вход по аккаунту


Catalytic Antibodies A New Class of Transition-State Analogues Used to Elicit Hydrolytic Antibodies.

код для вставкиСкачать
Catalytic Antibodies: A New Class of Transition-State Analogues
Used to Elicit Hydrolytic Antibodies
By Kevan M. Shokat, Marcia K. KO, Thomas S. Scanlan, Lynn Kochersperger,
Shirlee Yonkovich, Suvit Thaisrivongs, and Peter G. Schultz"
The design and generation of selective catalysts is an important aim of chemists and biologists.
A number of successful strategies have emerged, including the synthesis and derivatization of
synthetic hosts, the chemical modification and site-directed mutagenesis of enzymes, and the
attenuation of natural enzyme activities in organic solvents. Since 1986 several laboratories
have exploited the immune system to generate selective catalysts capable of catalyzing a wide
range of chemical transformations. These include acyl transfer, p-elimination, carbon-carbon
bond-forming, carbon -carbon bond-cleaving, porphyrin metalation, peroxidation, and redox
reactions. The variety and number of transformations catalyzed by antibodies in this short
period of time is testament to the versatility and power of the method in generating selective
catalysts for applications in chemistry, biology, and medicine. Here we report the use of a new
class of uncharged transition-state analogues for generating antibodies capable of catalyzing
ester and carbonate hydrolysis. These antibodies are compared to those raised against tetrahedral phosphate and phosphonate transition-state analogues
1. Introduction
The immune system has the potential to generate more
than 10" unique antibodies."] The ability to generate this
degree of diversity and to select molecules with the desired
biological properties from this enormous pool is unique to
biological systems. The mechanism by which the immune
system generates and screens this huge collection of antibodies is clonal selection. Each antibody molecule of the primary
immunological repertoire is stored on the surface of cells
called lymphocytes. When a cell with a unique antibody expressed on its surface encounters an antigen, it undergoes
division and differentiation and starts to produce soluble
antibody molecules. At the same time, the genes encoding
the binding site of the antibody undergo mutation, leading to
an increase in affinity for the hapten. Antibodies can be
selectively generated against virtually any biopolymer, small
synthetic molecule, o r natural product of interest. With the
ability to generate a binding site for virtually any potential
substrate, the challenge becomes one of designing chemical
strategies to introduce catalytic activity into these macromolecules. Enzymes utilize a variety of strategies simultaneously, including transition-state stabilization, general
acid-general base catalysis, cofactors, and proximity effects
to catalyze reactions. Each of these strategies has been successfully applied to the generation of antibody catalysts. Sev['I
Prof. P. G. Schultz, K . M. Shokat, M. K. KO, Dr. T. S . Scanlan
Department of Chemistry, University of California
Berkeley, CA 94720 (USA)
Dr. S. Thaisrivongs
Cardiovascular Diseases Research
The Upjohn Company
Dr. L. Kochersperger. S. Yonkovich
Department of Bioorganic Chemistry
Affymdx Research Institute
This work was supported by the Director, Office of Energy Research.
Office General Life Sciences, Structural Biology Division of the U. S.
Department of Energy (DE-AC03-76SF00098). and the Damon RunyonWalter Winchell Cancer Research Fund Fellowship (DRG-1016) (TS.S.1.
We thank Dr. Jejj Jacobs for helpful discussions in the early stages of this
1296 :C:.
VCH Verlugsgesellschuft mbH. 0-6940 kt't'inhKim, 1990
era1 comprehensive reviews on the design and characterization of catalytic antibodies have been written.[2-51
The first examples of antibody-catalyzed reactions were
based on the notion of transition-state stabilization. Jenck's
first proposed over twenty years ago that antibodies raised
against a transition-state analogue should bind the transition state of a reaction selectively over the substrate and
thereby catalyze the reaction (Fig.
In Fact, many en-
a u n c a t a l y z e d reaction
catalyzed reaction
Reaction Coordinate
Fig. 1. Reaction coordinate diagram for an antibody-catalyzed and an uncatalyzed reaction.
zymes have evolved to provide an active site that is sterically
and electronically complementary to the rate-determining
transition state.[7.*I Hydrolytic enzymes such as the zinc
peptidase thermolysin function in part by selectively stabilizing the negatively charged tetrahedral intermediate in peptide hydrolysis.[g1Analogues of the transition-state configuration have been shown to be potent enzyme inhibitors. Such
transition-state analogues seek to take advantage of the
complementarity of the enzyme to the transition-state configuration to achieve high binding affinities. One of the best
inhibitors of thermolysin is phosphonamidate 1, where a tetrahedral phosphorus is substituted for the scissile carbonyl
Angen,. Chem. I n t . Ed.
Enzl. 29 (1990) 1296-1303
carbon in the substrate.1g1Because a typical P - 0 single bond
is 10- 15 % longer than a C - 0 single bond, these analogues
are quite similar to the tetrahedral transition-state structure
which has longer, partially formed bonds.
have been characterized and several examples are reviewed
2.1. Phosphorylcholine-Binding Antibodies MOPC 167
and T15
Phosphonamidate 1 was shown to be a transition-state
analogue inhibitor by the synthesis of a series of analogues
of inhibitor 1 and the corresponding substrates with substitutions at the carboxy-terminal amino acid and measurement of K, and KM/k,,,values, respectively. The K, values
correlated not with KMvalues but with KM/k,,,values, suggesting that the factors that influenced binding of inhibitors
similarly affected binding of the transition state relative to
the substrate.
Another important class of peptidase inhibitors consists of
uncharged tetrahedral mimics of the transition state, including hydrated ketones,["] fluoroketones," and secondary
alcohols.['*] One such molecule is pepstatin, an inhibitor of
the aspartic protease
The key feature of this inhibitor is the novel amino acid statine, which contains an uncharged secondary alcohol substituted for the carbonyl of
the scissile amide bond in the substrate. An analysis similar
to that performed in the case of thermolysin compared a
series of inhibitor K, values and substrate KM/k,,,values for
pepsin, but yielded inconclusive results due to the fact that
two distinct rate constants contribute to the measured k,,,
values for this enzyme.["] The fact that pepstatin Kl values
do not correlate with KMor with KM/k,,,values suggests that
these inhibitors do not function exclusively as either transition-state analogue inhibitors or as multisubstrate inhibitors. Since statine-like compounds at least in part resemble the tetrahedral transition state for hydrolysis of an amide
bond, we reasoned that antibodies raised against a statinebased inhibitor might catalyze the hydrolysis of the corresponding ester or amide substrates. Antibodies elicited
against alcohol 2 do indeed catalyze the hydrolysis of paranitrophenyl acetate and para-nitrophenyl methyl carbonate.
This study extends the transition-state stabilization strategy
for generating catalytic antibodies to a new class of peptidase
inhibitors. The kinetic properties of antibodies elicited
toward statine-like inhibitor 2 are comparable with those
of the more numerous phosphonate-specific antibodies.
2. Phosphate- and Phosphonate-Specific Antibodies
A large number of antibodies elicited against compounds
that contain charged tetrahedral phosphorus and function as
analogues of the transition states in hydrolytic reactions
A n g e x Chem. I n i . Ed. Engl. 2Y (1990) 1296-1303
Two of the first catalytic antibodies characterized, MOPC
167 and T15, were not elicited against a synthetic immunogen but were members of a well-studied class of antibodies
binding phosphorylcholine (PC). These antibodies had been
previously characterized with regard to ligand-binding
' 1 kinetics, and biomolecular structure." 1'
Both are specific for para-nitrophenyl phosphorylcholine (3), which is an analogue of the transition-state 5 in
the hydrolysis of para-nitrophenyl phosphorylcholine carbonate (4). Indeed, MOPC 167 and TI 5 catalyze the hydrolysis of 4 with kinetics obeying the Michaelis-Menten rate
expression.[". '*I The Michaelis constant KMis equal to the
concentration of substrate, [S], that produces one-half the
maximal catalyzed rate, k,,, [E,], where [E,] is the total enzyme concentration and k,,, is the unimolecular rate constant for the catalytic step. The k,,, and KMvalues for MOPC
167 and TI5 are k,,, = 0.4min-', KM= 208 PM and
k,,, = 0.32 min- I , K M= 708 p ~ respectively.
The highly
specific nature of antibody-antigen recognition leads to
highly specific catalysts. For example, para-nitrophenyl ethyl carbonate is not hydrolyzed to any appreciable extent by
MOPC 167. The inhibition constants measured for MOPC
167 and T15 were 5 p~ and 55 p ~ respectively.
pH 6.25 and 8.0, both antibody-catalyzed reactions were
shown to be first order with respect to hydroxide ion. Comparison of the OH'-catalyzed rate k,,[OH@] (corrected for
intramolecular assistance by the tetraalkylammonium ion)
with the antibody-catalyzed rate k,,, affords rate accelerations of 1 1 500 for MOPC 167 and 9200 for T15. However,
the observed differential binding energy of the transitionstate analogue over the substrate ( K M / K ldoes
not account
for the entire rate acceleration, indicating that other Factors
must be contributing to the observed rate acceleration.
2.2. X-Ray Crystal Structure of the PhosphorylcholineSpecific Antibody McPC603
The X-ray crystal structure of McPC603, which is an antibody highly homologous to T15, has been solved.["I This
crystal structure makes possible direct identification of the
combining-site residues responsible for catalysis (Fig. 2) .I* 91
Phosphorylcholine is bound in the cavity of McPC603 with
the choline group deep in the interior and the phosphate
toward the exterior, in contact with aqueous solvent. The
heavy-chain residues, Tyr 33H and Arg 52H, which are invariant in all of the phosphorylcholine binding antibodies
sequenced to date, bind the phosphate via hydrogen bonding
and electrostatic interactions with the phosphoryl oxygen
atoms. The X-ray structure demonstrates that the combining
Glu 35H P O
Fig. 2.Schematic representation of binding interactions between phosphorylcholine and amino acid side chains in the combining site of McPC603.
site of McPC603 is both sterically and electronically complementary to the tetrahedral, negatively charged phosphate
moiety of phosphorylcholine. Inasmuch as this tetrahedral
phosphate mimics the transition state for OH@-catalyzed
hydrolysis of 4, the phosphorylcholine antibodies should be
capable of polarizing the bound carbonate for attack of hydroxide ion in the rate-determining step. Because the
ground-state structure of 4 differs substantially from the
transition-state configuration for hydrolysis, 5, the differential binding affinity of the antibody to these two species is
reflected in a lowered free energy of activation for the reaction.
was introduced into the myeloma cells by electroporation
using an azaserine/hypoxanthine selection. Large quantities
of the mutants (1 -50 mg) were isolated from ascites of immunosuppressed mice injected with the transfected cells.
Tyrosine 33H was mutated to Phe, GIu, Asp, and His in
the study. Without exception, all mutants exhibited only
slight increases in KM for4 compared to wild-type S107.
With the exception of the His mutant, the effect of the Tyr
mutations on the catalytic activity was small. These results
suggest that Tyr 33H is not involved in binding or catalysis
of the reaction being studied. By substituting His for Tyr at
position 33H, a general base is placed in close proximity to
the bound carbonyl carbon of 4 as shown by molecular modeling using the McPC603 crystal structure. Indeed, the
His 33H mutant was shown to accelerate the hydrolysis of 4
by a factor of 6700 compared to methylimidazole free in
solution. Alternatively, it might have been expected that
placement of a general acid in the combining site, such as the
@ or y carboxyl group of Asp or Glu, might also lead to
enhanced catalysis relative to the wild type. The failure of
these substitutions suggests that the formation of the negatively charged transition state in the hydrolysis of 4 is disfavored in the presence of the negatively charged carboxylates,
at neutral pH.
Arginine52H was mutated to Lys, Gin, and Cys. The
Lys 52H mutant showed only slightly lowered catalytic effciency compared to wild type S107. In contrast, the Gln 52H
and Cys 52H mutants showed considerably lowered catalytic
efficiencies, presumably due to the loss of positive charge
necessary for stabilization of the negatively charged oxyanion in the transition state of the reaction. Consistent with this
notion, the Gln 52H and Cys 52H mutants bound 3 much
more weakly than wild-type S107, whereas the mutants
showed little change in binding to the uncharged substrate 4.
The Cys mutation was introduced as a chemical handle to
enable derivatization with cofactors or reporter molecules to
generate semisynthetic catalytic antibodies.[*'. The mutagenesis study of S107 demonstrates the importance of electrostatic interactions in transition-state stabilization and also shows that the catalytic efficiency of a catalytic antibody
can be enhanced by rational design.
2.4. Generation of Antibodies to Phosphonate and
Phosphonamidate Transition-State AnaIogues
2.3. Site-Directed Mutagenesis of the PhosphorylcholineSpecific Antibody S107
A site-directed mutagenesis study of the PC-specific antibody S107, which has the same heavy-chain variable sequence as T15, has further delineated the catalytic roles of
residues Arg 52H and Tyr 33H.["] Both residues were substituted with a variety of amino acids with no charge, with
opposite charge, or with side chains designed to serve specific
catalytic functions such as general acid or general base catalysis. One mutant (Tyr-+His33H) displayed increased catalytic activity over the wild-type antibody. The increase in
activity is attributed to His 33H functioning as a general
The mutant S107 antibodies were expressed in myeloma
cells not expressing endogenous antibody. The mutant DNA
Besides antibodies that fortuitously bind transition-state
analogues for carbonate hydrolysis, catalytic antibodies
have been rationally designed to bind transition-state analogues for specific reactions. The hydrolysis of para-nitrophenyl methyl carbonate (6) proceeds through the tetrahedral negatively charged transition state 7. The corresponding
phosphonate transition-state analogue 8 was synthesized
and coupled to protein carrier keyhole limpet hemocyanin
Small molecules are not immunogenic by themselves and must be covalently attached to a protein in order
to elicit an antibody response. Fourteen distinct monoclonal
antibodies specific for8 were isolated in this way and assayed for catalysis of the hydrolysis of para-nitrophenyl
methyl carbonate (6) and para-nitrophenyl acetate (9). Two
of these antibodies catalyzed the reaction and were inhibited
Angew. Chem. Int. Ed. Engi. 29 (1990) 1296-1303
by free hapten 8. Inhibition by free hapten strongly suggests
that the catalysis observed is taking place in the antibody
combining site. Antibody 48G7-4A1 catalyzed hydrolysis of
ester 9 with k,,, = 7.4 min- and KM= 430 p ~ The
. kinetic
constants for the hydrofysis of the related carbonate 6 were
k,,, = 26.2 min-' and KM = 360 p ~ The
. rate enhancements
over the uncatalyzed rate are 1600 and 7300 for 9 and 6,
respectively. Inhibition studies demonstrated competitive inhibition by the free hapten 8 with K, = 330 nM.
O , N~ O X O /
a pre-steady-state burst of para-nitrophenolate and thus
ruled out the third mechanism involving acyl-enzyme formation. Chemical modification studies revealed the presence of
tyrosine, arginine, and histidine in the combining site of this
a n t i b ~ d y . " ~Interestingly,
antibody S107, which catalyzes
the hydrolysis ofpara-nitrophenyl choline carbonate (4) also
has an arginine and a tyrosine, which were identified by
chemical modification studies. Perhaps in analogy to S107,
arginine serves to stabilize the oxyanion formed in the transition state for ester or carbonate hydrolysis.
The largest rate acceleration by an esterolytic antibody
was achieved using a phosphonate transition-state analogue.
Lerner, Trarnontano et al. isolated twenty antibodies specific
for compound 10 and found five which catalyzed the hydrolysis of ester 12.[251The kinetic constants measured for the
most active antibody were k,,, = 20 s-' and KM = 1.5 mM.
The free hapten 11 binds extremely tightly with K, = 50 nM.
The large difference in affinity of the antibody for the substrate versus the inhibitor accounts for most of the large
rate acceleration achieved by this antibody, k,,,/k,,,,, =
6.25 x 1 0 5 .
0' e ' 0
Several possible mechanisms could account for the catalysis observed : direct hydroxide attack, general base catalysis,
or acyl-antibody formation followed by hydrolysis (Fig. 3).
10, R = (CH,),COON(COCH,),
11, R = CH,
Direct Hydroxide Attack
General Base Catalysts:
Covalent Acyl-Antibody Intermediate
Antibodies have been shown to catalyze the stereospecific
hydrolysis of unactivated esters. Antibodies elicited toward
phosphonates 13and 15were shown to catalyze the hydrolysis of 15 and 16, respectively, in a stereospecific manner. In
both cases, antibodies were elicited against a racemic mixture of the hapten. Of eighteen antibodies specific for 13,
nine catalyzed the hydrolysis of (R)-14 and two catalyzed the
hydrolysis of (s)-14.[261 Two antibodies, one of each
specificity, were further characterized. The rate acceleration
for the hydrolysis of (R)-14 by antibody 2H6 was 80000,
whereas the (S)-lCspecific antibody 21H3 only showed a
modest 1600-fold rate acceleration over the uncatalyzed reaction. The R / S selectivity is greater than 98 %.
Fig. 3. Possible mechanisms for hydrolysis of9 by antibody 48G7-4A1.
Ig = immunoglobulin, p = product.
y f 3
The pH dependence of antibody catalysis was shown to be
first order with respect to hydroxide between pH 7.5 and 9.4,
arguing for direct attack by hydroxide ion in this pH range.
If the general base mechanism were responsible for the catalysis observed, the pH dependence of k,,, would reflect a
dependence on the pK, of a specific residue in the combining
site. Note, however, that a general base could be present but
its p K , cannot lie in the range of 7.5-9.4. Kinetic analyses
using high antibody concentrations were not consistent with
Angen. C'hrm. In1 Ed. En$ 29 (1990) 1296-1303
Of 25 antibodies isolated that were specific for the tripeptide phosphonate 15, 18 accelerated the hydrolysis of the
structural feature of this hapten responsible for induction of
such acid/base catalysis is not apparent because the phosphonamidate N H group is not charged in the immunogen.
3. Statine-based Peptidase Inhibitors
In addition to negatively charged phosphonates, a number
of uncharged transition-state analogues exist for hydrolytic
enzymes. The most well characterized of these is the potent
pepsin inhibitor pepstatin, which contains the novel amino
acid analogue statine (Fig. 4). The measured Kl of pepstatin
corresponding ester substrates.[271It is interesting that an
exclusive preference for antibodies of one specificity was
observed. All 18 antibodies were selective for the D-phenylalanyl isomer of 16. The selectivity for D- over L-phenylalanine at this position was greater than 99.5% for three of the
five antibodies characterized. The modest rate accelerations
of SO-300-fold may be a result of the size of the immunogen.
The tetrahedral phosphonate contributes proportionally less
to the overall binding energy of hapten to antibody than it
does in smaller haptens. Note that substrate 16 also contains
fluorogenic and quenching groups at the amino and carboxy
termini, respectively. These groups allow hydrolysis of the
substrate to be monitored by observing the fluorescence increase which occurs when the fluorescent 2-aminobenzoyl
group is separated from the quenching 4-nitrobenzylamide
group in the reaction. This sensitive assay may allow direct
screening of ELISAi2'I plates or of 1 libraries expressing up
to 1O6 different Fabfragments,'291greatly facilitating the production of catalytic antibodies. These two examples suggest
that excellent stereoselectivities can be achieved by antibodies with large o r small substrates and chiral centers in the
alcohol portion or the acyl group portion of the substrate.
An antibody, NPN43C9, elicited against phosphonamidate 17 was shown to catalyze the hydrolysis of the activated
amide bond in 18.[301A rate acceleration of 2.5 x lo5 over
the uncatalyzed rate was reported. This rate acceleration is
again much higher than the differential binding of the transition state over the substrate for the reaction, AAG =
- 2.2 kcal mol- . This difference in binding energy only accounts for a 100-fold rate acceleration. Again other factors
such as acid/base catalysis must be responsible for the observed rate acceleration. In this context it was found that
150 mM NaCl completely inhibited antibody catalysis. The
Statine (Sta)
Fig. 4. Structures of pepstatin and the phosphindte analogue 19 of the novel
amino acid statine.
for pepsin is 46 PM,making it one of the most potent enzyme
inhibitors known. Statine is considered a dipeptide analogue
which contains a hydroxyl group substituted for the carbony1 group of the scissile amide bond in pepsin substrates.
The hydroxyl group is thought to displace an enzyme-bound
water molecule which is contacted by the two catalytically
important aspartate residues in pepsin. In the X-ray crystal
structure of the complex formed between pepstatin and Rhizopus chinensis aspartic protease, the hydroxyl group is
within hydrogen-bonding distance of the two aspartate
residues.[301Based on this structure, statine is considered to
be a "collected" substrate analogue possessing binding determinants of both the peptide substrate and the enzymebound water responsible for addition to the scissile amide
bond. The sp3-hybridized carbon attached to the hydroxyl
group mimics the tetrahedral transition state for amide bond
A comparison between the two classes of inhibitors, phosphorus-based versus statine-based, can be made directly because the phosphinate analogue of statine, 19, has been synthesized.["] The KI of the pepstatin-like peptide, where
statine is substituted with the phosphinate analogue 19 is
< 70 PM.Thus, the two classes of inhibitors display roughly
the same inhibitory potency.
In addition to peptidase inhibitors, uncharged transitionstate-analogue inhibitors of enzymes in nucleotide biosynthesis pathways have been isolated. Coformycin (20) inhibits
the enzyme adenosine deaminase and is one of the most
potent inhibitors known for any enzyme ( K , = 2.5 pM).[321
The enzyme converts adenosine into inosine, which involves
the hydrolysis of the amidine moiety of adenine. Tetrahydrouridine (21) is another member of this class of inhibitors."31 One interesting common feature of this class of
inhibitors, which is not understood, is the slow binding nature of the inhibition observed. The rate constant for formation of the tight enzyme-inhibitor complex is exceedingly
slow, the half-time for dissociation of coformycin from
Angeu. Chem. In!. Ed. EngI. 29 (1990) 1296-1303
adenosine deaminase is 24 h. Based on the tight binding constants of these transition-state analogues, haptens with similar structural features might be expected to elicit catalytic
4. Hydrolytic Antibodies Elicited toward an
Uncharged Hapten
In this study we have generated five antibodies toward
racemic 6-hydroxy-7-(4-nitrophenyl)heptanoic acid (2),
which is a "statine-like" transition-state-analogue inhibitor
for the hydrolysis of para-nitrophenyl acetate (9) and paranitrophenyl methyl carbonate (6). This hapten was chosen
because it is an uncharged hydroxyl-containing analogue of
hapten 8.[231
Direct comparison of the catalytic properties of
antibodies elicited toward these differenr haptens is straightforward since they are designed to catalyze the same reactions. Indeed, one antibody, 7K16.2, was isolated which catalyzed the hydrolysis of 6 and 9. This is the first example of
the use of an uncharged hapten to elicit an esterolytic antibody. This is a surprising result considering the importance
of charge -charge complementarity between the antibody
combining site and the transition state as demonstrated by
the mutagenesis study of the PC-binding antibody S107.[201
The hydroxy carboxylic acid 2 was prepared from adipic
acid monoethyl ester and ethyl para-nitrophenylacetate
via 22 and 23 (Scheme 1 ) .
Scheme 1
A n g i w C'liem. In[. Ed. Engl. 29 (1990) 1296-1303
Experimental Procedure
22: To a stirred solution of adipic acid monoethyl ester (102.3 mg, 0.59 mmol)
in 0.6 mL of tetrahydrofuran was added carbonyldiimidazole (100 mg,
0.62 mmol). The resulting mixture was allowed to stir at room temperature for
45 min. To a stirred solution of 0.62 mL (0.62 mmol) of 1 M sodium hexamethyldisilazide in tetrahydrofuran at - 78 "C under argon was slowly added a
solution of ethyl 4-nitrophenylacetate (123.3 mg, 0.59 mmol) in 0.5 mL of tetrahydrofuran to give a purple reaction mixture. After 5 min, the above solution of the imidazolide was slowly added. The resulting mixture was allowed to
warm to 0 "C. After 30 min, i t was allowed to warm to room temperature. After
an additional 30 min, the reaction mixture was partitioned between
dichloromethane and aqueous NH,CI. The organic phase was dried over
MgSO, and then concentrated. The resulting residue was chromatographed on
silica gel with 20% ethyl acetate in hexane to give 182mg (0.5 mmol, 85%)
of22. 1R (thin film): C[cm-'] = 3000,1730, 1650. 1600, 1520, 1350. 'H NMR
(250 MHZ, CDCI,) (equilibrium mixture of keto and enol tautomers): h = 1.20
(s, lH),7.33,7 55(d.2H,
J = 9 H z ) , 8.21, 8.24 (d, 2H, J = 9 H z ) . High-resolution FAB MS: mi366.1546 ( M + He).
23: A solution of 22 (145 mg, 0.4 mmol) in 2 mL of 40% aqueous H,SO, was
heated in an oil bath at 100 "C for 15 h. The cooled reaction mixture was diluted
with water and then extracted with dichloromethane. The organic phase was
washed with Saturated aqueous NaCI, dried over MgSO,, and then concentrated. The residue was chromarographed on silica gel with 5% methanol in ethyl
acetate to give 44mg (0.16mmo1, 42%) o f 2 3 IR (KBr): C[cm-'] = 2950.
1710, 1605, 1520, 1340. M.p. 103-105°C. 'H NMR (250 MHz. CDCI,):
d = 1.63 (m, 4H), 2.36 (t. 2H, J = 7 Hz), 2.55 (t, 2H, J = 7 Hz), 3.83 ( s , 2H),
7.36 (d, 2H, J = 9 Hz). 8.19 (d. 2H, J = 9 Hz). High-resolution FAB MS: m/z
266.1021 ( M + He).
2: To a stirred solution of 23 (44 mg, 0.16 mmol) in 1 mL of methanol at 0 'C
was added sodium borohydride (13 mg, 0.34mmol). After stirring at O'C for
1 h, the reaction mixture was allowed to warm to room temperature and then
partitioned between dichloromethane and aqueous KHSO,. The organic phase
wasdriedover MgS0,and then concentrated togive44 mg(0.16 mmol, 100%)
of2. IR(KBr): 3[cm-'] = 3000,1710,1700,1600,1510,1320. M.p. 81-84 C.
'HNMR(250MHz,CDCI,):b= 1.6(m,6H),2.37(t,2H,J=7Hr),2.79(dd,
l H , J = 8, 14Hz), 2.91 (dd. IH, J = 4, 14 Hz),3.90 (m. IH). 7.38 (d. 2H,
J = 9 Hz), 8.16 (d, 2H, J = 9 Hz). High-resolution FAB MS: m/r 268.1195
( M He).
4.1. Generation and Characterization of Antibody 7K16.2
The racemic hapten 2 was coupled to carrier proteins
KLH and bovine serum albumin (BSA) by treatment of the
N-hydroxysuccinimide-activatedester of 2 with the individual proteins at pH 9.0 in order to derivatize surface >;-amino
groups of surface lysine residues.[331The ratio of hapten to
protein carrier, calculated by using UV difference spectroscopy (2: E~~~ = 9100), was 7 per KLH monomer and 3 per
BSA monomer.
Five monoclonal antibodies were isolated from the fusion
of KLH -2 conjugate-immunized Swiss Webster mouse
spleenocytes with P3X63 myeloma cells.[341Monoclonal antibodies were produced in vivo and isolated by affinity chromatography on Protein-A-coupled Sepharose 4B.[351
Following Protein A affinity chromatography, the antibodies
were further purified by Pharmacia FPLC Mono S lOjl0
cation-exchange column chromatography. Antibody was
determined to be > 95% pure by 12.5% SDS polyacrylamide gel electrophoresis with Coomasie blue
Of these five antibodies one (7K16.2) accelerated the hydrolysis of para-nitrophenyl acetate (9) and para-nitrophenyl
methyl carbonate (6) and was inhibited by free hapten.
The 7K 16.2-catalyzed reactions obeyed classical
Michaelis-Menten kinetics. Lineweaver -Burke plots obtained at pH 7.5 were linear with kinetic constants k,,,(9) =
0.72 min-' KM(9)= 3.65 mM and k,,,(6) = 0.31 min-',
KM(6)= 3.33 mM (Fig. 5). The uncatalyzed rate constants
under the same conditions are k,,,,,(9) = 3.33 x
and k,,,,,(6) = 3.2 x
min-' . The antibody is competi1301
Fig. 5. Lineweaver-Burke plot for the hydrolysis of 6 ( 0 )and 9 (+) catalyzed
by antibody 7K16.2. Velocities were determined spectrophotometrically by
measuring the initial absorbance increase at 400 nm (para-nitrophenol
J.,,, = 400 nm ( E = 4.05)). The concentration of 7K16.2 was 2.5 PM as determinedbyabsorbanceat28OnmusingE(l cm,0.1%) = 1.37andMr = 150000
for IgG. Antibody 7K16.2 was preincubated at 30°C rn 10 mM sodium phosphate, 100 mM NaCI, pH 7.5. The reaction was initiated by adding 10 pL of a
stock solution of 6 or 9 in tetrahydrofuran (THF) to give a final T H F concentration of 2%. The uncatalyzed rate was measured under the same conditions,
min-I, k,,,,, (9) = 3.3 x
affording k,,,,, (6) = 3.2 x
tively inhibited by free hapten 2 with a Kl of 140 p~
(Fig. 6).c3*’As in other examples of hydrolytic catalytic antibodies, the differential binding energy between the substrate
and inhibitor, KM(6)/K1
= 24, KM(9)/Ki= 26, found with
antibody 7K16.2 does not account for the full rate accelera= 930; kca,(9)/k,,,,,(9) =
tion observed, kca,(6)/knnca1(6)
catalysts elicited toward different classes of inhibitors, large
numbers of antibodies must be analyzed. With recent 2library technology these statistically significant numbers of
different antibodies may be a c c e ~ s i b l e . [ ~ ~ ~
Catalysis by 7K16.2 of a more energetically demanding
hydrolytic reaction was also explored. The rate of hydrolysis
ofpara-nitrophenylacetanilideat pH 9.0 was not accelerated
by this antibody as determined by an HPLC assay, using
para-nitrobenzamide as an internal standard. In a separate
study, antibodies were elicited against a KLH -coformycin
conjugate. None of the eight coformycin-specific antibodies
accelerated the conversion of adenosine to inosine
(k,,,,, = 1.8 x l o - “ s - I ) as measured by an ammonia release assay.[391It is possible that further screening of antibodies specific for these haptens will yield catalytic antibodies.
5. Outlook
In the past five years the diversity of the immune system
has been exploited to generate catalysts capable of carrying
out many different transformations. Moreover, a number of
successful strategies have been developed to produce catalytic antibodies. Several strategies have yet to be demonstrated
such as the use of strain to destabilize substrates as well as
genetic screens and selections to identify more active catalysts generated by random and directed mutagenesis. The
efficient and selective hydrolysis of peptides, nucleic acids,
and carbohydrates continue to be major targets for catalysis
by antibodies. Clearly, the antibody molecule will play an
increasingly important role in chemistry. In addition, exciting new approaches for tapping the diversity of biological
systems are being d e v e l ~ p e d . [ ~Th
- ~advances
~ - ~ ~ ~will
no doubt lead to new molecules with novel biological properties in much the same fashion as the development of antibody catalysts.
Received: July 24, 1990 [A 793 IE]
German version: Angew. Chem. 102 (1990) 1339
[21 [pMlFig. 6. Dixon plot of inhibition of antibody 7K36.2 by 2. Antibody concentration was 2.5 p ~ Data
were obtained at two concentrations of substrate 9:
500 p~ (e)and 1 mM ( 0 ) .Buffer conditions are as described in Figure 5.
2250. For the identical reaction catalyzed by antibody 48674A1 raised against phosphonate 8, the differential binding
energy is KM(6)/K,(8)= 1300, compared to the observed rate
accelerations of kC,,(6)/k,,,,,(6) = 7300. However, the disparity between the two values is much more pronounced in
the case of antibody 7K16.2, suggesting that other factors
(such as immunological diversity) contributing to the catalysis besides transition-state stabilization are of greater relative
importance than was the case for antibody 48G7-4Al. In
order to more definitively determine the differences between
[l] M. M. Davis, P. J. Bjorkmann, Nature (London) 334 (1988) 395.
[2] P. G. Schultz, Angew. Chem. 101 (1989) 1336; Angew. Chem. Int. Ed. Engl.
28 (1989) 1283.
[3] P. G. Schultz, A r c . Chem. Res. 22 (1989) 287.
[4] K. M. Shokat, P. G. Schultz, Annu. Rev. Immunol. 8 (1990) 335.
[5] P. G. Schultz, R. A. Lerner, S. J. Benkovic, Chem. Eng. News 68 (1990)
No. 22, p. 26.
[6] W. P. Jencks: Catalysis in Chemistry and Enzymology, McGraw-Hill, New
York 1969.
[7] L. Pauling, Am. Sci. 36 (1948) 51.
[8] R. Wolfenden, Annu. Rev. Biophys. Bioeng. 5 (1976) 271.
191 P. A. Bartlett, C. K. Marlowe, Biochemisfry 22 (1983) 4618.
[lo] M. W Holladay, F. G. Salituro, D. H. Rich, J. Med. Chem. 30 (1987) 374.
[ l l ] K. N. Allen, R. H. Abeles, Biochemistry 28 (1989) 135.
[12] W M. Kati, D. T. Pals, S. Thaisrivongs, Biochemistry 26 (1987) 7621.
Med. Chem. 28 (1985) 263.
[13] D. R. Rich, .
[I41 A. M. Goetze, J. H. Richards, Biochemistry 16 (1977) 228.
[15] L. G. Bennett, C. P. Gbdudemdns, Biochemistry 18 (1979) 3337.
[16] D. M. Segal, E. A. Pddlan, G. H. Cohen, S. Rudikoff, D. Potter, D. R
Davies, Proc. Natl. Acud. Sci. USA 71 (1974) 4298.
[17] S. J. Pollack, J. W. Jacobs, P. G. Schultz, Science (Washinglon D.C.) 234
(1986) 1570.
[18] S . J. Pollack, P G. Schultz, Cold Spring Harbor Symp. Quant. Bid. 52
(1987) 97.
Angew Chem. I n ! . Ed. EngI. 29 (1990) /296-1303
[19] Y. Satow, G. H. Cohen, E. A. Padlan, D. R. Davies, J. Mol. Bid. 190
(19x6) 593.
[20] D . Y. Jackson. J. R. Prudent, E. P. Baldwin, P. G. Schultz, Proc. Nafl.
Acud. Sci. USA 87 (1990), in press.
(211 S. J. Pollack, G. R. Nakayama, P. G. Schultz. Science (Washington D.C.)
242 (1988) 1038.
[22] S . J Pollack, P. G. Schultz, J. Am. Chem. Soc. f / f (1989) 1929.
1231 J. W. Jacobs, P. G. Schultz, R. Sugasawara, M. Powell, J. Am. Chem. Soc.
/f)Y (1988) 2174.
[24] J. W. Jacobs. Dissertation, University of California, Berkeley 1989.
1251 A. Tramontdno, A. A. Ammann. R. A. Lerner. J. Am. Chem. Soc. 110
(1988) 2282.
1261 K . D. Janda, S. J. Benkovic, R. A. Lerner, Science ( WashingfonD.C.) 224
(1989) 437.
1271 S. J. Pollack, P. Hsuin, P. G. Schultz, J. Am. Chem. Soc. / / f (1989) 5961.
[28] E. Engvall, P. Perlmann, J. Immunol. f 0 9 (1972) 129.
1291 W. D. Huse, L. Sastry, S. A. Iverson, A. S. Kang, M. Alting-Mees, D. R.
Angrw. Chem. h i . Ed. Engl. 29 (1990) 1296-1303
Burton, S. J. Benkovic, R. A. Lerner, Science ( Wkshmgton D.C. j 246
(1989) 1275.
[30] K. D. Janda, D. Schloeder, S. J. Benkovic, R. A. Lerner, Science 241
(1988) 1188.
[3l] P. A. Bartlett, W B. Kezer, J. Am. Chrm. Soc. 106 (1984) 4282.
1321 L. Frick, J. P. Mac Neela, R. Wolfenden. Bioorg. Chem. 15 (1987)
1331 B. Erlanger, Mefhods Enzymol. 70 (1980) 85.
[34] R. Sugasawara, C. Prato, J. Sippel, Infect. Immun. 42 (1983) 863.
[35] G. Kronvall, H. Grey, R. Williams, J. Immunol. I05 (1 972) 11 16.
1361 C. Tuerk, L. Gold, Science (Washington D.C.) 249 (1990) 505.
[37] U. Laemmli, Nature (London) 227 (1970) 680.
[38] G. Dixon, Biochem. J. 55 (1953) 170.
1391 A. L. Chaney, E. P. Marbach, Clin. Chem. 8 (1962) 130.
[40] J. J. Devlin, L. C. Panganiban, P. E. Devlin, Scrence (Washington D.C.)
249 (1990) 404.
1411 J. K. Scott, G. P. Smith, Science (Wushinglon D.C.) 249 (1990) 386.
Без категории
Размер файла
766 Кб
class, antibodies, catalytic, state, transitional, used, hydrolytic, new, elicits, analogues
Пожаловаться на содержимое документа