Disulfide prodrugs of albitiazolium (T3/SAR97276): Synthesis and biological activities

Article abstract of DOI:10.1021/jm3000328

We report herein the design, synthesis, and biological screening of a series of 15 disulfide prodrugs as precursors of albitiazolium bromide (T3/SAR97276, compound 1), a choline analogue which is currently being evaluated in clinical trials (phase II) for severe malaria. The corresponding prodrugs are expected to revert back to the active bis-thiazolium salt through an enzymatic reduction of the disulfide bond. To enhance aqueous solubility of these prodrugs, an amino acid residue (valine or lysine) or a phosphate group was introduced on the thiazolium side chain. Most of the novel derivatives exhibited potent in vitro antimalarial activity against P. falciparum. After oral administration, the cyclic disulfide prodrug 8 showed the best improvement of oral efficacy in comparison to the parent drug.

Full text of DOI:10.1021/jm3000328

Article
pubs.acs.org/jmc
Disulfide Prodrugs of Albitiazolium (T3/SAR97276): Synthesis and
Biological Activities
Sergio A. Caldarelli,†,⊥ Matthieu Hamel,†,⊥ Jean-Frederic Duckert,† Mahama Ouattara,† Michele Calas,†
́
́
̀
Marjorie Maynadier,‡ Sharon Wein,‡ Christian Perigaud,† Alain Pellet,§ Henri J. Vial,*,‡
́
and Suzanne Peyrottes*,†
†Institut des Biomolec
Bataillon, 34095 Montpellier, France
́ ́
ules Max Mousseron (IBMM), UMR 5247 CNRS-UM1&2, Universite Montpellier 2, cc 1705, place E.
‡Dynamique des Interactions Membranaires Normales et Pathologiques (DIMNP), UMR 5235 CNRS-UM2, Universite
2, cc 107, place E. Bataillon, 34095 Montpellier, France
§Sanofi Research & Development, 195 route d’Espagne, BP 13669, 31036 Toulouse, France
́
Montpellier
S
* Supporting Information
ABSTRACT: We report herein the design, synthesis, and biological screening of a series of 15
disulfide prodrugs as precursors of albitiazolium bromide (T3/SAR97276, compound 1), a choline
analogue which is currently being evaluated in clinical trials (phase II) for severe malaria. The
corresponding prodrugs are expected to revert back to the active bis-thiazolium salt through an
enzymatic reduction of the disulfide bond. To enhance aqueous solubility of these prodrugs, an
amino acid residue (valine or lysine) or a phosphate group was introduced on the thiazolium side
chain. Most of the novel derivatives exhibited potent in vitro antimalarial activity against P.
falciparum. After oral administration, the cyclic disulfide prodrug 8 showed the best improvement of
oral efficacy in comparison to the parent drug.
Potency and specificity of these antiphospholipid effectors are
likely due to their unique property to accumulate in a
nonreversible way inside the intraerythrocytic parasite.11,13,14
The efficiency of albitiazolium comes from its dual mechanism
of action that involves, on one hand, the inhibition of the de
novo phosphatidylcholine biosynthesis11 and, on the other
hand, an interaction with the ferriprotoporphyrin IX (FPIX)
which leads to heme detoxification.13
Currently, albitiazolium is undergoing phase II clinical trials
to treat severe malaria by parenteral administration due to its
poor oral bioavailability. Most of the infections by malaria
parasites occur in tropical or subtropical countries where the
medical care systems are not always available. Consequently,
the way of administration of albitiazolium limits its therapeutic
use to the treatment of severe malaria and highlights the need
for an oral form to treat uncomplicated malaria on a large scale.
Owing to the presence of two cationic charges, bis-thiazolium
derivatives have greater difficulty crossing biological barriers,
especially the intestinal epithelium. Thus, we devoted our
recent efforts to the design of S-acyl prodrug approaches to
temporally mask the cationic charges, and it was anticipated
that the resulting lipophilic albitiazolium prodrugs would then
be able to cross the intestinal epithelium by passive
diffusion.11,15 So far, the best absolute bioavailability obtained
for a thioester-type prodrug was 15% in rat,16 and this modest
improvement was attributed to the early conversion of the
INTRODUCTION
■
Malaria is the most prevalent parasite disease, causing each year
250 million cases and 800 000 deaths, mostly in African
children, and 80% of these cases are located in sub-Saharan
Africa.1,2 Among the key interventions for controlling this
disease, the arsenal of antimalarial drugs is critical, but the
current choice of drugs is limited.3 The discovery and
development of the artemisinin derivatives in China have
provided a new class of highly effective antimalarials now used
as artemisinin-based combination therapy (ACTs) to overcome
the chemoresistance problem. However, artemisinin-resistant
parasites recently reported in Asia could seriously undermine
global malaria control.4,5 Plasmodium falciparum, the most
pathogenic human malaria parasite, is becoming pharmacor-
esistant to conventional as well as newly discovered drugs; thus,
the need for new antimalarial strategies involving novel targets
is as crucial as ever.6 Thus, various research groups are
developing a new family of derivatives differing in their
mechanisms of action.3,7 A decade ago, some of us contributed
to this challenge with a novel class of choline analogues. The
structure of these potent antimalarials is based on a long
lipophilic chain incorporating two thiazolium cationic
heads.8−12 One lead compound, namely albitiazolium bromide
(T3/SAR97276, compound 1, Figure 1) shows high efficacy in
vitro against P. falciparum and in vivo against P. vinckei in
mouse and primate malaria models8,10,11 and has fulfilled
multiple criteria required for its development. Available in a
single-step synthesis from commercial reactants, its preparation
is therefore adapted for large-scale and low-cost production.
Received: January 17, 2012
Published: May 16, 2012
© 2012 American Chemical Society
4619
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
Figure 1. Structures of albitiazolium (T3/SAR97276, compound 1) and related disulfide prodrugs.
a
Scheme 1. Synthesis of Disulfide Prodrugs 2a−d and 3
a
Reagents and conditions: (i) 2 M NaOH, sodium alkyl thiosulfate, CHCl3, 1−2 h; (ii) 2 M NaOH, tert-butyl nitrite, CH2Cl2, 10 h.
prodrug into the drug, occurring in the gastrointestinal tract
before absorption.
RESULTS AND DISCUSSION
■
Chemistry. Disulfide Prodrugs of Albitiazolium. Lipophilic
prodrugs of albitiazolium were obtained following a common
route used for the synthesis of mixed disulfides and based on
Herein, we described a new series of albitiazolium disulfide
prodrugs related to the approach previously developed for
thiamine (vitamin B1) and which led to orally bioavailable
precursors.17,18 Briefly, it consists of the thiazolium ring-
opening and concomitant trapping of the resulting thiol
function as a disulfide linkage (Figure 1). The release of the
parent drug could involve either an enzymatic reduction of the
disulfide bond (for which no evidence is presently available) or
a thiol−disulfide exchange reaction.11
In addition, to improve the physicochemical properties of the
albitiazolium disulfide prodrugs and especially their aqueous
solubility, we introduced various polar residues on the hydroxyl
function of the side chain of albitiazolium. In this respect,
aminoacyl or phosphate moieties were selected, as they have
been extensively used for this purpose.19 Thus, the resulting
pro-prodrugs were expected to be converted in vivo in two
steps to the active drug. First, the amino acid or phosphate pro-
moieties should be hydrolyzed through esterase or alkaline
phosphatase activities present in abundance on the brush
border surface of the gastrointestinal tract. The resulting
lipophilic disulfide prodrug generated nearby is expected to
cross the intestinal epithelium more easily than the parent bis-
cationic drug. This kind of approach has already proved its
efficiency20,21 and may help us to circumvent the low
bioavailability of albitiazolium after oral administration.
the reaction of a Bunte salt (previously prepared from an alkyl
̈
halide and sodium thiosulfate) with a mercaptan. The
procedure involved a two-step one-pot reaction (Scheme 1):
the thiazolium ring is opened in alkaline media, according to
the methodology previously reported for the synthesis of
thioester prodrugs of bis-thiazolium salts;15 then addition of the
sodium alkyl thiosulfate (Bunte salt) led to the desired disulfide
̈
prodrugs 2a−d in high yields. The different promoieties were
chosen to vary the steric and electronic environment in the
vicinity of the biolabile linkage, as well as the relative
lipophilicity of the resulting prodrugs.
In addition to minimizing the molecular weight and limiting
the molecular flexibility, we envisaged the intramolecular
cyclization of compound 1, bridging together the two thiol
functions of the molecule. This reaction is based on the
peculiarity of thionitrite which spontaneously decomposes in
solution into the corresponding disulfide and nitric oxide.22
Thus, the cyclic disulfide prodrug 3 was prepared in a single
step (Scheme 1): after thiazolium ring-opening in alkaline
media, the thiolate intermediate was activated by means of tert-
butyl nitrite, and in situ the thionitrite led to the formation of
intra- and intermolecular disulfide bonds. To limit the
polymerization of the thionitrite intermediate, the reaction
was carried under high dilution conditions (10 mM) and
compound 3 was isolated in 25% yield.
4620
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Aminoacyl 6a,b and 7a,b and Acetyl 8 Disulfide Prodrugs
a
Reagents and conditions: (i) Boc-Val or (Boc)2-Lys, DCC or EDC, DMAP or acetyl chloride, CH2Cl2, 2−3 h; (ii) HCl (4 M), 1,4-dioxane, 1 h.
a
Scheme 3. Side Reaction Observed during the Synthesis of Monophosphate Prodrug of 1
a
Reagents and conditions: (i) POCl3, (EtO)3PO, 15 h; then TEAB, 0 °C; (ii) NaOH, H2O/CH2Cl2, 0 °C, 30 min.
a
Scheme 4. Synthesis of Phosphorylated Disulfide Prodrugs 13b,d
a
Reagents and conditions: (i) Diethyl chlorophosphate, NEt3, DMAP (cat.), 1 h; (ii) POCl3, NEt3, CH2Cl2, −60 °C to room temperature, 1 h; (iii)
NEt3, DMAP, CH2Cl2, −60 °C; (iv) LiOH, CH3CN/H2O (1:1 v/v), 24 h.
Aminoacyl Disulfide Prodrugs. Between the different
disulfide prodrugs previously obtained, we selected the ones
with the lower molecular weights to prepare the pro-prodrugs,
i.e., compounds 2a and 3, respectively. Thus, the aminoacyl-
prodrugs were synthesized (Scheme 2) by coupling Boc-Val or
(Boc)2-Lys in the presence of DCC or EDC and a catalytic
amount of DMAP. Then removal of the Boc protecting groups
from intermediates 4a,b and 5a,b afforded compounds 6a,b and
7a,b as hydrochloride salts, respectively.
Finally, an O-acylated prodrug (compound 8, Scheme 2) was
prepared by treatment of derivative 3 with acetyl chloride. This
last prodrug was designed to evaluate the effect of masking the
two hydroxyl functions of prodrug 3 with the smallest (size and
molecular weight) esterase labile promoiety.
Phosphorylated Disulfide Prodrugs. To obtain the disulfide
prodrugs bearing a phosphate moiety, we first adapted the
synthetic pathway already described in the literature for the
preparation of benfotiamine (S-benzoylthiamine O-mono-
phosphate).23 Thus, preparation of the O-phosphorylated
4621
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
a
Table 1. In Vitro and in Vivo Antimalarial Activities and Oral Absorption Index after Administration in the Mouse
a
IC50 is the concentration to inhibit by 50% in vitro P. falciparum growth, ED50 is the efficient dose to inhibit by 50% in vivo P. vinckei growth
b
according to a 4-day suppressive test, ip means intraperitoneal administration, and po means per-os administration. clogP values were calculated
with Chemdraw. Compounds were administered in vivo in DMSO for both ip and po routes except for compounds noted (footnotes c and d).
cAdministered in 0.9% NaCl and water, for ip and po routes, respectively. dAdministered in a mixture of intralipid and DMSO, for ip and po routes,
e
respectively. The ip/po ratio is generally accepted as a raw indicator of oral bioavailability.
derivative of albitiazolium as an intermediate was first required
and followed by the ring-opening of the thiazolium to further
generate the disulfide in situ (Scheme 3).
At first we chose to investigate the diethylphosphoryl
protecting group, which is one of the most widely used for
phosphate protection and is usually cleaved with bromotrime-
thylsilane (Me3SiBr).27,28 Thus, the disulfide prodrug 2b was
converted to its phosphorylated analogue 11, by reaction of
diethyl chlorophosphate in the presence of triethylamine
(TEA) and a catalytic amount of DMAP (Scheme 4).
Unfortunately, all attempts to convert the phosphate diester
11 to its free acid using Me3SiBr failed. Therefore, we turned
our attention to the “CycloSal” protecting group, derived from
salicylic acid, which is described to be very labile when exposed
to smooth basic conditions and thereby could be eliminated
under soft conditions.29−31 To introduce the CycloSal group on
our disulfide derivatives, we used a chlorophosphate strategy
derived from the method of Casida et al.32,33 Our various assays
showed that without DMAP (at least 1 equiv), only the starting
material and the monophosphorylated product were recovered
from the reaction mixture. A large excess of the chlorophos-
phate of CycloSal was necessary to observe the conversion of
the disulfide prodrug to its phosphorylated analogue 12b or
12d. The first attempt to remove the CycloSal protecting group
from compound 12b was carried out following the conditions
described by Meier’s group and using a mixture of TEA/water
in acetonitrile.34 After 48 h at room temperature, the reaction
did not go further. Partially deprotected intermediates were
isolated and their structure assigned by 1H and 31P NMR (data
not shown). The selectivity of this reaction (P−O or C−O
bond cleavage) was rather poor, and increasing the reaction
time or the amount of TEA did not allow shifting the reaction
The two hydroxyl groups of starting material 1 were
phosphorylated using the Yoshikawa method.24,25 Thus,
albitiazolium was reacted with an excess of POCl3, leading to
a bis-phosphorodichloridate intermediate which was hydrolyzed
in situ with 1 M TEAB. The bis-phosphate intermediate 9
(Scheme 3) was isolated by reverse phase chromatography with
moderate to good yields. Unfortunately, attempts to obtain the
desired phosphorylated disulfide prodrugs by treatment of 9
(either one-pot reaction or after isolation) with sodium
hydroxide and rapid addition of an alkyl thiosulfate salt were
not successful. The sole product identified was the cyclic
derivative, bisthietane 10 (Scheme 3). We hypothesized that
once the thiolate is generated, the intramolecular nucleophilic
attack onto the phosphorylated side chain occurs rapidly, and
preferentially, compared to the disulfide bond formation.
Indeed, a test reaction was performed (phosphorylation and
subsequent addition of sodium hydroxide) in the absence of
any electrophile to trap the thiolate intermediate. Completion
of the reaction was observed after 30 min and led to the
isolation of derivative 10 in 73% yield.
Consequently, we envisaged the two previous steps in the
reverse order, i.e., the introduction of the phosphate group on
the preformed disulfide prodrug (Scheme 4). This method-
ology has been reported for the di- and triphosphorylation of
thiamine and one corresponding disulfide derivative.26
4622
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
in favor of the final compound 13b. Under microwave
irradiation, 13b was finally obtained in 35% yield. To optimize
the deprotection step, we tested various bases such as lithium
hydroxide, and compound 13b was then isolated with 75%
yield. When the same procedure was applied to protected S-
benzyl prodrug 12d, lower yields were observed (12%).
All the disulfide prodrugs and their aminoacyl and phosphate
conjugates were fully characterized by 1H NMR, eventually 31P
and 13C NMR, and HR-MS and then evaluated for their in vitro
and in vivo antimalarial activity. Note that the ring-opening
reaction of thiazolium ions under basic conditions, followed by
in situ trapping of the amidoenothiolates, leads to the formation
of a pair of amide rotamers. These two rotational isomers are
related by a slow rotation about the C−N bond of the amide
group and can be readily observed by NMR spectroscopy. This
phenomenom gave rise to broadening and/or duplication of the
formamide signal as well as the signals of vicinal groups.
Consequently, for all our prodrugs, two characteristic signals for
the N-formyl function (in the 7.5−8.0 ppm range) and for the
methyl group (∼2.0 ppm) appeared and were attributed to the
major and the minor conformers.
In the second set of compounds, we introduced amino acid
residues on the side chain of the thiazolium to improve aqueous
solubility of the resulting pro-prodrugs. Thus, the IC50 values of
corresponding aminoacyl disulfide prodrugs 6a,b and 7a,b
ranged from 120 to 590 nM. In that case, the polarity of the
second pro-moiety (amino acid) might delay the bioconversion
of the pro-prodrugs to the corresponding parent prodrug 2a
and then to albitiazolium. This postponed bioconversion might
be advantageous for oral absorption.
We also designed and synthesized O-phosphorylated
derivatives 9, 13b, and 13d, which led to much contrasted
results. Only, the bis-O-phosphorylated drug 9 (IC50 = 6.6 nM)
exhibited activity similar to that of the parent drug 1 (IC50
=
2.25 nM), proving that the phosphate group is efficiently
hydrolyzed in biological media to release the active parent drug.
However, phosphorylated disulfide prodrugs 13b and 13d
(which involved two enzymatic conversion steps presumably
mediated by alkaline phosphatase and reductase) showed very
low activity in vitro with an IC50 of 5.7 and 4.4 μM,
respectively. We hypothesized that in the presence of the
highly polar phosphate groups, the conversion of the pro-
prodrugs into the parent prodrugs 2b or 2d occurred very
slowly. Indeed, related prodrugs 2b and 2d exhibited an IC50 of
5.4 and 14.4 nM, respectively, indicating that in the absence of
the phosphate group the reduction of the disulfide bond
proceeded rapidly. Phosphorylated prodrug intermediates 11,
12b, and 12d were also screened and showed a modest to low
ability to inhibit P. falciparum growth (60 nM < IC50 < 900
nM), thus indicating that the disulfide promoieties were
hydrolyzed (more rapidly than for compounds 13b and 13d)
to generate the corresponding bis-thiazolium heads. However,
we have no evidence that the diethyl or the CycloSal
phosphoester groups were removed (or not) from the molecule
either through a chemical and/or an enzymatic process.
B. In Vivo Antimalarial Activity of Disulfide Prodrugs. All
derivatives were also evaluated in vivo against P. vinckei in mice
by both intraperitoneal (i.p.) and oral (p.o.) routes (Table 1).
When administered intraperitoneally once daily for 4 days,
most of the compounds exhibited very potent curative activity
against the pathogenic parasite. Efficient doses ED50 ranged
from 0.41 to 5 mg/kg, which are close to that of albitiazolium
(ED50 of 0.2 mg/kg) and comparable to that of chloroquine
(ED50 of 1.1 mg/kg). Consequently, both the disulfide
prodrugs (2a−d, 3) and related pro-prodrugs (6a and 6b, 7a
and 7b, 8, 12b, 13b, and 13d) are readily bioconverted. The
ED50 values for compounds 2a−d, 3, 8, and 9 are in agreement
with their potent in vitro antimalarial activity against P.
falciparum.
Surprisingly, the aminoacyl and phosphorylated disulfide pro-
prodrugs 6a and 6b, 7a and 7b, 12b, 13b, and 13d, which were
significantly less active in vitro with an IC50 ranging from 120 to
5700 nM, appeared as potent as the parent prodrugs with an
ED50 of 0.9 to 5 mg/kg. This observation may be attributed to
the higher and wide enzymatic activity present in animal
models than in the culture medium and corroborated the
hypothesis made above that the kinetic of the decomposition of
these pro-prodrugs was very slow in the culture model. Overall,
these results point out that the pro-prodrug approaches
envisaged are valuable ones. In addition, it is important to
note that, as expected, an improvement in the aqueous
solubility of the related compounds was observed. As an
example, aminoacyl derivatives 6b and 7b could be
administered as aqueous solutions.
We also calculated the logP value for final derivatives in order
to estimate their lipophilicity and compared compounds to
each other within the series (Table 1). Data corresponding to
chloroquine were introduced as a reference.
Biological Activities. A. In Vitro Studies of Disulfide
Prodrugs. The antimalarial activities of the compounds were
first evaluated in vitro against the blood stage of P. falciparum
(Table 1). The activities were determined using [3H]-
hypoxanthine incorporation to assess parasite growth after
contact of the compounds with the parasite for one parasite
cycle (48 h). Parasitic viability was expressed as IC50, the drug
concentration causing 50% parasite growth inhibition. IC50
values presented in Table 1 indicate that disulfide prodrugs
(2a−d, 3, and 8) have potent antimalarial activity against P.
falciparum with an IC50 in the very low nanomolar range. As the
prodrug scaffold is devoid of antimalarial activity,11 this attests
that all compounds were converted into the parent drug and
behave as prodrugs of albitiazolium. Remarkably, the linear
disulfide prodrugs 2a−c showed an IC50 lower than 10 nM, and
compound 2c (IC50 = 1.3 nM) was more potent than the
parent drug 1 (IC50 = 2.2 nM). The presence of the n-propyl
disulfide promoiety leads to a highly lipophilic derivative (clogP
= 7.4) which can easily penetrate into infected erythrocytes and
thereby give rise, after reduction of the disulfide bond, to a
higher intracellular concentration of the drug. The prodrug 2c
(R1 = n-propyl, IC50 =1.3 nM) exhibited a substantially
increased activity compared to 2d (R1 = benzyl, IC50 = 14.4
nM) despite a similar clogP, which may be attributed to an
increase in the enzymatic stability of the S-benzyl promoiety
(owing to steric hindrance and/or electronic effect) compared
to the n-propyl promoiety.
The cyclic disulfide prodrugs 3 and 8 showed an IC50 on the
same order of magnitude. However, the significantly lower IC50
value of the cyclic acetylated prodrug 8 (24 nM) in comparison
to 3 (IC50 = 65 nM) was unexpected, taking into account the
double conversion step (disulfide reduction and ester
hydrolysis) to give albitiazolium. This result suggests that, in
this case, the enzymatic conversion of the pro-prodrugs is rapid
and is not a limiting step, and that the pro-prodrugs might have
a few advantages to reach the targets (or easier access to the
intraparasitic target).
4623
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
levels were monitored at day 5 after four days of treatment.11 Results
were expressed as the dose inhibiting 50% of parasitemia (ED50,
efficient dose 50).
To further study the ability of our pro-prodrug approach to
improve the oral activity of albitiazolium, we orally adminis-
tered all the compounds studied to P. vinckei-infected mice.
Linear disulfide prodrugs (2a−d) bearing various lipophilic
moieties showed a good correlation between oral antimalarial
activity and their clogP, which was used for comparison of
lipophilicity within our prodrug series. Thus, compound 2a,
with a relatively low clogP value (5.3), was the more active
compound after oral administration with an efficient dose of 15
mg/kg close to that of the parent drug (1, ED50(po) = 13 mg/
kg), meaning that the prodrug is quickly converted in the
gastrointestinal tract. The cyclic prodrug 6 was the less
lipophilic prodrug in this series (clogP = 4.4) but still showed
interesting activity, with an ED50(po) = 22 mg/kg. Increasing
the lipophilicity of this cyclic derivative by acetylation of the
hydroxyl groups (compound 8, clogP = 6.2) led to a remarkable
enhancement of the oral activity of the prodrug (ED50(po) =
7.3 mg/kg). This result illustrates, for the first time in our series
of bis-thiazolium prodrugs, that the global shape of the
molecule also plays a crucial role in the absorption
phenomenon. One could easily imagine that the intramolecular
cyclization of the drug is “freezing” the 3D-structure of the
molecule and decreasing the number of the rotatable bounds in
comparison to the linear prodrugs.
Chemistry. General Information. All moisture-sensitive reactions
were carried out under rigorous anhydrous conditions and argon
atmosphere and using oven-dried glassware. Solvents were dried and
distilled prior to use, and solids were dried over P2O5 under reduced
1
pressure. H, 13C, and 31P NMR spectra were recorded at ambient
temperature on a Bruker 250 or 300 Avance. Chemical shifts (δ) are
quoted in parts per million (ppm) referenced to the residual solvent
peak (CDCl3 fixed at 7.26 ppm and 77 ppm, DMSO-d6 fixed at 2.49
ppm and 39.5 ppm) relative to tetramethylsilane. Deuterium exchange
and 2D-COSY experiments were performed in order to confirm
proton assignments. Coupling constants, J, are reported in hertz. ESI
mass and high resolution mass spectra were recorded in the positive or
negative-ion mode on a Micromass Q-TOF. Chromatography was
performed on Merck silica gel 60 (230−400 mesh ASTM). Analytical
HPLC traces were obtained using a Waters HPLC system (Separation
module 2695, 996 Photodiode Array Detector 2996) and a Waters
Symmetry Shield (50 × 4.6 mm, 3.5 μm) RP-18-column, with a 1 mL/
min flow rate The elution solvents were water containing 0.1% (v/v)
of TFA (solvent A) and acetonitrile containing 0.1% (v/v) TFA
(solvent B). A linear gradient was performed from 100% of solvent A
to 100% of solvent B over 15 min. Purity was determined by HPLC,
and most of the tested compounds were confirmed to have ≥95%
purity.
̈
Preparation of Alkyl Thiosulfate (Bunte salts). The appropriate
As the clogP and molecular weight of all disulfide prodrugs
were higher than the usually suggested cutoff for orally druglike
compounds, we also explored the effect of decreasing their
lipophilicity by appending amino acids and phosphate groups as
polar moieties. Among the disulfide pro-prodrugs incorporating
L-valine (6a and 7a), L-lysine (6b and 7b), and phosphate
groups (13b and 13d), only the L-lysine disulfide prodrugs were
soluble in water at the test concentration condition (27 mg/
mL). Unfortunately, their pharmacological activities were lower
than that of albitiazolium, meaning that improvement of the
aqueous solubility for the studied disulfide prodrugs has no
significant benefits.
alkyl halide (1 equiv) was dissolved in ethanol (0.5 mL/mmol of alkyl
halide), and sodium thiosulfate (1 equiv) in the minimum amount of
water was added. The reaction mixture was vigorously stirred until
completion of the reaction (ca. overnight) and concentrated in
vacuum. The Bunte salt was isolated as a white solid after freeze-
̈
drying. Following this procedure, methyl, ethyl, propyl, and benzyl
thiosulfate were isolated in quantitative yields and used in the next step
without further purification.
General Procedure A. Preparation of linear disulfide prodrugs:
Compound 1 was dissolved in NaOH (2 N) aqueous solution (5
equiv), and the mixture was stirred in an ice bath for 30 min.
Chloroform and then the appropriate alkyl thiosulfate (4−8 equiv)
were added, and stirring was continued for 1−2 h. The organic layer
was separated, and the aqueous phase was extracted with CH2Cl2. The
organic layers were combined, dried over MgSO4, filtered, and
concentrated in vacuum. The residue was purified by silica gel
chromatography (gradient: CH2Cl2 to CH2Cl2/MeOH, 95:5) to afford
the expected compound.
CONCLUSION
■
The difficult challenge in optimizing the orally druglike feature
of our promising antimalarial drug albitiazolium by means of a
prodrug approach is somehow limited by intrinsic nonfavorable
parameters of the parent drug, such as its high molecular weight
and lipophilicity associated with the flexible C12-alkyl chain.
Thus, the range of alternative structural modulation of the
prodrug moiety for optimizing oral bioavailability is very
narrow. Nevertheless, within the series of disulfide pro-
prodrugs designed and studied herein, the cyclic compound 8
was identified as the best derivative, exhibiting a substantial
increase in oral activity (ED50 of 7.3 mg/kg and ratio ED50(ip)/
ED50(po) of 13.7%), that is nearly twice more active than the
parent drug.
N,N′-(Dodecan-1,12-diyl)bis[(1Z)-4-hydroxy-1-methyl-2-
(methyldisulfanyl)buten-1-yl)]diformamide 2a. According to general
procedure A, the title compound (941 mg, 84%) was obtained, as a
yellow oil, from 1 (1.18 g, 1.92 mmol), NaOH (4.8 mL), methyl
1
thiosulfate (2.3 g, 7.68 mmol), and CHCl3 (10 mL). H NMR (250
MHz, CDCl3): δ 8.01 and 7.93 (2s, 2H, major and minor rotamers),
3.84 (t, J = 6.7, 4H), 3.45−3.36 (m, 4H), 2.96−2.88 (m, 4H), 2.38 and
2.37 (2s, 6H), 2.05−1.70 (m, 8H), 1.69−1.15 (m, 20H). 13C NMR
(75 MHz, CDCl3): δ 162.4, 161.7, 135.3, 133.0, 132.1, 131.1, 60.6,
60.1, 53.3, 48.1, 42.8, 32.9, 32.8, 29.3, 29.3, 28.9, 27.6, 27.1, 26.7, 23.4,
23.2, 18.9, 18.3. MS (ESI+): 581.0 [M + H]+. HRMS (TOF-ESI+)
calcd for C26H49N2O4S4 [M + H]+: 581.2575, found: 581.2566.
N,N′-(Dodecan-1,12-diyl)bis[(1Z)-2-(ethyldisulfanyl)-4-hydroxy-1-
methylbuten-1-yl)]diformamide 2b. According to general procedure
A, the title compound (1.14 g, 94%) was obtained, as a yellow oil, from
1 (1.22 g, 2 mmol), NaOH (5 mL), ethyl thiosulfate (1.57 g, 9.6
mmol), and CHCl3 (10 mL). Purity was determined by analytical
EXPERIMENTAL SECTION
■
Biology. In Vitro Antimalarial Activity. Drug effects on in vitro P.
falciparum growth (Nigerian strain) were measured in microtiter plates
according to a modified Desjardins test.35,36 Parasite growth was
assessed by measuring the incorporation of [3H]hypoxanthine into
nucleic acids after 48 h of contact with the drug, as previously
described.37 Results were expressed as the concentration resulting in
50% inhibition (IC50).
1
HPLC and was 92%. H NMR (300 MHz, CDCl3): δ 7.93 and 7.85
(2s, 2H, major and minor rotamers), 3.76−3.70 (m, 4H), 3.33 (t, J =
7.2, 4H), 2.86−2.78 (m, 4H), 2.62−2.55 (m, 4H), 1.94 and 1.92 (2s,
6H), 1.53−1.41 (m, 4H), 1.23−1.14 (m, 22H). 13C NMR (75 MHz,
CDCl3): δ 162.7, 161.9, 134.6, 132.8, 132.5, 131.8, 60.6, 60.1, 48.8,
42.9, 33.4, 33.2, 32.9, 29.4, 29.2, 29.1, 27.8, 27.2, 26.8, 19.1, 18.6, 14.2.
MS (ESI+): 609.4 [M + H]+. HRMS (TOF-ESI+) calcd for
C28H53N2O4S4 [M + H]+: 609.2888, found: 609.2880.
In Vivo Antimalarial Activity. Antimalarial activities were
determined against the Plasmodium vinckei petteri (279BY) strain in
female Swiss mice. Compounds were injected ip and orally in DMSO,
or 0.9% NaCl, or a mixture of intralipid/DMSO (90/10). Parasitemia
4624
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
N,N′-(Dodecan-1,12-diyl)bis[(1Z)-4-hydroxy-1-methyl-2-(propyl-
disulfanyl)-buten-1-yl)]diformamide 2c. According to general
procedure A, the title compound (1.8 g, 87%) was obtained, as a
yellow oil, from 1 (2 g, 3.25 mmol), NaOH (8.5 mL), n-butyl
thiosulfate (5.5 g, 19.5 mmol), and CHCl3 (15 mL). The compound
CDCl3): δ 7.93 and 7.79 (2s, 2H, major and minor rotamers), 4.98−
4.90 (m, 2H), 4.36−4.10 (m, 6H), 3.40−3.23 (m, 4H), 2.92 (t, J = 6.7,
4H), 2.29 (s, 6H), 2.14−1.95 (m, 8H), 1.60−1.10 (m, 38H), 0.91,
0.88, 0.84, 0.81 (2d, 12H). 13C NMR (75 MHz, CDCl3): δ 172.4,
172.3, 162.2, 161.3, 155.6, 136.2, 135.1, 130.4, 130.0, 79.9, 62.8, 60.4,
58.5, 53.4, 48.8, 42.9, 31.3, 29.5, 29.3, 29.3, 29.2, 28.3, 27.8, 27.2, 26.8,
23.6, 23.3, 19.0, 18.9, 18.7, 17.6, 17.5. MS (ESI+): 979.5 [M + H]+.
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-4-(formylamino)3-
(methyldisulfanyl)penten-3-yl] Bis[(2,6-bis(tert-butoxycarbonyl)-
diamino)hexanoate] 4b. According to general procedure B, the
title compound (665 mg, 68%) was obtained, as a white amorphous
solid, from 2a (453 mg, 0.78 mmol), (Boc)2-Lys (1.08 g, 3.11 mmol),
N,N′-dicyclohexylcarbodiimide (643 mg, 3.11 mmol), DMAP (57 mg,
0.467 mmol), and CH2Cl2 (10 mL). The compound was purified by
1
was purified by column chromatography (CH2Cl2/MeOH, 98:2). H
NMR (250 MHz, CDCl3): δ 7.93 and 7.85 (2s, 2H, major and minor
rotamers), 3.78−3.62 (m, 4H), 3.49−3.27 (m, 4H), 2.85 (t, J = 6.6,
4H), 2.57 (t, J = 7.2, 4H), 1.96 (s, 6H), 1.70−0.80 (m, 30H). 13C
NMR (75 MHz, CDCl3): δ 163.1, 162.3, 135.0, 133.3, 132.8, 132.2,
61.0, 60.5, 53.9, 48.8, 43.3, 41.9, 41.7, 33.3, 29.9, 29.7, 29.6, 29.5, 28.2,
27.6, 27.3, 22.7, 19.5, 19.0, 13.5. MS (ESI+): 637.4 [M + H]+. HRMS
(TOF-ESI+) calcd for C30H57N2O4S4 [M + H]+: 637.3201, found:
637.3205.
1
column chromatography (cyclohexane/AcOEt, 4:6). H NMR (300
N,N′-(Dodecan-1,12-diyl)bis[(1Z)-2-(benzyldisulfanyl)-4-hydroxy-
1-methyl-buten-1-yl)] diformamide 2d. According to general
procedure A, the title compound (1.11 g, 94%) was obtained, as a
yellow oil, from 1 (1 g, 1.62 mmol), NaOH (4.1 mL), benzyl
MHz, CDCl3): δ 7.92 and 7.79 (s, 2H, major and minor rotamers),
5.02 (bs, 2H), 4.52 (bs, 2H), 4.35−4.10 (m, 6H), 3.45−3.23 (m, 4H),
3.10−2.85 (2 m, 8H), 2.29 (s, 6H), 1.98 and 1.95 (2s, 6H), 1.80−1.10
(m, 66H). 13C NMR (75 MHz, CDCl3): δ 172.8, 162.3, 156.1, 136.2,
129.9, 80.0, 79.2, 62.9, 53.3, 48.7, 42.9, 40.0, 32.3, 29.7, 29.5, 29.3,
28.5, 28.3, 27.8, 27.2, 26.9, 23.3, 22.5, 18.9, 18.7. MS (ESI+): 1237.4
[M + H]+.
2,2′-((3Z,19Z)-5,18-Diformyl-4,19-dimethyl-1,2-dithia-5,18-dia-
zacycloicosa-3,19-diene-3,20-diyl)bis(ethane-2,1-diyl) Bis[(2-(tert-
butoxycarbonyl)amino)-3-methylbutanoate] 5a. According to gen-
eral procedure B, the title compound (381 mg, 70%) was obtained, as
a white amorphous solid, from 3 (300 mg, 0.616 mmol), Boc-Val (401
mg, 1.85 mmol), N,N′-dicyclohexylcarbodiimide (382 mg, 1.85 mmol),
4-(dimethylamino)pyridine (38 mg, 0.31 mmol), and CH2Cl2 (8 mL).
The compound was purified by column chromatography (cyclo-
hexane/AcOEt, 4:6). 1H NMR (300 MHz, CDCl3): δ 7.93 and 7.81 (s,
2H, major and minor rotamers), 5.0−4.83 (m, 2H), 4.35−4.02 (m,
6H), 3.55−3.12 (m, 4H), 3.02−2.75 (2 m, 4H), 2.15−1.80 (m, 8H),
1.50−1.10 (m, 22H), 0.90, 0.88, 0.83, 0.81 (2d, 12H). 13C NMR (75
MHz, CDCl3): δ 172.4, 162.0, 161.3, 155.6, 129.4, 79.9, 62.4, 58.5,
47.7, 42.5, 31.3, 30.1, 28.3, 28.0, 27.7, 27.5, 27.1, 27.0, 28.9, 26.8, 26.4,
19.018.5, 17.6. HRMS (TOF-ESI+) calcd for C34H61N4O6S2 [M − H
− 2Cl]+: 685.4033, found: 685.4003.
2,2′-((3Z,19Z)-5,18-Diformyl-4,19-dimethyl-1,2-dithia-5,18-dia-
zacycloicosa-3,19-diene-3,20-diyl)bis(ethane-2,1-diyl) Bis[(2,6-bis-
(tert-butoxycarbonyl)diamino)hexanoate] 5b. According to general
procedure B, the title compound (479 mg, 70%) was obtained, as a
white amorphous solid, from 3 (290 mg, 0.517 mmol), (Boc)2-Lys
(829 mg, 2.39 mmol), 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide (423 μL, 2.39 mmol), DMAP (37 mg, 0.30 mmol),
and CH2Cl2 (8 mL). The compound was purified by column
chromatography (cyclohexane/AcOEt, 4:6). 1H NMR (300 MHz,
CDCl3): δ 8.0−7.86 (m, 4H), 5.20−5.05 (m, 2H), 4.85−4.60 (m, 2H),
4.42−4.10 (m, 8H), 3.80−3.20 (m, 8H), 3.15−2.70 (m, 4H), 2.0−1.90
(m, 6H), 1.85−1.60 (m, 4H), 1.50−1.20 (m, 36H). 13C NMR (75
MHz, CDCl3): δ 172.8, 162.5, 162.4, 161.8, 156.3, 156.2, 155.5, 136.8,
136.2, 131.9, 130.6, 129.6, 80.1, 79.1, 62.6, 60.8, 60.4, 60.0, 53.3, 48.1,
47.5, 42.3, 42.2, 40.0, 34.0, 33.4, 32.3, 28.5, 28.4, 28.3, 28.2, 27.6, 27.3,
27.1, 27.0, 26.5, 22.5, 18.7, 18.6, 18.4, 17.8, 17.4. MS (ESI+): 1165.8
[M + Na]+.
General Procedure C. Deprotection of tBoc-aminoacyl disulfide
prodrugs: Protected pro-prodrug was dissolved in a cold hydrogen
chloride solution (4 M in 1,4-dioxane). The mixture was stirred for 45
min at 0 °C and then concentrated in vacuum. The residue was
triturated in diethyl ether. After removal of the diethyl ether, the
residue was dissolved in water and freeze-dried to afford the desired
compound as its hydrochloride salt.
1
thiosulfate (1.84 g, 8.13 mmol), and CHCl3 (10 mL). H NMR (300
MHz, CDCl3): δ 7.88 and 7.80 (2s, 2H, major and minor rotamers),
7.27−7.15 (m, 10H), 3.81 (s, 4H), 3.65−3.3.50 (m, 4H), 3.35−3.22
(m, 4H), 2.72−2.65 (m, 4H), 1.90 and 1.88 (2s, 6H), 1.47−1.38 (m,
4H), 1.25−1.10 (m, 18H). 13C NMR (75 MHz, CDCl3): δ 162.8,
161.9, 136.5, 136.3, 134.7, 132.9, 132.5, 131.6, 129.3, 129.2, 128.7,
128.6, 127.7, 127.6, 60.5, 60.2, 48.4, 44.6, 44.5, 43.1, 33.1, 29.5, 29.3,
29.2, 27.9, 27.2, 26.9, 19.3, 18.7. MS (ESI+): 733.4 [M + H]+. HRMS
(TOF-ESI+) calcd for C38H57N2O4S4 [M + H]+: 733.3201, found:
733.3203.
(3Z,19Z)-3,20-Bis(2-hydroxyethyl)-4,19-dimethyl-1,2-dithia-5,18-
diazacycloicosa-3,19-diene-5,18-dicarbaldehyde 3. Compound 1 (2
g, 3.27 mmol) was dissolved in a 2 N NaOH aqueous solution (8 mL,
16.33 mmol, 5 equiv), and the mixture was stirred in an ice bath for 30
min. Then dichloromethane (330 mL) was added to the reaction
mixture and, under vigorous stirring, tert-butyl nitrite (1.73 mL, 13.06
mmol, 4 equiv) was added dropwise. After 10 h stirring, the organic
layer was separated and washed with aqueous sodium bicarbonate
solution followed by brine. The organic layer was dried over
magnesium sulfate, filtered, and evaporated in vacuum. The residue
was purified by silica gel chromatography (gradient: CH2Cl2 to
CH2Cl2/MeOH, 95:5) to give compound 3 (400 mg, 25%) as an oily
1
yellow liquid. H NMR (300 MHz, DMSO-d6): δ 8.01 and 7.81 (2s,
2H, major and minor rotamers), 4.70 (m, 2H), 3.60−3.30 (m, 4H,
overlap with water), 2.75−2.55 (m, 4H), 2.10−1.80 (m, 8H), 1.45−
1.15 (m, 20H). 13C NMR (75 MHz, DMSO-d6): δ 162.1, 162.0, 161.9,
161.8, 161.5, 137.8, 137.7, 136.1, 129.7, 129.4, 129.3, 129.2, 66.3, 60.0,
59.9, 41.9, 41.6, 34.4, 29.0, 28.9, 28.8, 27.3, 27.2, 26.7, 26.3, 18.3, 18.1,
17.9. MS (ESI+): 487.5 [M + H]+. HRMS (TOF-ESI+) calcd for
+
C24H43N2O4S2 [M + H]+: 487.2648, found: 487.2664.
General Procedure B. Preparation of tBoc-aminoacyl disulfide
prodrugs: To an ice-cold solution of the protected amino acid (3 to 4.5
equiv) in dichloromethane was added N,N′-dicyclohexylcarbodiimide
(3 or 4 equiv) or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (4
or 4.5 equiv) and DMAP (0.5 or 0.6 equiv). The mixture was stirred
for 15 min, and a solution of the required disulfide prodrug (either 2a
or 3) in dichloromethane was added. After 2−3 h stirring at room
temperature, the solvent was then evaporated under reduced pressure.
The residue was taken up in dichloromethane, and the resulting
organic layer was washed successively with a 1 M aqueous KHSO4
solution, a saturated aqueous NaHCO3 solution, and brine and then
dried over Na2SO4. The solvent was removed in vacuum, and the
residue was purified by column chromatography to afford the expected
compound.
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-4-(formylamino)3-
(methyldisulfanyl)penten-3-yl] Bis[(2-(tert-butoxycarbonyl)amino)-
3-methylbutanoate] 4a. According to general procedure B, the title
compound (617 mg, 85%) was obtained, as a white amorphous solid,
from 2a (430 mg, 0.742 mmol), Boc-Val (483 mg, 2.22 mmol), N,N′-
dicyclohexylcarbodiimide (459 mg, 2.22 mmol), DMAP (45 mg, 0.371
mmol), and CH2Cl2 (8 mL). The compound was purified by column
chromatography (cyclohexane/AcOEt, 1:1). 1H NMR (300 MHz,
(3Z,3′Z)-4,4′-(Dodecan-1,12-diyl)bis[4-(formylamino)-3-
(methyldisulfanyl)penten-3-yl] Bis[(2-amino-3-methylbutanoate)
hydrochloride] 6a. According to general procedure C, the title
compound (406 mg, 78%) was obtained as a white powder from 4a
(597 mg, 0.610 mmol) and 4 N hydrogen chloride solution in 1,4-
1
dioxane (22 mL). H NMR (300 MHz, DMSO-d6): δ 8.72 (bs, 6H),
8.01 and 7.79 (2s, 2H, major and minor rotamers), 4.40−4.25 (m,
4H), 3.90−3.8 (m, 2H), 3.40−3.30 (m, 4H), 3.05−2.90 (m, 4H), 2.39
and 2.35 (2s, 6H), 2.30−2.15 (m, 2H), 2.02 and 1.94 (2s, 6H), 1.50−
4625
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
1.22 (m, 20H), 1.0, 0.98, 0.96, 0.94 (2d, 12H). 13C NMR (75 MHz,
DMSO-d6): δ 168.9, 168.8, 161.9, 161.5, 137.6, 135.2, 127.6, 63.2,
63.0, 57.3, 57.2, 47.3, 41.7, 29.2, 28.9, 28.7, 27.3, 26.6, 26.2, 22.9, 22.8,
yellow solution was hydrolyzed by adding TEAB (1M, 10 mL) until
pH ≈ 8. Triethyl phosphate was removed by extracting the aqueous
layer with Et2O. Water was evaporated, and free triethylammonium
phosphate salts resulting from the hydrolysis of excess POCl3 with
TEAB were precipitated from acetone. The residue was purified by
reverse phase RP18 chromatography, eluting with H2O/CH3CN
system (H2O to H2O/CH3CN, 85:15). A colorless oil (0.43 g, 49%)
was obtained after evaporation and freeze-drying. 1H NMR (300 MHz,
D2O): δ 9.85 (s, 2H, major and minor rotamers), 4.43 (t, J = 7.2, 4H),
3.90−3.75 (m, 4H), 3.38 (t, J = 6.0, 4H), 2.49 and 2.47 (2s, 6H),
1.90−1.75 (m, 4H), 1.30−1.0 (m, 16H). 13C NMR (75 MHz, D2O): δ
156.8, 156.2, 145.1, 143.5, 137.7, 136.3, 63.8 (d, JCP = 5.3), 55.4, 46.1,
30.9, 30.5, 30.4, 30.3, 30.0, 27.1, 27.0, 17.3 (d, JCP = 6.8), 13.1, 12.9.
31P NMR (121 MHz, D2O): δ 0.7. MS (ESI+): 730.9 [M + 2TFA]+,
+
18.5, 18.4. HRMS (TOF-ESI+) calcd for C36H67N4O6S4 [M − H −
2Cl]+: 779.3943, found: 779.3913.
(3Z,3′Z)-4,4′-(Dodecan-1,12-diyl)bis[4-(formylamino)-3-
(methyldisulfanyl)penten-3-yl] Bis[(2,6-diaminohexanoate) dihy-
drochloride] 6b. According to general procedure C, the title
compound (464 mg, 89%) was obtained as a white powder from 4b
(660 mg, 0.53 mmol) and 4 N hydrogen chloride solution in 1,4-
1
dioxane (10 mL). H NMR (300 MHz, DMSO-d6): δ 8.71 (bs, 6H),
8.13 (bs, 6H), 8.04 and 7.80 (s, 2H, major and minor rotamers), 4.40−
4.25 (m, 4H), 4.10−3.92 (m, 2H), 3.11−2.90 (m, 4H), 2.88−2.63 (m,
4H), 2.41 and 2.38 (2s, 6H), 2.04 and 1.96 (2s, 6H), 1.90−1.78 (m,
4H), 1.70−1.20 (m, 30H). 13C NMR (75 MHz, DMSO-d6): δ 169.4,
162.0, 137.6, 134.8, 128.4, 127.4, 63.4, 51.5, 47.1, 41.6, 29.2, 28.9, 28.7,
28.3, 27.3, 26.6, 26.1, 22.9, 22.7, 21.2, 18.3, 17.9. MS (ESI+): 837.2 [M
+ H]+. HRMS (TOF-ESI+) calcd for C38H72N6O6S4·2HCl [M − H −
2Cl]+: 837.4479, found: 837.4486.
307.0 [(M + 2H)/2]2+.
N,N′-(Dodecane-1,12-diyl)bis[(1Z)-1-(2-thietanylidene)ethyl]-
diformamide 10. To an ice-cooled solution of compound 9 (0.18 g,
0.29 mmol) in water (0.5 mL) was added a 2 N NaOH aqueous
solution (0.58 mL, 1.16 mmol, 4 equiv), the mixture was stirred for 10
min, and then 2 mL of dichloromethane was added. After 30 min
stirring at room temperature, the reaction mixture was taken up in
dichloromethane, and the resulting aqueous layer was extracted with
dichloromethane. The combined organic layers were dried over
Na2SO4 and concentrated. The residue was purified by column
chromatography (CH2Cl2/MeOH, 97:3), affording compound 10 (96
2,2′-((3Z,19Z)-5,18-Diformyl-4,19-dimethyl-1,2-dithia-5,18-dia-
zacycloicosa-3,19-diene-3,20-diyl)bis(ethane-2,1-diyl) Bis[(2-amino-
3-methylbutanoate) hydrochloride] 7a. According to general
procedure C, the title compound (218 mg, 95%) was obtained as a
white powder from 5a (267 mg, 0.301 mmol) and 4 N hydrogen
chloride solution in 1,4-dioxane (6 mL). 1H NMR (300 MHz, DMSO-
d6): δ 8.04 (bs, 6H), 7.84 and 7.79 (s, 2H, major and minor rotamers),
4.31−4.20 (m, 4H), 3.85−3.70 (m, 2H), 3.55−3.20 (m, 4H, overlap
with water), 3.0−2.80 (m, 4H), 2.20−2.10 (m, 2H), 2.05−1.90 (m,
6H), 1.40−1.15 (m, 20H), 1.05−0.90 (m, 12H). 13C NMR (75 MHz,
DMSO-d6): δ 169.1, 169.0, 161.9, 127.3, 66.3, 63.0, 57.3, 29.3, 29.0,
28.8, 27.3, 27.1, 26.8, 26.7, 26.6, 26.4, 26.3, 26.1, 25.6, 18.4, 18.3, 18.1,
17.9, 17.6. MS (ESI+): 685.6 [M + H]+. HRMS (TOF-ESI+) calcd for
C34H61N4O6S2.2HCl [M − H − 2Cl]+: 685.4033, found: 685.4003.
2,2′-((3Z,19Z)-5,18-Diformyl-4,19-dimethyl-1,2-dithia-5,18-dia-
zacycloicosa-3,19-diene-3,20-diyl)bis(ethane-2,1-diyl) Bis[(2,6-dia-
minohexanoate) dihydrochloride] 7b. According to general
procedure C, the title compound (138 mg, 85%) was obtained as a
white powder from 5b (207 mg, 0.181 mmol) and 4 N hydrogen
chloride solution in 1,4-dioxane (6 mL). 1H NMR (300 MHz, DMSO-
d6): δ 8.10−7.80 (m, 8H), 4.25 (m, 4H), 4.02 (m, 2H), 3.05−2.55 (m,
8H), 2.10−1.72 (m, 12H), 1.68−1.10 (m, 42H). 13C NMR (75 MHz,
DMSO-d6): δ 169.8, 162.1, 162.0, 138.0, 136.8, 129.4, 127.6, 63.2,
59.0, 58.9, 54.9, 51.8, 51.7, 41.1, 40.8, 38.1, 34.0, 33.5, 33.1, 29.4, 29.0,
28.4, 27.2, 27.0, 26.8, 26.5, 26.4, 26.3, 26.2, 25.7, 25.6, 22.4, 21.7, 21.2,
19.8, 19.2, 18.0, 17.9, 17.2, 14.1, 11.2. MS (ESI+): 743.5 [M + H]+.
HRMS (TOF-ESI+) calcd for C36H67N6O6S2 2HCl [M − H − 2Cl−]+:
743.4564, found: 743.4581.
1
mg, 73%) as a yellow oil. H NMR (300 MHz, CDCl3): δ 7.94 and
7.88 (2s, 2H, major and minor rotamers), 3.50−3.40 (m, 4H), 3.35−
3.20 (m, 4H), 3.10−3.05 (m, 4H), 1.66 and 1.63 (2s, 6H), 1.45−1.39
(m, 4H), 1.23−1.10 (m, 16H). 13C NMR (75 MHz, CDCl3): δ 161.5,
160.1, 130.1, 129.2, 122.3, 120.5, 45.9, 40.7, 33.2, 30.9, 29.3, 28.7, 28.5,
28.4, 28.2, 27.4, 26.0, 25.6, 20.2, 19.5, 14.3, 14.1. MS (ESI+): 453.2
[(M + 2H)/2]2+.
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-3-(ethyldisulfanyl)-4-
(formylamino)penten-3-yl] Bis[diethyl phosphate] 11. In a dried
round-bottom flask, the prodrug 2b (0.26 g, 0.43 mmol) was dissolved
in CH2Cl2 (10 mL). Diethyl chlorophosphate (0.15 mL, 1.07 mmol,
2.5 equiv), previously dried under vacuum, was added slowly at 0 °C
under neutral atmosphere, followed by TEA (0.17 mL, 1.20 mmol, 2.8
equiv) and DMAP (0.016 g, 0.13, 0.3 equiv). The ice bath was
removed, and the reaction mixture was stirred for 1 h. The solvent was
evaporated, and the crude residue was directly purified on silica gel
chromatography (CH2Cl2/MeOH, 97:3), affording a clear yellow oil
1
(0.28 g, 73%). H NMR (300 MHz, CDCl3): δ 7.94 and 7.83 (s, 2H,
major and minor rotamers), 4.23−4.00 (m, 12H), 3.34−3.29 (m, 4H),
2.94 (t, J = 5.7, 4H), 2.58 (q, J = 7.2, 4H), 1.94 (s, 6H), 1.48−1.43 (m,
4H), 1.34−1.12 (m, 34H). 13C NMR (75 MHz, CDCl3): δ 162.3,
161.2, 1135.9, 135.0, 130.2, 130.0, 65.0 (d, JCP = 6.7), 64.8 (d, JCP
=
(3Z,19Z)-3,20-Bis[2-(acetoxy)ethyl)]-4,19-dimethyl-1,2-dithia-
5,18-diazacycloicosa-3,19-diene-5,18-dicarbaldehyde 8. To a cold
solution (0 °C) of 3 (72.8 mg) and TEA (85 μL, 0.59 mmol, 4 equiv)
in dichloromethane (1.5 mL) was added dropwise acetyl chloride (50
μL, 0.59 mmol, 4 equiv). The mixture was stirred for 15 min at this
temperature and then allowed to warm to room temperature (ca. 1 h).
The mixture was diluted with dichloromethane, and the organic layer
was washed with water and finally dried over Na2SO4. The solvent was
removed in vacuum, and the residue was purified by column
chromatography (gradient: CH2Cl2 to CH2Cl2/MeOH, 98:2) to
afford 15 (80 mg, 94%) as an oily yellow liquid. 1H NMR (300 MHz,
CDCl3): δ 8.01 and 7.90 (s, 2H, major and minor rotamers), 4.22−
4.15 (m, 4H), 3.60−3.20 (m, 4H), 3.05−2.75 (m, 4H), 2.10−1.90 (m,
12H), 1.50−1.00 (m, 20H). 13C NMR (75 MHz, CDCl3): δ 170.7,
162.4, 162.1, 161.3, 136.9, 131.1, 129.9, 61.9, 61.8, 61.7, 47.7, 42.4,
42.3, 30.0, 29.7, 29.3, 28.3, 28.1, 27.8, 27.7, 27.6, 27.3, 27.1, 26.9, 26.6,
26.5, 20.9, 18.6, 18.4, 17.5. MS (ESI+): 571.4 [M + H]+. HRMS
(TOF-ESI+) calcd for C28H47N2O6S2 [M + H]+: 571.2870, found:
571.2876.
6.0), 63.8 (d, JCP = 7.2), 63.7 (d, JCP = 6.7), 48.9, 42.9, 33.2, 33.1, 30.6
(d, JCP = 7.5), 29.5, 29.2, 27.8, 27.2, 26.8, 19.0, 18.8, 16.1 (d, JCP = 6.8),
16.0 (d, JCP = 6.8), 14.1, 14.0. 31P NMR (121 MHz, CDCl3): δ −0.9
and −0.8. MS (ESI+): 903.4 [M + Na]+, 452.2 [(M + Na)/2]2+.
HRMS (TOF-ESI+) calcd for C36H71N2O10P2S4 [M + H]+: 881.3467,
found: 881.3484.
General Procedure D. Preparation of CycloSal phosphorylated
prodrug 12b and 12d: In a flame-dried round-bottom glass filled with
argon, anhydrous dichloromethane was cooled to −50 °C. Freshly
distilled POCl3 (15 equiv) was added, and a combination of dried 2-
hydroxybenzoic acid (15 equiv) and TEA (30 equiv) in dichloro-
methane (2 mL/mmol of POCl3) was slowly added in 2 h. After
addition, the reaction mixture was warmed to 0 °C for 1 h and cooled
down to −50 °C for the slow addition of a combination of the
disulfide prodrug (1 equiv), TEA (15 equiv) and DMAP (1−2 equiv)
in dichloromethane (6.5 mL/mmol of the prodrug). The mixture was
allowed to reach room temperature. The reaction was quenched after
overnight stirring by adding water. The organic layer was washed with
water, dried with MgSO4, filtered and concentrated to afford brown oil
which was further purified on silica gel chromatography, eluting with
CH2Cl2/MeOH system.
N,N′-(Dodecane-1,12-diyl)bis[2-(4-methyl-1,3-thiazol-3-ium-5-
yl)ethyl dihydrogen phosphate] 9. Compound 1 (0.63 g, 1.02 mmol)
was suspended in dry triethyl phosphate (4 mL), freshly distilled
POCl3 (0.38 mL, 4.07 mmol, 4 equiv) was added, and stirring was
maintained overnight at room temperature. The homogeneous clear
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-3-(ethyldisulfanyl)-4-
(formylamino)penten-3-yl] Bis[(2-oxo-4H-2l5-benzo[1,3,2]dioxa)-
4626
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
phosphate] 12b. According to procedure D, the title compound (0.92
g, 53%) was obtained as a clear yellow oil from 2-hydroxybenzoic acid
(3.43 g, 27.6 mmol), POCl3 (2.53 mL, 27.6 mmol), TEA (7.7 mL, 55.3
mmol), the disulfide prodrug 2b (1.12 g, 1.84 mmol), TEA (3.84 mL,
ASSOCIATED CONTENT
■
S
* Supporting Information
Spectral data (1H and 31P NMR, HRMS, HPLC traces) for final
compounds 2a−d, 3 and 6a,b, 7a,b, 8, 9, 11, 12b,d, 13b,d. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
27.66 mmol), and DMAP (0.25 g, 1.84 mmol). H NMR (300 MHz,
CDCl3): δ 7.92 and 7.72 (s, 2H, major and minor rotamers), 7.30−
7.20 (m, 2H), 7.15−6.95 (m, 6H), 5.37−5.20 (m, 4H), 4.31−4.23 (m,
4H), 3.32−3.20 (m, 4H), 3.02−2.90 (m, 4H), 2.55 (q, J = 7.2, 4H),
1.89 (s, 6H), 1.45−1.32 (m, 4H), 1.25−1.15 (m, 22H). 13C NMR (75
MHz, CDCl3): δ 162.2, 161.2, 150.1(d, JCP = 7.0), 136.2, 129.9, 129.7,
AUTHOR INFORMATION
■
Corresponding Author
129.6, 125.3 (d, JCP = 4.1), 124.4, 124.2, 120.6, 120.5, 118.6 (d, JCP
=
8.8), 68.6 (d, JCP = 6.9), 66.1 (d, JCP = 6.0), 49.0, 42.9, 33.3, 33.1, 30.7,
30.6, 29.5, 29.2 (d, JCP = 8.2), 27.9, 27.2, 26.8, 19.1, 14.2, 14.1. 31P
NMR (121 MHz, CDCl3): δ −9.4 and −9.5. MS (ESI+): 945.5 [M +
H]+. HRMS (TOF-ESI+) calcd for C42H63N2O10P2S4 [M + H]+:
945.2841, found: 945.2835.
*(S.P.): Phone: +33 467 144 964; fax: +33 467 042 029; e-
mail: suzanne.peyrottes@univ-montp2.fr. (H.J.V.): Phone: +33
467 143 745; fax: +33 467 144 286; e-mail: vial@univ-montp2.
fr.
Author Contributions
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-3-(benzyldisulfanyl)-4-
(formylamino)penten-3-yl] Bis[(2-oxo-4H-2l5-benzo[1,3,2]dioxa)-
phosphate] 12d. According to procedure D, the title compound
(1.26 g, 77%) was obtained as a clear yellow oil from 2-
hydroxybenzoic acid (3 g, 24.1 mmol), POCl3 (2.21 mL, 24.1
mmol), TEA (6.72 mL, 48.33 mmol), the disulfide prodrug 2d (1.24 g,
1.53 mmol), TEA (3.36 mL, 24.16 mmol) and DMAP (0.39 g, 3.22
⊥These authors have contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
1
mmol). H NMR (300 MHz, CDCl3): δ 7.92 and 7.69 (s, 2H, major
This work was supported by the European Community
Integrated Project AntiMal (no. IP-018834) and Sanofi.
M.H., S.C., and J.-F.D. are grateful to Sanofi and the European
Community for postdoctoral fellowships. The authors thank
M.-C. Bergogne for manuscript editing, and C. Rabeson and J.-
Y. Puy for technical assistance.
and minor rotamers), 7.30−6.90 (m, 18H), 5.32−5.20 (m, 4H), 4.25−
4.11 (m, 4H), 3.79 (s, 4H), 3.30−3.22 (m, 4H), 2.92−2.75 (m, 4H),
1.88, 1.85 (2s, 6H), 1.45−1.30 (m, 4H), 1.20−1.05 (m, 16H). 13C
NMR (75 MHz, CDCl3): δ 162.3, 161.3, 150.1 (d, JCP = 7.5), 150.0,
136.6, 136.2, 136.1, 129.9 (d, JCP = 1.5), 129.7, 129.6, 129.4, 129.2,
128.7, 128.6, 128.4, 127.8, 127.6, 125.4, 125.3, 124.4, 124.2, 120.7,
120.4, 118.5 (d, JCP = 8.3), 68.6 (d, JCP = 6.0), 66.0 (d, JCP = 6.0), 53.5,
48.9, 44.5, 44.3, 43.6, 43.0, 30.7, 30.6, 29.5, 29.3, 29.2 (d, JCP = 8.3),
27.9, 27.2, 26.8, 19.2, 18.8. 31P NMR (121 MHz, CDCl3): δ −9.6 and
−9.5. MS (ESI+): 1069.5 [M + H]+. HRMS (TOF-ESI+): calcd for
C52H67N2O10P2S4 [M + H]+: 1069.3154, found: 1069.3179.
General Procedure E. Removal of the CycloSal protecting groups:
the protected phosphorylated prodrug (1 equiv) was dissolved in a
mixture of acetonitrile and water (2:1, v/v, 12.5 mL/mmol of
prodrug). Lithium hydroxide monohydrate (5 equiv) was added, and
the reaction was stirred at 20 °C for 24 h. After concentration, the
yellow residue was purified on reverse phase RP18 chromatography,
eluting with H2O/CH3OH systems.
ABBREVIATIONS USED
■
ACTs, artemisinin-based combination therapy; FPIX, ferripro-
toporphyrin IX; DCC, dicyclohexylcarbodiimide; DMAP, 4-
(dimethylamino)pyridine; EDC, 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide; TEA, triethylamine;
TEAB, triethylammonium bicarbonate; DMSO, dimethyl
sulfoxide; TFA, trifluoroacetic acid; 2D-COSY, two-dimen-
sional correlation spectrocopy experiment
REFERENCES
■
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-3-(ethyldisulfanyl)-4-
(formylamino)penten-3-yl] Bis(dilithium phosphate) 13b. According
to procedure E, the title compound (72 mg, 75%) was obtained as a
white powder from derivative 12b (115 mg, 0.12 mmol), LiOH·H2O
(1) The RBM Partnership. Roll Back Malaria; The Global Malaria
Action Plan for a malaria-free world, 2008; http://www.
rollbackmalaria.org/gmap/index.html (Accessed January 2012).
(2) WHO. World Malaria Report, World Health Organization, 2010;
http://www.who.int/malaria/world_malaria_report_2010/en/index.
html (Accessed December 2011).
1
(25.5 mg, 0.6 mmol) and acetonitrile/water (1.5 mL). H NMR (300
MHz, D2O): δ 7.87 and 7.83 (2s, 2H, major and minor rotamers),
3.90−3.75 (m, 4H), 3.50−3.30 (m, 4H), 3.0−2.80 (m, 4H), 2.66 (q, J
= 6.9, 4H), 1.97 and 1.92 (2s, 6H), 1.50−1.40 (m, 4H), 1.25−1.10 (m,
22H). 13C NMR (75 MHz, D2O): δ 165.2, 163.4, 135.8, 132.2, 131.7,
131.0, 62.1 (d, JCP = 4.0), 48.3, 42.3, 33.4, 33.2, 31.3 (d, JCP = 3.5),
29.4, 29.3, 29.2, 28.4, 27.4, 27.2, 26.8, 17.8, 17.3, 14.1. 31P NMR (121
MHz, D2O): δ 3.5 and 3.4. MS (ESI+): 769.4 [M + H]+. HRMS
(TOF-ESI−): calcd for C28H53N2O10P2S4 [M − H]−: 767.2058,
found: 767.2050.
(3) Olliaro, P.; Wells, T. N. C. The Global Portfolio of New
Antimalarial Medicines Under Development. Clin. Pharmacol. Ther.
2009, 85, 584−595.
(4) Dondorp, A. M.; Nosten, F.; Yi, P.; Das, D.; Phyo, A. P.; Tarning,
J.; Lwin, K. M.; Ariey, F.; Hanpithakpong, W.; Lee, S. J.; Ringwald, P.;
Silamut, K.; Imwong, M.; Chotivanich, K.; Lim, P.; Herdman, T.; An,
S. S.; Yeung, S.; Singhasivanon, P.; Day, N. P. J.; Lindegardh, N.;
Socheat, D.; White, N. J. Artemisinin Resistance in Plasmodium
falciparum Malaria. New Engl. J. Med. 2009, 361, 455−467.
(5) White, N. J. Artemisinin resistance-the clock is ticking. Lancet
2010, 376, 2051−2052.
N,N′-(Dodecan-1,12-diyl)bis[(3Z)-3-(benzyldisulfanyl)-4-
(formylamino)penten-3-yl] Bis(dilithium phosphate) 13d. According
to procedure E, the title compound (83 mg, 12%) was obtained as a
white powder from derivative 12d (783 mg, 0.73 mmol) and
1
LiOH·H2O (153 mg, 3.65 mmol) and acetonitrile/water (9 mL). H
(6) Wells, T. N.; Poll, E. M. When is enough enough? The need for a
robust pipeline of high-quality antimalarials. Discovery Med. 2010, 9,
389−398.
(7) Schlitzer, M. Malaria chemotherapeutics part 1: History of
antimalarial drug development, currently used therapeutics, and drugs
in clinical development. ChemMedChem 2007, 2, 944−986.
(8) Hamze, A.; Rubi, E.; Arnal, P.; Boisbrun, M.; Carcel, C.; Salom-
Roig, X.; Maynadier, M.; Wein, S.; Vial, H.; Calas, M. Mono- and bis-
thiazolium salts have potent antimalarial activity. J. Med. Chem. 2005,
48, 3639−3643.
NMR (300 MHz, D2O): δ 7.80 and 7.66 (2s, 2H, major and minor
rotamers), 7.14−7.10 (m, 10H), 3.92 (bs, 4H), 3.76 (bs, 4H), 3.10 (bs,
2H), 2.80 (bs, 4H), 1.90 and 1.85 (2s, 6H), 1.30−1.20 (m, 4H), 1.10−
0.90 (m, 22H). 13C NMR (75 MHz, D2O): δ 164.9, 163.3, 136.7,
136.6, 136.3, 132.4, 131.7, 130.5, 129.2, 128.7, 127.6, 62.9, 48.8, 48.0,
44.1, 42.3, 31.0, 29.5, 29.3, 28.5, 27.7, 27.3, 26.9, 18.0, 17.4. 31P NMR
(121 MHz, D2O): δ 4.0 and 3.9. MS (ESI+): 893.4 [M + H]+. HRMS
+
(TOF-ESI+): calcd for C38H59N2O10P2S4 [M + H]+: 893.2528,
found: 893.2527.
4627
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628

Journal of Medicinal Chemistry
Article
(9) Salom-Roig, X. J.; Hamze, A.; Calas, M.; Vial, H. J. Dual
molecules as new antimalarials. Comb. Chem. High Throughput
Screening 2005, 8, 49−62.
(10) Vial, H. J.; Penarete, D.; Wein, S.; Caldarelli, S.; Fraisse, L.;
Peyrottes, S. Lipids as Drug Targets for Malaria Therapy. In
Apicomplexan Parasites: Molecular Approaches toward Targeted Drug
Development; Becker, K., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA:
Weinheim, Germany, 2011; p 544.
(28) Rabinowitz, R. Reactions of Phosphonic Acid Esters with Acid
Chlorides - a Very Mild Hydrolytic Route. J. Org. Chem. 1963, 28,
2975−2978.
(29) Meier, C. 2-Nucleos-5′-O-yl-4H-1,3,2-benzodioxaphosphinin-2-
oxides - A new concept for lipophilic, potential prodrugs of biologically
active nucleoside monophosphates. Angew. Chem., Int. Ed. 1996, 35,
70−72.
(30) Meier, C. CycloSal phosphates as chemical Trojan horses for
intracellular nucleotide and glycosylmonophosphate delivery -
Chemistry meets biology. Eur. J. Org. Chem. 2006, 1081−1102.
(31) Meier, C.; Balzarini, J. Application of the cycloSal-prodrug
approach for improving the biological potential of phosphorylated
biomolecules. Antiviral Res. 2006, 71, 282−292.
(32) Wu, S. Y.; Casida, J. E. Neuropathy Target Esterase Inhibitors -
2-Alkyl-, 2-Alkoxy-, and 2-(Aryloxy)-4h-1,3,2-Benzodioxaphosphorin
2-Oxides. Chem. Res. Toxicol. 1992, 5, 680−684.
(33) Wu, S. Y.; Casida, J. E. Neuropathy Target Esterase Inhibitors -
Enantiomeric Separation and Stereospecificity of 2-Substituted-4h-
1,3,2-Benzodioxaphosphorin 2-Oxides. Chem. Res. Toxicol. 1994, 7,
77−81.
(34) Ludek, O. R.; Meier, C. Synthesis of carbocyclic nucleotides as
potential substrates for thymidylate kinase. Nucleosides, Nucleotides
Nucleic Acids 2005, 24, 683−686.
(35) Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D.
Quantitative assessment of antimalarial activity in vitro by a
semiautomated microdilution technique. Antimicrob. Agents Chemother.
1979, 16, 710−718.
(36) Ancelin, M. L.; Calas, M.; Vidal-Sailhan, V.; Herbute, S.;
Ringwald, P.; Vial, H. J. Potent inhibitors of Plasmodium phospholipid
metabolism with a broad spectrum of in vitro antimalarial activities.
Antimicrob. Agents Chemother. 2003, 47, 2590−2597.
(11) Vial, H. J.; Wein, S.; Farenc, C.; Kocken, C.; Nicolas, O.;
Ancelin, M. L.; Bressolle, F.; Thomas, A.; Calas, M. Prodrugs of
bisthiazolium salts are orally potent antimalarials. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 15458−15463.
(12) Wein, S.; Calas, M.; Bressolle, F.; Herrera, S.; Thomas, A.; Vial,
H. Malaria: towards a new treatment. Med
341−343.
́
ecine/Sciences 2005, 21,
(13) Biagini, G. A.; Richier, E.; Bray, P. G.; Calas, M.; Vial, H.; Ward,
S. A. Heme binding contributes to antimalarial activity of bis-
quaternary ammoniums. Antimicrob. Agents Chemother. 2003, 47,
2584−2589.
(14) Wengelnik, K.; Vidal, V.; Ancelin, M. L.; Cathiard, A. M.;
Morgat, J. L.; Kocken, C. H.; Calas, M.; Herrera, S.; Thomas, A. W.;
Vial, H. J. A class of potent antimalarials and their specific
accumulation in infected erythrocytes. Science 2002, 295, 1311−1314.
(15) Caldarelli, S. A.; Boisbrun, M.; Alarcon, K.; Hamze, A.; Ouattara,
M.; Salom-Roig, X.; Maynadier, M.; Wein, S.; Peyrottes, S.; Pellet, A.;
Calas, M.; Vial, H. Exploration of potential prodrug approach of the
bis-thiazolium salts T3 and T4 for orally delivered antimalarials. Bioorg.
Med. Chem. 2010, 20, 3953−3956.
(16) Nicolas, O.; Margout, D.; Taudon, N.; Wein, S.; Calas, M.; Vial,
H. J.; Bressolle, F. Pharmacological properties of a new antimalarial
bisthiazolium salt, T3, and a corresponding prodrug, TE3. Antimicrob.
Agents Chemother. 2005, 49, 3631−3639.
(37) Calas, M.; Cordina, G.; Bompart, J.; BenBari, M.; Jei, T.;
Ancelin, M. L.; Vial, H. Antimalarial activity of molecules interfering
with Plasmodium falciparum phospholipid metabolism. Structure−
activity relationship analysis. J. Med. Chem. 1997, 40, 3557−3566.
(17) Greb, A.; Bitsch, R. Comparative bioavailability of various
thiamine derivatives after oral administration. Int. J. Clin. Pharm. Ther.
1998, 36, 216−221.
(18) Kawasaki, C. Modified Thiamine Compounds. Vitam. Horm.
1963, 21, 69−111.
(19) Jana, S.; Mandlekar, S.; Marathe, P. Prodrug Design to Improve
Pharmacokinetic and Drug Delivery Properties: Challenges to the
Discovery Scientists. Curr. Med. Chem. 2010, 17, 3874−3908.
(20) Brouwers, J.; Tack, J.; Augustijns, P. In vitro behavior of a
phosphate ester prodrug of amprenavir in human intestinal fluids and
in the Caco-2 system: Illustration of intraluminal supersaturation. Int. J.
Pharm. 2007, 336, 302−309.
(21) Yoshimi, A.; Hashizume, H.; Kitagawa, M.; Nishimura, K.;
Kakeya, N. Absorption Mechanism of 1,3-Bis(2-Ethoxycarbonylchro-
mon-5-Yloxy)-2-((S)-Lysyloxy)Propane Dihydrochloride (N-556), a
Prodrug for the Oral Delivery of Disodium-Cromoglycate. Biol. Pharm.
Bull. 1993, 16, 375−378.
(22) Oae, S.; Shinhama, K. Organic Thionitrites and Related
Substances - a Review. Org. Prep. Proced. Int. 1983, 15, 165−198.
(23) Ito, A.; Hamanaka, W.; Takagi, H.; Wada, T.; Kawada, K. S-
benzoylthiamine O-monophosphate and a process for preparing the
same. US Patent 3064000, Nov 13, 1962.
(24) Gaudino, J. J.; Wilcox, C. S. A Concise Approach to
Enantiomerically Pure Carbocyclic Ribose Analogues - Synthesis of
(4s,5r,6r,7r)-7-(Hydroxymethyl)Spiro[2.4]Heptane-4,5,6-Triol 7-O-
(Dihydrogen Phosphate). J. Am. Chem. Soc. 1990, 112, 4374−4380.
(25) Yoshikawa, M.; Kato, T.; Takenish, T. Studies of Phosphor-
ylation. 3. Selective Phosphorylation of Unprotected Nucleosides. Bull.
Chem. Soc. Jpn. 1969, 42, 3505−3512.
(26) Klein, E.; Nghiem, H. O.; Valleix, A.; Mioskowski, C.; Lebeau, L.
Synthesis of stable analogues of thiamine di- and triphosphate as tools
for probing a new phosphorylation pathway. Chem.Eur. J. 2002, 8,
4649−4655.
(27) Mckenna, C. E.; Higa, M. T.; Cheung, N. H.; Mckenna, M. C.
Facile Dealkylation of Phosphonic Acid Dialkyl Esters by Bromo-
trimethylsilane. Tetrahedron Lett. 1977, 18, 155−158.
4628
dx.doi.org/10.1021/jm3000328 | J. Med. Chem. 2012, 55, 4619−4628