Antibody-catalyzed decarboxylative oxidation of vanillylmandelic acid

Article abstract of DOI:10.1021/ja9724811

The most important industrial process for the synthesis of vanillin is performed in two steps involving an electrophilic aromatic substitution of glyoxylic acid on guaiacol followed by an oxidative decarboxylation of the intermediary α-hydroxy acids formed, thereby producing not only vanillin, but also byproducts which have to be eliminated. In the present study, we took advantage of the high specificity of catalytic antibodies to improve the synthesis of vanillin. Among 11 monoclonal antibodies elicited against the quaternary ammonium hapten H3, antibody H3-12 was found to catalyze the oxidative decarboxylation of vanillylmandelic acid (VMA), the precursor of vanillin, in the presence of sodium metaperiodate. The kinetic data of the antibody-catalyzed reaction are consistent with an ordered binding mechanism. At pH 9.0, H3-12 catalyzed the transformation of VMA into vanillin with a k(cat) of 2.70 min^-1, a Michaelis-Menten constant K(a) for the binary complex of 260 μM, and a K(b) for the ternary complex of 2100 μM. The catalyzed reaction was fully inhibited by a hapten analogue with a K(i) of 10 μM. The fine specificity of anti-H3 monoclonal antibodies was determined using H3-related compounds with a competitive enzyme immunoassay. Controls demonstrating that catalytic activity is actually related to antibody binding, and mechanistic studies, are also presented.

Full text of DOI:10.1021/ja9724811

3332
J. Am. Chem. Soc. 1998, 120, 3332-3339
Antibody-Catalyzed Decarboxylative Oxidation of Vanillylmandelic
Acid
F. Taran,† P. Y. Renard,†,‡ H. Bernard,§ C. Mioskowski,*,† Y. Frobert,§ P. Pradelles,*,§ and
J. Grassi§
Contribution from the CEA, SerVice des Mole´cules Marque´es, DBCM, CEA,
SerVice de Pharmacologie et d’Immunologie, DRM, CE Saclay 91191 Gif sur YVette Cedex,
France, and CEB, BP 3, 91710 Vert le Petit, France
ReceiVed July 22, 1997
Abstract: The most important industrial process for the synthesis of vanillin is performed in two steps involving
an electrophilic aromatic substitution of glyoxylic acid on guaiacol followed by an oxidative decarboxylation
of the intermediary R-hydroxy acids formed, thereby producing not only vanillin, but also byproducts which
have to be eliminated. In the present study, we took advantage of the high specificity of catalytic antibodies
to improve the synthesis of vanillin. Among 11 monoclonal antibodies elicited against the quaternary ammonium
hapten H3, antibody H3-12 was found to catalyze the oxidative decarboxylation of vanillylmandelic acid (VMA),
the precursor of vanillin, in the presence of sodium metaperiodate. The kinetic data of the antibody-catalyzed
reaction are consistent with an ordered binding mechanism. At pH 9.0, H3-12 catalyzed the transformation
of VMA into vanillin with a kcat of 2.70 min-1, a Michaelis-Menten constant Ka for the binary complex of
260 µM, and a Kb for the ternary complex of 2100 µM. The catalyzed reaction was fully inhibited by a
hapten analogue with a Ki of 10 µM. The fine specificity of anti-H3 monoclonal antibodies was determined
using H3-related compounds with a competitive enzyme immunoassay. Controls demonstrating that catalytic
activity is actually related to antibody binding, and mechanistic studies, are also presented.
Introduction
fully controlled and two byproducts, 2-hydroxy-3-methoxy-
mandelic acid (o-VMA) and 4-hydroxy 5-methoxy-1,3-diman-
delic acid (di-VMA), are generated in a total 15% yield. The
oxidative decarboxylation, occurring in the second step of this
synthesis, leads to the formation of three aldehydes. To obtain
pure vanillin, a time-consuming separation of closely related
products must be achieved.
We focused our efforts on the second step of vanillin
synthesis, with a view to selective oxidative decarboxylation
of vanillylmandelic acid (VMA). So far, only two carbon-
carbon bond cleavage reactions catalyzed by abzymes have been
described, namely, retroaldolization7,28 and decarboxylation
Among the tools used to catalyze organic reactions, enzymes
are playing an increasingly important role due to their high
efficiency and selectivity. In addition, when compared to
conventional catalytic processes, enzymatic catalysis represents
an environmentally compatible and inexpensive approach likely
to replace chemical catalysis. However, the main drawback of
enzymes is the limited number of accessible reactions and
substrates.
To evaluate the potential of catalytic antibodies1-5 to mediate
transformations of a type that are of interest industrially, we
examined the concrete case of the industrial synthesis of vanillin
and sought to improve this process by the use of catalytic
antibodies. Vanillin, the well-known flavor, is a widely used
additive in the food and beverage industry. Although it can be
obtained from natural sources, the main supply (8000 tons per
year) comes from synthetic material. The industrial synthesis
of vanillin is currently realized from guaiacol through a two-
step procedure developed by Rhoˆne-Poulenc6 as displayed in
Scheme 1.
(7) (a) Flanagan, M. E.; Jacobsen, J. R.; Sweet, E.; Scultz, P. G. J. Am.
Chem. Soc. 1996, 118, 6078-6079. (b) Reymond, J. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2285-2287.
(8) Tarasow, T. M.; Lewis, C.; Hilvert, D. J. Am. Chem. Soc. 1994, 116,
7959-7963.
(9) Ashley, J. A.; Lo, C.-H. L.; McElhaney, G. P.; Wirsching, P.; Janda,
K. D. J. Am. Chem. Soc. 1993, 115, 2515-2516.
(10) Pisano, J. J.; Crout, J. R.; Abraham, D. Clin. Chem. Acta 1962, 7,
285-291.
(11) Wybenga, D.; Pileggi, V. J. Clin. Chem. Acta 1967, 16, 147-154.
(12) Felice, L. J.; Kissinger, P. T. Clin. Chem. Acta 1977, 76, 317-320.
(13) Wilson, M. B.; Nakane, P. K. In Immunofluorescence related
staining techniques; Knapp, W., Holubar, K., Wick, G., Eds.; Elsevier:
Amsterdam, 1978; p 215.
In the first step of this synthesis, the regioselectivity of the
electrophilic substitution of glyoxylic acid on guaiacol is not
† CEA, Service des Mole´cules Marque´es.
‡ CEB.
(14) Hsieh, L. C.; Stephans, J. C.; Schultz, P. G. J. Am. Chem. Soc.
1994, 116, 2167-2168.
(15) Courtois, J. Ann. Pharm. 1944, 2, 148.
(16) Yanaka, Y.; Katz, R.; Sarel, S. Tetrahedron Lett. 1968, 14, 1725-
1728.
§ CEA, Service de Pharmacologie et d’Immunologie.
(1) Pollack, S. J.; Jacobs, J. W.; Schultz, P. G. Science 1986, 234, 1570-
1573.
(2) Tramontano, A.; Janda, K. D.; Lerner, R. A. Science 1986, 234,
1566-1573.
(17) Fleury, M. P.; Courtois, J. Bull. Soc. Chem. Fr. 1948, 40, 190-
194.
(18) Fleury, M. P.; Courtois, J.; Perles, R.; Le Dizet, L. Compt. Rend.
Acad. Sci. 1953, 237, 1019-1021.
(19) Adler, E.; Falkehad, I.; Smith, B. Acta Chem. Scand. 1962, 16, 529-
540.
(3) Kirby, A. J. Angew. Chem., Int. Ed. Engl. 1996, 35, 707-727.
(4) Mader, M. M.; Bartlett, P. A. Chem. ReV. 1997, 97, 1281-1301.
(5) Smithrud, D. B.; Benkovic, S. J. Curr. Opin. Biotechnol. 1997, 8,
459-466.
(6) Bauer, K.; Garbe, P. Common fragance and flaVor materials,
preparation, properties and uses; VCH: Lausanne, 1985.
(20) Pradelles, P.; Antoine, C.; Maclouf, J. Methods Enzymol. 1990, 187,
24-34.
S0002-7863(97)02481-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/28/1998

DecarboxylatiVe Oxidation of Vanillylmandelic Acid
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3333
Scheme 1. Rhoˆne-Poulenc Process for the Industrial Synthesis of Vanillin
reactions,8,9 for which enzymes are known to be efficient. To
our knowledge, this is not the case for the decarboxylative
oxidation we are considering.
The velocity of this reaction with VMA increases as the pH
rises as a function of the proportion of phenoxide ions, which
are very reactive toward NaIO4 (kuncat ) 0.352 ( 0.004 × 10-5
µM-1 min-1 at pH 7.4; kuncat ) 1.029 ( 0.010 × 10-5 µM-1
min-1 at pH 9.0).
These observations led us to postulate that mechanism A
prevails in this reaction in neutral or alkaline media.
Hapten Design and Synthesis. The key point in the
generation of catalytic antibodies is the design of the hapten
used to elicit antibodies. According to the previously described
mechanisms of the oxidation of VMA by NaIO4, two approaches
are conceptually possible: either trigger mechanism B through
activation of the R-hydroxy acid moiety, with mimicry of the
cyclic intermediate, or enhance the favored mechanism A. To
avoid the oxidation of the byproducts also bearing R-hydroxy
acid, we chose the second strategy.
Such an oxidation can be performed with easily available
periodinate NaIO4. The aqueous oxidative decarboxylation of
VMA leading to vanillin has been widely used10-12 to measure
VMA concentrations in biological media by measuring the UV
absorbance of vanillin (ꢀ348 ) 19 000; pH 9). In addition,
M
NaIO4 does not alter the antibodies’ recognition activity, since
the oxidation of the sugar moiety of monoclonal antibodies is
one of the commonest methods of preparing antibody-enzyme
conjugates.13 Furthermore, a catalytic antibody has already been
described which achieves NaIO4-mediated sulfur oxidation into
sulfoxide.14
Results and Discussion
We thus chose the hapten H3 (see Scheme 2) which presents
the following required characteristics:
(1) H3 keeps the global structural features of VMA, with
the R-hydroxy acid moiety in the para position toward the
phenol, to discriminate between the regioisomers (o-VMA and
di-VMA).
(2) The partially positively charged oxygen atom in the
transition state is mimicked by the positive charge on the
tetrasubstituted nitrogen atom.
(3) In addition, this positively charged nitrogen atom could
induce, in the antibody combining site, a general base that would
assist the deprotonation to generate the more reactive phenate
form of VMA.
Determination of the Reaction Mechanism. Although the
action of NaIO4 on VMA is well-known, its precise mechanism
remains unclear. Study of the uncatalyzed reaction mechanism
was thus necessary before production of catalytic antibodies.
Two mechanisms can be proposed (Scheme 2), based on the
well-known action of NaIO4 on R-hydroxy acid,15-17 glyoxylic
acid,18 and phenols.19
In both cases, the oxidative decarboxylation would proceed
through a concerted mechanism, after the formation of a reaction
intermediate in which the hypervalent iodine atom is covalently
attached to one (IA) or two (IB) oxygen atoms.
Several experimental arguments allowed us to favor mech-
anism A. In a first series of experiments, we studied the kinetics
of the oxidation of different VMA-related R-hydroxy acids by
NaIO4. The phenol function plays a predominant role in this
oxidation, since the rate of the reaction is significantly decreased
in its absence or when it is substituted by a methoxy group
(Table 1).
Hapten H3 was synthesized in eight steps with a 33% overall
yield from commercially available 4-amino-3-methoxybenzoic
acid as depicted in Scheme 3.
The pivotal point in the synthesis of H3 was to discriminate
the acidic functions. Therefore, we decided to perform a one-
step introduction of an already functionalized linker bearing an
N-hydroxysuccinimide-activated acid moiety. Compound 3C
was easily obtained via classical chemical transformations. The
unprecedented one-step coupling of a functionalized primary
alkyl iodide on a polysubstituted aromatic amine was then
efficiently performed using silver trifluoromethanesulfonate,
yielding essentially pure H3 in 65% yield after one single
precipitation in diethyl ether.
(21) Pradelles, P.; Antoine, C.; Lellouche, J. P.; Maclouf, J. Methods
Enzymol. 1990, 187, 82-89.
(22) Koler, G.; Milstein, C. Nature 1975, 256, 495-497.
(23) Grassi, J.; Frobert, Y.; Lamourette, P.; Lagoutte, B. Anal. Biochem.
1988, 138, 436-450.
(24) Frobert, Y.; Grassi, J. Methods Mol. Biol. 1992, 10, 65-78.
(25) Segel, I. H. Enzymes kinetics; Wiley: New York, 1975.
(26) HPLC results were in total agreement with those obtained with direct
absorbance measurements at 348 nm.
(27) Fenniri, H. Chemtech 1996, 15-25 and references therein.

3334 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Taran et al.
Scheme 2. Possible Mechanisms of Oxidation of VMA
Table 1. Initial Velocities (Vi) of the Oxidation of Various
R-Hydroxy Acids (70 µM) with NaIO4 (1220 µM)a
say (see the Experimental Section), by testing series of
compounds structurally related to H3. The synthesis of these
analogues is described in the Experimental Section. Standard
curves were established with each of these compounds for each
monoclonal antibody, and the corresponding 50% B/Bo values
were compared to that of H3, to evaluate the relative affinity
of the analogues (see the Experimental Section). It is worth
noting that this is only a rough estimation of relative affinities,
and that only 10-fold differences can be unambiguously
interpreted. The results of these experiments are given in Table
2.
These results showed that the aliphatic linker was poorly
involved in hapten-antibody binding for most of monoclonal
antibodies excepting monoclonal antibodies H3-12 and H3-
15, since compound I3-1 was recognized with equivalent or
even better affinity than H3. On the other hand, removal of
the methoxy group (compound I3-2) was highly detrimental for
all monoclonal antibodies (excepting monoclonal antibody H3-
6). The presence of the R-hydroxy acid moiety appeared still
more important, since compounds I3-3 and I3-4 were almost
unrecognized by all antibodies. The importance of the am-
monium group was demonstrated by the very low cross-
reactivities observed with 3-methoxymandelic acid (Table 2).
Taken together, these results demonstrated that the binding sites
of most of the antibodies cover the entire surface of the haptenic
molecule, excepting the aliphatic linker. This is not surprising,
taking into account the small size of the molecule. The critical
contribution of the R-hydroxy acid group is a direct consequence
a The reactions were carried out for 15 min at pH 9.0 and 25 °C in
80 mM potassium carbonate, 50 mM potassium phosphate buffer. The
reaction products (aldehydes) were determined by absorbance measure-
ments using a UV-vis spectrophotometer.
Hapten H3 was covalently coupled to KLH (keyhole limpet
hemocyanin, for immunization) or AChE (acetylcholinesterase,
to be used as enzymatic tracer in enzyme immunoassays -EIA-
experiments) as described in the Experimental Section.20,21 Three
Biozzi mice were immunized, and monoclonal antibodies were
produced using conventional procedures.22,23 Hybridoma su-
pernatants were screened by testing their capacity to bind H3-
AChE conjugate as previously described for other haptens.24
Finally, 11 hybridoma cell lines were stabilized and expended
as ascitic fluids.
Monoclonal Antibodies Characterization. To characterize
in detail the corresponding monoclonal antibodies, their binding
specificity was studied using a competitive enzyme immunoas-

DecarboxylatiVe Oxidation of Vanillylmandelic Acid
Scheme 3. Synthesis of Hapten H3
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3335
Table 2. Relative Affinities of the Different Monoclonal Antibodies with Regard to Series of H3 Analogs
of choosing the quaternary ammonium as coupling point for
the preparation of the immunogen.
complex, while vanillin formed should be released thus prevent-
ing any product inhibition.
This, however, does not guarantee the actual catalytic activity
of the antibodies, since the catalytic process is not an obligatory
consequence of the binding of the substrate H3.
After having thoroughly studied the specificity of these
antibodies, we checked their affinities for the product (vanillin)
and the substrates (VMA and NaIO4) of the reaction. Most of
the monoclonal antibodies (excepting antibody H3-16) have an
ideal specificity profile, affinity for hapten > affinity for
substrate > affinity for product, which is compatible with a
catalytic activity and even at high concentrations (10 mM)
NaIO4 was unable to bind any of the monoclonal antibodies.
Indeed, the catalytic act itself (linked to anti-hapten activity)
has to be preceded by the formation of an antibody-substrate
Catalytic Activity. The next step was to check the capacity
of all monoclonal antibodies to catalyze the oxidation of VMA
by NaIO4 at pH 7.4 (phosphate buffer) and pH 9.0 (phosphate/
carbonate buffer). The reaction was monitored by measuring
the formation of vanillin by means of UV absorbance measure-
ments (ꢀ348 ) 17 000; pH 7.4; ꢀ348 ) 19000; pH 9.0). No
M
M
catalytic activity could be detected at pH 7.4 for any of the

3336 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Taran et al.
Figure 1. Lineweaver-Burk plots for vanillin formation: (A) as a function of VMA concentration at various concentrations of NaIO4 (170, 340,
1000, and 1760 mM) and (B) as a function of NaIO4 concentration at various concentrations of VMA (30, 50, 75, 100, and 200 µM). Assays were
conducted in 80 mM potassium carbonate, 50 mM potassium phosphate buffer, pH 9.0 at 25 °C. The antibody concentration was 3 µM. The
reaction was monitored at 348 nm with a UV-vis spectrophotometer.
monoclonal antibodies, but monoclonal antibody H3-12 sig-
nificantly enhanced the rate of formation of vanillin at pH 9.0.
We therefore selected this monoclonal antibody H3-12 for
further studies. Catalysis obeyed Michaelis-Menten kinetics
(Figure 1), with a kcat of 2.70 min-1 and catalytic efficiencies
(kcat/Ka)/kuncat ) 815 and (kcat/Kb)/kuncat ) 100.
Changes in the Lineweaver-Burk plots as a function of VMA
and NaIO4 concentrations corresponded to an ordered substrate
association.25 First, antibody H3-12 binds VMA to form a
binary complex (Ka ) 260 ( 15 µM). It is only from this
complex that an NaIO4 molecule can be recognized as a second
substrate. This step leads to the formation of a ternary complex
(Kb ) 2100 ( 380 µM) which results in the formation of the
products. This catalytic mechanism perfectly corresponds to
the specificity of H3-12, which has a marked affinity for VMA
but none for NaIO4.
The Ka value (260 ( 15 µM) obtained from the Lineweaver-
Burk plots is in close agreement with the dissociation constant
of the antibody-VMA complex as estimated by competitive
EIA. This correlates with the catalytic mechanism mentioned
above, and suggests that the catalytic reactions take place in
the combining site of the antibody.
Further experiments were performed to confirm that the
observed catalytic activity was actually related to the presence
and the binding capacity of the monoclonal antibody H3-12.
(1) Monoclonal antibody H3-12 from various batches of
ascitic fluids was affinity purified one to three times on a protein
A column and successively purified on a protein A column
followed by size exclusion chromatography with ACA 44 gel.
Whatever the preparation, activity was recovered together with
the antibody protein yielding solutions with equivalent specific
activities (results not shown). The purity of these preparations
was attested by SDS-PAGE analysis in reducing conditions
(results not shown).
of I3-1 allowed us to calculate an inhibition constant (Ki ) 10
( 13 µΜ). This inhibition constant is consistent with the
apparent affinity determined in EIA experiments (50% B/Bo
) 4.6 µΜ). Nevertheless, it is worth noting that in these
Lineweaver-Burk experiments the Ki measured is limited by
the amount of antibody used (3 µM here), and thus the apparent
inhibition constant calculated here has to be carefully interpreted,
taking into account the active sites concentration.
(6) Both the catalytic activity and the binding properties of
monoclonal antibody H3-12 were unchanged after prolonged
treatment with excess NaIO4 (results not shown).
Further investigations were performed in order to characterize
the specificity of the monoclonal antibody H3-12-catalyzed
reaction. Oxidation was carried out with various substrates
including o-VMA, di-VMA and mandelic acid derivatives and
was monitored by HPLC detection of the aldehydes formed.
As expected, only oxidation of VMA26 and, to a lesser extent,
of 4-hydroxymandelic acid (both accepted as substrates) was
significantly accelerated by monoclonal antibody H3-12. Be-
cause of the structural differences between di-VMA, o-VMA,
and hapten H3, antibody H3-12 does not bind to these
molecules. As a consequence, these two byproducts of the first
step of vanillin synthesis are not accepted as a substrate by H3-
12 and their oxidation was not catalyzed. On the other hand,
mandelic acid, 3-methoxymandelic acid and 3,4-dimethoxy-
mandelic acid display a marked affinity for H3-12 (even higher
than VMA), but their oxidation is not catalyzed by antibody
H3-12. It is noteworthy that for these compounds, oxidative
decarboxylation can only occur via NaIO4 addition on the
R-hydroxy acid moiety (mechanism B). These data support the
hypothesis that the monoclonal antibody H3-12 catalytic oxida-
tion process occurs through mechanism A. These results, and
the pH-dependence of the catalysis (as far as VMA is concerned,
no catalytic activity was detected at pH 7.4, and a slight, but
noticeable activity was detected at pH 8.0; the optimum activity
occurred between pH 8.7 and 9.0), are consistent with the
presence of an anionic group in the binding site of the antibody
close to the phenol, as expected, due to the presence of the
charged nitrogen atom on hapten H3.
(2) When kinetic measurements were performed without one
of the two substrates, no detectable vanillin was formed.
(3) None of the control reactions performed with other anti-
H3 monoclonal antibodies and with irrelevant monoclonal
antibodies purified under the same conditions as H3-12 dis-
played any catalytic activity.
(4) The initial rate for the catalyzed reaction is proportional
to the antibody concentration (in the 1-10 µM range, results
not shown).
(5) Catalysis was totally inhibited by compound I3-1, which
binds monoclonal antibody H3-12 with micromolar affinity but
is not subject to NaIO4 oxidation in the reaction period used.
Lineweaver-Burk plots recorded with variable concentrations
We took advantage of the selectivity of monoclonal antibody
H3-12-induced catalysis to perform the decarboxylative oxida-
tion of the crude reaction mixtures from industrial batches.
Reaction progress was monitored by HPLC. In every case,
formation of p-vanillin was significantly enhanced compared
to o-vanillin and other byproducts. A typical experiment
performed with Rhoˆne-Poulenc crude condensation mixture is
shown in Table 3. The results showed that the rate of formation

DecarboxylatiVe Oxidation of Vanillylmandelic Acid
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3337
Table 3. Improvement of p-Vanillin Formation in the Presence of
antibody H3-12 also provide information on the uncatalyzed
reaction mechanism. Indeed, the fact that an antibody produced
against hapten H3 can accelerate the oxidative decarboxylation
of VMA, with an abzymatic mechanism that favors mechanism
A constitutes, in our minds, an additional indication in favor of
the predominance of this mechanism A in the uncatalyzed
reaction. It is, to our knowledge, the first time that such a
fundamental issue has been addressed by the use of a catalytic
antibody.
Monoclonal Antibody H3-12a
a The amounts of the various aldehydes formed in the presence (cat)
or absence (ncat) of antibody H3-12 were determined after oxidation
by 1500 µM NaIO4 of a 100 µM crude reaction mixture of the
condensation of glyoxylic acid on guaiacol obtained from Rhoˆne-
Poulenc industrial batches. Assays were conducted in 80 mM potassium
carbonate, 50 mM potassium phosphate buffer, pH 9.0 at 25 °C. The
antibody concentration was 6 µM, the reaction was stopped after 45 s
by addition of a 10-fold excess of ethylene glycol, and the amounts of
aldehydes formed were determined after HPLC separation and UV
absorbance measurements (see the Experimental Section). Results are
expressed as the ratios of the amount formed in the presence or absence
of monoclonal antibody H3-12.
Experimental Section
A. Synthesis. General Remarks. Reagents were from Aldrich.
All chromatography (flash) was performed with Merck Silicagel 60
(0.02-0.04 mm). TLC was performed with fluorescent Merck F254
glass plates. NMR spectra were recorded on a Brucker AC-300 MHz.
Chemical shifts (δ) are given in ppm, and the coupling constant J is
expressed in hertz. IR spectra were recorded in a Perkin-Elmer
spectrophotometer 2000-FT/IR, and MS were obtained with a Finnigan-
Mat 4600 quadrupole system. Elemental analysis was done by the
“Institut des Substances Naturelles” in Gif sur Yvette.
of p-vanillin increased more than 2-fold while the production
of all other aldehydes was significantly reduced (up to 4-fold).
Hapten H3 was prepared from commercial 4-amino-3-methoxyben-
zoic acid in an 8-step sequence as outlined in Scheme 4.
4-(Dimethylamino)-3-methoxyethyl Benzoate (3A). 4-amino-3-
methoxybenzoic acid (4.54 g, 23.28 mmol) was dissolved in 100 mL
of methanol. Concentrated sulfuric acid (11 mL, 232.8 mmol) was
slowly added. The mixture was then refluxed for 6 h, cooled to room
temperature, neutralized with a saturated K2CO3 aqueous solution, and
extracted three times with Et2O. Organic phases were then dried over
Na2SO4, filtered, and concentrated. The resultant orange oil was
chromatographed on silica gel (hexane/AcOEt, 8/2). The white solid
was further purified by three successive crystallizations, yielding 4.18
g (92%) of 4-amino-3-methoxyethyl benzoate (mp 77 °C). 1H NMR
(300 MHz, CDCl3): δ 1.37 (t, J ) 7 Hz, 3H), 3.9 (s, 3H), 4.31 (q J )
7 Hz, 2H), 6.66 (d, J ) 9 Hz, 1H), 7.46 (d, J ) 1.5 Hz, 1H) 7.56 (dd,
J ) 1.5 and 9 Hz, 1H). 13C NMR: δ 15.4, 56.5, 61.4, 112.0, 114.0,
120.7, 124.9, 142.0, 147.0. IR (neat): 3366, 3010, 1685, 1522, 1295.
MS (CI, NH3): 196 ([M + H]+), 213 ([M + NH4]+). Anal. Calcd
for C10H13NO3: C, 61.53; H, 6.71; N, 7.17. Found: C, 61.57; H, 6.71;
N,7.17.
A mixture of 4-amino-3-methoxyethyl benzoate (1.8 g, 9.23 mmol)
and NaBH4 (2 g, 56.5 mmol), suspended in 30 mL of THF, was added
dropwise to a 37% aqueous formaldehyde solution (4.5 mL, 48.4 mmol)
and 3 M sulfuric acid (3.3 mL, 8.07 mmol) in 50 mL of THF. When
half of the suspension had been added, an extra 3.3 mL of 3 M sulfuric
acid was added, and the mixture was vigorously stirred for 1 h,
neutralized with a saturated aqueous solution of K2CO3, and extracted
three times with Et2O. The organic phases were dried over Na2SO4,
filtered, and concentrated. Chromatography on silica gel (hexane/
AcOEt, 9/1) gave 1.82 g (88%) of dimethylamine 3A as a yellow oil.
1H NMR (CDCl3): δ 1.24 (t, J ) 8 Hz, 3H), 2.73 (s, 6H), 3.78 (s,
3H), 4.21 (q, J ) 8 Hz, 2H), 6.7 (d, J ) 8 Hz, 1H), 7.51 (d, J ) 2 Hz,
1H), 7.56 (dd, J ) 2 and 8 Hz, 1H). 13C NMR: δ 15.3, 43.6 (2C),
56.4, 61.5, 112.85, 117.6, 124.0, 124.2, 147.5, 152.0. IR (neat): 2950,
1710, 1514, 1293, 1124. MS (CI, NH3): 224 ([M + H]+). Anal. Calcd
for C12H17NO3: C, 64.55; H, 7.67; N, 6.27. Found: C, 64.58; H, 7.57;
N, 6.41.
Conclusion
We have shown that the monoclonal antibody H3-12 elicited
against hapten H3 significantly increases the rate of oxidative
decarboxylation of VMA in the presence of NaIO4. This
reaction was chosen as a typical example of an industrial process
which could be improved by the use of catalytic antibodies.
The catalytic activity of this antibody remains relatively low
(2.70 min-1), so that it cannot be seriously envisaged for use
in an actual production process. However, it displayed several
particularly interesting and original features, as a remarkable
substrate specificity which enabled the antibody to favor VMA
oxidation with regard to the other regioisomers o-VMA and
di-VMA formed during the first step of the industrial synthesis
of vanillin.
Moreover, we have shown that simple immunological meth-
ods (here a competitive enzyme immunoassay) are well suited
to precise determination of the binding properties of the different
monoclonal antibodies, using series of relevant H3 analogues.
Even if these properties are not directly related to the catalytic
potency of the monoclonal antibodies, a good knowledge of
their fine specificity may help in understanding the catalytic
process and/or the reaction mechanism.
In addition, the fact that H3-12 is the only monoclonal
antibody (with H3-15) exhibiting significant recognition of the
linker (Table 2) may be related to the catalytic capacity,
assuming that the part of the antibody binding site in contact
with the linker could be involved in NaIO4 binding to form the
tertiary complex, a step required for catalytic activity.
It is also noteworthy that monoclonal antibody H3-12, the
only anti-H3 antibody with catalytic activity, has the lowest
affinity for the hapten (0.5 µM) of all 11 monoclonals tested.
Clearly, the conventional procedure, which consists of selecting
antibodies according to their binding affinity for the hapten,
would not have proven felicitous in this case. This illustrates
the importance of an early detection of the catalytic activity of
an antibody, a question which has been widely discussed in
previous papers. Finally, the results observed with monoclonal
4-(Dimethylamino)-3-methoxybenzaldehyde (3B). Dimethylamine
3A (1.7 g, 7.62 mmol) was dissolved in 100 mL of hexane. The
solution is cooled to -78 °C, and 1 M DiBAl-H (17 mL, 16.75 mmol)
in toluene was added dropwise. The solution was then stirred at -78
°C for 30 min, and excess hydride was destroyed with successive
additions of methanol and water. The resultant suspension was filtered
on Celite and extracted four times with AcOEt, dried, and concentrated.
The viscous oil is chromatographed on silica gel (hexane/AcOEt, 2/8)
(28) Wagner, J.; Lerner, R. A.; Barbas, C. F., III Science 1995, 270,
1797-1800.
(29) Tawfik, D. S.; Lindner, A. B.; Chap, R.; Kim, S. H.; Green, B. S.;
Eshhar, Z. In Immunology Methods Manual- the comprehensiVe Sourcebook
of Techniques; Lefkovits, I., Ed.; Academic Press: San Diego, CA, 1997;
Vol. 1, pp 551-560.
(30) Hua, T. D.; Rolland-Fulcrand, V.; Lazaro, R.; Viallefont, P.; Lefranc,
M. P.; Weill, R. Tetrahedron Lett. 1996, 37, 175-178.
(31) Lewis, C.; Kra¨mer, T.; Robinson, S.; Hilvert, D. Science 1991, 253,
1019.
(32) Smiley, J. A.; Benkovic, S. J. Proc. Natl. Acad. Sci. U.S.A. 1994,
91, 8319-8323.

3338 J. Am. Chem. Soc., Vol. 120, No. 14, 1998
Taran et al.
yielding 1.24 g (90%) of (4-(dimethylamino)-3-methoxyphenyl)-
methanol as a pale yellow oil. 1H NMR (CDCl3): δ 2.72 (s, 6H),
3.78 (s, 3H), 3.82 (s, 3H), 4.54 (s, 2H), 6.84 (s, 3H). 13C NMR: δ
44.3 (2C), 56.2, 65.95, 111.1, 118.8, 120.25, 136.5, 142.7, 153.4. IR
(neat): 3386, 2950, 1513, 1258, 1121. MS (CI, NH3): 182 ([M +
H]+).
4-Iodobutyric acid (1 g, 4.67 mmol), N-hydroxysuccinimide (0.54
mg, 4.69 mmol), and dicyclohexylcarbodiimide (0.96 g, 4.66 mmol)
were dissolved in 10 mL of AcOEt. The reaction mixture was stirred
at room temperature for 16 h. The dicyclohexyl urea formed was
discarded by filtration over Celite, and the crude material was
concentrated and chromatographed on silica gel (petroleum ether/
AcOEt, 1/1). The resultant yellow powder was recrystallized from
petroleum ether/EtOH (9/1) yielding activated ester 3D (1.38 g, 95%)
as white crystals (mp ) 86 °C). 1H NMR (CDCl3): δ 2.24 (quint, J
) 6.5 Hz, 2H), 2.76 (t, J ) 6.5 Hz, 2H), 2.83 (s, 4H), 3.28 (t, J ) 6.5
Hz, 2H). 13C NMR: δ 3.5, 25.3 (2C), 28.0, 31.6, 168.7 (3C). IR (neat)
1803, 1734, 1201. MS (CI, NH3): 329 ([M + NH4]+). Anal. Calcd
for C8H10INO4: C, 30.89; H, 3.24; N, 4.5. Found: C, 30.66; H, 3.17;
N, 4.36.
[4-(1-Carboxy-1-hydroxymethyl)-2-methoxyphenyl]-[3-((2,5-di-
oxopyrrolidin-1-yloxy)carbonyl)propyl]dimethylammonium Tri-
fluoromethanesulfonate (H3). R-Hydroxy acid 3C (0.040 g, 0.18
mmol), activated ester 3D (0.055 g, 0.18 mmol), and silver trifluo-
romethanesulfonate (0.046 g, 0.18 mmol) were dissolved in 2 mL of
anhydrous acetone. The solution was stirred for 16 h at room
temperature. The silver iodide formed was discarded by filtration over
Celite, and the solvent was evaporated. The resultant orange oil was
dissolved in 1 mL of MeOH, and hapten H3 was precipitated by
addition of 10 mL of Et2O. The white powder (53 mg, 54% yield)
was collected and used without further purification. 1H NMR (DMSO-
d6): δ 2.58 (quint, J ) 9 Hz, 2H), 2.74 (t, J ) 9 Hz, 2H), 2.87 (s, 4H),
3.89 (s, 6H), 4.09 (s, 3H), 4.21 (t, J ) 9 Hz, 2H), 5.44 (s, 1H), 7.33
(dd, J ) 2 and 10.5 Hz, 1H), 7.56 (d, J ) 2 Hz, 1H), 7.86 (d, J ) 10.5
Hz, 1H). 13C NMR: δ 24.35, 25.5 (2C), 27.7, 55, 57.2 (2C), 64.4,
72.8, 113.5, 118.5, 120.2, 123.8, 124.9, 130.9, 168.8; 170.5 (2C), 172.6.
IR (neat): 3448, 2955, 1815, 1783, 1731, 1418, 1267, 1166. MS
(EI): 408 (m/z).
(4-(Dimethylamino)-3-methoxyphenyl)methanol (1.06 g, 5.85 mmol)
was dissolved in 100 mL of freshly distilled CH2Cl2. MnO2 (10 g,
115 mmol) was added, and the black suspension was stirred at room
temperature for 16 h. The suspension was then filtered over Celite,
concentrated, and chromatographed on silica gel (hexane/AcOEt, 8/2),
yielding 0.9 g (86%) of aldehyde 3B as a pale yellow oil. 1H NMR
(CDCl3): δ 2.9 (s, 6H), 3.93 (s, 3H), 6.9 (d, J ) 8 Hz, 1H), 7.30 (dd,
J ) 2 and 8 Hz, 1H), 7.36 (d, J ) 2 Hz, 1H), 9.8 (s, 1H). 13C NMR:
δ 43.5 (2C), 56.5, 110.1, 117.3, 127.7, 130.7, 149.2, 152.5, 191. IR
(neat): 2950, 1678, 1515, 1255, 1124. MS (CI, NH3): 180 ([M +
H]+). Anal. Calcd for C12H17NO3: C, 67.02; H, 7.31; N, 7.81.
Found: C, 66.89; H, 7.28; N, 7.77.
1-(4-(Dimethylamino)-3-methoxyphenyl)hydroxyacetic Acid (3C).
Aldehyde 3B (850 mg, 4.75 mmol) was dissolved in 150 mL of
anhydrous CH2Cl2. The solution was cooled to 0 °C, and TMSCN
(1.3 mL, 9.5 mmol) and Et3N (0.1 mL, 0.95 mmol) were added
dropwise. The solution was stirred for 8 h at 0 °C and then brought to
room temperature overnight. Evaporation of the solvent yielded 1.32
g (99%) of (4-(dimethylamino)-3-methoxyphenyl)trimethylsilanoxy)-
acetonitrile as a colorless oil, which could be used without further
purification. 1H NMR (CDCl3): δ 0.23 (s, 9H), 2.80 (s, 6H), 3.93 (s,
3H), 5.44 (s, 1H), 6.92 (d, J ) 8.5 Hz, 1H), 6.96 (d, J ) 3 Hz, 1H),
6.97 (dd, J ) 3 and 8.5 Hz, 1H). 13C NMR: δ 0.8 (3C), 44.1 (2C),
56.4, 64.6, 110.05, 118.9, 120.1, 120.3, 130.8, 144.5, 153.5. IR
(neat): 2956, 1515, 1458, 1253, 1126. MS (CI, NH3): 279 ([M +
H]+).
[4-(1-carboxy-1-hydroxymethyl)-2-methoxyphenyl]trimethylam-
monium Iodide (I3-1). (4-(Dimethylamino)-3-methoxyphenyl)hy-
droxymethyl acetate (0.254 g, 1.06 mmol) and methyl iodide (0.2 mL,
3.2 mmol) are dissolved in 10 mL of acetone. The solution was
refluxed for 16 h. After cooling, the ammonium salt was precipitated
by addition of Et2O. The white solid was washed with Et2O, yielding
0.36 g (89%) of [4-(1-carbxy-1-hydroxymethyl)-2-methoxyphenyl]-
trimethylammonium iodide as a white powder (mp ) 142 °C). 1H
NMR (CD3OD): δ 3.71 (s, 3H), 3.76 (s, 9H), 4.11 (s, 3H), 5.35 (s,
1H), 7.25 (dd, J ) 2 and 12 Hz, 1H), 7.5 (d, J ) 2 Hz, 1H), 7.81 (d,
J ) 12 Hz, 1H). 13C NMR: δ 53, 56.65 (3C), 57.2, 73.1, 114, 120.7,
122.55, 134.55, 145.2, 153.2, 173. IR (neat): 3310, 3028, 1748, 1419,
1084. Anal. Calcd for C13H20INO4: C, 40.96; H, 5.29; N, 3.67.
Found: C, 40.79; H, 5.14; N, 3.63.
This ammonium salt (0.17 g, 0.45 mmol), K2CO3 (0.06 g, 0.44
mmol), and water (0.04 mL) were dissolved in 3 mL of THF. The
solution was stirred for 18 h at room temperature and concentrated.
The crude material was dissolved in 5 mL of methanol, and the
ammonium salt was precipitated with Et2O; 165 mg (99%) I3-1 was
then collected as a white powder (mp ) 155 °C). 1H NMR (D2O): δ
3.73 (s, 9H), 4.15 (s, 3H), 4.81 (s, 1H), 6.84 (d, J ) 12 Hz, 1H), 7.14
(s, 1H), 7.48 (d, J ) 12 Hz, 1H). 13C NMR: δ 56.45 (3C), 56.9, 75.2,
113.9, 120.5, 121.6, 133.6, 148.8, 152.8, 177.9. IR (neat): 3421, 1608,
1384, 1088.
(2-Methoxyphenyl)trimethylammonium iodide (I3-3). 2-Ami-
nophenol (2 g, 18.3 mmol) was dissolved in 50 mL of anhydrous THF
under nitrogen. Sodium hydride (1.65 g, 80% dispersion in mineral
oil, 55 mmol) was added in one portion. The solution was stirred for
15 min at room temperature, and methyl iodide (5.7 mL, 91.5 mmol)
was added. The solution was refluxed for 16 h, cooled, and filtered.
The yellow powder was then washed with Et2O, yielding 3.55 g (70%)
of ammonium salt I3-3. 1H NMR (D2O): δ 3.57 (s, 9H), 3.91 (s, 3H),
7.02 (t, J ) 8.5 Hz, 1H), 7.21 (d, J ) 8.5 Hz, 1H), 7.46 (t, J ) 8.5 Hz,
1H) 7.49 (d, J ) 8.5 Hz, 1H).
(4-(Hydroxymethyl)-2-methoxyphenyl)trimethylammonium Io-
dide (I3-4). (4-(Dimethylamino)-3-methoxyphenyl)methanol (0.17 mg,
0.94 mmol) and methyl iodide (0.3 mL, 4.82 mmol) were dissolved in
10 mL of anhydrous acetone. The solution was refluxed for 16 h and
cooled, and the quaternarized amine was precipitated with 100 mL of
Cyanhydrine (1.32 g, 4.75 mmol) was placed in 120 mL of 1 N
HCl solution. The solution was refluxed for 7 h, cooled, and then
lyophilized. The resultant yellow paste was dissolved in 100 mL of
methanol, 0.05 mL of concentrated sulfuric acid was added, and the
reaction mixture was stirred at room temperature for 14 h. The solution
was then neutralized with saturated aqueous NaHCO3 solution and
extracted three times with AcOEt. The organic phase was dried over
Na2SO4 and concentrated. The resultant yellow oil is chromatographed
on silica gel (hexane/AcOEt, 1/1) to yield (4-(dimethylamino)-3-
methoxyphenyl)hydroxymethyl acetate (1.01 g, 89%) as a white powder
(mp: 82 °C). 1H NMR (CDCl3): δ 2.73 (s, 6H), 3.71 (s, 3H), 3.84 (s,
3H), 5.08 (s, 1H) 6.87 (s, 3H). 13C NMR: δ 44.0 (2C), 53.8, 56.3,
73.8, 110.1, 118.9, 120.1, 133.3, 143.5, 153.3, 175.1. IR (neat): 3431,
2951, 1742, 1512, 1460, 1252, 1123. MS (CI, NH3): 240 ([M + H]+).
Anal. Calcd for C12H17NO4: C, 60.23; H, 7.16; N, 5.85. Found: C,
59.99; H, 6.94; N, 5.81.
(4-(Dimethylamino)-3-methoxyphenyl)hydroxymethyl acetate (335
mg, 1.4 mmol) was dissolved in 5 mL of methanol, and 3 M aqueous
KOH solution (0.5 mL, 1.4 mmol) was added. The solution was stirred
for 1 h at room temperature and then neutralized with dilute HCl. Water
was evaporated, and the crude material was chromatographed on silica
gel (CH2Cl2/MeOH, 1/1) to yield R-hydroxy acid 3C (0.29 g, 92%) as
a white powder (mp ) 183 °C). 1H NMR (CD3OD): δ 2.71 (s, 6H),
3.88 (s, 3H), 4.82 (s, 1H), 6.92 (d, J ) 12 Hz, 1H), 6.96 (d, J ) 3 Hz,
1H), 7.02 (dd, J ) 3 and 12 Hz, 1H). 13C NMR: δ 44.5 (2C), 55.8,
76.7, 111.7, 118.9, 120.9, 138.35, 142.3, 153.5, 180.4.
4-Iodo-[2,5-dioxopyrrolidin-1-yl] Butyrate (3D). 4-Bromobutyric
acid (5 g, 29.94 mmol) and sodium iodide (13.35 g, 86.72 mmol) were
dissolved in 100 mL of anhydrous acetone. The solution was stirred
for 16 h under argon, and 400 mL of water were added. The aqueous
phase was then extracted three times with 200 mL of Et2O, and the
organic phase was washed with 400 mL of 10% aqueous Na2S2O3
solution, dried over MgSO4, and concentrated. The crude material was
then chromatographed on silica gel (CH2Cl2/EtOH, 98/2) yielding 5.45
g (85%) of 4-iodobutyric acid as a colorless oil. 1H NMR (CDCl3):
δ 2.12 (quint, J ) 6.5 Hz, 2H), 2.51 (t, J ) 6.5 Hz, 2H), 3.84 (t, J )
6.5 Hz, 2H). 13C NMR: δ 5.3, 27.9, 34.3, 178.4. MS (CI, NH3): 232
([M + NH4]+).

DecarboxylatiVe Oxidation of Vanillylmandelic Acid
J. Am. Chem. Soc., Vol. 120, No. 14, 1998 3339
Et2O. The white solid was washed with Et2O, yielding 243 mg (80%)
of I3-4 (mp ) 145 °C). 1H NMR (D2O): δ 3.52 (s, 9H), 3.87 (s, 3H),
4.52 (s, 2H), 6.94 (d, J ) 8.5 Hz, 1H), 7.14 (s, 1H), 7.51 (d, J ) 8.5
Hz, 1H). 13C NMR: δ 59.1 (3C), 59.6, 66.15, 116.6, 122.6, 124.4,
136, 148.4, 155.4, 173.. IR (neat): 3341, 3041, 1426, 1086. Anal.
Calcd for C11H18INO2: C, 40.88; H, 5.61; N, 4.33. Found: C, 40.44;
H, 5.39; N, 4.53.
B. Antibody Assays. General Remarks. Immunogen Prepara-
tion. The conjugate H3-KLH was prepared by adding 30 µmol of
hapten H3 in 100 µL of dimethylformamide (DMF) to 5 mg of KLH
in 10 mL of 0.1 M borate buffer pH 9.0. After 1 h of stirring at room
temperature, the mixture was dialyzed against 0.1 M phosphate buffer
pH 7.4 at 4 °C. Immunogen H3-KLH was stored at -20 °C until
use.
Enzymatic Tracer Preparation. Enzymatic tracer was prepared
by covalent linkage of hapten H3 to the G4 form of AChE as follows:
30 µL (10 nmol) of hapten H3 in DMF were added to 500 µL of 0.1
M pH 9.0 borate buffer containing 100 µg (0.32 nmol) of AChE. The
reaction was allowed to proceed for 30 min at room temperature and
was then stopped by addition of 500 µL of EIA buffer (0.1 M phosphate
buffer (pH 7.4) containing 0.15 M NaCl, 10-3 M EDTA, 0.1% bovine
serum albumin, and 0.01% sodium azide). Enzymatic tracer was
purified and stored as described elsewhere.33
Competitive EIA Procedure. Antibody Affinity Measurement.
Competitive EIA was performed as previously described34 using 0.1
M phosphate buffer (pH 7.4) containing 0.15 M NaCl, 10-3 M EDTA,
0.1% bovine serum albumin, and 0.01% sodium azide (EIA buffer).
Briefly, 96-well microtiter plates coated with affinity-purified goat
polyclonal anti-mouse IgGs were used to ensure separation between
bound and free moieties of the tracer. The reagents were dispensed in
the following order: 50 µL of standard or buffer, 50 µL of enzymatic
tracer, and 50 µL of diluted selected antibody. The plates were reacted
for 18 h at 4 °C and washed before 200 µL of Ellman’s reagent were
dispensed into each well. After 2 h of enzymatic reaction, the
absorbance at 414 nm of each well was measured. Results were
expressed in terms of B/Bo (%), where B and Bo represents the amount
of solid phase-bound tracer in the presence or absence of competitor,
respectively. 50% B/Bo values correspond to the concentration of
competitor for which the solid phase-bound tracer activity is half of
Bo. These values can be considered as a good approximation to the
K
D value of the desired product toward the antibody when KD > [Ab]
> [tracer].
UV-Monitored Kinetic Assays. The rates were determined by
measuring the initial change in absorbance (<5% substrate depletion)
at 348 nm reflecting vanillin release. The background rate of oxidation
of VMA was also measured each time and subtracted from the velocities
noted in the presence of antibody H3-12.
Monoclonal Antibodies. Generation and Purification. Three
mice each received a subcutaneous injection of 100 µL of 0.5 mg/mL
H3-KLH conjugate emulsified in complete Freund’s adjuvant. Boost-
ers were given at two-week intervals. After receiving a final
intravenous injection of 100 µg of H3-KLH, the spleen cells of the
mouse presenting the highest antiserum titer for H3-AChE conjugate
were fused with NS1 myeloma cells as described previously.20
Hybridoma supernatants were screened for their capacity to bind H3-
AChE conjugate using EIA as described previously.21 After selected
hybridoma cells were cloned, monoclonal antibodies were expanded
as ascitic fluids in mice. Monoclonal antibodies were purified with
standard protein A chromatography: to a 5 mL sample of ascitic fluid
was added 20 mL of binding buffer (aqueous 0.1 M borate, 0.1 M
NaCl, pH 9.0) and 1.0 g of silica gel immobilized protein A. This
solution was applied to a column and washed with binding buffer until
the absorbance of the eluant was <0.05 absorbance units (at 280 nm).
Elution buffer (0.1 M citrate buffer, pH 3.5) was applied to the column,
and 2 mL fractions were collected. IgG-containing fractions were
pooled and dialyzed against 0.1 M phosphate buffer. Purified mono-
clonal antibody solutions were stored at -20 °C until use.
HPLC-Monitored Kinetics Assays. HPLC separation is achieved
on a C18 reverse-phase nucleosil 5 µm column. Flow rate: 1 mL/min
Column temperature: 40 °C. Elution conditions: 0-10 min, water/
AcOH 98/2 (solvent A); 10-30 min, gradient up to 100% solvent B
(water/MeOH/AcOH 69/29/2); 30-50 min 100%, solvent B. Alde-
hydes were quantified by UV absorbance measurement at 310 and 255
nm.
Acknowledgment. This study was conducted as part of the
BIOAVENIR program financed by Rhone-Poulenc and the CEA
with a contribution from the “Ministe`re de l’Education Natio-
nale, de l’Enseignement Supe´rieur et de la Recherche”. The
authors are indebted to P. Lamourette and M. Plaisance for
expert technical assistance in monoclonal antibody preparation
and to A. Valleix for HPLC advice, and also thank Dr. D. Pe´tre´
(Rhoˆne-Poulenc) for early discussions of this work.
JA9724811
(33) Pradelles, P.; Grassi, J.; Maclouf, J. Anal. Chem. 1985, 57, 1170-
1173.
(34) Pradelles, P.; Grassi, J.; Charbardes, D.; Guiso, N. Anal. Chem. 1989,
61, 447-453.