6
510
J. Am. Chem. Soc. 1996, 118, 6510-6511
P1-S1 Interactions Control the Enantioselectivity
and Hydrolytic Activity of the Norleucine
Phenylesterase Catalytic Antibody 17E8
Herschel Wade and Thomas S. Scanlan*
Departments of Pharmaceutical Chemistry and
Cellular and Molecular Pharmacology
UniVersity of California
San Francisco, California 94143-0446
ReceiVed February 20, 1996
The catalytic antibody 17E8 is an esterase that catalyzes the
1
hydrolysis of N-acyl amino acid phenyl esters. This catalytic
antibody was generated against the norleucine phenyl phos-
phonate analog 1, and one of the best substrates for the enzyme
is N-formyl norleucine phenyl ester 2 (Figure 1). The crystal
structure of the 17E8 Fab complexed with the phosphonate
analog 1 has been solved to 2.5 Å resolution.2 This structure
shows that 17E8 has similar active site features to natural triad-
based hydrolases. In addition, the structure shows that 1 is
buried deeply in the antigen combining site, and there are
separate and well-defined binding pockets for the phenyl group
and n-butyl side chain of 1. The system is thus a simple model
of proteolytic enzymes containing a single, isolated S-subsite
pocket for recognition of a P1 hydrophobic side chain of the
amino acid substrate.
Figure 1. Schematic of 17E8-1 active site structure and hydrolysis
reaction catalyzed by 17E8.
Table 1. Enantioselectivity of 17E8 Catalysis
CR
substrate
configuration [R] a
D
k
cat (s-1)b
K
M
(mM)b
N-for-Nle-OPh (2) S
-14.0° 2.1 ( 0.1 0.10 ( 0.01
+13.7°
R
racemic
0
2.0 ( 0.1 0.18 ( 0.04
N-for-Ala-OPh (3) S
-36.0° 0.9 ( 0.1 11 ( 4
+ 35.6°
R
racemic
0
0
0
0
0.8 ( 0.1 9 ( 2
N-for-Aib-OPh (4) nonstereogenic
N-for-Gly-OPh (5) nonstereogenic
In our initial kinetic studies with 17E8, we observed that the
hydrolysis reaction was catalyzed in an enantioselective man-
phenyl hexanoatec nonstereogenic
2.2 ( 0.6 0.6 ( 0.2
a
1
Optical activity measurements were performed on a Perkin-Elmer
41 polarimeter. All samples were prepared as dichloromethane
ner. The synthesis of substrate 2 resulted in CR racemization,
2
making it unsuitable for preparing enantiomerically pure
substrates with either R or S CR configuration. However, we
were able to obtain enantiomerically enriched CR R and S
samples and found that the S-enriched substrate afforded a higher
specific activity than the R-enriched substrate by the expected
value assuming complete S-enantioselectivity for 17E8-catalyzed
hydrolysis. Additional support for the hydrolytic enantioselec-
tivity was obtained from the 17E8-1 crystal structure which
showed that the enantiomer of 1 with the same CR configuration
as the S substrate was bound in the active site even though
racemic 1 was used in the crystallization experiment.2
An alternative synthetic route involving esterification of
N-Boc amino acids was developed to prepare enantiomerically
pure 17E8 substrates.3 We now report results that unambigu-
ously confirm the 17E8 S-enantioselectivity of hydrolysis and
show that enantioselectivity and catalytic activity are controlled
by hydrophobic interactions between the substrate P1 side chain
and enzyme S1 pocket.
The enantioselectivity of the 17E8-catalyzed hydrolysis
reaction was examined using the S and R substrate as well as
the racemic mixture (Table 1). The (S)-2 and racemic 2 are
effective substrates for 17E8, whereas the hydrolysis of (R)-2
was not catalyzed by 17E8. A steady-state kinetic analysis was
performed with (S)-2 and racemic 2 to obtain the kinetic
constants kcat and KM.4 The magnitudes of these constants are
consistent with an S-enantioselective mode of hydrolysis: the
solutions, and optical density measurements were made at 25 °C.
Activity assays were performed in 50 mM borate, 150 mM NaCl, pH
b
8.7 at 24.5 °C. Steady-state kinetic constants were obtained from
nonlinear fits of V vs [S] data to the Michaelis-Menten equation using
the Kaliedagraph program. c The kinetic data for this substrate were
obtained at pH 9.5. For comparison, the racemate of the natural
-
1
M
substrate 2 gives a kcat of 3.7 s and K of 0.4 mM at pH 9.5.
same kcat value is observed for (S)-2 and racemic 2, but the KM
value for racemic 2 is approximately twice that of (S)-2. Taken
together, these results unambiguously establish that 17E8
catalyzes the enantioselective hydrolysis of norleucine phenyl
substrates that have the S configuration at CR, and the
enantiomeric R substrates are not processed by the enzyme.
The serine proteases show similar S-enantioselectivity of
5
hydrolysis. A simple explanation to account for the enantio-
selectivity of serine proteases is a three-locus model where the
three non-hydrogen groups attached to the stereogenic R-carbon
make stabilizing interactions with specifically oriented active
6
site groups in the hydrolysis transition state. In the case of
chymotrypsin, an H-bond between the NH of the R-N-acyl group
and an active site acceptor has been identified as a critical
7
interaction for controlling enantioselectivity. Replacement of
the N-acetyl group with a hydrogen atom in phenylalanine ester
substrates leads to a 4000-fold decrease in kcat and a 9000-fold
decrease in kcat/KM. In contrast, the analogous experiment with
1
7E8 shows that the substrate N-acyl group is not particularly
(
1) Guo, J.; Huang, W.; Scanlan, T. S. J. Am. Chem. Soc. 1994, 116,
062.
2) Zhou, G. W.; Guo, J.; Huang, W.; Fletterick, R. J.; Scanlan, T. S.
Science 1994, 265, 1059.
3) Castro, B.; Evin, G.; Selve, C.; Seyer, R. Synthesis 1977, 413. Both
important in catalysis. Phenyl hexanoate, which contains a
hydrogen atom in place of the N-formyl group, gives values
for kcat and KM similar to those of the norleucine substrate 2
6
(
(Table 1), suggesting that the enantioselectivity of 17E8 is not
(
the R and S enantiomers of the norleucine (2) and alanine (3) were prepared
by this route. The R-amino isobutyric acid (Aib, 4) and glycine (5) substrates,
which are nonstereogenic at CR, were prepared by direct esterification of
the corresponding N-formyl amino acid. Phenyl hexanoate, a substrate that
possesses the n-butyl P1 side chain but does not contain the R-formylamino
substituent, was prepared by esterification of hexanoic acid.
controlled by an H-bonding interaction between the R-N-acyl
group of substrate and an active site acceptor.
The alanine, Aib, and glycine substrates (3, 4, and 5) were
examined to probe the contribution of the substrate P1 side chain
(
4) The kinetic constants kcat and KM were obtained from a nonlinear fit
(5) Stroud, R. M. Sci. Am. 1974, 231, 74.
(6) Ingles, D. W.; Knowles, J. R. Biochem. J. 1967, 104, 369.
(7) Ingles, D. W.; Knowles, J. R. Biochem. J. 1968, 108, 561.
of the V vs [S] data to the Michaelis-Menten equation using the
Kaleidagraph program.
S0002-7863(96)00532-X CCC: $12.00 © 1996 American Chemical Society