Application of inverse substrates to trypsin-catalyzed peptide synthesis

Article abstract of DOI:10.1006/bioo.1996.0007

Trypsin-catalyzed peptide synthesis has been studied by using 'inverse substrate,' i.e., p-amidinophenyl ester derived from α-amino acid derivative as an acyl donor component. Inverse substrate can afford acyl trypsin in a very specific manner, liberating

Full text of DOI:10.1006/bioo.1996.0007

BIOORGANIC CHEMISTRY 24, 59–68 (1996)
ARTICLE NO. 0007
Application of Inverse Substrates to Trypsin-Catalyzed
Peptide Synthesis
KUNIHIKO
I
TOH, HARUO
S
EKIZAKI, EIKO
T
OYOTA, NORIHISA
F
UJIWARA AND
,
1
K
AZUTAKA ANIZAWA
T
Faculty of Pharmaceutical Sciences, Health Sciences University of Hokkaido, Ishikari-Tobetsu,
Hokkaido 061-02, Japan
Received May 25, 1995
Trypsin-catalyzed peptide synthesis has been studied by using ‘‘inverse substrate,’’ i.e., p-
amidinophenyl ester derived from Ͱ-amino acid derivative as an acyl donor component.
Inverse substrate can afford acyl trypsin in a very specific manner, liberating the site-specific
p-amidinophenyl moiety as the leaving group. Thus a variety of Ͱ-amino acid residues which
are a part of p-amidinophenyl ester can be involved in the trypsin-catalyzed coupling reaction.
The method has been shown to be successful as expected. In conclusion, the method was
proposed as a new procedure which overcomes the disadvantage of enzymatic peptide
synthesis.
1996 Academic Press, Inc.
INTRODUCTION
It is recognized that enzyme-catalyzed peptide synthesis is more advantageous
than chemical synthesis in many respects (1, 2). The enzymatic method is highly
stereoselective, racemization-free, and requires minimal side-chain protection. The
enzymatic method is at an advantage when discriminating enantiomers and amino
acid side-chain residue and at a disadvantage as a general procedure of peptide
synthesis. The reactants used as acyl coupling component are limited. Only the
amino acid or peptide derivatives which meet the substrate specificity of the enzyme
are applicable to the coupling reaction. This is because the enzymatic peptide
coupling is considered to proceed via enzyme–substrate complex and subsequent
acyl enzyme formation. In the case of trypsin-catalyzed reaction, for example, the
acyl donor is limited to positively charged arginine or lysine residue at the carboxyl
end position.
In our previous work (3), it has been shown that esters of p-amidinophenol are
specifically hydrolyzed by trypsin and trypsin-like enzymes. In these esters the site-
specific group for the enzyme, a charged amidinium group, is liberated during
acylation to produce an acyl enzyme intermediate. Kinetic analyses showed that
the binding affinity and catalytic efficiency at the acylation step of these esters are
comparable to those for normal-type specific substrates, and so a new term, ‘‘inverse
substrates,’’ was proposed for these esters. Thus the compounds provide a general
method for efficient synthesis of acyl-enzymes without recourse to the structure of
1 To whom correspondence should be addressed.
59
0045-2068/96 $18.00
Copyright 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

60
ITOH ET AL.
the acyl component. The compounds are applicable to trypsin-catalyzed peptide
synthesis for the same reason. The method is extremely useful since the acyl compo-
nent is not restricted to cationic residues such as lysine and arginine.
Noting the characteristic feature of inverse substrate Schellenberger et al. (4)
have tested the applicability of p-guanidinophenyl ester as tool for trypsin-catalyzed
peptide synthesis. The paper prompted us to report the results on the general
use of our inverse substrate, p-amidinophenyl esters, to tryptic peptide synthesis
in detail.
EXPERIMENTAL PROCEDURES
Synthetic Chemistry
General. Dimethylformamide (DMF), ethyl acetate, and diethyl ether were ob-
tained as the anhydrous solvents by usual manner. N-tert-butyloxycarbonylamino
(Boc-amino acid) and N,NЈ-dicyclohexylcarbodiimide(DCC) were purchased
acid
Peptide Institute, Inc. All other reagents were used without further purification
from
from Aldrich Chemical Co., Inc., Wako Pure Chemical Industries, LTD., and Kanto
Chemical Co., Inc. Melting points were determined on a Yanaco MP-500D appara-
tus and are uncorrected. Infrared spectra were recorded on a JASCO VALOR-III
FT-IR spectrometer. NMR spectra were recorded on a JEOL JNM-FX400 FT
NMR spectrometer.
General procedure for synthesis of inverse substrates. Inverse substrates, i.e., Boc-
amino acid p-amidinophenyl esters, were prepared as follows according to our
previous paper (5). DCC (454 mg, 2.2 mmol) was added to a solution of Boc-amino
acid (2.0 mmol), N-benzyloxycarbonyl-p-amidinophenol (5) (540 mg, 2.0 mmol),
and 4-dimethylaminopyridine (24 mg, 0.2 mmol) in 3 ml of DMF and 6 ml of ethyl
acetate. The reaction mixture was kept at 0ЊC for 30 min and subsequently at room
temperature overnight. The precipitated dicyclohexylurea was removed by filtration
and the filtrate was evaporated in vacuo. The resulting N-benzyloxycarbonyl-p-
amidinophenyl ester was recrystallized from benzene–hexane. Isolation yields of
64–80% resulted. Structure of the ester was confirmed by NMR spectra, ir spectra,
and elemental analysis. Optical purity of each enantiomer was analyzed by optical
rotation measurement. The ester (1.0 mmol) was dissolved in methanol (20 ml)
containing p-toluenesulfonic acid monohydrate (190 mg, 1.0 mmol) and hydroge-
nated in the presence of 10% Pd-charcoal (10 mg). After the catalyst was removed,
the filtrate was evaporated to dryness in vacuo and the residue was washed with
dry ether. Recrystallization from ethanol–ether gave pure p-amidinophenyl ester to-
sylate.
p-Amidinophenyl esters prepared according to the general procedure are summa-
rized in Table 1.
Enzyme Assay
Substrates and enzymes. Alanine p-nitroanilide (Ala-pNA) and leucine p-nitroani-
lide (Leu-pNA) were purchased from Peptide Institute, Inc. Glycine p-nitroanilide
(Gly-pNA),
D-alanine p-nitroanilide (
D-Ala-pNA), and
D-leucine p-nitroanilide
(D-Leu-pNA) were prepared following the reported procedure (6–8).

SUBSTRATE FOR ENZYMATIC PEPTIDE SYNTHESIS
TABLE 1
61
Physical Data of the Inverse Substrates (N-tert-Butyloxycarbonylamino Acid p-Amidinophenyl Ester
Tosylate) Used for Trypsin-Catalyzed Coupling Reaction
Analysis:
calculated(found)
Inverse substrate
tosylate
[Ͱ]2D0
mp(ЊC) (c ϭ 1, methanol)
Formula
C
H
N
S
Boc-GlyOAm
Boc-AlaOAm
170–172
C21H27N3O6S 4/3H2O
52.65 6.00 8.77 6.69
(52.37 5.70 8.53 7.04)
162–164
163–164
161–162
Ϫ41.8
ϩ41.2
Ϫ7.2
C22H29N3O7S
55.10 6.10 8.76 6.69
(55.04 6.07 8.72 6.71)
Boc-D-AlaOAm
C22H29N3O7S 1/2H2O
C28H33N3O7S 1/2H2O
C28H33N3O7S 1/2H2O
C25H35N3O7S 2/3H2O
C25H35N3O7S 2/3H2O
54.09 6.19 8.60 6.56
(54.00 5.94 8.48 6.69)
Boc-PheOAm
59.56 6.07 7.44 5.68
(59.56 5.80 7.46 5.96)
Boc-D-PheOAm 161–163
ϩ7.0
59.56 6.07 7.44 5.68
(59.68 5.93 7.49 5.99)
Boc-LeuOAm
153–154
Ϫ40.0
ϩ39.2
56.27 6.86 7.65 6.01
(56.24 6.66 7.65 6.26)
Boc-D-LeuOAm 153–155
56.27 6.86 7.65 6.01
(56.37 6.63 7.71 6.43)
Bovine pancreatic trypsin (EC 3.4.21.4) purchased from Worthington Biochemi-
cal Corp. (twice recrystallized, lot TRL) was further purified by affinity chromatogra-
phy (9), using Benzamidine Sepharose 6B obtained from Pharmacia. Enzyme con-
centration was determined from active site titration using p-nitrophenyl p-
guanidinobenzoate (10).
Trypsin-catalyzed peptide coupling reaction. Peptide coupling reaction was carried
out at 25ЊC in 50 m
containing dimethylsulfoxide (DMSO). Concentrations of acyl donor (inverse sub-
strate), acyl acceptor (amino acid p-nitroanilide), and trypsin were 1 m , 20 m
and 5Ȑ , respectively. The progress of the coupling reaction was monitored by
M
4-morpholinepropanesulfonic acid buffer (MOPS), pH 8.0,
M
M
,
M
HPLC on a Hitachi Model L-6200 equipped with a Hitachi UV-VIS detector. The
chromatograph was equipped with a Wakosil 5C18-200 column (4.0 ϫ 250 mm).
The mobile phase was acetonitrile-0.1% aqueous trifluoroacetic acid and the flow
rate was 1.0 ml/min. Analysis was performed using isocratic system. An aliquot of the
reaction mixture was injected and peaks were detected at 310 nm for p-nitroanilide
moiety. Peak identification was made by correlation with authentic samples which
were chemically synthesized (11, 12). Peak intensities were used to calculate relative
concentration.

62
ITOH ET AL.
RESULTS
Trypsin-catalyzed coupling reaction of N-tert-butyloxycarbonylalanine p-amidi-
nophenyl ester (Boc-Ala-OAm) and Ala-pNA to give Boc-Ala-Ala-pNA was deter-
mined in aqueous DMSO and aqueous DMF. The coupling reaction was instanta-
neous in the media less than 60% organic solvent and was completed within several
minutes. During the reaction period the peak of Boc-Ala-OAm completely disap-
peared from the HPLC elution diagram. Only two peaks which were assigned as
the coupling product and the hydrolysate newly emerged. Control experiments in
the absence of trypsin showed that hydrolysis of Boc-Ala-OAm is so slow that no
appreciable amount of the ester is consumed during the period that enzymatic
reaction is complete. The rate constant for the spontaneous hydrolysis at pH 8.0
5
in 50% DMSO was determined to be 1.3 ϫ 10Ϫ sϪ1 which is equal to 14.8 h of half-
life time. In the presence of the acyl acceptor aminolysis takes place and affords
coupling product even in the absence of the enzyme though the extent of the
reaction is not large. It was found that the nonenzymatic aminolysis is sufficiently
less than the enzymatic reaction; i.e., coupling yield of Boc-Ala-Ala-pNA in the
absence of trypsin is 3% after 0.1 h. Thus the enzymatic hydrolysis of acyl donor
(inverse substrate) is the only reaction competitive to the peptide synthesis, and
the rest of the coupling is nearly equal to the part of hydrolysis.
Effects of DMSO and DMF concentration on coupling yields were shown in Fig.
1. Coupling yields higher than 65% were observed at the DMSO concentration
range of 20–60%, and the best yield (85%) was obtained at 50% DMSO. Effect of
DMF was nearly the same as that of DMSO but the coupling yield in aqueous
DMF was less than that in aqueous DMSO. The diminished coupling yield at the low
concentration of organic solvent could be due to the enzyme-catalyzed hydrolysis of
acyl donor as mentioned previously. A higher concentration of organic solvent is
not completely advantageous for the coupling, although it is expected to retard the
hydrolysis. The high concentration of organic solvent results in the separation of
the catalyst as well as the decrease of enzymatic activity.
The effect of pH of the buffer component in the medium on the coupling yields
was analyzed. Reaction yields were determined for 50% DMSO solutions, changing
pH of the buffer solution. The pH dependency of the yield at the reaction period
of 10 min was determined as shown in Fig. 2. The pH dependency was also
determined for 50% DMF in a similar manner. The pH dependency was not different
for both cases though the reaction yield in 50% DMF was relatively low. The
observed dependency was similar to that of trypsin-catalyzed hydrolysis of the
specific ester substrates as reported previously in which the catalytic rate was
increased at the higher pH and reached the limit at around pH 9 (13). Effect of
cosolvent on the coupling yield was analyzed. Among six solvents so far tested,
DMF and especially DMSO were found to be advantageous, as shown in Table 2.
The analysis was made in MOPS buffer (pH 8.0) containing 50% organic solvent.
For each case the coupling reaction was completed within 2 h. As shown in Table
2 the best result was obtained with DMSO after all. The observation was not
different from that of enzyme-catalyzed coupling procedure by means of conven-
tional substrates (1, 2).

SUBSTRATE FOR ENZYMATIC PEPTIDE SYNTHESIS
63
F
IG. 1. Effect of organic solvent on trypsin-catalyzed condensation of t-butyloxycarbonylalanine p-
amidinophenyl ester and alanine p-nitroanilide. Reaction was carried out in 50 m MOPS buffer (pH
8.0) containing DMSO (᭺) or DMF (᭹) at 25ЊC. Product yield was analyzed after reaction period of
10 min in which the coupling was completed. Boc-Ala-OAm, 1 m ; Ala-pNA, 20 m ; trypsin, 5 Ȑ
M
M
M
M.
F
IG. 2. pH dependency of trypsin-catalyzed condensation. Boc-Ala-OAm and Ala-pNA were reacted
in 50 m 4-morpholineethanesulfonic acid (MES) (᭿), 50 m MOPS (᭹), and 50 m Tris (᭺) buffers
containing 50% DMSO at 25ЊC. Boc-Ala-OAm, 1 m ; Ala-pNA, 20 m ; trypsin, 5 Ȑ . Coupling yield
was determined after a reaction period of 10 min in which the coupling was completed.
M
M
M
M
M
M

64
ITOH ET AL.
TABLE 2
Effect of Water-Miscible Or-
ganic Cosolvent on Trypsin-Cata-
lyzed Peptide Synthesisa
Cosolvent
Yield (%)
DMF
DMSO
Dioxane
Acetonitrile
THFb
29
58
9
18
5
2-Propanol
10
a Boc-Ala-OAm, 1 m
NA, 10 mM; trypsin, 10 ȐM
vent: MOPS(50 m
M
; Ala-p-
; cosol-
M, pH 8.0), 1:1,
at 25ЊC for 2 h.
b Precipitation of Ala-pNA re-
sulted.
Acyl acceptor concentration influenced the coupling yield. The reaction was
determined in aqueous DMSO (Fig. 3). The reaction yield was increasing to 80%
when 40 m acyl acceptor was used.
Coupling reactions were carried out for various combinations of acyl donor and
acyl acceptor. As summarized in Table 3, the reaction involving -acyl acceptor is
not favorable. Reaction yields are very low for cases in which either - or
M
D
L
D-
enantiomer was used as acyl donor. In contrast, the structure of acyl donor does
not markedly affect the reaction yield. Acyl donors derived from Gly, Ala, Leu,
and Phe react with Ala-pNA with comparable rates. Reaction rate for acyl donor
F
IG. 3. Effect of acyl acceptor concentration on coupling reaction. Reactions were carried out in 50
m
m
M
MOPS (pH 8.0) containing 50% DMSO at 25ЊC, using different concentration of Ala-pNA (1–80
). Boc-Ala-OAm, 1 mM; trypsin, 5 ȐM.
M

SUBSTRATE FOR ENZYMATIC PEPTIDE SYNTHESIS
65
TABLE 3
Trypsin-Catalyzed Peptide Coupling by Use of Inverse Substrate as Acyl Donor Componenta
Acyl donor
Acyl acceptor
Product
Reaction time (h)
Yield (%)
Boc-Gly-OAm
Boc-Ala-OAm
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Ala-pNA
Leu-pNA
Leu-pNA
Gly-pNA
Gly-pNA
Boc-Gly-Ala-pNA
Boc-Ala-Ala-pNA
0.1
0.1
1
0.1
2
0.1
2
2
12
0.1
0.5
0.1
6
0.1
3
1
77
77
65
85
68
82
64
80
76
74
54
47
53
8
7
Boc-
Boc-Leu-OAm
Boc- -Leu-OAm
Boc-Phe-OAm
Boc- -Phe-OAm
Boc-Pro-OAm
Boc- -Pro-OAm
Boc-Ala-OAm
Boc- -Ala-OAm
Boc-Ala-OAm
Boc- -Ala-OAm
Boc-Ala-OAm
Boc- -AlaOAm
Boc-Ala-OAm
Boc- -Ala-OAm
D-Ala-OAm
Boc-
Boc-Leu-Ala-pNA
Boc- -Leu-Ala-pNA
Boc-Phe-Ala-pNA
Boc- -Phe-Ala-pNA
Boc-Pro-Ala-pNA
Boc- -Pro-Ala-pNA
Boc-Ala-Leu-pNA
Bod- -Ala-Leu-pNA
Boc-Ala-Gly-pNA
Bod- -Ala-Gly-pNA
Boc-Ala- -Ala-pNA
Boc- -Ala- -Ala-pNA
Boc-Ala- -Leu-pNA
Boc- -Ala- -Leu-pNA
Boc-ͱ-Ala-Ala-pNA
D
-Ala-Ala-pNA
D
D
D
D
D
D
D
D
D
D
D-Ala-pNA
D-Ala-pNA
D-Leu-pNA
D-Leu-pNA
D
D
D
D
D
Trace
Trace
8
D
D
D
4
2
Boc-ͱ-Ala-OAm
Ala-pNA
a Acyl donor, 1 m
M
; acyl acceptor, 20 m
M; trypsin, 5 Ȑ
M
; DMSO: MOPS(50 m
M
, pH 8.0), 1:1.
derived from
D
-amino acid is more or less slower than that of corresponding -
L
enantiomer and the difference is reflected in the final reaction yields for both cases.
As previously stated, the enzymatic hydrolysis of inverse substrate is the only
process competitive to the peptide coupling, and the rate ratios of coupling reaction
over hydrolysis directly determine the coupling yield. The observed small difference
in the coupling yields for the reactions of the enantiomeric pair of acyl donor could
be due to the respective rate ratio.
In Fig. 4 time courses of the reaction of Ala-pNA with Boc-Leu-OAm or Boc-
Leu-OAm were shown. Both reactions were compared with nonenzymatic coupling
reactions. It was shown that -acyl donors are efficient substrates for the enzymatic
coupling reaction though they are relatively less efficient than -acceptors. The
-acyl donors.
D-
D
L
behavior of the
D-acyl acceptors is distinct from that of
D
DISCUSSION
Our results in Table 3 show that the enantiomeric preference of acyl acceptor is
rather strict. This is in contrast to the result obtained from the conventional enzy-
matic method. It is reported that the enantiomeric preference of acyl acceptor for
the trypsin-catalyzed and chymotrypsin-catalyzed coupling reactions was less strict
and substantial reaction yield resulted in the reaction involving
D-acyl acceptor
(14, 15).

66
ITOH ET AL.
F
IG. 4. Time course of the peptide coupling reactions. Reactions were carried out in 50 m
M
MOPS
(᭺),
. In the absence of trypsin the same reactions
-Leu-OAm as acyl donor (ϫ).
(pH 8.0) containing 50% DMSO at 25ЊC. Ala-pNA, 20 m
or Boc- -Leu, 1 m (᭹) in the presence of trypsin, 5 Ȑ
were carried out using Boc-Leu-OAm or Boc-
M was reacted with Boc-Leu-OAm, 1 mM
D
M
M
D
A variety of attempts have been carried out to overcome the shortcomings of
enzymatic peptide synthesis. A narrow substrate specificity including -specificity
restricts its wide application to peptide synthesis. Margolin et al. (16) reported that
-specificity of subtilisin for acyl donor relaxed in anhydrous organic solvent. It
was reported that coupling product was obtained in good yield from -acyl donor
L
L
D
in anhydrous tert-amyl alcohol. The method, however, is inadequate for a general
procedure of peptide synthesis. The method requires use of high concentrations of
acyl donor, acceptor, and subtilisin as well.
Another drawback of the enzymatic method is an undesirable proteolysis of
growing polypeptide chain. Synthesis by use of nonprotease, lipase, was proposed
for this solution (17–19). The method also requires use of a high concentration of
the enzyme. The use of chemically modified enzymes was another approach to
prevent undesirable proteolysis of the growing peptide chain. Thiolsubtilisin (20),
methylchymotrypsin (21), and methylsubtilisin (22) are modified proteases in which
the active site serine or histidine residue is modified. All the modified enzymes
were analyzed to retain the catalytic activity for peptide coupling to some extent.
In these instances, substantial improvement of the ratio of peptide coupling over
hydrolysis resulted, though the enzymatic activity decreased greatly on the basis of
the modification (21).
It should be noticed that in our case protease itself was used as a catalyst and
no hydrolysate originating from the coupling product was included in the reaction
mixture. This is another characteristic of the present method in addition to its
remarkable feature of being free from the narrow substrate specificity. Inverse
substrates exhibit strong binding affinity toward trypsin, so that the substrates

SUBSTRATE FOR ENZYMATIC PEPTIDE SYNTHESIS
67
facilitate the formation of the coupling product. Furthermore, the resulting peptides
lacking the site-specific group are no longer involved in its reverse process, thus
hydrolysis of the resulting peptides are retarded.
It is necessary to compare our result with the finding of Schellenberger et al. (4)
who noticed the versatility of inverse substrate as a tool for peptide synthesis and
attempted the use of p-guanidinophenyl ester. Schellenberger’s work, however, did
not include any information on the reaction rate. Therefore the direct comparison
of the practical applicabilities of both methods is inadequate. It was noticed that
the experiments were carried out using much higher concentrations of enzyme and
acyl acceptor than those of ours, 10-fold and 3-fold, respectively. The paper mainly
concerned determination of ‘‘partition constant,’’ the rate ratio between hydrolysis
and coupling. Calculation of the partition constants was carried out for our case.
The values were determined to be 8.6–5 m
M
for the typical examples in Table 3
(for cases which afford 70–80% yield), and these values were 10-times smaller than
those of Schellenberger et al. (4); i.e., ours are preferable for peptide synthesis.
This is the reason why in our case a low concentration of acyl acceptor (one-third
of that used by Schellenberger et al.) is applied to the coupling. Thus it may be
concluded that p-amidinophenyl esters are much more suitable as a tool for peptide
synthesis. One possible reason for this would be the difference in the acylation
efficiencies of their respective substrates. An adequate explanation for this has to
await detailed kinetic analysis and the study is now in progress.
ACKNOWLEDGMENTS
This work was supported in part by grants from The Akiyama Foundation and The Fugaku Trust
for Medicinal Research.
REFERENCES
1. SCHELLENBERGER, V., AND
J
AKUBKE, H.-D. (1991) Angew. Chem. Int. Ed. Engl. 30, 1437–1449.
2. WONG, C.-H. (1989) Science 244, 1145–1152.
3. TANIZAWA, K., KASABA, Y., AND ANAOKA, Y. (1977) J. Am. Chem. Soc. 99, 4485–4488.
K
4. SCHELLENBERGER, V., JAKUBKE, H.-D., ZAPEVALOVA, N. P., AND
MITIN, Y. V. (1991) Biotechnol.
Bioeng. 38, 104–108.
5. FUJIOKA, Y., TANIZAWA, K., NAKAYAMA, H. AND
1905.
KANAOKA, Y. (1980) Chem. Pharm. Bull. 28, 1899–
6. OYAMADA, H., SAITO, T., INABA, S., AND
U
EKI, M. (1991) Bull. Chem. Soc. Jpn. 64, 1422–1424.
ALIS, F. (1976) Eur. J. Biochem. 69, 1–13.
WINTERSBERGER, E. (1962) Hoppe-Seyler’s Z. Physiol. Chem.
7. KASAFIREK, E., FRIC, P., SLABY, J., AND
8. TUPPY, H., WIESBAUER, U., AND
M
329, 278–288.
9. HIXSON, H. F., J
10. CHASE, T., J
11. BIETH, J., AND
R
., AND
., AND
EREMATH, C. G. (1970) Biochem. Biophys. Res. Commun. 53, 383–390.
N
ISHIKAWA, A. H. (1973) Arch. Biochem. Biophys. 154, 501–509.
R
SHAW, E. (1967) Biochem. Biophys. Res. Commun. 29, 508–514.
W
12. OKADA, Y., TSUDA, Y., HIRATA, A., NAGAMATSU, Y., AND
Bull. 31, 4060–4068.
OKAMOTO, U. (1982) Chem. Pharm.
13. GUTFREUND, H. (1955) Trans. Faraday Soc. 51, 441–446.
14. OKA, T., AND
MORIHARA, K. (1977) J. Biochem. 82, 1055–1062.

68
ITOH ET AL.
15. WEAST, J. B., AND
W
ONG, C.-H. (1986) J. Org. Chem. 51, 2728–2735.
16. MARGOLIN, A. L., TAI, D. F., AND
KLIBANOV, A. M. (1987) J. Am. Chem. Soc. 109, 7885–7887.
17. WEST, J. B., HENNEN, W., LALONDE, J. L., BIBBS, J. A., ZHONG, Z., MEYER, E. F., JR., AND WONG,
C.-H. (1990) J. Am. Chem. Soc. 112, 5313–5310.
18. GARDOSSI, L., BIANCHI, D., AND
KLIBANOV, A. M. (1991) J. Am. Chem. Soc. 113, 6328–6329.
19. KAWASHIRO, K., KAISO, K., MINATO, D., SUGIYAMA, S., AND
HAYASHI, H. (1993) Tetrahedron
49, 4541–4548.
20. NAKATUKA, T., SASAKI, T., AND
KAISER, E. T. (1987) J. Am. Chem. Soc. 109, 3808–3810.
21. WEST, J. B., SHOLTEN, J., STOLOWICH, N. J., HOGG, J. L., WEST, A. I., AND
WONG, C.-H. (1988) J.
Am. Chem. Soc. 110, 3709–3710.
22. ZHONG, Z., BIBBS, J. A., YUAN, W., AND
WONG, C.-H. (1991) J. Am. Chem. Soc. 113, 2259–2263.