear relationships and yielded the Michaelis-Menten parameters (kcat and Km). Table 3 summarizes those parameters for all of the modified myoglobins. The following points in Table 3 should be noted. (i) Most importantly, the net activity (i.e., kcat/ Km) for Mb(PhBOH)2 and Mb(m-Bphe)2 was facilitated by D-fructose (7.7- and 3.5-fold, respectively), whereas the other myoglobins were unresponsive to D-fructose. (ii) In terms of both the kcat value and the net activity, Mb(m-Bphe)2 in the presence of D-fructose displayed the greatest hydroxylase activity of all the myoglobin derivatives. (iii) Apart from the net activity, the responsiveness

Bioconjugate Chem., Vol. 8, No. 6, 1997 865

Semisynthetic Enzymes with Allosteric Properties

Table 3. Kinetic Parameters of the Aniline Hydroxylase Activity of the Myoglobin Derivatives none Mb(PhBOH)2 Mb(PhH)2a Mb(m-Bphe)2 Mb(Phe)2 native Mb a

D-fructose

kcat (× 10-2 min-1)

Km (mM)

kcat/Km (min-1 M-1)

kcat (× 10-2 min-1)

Km (mM)

kcat/Km (min-1 M-1)

0.8 ( 0.2

9.4 ( 2.6

0.9 ( 0.3

2.4 ( 0.7

3.5 ( 1.0

6.9 ( 1.9

2.1 ( 0.7 1.7 ( 0.5 5.6 ( 0.4

3.5 ( 1.1 10.8 ( 3.2 8.8 ( 0.9

6.0 ( 2.0 1.6 ( 0.5 6.4 ( 0.5

7.0 ( 0.5 1.7 ( 0.4 4.8 ( 0.4

3.4 ( 0.4 9.9 ( 3.0 4.8 ( 0.6

21 ( 1.0 1.7 ( 0.4 5.0 ( 0.8

The kinetic parameters could not be determined because of its low activity.

Figure 4. Sugar structure dependence of aniline hydroxylase activity of Mb(m-Bphe)2. The amount of the generated paminophenol (reaction time ) 20 min) was determined according to the phenol-indophenol method (Mieyal et al., 1976; Kokubo et al., 1987; Hamachi et al., 1995). Reaction conditions were described under Experimental Procedures.

[i.e., (kcat/Km)D-fru/(kcat/Km)no D-fru] of Mb(PhBOH)2 was more efficient than that of Mb(m-Bphe)2. This is mainly ascribed to the fact that the binding affinity (1/Km) was enhanced 3-fold by D-fructose for only Mb(PhBOH)2, while kcat values were accelerated ∼3-fold by D-fructose for both Mb(PhBOH)2 and Mb(m-Bphe)2.

The aniline hydroxylase activity of Mb(m-Bphe)2 also depended on the sugar’s structure (Figure 4). The order of the sugar-facilitated reaction rates (D-fructose > Darabinose > D-mannose > D-saccharose > D-glucose) corresponds to the binding selectivity of the phenylboronic acid unit to sugar derivatives (Lorand and Edwards, 1959). In addition, the reaction rate shows a saturation curve with respect to the concentration of D-fructose, which corresponded well to the saturation behavior for the preceding D-fructose-induced spectral changes. These results clearly indicate that the sugar molecules bound to phenylboronic acid moieties play an essential role in the enhanced aniline hydroxylase activity. DISCUSSION

The present results establish that an incorporation of the artificial receptor molecule can be a valid method for the design of stimuli-responsive semiartificial enzymes. Although the Mb-catalyzed aniline hydroxylation reaction is mechanistically complex, the reductively activated O2 that is bound to the heme oxidizes aniline according to Mieyal’s mechanism. The pKa-kcat profile (Figure 5a) shows that the kcat value was promoted along with an

Figure 5. Relationship of the kinetic parameter kcat with (a) pKa of the axial H2O or (b) the Mb structure stability: Mb(PhBOH)2 (1), Mb(m-Bphe)2 (3), Mb(Phe)2 (5) and native Mb (7) in the absence of D-fructose, respectively; Mb(PhBOH)2 (2), Mb(m-Bphe)2 (4), Mb(Phe)2 (6) and native Mb (8) in the presence of 0.1 M D-fructose, respectively.

increase in the pKa of the coordinated water (log kcat ) 0.73pKa - 7.8; correlation coefficient ) 0.96). The effect of D-fructose on Mb(PhBOH)2 and Mb(m-Bphe)2 is in this correlation line, implying that the bound D-fructoses induced the considerable pKa shift, and as a result, the catalytic efficiency of the myoglobin derivatives was enhanced. In addition, rough correlation (correlation coefficient ) 0.82) between ∆GDH2O and kcat was observed as shown in Figure 5b. Apparently, the activity increases with protein stability. It is conceivable sugar molecules operate as an active effector on the enzymatic activity, as well as structural stability (see Figure 6). There are many allosteric enzymes, the activity of which can be regulated by various effectors. Proton and bisphosphoglycerate, for example, are known to lower the dioxgen affinity of hemoglobin (Hb) by factors of 9 and 3, respectively (Imai, 1979; Benesch and Benesch, 1969). The activity of aspartate transcarbamylase is regulated by adenine triphosphate (ATP; 1.5-fold) and cytidine triphosphate (CTP; 0.5-fold) (Kantrowitz et al., 1980). The magnitude of the activity enhancement (3.5-7.7-fold) in

866 Bioconjugate Chem., Vol. 8, No. 6, 1997

Hamachi et al.

Figure 6. Schematic illustration of our sugar responsive enzymes bearing phenylboronic acid moieties.

our system is not sufficient for practical use as a biosensor, but the activity enhancement is in the same range as that of allosteric native enzymes. Hemoprotein engineering has been performed primarily by the exchange of axial ligands (to change the electronic state of the heme iron) or the replacement of amino acids located in the active site (to modify its microenvironment) (Egeberg et al., 1990; Raphael and Gray, 1991; Bren and Gray, 1993; Pin et al., 1994; Qin et al., 1994; Adachi et al., 1993; DePillis et al., 1994; Lloyd et al., 1995). In sharp contrast, the present example is unique in its dynamic modulation of activity using the external stimuli that act on a receptor located on the protein surface, not at the active site. This effector-triggered modulation of cofactor-apoenzyme interactions can also be applied to other kinds of cofactordependent enzymes. Although the interaction between boronic acid and sugars is employed as a typical model in this study, we believe that a wide variety of artificial molecular receptors can play a crucial role in the efficient modulation of native enzyme properties (Zuckermann et al., 1988; Corey et al., 1989; Reinhoudt et al., 1989). EXPERIMENTAL PROCEDURES

Materials. Protoporphyrin IX (PP IX) was purchased from Aldrich. Myoglobin (horse heart) was purchased from Sigma. All chemicals were used without further purification. General Procedures. Thin-layer chromatography (TLC) was carried out on aluminum sheets coated with silica gel (Merck 5554). Column chromatography was performed on silica 60 (Merck 9385, 230-400 mesh). Melting points were determined on a Micro Melting Point Apparatus Yanoco MP-500D and are uncorrected. UVvisible spectra were recorded on a Shimadzu UV-3000 spectrophotometer. 1H-NMR spectra were recorded on either a Bruker AC-250P (250MHz) or a JEOL GSX-400 (400 MHz) spectrophotometer. IR spectra were recorded on a Jasco A-100 spectrophotometer. CD spectra were recorded on a Jasco J-720 spectrophotometer. Synthesis. Chemically modified heme derivatives were synthesized according to Scheme 1. m-Boronophenylalanine ethyl ester (5) was prepared according to the slightly modified procedure of the para derivatives reported previously (Snyder et al., 1958). Synthesis of a m-Boronophenylalanine-Appended Porphyrin (3a). Under N2 atmosphere, oxalyl

chloride (0.2 mL) was added dropwise to a suspended dry CH2Cl2 solution (12 mL) containing PP IX (100 mg, 0.18 mmol) with ice cooling. After 1 h of stirring at room temperature, the mixture was concentrated and dried in vacuo. The residual dark green solid dissolved in dry CH2Cl2 (8 mL) was added dropwise to an anhydrous mixed solvent [pyridine (3 mL)/CH2Cl2 (5 mL)] containing 5‚HCl salt (325 mg, 1.2 mmol). The reaction mixture was stirred at room temperature overnight and then concentrated in vacuo. The residue dissolved in CHCl3 (100 mL) was washed with acidic water (pH 3) and then with saturated NaHCO3 aqueous solution. The organic layer was concentrated and applied to column chromatography [silica gel, 3 cm × 15 cm; solvent, CHCl3/MeOH ) 30/1 (v/v)] to yield 6 (58 mg, 33%): mp 210 °C dec; IR (KBr, cm-1) 3400 (OH), 1735 (ester CdO), 1650 (amide CdO); 1H NMR (CDCl ) δ -3.96 (2H, s, NH), 0.68 [6H, m, 3 CH3(ester)], 2.69 [4H, m, CH2(benzyl)], 3.08 (4H, m, CH2CO), 3.42 [4H, m, CH2(ester)], 3.65 (12H, m, CH3), 4.27 (2H, m, CH), 4.38 (4H, m, CH2), 6.22, 6.45 (2H each, d each, dCH2), 6.78, 7.42, 7.53 (4H, 2H, and 2H, respectively; m, t, and s, respectively, ArH), 7.92 (4H, s, OH), 8.53 (4H, m, NH and CHd), 10.24 [4H, m, CH(meso position)]. Anal. Found: C, 65.31; H, 6.01; N, 7.86. Calcd for C56H62N6O10B2‚1.5H2O: C, 65.44; H, 6.29; N, 8.18. Synthesis of Iron(III) Complex (3b). FeCl2‚4H2O (80 mg, 0.4 mmol) and 3a (40 mg, 0.04 mmol) were mixed in dry DMF (20 mL) and stirred at 65 °C for 8 h in the dark. DMF was evaporated off, and the residue was washed with dilute HCl (pH 3, four times). The dark brown solid was purified by reprecipitation (CHCl3/ MeOH/hexane) to yield 3b (35 mg, 85%): mp 250 °C dec; IR (KBr, cm-1) 3400 (OH), 1735 (ester CdO), 1650 (amide CdO); UV-vis (MeOH, λmax) 502 and 625 nm. Anal. Found: C, 59.79; H, 5.48; N, 7.28. Calcd for C56H60N6O10B2FeCl‚2H2O: C, 59.73; H, 5.73; N, 7.46. Synthesis of Heme 3. A mixed solution [THF (10 mL) and MeOH (4 mL)] containing 3b (60 mg, 0.06 mmol) and 1 N aqueous NaOH (0.17 mL) was stirred at room temperature overnight. After solvents were evaporated off, the residue was dissolved in 20 mL of H2O. A black solid was precipitated by acidification (pH 3), and it was further purified by reprecipitation (CHCl3/MeOH/hexane) to afford 3 (20 mg, 35%): mp 250 °C dec; IR (KBr, cm-1) 3300 (OH), 1710 (calboxylic acid CdO), 1650 (amide C)O). Anal. Found: C, 60.63; H, 5.11; N, 7.98. Calcd for C52H52N6O10B2FeCl: C, 60.41; H, 5.07; N, 8.13.

a

(i) (COCl)2/dry CH2Cl2; then R-H, pyridine/dry CH2Cl2, (ii) FeCl2/dry DMF, (iii) OH-/THF, H2O.

Scheme 1a

Semisynthetic Enzymes with Allosteric Properties

Bioconjugate Chem., Vol. 8, No. 6, 1997 867

868 Bioconjugate Chem., Vol. 8, No. 6, 1997

Other heme derivatives (1, 2, and 4) were prepared in the same manner. Their analytical data are as follows. Heme 1. Anal. Found: C, 62.35; H, 5.00; N, 9.34. Calcd for C46H44N6O6B2FeCl: C, 62.09; H, 4.98; N, 9.44. Heme 2. Anal. Found: C, 67.56; H, 5.35; N, 10.06. Calcd for C46H42N6O2FeCl‚H2O: C, 67.36; H, 5.41; N, 10.25. Heme 4. IR (KBr, cm-1) 1720 (carboxylic acid CdO), 1640 (amide CdO). Anal. Found: C, 66.12; H, 5.41; N, 8.80. Calcd for C53H52O6N6FeCl: C, 66.29; H, 5.46; N, 8.75. Reconstitution of Apo-Mb with Various Heme Derivatives. Chemically modified hemes were incorporated into the heme pocket of apo-Mb according to a slightly modified procedure described previously (Asakura and Yonetani, 1969; Asakura, 1978). Molar extinction coefficients of reconstituted myoglobins were determined by following the conventional pyridine-hemochromogen method (Paul et al., 1953) [409 for Mb(PhBOH)2 ) 147 mM-1 cm-1, for Mb(PhH)2 ) 138, for Mb(m-Bphe)2 ) 197, for Mb(Phe)2 ) 151, and for native Mb ) 188]. Myoglobins’ concentrations were spectrophotometrically determined using the corresponding values. Denaturation of Mbs with Urea. Denaturation experiments of the myoglobin derivatives with increasing amounts of urea were conducted according to the reported method (Puett, 1973; Pace, 1986) at [Mb] ) 8 µM, 50 mM phosphate buffer (pH 7.5) and 25 °C. pH Titration of Met-Mbs. Spectrophotometric pH titration experiments were done according to the standard procedure. The pH of the solution was controlled by the addition of an appropriate amount of 1 N aqueous NaOH. Assay of Aniline Hydroxylation by Mb. Aniline hydroxylase activities were estimated by measuring the amount of produced p-aminophenol according to the phenol-indophenol method (Mieyal et al., 1976; Kokubo et al., 1987; Hamachi et al., 1995). The reaction mixture consisted of a total volume of 1 mL containing appropriate amounts of aniline, 1 mM NADH, 500 µM FMN, 8 µM Mb, and 0 or 0.1 M D-fructose in 50 mM phosphate buffer (pH 7.5). The reaction was initiated by the addition of NADH at 37 °C and terminated by the extraction of the mixture with 1 mL of ethyl ether (three times). The ether layers were combined and evaporated under nitrogen gas at room temperature. The residue dissolved in 1 mL of 0.1 N HCl was mixed with 0.2 mL of 2.5 N Na2CO3 and with 5% (w/v) phenol in 2.5 N NaOH. The mixture was then allowed to stand at 37 °C for 30 min. Absorbance of the resultant solution was measured at 630 nm, the absorption maximum of the indophenol derivative (1 nmol of p-aminophenol in the standard assay mixture gave a ∆A630nm value of 0.018). LITERATURE CITED Adachi, S., Nagano, S., Ishimori, K., Watanabe, Y., and Morishima, I. (1993) Biochemistry 32, 241. Asakura, T. (1978) In Methods in Enzymology (S. Fleiser and L. Packer, Eds.) Part C, p 446, Academic Press, New York.

Hamachi et al. Asakura, T., and Yonetani, T. (1969) J. Biol. Chem. 244, 537. Bell, I. B., and Hilvert, D. (1994) In The Lock-and-Key Principles, (J.-P. Behr, Ed.) P 73, Wiley, New York. Benesch, R., and Benesch, R. E. (1969) Nature 221, 618. Benesi, H., and Hildebrand, J. H. (1949) J. Am. Chem. Soc. 71, 2703. Bren, K. L., and Gray, H. B. (1993) J. Am. Chem. Soc. 115, 10382. Brunori, M., Amiconi, G., Antonini, E., Wyman, J., Zito, R., and Rossi Fanelli, A. (1968) Biochim. Biopys. Acta 154, 315. Corey, D. R., Pei, D., and Schultz, P. G. (1989) J. Am. Chem. Soc. 111, 8523. Cornish, V. W., Mendel, D., and Schultz, P. G. (1995) Angew. Chem., Int. Ed. Engl. 34, 621. DePillis, G. D., Decatur, S. M., Barrick, D., and Boxer, S. G. (1994) J. Am. Chem. Soc. 116, 6981. Egeberg, K. D., Springer, B. A., Martinis, S. A., and Sliger, S. G. (1990) Biochemistry 29, 9783. Hamachi, I., Tajiri, Y., and Shinkai, S. (1994a) J. Am. Chem. Soc. 116, 7437. Hamachi, I., Tajiri, Y., Murakami, H., and Shinkai, S. (1994b) Chem. Lett., 575. Hamachi, I., Fujita, A., and Kunitake, T. (1995) Chem. Lett., 657. Imai, K. (1979) J. Mol. Biol. 133, 233. James, T. D., Linnane, P., and Shinkai, S. (1996) J. Chem. Soc., Chem. Commun., 281 and references cited therein. Kaiser, E. T. (1988) Angew. Chem., Int. Ed. Engl. 27, 913 and references cited therein. Kaiser, E. T., and Lawrence, D. S. (1984) Science 266, 505. Kantrowitz, E. R., Pastr-Landis, S. C., and Lipscomb, W. N. (1980) Trends Biochem. Sci. 5, 124. Kokubo, T., Sassa, S., and Kaiser, E. T. (1987) J. Am. Chem. Soc. 109, 606. Lloyd, E., Hildebrand, D. P., Tu, K. M., and Mauk, A. G. (1995) J. Am. Chem. Soc. 117, 6434. Lorand, J. P., and Edwards, J. O. (1959) J. Org. Chem. 24, 769. Mieyal, J. J., Acherman, R. S., Blumer, J. L., and Freeman, L. S. (1976) J. Biol. Chem. 251, 3436. Pace, C. N. (1986) Methods Enzymol. 131, 206. Paugam, M.-F., Valencia, L. S., Boggess, B., and Smith, B. D. (1994) J. Am. Chem. Soc. 116, 11203. Paul, K. G., Theorell, H., and A° keson, A° . (1953) Acta Chem. Scand. 7, 1284. Petersen, E. B., and Hilvert, D. (1995) Biochemistry 34, 6616. Pin, S., Alpert, B., Cortes, R., Ascone, I., Chiu, M. L., and Silgar, S. G. (1994) Biochemistry 33, 11618. Puett, D. (1973) J. Biol. Chem. 248, 4623. Qin, J., La Mar, G. N., Dou, Y., Admiraal, S. J., and Ikeda-Saito, M. (1994) J. Biol. Chem. 269, 1083. Raphael, R., and Gray, H. B. (1991) J. Am. Chem. Soc. 113, 1038. Reinhoudt, D. N., Egendebak, A. M., Nigenhuis, W. F., Verboom, W., Kloosterman, M., and Schoemaker, H. E. (1989) J. Chem. Soc., Chem. Commun., 399. Snyder, H. R., Reedy, A. J., and Lennarz, W. J. (1958) J. Am. Chem. Soc. 80, 835. Tamura, M., Asakura, T., and Yonetani, T. (1973) Biochim. Biophys. Acta 295, 467. Wu, Z.-P., and Hilvert, D. (1989) J. Am. Chem. Soc. 111, 4513. Yoon, J., and Czarnik, A. W. (1992) J. Am. Chem. Soc. 114, 5874. Zuckermann, R. N., Corey, D. R., and Schultz, P. G. (1988) J. Am. Chem. Soc. 110, 1614.

BC970055Z