Synthesis of 2-substituted β-cyclodextrin derivatives with a hydrolytic activity against the organophosphorylester paraoxon

Article abstract of DOI:10.1016/j.ejmech.2005.02.008

β-Cyclodextrin was substituted by an iodosobenzoic acid derivative to create a catalytic hydrolytic activity against neurotoxic organophosphorus agents. The catalytic moiety was introduced on a secondary hydroxy group at the position 2 of a glucose unit.

Full text of DOI:10.1016/j.ejmech.2005.02.008

European Journal of Medicinal Chemistry 40 (2005) 615–623
www.elsevier.com/locate/ejmech
Original article
Synthesis of 2-substituted b-cyclodextrin derivatives
with a hydrolytic activity against the organophosphorylester paraoxon
Nicolas Masurier a, François Estour a, Marie-Thérèse Froment b, Bertrand Lefèvre c,
Jean-Claude Debouzy d, Bernard Brasme c, Patrick Masson b, Olivier Lafont a,*
a Laboratoire de Pharmacochimie, Département de Chimie Organique Pharmaceutique, Faculté de Médecine et de Pharmacie de Rouen, 22 Boulevard
Gambetta, 76183 Rouen cedex 1, France
b Unité d’Enzymologie, Centre de Recherches du Service de Santé des Armées, 24, avenue des Maquis du Grésivaudan, B.P. 87,
38702 La Tronche cedex, France
c Service de Biospectrométrie, Centre de Recherches du Service de Santé des Armées, 24, avenue des Maquis du Grésivaudan, B.P. 87,
38702 La Tronche cedex, France
d Unité de Biophysique Cellulaire et Moléculaire, Centre de Recherches du Service de Santé des Armées, 24, avenue des Maquis du Grésivaudan, B.P. 87,
38702 La Tronche cedex, France
Received 29 October 2004; received in revised form 17 February 2005; accepted 24 February 2005
Available online 01 April 2005
Abstract
b-Cyclodextrin was substituted by an iodosobenzoic acid derivative to create a catalytic hydrolytic activity against neurotoxic organophos-
phorus agents. The catalytic moiety was introduced on a secondary hydroxy group at the position 2 of a glucose unit. Several b-cyclodextrin
derivatives were obtained. In these derivatives, the methylene linker occupied all potential positions on the aromatic ring. Kinetic assays were
carried out with paraoxon as organophosphate model. Three regioisomers hydrolyzed paraoxon, although the paraoxon-leaving group, para-
nitrophenol, was not released from the b-cyclodextrin torus.
© 2005 Elsevier SAS. All rights reserved.
Keywords: Cyclodextrin; Iodosobenzoic acid; Organophosphorus agents; Catalyst; Paraoxon
1. Introduction
can be achieved through the use of scavengers in active topi-
cal skin protectants, in decontaminant devices and in prophy-
lactic injectable solutions [4,5]. Bioscavengers are enzymes
that have the ability to react with OPs [6–8]. However,
enzymes cause numerous problems: large-scale production,
pharmaceutical formulations, cost, storage stability, immu-
nogenicity and residence time in the bloodstream. Thus, the
use of chemically accessible artificial enzymes as OPs scav-
engers for decontamination offer an attractive approach.
Cyclodextrins (CD) derivatives are potentially interesting
for skin protection. Their chemical structure with a hydro-
phobic cavity is known to mimic the binding and catalytic
activity of various enzymes [9–11]. Various esters showed
the ability to bind into the CD cavity and then to react with a
CD hydroxy group [12–18]. In the specific case of phosphate
esters, an oxyanion obtained from a CD hydroxy group,
reacted with the phosphate function, leading to phosphory-
lated CDs [19]. CD itself increased the cleavage of bis(para-
nitrophenyl)phosphate by almost two orders of magnitude
The acute toxicity of neurotoxic organophosphorus (OP)
agents is mainly due to their ability to inhibit cholinesterases
[1]. These enzymes play a key role in the cholinergic system
in terminating the action of the neurotransmitter acetylcho-
line. Therefore, intoxication by OPs leads to a severe cholin-
ergic syndrome which can cause seizures and the death by
respiratory failure [2]. Although prophylaxis and treatment
of intoxication by OPs have been improved in the past
20 years, they are imperfect and can induce side effects [3].
To overcome these disadvantages, the concept of scavenger
capable of degrading OPs has emerged as a new approach to
reduce toxicity of chemical neurotoxic agents. The support-
ing idea was to neutralize OP molecules before they reached
their neurological targets. Protection against OP poisoning
* Corresponding author. Tel.: +33 2 35 14 86 51; fax: +33 2 35 14 84 00.
E-mail address: Olivier.Lafont@univ-rouen.fr (O. Lafont).
0223-5234/$ - see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2005.02.008

616
N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
[20]. Some improvements could also be obtained by capping
[21,22]. But, the best method to obtain a suitable enzymomi-
metic model consisted in introducing a reactive functional
group on the CD torus [23–25]. Even if different CDs were
efficient, the choice of b-CD was justified by the fact that
soman and sarin showed a better affinity for b-CD than for a-
or c-CD [26]. It had also been proved that some fluorophos-
phates were able to acetylate rapidly b-CD [27–29].
The present study was especially based on interactions of
b-CD with soman and paraoxon. Generally, intoxications
caused by soman are highly dangerous because they are resis-
tant to usual treatment of anticholinesterase poisoning. In
order to test the activity of protectants against soman, the most
commonly used model is a less toxic organophosphorylester:
paraoxon (Fig. 5), the hydrolysis of which can easily be moni-
tored by UV spectroscopy.
Fig. 1. The three categories of hydroxy groups in b-CD glucosyl unit.
The essential properties of the inclusion into a cyclodex-
trin catalyst of an organophosphonate were explained in terms
of the geometries of the CD–OP inclusion complexes [30].
Data in relation with these parameters had been already pub-
lished concerning soman and paraoxon. The hydrophobic
nature of soman and its size compatible with b-CD made it a
substrate for the oligosaccharide molecule [31]. Soman reac-
tion with a secondary hydroxy group, leading to its inclusion
in the hydrophobic cavity of the b-CD, had been described
[32]. Soman and b-CD form a 1:1 complex [31]. Moreover, it
had been observed that soman is hydrolyzed about 2500 times
faster by the monoanion formed from a secondary CD
hydroxy group than by the hydroxide ion [31]. Concerning
paraoxon, its promotive catalysis by b-CD is probably caused
by sufficient proximity of the paraoxon phosphorus atom to
the secondary hydroxy groups surrounding the top of b-CD
torus [30]. It had been proved that the para-nitro phenoxy
ring of paraoxon was axially included in b-CD with the nitro-
end first, and with its phosphate group exposed in solvent
[33]. Moreover, the catalytic effect of cyclodextrin is essen-
tially dependent on the hydrogen-bonding interaction between
the secondary hydroxy groups of paraoxon [30].
Ortho-iodosobenzoic acid (IBA) is known to react as an
anionic nucleophile on phosphorus atom of various OPs
[34–38], including nerve active agents [39]. Moreover this
compound is efficient in neutral aqueous media. The choice
of IBA as catalytic system was then justified. In the
b-cyclodextrin (b-CD) structure, three hydroxy groups can
be potentially used to introduce IBA system, i.e. positions 2,
3 and 6 (Fig. 1). b-CD presenting a functional group branched
on a secondary alcohol (C-2 or C-3) or on a primary alcohol
(C-6) had shown different catalytic effects [40]. It was then
necessary to choose the best position to branch IBA. Intro-
duction of IBA as a catalytic moiety on a secondary hydroxy
group had already been made it a catalyst for soman inacti-
vation. But, these results were only partially published with-
out description of synthesis protocols, separation methods,
identification and structure of compounds [41].
that position being more reactive (vide infra) than the hydroxy
group in positon 3. In that purpose, methyl bromomethyl-2-
iodobenzoate derivatives were synthesized. Starting materi-
als were 2-iodomethyl-benzoic acid isomers with the methyl
group in various positions. The carboxylic function was pro-
tected as a methyl ester, then the methyl group was converted
in a bromomethyl group by treatment by N-bromosuccinimide
(NBS). Sodium hydride was first used to generate mono
sodium salt of b-CD which reacted with the bromomethyl
compound to give the alkylated b-CD. This latest was then
oxidized by peracetic acid, to yield the iodoso derivative. The
synthetic pathway is described in Scheme 1.
2. Synthesis
Each of the various steps of this synthesis made difficul-
ties appear. The first step consisted in the synthesis of vari-
ously substituted precursors of IBA. In order to study the influ-
ence of the position of the linker between IBA and b-CD on
the OP hydrolytic activity, the bromoalkyl group was succes-
sively introduced in the four free positions (3–6) of the methyl
2-iodomethylbenzoate ring (Fig. 2).
In order to functionalize b-CD by IBA, we decided to intro-
duce this reactive moiety in position 2, the hydroxy group in
Scheme 1. Synthetic pathway for alkylation of b-CD.

N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
617
Fig. 2. The free positions of the methyl 2-iodobenzoate ring.
2.1. Synthesis of bromomethyl aromatic precursors of IBA
For isomeric positions 3, 5 and 6, iodination of the aro-
matic ring was carried out via a diazoic intermediate by treat-
ment of the corresponding 2-aminomethylbenzoic acid 1–3
by sodium nitrite in acidic medium and then by potassium
iodide to give compounds 4–6 (Scheme 2) [42].
For the 4-bromomethyl derivative, another pathway had
to be used. The starting material was then 4-methylbenzoic
acid. Iodination was performed by thallation followed by sub-
sequent halogenation with potassium iodide and led to com-
pound 7 (Scheme 3) [43].
All 2-iodo-methylbenzoic derivatives 4–7 were then esteri-
fied by methanol in acidic medium. Only in the case of com-
pound 6, the reaction needed an acyl chloride intermediate
before reaction with methanol. For all compounds 8–11, radi-
cal bromination of the methyl group led, in a last step, to the
corresponding bromomethyl aromatic precursors of IBA
12–15 (Scheme 4).
Scheme 4. Synthesis of bromomethyl precursors of catalytic moiety.
is the structure, which makes its access difficult. The choice
to substitute position 2 in alkaline medium was confirmed by
the fact that it was the best possibility to obtain a monosub-
stitution. This choice was strengthened by the strategy devel-
oped by Rong and D’Souza: deprotonation of a single hydroxy
group to afford an oxyanion followed by nucleophilic substi-
tution of a halogeno derivative. The degree of functionaliza-
tion was controlled by the amount of alkali [44].
In order to synthesize 2-substituted derivatives, b-CD was
then monosubstituted by the bromomethyl precursors (12–
15) (Scheme 5).
2.2. Substitution of b-CD
The three sorts of hydroxy groups in b-CD present differ-
ent behaviors in alkaline medium. The hydroxy group in posi-
tion 6 which corresponds to primary alcohol is the less acidic
(pKa ~ 15–16) [44].Amongst the two secondary alcohols, the
most acidic is situated in position 2 (pKa = 12.1) [44]. In the
case of hydroxy group situated in position 3, the main factor
This reaction led predominantly, as expected, to the sub-
stitution of a hydroxy group in position 2 (Table 1). The best
selectivity was detected with methyl 6-bromomethyl-2-
iodobenzoate 14. In that case the reaction led to the isolation
of one single compound monoalkylated in position 2 of a glu-
cose unit. In the case of methyl 3-bromomethyl-2-
iodobenzoate 12, two regioisomers were obtained. The
2-substituted compound was the major product, and the
by-product was monosubstituted in position 3. The 4-bromo-
methyl analog 15 led also to two regioisomers, but in that
Scheme 2. Iodination of the aromatic ring via diazotation.
Scheme 3
.
Iodination of the aromatic ring via thallation.
Scheme 5. Alkylation of b-CD.

618
N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
Table 1
Yields in 2-substituted compounds
Bromomethyl precursor
Compound 12
Compound 13
Compound 14
Compound 15
Total yield (%)
Analytic evaluation of 2-substitution rate (%)
Isolated 2-substituted compound yield
18% (compound 16)
Not isolated (compound 17)
16% (compound 18)
24
37
16
43
77
57
100
77
31% (compound 19)
Conditions for substitution of b-cyclodextrin: NaH in DMF.
case, the by-product was monosubstituted in position 6. The
lowest selectivity was observed in the case of the
5-bromomethyl compound 13 and then three regioisomers,
respectively, monosubstituted in positions 2, 3 and 6 were
formed.
These differences in the reactivity could be explained by
two phenomenons. First of all sodium hydride is a strong base
and the differences in acidity of hydroxy groups were not in
the best conditions to be revealed. Then, the structures of the
variously substituted benzyl bromides were not equivalent.
For example, in the case of compound 18, the linker in ortho
position of the carboxymethyl group could orient its reactive
site towards a hydroxy group in position 2. On the other hand,
in the case of compound 17, the linker in meta position of the
carboxymethyl group could allow the nucleophilic attack by
an anion of the b-CD situated either in position 2, 3 or 6.
Anyway substitution in position 2, 3 or 6 of b-CD with this
base–solvent system, i.e. sodium hydride in dimethylforma-
mide, had already been described [45,46].
Scheme 6. Conversion into iodoso derivatives.
Fig. 3. Active form of 2-iodosobenzoic acid.
In order to get reproducible kinetic assays and to compare
OP-hydrolyzing efficiency of the different derivatives, it was
essential to test hydrolytic activity on a single derivative and
not on a mixture of regioisomers. Compound 18 being
devoided of any side product, it was then easily isolated by
column chromatography. For compounds 16 and 19, they were
separated from the respective side products by column chro-
matography. No condition for chromatographic separation
could be found in the case of compound 17. It was then nec-
essary to optimize the experimental conditions of b-CD sub-
stitution in order to obtain this product as a single isomer. To
prevent a deprotonation of the three kinds of alcohols, the
reaction was realized with sodium hydroxide which is a
weaker base than sodium hydride. No substitution was
observed in dimethylformamide, but the same experiment
with dimethylsulfoxide as solvent allowed to obtain com-
pound 17 as a single 2-substituted b-CD isomer. With sodium
hydroxide the yields were lower than with sodium hydride,
but regioselectivity was better.
b-CD derivatives monoalkylated in position 2 were then
converted into corresponding iodoso derivatives by peracetic
acid oxidation in acetic acid medium (Scheme 6). The lack of
reactivity of alcohols in the b-CD with peracetic acid was
proved by treatment of b-CD itself under the same condi-
tions. It was known that ortho-iodoso carboxylic acid deriva-
tives underwent an equilibrium leading to a cyclic structure
which was the active form (Fig. 3) [39].
Fig. 4. Tested compounds against paraoxon.
3. Protective activity
Hydrolysis assays were carried out with paraoxon as orga-
nophosphate model (Figs. 5 and 6). Compound 21 was active
against paraoxon. Compounds 22 and 23 showed a weak
hydrolytic activity. Compound 20 was inactive. This pro-
The four iodoso regiosisomers 20–23 which were tested
are presented in Fig. 4 under this cyclic form.

N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
619
Fig. 5. Chemical structure of paraoxon.
Fig. 7. Effect of competitors on hydrolysis of paraoxon by b-CD derivative
22.
nols substituted by electron-withdrawing groups, acetylsali-
cylic acid, ortho- and meta-nitrophenol caused increasing
inhibition of paraoxon hydrolysis. It is important to notice
that para-nitrophenol led to complete inhibition of hydroly-
sis. These observations provided evidence that para-
nitrophenol formation during paraoxon hydrolysis limitated
the efficiency of b-CD catalyst. The complex formation con-
stants were then determined for both protonated and depro-
tonated forms of phenol, ortho-, meta- and para-nitrophenols
with b-CD. In the single case of the para isomer of nitrophe-
nol, the acid-base (phenol-phenolate) equilibrium was shifted
toward the ionic form by addition of b-CD in the media [49].
The particular high stability (410 40 dm3 mol–1) of the com-
plex between para-nitrophenolate anion and b-CD could be
explained by polarizability and the resonance delocalization
in this anion which increased the electron density of the
nucleus. This concept allowed to explain the difficulty of the
b-CD catalyst to release the paraoxon leaving group.
Fig. 6. Hydrolysis of paraoxon by iodoso-substituted b-CD derivatives.
vided evidence that hydrolytic activity of b-CD derivatives
was regio-controlled. The relative position of the ether linker
between the IBA and b-CD (ortho, meta or para to the iodoso
group) had an influence. No activity was observed when the
linker was in ortho to the iodoso group, the steric hindrance
being maximum.An intermediate activity was observed when
the linker was in meta-position but the best hydrolytic activ-
ity was obtained when iodoso group was in para-position. It
is likely that the ether linker in para-position of the iodoso
group optimized interactions between the catalytic moiety and
the OP compound scavenged into the hydrophobic cavity of
b-CD.
The paraoxon molecule being included in CD cavity in
the case of the four isomers, it was then proved that hydro-
lytic activity against paraoxon was depended only on the ori-
entation of the iodoso group in the surroundings of b-CD,
and on the flexibility of the linker. Nevertheless, hydrolysis
of paraoxon showed saturation kinetics with the three active
compounds 21–23. The higher the paraoxon concentration,
the earlier the saturation plateau appeared (Fig. 6). Hypoth-
esis was made that the b-CD catalyst could be poisoned by
the paraoxon leaving group, para-nitrophenol which was not
released from the torus of the CD molecule. The size of para-
nitrophenol molecule is suitable for being included in b-CD,
and it is known that phenol derivatives can form more or less
stable complexes with b-CD [47,48]. Kinetic measurements
were then performed with compound 22, using 3 mM
paraoxon in the presence of various potential competitors
(final concentration: 0.5 mM) of paraoxon (Fig. 7). Phenol
itself did not alter CD-catalyzed hydrolysis of paraoxon. Phe-
4. Conclusion
Different b-CD derivatives, monosubstituted by IBA on a
secondary hydroxy group, in position 2 of a glucose unit were
prepared. The reactivity of hydroxy groups in position 2, 3,
and 6 being different, it was possible to substitute mainly posi-
tion 2 with sodium hydride as a base and bromobenzyl deriva-
tives as nucleophile acceptors. But, when by-products were
difficult to separate from the 2-substituted derivative, sodium
hydroxide allowed a better regioselectivity. Their capacity to
hydrolyze OP was tested with paraoxon. Several IBA-
monosubstituted b-CDs displayed a hydrolytic activity against
paraoxon. It was proved that catalytic activity was strongly
dependent on the position of the linker between IBA and b-CD
on the aromatic moiety. Compound 21 substituted on posi-

620
N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
tion 5′ of the nucleus proved to be the most potent com-
pound. The efficacy of the OP-hydrolyzing was limited due
to poisoning of the catalyst by para-nitrophenol, the paraoxon-
leaving group. This competitive inhibition is related to
paraoxon use as model OP and should not be observed with
OPs devoided of similar leaving group.
gel; solvent: cyclohexane/ethyl acetate (4:1; v/v). Compound
10 was obtained as a yellow oil (yield 70%); IR (NaCl): m
(cm–1) 1740; 1H-NMR (CDCl3): d 2.34 (s, 3H, CH3), 3.96 (s,
3H, OCH3), 6.99 (dd, 1H, J = 7.9 Hz, J = 7.4 Hz, H-4), 7.17
(d, 1H, J = 7.4 Hz, H-5), 7.64 (d, 1H, J = 7.9 Hz, H-3); 13C-
NMR (CDCl3): d 20.0 (CH3), 52.5 (OCH3), 91.8 (C-2), 129.6
(C-5), 130.6 (C-4), 136.3 (C-3), 136.5 (C-1), 140.1 (C-6),
169.4 (CO). Anal. (C9H9O2I): C, H.
5. Experimental protocols
5.1.2. Methyl 2-iodo-4-methylbenzoate (11)
To a solution of 3 g (11.4 mmol) of 2-iodo-4-methylbenzoic
acid (7) in 15 ml (370 mmol) of methanol, were added 1.5 ml
of concentrated sulfuric acid. The mixture was refluxed for
5 h. After cooling, the reaction mixture was neutralized with
aqueous NaHCO3 solution, first extracted by ethyl acetate
(50 ml), then extracted again by ethyl acetate (20 ml). The
organic layers were collected, washed with a 20% aqueous
solution of sodium thiosulfate, then with distilled water
(50 ml). The organic solution was separated and dried over
sodium sulfate. The solvent was evaporated under reduced
pressure and the residual oil was chromatographied on silica
gel; solvent: chloroform. Compound 11 was obtained as a
yellow oil (yield 85.5%); IR (NaCl): m (cm–1) 1750; 1H-NMR
(CDCl3): d 2.25 (s, 3H, CH3), 3.83 (s, 3H, OCH3), 7.11 (d,
1H, J = 7.9 Hz, H-5), 7.66 (d, 1H, J = 7.9 Hz, H-6), 7.76 (s,
1 H, H-3); 13C-NMR (CDCl3): d 20.9 (CH3), 52.6 (OCH3),
94.7 (C-2), 129.0 (C-5), 131.3 (C-6), 131.9 (C-1), 142.3 (C-3),
143.9 (C-4), 167.0 (CO). Anal. (C9H9O2I): C, H.
5.1. Chemistry
Melting points were determined on a Kofler hot-plate melt-
ing point apparatus and were not corrected. IR spectra were
obtained on a Shimadzu IR-408 spectrometer. Absorption
bands are expressed in cm–1 and only noteworthy absorp-
tions are listed. 1H- and 13C-NMR spectra were recorded on
a Bruker DPX 300 spectrometer working at 300 MHz (1H-
NMR) and 75 MHz (13C-NMR). Chemical shifts are reported
in ppm downfield d from tetramethylsilane. The electrospray
ionization mass spectrometry (ESI-MS) experiments were
performed on a MICROMASS Quattro II (altrincham, G.B.).
Solutions were introduced through a Z-spray source. This one
was continuously fed (5 µl min–1) with a solution (1 mg ml–1)
of the studied compound which was diluted in a mixture of
acetonitrile/water/formic acid (60:40:0.05; v/v/v). This solu-
tion was delivered by a syringe pump HARVARD APPARA-
TUS type 22A. The temperature of the source was set at 80 °C.
The electrospray probe (capillary) voltage was optimized in
the range 2.5–4 kV for positive or negative ion electrospray.
The sample cone was set within the range 40–160 V. Mi-
croanalyses are indicated only by symbols of the elements
analyzed. The results obtained had a maximum deviation of
0.4% from the theoretical value. For iodoso derivatives 20–23
no correct analytical data were obtained, probably as a result
of their highly hygroscopic nature and their complexation with
solvents: their purity was established by TLC in two different
systems and NMR spectroscopy. 2-Iodomethylbenzoic acids
4–7 [42,43,50], methyl 2-iodomethylbenzoate 8 [50] and 9
[38] and methyl 5-bromomethyl-2-iodobenzoate 13 [38] were
synthesized as described in the literature.
5.1.3. Methyl bromomethyl-2-iodobenzoate (12, 14 and 15)
Seven hundred thirty milligrams (41 mmol) of NBS and
two drops of bromine were added to a solution of 1.2 g
(4.35 mmol) of the corresponding methyl 2-iodo-
methylbenzoate (8–11) in 45 ml of carbon tetrachloride. Then
310 mg (1.3 mmol) of benzoyl peroxide were added to the
medium. The mixture was refluxed for 24 h. After cooling
and filtration, the solution was washed with 20% aqueous
sodium thiosulfate (2 × 50 ml), then with brine (50 ml), and
finally with distilled water (50 ml). The organic layer was
dried over sodium sulfate. The solvent was evaporated under
reduced pressure. The crude product was chromatographied
on silica gel; solvent: cyclohexane/tetrahydrofuran (14:1; v/v).
Methyl 3-bromomethyl-2-iodobenzoate (12) was obtained
as a yellow oil (yield 34.5%); IR (NaCl): m (cm–1) 1740;
1H-NMR (CDCl3): d 3.94 (s, 3H, CH3), 4.69 (s, 2H, CH2),
7.34 (t, J = 7.5 Hz, H-5), 7.48 (dd, J = 7.5, 1.5 Hz, H-4), 7.57
(dd, J = 7.5, 1.5 Hz, H-6); 13C-NMR (CDCl3): d 39.7 (CH2),
52.8 (OCH3), 99.3 (C-2), 129.7 (C-5), 130.6 (C-6), 132.5
(C-4), 139.3 (C-1), 142.1 (C-3), 168.2 (CO). Anal.
(C9H8O2BrI): C, H.
5.1.1. Methyl 2-iodo-6-methylbenzoate (10)
A solution of 2.8 g (10.7 mmol) of 2-iodo-6-methylbenzoic
acid (6) in 3 ml (41 mmol) of thionyl chloride was refluxed
for 2 h. Twenty-five milliliters (617 mmol) of methanol were
added and the mixture was refluxed for 3 additional h. After
cooling, the reaction mixture was neutralized with aqueous
NaHCO3 solution, first extracted by ethyl acetate (50 ml),
then extracted again by ethyl acetate (20 ml). The organic
layers were collected, washed with a 20% aqueous solution
of sodium thiosulfate, then with a 10% aqueous solution of
sodium hydroxide (50 ml), and finally with distilled water
(50 ml). The organic solution was separated and dried over
sodium sulfate. The solvent was evaporated under reduced
pressure and the residual oil was chromatographied on silica
Methyl 6-bromomethyl-2-iodobenzoate (14) was obtained
as a yellow solid (yield 45%); m.p. 67 °C; IR (nujol): m (cm–1)
1
1755; H-NMR (CDCl3): d 3.99 (s, 3H, CH3), 4.47 (s, 2H,
CH2), 7.09 (dd, 1H, J = 8.0, 7.5 Hz, H-4), 7.40 (d, 1H,
J = 7.5 Hz, H-3), 7.78 (d, 1H, J = 8.0 Hz, H-5); 13C-NMR
(CDCl3): d 30.0 (CH2); 52.9 (OCH3); 93.0 (C-2); 130.0 (C-5);

N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
621
131.3 (C-4); 136.8 (C-1); 139.5 (C-3); 140.1 (C-6); 175.4
(CO). Anal. (C9H8O2BrI): C, H.
4.54 (m, 7H, OH-6), 4.79 (bs, 7H, H-1), 5.95 (m, 13H,
OH-2 and 3), 7.51 (d, 1H, J = 7.5 Hz, H-6Ar), 7,66 (d, 1H,
J = 7.5 Hz, H-5Ar), 7.99 (s, 1H, H-2Ar); 13C-NMR (CDCl3):
d 52.6 (OCH3), 55.1 (OCH2), 59.9 (C-6), 71.7 (C-3′), 72.1
(C-5), 72.4 (C-2), 73.1 (C-3), 79.5 (C-2′), 81.6 (C-4), 81.9
(C-4′), 94.6 (C-I), 101.7 (C-1′), 101.9 (C-1), 127.4, 130.3,
134.8, 139.5, 143.3 (5× C-Ar), 167 (CO); MS ES+: 1431 (M
+ Na)+; MS ES–: 1407 (M – H)–, 1443 (M + Cl)–, 1445 (M +
Cl)–. Anal. (C51H77O37I): C, H.
Methyl 4-bromomethyl-2-iodobenzoate (15) was obtained
as a yellow solid (yield 36%); m.p. 98 °C; IR (nujol): m (cm–1)
1
1740; H-NMR (CDCl3): d 3.93 (s, 3H, CH3), 4.39 (s, 2H,
CH2), 7.42 (dd, 1H, J = 8.0, 1.0 Hz, H-5), 7.78 (d, 1H,
J = 8.0 Hz, H-6), 8.01 (d, 1H, J = 1.0 Hz, H-3); 13C-NMR
(CDCl3): d 30.7 (CH2), 52.7 (OCH3), 94.4 (C-2), 128.6 (C-5),
131.3 (C-6), 134.7 (C-1), 141.7 (C-3), 142.6 (C-4), 166.4
(CO). Anal. (C9H8O2BrI): C, H.
Compound 17 (from 13) could not be isolated from the
mixture and another method was used.
5.1.4. 2-Substitution of b-CD
5.1.5. 2-O-(3-Carboxymethyl-4-iodo)benzyl-b-cyclodextrin
(17)
To a solution of 1 g (0.88 mmol) of b-cyclodextrin (previ-
ously dried for 48 h under vacuum at 120 °C) in 40 ml of dry
DMF, were added 35 mg (0.88 mmol) of sodium hydride (60%
in mineral oil, in toluene and DMF). The solution was stirred
under nitrogen overnight. A solution of 320 mg (0.90 mmol)
of the corresponding methyl bromomethyl-2-iodobenzoate
(12–15) in 5 ml of dry DMF was then added drop wise in the
medium. The mixture was stirred for 9 additional h. Acetone
(500 ml) was added and the crude product precipitated. After
filtration, the solid residue was chromatographied on silica
gel; solvent: ethyl acetate/isopropanol/water (12:7:4; v/v/v).
Identification of by-products was realized by analysis of 13C-
NMR spectra of their mixture and their ratio was evaluated
by comparison of the integrations of their respective aro-
To a solution of 1 g (0.88 mmol) of b-cyclodextrin (previ-
ously dried for 48 h under vacuum at 120 °C) in 40 ml of dry
DMSO, were added 88 mg (2.20 mmol) of sodium hydrox-
ide. The solution was stirred under nitrogen for 48 h. A solu-
tion of 320 mg (0.90 mmol) of methyl 2-iodo-5-bromo-
methylbenzoate (13) in 5 ml of dry DMSO was then added
drop wise to the media. The mixture was stirred for 24 addi-
tional h. Acetone (500 ml) was added and the crude product
precipitated.After filtration, the solid residue was chromatog-
raphied on silica gel; solvent: ethyl acetate/isopropanol/
water (12:7:4; v/v/v). Compound 17 was obtained as a yel-
low solid (yield 16%); IR (nujol): m (cm–1) 3400, 3100, 1760;
1H-NMR (DMSO-d6): d 3.40–3.67 (m, 93H, 56H from H2O),
3.80 (s, 3H, OCH3), 4.51 (m, 7H, OH-6), 4.78 (bs, 7H, H-1),
6.17 (m, 13H, OH-2 and 3), 7.38 (dd, 1H, J = 8.0, 2.0 Hz,
H-6Ar), 7.76 (d, 1H, J = 2.0 Hz, H-2Ar), 7.99 (d, 1H,
J = 8.0 Hz, H-5Ar); 13C-NMR (DMSO-d6): d 52.4 (OCH3),
55.8 (OCH2), 59.7 (C-6), 71.5 (C-3′), 71.8 (C-5), 72.2 (C-2),
72.9 (C-3), 79.7 (C-2′), 81.3 (C-4), 81.9 (C-4′), 93.2 (C-I),
99.8 (C-1′), 101.8 (C-1), 129.4, 132.2, 135.3, 138.3, 140.5
(5× C-Ar), 166.7 (CO); MS ES+: 1431 (M + Na)+; MS ES–:
1407 (M – H)–, 1443 (M + Cl)–, 1445 (M + Cl)–. Anal.
(C51H77O37I): C, H.
1
matic signals on H-NMR spectra. The following com-
pounds were isolated.
2-O-(3-carboxymethyl-2-iodobenzyl)-b-cyclodextrin (16)
(from 12) was obtained as a yellow solid (yield 18%); IR
(nujol): m (cm–1) 3400, 3100, 1740; 1H-NMR (DMSO-d6): d
3.31–3.62 (m, 83H, 46H from H2O), 3.85 (s, 3H, OCH3),
4.50 (m, 7H, OH-6), 4.82 (bs, 7H, H-1), 6.02 (m, 13H,
OH-2 and 3), 7.23–8.00 (m, 3H, ArH); 13C-NMR (DMSO-
d6): d 52.3 (OCH3), 54.7 (OCH2), 59.7 (C-6), 70.1 (C-3′),
71.8 (C-5), 72.1 (C-2), 72.9 (C-3), 79.0 (C-2′), 81.5 (C-4),
81.7 (C-4′), 93.1 (C-I), 99.8 (C-1′), 101.7 (C-1), 129.4, 132.1,
135.3, 138.3, 140.4 (5× C-Ar), 166.6 (CO); MS ES+: 1431
(M + Na)+; MS ES–: 1407 (M – H)–, 1443 (M + Cl)–, 1445
(M + Cl)–. Anal. (C51H77O37I): C, H.
2-O-(2-carboxymethyl-3-iodobenzyl)-b-cyclodextrin (18)
(from 14) was obtained as a yellow solid (yield 16%); IR
(nujol): m (cm–1) 3400, 3100, 1735; 1H-NMR (DMSO-d6): d
3.31–3.45 (m, 67H, 30H from H2O), 3.87 (s, 3H, OCH3),
4.40 (m, 7H, OH-6), 4.64 (bs, 7H, H-1), 6.17 (m, 13H,
OH-2 and 3), 7.22–7.85 (m, 3H, ArH); 13C-NMR (DMSO-
d6): d 48.5 (OCH3), 52.8 (OCH2), 60.0 (C-6), 69.0 (C-3′),
71.8 (C-5), 72.5 (C-2), 73.2 (C-3), 79.0 (C-2′), 81.6 (C-4),
82.0 (C-4′), 93.3 (C-I), 100.0 (C-1′), 102.0 (C-1), 128.8, 131.4,
136.7, 138.3, 138.8 (5× C-Ar), 167.3 (CO); MS ES+: 1431
(M + Na)+; MS ES–: 1407 (M – H)–, 1443 (M + Cl)–, 1445
(M + Cl)–. Anal. (C51H77O37I): C, H.
5.1.6. 2-O-(Carboxy-iodosobenzyl)-b-cyclodextrin (20–23)
Sixteen milliliters (204 mmol) of peracetic acid were added
over a period of 1 h to a solution of 300 mg (0.20 mmol) of
the corresponding 2-O-(carboxymethyl-iodobenzyl)-b-
cyclodextrin (16–18 or 19) in 3 ml of DMF and 20 ml of
glacial acetic acid. The solution was stirred at room tempera-
ture for 24 h.Acetone (500 ml) was added and the crude prod-
uct precipitated. The solution was filtered, and the precipitate
was washed with acetone (20 ml), then with diethyl ether
(20 ml).
2-O-(3-carboxy-2-iodosobenzyl)-b-cyclodextrin (20) was
obtained as a white solid (yield 78%); IR (nujol): m (cm–1)
1
3500, 3250, 1770; H-NMR (DMSO-d6): d 3.35–3.85 (m,
85 H, 48H from H2O), 4.49 (m, 7H, OH-6), 4.82 (bs, 7H,
H-1), 5.73 (m, 13H, OH-2 and 3), 7.46–7.81 (m, 3H, H-Ar);
13C-NMR (DMSO-d6): d 52.7 (OCH2), 60.0 (C-6), 69.0
(C-3′), 72.1 (C-5), 72.5 (C-3), 73.4 (C-2), 73.1 (C-2′), 81.6
(C-4), 102.0 (C-1′), 102.6 (C-1), 119.3 (C-I), 128.3, 130.8,
131.8, 134.2, 138.9 (5× C-Ar), 168.5 (CO).
2-O-(4-carboxymethyl-3-iodo)benzyl-b-cyclodextrin (19)
(from 15) was obtained as a yellow solid (yield 31%); IR
(nujol): m (cm–1) 3450, 3150, 1730; 1H-NMR (DMSO-d6): d
3.43–3.53 (m, 103H, 66H from H2O), 3.80 (s, 3H, OCH3),

622
N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
2-O-(3-carboxy-4-iodosobenzyl)-b-cyclodextrin (21) was
References
obtained as a white solid (yield 61%); IR (nujol): m (cm–1)
1
[1] T.C. Mars, R.L. Maynard, F.R. Sidell, Chemical warfare agents:
toxicology and treatment, Wiley, Chichester, 1996.
[2] C. Ballantyne, T.C. Mars, Clinical and experimental toxicology of
organo-phosphates and carbamates, Butterworth-Heinemann,
Oxford, UK, 1992.
[3] L. Leadbeater, R.H. Inns, J.M. Rylands, Toxicol. Appl. Pharmacol. 5
(1985) S225–S231.
[4] E.H. Braue, S.T. Hobson, C. Govardhan, US Patent Number
6,410,604; June 25, 2002.
[5] Y. Ashani, S. Pistinner, Toxicol. Sci. 77 (2004) 358–367.
[6] M.A. Sogorb, E. Vilanova, Toxicol. Lett. 128 (2002) 215–228.
[7] P. Masson, D. Josse, O. Lockridge, N. Viguié, C. Taupin, C. Buhler, J.
Physiol. (Paris) 92 (1998) 357–362.
[8] F.M. Raushel, Curr. Opin. Microbiol. 5 (2002) 288–295.
[9] R. Breslow, S.D. Dong, Chem. Rev. 98 (1998) 1997–2011.
[10] E. Rizzarelli, G. Vecchio, Coord. Chem. Rev. 188 (1999) 343–364.
[11] E. Engeldinger, D. Armspach, D. Matt, Chem. Rev. 103 (2003) 4147–
4174.
[12] M. Komiyama, S. Inoue, Bull. Chem. Soc. Jpn. 53 (1980) 3266–3269.
[13] M. Komiyama, S. Inoue, Chem. Lett. (Jpn.) 9 (1979) 1101–1104.
[14] T.A. Gadosy, O.S. Tee, Can. J. Chem. 74 (1996) 745–752.
[15] O.S. Tee, M. Bozzi, N. Clement, T.A. Gadosy, J. Org. Chem. 60
(1995) 3509–3517.
3550, 3300, 1770; H-NMR (DMSO-d6): d 3.34–3.85 (m,
85H, 48H from H2O), 4.49 (m, 7H, OH-6), 4.82 (bs, 7H, H-1),
5.71 (m, 13H, OH-2 and 3), 7.83–8.04 (m, 3H, H-Ar); 13C-
NMR (DMSO-d6): d 58.2 (OCH2), 60.0 (C-6), 71.6 (C-5),
73.4 (C-3), 73.5 (C-2), 80.8 (C-2′), 82.4 (C-4), 82.8 (C-4′),
99.2 (C-1′), 101.1 (C-1), 121.0 (C-I), 127.1, 131.2, 131.4,
140.0, 145.4 (5× C-Ar), 159.8 (CO).
2-O-(2-carboxy-3-iodosobenzyl)-b-cyclodextrin (22) was
obtained as a white solid (yield 65%); IR (nujol): m (cm–1)
1
3500, 3200, 1770; H-NMR (DMSO-d6): d 3.38–3.61 (m,
81H, 44H from H2O), 4.51 (m, 7H, OH-6), 4.85 (bs, 7H, H-1),
5.77 (m, 13H, OH-2 and 3), 7.20–7.97 (m, 3H, H-Ar); 13C-
NMR (DMSO-d6): d 53.1 (OCH2), 60.2 (C-6), 71.1 (C-3′),
72.4 (C-5), 72.8 (C-3), 73.4 (C-2), 81.8 (C-4), 100.3 (C-1′),
102.2 (C-1), 122.5 (C-I), 130.0, 131.7, 137.0, 138.6, 139.7
(5× C-Ar), 167.0 (CO).
2-O-(4-carboxy-3-iodosobenzyl)-b-cyclodextrin (23) was
obtained as a white solid (yield 81%) IR (nujol): m (cm–1)
1
3400, 3100, 1730; H-NMR (DMSO-d6): d 3.39–3.70 (m,
[16] O.S. Tee, T.A. Gasoy, J. Chem. Soc., Perkin Trans. 2 11 (1994)
2307–2311.
[17] T.A. Gadosy, O.S. Tee, J. Chem. Soc., Perkin Trans. 2 1 (1995) 71–76.
[18] A.K. Yatsimirsky, K.Y. Bezoudnova, Russ. Chem. Bull. 45 (1996)
2517–2524.
[19] F. Cramer, W. Kampe, J. Am. Chem. Soc. 87 (1965) 1115–1120.
[20] A.C. Hengge, W.W. Cleland, J. Org. Chem. 56 (1991) 1972–1974.
[21] R. Breslow, M.F. Czarniecki, J. Emert, H. Hamaguchi, J. Am. Chem.
Soc. 102 (1980) 762–770.
77H, 40H from H2O), 4.57 (m, 7H, OH-6), 4.90 (bs, 7H, H-1),
5.75 (m, 13H, OH-2 and 3), 7.61–8.1 (m, 3H, H-Ar); 13C-
NMR (DMSO-d6): d 48.2 (OCH2), 60.2 (C-6), 72.3 (C-5),
72.6 (C-3), 73.3 (C-2), 79.6 (C-2′), 80.3 (C-4), 82.8 (C-4′),
100.2 (C-1′), 102.2 (C-1), 121.0 (C-I), 127.1, 131.2, 131.4,
140.0, 145.4 (5× C-Ar), 168.0 (CO).
[22] J. Emert, R. Breslow, J. Am. Chem. Soc. 97 (1975) 670–672.
[23] H. Ikeda, R. Kojin, C.J. Yoon, T. Ikeda, F. Toda, J. Inclusion Phenom.
Mol. Recognit. Chem. 7 (1989) 117–124.
5.2. OP hydrolytic activity of b-CD derivatives
[24] K.R. Rao, T.N. Srinivasan, N. Bhanumathi, P.B. Sattur, J. Chem. Soc.
Chem. Commun. 1 (1990) 10–11.
[25] D. Yuan, K. Ohta, K. Fujita, J. Chem. Soc. Chem. Commun. 7 (1996)
821–822.
[26] B. Desiré, S. Saint-André, Experientia 43 (1987) 395–397.
[27] C. Van Hooidonk, J.C.A.E. Breebaart-Hansen, Recl. Trav. Chim. Pays
Bas 89 (1970) 289–299.
[28] J. Cabal, Collect. Czech. Chem. Commun. 60 (1995) 1162–1169.
[29] C. Van Hooidonk, C. Groos, Recl. Trav. Chim. Pays Bas 89 (1970)
845–846.
[30] M. Kamiya, S. Mitsuhashi, M. Makino, Chemosphere 25 (1992)
1783–1796.
[31] B. Désiré, S. Saint-André, Fundam. Appl. Toxicol. 7 (1986) 646–657.
[32] S. Saint-André, B. Désiré, C.R. Acad. Sci. Paris t. 301 (3) (1985)
67–72 Serie III.
[33] S. Ishiwata, M. Kamiya, Chemosphere 41 (2000) 701–704.
[34] R.A. Moss, K.W. Alwis, G.O. Bizzigotti, J. Am. Chem. Soc. 105
(1983) 681–682.
[35] R.A. Moss, K.W. Alwis, J.S. Shin, J. Am. Chem. Soc. 106 (1984)
2651–2655.
Paraoxon was the organophosphate model: its stock solu-
tion was 100 mM in anhydrous methanol. Stock solutions of
b-CD derivatives (5 mM) were in a mixture of 10% v/v DMSO
in 20 mM sodium phosphate buffer pH 7.65. Kinetic assays
were carried out with paraoxon (3, 1, 0.5 and 0.1 mM) at
25 °C, in 20 mM phosphate buffer pH 7.65 in the presence of
13 mM cetyltrimethylammonium chloride (CTAC) [34]. The
final concentration of methanol in assays was 3% v/v and
that of DMSO was 2% v/v. The final concentration of b-CD
derivatives in assays was 1 mM. Hydrolysis of paraoxon was
monitored up to 120 min by following the release of the leav-
ing group para-nitrophenol (k = 400 nm; double beam spec-
trophotometer). The control cuvette contained all assay com-
ponents except the b-CD derivative.
[36] R.A. Moss, K.Y. Kim, S. Swarup, J. Am. Chem. Soc. 108 (1986)
788–793.
[37] R.A. Moss, S. Chatterjee, B. Wilk, J. Org. Chem. 51 (1986) 4303–
4307.
Acknowledgements
[38] R.A. Moss, Y.C. Chung, J. Org. Chem. 55 (1990) 2064–2069.
[39] P.S. Hammond, J.S. Forster, C.N. Lieske, H.D. Durst, J. Am. Chem.
Soc. 111 (1989) 7860–7866.
[40] R. Breslow, A.W. Czarnik, J. Am. Chem. Soc. 105 (1983) 1390–1391.
[41] H.H. Seltzman, M.S. Lonikar, In Proceedings of the 1993 US Army
Medical Defense, Bioscience Review 3 (1993) 1075–1083.
Supported by DGA: a) DGA/CNRS fellowship to N.M. b)
DSP/STTC contract PEA010807 to F.E./O.L. and P.M. The
authors thank Florence Arpaillange for excellent technical
assistance.

N. Masurier et al. / European Journal of Medicinal Chemistry 40 (2005) 615–623
623
[42] A.R. Katritzky, G.P. Savage, J.K. Gallos, Org. Prep. Proc. Int. 21
(1989) 157–162.
[47] Y. Inoue, T. Okuda, Y. Miyata, R. Chujo, Carbohydr. Res. 125 (1984)
65–76.
[43] W. Carruthers, R. Pooranamoorthy, J. Chem. Soc., Perkin Trans. 1 20
(1974) 2405–2409.
[48] S. Kýsl, E. Smolková-Keulemansová, J. Chromatogr. A. 349 (1985)
167–172.
[44] D. Rong, V.T. D’Souza, Tetrahedron Lett. 31 (1990) 4275–4278.
[45] M.A. Mortellaro, W.K. Hartmann, D.G. Nocera, Angew. Chem. Int.
Ed. Engl. 35 (1996) 1945–1946.
[49] A. Buvári, L. Barcza, J. Chem. Soc., Perkin Trans. 2 4 (1988) 543–
545.
[50] S. Kanoh, H. Muramoto, N. Kobayashi, M. Motoi, H. Suda, Bull.
Chem. Soc. Jpn. 60 (1987) 3659–3662.
[46] F.M. Menger, M.A. Dulany, Tetrahedron Lett. 26 (1985) 267–270.