Pivalase catalytic antibodies: Towards abzymatic activation of prodrugs

Article abstract of DOI:10.1002/1521-3765(20011105)7:21<4604::AID-CHEM4604>3.0.CO;2-Z

Screening of monoclonal-an-tibody libraries generated against the tert-butyl phosphonate hapten 2 and the chloromethyl phosphonate hapten 3 with pivaloyloxymethyl-umbelliferone 1 as a fluorogenic substrate led to the isolation of eleven catalytic antibodies with rate accelerations around k_cat/k_uncat=10³. The antibodies are not inhibited by the product and accept different acyloxymethyl derivatives of acidic phenols as substrates. The highest activity was found for the bulky, chemically less-reactive pivaloyloxymethyl group; there is no activity with acetoxymethyl or acetyl esters. This difference might reflect the preference of the immune system for hydrophobic interactions in binding and catalysis. Pivalase catalytic antibodies might be useful for activating orally available pivaloyloxymethyl prodrugs.

Full text of DOI:10.1002/1521-3765(20011105)7:21<4604::AID-CHEM4604>3.0.CO;2-Z

FULL PAPER
Pivalase Catalytic Antibodies: Towards Abzymatic Activation of Prodrugs
Nicolas Bensel, Martine T. Reymond, and Jean-Louis Reymond*[a]
Abstract: Screening of monoclonal-an-
tibody libraries generated against the
tert-butyl phosphonate hapten 2 and the
hibited by the product and accept differ-
ent acyloxymethyl derivatives of acidic
phenols as substrates. The highest activ-
ity was found for the bulky, chemically
less-reactive pivaloyloxymethyl group;
there is no activity with acetoxymethyl
or acetyl esters. This difference might
reflect the preference of the immune
system for hydrophobic interactions in
binding and catalysis. Pivalase catalytic
antibodies might be useful for activating
orally available pivaloyloxymethyl pro-
drugs.
chloromethyl phosphonate hapten
3
with pivaloyloxymethyl-umbelliferone
1 as a fluorogenic substrate led to the
isolation of eleven catalytic antibodies
with rate accelerations around kcat/
kuncat ˆ 103. The antibodies are not in-
Keywords: catalytic antibodies ´ flu-
orogenic substrates
´ hydrolysis ´
immunochemistry ´ prodrugs
catalysis.[7] Herein we report on the activation of pivalate
prodrugs by an esterolytic reaction that makes use of catalytic
antibodies.
Introduction
Catalytic antibodies are obtained by immunization against
stable transition-state analogues of chemical reactions. This
technique allows, in principle, the installation of any catalytic
activity in the combining site of immunoglobulins.[1] On
account of their excellent long-term stability in human blood
and their ability to escape immune rejection, immunoglobu-
lins are also an outstanding framework for designing whole-
protein drugs. Several monoclonal-antibody drugs have
reached the market recently; in particular, antibodies that
bind to tumor antigens and that are thus effective as antitumor
agents.[2] Similarly, one of the best applications envisioned for
catalytic antibodies lies in the medicinal area, particularly for
drug detoxification[3] and prodrug activation.[4] Provided that
a useful catalytic activity is available in an antibody binding
pocket, its adaptation to a tool for prodrug activation follows
a straightforward sequence that has been well studied owing
to work with antibody ± enzyme conjugates[5] and therapeutic
antibodies.[2] In particular, the ªhumanizationº of the starting
antibody, which is essential to ensure the immunocompati-
bility of an antibody, can be done by CDR (complementarity-
determining regions) grafting without altering the specificity
of the antibodyꢀs binding pocket.[6] One of the critical factors
that remains today is the identification of prodrug-activating
reactions that are both selective and suitable for antibody
Results and Discussion
Design: The reaction catalyzed most often by catalytic
antibodies is ester hydrolysis. Catalytic antibodies, however,
most often hydrolyze reactive esters, which are chemically too
unstable to be suitable for prodrug-masking chemistry. We
reasoned that a useful catalytic antibody for prodrug activa-
tion should be able to cleave less-reactive esters, for example,
sterically hindered esters such as pivalate esters. The piv-
aloyloxymethyl group seemed interesting since it can be used
to protect a variety of functional groups found in drugs, such
as phenols, amides, amines, pyridines, or phosphates, and is
frequently used in prodrug formulations to make drugs orally
available.[8] We therefore set out to investigate the prepara-
tion of ªpivalaseº catalytic antibodies (Scheme 1).
[a] Prof. Dr. J.-L. Reymond, Dr. N. Bensel, M. T. Reymond
Departement für Chemie und Biochemie, Universität Bern
Freiestrasse 3, 3012 Bern (Switzerland)
Fax : (‡41)31-631-80-57
E-mail: jean-louis.reymond@ioc.unibe.ch
Scheme 1. Principle of pivalase catalytic antibodies. Pivaloyloxymethyl-
protected drugs (XH) become selectively activated by pivalase catalytic
antibodies.
Supporting information for this article is available on the WWW under
http://wiley-vch.de/home/chemistry/ or from the author.
4604
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4604 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 21

4604±4612
The pivaloyloxymethyl derivative of 4-carboxymethyl um-
belliferone 1 was selected as a convenient fluorogenic prodrug
model (Scheme 2). Fluorescence screening with substrate 1
would allow us to test catalysis early in the process of cell
culture that leads from immunized mice to the stable hybrid-
oma cell lines that produce catalytic antibodies.[9] Phospho-
nate-type haptens 2 and 3 were chosen to elicit a catalytic
immune response. tert-Butyl phosphonate 2 has a phospho-
nate group that corresponds to the hydrolyzing pivalate and
also the correct spacing to the anilide aromatic group. We
selected this anilide group and its adjacent glutaramide linker
because this immunogenic combination is often found in
haptens that lead to catalytic antibodies. The discrepancy
between the anilide group in the hapten and the coumarin
nucleus in the assay substrate was intended to enforce the
selection of catalytic antibodies with a broad substrate
tolerance. In addition to the tert-butyl phosphonate 2, we
also prepared the chloromethyl phosphonate hapten 3, which
might trigger a process of reactive immunization, whereby a
nucleophilic residue in the antibody binding pocket would
displace the reactive chloride and become covalently attached
to hapten 3.[10] This nucleophilic residue would then assist the
hydrolysis of the substrate.
Scheme 3. Synthesis of assay substrate 1 and haptens 2 and 3. The haptens
were conjugated to carrier proteins KLH for immunization and to BSA for
ELISA assays. i) NaH, ICH2O2CtBu, DMF, 608C to RT, 10 h; ii) NH3/
NH4Cl in MeOH, 208C, 6 h (5%). iii) NaH, dioxane, tBuPOCl2, 508C, 1 h
(35%); iv)a) 1 atm H2, Pd/C, MeOH, 258C, 2 h; b) glutaric anhydride,
CH2Cl2, 258C, 3 h (80%). v) ClCH2POCl2, iPrEt2N, cat. tetrazole, 258C,
8 h (20%); vi)a) 1 atm H2, Pd/C, MeOH, 258C, 3 h; b) glutaric anhydride,
DMF/CH2Cl2, 258C, 3 h (27%).
was prepared by esterification of tert-butyldichloro phosphor-
idate[11] with the sodium salt of 4-nitrophenethyl alcohol,[12]
followed by hydrogenation of the nitro group and amidation
with glutaric anhydride. Hapten 3 was obtained by a similar
sequence by means of the tetrazole-catalyzed phosphonate
synthesis described by Landry et al.[13] The haptens were
conjugated to carrier proteins KLH (keyhole limpet hemo-
cyanin) and BSA (bovine serum albumin) through their N-
hydroxysuccinimide esters.
Synthesis and immunization: Alkylation of 4-carboxymethyl
umbelliferone 4 with iodomethyl pivalate under optimized
conditions gave an equal mixture of the desired pivaloyloxy-
methyl derivate 1 together with the corresponding pivalate
derivative. Selective cleavage of this more reactive pivalate
ester with buffered ammonia in methanol and subsequent
purification by preparative reverse-phase HPLC allowed the
isolation of substrate 1 in a pure (>99%) form, as required
for high-throughput screeening (HTS) (Scheme 3). Hapten 2
Two series of four 129GIX ‡ mice were immunized with
the KLH conjugate of either hapten 2 or hapten 3, according
to standard procedures.[14] The
spleen cells of the immunized
mice were fused with NS1
myeloma cells, and the resulting
hybridoma plated out in 96-well
microtiter cell-culture plates.
Wells containing cells that were
producing hapten-specific anti-
bodies against the hapten ±
BSA conjugates, as judged by
ELISA (enzyme-linked immu-
no-sorbens assay),[15] were
propagated to
a volume of
5 mL. At that stage, the anti-
bodies present in the superna-
tant of each cell-culture sample
were isolated by a short pro-
tein-G-affinity column and as-
sayed for catalysis. Catalysis
was assayed in each antibody
sample by measuring the rate of
hydrolysis of fluorogenic sub-
strate 1, by taking the identical
sample with added hapten 2 or
3 as inhibitor as reference for
the uncatalyzed reaction. This
Scheme 2. Hapten design and HTS assay for pivalase catalytic antibodies. Antibodies raised against hapten 2
(exact match) or 3 (reactive immunization) might catalyze the hydrolysis of fluorogenic prodrug model 1 to
release pivalate, formaldehyde, and the strongly fluorescent umbelliferone 4.
Chem. Eur. J. 2001, 7, No. 21
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4605 $ 17.50+.50/0
4605

J.-L. Reymond et al.
FULL PAPER
Table 2. Catalytic anti-3 antibodies: immunological data and kinetic
parameters with substrate 1.
procedure allowed the unequivocal identification of any
catalysis occurring specifically in the combining site of the
antibodies. This procedure was used systematically to confirm
all ELISA data for hapten binding, and proved indispensable
in the further steps of subcloning for distinguishing cell
colonies producing catalytic antibodies from those producing
only hapten-binding antibodies.
The overall experiment lead to the isolation of 24 hybrid-
oma (Table 1). The monoclonal antibody expressed by each
cell line was produced in 5 to 50 mg amounts by cell culture,
and purified to homogeneity by ammonium sulfate precip-
itation and protein-G-affinity chromatography. Immunization
[d]
Antibody[a] Amount Isotype[c] kcat
KM
[mm][d]
kcat/kuncat
KTS
[mm][d]
1
[mg][b]
[10 3 s
]
8D5
8A9
7H11
8A11
2G8
8E1
18.5
14.8
15.5
53.7
15.3
11.5
18
15.5
39
23
kg2a
kg1
kg1
kg1
kg1
kg1
kg1
kg1
kg2a
kg1
kg2a
0.12
0.17
0.08
0.11
0.07
0.12
0.09
0.09
0.07
0.05
0.07
43
270
57
47
69
1700
2450
1120
1900
1020
2100
1280
1260
990
0.025
0.11
0.05
0.02
0.07
0.04
0.14
0.10
0.19
0.09
0.15
83
8C5
185
130
185
60
9D12
8F10
6F4
700
980
1C12
15.8
150
[a] The code refers to the numbering of the original fusion cell-culture plate
followed by the well code (A1 ± H12). [b] Amount of pure antibody
isolated from 1 L of culture of the corresponding hybridoma cell line.
[c] Isotype, as determined by immunological isotyping. All isotypes were
cleanly unique and establish the monoclonal nature of the cell lines.
[d] Measured at 308C in aq. bis-tris buffer (20mm), NaCl (150mm),
pH 7.55, 5% v/v DMF, with antibody (0.2 mgmL 1). The uncatalyzed
Table 1. Data for immunization with haptens 2 and 3 and catalysis HTS with
substrate 1.
hapten
mouseL[a] mouseR[a] mouseLL[a] mouseLR[a]
2 (tBu)
serum titer[b]
no. of binders[c]
monoclonals[d]
catalytic[e]
12800
6400
47
1
25600
25600
13
1
0
55
3
0
75
2
0
1
reaction is 7 Â 10 8 s under these conditions. See also Experimental
0
Section. Error margins are Æ10%.
3 (ClCH2) serum titer[b]
no. of binders[c]
monoclonals[d]
12800
19200
25600
160
16
25600
36
0
0
45
0
0
50
1
0
The hydrolysis of 1 was proportional to the hydroxide ion
concentration between pH 6.5 and 9.5, in agreement with the
generally accepted mechanism of hydrolysis of acyloxymethyl
esters, as shown in Scheme 2.[16] A pH-profile analysis with
catalytic antibody 8A11 showed that its catalytic efficiency
was also proportional to the concentration of hydroxide
between pH 6.5 and 9.5. This suggests that the antibody-
catalyzed process also used the classical esterolytic mecha-
nism, in agreement with the phosphonate transition-state
analogues used for immunization. This suggested that no
protein sidechain ionizable in that range participated in the
catalysis, and that reactive immunization had not occurred
with chloromethyl phosphonate hapten 3. Indeed, while all
the antibodies lost their catalytic activity in the presence of
hapten 3, the activity returned in all cases by repurifying the
antibodies by protein-G-affinity chromatography. This proce-
dure included an extensive washing step in which the antibody
was bound to the column (at neutral pH) to remove the
hapten before acidic elution. The inhibition by hapten 3 was
thus reversible and most likely noncovalent. The catalytic
activity of these antibodies probably reflects simple transi-
tion-state stabilization. The absence of catalytic antibodies in
the immunization with hapten 2 is unclear, especially in view
of the fact that all the anti-3 catalytic antibodies are also
inhibited by hapten 2.
catalytic[e]
11
[a] Mouse code according to ear marks. [b] Dilution factor of blood serum
(after immunization with hapten ± KLH conjugates) for 50% reduction of
ELISA signal against the hapten ± BSA conjugate. A high number indicates
strong immune response. [c] Number of hybridoma cell lines that test positive
for binding by ELISA against the hapten ± BSA conjugate after fusion and that
were grown up to 5 mL for catalysis testing. [d] Number of fully subcloned and
stabilized hybridoma. [e] Number of fully subcloned hybridoma producing
catalytic antibodies.
with tert-butylphosphonate hapten 2 did not produce any
clones that tested positive for catalysis. Nevertheless, seven
clones were isolated as controls. None of these showed any
catalytic activity in the purified form; this showed that the
HTS fluorescence assay did not produce false negatives. The
immunization with chloromethylphosphonate hapten 3, by
contrast, showed a number of ªhitsº. Seventeen of the initial
hits were successfully subcloned twice to stable hybridoma.
Eleven of these showed confirmed catalytic activity after two
subclonings and large-scale production and purification. All
catalytic antibodies were obtained from a single mouse
(Table 1).
Kinetic analysis: All eleven catalytic antibodies showed
Michaelis ± Menten kinetic behavior for the hydrolysis of 1
(Table 2). All the antibodies also bound to hapten 3 tightly,
and their activity could be titrated quantitatively to two
equivalent binding sites per antibody molecule. The rate
accelerations kcat/kuncat observed with substrate 1 are in the
range of three orders of magnitude, a typical number for
esterolytic catalytic antibodies. The transition-state dissocia-
tion constants KTS ˆ KM/(kcat/kuncat) are in the range of 10 7 m,
as found for most known catalytic antibodies. Remarkably,
the catalytic antibodies were not susceptible to any product
inhibition by pivalic acid or umbelliferone.
Substrate specificity: The substrate specificity of the catalytic
antibodies was then investigated. The simple fluorogenic
umbelliferone derivative 5 and the chromogenic nitrophenyl
derivative 6 were both accepted as substrates, with kinetic
parameters almost equivalent to those of substrate 1 for all
catalytic antibodies. Derivatives of substrate 5, whose syn-
thesis was less problematic than that of substrate 1, were thus
prepared (Scheme 4). Gradual removal of the pivalateꢀs
methyl substituents to give isobutyryloxymethyl umbellifer-
one 7 and propionyloxymethyl umbelliferone 8, resulted in a
4606
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4606 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 21

Pivalase Catalytic Antibodies
4604±4612
Scheme 4. Substrates for anti-3 catalytic antibodies.
decrease in the catalytic activity, as measured by kcat/kuncat with
all catalytic antibodies (Table 3). Only two of the antibodies
catalyzed the hydrolysis of acetoxymethyl substrate 9 (8A9,
kcat/kuncat ˆ 200; 8E1, kcat/kuncat ˆ 300). The antibodies also
hydrolyzed the umbelliferyl esters 10 ± 12, but showed no
activity with umbelliferyl acetate 13. The highest turnover
numbers kcat were observed with propionate derivative 12, but
these reflect its high spontaneous reactivity. Overall, the
pivaloyloxymethyl (POM) derivatives 5 and 6 were the best
substrates in terms of rate acceleration kcat/kuncat and transi-
tion-state binding KTS; this reflects the antibody-specific
contributions to catalysis.
With the POM group providing the optimal functional
group for our antibodies, we next investigated whether POM-
protected derivatives other than umbelliferone would react
with our antibodies. While the chromogenic POM derivative 6
of 4-nitrophenol was a good substrate with all antibodies, the
corresponding POM-protected 2-nitrophenol 14 (Scheme 5)
was not accepted as a substrate by any of the catalytic
antibodies. Competitive inhibition experiments with this and
other POM derivatives of ortho-substituted phenols, such as
the vanilline derivative 15 (Scheme 5), showed that such
derivatives were not recognized by the catalytic antibodies;
this points to strict steric requirements for access to the
antibodyꢀs binding pocket. Among the sterically less-demand-
ing para-substituted phenols, only 4-acetylphenol 16 and
4-cyanophenol 17 showed marginal activity with one of the
antibodies, while the others, such as the 4-chlorophenol
derivative 18, paracetamol derivative 19, or fluoresceine
derivative 20, all acted as competitive inhibitors of the
catalytic antibodies but did not undergo any reaction.
Pivaloyloxymethyl pyridinium trifluoroacetate (21) was ac-
cepted weakly by one of the antibodies. These observations
Scheme 5. Some POM derivatives tested with anti-3 catalytic antibodies.
showed that a strongly acidic leaving group was required
for the antibodies to hydrolyze the pivaloyloxymethyl
prodrugs. It should be noted that the Boc-protected
umbelliferone 22 was not accepted as a substrate by any of
the catalytic antibodies, despite its high spontaneous reac-
tivity.
Interestingly, the difficulty encountered with our catalytic
antibodies for hydrolyzing the bulky and unreactive POM
derivatives when attached to nonacidic leaving groups was
also observed with esterolytic enzymes. Thus the caged NO
prodrug 23, which we had synthesized from DEA NONOate
(2-(N,N-diethylamino)diazenolate-2-oxide sodium salt) by
reaction with chloromethyl pivalate, was not hydrolyzed by
the antibodies; it was also resistant to a range of esterases,
lipases, and to exposure to mouse serum. This is in contrast
to the corresponding acetoxymethyl derivative.[17] Similarly,
while esterolytic enzymes cleaved our fluorogenic umbellifer-
one substrates (horse liver esterase and Pseudomonas
fluorescens lipase were used as representative cases),
they were not capable of cleaving the POM derivate of the
less acidic fluorescein 20. This showed that a bulky pivalate
group combined with an unreactive leaving group poses
particular problems for enzymes and catalytic antibodies
alike.
Chem. Eur. J. 2001, 7, No. 21
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4607 $ 17.50+.50/0
4607

J.-L. Reymond et al.
FULL PAPER
Table 3. Kinetic parameters for anti-3 catalytic antibodies.[a]
Substrate
5
6
7
8
10
11
12
1
background
Ab8D5
kuncat [10 5 s
]
0.007
12
1770
65
0.04
60
1580
120
0.3
0.4
0.06
26
440
100
0.2
22
110
0.6
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
kcat [10 5 s
kcat/kuncat
KM [mm]
KTS [mm]
]
]
]
]
]
]
]
]
]
]
]
180
610
245
115
285
100
255
430
225
1
1040
0.037
0.07
0.4
0.36
0.23
0.93
22
110
0.52
1
1
1
1
1
1
1
1
1
1
Ab 8A9
Ab 7H1
Ab 8A11
Ab 2G8
Ab 8E1
Ab 8C5
Ab 3E2
Ab 8F10
Ab 6F4
Ab 1C12
17
2450
265
37
940
200
154
515
70
280
710
100
5
80
35
195
325
155
1040
0.11
0.21
0.14
0.14
0.42
0.93
14
70
120
1.72
15
80
220
0.48
8
1125
60
50
1240
265
133
445
70
106
265
80
10
165
70
485
810
350
0.05
0.2
0.16
0.3
0.42
0.43
11
1590
95
50
1250
135
125
415
95
83
210
45
27
445
282
266
445
165
0.06
0.11
0.2
0.22
0.63
2.87
0.37
7
1015
70
37
940
200
110
370
70
180
460
240
5
80
35
40
200
310
300
500
265
0.07
0.21
0.19
0.52
0.41
1.54
0.52
12
1740
95
37
940
90
70
225
55
135
340
135
13
215
70
40
200
310
170
280
415
0.055
0.09
0.25
0.39
0.33
1.54
1.48
9
1280
185
12
310
65
65
215
90
95
240
120
5
90
60
70
355
1440
4
21
105
165
1.52
14
70
255
3.55
275
460
420
0.14
0.21
0.42
0.49
0.65
0.91
9
1260
130
19
470
100
57
190
255
56
140
65
6
100
65
224
375
240
0.1
0.21
1.33
0.46
0.65
0.64
7
990
185
19
470
150
45
150
52
175
440
130
7
125
130
1
5
90
30
415
695
310
0.19
0.32
0.35
0.3
0.45
5
700
60
20
470
100
45
145
50
175
435
175
34
170
215
162
270
165
0.09
0.21
0.34
0.4
0.34
1.25
0.62
7
985
150
19
470
200
30
100
50
150
380
70
5
90
30
180
890
600
440
735
485
0.15
0.43
0.5
0.18
0.34
0.67
0.65
[a] Measured at 318C in 20mm bis-tris pH 7.55, see Experimental Section for details. The background reaction was measured in the presence of catalytic
antibody with excess (20mm) hapten 3. Error margins are Æ10%.
analogues might reflect the strong preference of the immune
response for exploiting hydrophobic interactions as a means
Conclusion
Immunization with chloromethyl phosphonate hapten 2 has
yielded the first catalytic antibodies capable of hydrolyzing
bulky pivalate esters, in particular the rather stable POM-
protected derivatives, which can be incorporated as elements
of orally available prodrugs. The catalytic antibodies obtained
appear to be limited to POM derivatives of acidic phenols and
do not cleave POM derivatives with less acidic phenol leaving
groups. The limitation to activated leaving groups is often
encountered with esterolytic catalytic antibodies. In the
present case, the reactivity pattern also parallels that observed
with esterases and lipasesÐwith the enzyme-resistant POM
derivatives, which are the most interesting in prodrug
applications, being presently left untouched. The preference
of our catalytic antibodies for pivalate esters over less bulky
of achieving high binding affinity to antigens. Further
immunizations or manipulation of the current set of catalysts
by recombinant methods might lead to pivalase catalytic
antibodies with improved kinetic properties.
Experimental Section
Synthesis: All reagents were either purchased from Aldrich or Fluka or
synthesized following literature procedures. Chromatography (flash) was
performed with Merck silica gel60 (0.040 ± 0.063 mm). Preparative HPLC
was performed with HPLC-grade acetonitrile and MilliQ deionized water
in a Waters prepak cartridge 500 g (RP-C18 20 mm, 300 Š pore size)
installed on a Waters Prep LC4000 system from Millipore (flow rate
1
100 mLmin 1, gradient ‡0.5%min CH3CN) following the conditions
4608
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4608 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 21

Pivalase Catalytic Antibodies
4604±4612
given in Table 4. TLC was performed with fluorescent F254 glass plates.
MS was provided by Dr. Thomas Schneeberger (University of Bern,
Switzerland).
C: KLH (5 mgmL 1) in phosphate buffer saline (PBS: Na2HPO4 (10mm),
NaCl (160mm), pH 7.4).
D: BSA (5 mgmL 1) in PBS.
A solution of hapten 2 (10 mg, 0.027 mmol) in DMF (0.15 mL) was treated
with EDCI (soln A, 0.045 mL, 2.5 equiv) and NHS (soln B, 0.042 mL,
2.5 equiv). After 24 h, HPLC analysis (Table 4) showed that 2 (tR ˆ 6.0 min)
was >90% converted to the active ester (tR ˆ 12.9 min). 0.075 mL of the
active solution was then added to KLH (soln C, 0.5 mL) and 0.15 mL to
BSA (soln D, 1 mL). Aq. sat. NaHCO3 (0.07 mL) was added to each
solution. After 24 h at 48C, HPLC analysis indicated that the active ester
had been consumed. No free acid was detected. The carrier-protein
solutions were used without further purification.
Table 4. Reverse-phase HPLC conditions for analysis and purification.[a]
Compounds
A [%]
B [%]
tR [min]
Preparative conditions[b]
1
2
3
5
6
7
8
9
10
12
13
21
24
25
30
40
70
40
15
50
50
75
40
50
65
90
20
60
70
60
30
60
85
50
50
25
60
50
35
10
80
40
18.3
6
4.8
24.9
13.5
26.9
13.5
15.6
27
10.5
10
14.9
8.3
50 ± 70% B in 20 min
[c]
30 ± 50% B in 20 min
[c]
[c]
45 ± 75% B in 30 min
40 ± 60% B in 20 min
4-Nitrophenethyl chloromethyl phosphonate (25): A solution of nitro-
phenethyl alcohol (100 mg, 0.6 mmol) in toluene (3 mL) and anhydrous
DMF (0.5 mL) at 08C was treated with 1H-tetrazole (4 mg, 0.1 equiv), N,N-
diisopropylethylamine (0.15 mL, 2.2 equiv), and chloromethyl dichloro-
phosphoridate (100 mg, 0.062 mL, 1 equiv). The temperature was slowly
raised to 258C. After 8 hours, water (1 mL) was added. After 45 minutes,
the reaction was acidified (50 mL HCl, 1n) and extracted twice with
CH2Cl2. The organic phase was dried over Na2SO4 and concentrated, and
the residue was purified by preparative reverse-phase HPLC (gradient 10 ±
20% acetonitrile in 30 minutes). The main fraction was lyophilyzed to give
the phosphonate 25 as a colorless solid. Yield: 50 mg (0.18 mmol, 30%);
10 ± 40% B in 30 min
[c]
40 ± 60% B in 20 min
40 ± 60% B in 20 min
0 ± 20% B in 20 min
60 ± 80% B in 20 min
20 ± 40% B in 30 min
3.8
[a] Isocratic conditions for analytical RP-HPLC on
a Vydac 218-
TP54C18RP, particle size 5 mm, pore size 300 Š, 0.45  22 cm, flow
1
m.p. 1018C; H NMR (300 MHz, CDCl3/CD3OD): d ˆ 8.14 (d, J ˆ 6.6 Hz,
1
1.5 mLmin
,
A ˆ 0.1% CF3COOH in water, B ˆ acetonitrile/water
2H), 7.45 (d, J ˆ 6.6 Hz, 2H), 4.32 (q, J ˆ 6.3 Hz, 2H), 3.47 (d, J ˆ 10.7 Hz,
2H), 3.09 (t, J ˆ 6.3 Hz, 2H); 13C NMR (75 MHz, CD3OD/CDCl3): d ˆ
147.5, 146.2, 130.6, 124.1, 66.9 (d, J ˆ 6.1 Hz), 37.2 (d, J ˆ 6.6 Hz), 34.0 (d,
J ˆ 156 Hz); IR (CHCl3): nÄ ˆ 2956, 2322, 1598, 1511, 1394, 1341, 1270, 1046,
(1:1). [b] Preparative HPLC was performed by gradient elution on a
Waters prepak cartridge (500 g), flow rate 100 mLmin 1, with the same
eluents as indicated. [c] This compound was purified by flash chromatog-
raphy on silica gel.
‡
1035, 852 cm 1; MS (SIMS): 281 [M‡H] , 267, 221, 207, 154, 147, 136, 120;
elemental analysis calcd (%) for C9H11ClNO5P (279.01): C 38.71, H 3.97;
found: C 38.78, H 3.94.
4-Nitrophenethyl-tert-butylphosphonate (24): A solution of nitrophenethyl
alcohol (100 mg, 0.57 mmol) in dry dioxane (2 mL) at 208C was treated
with NaH (23 mg, 55% in oil, 1.1 equiv). After 5 minutes at 258C, tert-butyl
phosphonyl dichloride[11] (100 mg, 0.57 mmol) was added, and the reaction
was heated to 508C for 1 h. Water (1 mL) was added, and the reac-
tion mixture was heated for 45 minutes at 508C. The reaction was
then acidified with aq. 1n HCl (50 mL) and extracted twice with ethyl
acetate. The organic phase was dried (Na2SO4) and concentrated, and the
residue purified by preparative reverse-phase HPLC (gradient 30 ± 40%
acetonitrile in 20 minutes). The main fraction was lyophilized to give
compound 24 as a yellow solid. Yield: 55 mg (0.19 mmol, 35%); 1H NMR
(200 MHz, CDCl3): d ˆ 8.20 (d, J ˆ 8.5 Hz, 2H), 7.42 (d, J ˆ 8.5 Hz, 1H),
4.45 (m, 2H), 3.17 (t, 2H), 1.23 (d, J ˆ 21 Hz, 9H); 13C NMR (75 MHz,
CDCl3): d ˆ 144.7, 141.2, 129.8, 123.7, 65.8 (d, J ˆ 36.4 Hz), 37.8 (d, J ˆ
466 Hz), 36.1 (d, J ˆ 26.7 Hz), 24.2 (d, J ˆ 4.8 Hz); 31P NMR (81 MHz,
CDCl3): d ˆ 57.2.
Chloromethylphosphonic acid 2-[4-(5-hydroxy-1,5-dioxopentalamino)phe-
nyl]ethyl ester (3): A solution of chloromethyl phosphonate 24 (100 mg,
0.36 mmol) in methanol (3 mL) was stirred with Pd/C (8 mg) under 1 atm
H2 for 3 h, filtered over celite, and concentrated. The residue (86 mg of a
pale yellow oil) was dissolved in anhydrous CH2Cl2 (3 mL) and dry DMF
(3 mL); then glutaric anhydride (80 mg, 2 equiv) was added. After 3 h at
258C, the solution was concentrated under vacuum, and the residue
purified by preparative reverse-phase HPLC. Lyophilization of the main
fraction yielded hapten 3 as a yellow oil. Yield: 37 mg (0.1 mmol, 27%);
1H NMR (200 MHz, CD3OD): d ˆ 7.43 (d, J ˆ 8.2 Hz, 2H), 7.14 (d, J ˆ
8.2 Hz, 1H), 4.21 (q, J ˆ 6.9 Hz, 2H), 3.43 (d, J ˆ 10.2 Hz, 2H), 2.92 (t, J ˆ
6.9 Hz, 2H), 2.37(m, 4H), 1.98 (q, 2H); 13C NMR (75 MHz, CDCl3): d ˆ
176.2, 172.5, 137.3, 133.3, 129.7, 120.6, 67.6 (d, J ˆ 6.1 Hz), 36.7 (d, J ˆ
5.5 Hz), 36.2, 34.0 (d, J ˆ 119.5 Hz), 33.5, 21.2; IR (CHCl3): nÄ ˆ 3002,
1
1712, 1673, 1518, 1388, 1093 cm
.
Carrier protein conjugates of hapten 3: The same procedure as for hapten 2
was applied starting with hapten 3 (10 mg, 0.028 mmol).
4-{4-[2-(tert-Butylphosphonyloxy)ethyl]phenylcarbamoyl} butyric acid (2):
A solution of tert-butylphosphonate 24 (90 mg, 0.31 mmol) in methanol
(2 mL) was stirred with Pd/C (9.5 mg) under 1 atm H2 for 2 h, filtered over
celite, and concentrated. The residue (80 mg of a colorless oil) was
dissolved in anhydrous CH2Cl2 (3 mL), and glutaric anhydride (53 mg,
1.5 equiv) was added at 08C. After 3 h at 258C, the solution was washed
with 1n HCl, dried with Na2SO4, and concentrated. The residue was
purified by flash chromatography [i) hexane/ethyl acetate 3:7, ii) ethyl
acetate, iii) CH2Cl2/MeOH 7:3] to give hapten 2 as a yellow oil. Yield:
92 mg (0.25 mmol, 80%); Rf ˆ 0.15 (CH2Cl2/MeOH 7:3); 1H NMR
(200 MHz, CDCl3): d ˆ 8.28 (s, 1H), 7.49 (d, J ˆ 8.4 Hz, 2H), 7.13 (d, J ˆ
8.4 Hz, 2H), 4.36 (m, 2H), 2.98 (t, 2H), 2.42 (m, 4H), 2.04 (qu, 2H), 1.25 (d,
J ˆ 20.7 Hz, 9H); 13C NMR (50 MHz, CDCl3): d ˆ 177.3, 171.2, 136.8, 132.5,
129.4, 120.2, 67.1 (d, J ˆ 9.4 Hz), 37.1 (d, J ˆ 704 Hz), 33.1, 24.3, 20.7; 31P
NMR (81 MHz, CDCl3): d ˆ 55.8; IR (CHCl3): nÄ ˆ 3021, 1711, 1604, 1519,
7-[(Pivalyloxymethyl)oxy]-4-(carboxymethyl)-2H-benzopiran-2-one (1):
7-hydroxy-4-carboxymethyl-2H-benzopiran-2-one 4 (150 mg, 0.68 mmol)
was dissolved in anhydrous DMF (3 mL), and NaH (35 mg , 55% in oil, 2.1
Â
eq.) was added at 208C. The temperature was raised to 258C for 30 min,
then cooled again to 608C, and a solution of iodomethyl pivalate (250 mg,
1.5 equiv; prepared from the chloride by reaction with NaI and purified by
distillation) in anhydrous THF (1 mL) was added dropwise. The temper-
ature was maintained at 608C for 2 h, then the mixture was stirred for a
further 10 h up to 258C. The reaction was poured into aq. 0.1n HCl and
extracted twice with ethyl acetate. The organic phase was dried (Na2SO4)
and concentrated, and the residue dissolved in MeOH (1 mL) buffered with
NH3/NH4Cl. After 6 h at 258C, the solution was evaporated and the residue
purified by preparative reverse-phase HPLC (gradient 25 ± 35% acetoni-
trile in 20 minutes). After lyophylization, the main fraction gave the
desired product 1. Yield: 6.5 mg (5%); 1H NMR (300 MHz, CDCl3): d ˆ
7.55 (d, J ˆ 8.8 Hz, 1H), 7.01 (dd, J ˆ 8.8 Hz, J ˆ 2.2 Hz, 1H), 6.97 (d, J ˆ
2.2 Hz, 1H), 6.20 (s, 1H), 5.82 (s, 2H), 2.43 (s, 2H), 1.22 (s, 9H); 13C NMR
(75 MHz, CDCl3): d ˆ 174.2, 173.4, 160.9, 159.7, 155.0, 152.2, 125.8, 115.2,
113.3, 113.0, 103.5, 85.0, 39.0, 26.9, 18.7; IR (CHCl3): nÄ ˆ 3265, 2985, 1754,
‡
‡
1414, 1216, 760 cm 1; MS (SIMS): 371 [M] , 297 [M tBuOH] , 234 [M
tBuPO3H] , 120.
‡
Carrier-protein conjugates of hapten 2: The following four solutions were
prepared:
A: EDCI (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochlor-
ide, 44 mg) in DMF (0.125 mL) and water (0.025 mL), c ˆ 1.5 m.
B: N-Hydroxysuccinimide (25 mg) in DMF (0.09 mL) and water
(0.04 mL), c ˆ 1.6 m.
‡
1724, 1619, 1243, 1109, 997 cm 1; MS (EI): 290 [M CO2] , 259, 174, 147,
57.
Chem. Eur. J. 2001, 7, No. 21
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4609 $ 17.50+.50/0
4609

J.-L. Reymond et al.
FULL PAPER
General procedure for the synthesis of oxymethyl esters of 7-hydroxy-2H-
7-(Isobutyroyl)oxy-2H-1-benzopiran-2-one (11): Colorless solid; yield:
180 mg (0.77 mmol, 63%); m.p. 102 ± 1038C; Rf ˆ 0.3 (hexane/AcOEt
8:2); 1H NMR (300 MHz, CDCl3): d ˆ 7.69 (d, J ˆ 9.6 Hz, 1H), 7.48 (d, J ˆ
8.4 Hz, 1H), 7.08 (d, J ˆ 2.2 Hz, 1H), 7.04 (dd, J ˆ 8.4, 2.2 Hz, 1H), 6.40 (d,
J ˆ 9.6 Hz, 1H), 2.84 (hept, J ˆ 7.0 H, 1Hz), 1.34 (s, J ˆ 7.0 Hz, 6H);
13C NMR (50 MHz, CDCl3): d ˆ 174.6, 160.8, 159.7, 155.5, 143.1, 129.0,
1-benzopiran-2-one or 4-nitrophenol:
A solution of 7-hydroxy-2H-1-
benzopiran-2-one (300 mg, 1.85 mmol) in dry DMF (3 mL) was treated
with NaH (118 mg, 55% suspension in oil, 1.5 equiv). After 30 min at 258C,
the reaction was cooled to 608C, and the alkylating agent was added as a
solution in dry THF (1 mL). The temperature was maintained at 608C for
2 h, then slowly raised to 258C overnight. Subsequently, the reaction
mixture was poured into aq. 1n HCl (50 mL) and extracted twice with
CH2Cl2. After evaporation of the organic phase, the residue was purified by
preparative reverse-phase HPLC.
114.3, 114.0, 113.5, 103.5, 33.6, 20.4; IR (CHCl3): nÄ ˆ 2976, 1754, 1716, 1616,
1
1399, 1266, 1145, 1130, 1095, 990, 832 cm
.
7-(Propanoyl)oxy-2H-1-benzopiran-2-one (12): Colorless solid; yield:
226 mg (1.0 mmol, 84%); m.p. 948C; 1H NMR (300 MHz, CDCl3): d ˆ
7.70 (d, J ˆ 9.6 Hz, 1H), 7.49 (d, J ˆ 8.4 Hz, 1H), 7.12 (d, J ˆ 2.2 Hz, 1H),
7.05 (dd, J ˆ 8.4, 2.2 Hz, 1H), 6.40 (d, J ˆ 9.6 Hz, 1H), 2.64 (q, J ˆ 7.4 Hz,
2H), 1.29 (t, J ˆ 7.4 Hz, 3H); 13C NMR (50 MHz, CDCl3): d ˆ 172.2, 160.4,
154.7, 153.3, 142.8, 128.5, 118.4, 116.5, 116.0, 110.4, 27.7, 8.9; IR (CHCl3):
nÄ ˆ 2976, 1754, 1716, 1616, 1399, 1266, 1145, 1130, 1095, 990, 832 cm 1; MS
The alkylating agents bromomethyl acetate (distilled at 808C, 180 mbar),
bromomethyl propanoate (distilled at 708C, 190 mbar), and chloromethyl
isobutyrate (distilled at 858C, 170 mbar) were obtained as pure compounds
in 30 ± 50% yield by adding the acyl chloride (20 g) dropwise over 1 h to an
ice-cold mixture of paraformaldehyde (1.0 equiv) and ZnCl2 (0.02 equiv),
then heating the reaction at 558C for 15 h, and distilling the product under
reduced pressure.[18] Iodomethyl isobutyrate was obtained from chloro-
methyl isobutyrate by reaction with NaI in acetonitrile (24 h).
‡
(EI): 218 [M] , 162, 134, 105, 77, 57.
7-Acetoxy-2H-1-benzopiran-2-one (13): Colorless solid; yield: 195 mg
(0.95 mmol, 79%); m.p. 1408C; 1H NMR (300 MHz, CDCl3): d ˆ 7.69 (d,
J ˆ 9.6 Hz, 1H), 7.48 (d, J ˆ 8.4 Hz, 1H), 7.12 (d, J ˆ 2.2 Hz, 1H), 7.05 (dd,
J ˆ 8.4, 2.2 Hz, 1H), 6.39 (d, Jˆ 9.6 Hz, 1H), 2.34 (s, 3H); 13C NMR
(50 MHz, CDCl3): dˆ 169.4, 161.0, 155.3, 153.8, 143.5, 129.2, 119.7, 117.3,
116.7, 111.1, 21.7; IR (CHCl3): nÄ ˆ 3079, 1740, 1620, 1565, 1506, 1427, 1400,
7-[(Pivalyloxymethyl)oxy]-2H-benzopiran-2-one (5): Colorless solid;
yield: 228 mg (0.82 mmol, 44%); Rf ˆ 0.5 (hexane/AcOEt 7:3); 1H NMR
(300 MHz, CDCl3): d ˆ 7.66 (d, J ˆ 9.6 Hz, 1H), 7.42 (d, J ˆ 8.4 Hz, 1H),
7.02 (d, 4J ˆ 2.6 Hz, 1H), 6.95 (dd, J ˆ 2.6, 8.4 Hz, 1H), 6.31 (d, J ˆ 9.6 Hz,
1H), 5.82 (s, 2H), 1.20 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 177.0, 160.8,
159.8, 155.5, 143.1, 128.9, 114.3, 114.0, 113.6, 103.4, 84.9, 38.9, 26.9; IR
‡
1372, 1198, 1121, 1011, 988, 906 cm 1; MS (EI): 204 [M] , 162, 134, 105, 78, 43.
Synthesis of pivaloyloxymethyl derivatives of phenols: A solution of phenol
(300 mg) in dry DMF (4 mL) was treated with NaH (55% suspension in oil,
1.5 equiv) at 08C. After 10 min, the temperature was lowered to 608C,
and iodomethyl pivalate (1.5 equiv) as a solution in dry THF (4 mL) was
slowly added. The temperature was allowed to rise to 258C over 2 hours.
The mixture was poured into aq. 1n NaOH and extracted twice with ethyl
acetate. The organic phase was dried over Na2SO4 and concentrated, and
the residue purified by flash chromatography.
‡
(CHCl3): nÄ ˆ 3020, 1736, 1618, 1216, 762 cm 1; MS (EI): 276 [M] ; 175; 162;
134; 85; 57.
7-(iso-Propylcarbonyloxymethyl)oxy-2H-1-benzopiran-2-one (7): Color-
less solid; yield: 107 mg (0.41 mmol, 30%); m.p. 548C; 1H NMR
(300 MHz, CDCl3): d ˆ 7.68 (d, J ˆ 9.6 Hz, 1H), 7.42 (d, J ˆ 8.5 Hz, 1H),
7.02 (d, J ˆ 2.2 Hz, 1H), 6.95 (dd, J ˆ 8.5, 2.2 Hz, 1H), 6.31 (d, J ˆ 9.6 Hz,
1H), 5.82 (s, 2H), 2.62 (hept, J ˆ 7.0 Hz, 1H), 1.19 (d, J ˆ 7.0 Hz, 6H);
13C NMR (75 MHz, CDCl3): d ˆ 175.6, 160.8, 159.7, 155.5, 143.1, 128.9,
114.3, 114.0, 113.5, 103.4, 84.7, 33.9, 18.7; IR (CHCl3): nÄ ˆ 2977, 1732, 1616,
2,2-Dimethylpropionic acid (4-nitrophenoxy)methyl ester (6): Colorless
solid; yield: 450 mg (1.78 mmol, 82%); m.p. 135 ± 1378C; Rf ˆ 0.8 (hexane/
AcOEt 1:1); 1H NMR (300 MHz, CDCl3): d ˆ 8.23 (d, J ˆ 9.2 Hz, 2H), 7.13
(d, J ˆ 9.2 Hz, 2H), 5.85 (s, 2H), 1.21 (s, 9H); 13C NMR (50 MHz, CDCl3):
d ˆ 176.9, 161.5, 142.9, 125.8, 115.9, 84.6, 38.9, 26.8; IR (CHCl3): nÄ ˆ 3120,
‡
1506, 1397, 1343, 1278, 976, 759 cm 1; MS (EI): 262 [M] , 232, 175, 163, 134,
71, 43; elemental analysis calcd (%) for C14H14O5 (262.26): C 64.12, H 5.38;
found: C 64.02, H 5.32.
7-[(Ethylcarbonyloxymethyl)oxy]-2H-1-benzopiran-2-one (8): Colorless
solid; yield: 110 mg (0.44 mmol, 24%); m.p. 838C; 1H NMR (300 MHz,
CDCl3): d ˆ 7.67 (d, J ˆ 9.6 Hz, 1H), 7.42 (d, J ˆ 8.4 Hz, 1H), 7.03 (d, J ˆ
2.6 Hz, 1H), 6.95 (dd, J ˆ 8.4, 2.6 Hz, 1H), 6.31 (d, J ˆ 9.6 Hz, 1H), 5.82 (s,
2H), 2.42 (q, J ˆ 7.4 Hz, 2H), 1.19 (t, J ˆ 7.4 Hz, 3H); 13C NMR (75 MHz,
CDCl3): d ˆ 173.1, 160.8, 159.7, 155.5, 143.1, 128.9, 114.2, 113.5, 113.9, 103.4,
84.7, 27.4, 8.7; IR (CHCl3): nÄ ˆ 3080, 1759, 1712, 1621, 1507, 1413, 1345, 1280,
‡
2979, 1744, 1594, 1348, 1135, 850 cm 1; MS (EI): 253 [M] , 223, 180, 152, 85,
57.
2,2-Dimethylpropionic acid (2-nitro-phenoxy)methyl ester (14): Colorless
solid; yield: 431 mg (1.78 mmol, 79%), Rf ˆ 0.7 (hexane/AcOEt 1:1);
1H NMR (300 MHz, CDCl3): d ˆ 7.84 (dd, J ˆ 8.1, 1.5 Hz, 1H), 7.55 (ddd
J ˆ 8.5, 8.4, 1.5 Hz, 1H,), 7.28 (dd, J ˆ 8.5, 1.1 Hz, 1H), 7.16 (ddd, J ˆ 8.4, 8.1,
1.1 Hz, 1H), 5.85 (s, 2H), 1.21 (s, 9H); 13C NMR (50 MHz, CDCl3): d ˆ
173.4 (C3); 149.9, 140.7, 133.9, 125.5, 122.9, 117.8, 86.3, 38.8, 26.8; IR
(CHCl3): nÄ ˆ 3028, 2979, 1752, 1609, 1532, 1482, 1356, 1244, 1218, 1147,
‡
1162, 1044, 959 cm 1; MS (EI): 248 [M] , 218, 175, 162, 134, 105, 57.
7-[(Acetoxymethyl)oxy]-2H-1-benzopiran-2-one (9): Colorless solid;
yield: 110 mg (0.43 mmol, 25%); m.p. 1228C; 1H NMR (300 MHz, CDCl3):
d ˆ 7.67 (d, J ˆ 9.6 Hz, 1H), 7.42 (d, J ˆ 8.4 Hz, 1H), 7.03 (d, J ˆ 2.6 Hz,
1H), 6.95 (dd, J ˆ 8.4, 2.6 Hz, 1H), 6.31 (d, J ˆ 9.6 Hz, 1H), 5.82 (s, 2H),
2.14 (s, 3H); 13C NMR (50 MHz, CDCl3): d ˆ 169.6, 160.8, 159.7, 155.5,
143.1, 129.0, 114.3, 114.0, 113.5, 103.5, 84.7, 27.4; IR (CHCl3): nÄ ˆ 3084, 1763,
‡
1122, 1092, 1023, 983, 858, 772, 747 cm 1; MS (EI): 253 [M] , 223, 152, 139,
122, 85, 57.
2,2-Dimethylpropionic acid (2-methoxy-4-formylphenoxy)methyl ester
(15): Prepared from vanillin (200 mg, 1.3 mmol). Colorless solid; yield:
220 mg (0.83 mmol, 63%); Rf ˆ 0.45 (hexane/AcOEt 9:1); 1H NMR
(300 MHz, CDCl3): d ˆ 9.70 (s, 1H), 7.27 (m, 2H), 7.03 (d, J ˆ 8.8 Hz,
1H), 5.70 (s, 2H), 3.75 (s, 3H), 1.02 (s, 9H); 13C NMR (50 MHz, CDCl3): d ˆ
190.6, 176.8, 151.1, 150.2, 131.9, 125.5, 115.1, 110.0, 85.5, 55.8, 38.7, 26.7; IR
(CHCl3): nÄ ˆ 3026, 2978, 1749, 1687, 1594, 1509, 1466, 1425, 1395, 1268, 1227,
‡
1720, 1621, 1508, 1430, 1349, 1221, 1169, 1052, 970 cm 1; MS (EI): 234 [M] ,
204, 175, 162, 134, 105, 43.
General procedure for the synthesis of 7-hydroxy-2H-1-benzopiran-2-one
esters:
A
solution of 7-hydroxy-2H-1-benzopiran-2-one (200 mg,
1.23 mmol) in dry DMF (3 mL) was treated with NaH (118 mg, 55%
suspension in oil, 2.3 equiv). After 30 minutes at 258C, the reaction was
cooled to 08C, and the acyl chloride (1.5 equiv) as a solution in dry THF
(1 mL) was added dropwise. After 2 hours at 258C, the reaction was poured
into aq. 1n HCl (50 mL) and extracted twice with CH2Cl2. The organic
phase was dried over Na2SO4, and the residue purified either by flash
chromatography or by reverse-phase HPLC.
‡
1112, 1025, 980, 768, 752, 668 cm 1; MS (EI): 266 [M] , 236, 165, 152, 85, 57.
2,2-Dimethylpropionic acid (4-methyl-phenoxy)methyl ester (16): Color-
less oil; yield: 358 mg (1.43 mmol, 65%); Rf ˆ 0.4 (hexane/AcOEt 95:5);
1H NMR (300 MHz, CDCl3): d ˆ 7.95 (d, J ˆ 8.8 Hz, 2H), 7.07 (d, J ˆ
8.8 Hz, 2H), 5.82 (s, 2H), 1.20 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ
196.6, 177.1, 160.5, 131.8, 130.5, 115.5, 84.7, 38.9, 26.8, 26.3; IR (CHCl3): nÄ ˆ
‡
2981, 1749, 1679, 1603, 1510, 1273, 1236, 1123, 1028, 757 cm 1; MS (FAB ):
7-Pivaloyloxy-2H-benzopiran-2-one (10): Colorless solid; yield: 228 mg
(0.97 mmol, 79%); m.p. 1398C; Rf ˆ 0.5 (hexane/AcOEt 7:3); 1H NMR
(300 MHz, CDCl3): d ˆ 7.70 (d, J ˆ 9.6 Hz, 1H), 7.49 (d, J ˆ 8.4 Hz, 1H),
7.10 (d, J ˆ 2.6 Hz, 1H), 7.02 (dd, J ˆ 8.4, 2.6 Hz, 1H), 6.40 (d, J ˆ 9.6 Hz,
1H), 1.38 (s, 9H); 13C NMR (50 MHz, CD3OD): d ˆ 176.4, 160.3, 154.7,
153.7, 142.8, 128.4, 118.3, 116.2, 115.9, 110.3, 39.2, 27.0; IR (CHCl3): nÄ ˆ
‡
251 [M‡H] , 221, 165, 137.
2,2-Dimethylpropionic acid (4-cyanophenoxy)methyl ester (17): Colorless
solid; yield: 387 mg (1.66 mmol, 66%); Rf ˆ 0.6 (hexane/AcOEt 4:1);
1H NMR (300 MHz, CDCl3): d ˆ 7.62 (d, J ˆ 8.8 Hz, 2H), 7.10 (d, J ˆ
8.8 Hz, 2H), 5.81 (s, 2H), 1.20 (s, 9H); 13C NMR (50 MHz, CDCl3): d ˆ
177.0, 159.9, 134.1, 118.7, 116.5, 106.1, 84.5, 38.9, 26.9; IR (CHCl3): nÄ ˆ 3024,
2979, 2230, 1751, 1608, 1510, 1241, 1119, 1027, 983, 839, 765, 756 cm 1; MS
‡
2977, 1744, 1723, 1621, 1264, 1232, 989 cm 1; MS (EI): 246 [M] , 162, 134,
105, 85, 77, 57; elemental analysis calcd (%) for C14H14O4 (246.26): C 68.28,
H 5.73; found: C 68.28, H 5.69.
‡
‡
(FAB ): 234 [M‡H] , 149, 135, 123, 119.
4610
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4610 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 21

Pivalase Catalytic Antibodies
4604±4612
2,2-Dimethylpropionic acid (4-chlorophenoxy)methyl ester (18): Colorless
solid; yield: 387 mg (1.66 mmol, 58%); Rf ˆ 0.7 (hexane/AcOEt 95:5);
1H NMR (300 MHz, CDCl3): d ˆ 7.27 (d, J ˆ 8.8 Hz, 2H), 6.96 (d, J ˆ
8.8 Hz, 2H), 5.75 (s, 2H); 1.21 (s, 9H); 13C NMR (50 MHz, CDCl3): d ˆ
177.2, 155.5, 129.5, 127.7, 122.8, 85.7, 38.9, 26.9; IR (CHCl3): nÄ ˆ 3034, 2978,
1748, 1493, 1281, 1231, 1156, 1157, 1122, 1055, 1029, 973, 829 cm 1; MS
(SigmaA5284) as a selective agent. A minimum of 107 myeloma cells were
used to fuse one spleen. Three days before the scheduled fusion, and at
least 40 days after the second hapten ± KLH injection, four mice were given
a final intravenous injection of 50 mg KLH ± conjugate in PBS. On the
fusion day, the spleen cells of each mouse were collected and washed with
no-serum media, together with the myelomas. The spleen cells and the
myelomas were then fused with PEG1500 (Roche783641). The fused cells
were diluted in 150 mL media containing HAT (H ˆ hypoxanthine
9  10 3 m, A ˆ aminopterin 2  10 5 m, Tˆ thymidine 8  10 4 m)
(SigmaH0262) and plated out in (10) 96-well cell-culture plates (Corning
Costar3598).
‡
‡
(EI ): 242 [M] , 212, 141, 128, 111, 85, 57.
2,2-Dimethylpropionic acid (4-(N-acetylamino)phenoxy)methyl ester (19):
Colorless solid; yield: 187 mg (0.70 mmol, 71%); Rf ˆ 0.3 (hexane/AcOEt
9:1); 1H NMR (300 MHz, CDCl3): d ˆ 7.43 (d, J ˆ 8.8 Hz, 2H), 7.00 (d, J ˆ
8.8 Hz, 2H), 5.74 (s, 2H), 2.17 (s, 3H), 1.21 (s, 9H); 13C NMR (75 MHz,
CDCl3): d ˆ 177.4, 168.5, 153.5, 133.0, 121.6, 116.6, 86.1, 38.8, 26.9, 24.2; IR
(CHCl3): nÄ ˆ 2979, 1744, 1659, 1608, 1505, 1412, 1318, 1216, 1125, 1057, 975,
Cell culture: The fused cells were fed twice a week with media containing
HT (H ˆ hypoxanthine 9  10 3 m, Tˆ thymidine 8  10 4 m) (Sigma
H0137) for two weeks. After two weeks, the plates were tested by ELISA
to identify those hybridomas that were producing antigen binding.
Preclones were also selected for their ability to catalyze the hydrolysis of
1 (see below). Positive preclones were cultured further and cloned twice by
limited dilution, and finally produced in 1 L cultures or with Integra
CL1000 (Integra Biosciences 90005). After removal of the cells by
centrifugation, antibodies were purified from the cell-culture media by
ammonium sulfate precipitation and protein-G-affinity chromatography
(Amersham Pharmacia Biotech 27-0886-01). The antibodies were finally
dialyzed into PBS and concentrated to 3 ± 7 mgmL 1. Although repeated
freeze ± thaw cycles may be damaging, the antibodies can be stored in this
form at 208C without loss of activity for at least several months. Isotypes
were determined by means of an isotyping kit (Pierce no. 37501).
‡
‡
824 cm 1; MS (EI ): 265 [M] , 235, 164, 151, 122, 109, 85, 57.
(Pivaloyloxymethyl)fluorescein (20): A solution of fluorescein (200 mg,
0.45 mmol) in dry DMF (2 mL) and dry THF (2 mL) at 08C was treated
with NaH (29 mg, 55% suspension in oil, 1.5 equiv). After 10 min, the
temperature was lowered to 608C, and a solution of iodomethylpivalate
(174 mg, 1.2 equiv) in dry THF (4 mL) was added dropwise. The temper-
ature was raised to 258C over 2 hours, after which the solution was poured
into aq. 1n HCl (50 mL) and extracted twice with ethyl acetate. The organic
phase was dried over Na2SO4, and the residue purified by flash
chromatography to give 20 as a yellow solid. Yield: 80 mg (0.18 mmol,
30%); Rf ˆ 0.7 (hexane/AcOEt 1:9); 1H NMR (300 MHz, CDCl3): d ˆ 8.03
(m, 1H), 7.66 (m, 2H), 7.26 (m, 1H), 6.72 (m, 3H), 6.55 (m, 1H), 5.78 (s,
2H), 1.22 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 177.5, 176.5, 170.0, 158.5,
158.2, 153.0, 152.4, 135.2, 129.8, 129.3, 126.6, 125.1, 124.0, 113.2, 110.7, 103.7,
Screening for catalysis: A fluorescence assay for catalysis with substrate 1
was performed on all preclones when the cell culture reached a volume of
5 mL, before the first subcloning dilution. Thus 5 mL of cell-culture
supernatant were passed on 100 mg of protein-G gel (Gammabind Plus
sepharose, Pharmacia Biotech) in a cotton-plugged Pasteur pipette. The gel
was washed with PBS (2 Â 1 mL) and 0.1m aq. NaCl (2 Â 1 mL), and finally
eluted with citrate (50mm, 2 Â 150 mL, pH 2.7). The eluted acidic buffer
containing the antibody, was neutralized with Tris-base (tris(hydroxy-
methyl)aminomethane; 1m, 40 mL). 90 mL of this antibody-containing
solution were transferred to a well of a 96-well microtiter plate and mixed
with borate buffer (0.5m, 5 mL, pH 8.8) and substrate 1 (5 mL of a 2mm
stock solution in DMF). 50 mL of the solution was immediately transferred
to another well containing a solution of hapten 2 or 3 (0.5 mL, 2mm in PBS).
The time dependence of fluorescence increase was then followed over the
next 3 h. The difference in apparent rate between the antibody and the
antibody ‡ inhibitor samples was used as a criterion for specific catalysis.
The purification/assay procedure was performed easily in parallel with up
to 40 different cell-culture samples, and was performed repeatedly with
each individual hybridoma at all stages of cell culture.
‡
‡
103.1, 85.3, 38.9, 26.9; MS (FAB ): 447 [M‡H] , 417, 333, 287, 271, 119.
(N-Pivaloyloxymethyl) pyridinium trifluoroacetate (21): Pyridine (300 mg)
was treated dropwise under stirring with a solution of iodomethyl pivalate
(1.5 equiv) in hexane (5 mL). After 4 hours at 258C, the precipitate was
filtered off and purified by preparative RP-HPLC to give 21 as a pale
yellow solid. Yield: 580 mg (1.89 mmol, 50%); m.p. 648C; 1H NMR
(300 MHz, CDCl3): d ˆ 9.15 (d, J ˆ 7.5 Hz, 2H), 8.58 (t, J ˆ 7.5 Hz, 1H),
8.13 (t, J ˆ 7.5 Hz, 2H), 6.53 (s, 2H); 1.21 (s, 9H); 13C NMR (50 MHz,
CDCl3): d ˆ 175.7, 148.1, 145.8, 128.9, 80.8, 39.6, 27.3; IR (CHCl3): nÄ ˆ 3096,
2990, 1754, 1634, 1504, 1484, 1410, 1200, 1137, 1111, 1049 cm 1; HRMS calcd
for C11H16NO2: 194.118104, found: 194.117460
7-(tert-Butoxycarbonyl)oxy-2H-1-benzopiran-2-one (22):
A solution of
umbelliferone (215 mg, 1.34 mmol) in dry CH2Cl2 (1 mL) and dry DMF
(4 mL) at 08C was treated with di-tert-butyldicarbonate (440mg, 1.5 equiv)
and pyridine (0.16 mL, 1.5 equiv). After 2 h at 258C, the reaction mixture
was directly purified by flash chromatography (hexane/ethyl acetate 6:4) to
give 22 as a white solid. Yield: 190 mg (0.725 mmol, 55%); m.p. ˆ 1098C;
1H NMR (300 MHz, CDCl3): d ˆ 7.69 (d, J ˆ 9.6 Hz, 1H), 7.48 (d, J ˆ
8.5 Hz, 1H), 7.21 (d, J ˆ 2.6 Hz, 1H), 7.14 (dd, J ˆ 9.6, 2.6 Hz, 1H), 6.40
(d, J ˆ 9.6 Hz, 1H), 1.58 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 160.2,
154.6, 153.4, 150.7, 142.7, 128.4, 117.8, 116.4, 115.9, 109.9, 84.4, 27.5; IR
(CHCl3): nÄ ˆ 3021, 1741, 1623, 1285, 1250, 1218, 1144, 1125, 775, 744,
Kinetics: Reactions were followed either by fluorescence in round-bottom
polypropylene 96-well plates (Costar) with a Cytofluor II Fluorescence
Plate Reader (Perceptive Biosystems, filters lex ˆ 360 Æ 20 nm, lem ˆ 460 Æ
20 nm), or by UV/Vis spectroscopy at l ˆ 405 nm (for nitrophenyl substrate
6) in half-area flat-bottom clear polystyrene cell-culture plates (Corning ±
Costar) on a UV Spectramax250 instrument from Molecular Devices.
Solutions containing antibodies (90 mL at 0.2 mgmL 1 in either 20mm bis-
tris [bis(hydroxyethyl)-tris(hydroxymethyl)aminomethane] pH 7.55 or
20mm borate pH 8.8) were thermostated at 318C. The reactions were
initiated by addition of DMF (5 mL) and of a properly prediluted solution
(5 mL) of substrate in DMF/water (1:1). The activity screening concen-
tration was 100mm, kinetic measurement concentrations were 17, 26, 39, 59,
89, 133, 200, and 300mm (as obtained by a 2ꢂ3 serial dilution of substrate in
the 20X stock starting with 6mm). The fluorescence or UV signals were
converted to product concentration according to a calibration curve with
either 4, umbelliferone, or 4-nitrophenol. The net reaction rate of
background was used to calculate the kinetic parameters according to the
Michaelis ± Menten model, as already reported.[20] The lines obtained from
8 datapoints had correlation factors R2 > 0.995. Systematic errors of Æ10%
must be assumed as a result of uncertainty in the antibody concentrations
and pipetting inaccuracy of small volumes.
‡
‡
670 cm 1; MS (EI ): 262 [M] , 247, 203, 162, 145, 134, 117, 57.
1-(Pivalyloxymethyl)-2-oxo-3,3-diethyltriazene (23):
A
suspension of
1-(N,N-diethylamino)diazen-1-ium-1,2-diolate (sodium salt; DEA/NO;
100 mg, 0.64 mmol) in DMF (2 mL) at 08C was treated with N,N-
diisopropylethylamine (0.05 mL) and chloromethyl pivalate (0.1 mL,
1.1 equiv). The solution was stirred for 30 min at 08C and for 12 h at
258C, then directly deposited onto silica gel. Elution with hexane/ethyl
acetate (95:5) gave product 23 as a colorless solid. Yield: 30 mg (0.12 mmol,
18%); Rf ˆ 0.3 (hexane/AcOEt 6:4); 1H NMR (300 MHz, CDCl3): d ˆ 5.82
(s, 2H), 3.18 (q, 4H, J ˆ 7.35 Hz), 1.19 (s, 9H), 1.1 (t, 6H, J ˆ 7.35 Hz);
13C NMR (75 MHz, CDCl3): d ˆ 176.7, 87.5, 48.3, 38.7, 26.8, 11.4; IR
(CHCl3): nÄ ˆ 2983, 1752, 1519, 1481, 1386, 1281, 1238, 1168, 1131, 1054,
‡
1033, 963, 759 cm 1; MS (SIMS): 248 [M‡H] , 187.
Immunization:[19] Two sets of four mice (129GIX/boy ‡ ) were immunized
with the KLH conjugates of either hapten 2 or 3. Each mouse received two
intraperitoneal injections of hapten ± KLH conjugate (100 mg in 200 mL
PBS emulsified with MPL ‡ TDM adjuvent (SigmaM6536)) one on day 1
and one on day 14. Samples of serum were taken by means of a tail bleed on
day 21 to estimate the titer.
The catalytic constants kcat are reported for one active site, assuming a
molecular weight of 150 kDa and two catalytic sites for each antibody.
Exact active-site titrations were carried out with all antibodies by using
hapten 3 as a tight binding ligand. These titrations showed the expected
active-site concentration within 10%.
Fusion: Myelomas (NS1) were grown for a week in RPMI-1640 (Sigma
R0883) with 10% FCS (Fetal Calf Serum) (SigmaF2442) and 8-azaguanine
Chem. Eur. J. 2001, 7, No. 21
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4611 $ 17.50+.50/0
4611

J.-L. Reymond et al.
FULL PAPER
[9] N. Bensel, N. Bahr, M. T. Reymond, C. Schenkels, J.-L. Reymond,
Helv. Chim. Acta 1999, 82, 44 ± 52.
Acknowledgement
[10] a) P. Wirsching, J. A. Ashley, C. H. Lo, K. D. Janda, R. A. Lerner,
Science 1995, 270, 1775 ± 1782; b) J. Wagner, R. A. Lerner, C. F.
Barbas, Science 1995, 270, 1797 ± 1800.
This work was supported by the Swiss National Science Foundation and the
University of Bern, Switzerland.
[11] a) D. Seyferth, R. S. Marmor, J. Organomet. Chem. 1973, 59, 237 ± 245;
b) J. Clay, J. Am. Chem. Soc. 1951, 73, 892 ± 894.
[12] C. Yuan, S. Li, Z. Chen, Synthesis 1988, 3, 186 ± 189.
[13] a) G. Yang, K. Zhao, D. W. Landry, Tetrahedron Lett. 1998, 39, 2449 ±
2450; b) G. Sosnovsky, N. U. M. Rao, S. W. Li, H. M. Swartz, J. Org.
Chem. 1989, 54, 3667 ± 3674.
[14] G. Kohler, C. Milstein, Nature 1975, 256, 495 ± 497.
[15] P. Tijssen in ªPractice and Theory of Enzyme Immunoassaysº (Eds.:
R. H. Burdon, P. H. Knippenberg), Elsevier, Amsterdam, 1985.
[16] C. D. Hall, V. T. Le, J. Chem. Soc. Perkin Trans. 2 1998, 1483 ± 1488,
and references therein. These authors report that the hydrolysis of
acyloxyalkyl esters related to 1 by cleavage of the acetal bond occurs
in an SN1 manner under acid-catalyzed conditions, but requires a very
low pH (<1) for derivatives of acidic phenols. An SN2-type reaction of
hydroxide at the acetal carbon has never been observed with
acyloxymethyl esters.
[1] a) P. G. Schultz, R. A. Lerner, Science 1995, 269, 1835 ± 1842; b) R. A.
Lerner, S. J. Benkovic, P. G. Schultz, Science 1991, 252, 659 ± 667; c) J.-
L. Reymond, Top. Curr. Chem. 1999, 200, 59 ± 93; d) J. D. Stevenson,
N. R. Thomas, Nat. Prod. Rep. 2000, 535 ± 577; e) B. Golinelli-
Pimpaneau, Curr. Opin. Struct. Biol. 2000, 10, 697 ± 708; f) D. Hilvert,
Annu. Rev. Biochem. 2000, 69, 751 ± 793.
[2] P. J. Hudson, Curr. Opin. Immunol. 1999, 11, 548 ± 557.
[3] a) B. Mets, G. Winger, C. Cabrera , S. Seo, S. Jamdar, G. Yang, K.
Zhao, R. J. Briscoe, R. Almonte, J. H. Woods, D. W. Landry, Proc.
Natl. Acad. Sci. USA 1998, 95, 10176 ± 10181; b) P. Vayron, P. Y.
Renard, A. Valleix, C. Mioskowski, Chem. Eur. J. 2000, 6, 1050 ± 1063.
[4] Overview: C. Rader, B. List, Chem. Eur. J. 2000, 6, 2091 ± 2095.
[5] K. D. Bagshawe, S. K. Sharma, P. J. Burke, R. G. Melton, R. J. Knox,
Curr. Opin. Immunol. 1999, 11, 579 ± 583.
[6] J. McCafferty, D. R. Glover, Curr. Opin. Struct. Biol. 2000, 10, 417 ±
420.
[17] J. E. Saavedra, P. J. Shami, L. Y. Wang, K. M. Davies, M. N. Booth,
M. L. Citro, L. K. Keefer, J. Med. Chem. 2000, 43, 261 ± 269.
[18] R. P. Iyer, D. Yu, N. Ho, S. Agrawal, Synth. Commun. 1995, 25, 2739 ±
2749.
[19] E. Harlow, D. Lane in ªAntibodies, a Laboratory Manualº, Cold
Spring Harbor Laboratory, 1988.
[7] a) H. Miyashita, Y. Karaki, M. Kikuchi, I. Fujii, Proc. Natl. Acad. Sci.
USA 1993, 90, 5337 ± 5340; b) D. A. Campbell, B. Gong, L. M.
Kochersperger, S. Yonkovich, M. A. Gallop, P. G. Schultz, J. Am.
Chem. Soc. 1994, 116, 2165 ± 2166; c) P. Wentworth, A. Datta, D.
Blakey, T. Boyle, L. J. Partridge, G. M. Blackburn, Proc. Natl. Acad.
Sci. USA 1996, 93, 799 ± 803; d) O. Kristensen, D. G. Vassylyev, F.
Tanaka, K. Morikawa, I. Fujii, J. Mol. Biol. 1998, 281, 501 ± 511; e) D.
Shabat, C. Rader, B. List, R. A. Lerner, C. F. Barbas, Proc. Natl. Acad.
Sci. USA 1999, 96, 6925 ± 6930.
[20] J.-L. Reymond, G. K. Jahangiri, C. Stoudt, R. A. Lerner, J. Am. Chem.
Soc. 1993, 115, 3909 ± 3917.
[8] See, for example: R. Niemi, J. Vepsalainen, H. Taipale, T. J. Jarvinen,
J. Med. Chem. 1999, 42, 5053 ± 5058.
Received: April 17, 2001 [F3206]
4612
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0721-4612 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 21