Original preparation of conjugates for antibody production against Amicoumacin-related anti-microbial agents

Article abstract of DOI:10.1016/j.bmc.2008.08.017

Amicoumacins are natural products with potent anti-ulcerogenic and anti-bacterial activities, and have been isolated from different Bacillus genera. They belong to a family of 3,4-dihydroisocoumarin derivatives bearing hydroxylated amino acid side chains.

Full text of DOI:10.1016/j.bmc.2008.08.017

Bioorganic & Medicinal Chemistry 16 (2008) 9383–9391
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Original preparation of conjugates for antibody production against
Amicoumacin-related anti-microbial agents
Svitlana Shinkaruk a,b, Bernard Bennetau c,d, , Pierre Babin c,d, Jean-Marie Schmitter e,
*
Valerie Lamothe a,b, Catherine Bennetau-Pelissero a,b, Maria C. Urdaci b,e,
*
a Université Bordeaux 1, EA 2975 UBX1-UBX2-ENITAB, F-33405 Talence Cedex, France
b ENITA de Bordeaux, UMRS, 1 Cours du Général de Gaulle, F-33175 Gradignan, France
c Université Bordeaux 1, ISM UMR 5255, 351 Cours de la Libération, F-33405 Talence Cedex, France
d CNRS, ISM UMR 5255, 351 Cours de la Libération, F-33405 Talence Cedex, France
e Institut Européen de Chimie et Biologie UMR 5248 CNRS, Université Bordeaux 1- ENITAB 2, Rue Robert Escarpit, F-33607 Pessac, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Amicoumacins are natural products with potent anti-ulcerogenic and anti-bacterial activities, and have
been isolated from different Bacillus genera. They belong to a family of 3,4-dihydroisocoumarin deriva-
tives bearing hydroxylated amino acid side chains. The 3,4-dihydroisocoumarin moiety of Amicoumacins
has been obtained in two steps from 2-methoxybenzoic acid by combining directed and benzylic meta-
lation strategies. The use of s-BuLi in both steps gave satisfactory and reproducible yields. For the devel-
opment of an immunoassay (ELISA) of Amicoumacin-related compounds in biological media, the
deprotected 3,4-dihydroisocoumarin moiety has been coupled to the BSA carrier protein via a homobi-
Received 14 March 2008
Revised 31 July 2008
Accepted 4 August 2008
Available online 7 August 2008
Keywords:
Immunoassay
Amicoumacin
functional linker deriving from D-tartaric acid. This approach enabled to introduce the hydroxylated por-
tion of Amicoumacin directly during the preparation of hapten–protein conjugates. The coupling ratio
was evaluated by mass spectrometry. The hapten–protein conjugate showing the best coupling ratio
was used to generate polyclonal immunosera in rabbits. After immunoserum titration, ELISA conditions
were set up and specificity tests were performed on solutions of pure parent compounds, semi-purified
Amicoumacin B as well as on culture supernatants of strains known for their Amicoumacin production.
This immunoassay is suitable for a rapid and simple screening test for the production of Amicoumacins
and its related compounds by bacterial strains.
3,4-Dihydroisocoumarin
Homobifunctional coupling agent
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
B (or AI-77 A and B) are the major members of the family. It
remains unclear whether the minor structural analogues are
Amicoumacin family members represent an unique class of
drugs because of their potent anti-ulcerogenic action without any
anti-cholinergic and anti-histaminergic effects.1 Moreover, these
compounds possess anti-bacterial, anti-inflammatory, and anti-tu-
moral activities.2,3 Amicoumacins belong to a family of 3,4-dihy-
droisocoumarin derivatives (West block) bearing hydroxylated
amino acid side chains (East block) (Fig. 1). Recently, we have dem-
onstrated that the probiotic properties of a Bacillus subtilis strain
are due to the secretion of Amicoumacin B.4 The production of
Amicoumacin antibiotics by B. subtilis is a common characteristic
of individual strains that present genetic and physiological homo-
geneity.5 Compounds having the same backbone as well as another
minor structural analogues have been isolated from other bacterial
sources and called AI-77 A, B, C, D, E, and F.6–9 Amicoumacin A and
natural bacterial metabolites or products of chemical degradation
occurring during the isolation of the major compounds. However,
it is now well established that Amicoumacin A and B exhibit the
highest biological activity.
Amicoumacin B was the principal target of different synthetic
approaches and several alternative total syntheses of this com-
pound have been proposed.10–18 In most of the approaches, the
construction of the 3,4-dihydroisocoumarin skeleton was achieved
by lithiation at the benzylic position of different o-toluic acid
derivatives, such as esters10–13 or oxazolines,14,15 followed by
quenching with N-protected leucinal. In an alternative synthesis,
West and East blocks were both assembled from D-ribose as the
common chiral source.16 A convergent and highly stereocontrolled
total synthesis of Amicoumacin A, reported by Ghosh, included the
preparation of the 3,4-dihydroisocoumarin fragment by a regio-
specific Diels–Alder reaction as well as the synthesis of East block
by an ester-derived titanium-enolate-mediated syn-aldol reaction,
* Corresponding authors.
E-mail addresses: b.bennetau@ism.u-bordeaux1.fr (B. Bennetau), m-urdaci@
enitab.fr (M.C. Urdaci).
a
Curtius rearrangement, and the application of Dondoni’s
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.08.017

9384
S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
media of different B. subtilis strains that are known to produce
Amicoumacin Family
OH
these compounds.5 This rapid and low cost assay could serve as
an indicator of bacterial strains producing Amicoumacins as well
as for in vivo studies.
O
O
H
N
H
R
2. Results and discussion
O
2.1. Immunogen design and synthesis
West block East block
We were interested in the simultaneous immunodetection of all
Amicoumacin analogues in microbiological media. Amicoumacins
are low weight molecules, and thus, cannot act as immunogens.
The design and synthesis of the corresponding haptens and the
hapten–protein conjugates is a crucial step for successful antibody
generation. The main structural differences between all Amicou-
macins are found on the tail end of the hydroxylated b-amino acid
side chain that constitutes the East block. The West block formed
by the 3,4-dihydro-8-hydroxyisocoumarin moiety with a leucine
analogue at position 3 linked via the amide bond to the part of East
Block including the first hydroxy group are common for all major
and minor Amicoumacins (Fig. 1).
Amicoumacin (AI-77)
B
A
C
OH NH2
COOH
OH
NH2
O
OH NH2
CONH2
OH
O
OH
AI-77
D
C'
F
In order to address the recognition versus the common frag-
ment-derived moiety and to improve the cross-reactivity pattern
of the antibodies for the whole Amicoumacin family we designed
the immunizing hapten–protein conjugates.
OH
OH
OH
NHCOCH3
NHCOC2H5
O
O
O
2.1.1. Hapten synthesis
O
O
O
Most of the synthetic approaches toward the West block of Ami-
coumacins involve the addition of o-toluic anions of esters10–13 or
oxazolines14,15 to N-protected leucinal. These syntheses are time
consuming (up to 10 steps including the synthesis of the 2-meth-
oxy-6-methyl-benzoic acid28) with moderate yields. Our first at-
tempts concerning the West block synthesis, similar to the
original total synthesis reported by Hamada11 or Ward13, gave
low yields, were often difficult to reproduce, and led to the self-con-
densation product of benzoic ester, as also mentioned by Kessar.29
As underlined by Schlosser and Geneste, ‘The best protective
group is the one which can be omitted’.30 Directed ortho-metalation
of unprotected benzoic acids has become an important and useful
method for the preparation of different polyfunctional aromatic
systems.31–35 On the other hand, reactions of metalated toluic acids
with different electrophiles have been less explored.36–39
We decided to explore the combined directed and benzylic
metalation strategy on unprotected benzoic acids for the synthesis
of the West block of Amicoumacins. The 3,4-dihydroisocoumarin
moiety was obtained in only two steps from commercially avail-
able 2-methoxybenzoic acid 1 by a regioselective ortho-lithiation
reaction,31,32 followed by the generation of the benzylic carbanion
of 2, and then by the addition to N-Boc-protected leucinal 3 (pre-
pared as described by Dondoni et al.40) with a spontaneous lactone
formation during a mild acid hydrolysis (Fig. 2).
Figure 1. Amicoumacin family compounds.
aldehyde homologation.17–19 In addition, Superchi et al. described
an alternative asymmetric route to the West block based on the
regioselective anionic allylation of N,N0-diethylbenzamides and
on the Sharpless asymmetric dihydroxylation.19,20 Finally, several
methodologies have been reported to obtain the East block (three
asymmetric carbons).21–24 However, the total synthesis of Amicou-
macin B turned out to be very difficult, expensive, and time con-
suming, and thus, limited the development of its therapeutic
applications. The utilization of bacterial strains producing Amicou-
macins may be a good alternative that requires large-scale studies
in order to characterize the Amicoumacin production by different
bacteria using a simple detection system. However, a reliable and
simple system for the quantitative detection of this product is
not currently available.
Our work was focused on the development of an easy and rapid
detection method for these compounds in biological media. Immu-
nodetection based methods are successfully used for the routine
detection of small molecules.25 ELISA based immunoassays have
been previously developed in our Laboratories for the detection
of phyto-estrogens in biological media.26,27
In the benzylic metalation step, the lithio species was generated
by treatment of 2 with different lithium bases (Table 1 in
Supplementary data). The use of s-BuLi gave satisfactory and
reproducible yields (39%, 4 + 5) as well as reasonable diastereose-
lectivity comparable with previous syntheses. The use of 3 equiv
of s-BuLi seems to be crucial for the good deprotonation of 2. The
yield of the reaction was not improved by addition of TMEDA
and the utilization of sparteine41 did not increase the diastereose-
lectivity. It is worthy to note that the use of o-toluic acid instead of
2 under the same conditions as entry 8 (Table 1 in Supplementary
data) led to dihydroisocoumarin formation only in a 24% yield. This
observation is in good agreement with the study of possible com-
petition between ortho- and lateral-lithiation of o-toluic acid
derivatives.42
In this paper, we describe an original preparation of hapten–
protein conjugates for the production of specific antibodies against
the Amicoumacin family compounds. A short and efficient synthe-
sis of the West block is reported. The deprotected West block was
coupled to bovine serum albumin (BSA) via a homobifunctional
linker deriving from D-tartaric acid. This approach enabled to intro-
duce the common hydroxylated portion of the East block directly
during the preparation of the hapten–protein conjugates. The cou-
pling efficiency was examined by various methods, including
UV-spectroscopy, SDS–PAGE and MALDI mass spectrometry. The
antibodies raised against the hapten–protein conjugate with the
best coupling ratio were evaluated and competitive ELISA in a
semi-quantitative format was validated by the detection of semi-
purified Amicoumacin B and detection of Amicoumacins in culture

S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
9385
O
O
O
O
1) 2eq. s-BuLi,
O
O
O
O
TMEDA,
1) RLi (Table 1),
THF, -90 ºC
THF, -90 ºC
O
O
OH
OH
+
H
H
2) 4eq. MeI
2)
tBuMgCl
NHBoc
NHBoc
O
NHBoc
1
2
4
5
3
ref 15
Ref 13
Figure 2. Synthesis of the protected West block 4 of Amicoumacins.
The preliminary treatment of 3 with t-BuMgCl was crucial to
obtain a reproducibility of the benzylic anion addition. The N-
deprotonation of 3 is presumed to stabilize the chelation model
according to Cram’s rule and explains the reasonable diastereose-
lectivity. The conversion of 3 to the N-lithio derivative prior to
the addition of the benzylic anion, as described for N-benzylsulfo-
nyl-leucinal,11 did not lead to the expected product.
able to mimic the common portion of all compounds of the
Amicoumacin family and to increase the sensitivity of the immu-
noresponse with no discrimination of particular members of the
family.
Compound 8 was freshly prepared from D-tartaric acid with a
good yield (80%) using the typical procedure for N-succinimidyl es-
ter synthesis described by Anderson et al.43 The synthesis of the
West block-BSA conjugates was performed at different molar ratios
of 6/BSA 50:1 (9a) and 100:1 (9b) to determine the optimal ratio
for further routine use in immunogen and coating antigen prepara-
tion. The conjugation reaction is crucial for the success of an
immunization, and the use of large excess of hapten ensures the
high substitution of the lysine side chains of the carrier protein.
On the other hand, the hapten overload is not reasonable (unre-
acted West block is lost during the conjugate purification) because
of the complexity of its synthesis that requires the utilization of
organolithium reagents.
The stereochemistry of the major product 4 was confirmed by
comparison of its physical characteristics with literature data13
([
a]
D
ꢀ118° (c 1, MeOH); lit. [
a]
ꢀ122.8° (c 2.02, CHCl3); 1H
D
RMN spectrum and IR spectrum). The compound 5 with the unde-
sirable configuration was converted into 4 as previously re-
ported.13 The deprotection of the Boc function and demethylation
of 4 was achieved by reaction with boron tribromide in dichloro-
methane at ꢀ78 °C (Fig. 3).10 The basic treatment of 6ꢁHBr gives
the free amine 6, which easily isomerizes to the seven-membered
lactam 7, as previously noted in the degradation study of natural
Amicoumacins.9 The lactam 7 can be transformed to 6ꢁHCl only
at high temperature.
2.1.3. Analysis of the BSA conjugates
In summary, we have developed a new synthetic pathway to
the West block of Amicoumacins with major advantages compared
to previous syntheses11,13,15 (see Table 2 in Supplementary data).
Considering the three-dimensional structure of BSA, only 26
-NH2 groups of surface lysines (out of 59 lysine residues) of BSA
e
are theoretically available for coupling with haptens.44 A high ratio
of hapten per molecule of BSA ensures the efficiency of the cou-
pling reaction. Therefore, it is very important to verify the effi-
ciency of coupling before its administration to rabbits for
antibody generation because the use of homobifunctional linkers
increases the risk of undesired cross-linking between the same
molecules. The conjugates were analyzed by three independent
methods: UV-spectroscopy, sodium dodecyl sulfate–polyacryl-
amide gel electrophoresis (SDS–PAGE) and matrix-assisted laser
desorption–ionization mass spectrometry (MALDI-MS) (Table 1).
2.1.2. Conjugation to the carrier protein
For the production of specific antibodies against the Amicou-
macin family, the deprotected 3,4-dihydroisocoumarin moiety
6ꢁHBr was coupled to the bovine serum albumin (BSA) carrier pro-
tein via a homobifunctional linker deriving from D-tartaric acid 8
(Fig. 4). By this method, part of the hydroxylated portion of the
Amicoumacin East block is introduced directly during the prepara-
tion of the hapten–protein conjugates. This approach should en-
O
OH
OH
O
1) 10eq. BBr3, CH2Cl2,
-78 ºC to RT
1) NaHCO3 (sat), RT,
CHCl3 extraction
NH
+
O
6*HCl
4
H
2) 1N HCl, RT,
2) H2O
NH2*HBr
OH
7
10N HCl, 100 ºC
6*HBr
Figure 3. Removal of the protecting groups of the West block.
OH
O
O
O
H2N NH2
OH
O
0.1M NaHCO3,
pH8.5, DMF
H2N
+
O
OH O
n 6*HBr
n
+
N O
H
N
H
N
O
BSA
BSA
N
H
O
OH
O
H2N
NH2
O
OH
O
8
9a: n=50
9b: n=100
Figure 4. Preparation of the Amicoumacin–protein conjugates.

9386
S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
Table 1
Comparison of hapten incorporation on BSA conjugates determined by SDS–PAGE electrophoresis and MALDI-MS
Conjugate
Hapten:BSAa
Calculated hapten number per molecule of BSA (MWconjugate ꢀ MWBSA)/MWhapten
SDS–PAGEb
MALDI-MS
Exp. 1
Exp. 2
Matrix 1c
Matrix 2d
Average
Ion 2+
Ion 1+
Ion 2+
Ion 1+
9a
9b
50: 1
100: 1
8.3
6.4
4.5
7.5
9.3
10.5
9.25
10.3
10.7
11.8
10.6
11.4
10.0
11.0
a
Ratio used in conjugate preparation.
b
c
Molecular weights were calculated from peak centroids generated from SDS–PAGE bands with BioRad GS-800 Quantity One program from two independent experiences.
-Cyano-4-hydroxy-cinnamic acid.
a
d
Sinapinic acid.
The UV spectra display the differences between the carrier pro-
Noticeably, around 10–11 hapten molecules were bound to BSA,
and that this number did not change when the hapten to protein
coupling ratio used was increased. This observation was confirmed
by both SDS–PAGE and MALDI-MS. Thus, only part of the 26 theo-
retically available surface lysines of BSA is accessible for coupling
with haptens under our conditions and the utilization of 50 equiv
of hapten is sufficient enough to saturate available sites on protein
surface.
tein and the conjugates in the regions of the maximum absorbance
of the hapten, and provide a qualitative characterization of the
coupling efficiency (see Supplementary data). Conjugates were also
analyzed by SDS–PAGE. For the conjugates, smeared bands along
with weak bands for polymerized BSA were observed (see Supple-
mentary data). The molecular weight changes were determined for
each conjugate and the calculated hapten number per molecule of
BSA is presented in Table 1. The small changes detected in the
molecular weight can be explained by the poor accuracy of SDS–
PAGE (5%), moreover, the hapten density determined by this meth-
od varies a lot for the same sample preparation loaded in two
different gels. SDS–PAGE does not reflect the real degree of hapten
incorporation because it is insensitive to small changes in molecu-
lar weight. The main advantage of this method is the detection of
BSA self cross-linking as a side product during the conjugation
reaction.
2.2. ELISA development
2.2.1. Antibody titer determination
The hapten–protein conjugate 9a was used to generate poly-
clonal antibodies in two germ-free, pathogen-free rabbits. Compar-
ison of the binding of the two polyclonal immunosera with the
immobilized antigen coated onto the plates demonstrated that
immunosera 4676 was twice more concentrated than 4677
(titer ꢂ 2.16). Therefore, the former was retained for further tests,
and Figure 5 shows the results of the direct revelation test. It must
be noted that 1/1000 dilutions of each immunoserum were out of
range indicating a good response to the injection of the hapten–
protein conjugate. The coating with hapten–Thyr conjugates at
MALDI-MS gave the best determination of hapten conjugation.
The average molecular weight of each conjugate was calculated
from both singly (M+H)+ and doubly (M+2H)2+ charged peaks using
two different matrices (Table 1). This method enabled to determine
the average number of bound molecules with high reproducibility
and accuracy.
0.25
lg/mL and the 1/5000 dilution of antisera (final well dilution
3.50
1/2500
1/5000
3.00
1/10000
2.50
2.00
1.50
1.00
0.50
0.00
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Coating Concentrations (µg/mL)
Figure 5. Cross-test for titer determination applied to serum 4676. This is a direct measurement of the maximal binding for each couple of antisera and immobilized antigen
concentrations. For each data n = 4 and they are presented SD.

S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
9387
1/10,000) were used as standard conditions for preliminary com-
petitive ELISA tests.
pounds. When the unknown compound is not recognized by the
antibody there is no interference with the immobilized antigen
and the binding is close to 100%. When there is affinity of the
unknown compound for the antibody, the latter reacts preferably
with the free compound in solution rather than with the immo-
bilized hapten and the percentage of binding is reduced. When a
large set of dilution of the antigen is challenged with the immo-
bilized hapten for the antibody, a classical sigmoid curve is ob-
tained (data not shown). Classically, biochemical significance is
obtained for percentages of binding lying in the linear portion
of the sigmoid (i.e., between 30% and 75% of binding). Over
and below these values since the values, are no longer in the lin-
ear portion of the curve, the binding is no more proportional to
the affinity of the antibody for the unknown compound. The per-
centages of binding obtained in these conditions allow classifying
the affinity of the unknown samples for the polyclonal antibody
generated. To determine the specificity of the antibody, several
parent compounds 10–13 (entries 10–13, Table 2) were specially
synthesized by direct benzylic metalation of corresponding
unprotected benzoic acids (see Supplementary data). Semi-puri-
fied Amicoumacin B and culture supernatants of Amicoumacin
producing strains were used as a positive control. Compounds
6ꢁHCl and 6ꢁHBr, representing a major part of the Amicoumacin
West block, induce a significant displacement of the binding of
the antibody to the immobilized antigen. For 6ꢁHCl the displace-
ment reaches 35%. The 3-isobutyl-8-hydroxy-3,4-dihydroiso-
coumarin 10, which is a non-natural compound with a high
degree of homology to the West block, displaces the antibody
in the same manner as 6ꢁHBr. The fully protected West block
precursor 4 (entry 9, Table 2) was not recognized. No significant
cross-reactivity was observed with other synthetic 3,4-dihydro-
isocoumarin derivatives tested (compounds 11, 12, and 13).
These data indicate a good specificity of antibodies that are able
to selectively recognize the 8-hydroxy-3,4-dihydroisocoumarin
moiety. The compounds with another substitution in position 8
were not discriminated. The presence of an alkyl group in posi-
tion 3 is also important for good recognition.
2.2.2. Cross-reactivity
Specificity of the immunoassay was assessed by measuring
the interference of some structurally related substances or micro-
biological preparations from B. subtilis strains with the immobi-
lized hapten (hapten–Thyr) for the reaction with the polyclonal
antibody (Table 2). It is a competitive assay and values are ex-
pressed as percentages of total binding (Bo) obtained when the
antibody is only challenged with hapten immobilized on the
wells and without any competitive solution. Bi is the OD ob-
tained when the antibody is challenged with both the immobi-
lized hapten and the unknown solution. Note that the non-
specific binding is taken into account in Blank wells where the
antibody was present without the immobilized antigen. There-
fore data are expressed as a percentage of maximum binding,
that is (Bi ꢀ Blank)/(Bo ꢀ Blank). These transformations of the
data amplify the differences observed between the different com-
Table 2
Validation of the ELISA test: reactivity of antiserum with structurally related
chemicals and microbiological samples
Entry
Analyte
% of maximum binding
to coated antigen:
*
(Bi ꢀ Blank)/(Bo ꢀ Blank) 100a
1
Semi-purified Amicoumacin B
B. subtilis BS3
B. subtilis BS3 (diluted 1024)
B. subtilis M1-2
B. subtilis 168
B. subtilis BM15
6ꢁHCl
3.2
7.8
41
20
77
83
65
75
97
2b
3b
4b
5b
6b
7
8
9
6ꢁHBr
4
OH
O
O
O
10c
73
H
H
In summary, the cross-reactivity observed with different chem-
ical analogues of Amicoumacins confirmed that our immunogen
provided a favorable antigen presentation on the protein surface
and the specificity of the antibodies raised against the conjugate
9a is sufficient for qualitative determination of Amicoumacin
family.
10
OMe O
11c
95
91
2.2.3. ELISA validation for the detection of Amicoumacins
The ELISA test was validated by using a quantitative analysis of
Amicoumacins in culture supernatants of Bacillus strains that were
previously identified as positive or negative producers of these
compounds.5 Culture supernatants from negative Amicoumacins
production strains B. subtilis 168 and BM15, as well as bacterial cul-
ture media did not significantly displace the antibody from its
binding to the immobilized antigen (Table 2). In contrast, positive
results in the ELISA test were obtained with semi-purified Amicou-
macin B, and strains BS3 and M1-2 known for their production of
Amicoumacins.
11
O
O
O
12c
12
O
13c
100
87
Pinchuk et al.5 demonstrated that anti-microbial activity of
BS3 strain against S. aureus is related to the Amicoumacin B pro-
duction. Titer of BS3 supernatant against S. aureus was obtained
using a classical half dilution method as described by Pinchuk
et al.4,5 Using this method, we detected 64 Arbitrary Units
(AU) in the BS3 supernatant that correspond to a supernatant di-
luted 64 times (data not shown). In contrast, the ELISA devel-
oped enabled to detect Amicoumacin in BS3 culture
supernatant diluted 1024 times (1024 AU). Thus, the developed
ELISA test is about 20 times more sensitive than the Amicou-
macin detection previously described.5
13
14
Culture media
a
Bi, Bo, and Blank are the OD measured in the well at 490 nm. Bo is maximum
binding to the immobilized hapten-Thyr without any other compound added, Bi is
the OD obtained when the unknown compound is added, Blank is the non-specific
binding. When the unknown compound is not recognized by the antibody there is
no interference with the immobilized antigen and the binding is close to 100%.
b
Culture supernatant of Bacillus strain.
Chemical synthesis and structural characteristics are given in Supplementary
c
data.

9388
S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
(12.8 g, 16.6 ml, 0.11 mol, bp 120.5–121.0 °C) and 100 mL of dry
3. Conclusions
THF were added. The mixture was stirred magnetically and cooled
to ꢀ90 °C. After consecutive addition of s-BuLi (0.11 mol, 1.3 M in
cyclohexane/hexane, 85 mL) and 2-methoxybenzoic acid (7.6 g,
0.05 mol) in 30 ml of dry THF at ꢀ90 °C, the orange solution was
stirred for an additional hour. Methyl iodide (28.4 g, 12.45 ml,
0.20 mol) was added dropwise for 30 min. After stirring for 2 h,
the mixture was then treated with water, washed with diethyl
ether, and shaken, and the aqueous layer was acidified with 1 N
HCl down to pH 2. The mixture was diluted with diethyl ether,
and the organic layer was separated and dried with MgSO4. Filtra-
tion, concentration in vacuo, and recrystallization (diethyl ether/
pentane) gave 2 (4.15 g, 25 mmol, 50%). mp 138–140 °C (lit. 139–
140 °C45). 1H NMR (CDCl3) dH 11.58 (1H, OH); 7.30 (t; 1H; J = 8.1,
ArH); 6.88 (d; 1H, J = 7.6, ArH); 6.83 (d; 1H, J = 8.2, ArH); 3.91 (s,
3H, –O–CH3); 2.49 (s, 3H, Ar-CH3). NMR-13C (CDCl3) dC 170.3,
The goal of this work was the development of an ELISA test for
the detection of Amicoumacins for rapid and easy screening of bac-
terial producers and for future in vivo studies. The West block was
selected as principal common epitope for raising antibodies. A new
short and efficient synthesis of the Amicoumacin West block and
an original strategy for the preparation of Amicoumacin-protein
conjugates are described. As expected, the use of a D-tartaric acid
derivative as a coupling reagent enabled to mimic the common
portion of all of the compounds of the Amicoumacin family and
to increase the sensitivity of the immunoresponse without dis-
criminating any particular member of the family. MALDI-MS was
a very useful method for precise measurement of the hapten/BSA
ratio.
Our ELISA test for rapid and low cost detection of Amicoumacin
in biological samples uses a four-step procedure, and compared to
the classic method5 produces results in 3 h with a 20 times higher
sensitivity. The assay has applications as a specific screen for Ami-
coumacin producers among different bacterial strains, as well as to
study in vivo metabolism and bioavailability of these compounds.
155.8, 137.4, 130.2, 122.3, 107.7, 107.8, 55.8, 19.2. IR
m )
(cmꢀ1
3007, 1697, 1595, 1470, 1293, 1269, 1092, 1073, 915, 788. Spectral
characteristics were identical to those previously published.31
4.3. (3S)-3-[(10S)-tert-butyloxycarbonylamino-30-methylbutyl]-
8-methoxy-3,4-dihydroisocoumarin (4)
4. Experimental
s-Butyllithium (l4.4 mmol, 1.3 M in cyclohexane/hexane,
11.1 ml) was added dropwise to a solution of 2 (0.77 g, 4.64 mmol)
in 30 ml of dry THF at ꢀ90 °C. A bright red color appeared and the
solution was stirred for 45 min. In another flask, tert-butylmagne-
sium chloride (14.5 mmol, 2 M in diethyl ether, 7.3 ml) was added
4.1. Chemicals and instruments
2-Methoxybenzoic acid (99%), TMEDA (99%), methyl iodide
(99%), tert-butylmagnesium chloride (2.0 M in diethyl ether), and
boron tribromide (99%) were purchased from Sigma Aldrich Chem-
ical Co. (Saint Quentin Fallavier, France), while 4-hydroxyphenyl-
acetic acid (98%) was purchased from Acros Organics France
(Noisy, France). All other chemical reagents used were from Lan-
caster Synthesis Ltd (Bishheim, France).
to
(S)-2-tert-butyloxycarbonylamino-4-methyl-1-pentanal
3
(3.10 g, 14.4 mmol) in 10 ml of dry THF at ꢀ78 °C and stirred for
5 min before addition via cannula to the lithiated compound 2.
The bright red coloration disappeared progressively. The mixture
was stirred for a further 2 h at ꢀ90 °C and allowed to warm to
room temperature. Saturated aqueous ammonium chloride solu-
tion (100 ml) was added and pH was adjusted to pH 2–3 with
1 M hydrochloric acid. The vigorous stirring was maintained for
1 h 30 min at room temperature, then the solution was extracted
with ethyl acetate (3ꢂ 50 ml). The combined organic extracts were
washed with water (50 ml) and brine (50 ml), dried (MgSO4), and
concentrated under reduced pressure. Flash chromatography using
pentane–diethyl ether (4:1) as an eluent gave the desired com-
pound 4 as a pale yellow oil (0.51 g, 1.40 mmol, 30%, Rf = 0.08).
1H NMR (CDCl3) dH 7.42 (t, 1H, J = 8.3, ArH); 6.88 (d, 1H, J = 8.6;
ArH); 6.79 (d, 1H, J = 7.5, ArH); 4.85 (d, 1H, JNH–CH = 9.8, –NH–
CH–); 4.34 (d, 1H, J = 12.4, –O–CH–); 3.91 (s, 3H, Ar-OCH3); 3.65–
3.60 (m, 1H, –NH–CH–); 3.02 (dd, 1H, J = 16, 11.4, H4b); 2.75 (dd,
Melting points were determined with a Stuart Scientific melting
point apparatus SMP3 and are uncorrected. 1H and 13C NMR spec-
tra were recorded with a Bruker AC-300 FT (1H: 300 MHz, 13C:
75 MHz) using TMS as an internal standard. The chemical shifts
(d) and coupling constants (J) are, respectively, expressed in ppm
and Hz. IR spectra were recorded with a Perkin-Elmer paragon
1000 FT-IR spectrophotometer. Optical rotations were measured
on Bellingham Stanley Polarimeter ADP220 at ambient tempera-
ture. Mass spectra (both high and low resolution) were acquired
on a QStar Elite mass spectrometer (Applied Biosystems). The
instrument was equipped with an electrospray ionization (ESI)
source and spectra were recorded in the positive mode. The elec-
trospray needle was maintained at 4500 V and operated at room
temperature. Samples were introduced by injection through a
0
0
1H, J = 16, 2, H4a); 1.73–1.60 (m, 1H, H2 b; H3 ); 1.43–1.50 (m, 1H,
H2a ); 1.41 (s, 9H, t-Bu), 0.91 (d, 6H, J = 8.0, CH-(CH3)2). 13C NMR
0
10
the LC pump. Thin-layer chromatography (TLC) was performed
using SDS TLC plates, 0.25 mm, particle size 15 m, pore size
lL sample loop into a 200 l
L minꢀ1 flow of methanol from
(CDCl3)dC 170.1, 161.3, 156.6, 142.2, 134.6, 118.5, 113.4, 107.7,
79.7, 79.5, 55.9, 50.5, 41.0, 31.7, 28.3, 24.7, 22.9, 22.0. IR m
, cmꢀ1
l
3438, 3325, 2959, 2932, 2870, 1707, 1599, 1507, 1477, 1367,
1277, 1238, 1169, 910, 733. MS (ESI+) m/z 749 [2M+Na]+, 386
[M+Na]+; HR-MS (ESI+) m/z calcd for C20H29NO5, 386.1937
60 Å. Merck silica gel 60 (70–230 mesh) and (0.063–0.200 mm)
were used for flash chromatography. The spots were visualized
with UV lamp or revealed with ninhydrin. Diethyl ether and THF
were distilled over sodium wire and benzophenone under argon
atmosphere immediately before use. Dichloromethane was dis-
tilled over CaH2. n-BuLi (2.5 M solution in hexane) and s-BuLi
(1.3 M solution in cyclohexane/hexane) were purchased from Ac-
ros Chemical Co., Inc. and used after titration with diphenylacetic
acid. All moisture-sensitive reactions were performed under argon
atmosphere in oven-dried or flame-dried glassware.
[M+Na]+; found, 386.1941. [
a]
ꢀ118° (c 1, MeOH).
D
The followed fraction gave the minor diastereoisomer 5 (3R)-3-
0
[(1 S)-tert-butyloxycarbonylamino-30-methylbutyl]-8-methoxy-3,4-
dihydroisocoumarin. This product was isolated as a pale yellow oil
(0.17 g, 0.464 mmol, 9%, Rf = 0.05). 1H NMR (CDCl3) dH 7.43 (t, 1H,
J = 8.3, ArH); 6.89 (d, 1H, J = 8.6, ArH); 6.81 (d, 1H, J = 7.6, ArH);
4.89 (br d, 1H, JNHꢀCH = 9.8, –NH–CH–); 4.36 (d, 1H, J = 9.8, –O–
CH–); 3.91 (s, 3H, Ar-OCH3), 3.72–3.69 (m, 1H, –NH–CH–); 2.96
(br d, 1H, J = 12.4, H4b); 2.77 (dd, 1H, J = 16, 2.5, H4a); 1.47–1.80
4.2. 2-Methoxy-6-methyl-benzoic acid (2)
0
0
0
(m; 3H; H2 a; H2 b; H3 ); 1.42 (s, 9H, t-Bu), 0.93 (d, 3H, J = 7.2; –
CH–(CH3)2); 0.91 (d, 3H, J = 7.2; –CH–(CH3)2). 13C NMR(CDCl3) dC
169.3, 160.1, 154.6, 141.3, 137.1, 122.1, 118.4, 112.7, 79.4, 78.8,
To a 500 mL round-bottomed flask equipped with a condenser
and a rubber septum under argon, freshly redistilled TMEDA
55.1, 49.9, 40.1, 30.7, 27.3, 23.7, 22.7, 20.3. IR m
/cmꢀ1 3326, 2959,

S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
9389
1707, 1596, 1508, 1477, 1238, 1169, 911, 730. MS (ESI+) m/z 386
and used rapidly. 1H NMR (DMSO-d6) dH 6.73 (d, 2H, J = 7.5, OH);
4.89 (d, 2H, J = 7.5, CHOH); 2.83 (s, 8H, –CH2–CH2–). 13C NMR
(DMSO-d6) dC 170.1, 167.0, 61.9, 26.3. Spectral characteristics were
identical to those previously published.46
[M+Na]+, 364 [M+H]+.
4.4. (3S)-3-[(10S)-t-butyloxycarbonylamino-30-methylbutyl]-8-
hydroxy-3,4-dihydroisocoumarin hydrobromide (6ꢁHBr)
4.7. Preparation of hapten–BSA conjugates (9a and 9b)
Boron tribromide (400 mg, 150 ll, 1.6 mmol) was added drop-
wise to a solution of 4 (140 mg, 0.4 mmol) in 30 ml of dichloro-
methane at ꢀ78 °C over 5 min. The mixture was stirred for
30 min at ꢀ78 °C. After addition of distilled water (10 ml), the mix-
ture was allowed to warm to room temperature and extracted with
water (3ꢂ 10 ml). The combined aqueous extracts were lyophilized
to give a light brown powder (110 mg, 85%). After recrystalization
from ethanol light brown crystals were obtained. 1H NMR (MeOH-
d4) dH 7.53 (t, 1H, J = 8.4, ArH); 6.92 (d, 1H, J = 8.6, ArH); 6.89 (d, 1H,
J = 8.2, ArH); 4.76 (m, 1H, –O–CH–); 3.60–3.57 (m, 1H, –NH–CH–);
Conjugates were prepared at two different ratios of hapten to
BSA (50:1 for 9a and 100:1 for 9b). BSA (33.5 mg, 0.5
dissolved in 5 ml of 0.1 M NaHCO3 (pH 8.5) and cooled to +4 °C.
6ꢁHBr (8.3 mg, 25 mol for 9a and 16.5 mg, 50 mol for 9b) in
DMF (0.5 ml) was added. Freshly prepared 8 (8.6 mg, 25 mol for
9a and 17.2 mg, 50 mol for 9b) was dissolved in dry DMF
lmol) was
l
l
l
l
(0.2 ml) and added to the mixture. After stirring at +4 °C for 4 h,
the mixture was dialyzed (molecular weight cut off 6000–
8000 Da) against distilled water (2ꢂ 1000 ml) at +4 °C for 48 h
and lyophilized to afford a white powder (35 mg for 9a and
36 mg for 9b). The products were stored at ꢀ20 °C.
0
3.20–3.05 (m, 2H, H4); 1.90–1.75 (m, 1H, H3 ); 1.71–1.60 (m, 2H,
0
0
H
2 a; H2 b); 1.04 (d, 3H, J = 6.3, –CH–(CH3)2); 1.02 (d, 3H, J = 6.3,
–CH–(CH3)2). 13C NMR (MeOH-d4) dC 169.8, 159.9, 139.6, 136.3,
120.6, 116.1, 108.4, 83.5, 52.0, 42.6, 30.0, 24.5, 23.6, 21.5. IR
m,
4.8. Coating antigens
cmꢀ1 3364, 2956, 1685, 1462, 1259, 1025, 799. HR-MS (ESI+) m/z
calcd for C14H19NO3, 250.1437 [M+H]+; found, 250.1431.
The swine thyroglobulin (Thyr) was used for the preparation
of the coating antigen. The compound 6ꢁHCl was attached to Thyr
by the same method as described above for 9a and 9b except for
(1) the stoichiometric ratio hapten/Thyr was 200:1 and (2) the pro-
tein solution was prepared in borate–boric buffer (0.2 M, pH 8.7).
4.5. 6ꢁHCl and (3S,4S)-4,9-Dihydroxy-3-isobutyl-2,3,4,5-
tetrahydro-benzo[c]azepin-1-one (7)
The lyophilized crude 6ꢁHBr (100 mg, 0.3 mmol) was treated
with saturated aqueous sodium hydrocarbonate (15 ml). The solu-
tion was extracted with dichloromethane (3ꢂ 25 ml), then the
combined organic layers were dried over MgSO4. After filtration,
the organic layer was concentrated under reduced pressure to give
a 3/1 mixture of 6 and 7 as a brown oil, which was dissolved in
10 ml of MeOH and treated with 5 ml of 1 N HCl. After concentra-
tion under reduced pressure, chromatography of the residue using
hexane–ethyl acetate (3:7) as an eluent followed by elution with
methanol gave 6ꢁHCl (65 mg, 65%) (fraction eluted with MeOH)
as a white solid and the compound 7 as a colorless oil (22 mg,
22%). The product 7 was quantitatively transformed into 6ꢁHCl by
heating with concd HCl at 100 °C for 2 h.
4.9. Characterization of conjugates
The conjugates 9a and 9b were analyzed by UV-spectroscopy,
sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–
PAGE), and matrix-assisted laser desorption–ionization mass spec-
trometry (MALDI-MS). UV-spectroscopy: the UV spectra were run
with a UV/vis spectrophotometer (Hitachi U-3300). The samples
were prepared at a concentration of 1 mg/mL in distilled water.
SDS–PAGE: conjugates were dissolved in the SDS–PAGE sample
buffer (10 mM Tris/HCl, 1 mM EDTA, 2.5% SDS, and 5.0% b-mercap-
toethanol, pH 8.0) to 1 mg/mL and heated at 100 °C for 5 min. Gels
were loaded with 2–3
ll of the sample and run under denaturing
6ꢁHCl: physical and NMR characteristics were identical to those
conditions then were stained with Coomassie Blue according to
the manufacturers’ recommendations. The resulting gels were
evaluated first by scanning and second by analyzing the gel image,
with BioRad GS-800 Quantity One program. From the gel image a
graphical presentation of the gel bands observed was produced
and the molecular weights from peak centroids were calculated.
MALDI-MS: mass spectra were recorded in the linear mode with a
Bruker Reflex III mass spectrometer equipped with a UV laser
previously published.11,13
4.5.1. (3S,4S)-4,9-dihydroxy-3-isobutyl-2,3,4,5-tetrahydro-
benzo[c]azepin-1-one (7)
1H NMR (MeOH-d4) dH 7.27 (t, 1H, J = 8.3, ArH); 6.83 (d, 1H,
J = 8.3, ArH); 6.78 (d, 1H, J = 8.3, ArH); 4.03 (m, 1H, –C(4)HOH–);
3.21–2.95 (m, 2H, –NH–C(3)H–; H5a); 2.72 (dd, 1H, J = 13, 9.8,
0
0
0
H5b); 1.81–1.55 (m, 2H, H2 a; H3 ); 1.47–1.28 (m, 1H, H2 b) 0.91 (d,
3H, J = 6.8, –CH–(CH3)2); 0.81 (d, 3H, J = 6.4, –CH–(CH3)2). 13C
NMR (MeOH-d4) dC 174.2, 159.9, 137.6, 132.4, 120.1, 116.2,
(337 nm). Samples (10
tions (sinapinic acid or
in water/acetonitrile) and 1
stainless steel target and left to dry at room temperature. Spectra
were acquired in the external calibration mode using BSA as a
reference.
l
a
M) were mixed (1/1 v/v) with matrix solu-
-cyano-4-hydroxy-cinnamic acid, 10 mM
l of the mixture was applied on a
l
115.1, 75.3, 54.5, 40.4, 37.8, 24.3, 21.9, 20.8. IR
m
, cmꢀ1 3421,
2956, 2526, 1635, 1457, 1212, 1115, 971. MS (ESI+) m/z 250
[M+H]+.
4.6. (2S,3R)-2,3-Dihydroxy-succinic acid bis-(2,5-dioxo-
4.10. Antibody production
pyrrolidin-1-yl) ester (8)
The polyclonal antibodies were raised in two different rabbits
against the BSA–hapten 9a coupled in a borate–boric buffer with
a coupling ratio of 1/100. Immunizations were achieved in regular
house holding conditions and on sanitary controlled New Zealand
rabbits from Millegen (France). Rabbits were first ear-sampled for
Dry
D
-(ꢀ)-tartaric acid (1.5 g, 10 mmol) and N-hydroxy-
succinimide (2.5 g, 22 mmol) were dissolved in 30 ml of dry THF
and cooled to 0 °C. A solution of dicyclocarbodiimide (4.5 g,
22 mmol) in 20 ml of dry THF was added dropwise. The mixture
was allowed to warm to room temperature then stirred for 16 h.
The precipitated dicyclohexylurea was filtered off, the filtrate
was evaporated under reduced pressure and a residue was tritu-
rated with dry EtOH to afford 8 as a white solid (2.4 g, 70%, mp
198 °C). The product was very hydroscopic, was kept in desiccator
the pre-immunoserum test. Hapten–BSA conjugates (500 lg) were
injected each time. All injections were performed at multiple
points. A program of five consecutive injections was carried out.
For the two first injections separated by a 2-week period, the
conjugate was dissolved in 1 mL PBS-complete Freund’s adjuvant

9390
S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
(v–v). After the first two injections, three additional injections
were performed in PBS-incomplete Freund adjuvant (v–v) at a
3-week interval. Five days after the fifth injection, a test was per-
formed on a 5 mL blood sample to check the titers of the two
immunsera. Fifty milliliters of the sample was collected 1 week
after the test. Serum was obtained after blood clotting at 4 °C, for
24 h and centrifuged at 3000g for 10 min at 4 °C. Sera were stored
at ꢀ20 °C in small aliquots. The efficiency of the immunization was
tested in ELISA (as described below) by a direct binding of the anti-
body onto the coated conjugate.
coated in the wells without any other compound added. When un-
known samples are added they interfere with the immobilized
antigen and fix a part of the antibodies available. Therefore,
results are expressed as a percentage of maximum binding, that
is (Bi ꢀ Blank)/(Bo ꢀ Blank), where Bi is the OD obtained when
the unknown compound is added (Bo and Bi are means of homog-
enous duplicates performed twice, i.e., four independent determi-
nations). In these conditions supernatants of bacterial cultures,
semi-purified Amicoumacin B, and solutions of pure synthesized
compounds (100 lg/mL) were tested.
Different Bacillus strains were selected in relation to our previ-
ous studies.4,5 B. subtilis M1-2 and the probiotic strain BS3 were se-
lected as Amicoumacin producing strains.5 The B. subtilis 168 type
strain and BM15 were selected as a negative control, because these
strains do not produce Amicoumacins. Bacterial strains were cul-
tured in Mueller Hinton (MH) starch broth (Difco Lab, Detroit,
USA) at 30 °C during 72 h in an aerobic atmosphere as described
previously.5 Supernatants were obtained by centrifugation of the
culture broth at 10,000 rpm during 10 min and serial dilutions ef-
fected in PBS buffer.
4.11. Analysis of the antibody titer
These tests were performed with each of the antisera 4676 and
4677 as a direct revelation of the antibody binding for several dilu-
tions to different concentrations of immobilized antigen. Coating of
the microtitration plates was performed with the hapten–Thyr
conjugates (200
ate buffer (0.05 M, pH 9.6) at 4 °C, overnight. Concentrations of
coating-conjugates ranged from 10 g/mL to 0.312 g/mL with 1/
lL/well) (as the immobilized antigen) in a carbon-
l
l
2 dilution steps. The wells were then saturated with PBS-T-BSA-
DMSO (PBS containing 1 mg/mL BSA, 0.05% Tween 20, and 1%
DMSO) at 37 °C for 30 min. Plates were rinsed three times with
PBS-T-DMSO (PBS, 0.05% Tween 20, 1% DMSO). Serial dilutions
(1/1000; 1/2500; 1/5000; 1/10,000) of the specific antibodies were
Acknowledgments
We are grateful to ENITA de Bordeaux and Région Aquitaine for
funding immunological equipment.
then added (200
The plates were washed three times with PBS-T-DMSO. Then
200 L/well of the second antibody (swine anti-rabbit immuno-
lL/well). The incubation lasted for 2 h at 37 °C.
Supplementary data
l
globulin coupled to Peroxidase from Dako, France) was added in
PBS-T-BSA-DMSO. The incubation was performed at 37 °C for
Supplementary material contains some details of the hapten–
BSA conjugate characterization by UV-spectroscopy, SDS–PAGE
and MALDI-TOF spectrometry as well as the experimental proce-
dures and spectral characterization of synthetic parent compounds
of Amicoumacins, which were used for cross-reactivity tests (Table
2, entries 10–13). Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/j.bmc.
2008.08.017.
30 min. To measure the peroxidase activity, 200
strate solution containing 0.005 M o-phenylenediamine (10 mg in
20 mL) and 0.00025% H2O2 30% (5 L in 20 mL) in citrate-phos-
phate buffer (0.15 M, pH 5.0) was added. The reaction took place
lL/well of sub-
l
at room temperature for 30 min and was stopped with 50
of H2SO4 4 M. ODs were read at 490 nm.
lL/well
4.12. ELISA procedure and analysis of the microbiological
media
References and notes
1. Shimojima, Y.; Shirai, T.; Baba, T.; Hayashi, H. J. Med. Chem. 1985, 28, 3.
2. Itoh, J.; Omoto, S.; Shomura, T.; Nishizawa, N.; Miyado, S.; Yuda, Y.; Shibata, U.;
Inouye, S. J. Antibiot. (Tokyo) 1981, 34, 611.
3. Canedo Hernandez, L.; Acebal Sarabia, C.; Garcia Gravalos, D. Antitumor
isocoumarins. US patent 5,925,671, 1996; WO 9627594; Chem. Abstr. 1996, 125,
273726.
4. Pinchuk, I. V.; Bressollier, P.; Verneuil, B.; Fenet, B.; Sorokulova, I. B.; Megraud,
F.; Urdaci, M. C. Antimicrob. Agents Chemother. 2001, 45, 3156.
5. Pinchuk, I. V.; Bressollier, P.; Sorokulova, I. B.; Verneuil, B.; Urdaci, M. C. Res.
Microbiol. 2002, 153, 269.
6. Itoh, J.; Shomura, T.; Omoto, S.; Miyado, S.; Yuda, Y.; Shibata, U.; Inouye, S.
Agric. Biol. Chem. 1982, 46, 1255.
7. Shimojima, Y.; Hayashi, H.; Ooka, T.; Shibukawa, M. Agric. Biol. Chem. 1982, 46,
1823.
8. Shimojima, Y.; Hayashi, H.; Ooka, T.; Shibukawa, M.; Iitaka, Y. Tetrahedron Lett.
1982, 23, 5435.
9. Shimojima, Y.; Hayashi, H.; Ooka, T.; Shibukawa, M.; Iitaka, Y. Tetrahedron
1984, 40, 2519.
10. Hamada, Y.; Kawai, A.; Kohno, Y.; Hara, O.; Shiori, T. J. Am. Chem. Soc. 1989, 111,
1524.
The ELISA procedure described here is a competition between a
free compound and an immobilized antigen coated on the wells for
the polyclonal antibody. According to the previous tests performed
the following ELISA conditions were established. The 96-well ELISA
plates (Greiner bio-one, Les Ulis, France) were coated overnight at
4 °C with hapten–Thyr conjugates at 0.25 lg/mL (200 lL/well) in
carbonate buffer (0.05 M, pH 9.6). The wells were saturated with
PBS-T-BSA-DMSO (PBS containing 1 mg/mL BSA, 0.05% Tween 20,
and 1% DMSO) at 37 °C for 30 min and then washed three times
with PBS-T-DMSO (PBS, 0.05% Tween 20, and 1% DMSO).
For the analysis of the microbiological media, an unknown solu-
tion disposition (100 lL/wells) together with 100 lL/well of anti-
body 4676 1/5000 dilution in PBS-T-BSA-DMSO were incubated
for 2 h at 37 °C. The wells were washed three times with PBS-T-
DMSO. The secondary antibodies (200
DMSO were added and incubated at 37 °C for 30 min. Revelation
was carried out with 200 L/well of substrate solution containing
lL/well) in PBS-T-BSA-
11. Hamada, Y.; Hara, O.; Kawai, A.; Kohno, Y.; Shioiri, T. Tetrahedron 1991, 47,
8635.
12. Ward, R. A.; Procter, G. Tetrahedron Lett. 1992, 33, 3359.
13. Ward, R. A.; Procter, G. Tetrahedron 1995, 51, 12301.
14. Broady, S. D.; Rexhausen, J. E.; Thomas, E. J. J. Chem. Soc., Chem. Commun. 1991,
708.
l
0.005 M o-phenylenediamine and 0.00025% H2O2 30% in citrate-
phosphate buffer (0.15 M, pH 5.0). The reaction was performed at
room temperature for 30 min and was stopped with 50 lL/well
15. Broady, S. D.; Rexhausen, J. E.; Thomas, E. J. J. Chem. Soc., Perkin Trans. 1999, 1,
1083.
of 4 M H2SO4. ODs were read at 490 nm. The interpretation of
the data is performed taking into account the non-specific reac-
tions measured in Blanks (obtained in the presence of the antibody
but without the immobilized hapten) and with regard to the max-
imum of binding, namely Bo. Bo is obtained as in the titer test, by
allowing the antibody to react onto the immobilized hapten–Thyr
16. Kotsuki, H.; Araki, T.; Miyazaki, A.; Iwasaki, M.; Datta, P. K. Org. Lett. 1999, 1,
499.
17. Ghosh, A. K.; Bischoff, A.; Cappiello, J. Org. Lett. 2001, 3, 2677.
18. Ghosh, A. K.; Bischoff, A.; Cappiello, J. Eur. J. Org. Chem. 2003, 5, 821.
19. Superchi, S.; Minutolo, F.; Pini, D.; Salvadori, P. J. Org. Chem. 1996, 61, 3183.
20. Ghosh, A. K.; Cappiello, J. Tetrahedron Lett. 1998, 39, 8803.
21. Davies, G.; Russell, A. T. Tetrahedron Lett. 2002, 43, 8519.

S. Shinkaruk et al. / Bioorg. Med. Chem. 16 (2008) 9383–9391
9391
22. Shinozaki, K.; Mizuno, K.; Wakamatsu, H.; Masaki, Y. Chem. Pharm. Bull. 1996,
44, 1823.
23. Mukai, C.; Miyakawa, M.; Hanaoka, M. J. Chem. Soc., Perkin Trans. 1 1997, 6, 913.
24. Kotsuki, H.; Miyazaki, A.; Ochi, M. Tetrahedron Lett. 1991, 32, 4503.
25. Wheeler, M. J. Methods Mol. Biol. 2006, 324, 1.
26. Shinkaruk, S.; Lamothe, V.; Schmitter, J.-M.; Fructus, A.; Sauvant, P.; Vergne, S.;
Degueil, M.; Babin, P.; Bennetau, B.; Bennetau-Pelissero, C. J. Agric. Food Chem.
2008, 56, 6809.
27. Le Houerou, C.; Bennetau-Pelissero, C.; Lamothe, V.; Le Menn, F.; Babin, P.;
Bennetau, B. Tetrahedron 2000, 56, 295.
34. Gohier, F.; Castanet, A.-S.; Mortier, J. J. Org. Chem. 2005, 70, 1501.
35. Nguyen, T.-H.; Chau, N. T. T.; Castanet, A.-S.; Nguyen, K. P. P.; Mortier, J. Org.
Lett. 2005, 7, 2445.
36. Hauser, F. M.; Rhee, R. J. Am. Chem. Soc. 1977, 99, 4533.
37. Williams, D. R.; Barner, B. A.; Nishitani, K.; Phillips, J. G. J. Am. Chem. Soc. 1982,
104, 4708.
38. Thompson, R. C.; Kallmerten, J. J. Org. Chem. 1990, 55, 6076.
39. Roux, M.-C.; Paugam, R.; Rousseau, G. J. Org. Chem. 2001, 66, 4304.
40. Dondoni, A.; Perrone, D.; Merino, P. J. Org. Chem. 1995, 60, 8074.
41. Dieter Hoppe, T. H. Angew. Chem. Int. Ed. 1997, 36, 2282.
42. Tahara, N.; Fukuda, T.; Iwao, M. Tetrahedron Lett. 2002, 43, 9069.
43. Anderson, G. W.; Zimmerman, J. E.; Callahan, F. M. J. Am. Chem. Soc. 1964, 86,
1839.
28. Hauser, F. M.; Pogany, S. A. Synthesis 1980, 814.
29. Kessar, S. V.; Singh, P.; Vohra, R.; Kaur, N. P.; Vanugopal, D. J. Org. Chem. 1992,
57, 6716.
30. Schlosser, M.; Geneste, H. Tetrahedron 1998, 54, 10119.
31. Bennetau, B.; Mortier, J.; Moyroud, J.; Guesnet, J. J. Chem. Soc., Perkin Trans. 1
1995, 10, 1265.
32. Mortier, J.; Moyroud, J.; Bennetau, B.; Cain, P. A. J. Org. Chem. 1994, 59, 4042.
33. Moyroud, J.; Guesnet, J.-L.; Bennetau, B.; Mortier, J. Bull. Soc. Chim. Fr. 1996, 133,
133.
44. Singh, K. V.; Kaur, J.; Varshney, G. C.; Raje, M.; Suri, C. R. Bioconjug. Chem. 2004,
15, 168.
45. Kanakam, C. C.; Mani, N. S.; Ramanathan, H.; Subba Rao, G. S. R. J. Chem. Soc.,
Perkin Trans. 1 1989, 11, 1907.
46. Smith, R. J.; Capaldi, R. A.; Muchmore, D.; Dahlquist, F. Biochemistry 1978, 17,
3719.