The α-ketoamide group: A new motif for the elicitation of catalytic antibodies for acyl-transfer reactions

Article abstract of DOI:10.1039/a902475a

An α-ketoamide hapten elicited a monoclonal antibody that catalyzed the hydrolyses of esters and carbamates via either a covalent intermediate or direct addition of water.,.

Full text of DOI:10.1039/a902475a

View Article Online / Journal Homepage / Table of Contents for this issue
The á-ketoamide group: a new motif for the elicitation of catalytic
antibodies for acyl-transfer reactions
Matthew J. Taylor, Jari T. Yli-Kauhaluoma, Jon A. Ashley, Peter Wirsching,
Richard A. Lerner and Kim D. Janda*
Departments of Chemistry and Molecular Biology, The Scripps Research Institute and
the Skaggs Institute for Chemical Biology, 10550 N. Torrey Pines Road, La Jolla, California
9
2037, USA. E-mail: kdjanda@scripps.edu
Received (in Cambridge) 29th January 1999, Accepted 29th March 1999
An á-ketoamide hapten elicited a monoclonal antibody
that catalyzed the hydrolyses of esters and carbamates via
either a covalent intermediate or direct addition of water.
O
O
*
N
R
O
N
O
OR
O
The most successful haptens for the induction of catalytic
O
antibodies with hydrolytic activity have been transition-state
N
N
1
H
analogues based on phosphonate structures. According to
hypothesis and design, the phosphonate approach programmed
the direct addition of a water molecule to appropriate sub-
strates. However, in some cases, antibodies with nucleophilic
side chains used for covalent catalysis were serendipitously
H
1
R = CH3
2 R = CH3 * = 13C
3 R = (CH ) CO H
4
R = p-nitrophenyl
2
3
2
5 R = CH3
2
O
elicited by phosphonate haptens. Recently, a rational approach
O
to the design of haptens for eliciting covalent catalysis
was developed and coined “reactive immunization”. The
O
O
3
N
H
O
methodology enabled the development of catalytic antibodies
for enantioselective ester hydrolyses with high catalytic
O
6
N
NO2
4
H
7
proficiencies.
The α-ketoamide-containing natural product cyclotheon-
amide A inactivated the serine protease α-thrombin through
formation of a hemiketal adduct between the hydroxy group of
H
O
N
O
5
the catalytic serine and the ketone of the inhibitor. Protease
O
N
NO2
and esterase inactivation via such a mechanism was also
H
6
8
observed using various trifluoromethyl ketones. Reactive
immunization using an α-ketoamide should elicit antibodies
capable of operating by covalent catalysis through a discrete
acyl-antibody intermediate. However, an additional factor
generally encountered in the α-ketoamide structure is a ~90Њ
twist angle about the carbonyl bonds that might mimic charac-
Fig. 1
O
HO X-Ab
NR2
Ab-XH
NR2
7
teristics of a tetrahedral transition state. This feature could
O
O
make contributions to catalysis of covalent intermediate
formation or direct hydrolysis.
1) selection of antibody nucleophile during
immunization (reactive immunization)
8
We previously determined the X-ray structure of 1 that
revealed the presence of the keto form in the solid state with a
dihedral angle of 92.6Њ about the two carbonyls similar to the
2
) antibody nucleophile used for
covalent catalysis of substrates
7,9
‡
angle observed in other α-ketoamides. Under physiological
O
HO X-Ab
OR
O
Ab-XH
conditions (deutero-PBS, pH 7.4, 10% DMSO-d for solubil-
6
13
OR
X-Ab
ity), [ C]-labeled hapten analog 2 was found to exist pre-
dominantly (>97%) in the keto form (ketone, δ 192.7 ppm;
hydrate, δ 94.5 ppm). According to the precedent for enzymatic
inactivation, hapten 3 should be suitable for eliciting antibody
catalysts for acyl-transfer reactions via a mechanism operating
through a covalent intermediate.
H2O
O
+
Ab-XH
OH
A panel of 22 mAbs (monoclonal antibodies) was derived
from immunization of mice with hapten 3 coupled to KLH
Scheme 1
(
keyhole limpet hemocyanin). Antibodies were screened†
for their ability to catalyze the hydrolysis of esters 4 and 6.
Two antibodies were found to accelerate ester hydrolysis and
one antibody, mAb 3H11, was studied in detail. BIAcore
kinetic profile having a burst of 4-isobutyramidophenol release
followed by a slower steady-state phase (k /k = 10). For
cat uncat
mechanisms invoking an acyl intermediate with deacylation as
the rate-determining step, addition of a good nucleophile
(
Pharmacia) analysis using hapten 3 tethered with an amine-
9
10
linker was used to determine k and koff that provided a value
should increase the rate of deacylation and therefore the kcat.
on
of K = 10 nM and indicated excellent recognition and affinity
Consistent with this, addition of hydroxylamine increased the
steady-state rate of the phenol release three-fold. The results
suggested that reactive immunization had positioned a nucleo-
phile appropriate for acylation reactions in the combining site
of the antibody.
d
of the α-ketoamide structure. The position of the substrate
carbonyl group and other chemical features were varied and
this afforded other substrates 5, 7 and 8.
The acetate ester 6 in the presence of mAb 3H11 produced a
J. Chem. Soc., Perkin Trans. 1, 1999, 1133–1134
1133

View Article Online
Initially, the reaction of mAb 3H11 with p-nitrophenyl ester 4
to be determined, here the elimination mechanism might be
favored since hapten design elements to dictate otherwise were
not incorporated. Hence, the active-site nucleophile participat-
ing in acylation with esters 4 and 6 might act as a general base
for carbamate hydrolysis.
In the light of the known structure and reactivity of α-keto-
amides, the data supported a reactive immunization response to
the hapten along with possible modes of transition-state stabil-
ization. However, further studies will be required to unequivo-
cally address this issue. Given the successful results, a new
model is now available for obtaining antibody catalysts for ester
and carbamate hydrolyses, and for the exploration of other
acyl-transfer reactions.
(
5 µM mAb, 100 µM substrate) was observed to produce a burst
Ϫ1
of p-nitrophenol (vobs = 1.0 µM min ) with a corresponding
Ϫ1
lag in acid production (vobs = 0.02 µM min ). A complete
Michaelis–Menten analysis was not feasible due to substrate
solubility limitations, but the initial rates in the concentration
range studied were linear with antibody concentration and the
reaction was inhibited by hapten analog 1. Again, the kinetic
profile suggested the presence of an acyl-antibody in which
breakdown of the intermediate was the rate-determining step.
However, HPLC experiments indicated a discrepancy in the
amount of p-nitrophenol and 4-isobutyramidobenzoic acid
produced in the antibody-catalyzed reaction. The absence of
one-half equivalent of acid suggested either covalent or tight
noncovalent binding of the acyl portion to the antibody.
Quenching the reaction under strongly denaturing conditions
Acknowledgements
(
6 M guanidinium hydrochloride) gave no increase in the
Mrs Marie-Rose Benitez for antibody production, Mrs Anja
Salakari for technical assistance, and Shenlan Mao for BIAcore
analysis. J. Y.-K. thanks TEKES (Technology Development
Centre, Finland). The Skaggs Institute for Chemical Biology,
the National Institute of Health (GM-43858) and The Scripps
Research Institute.
amount of acid so that the data were in accord with covalent,
irreversible acylation of the antibody.
The above “suicide” type of inhibition had been observed
11
previously for esterolytic antibody catalysts. Essentially, the
ester 4 was a mechanism-based inactivator of mAb 3H11 where
antibody acylation (kacyl) was in competition with substrate
hydrolysis (k ). A qualitative comparison of the rates of
p-nitrophenol and acid production indicated that antibody
cat
Notes and references
acylation was the predominant pathway. A comparison of the
† Determined at 22 ЊC in 50 mM phosphate–100 mM NaCl, pH 8.0,
with 5% DMSO as cosolvent for 5 and 6, and 4% DMF, 1% dioxane,
Ϫ5
estimated k and kcat for 4 to the measured kuncat (6.7 × 10
acyl
Ϫ1
1
‡
% Tween-80 for 4, 7 and 8.
min ) provided a lower limit for the rate accelerations of 1490
BIAcore analysis is a system for real time bimolecular interaction
and 30, respectively. The dual activity observed could result
from alternative binding modes as determined by the two
twisted atropisomers possible for hapten 3, yielding different
mechanistic pathways, or competition in a single binding mode
between addition of an antibody nucleophile and of water.
Interestingly, the methyl ester 5 with the carbonyl and
analysis (BIA) using surface plasmon resonance technology.
1
2
P. G. Schultz and R. A. Lerner, Science, 1995, 269, 1835; B. J. Lavey
and K. D. Janda, Antibody Expression and Engineering, ACS
Symposium Series 604, 1995, ch. 10.
J. Guo, W. Huang and T. S. Scanlan, J. Am. Chem. Soc., 1994,
116, 6062; M. T. Martin, A. D. Napper, P. G. Schultz and A. R.
Rees, Biochemistry, 1991, 30, 9757; P. Wirsching, J. A. Ashley, S. J.
Benkovic, K. D. Janda and R. A. Lerner, Science, 1991, 252, 680.
P. Wirsching, J. A. Ashley, C.-H. L. Lo, K. D. Janda and R. A.
Lerner, Science, 1995, 270, 1775.
4
-isobutyramidophenyl group located identically to that of 4
Ϫ4 Ϫ1
was also a substrate for mAb 3H11 (kcat = 2.0 × 10 min ,
K = 280 µM, k /k = 100). Notably, the antibody was not
m
cat uncat
3
4
5
inactivated during reaction with 5, so that now simple
hydrolysis was the sole mechanistic pathway. Similarly, in the
reaction with the p-nitrophenol ester 7, no burst of p-nitro-
phenol was detected and equimolar amounts of acid and
p-nitrophenol were produced. Hence, 7 did not irreversibly
acylate the antibody as for 4, although the possibility of an acyl
intermediate could not be excluded. The varied and unusual
reactivity of the closely related activated esters 4 and 7 demon-
strated the subtle differences in kinetic behavior afforded in an
antibody elicited by the α-ketoamide hapten.
C.-H. L. Lo, P. Wentworth, Jr., K. W. Jung, J. Yoon, J. A. Ashley and
K. D. Janda, J. Am. Chem. Soc., 1997, 119, 10251.
S. D. Lewis, B. J. Lucas, S. F. Brady, J. T. Sisko, K. J. Cutrona, P. E. J.
Sanderson, R. M. Freidinger, S.-S. Mao, S. J. Gardell and J. A.
Shafer, J. Biol. Chem., 1998, 273, 4843; B. E. Maryanoff, X. Qiu,
K. P. Padmanabhan, A. Tulinsky, H. R. Almond, Jr., P. Andrade-
Gordon, M. N. Greco, J. A. Kauffman, K. C. Nicolaou, A. Liu,
P. H. Brungs and N. Fusetani, Proc. Natl. Acad. Sci. USA, 1993,
9
0, 8048.
6
K. Brady, A. Wei, D. Ringe and R. H. Abeles, Biochemistry, 1990,
29, 7600; K. N. Allen and R. H. Abeles, Biochemistry, 1989, 28,
The aryl carbamate 8 was a substrate for antibody mAb
Ϫ1
3
H11 (kcat = 0.20 min , K = 120 µM, k /k
= 480). The
m
cat uncat
8
466; B. Imperiali and R. H. Abeles, Biochemistry, 1986, 25, 3760.
reaction yielded multiple turnovers, no indication of product
7
8
R. D. Bach, I. Mintcheva, W. J. Kronenberg and H. B. Schlegel,
J. Org. Chem., 1993, 58, 6135.
J. T. Yli-Kauhaluoma and K. D. Janda, Bioorg. Med. Chem., 1994,
2, 521.
inhibition and was competitively inhibited by 1 (K = 0.5 µM).
i
No burst was observed and the rates of 4-isobutyramidophenol
and 4-nitroaniline formation were identical by HPLC analysis
which indicated production of the phenol was in the steady-
state. However, both the log Vmax and log Vmax/Km versus pH
9 M. J. Taylor, T. Z. Hoffman, J. T. Yli-Kauhaluoma, R. A. Lerner
and K. D. Janda, J. Am. Chem. Soc., 1998, 120, 12783.
1
1
1
1
0 A. Fersht, Enzyme Structure and Mechanism, W. H. Freeman, NY,
profiles showed an apparent pK = 7.5 where the deprotonated
a
1
985, 2nd edn.
1 T. S. Angeles and M. T. Martin, Biochem. Biophys. Res. Commun.,
993, 197, 696.
2 A. F. Hegarty and L. N. Frost, J. Chem. Soc., Perkin Trans. 2, 1973,
719.
form of an amino acid residue was catalytically active. The
energetically favored, uncatalyzed hydrolytic pathway for aryl
carbamates is believed to proceed by an elimination (E1cB)
mechanism forming an alcohol and isocyanate in which the
latter then reacts with water and decarboxylates to form the
1
1
3 P. Wentworth, Jr., A. Datta, S. Smith, A. Marshall, L. J. Partridge
12
amine and carbon dioxide. The disfavored pathway is an
addition–elimination (BAC2) mechanism proceeding through
a tetrahedral intermediate. Antibodies have been shown to
and G. M. Blackburn, J. Am. Chem. Soc., 1997, 119, 2315.
13
catalyze the highly disfavored BAC2 process. While it remains
Communication 9/02475A
1
134
J. Chem. Soc., Perkin Trans. 1, 1999, 1133–1134