Synthesis of α-substituted fosmidomycin analogues as highly potent Plasmodium falciparum growth inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.12.082

In view of the promising antimalarial activity of fosmidomycin or its N-acetyl homologue FR900098, the objective of this work was to investigate the influence of aromatic substituents in the α-position of the phosphonate moiety. The envisaged analogues we

Full text of DOI:10.1016/j.bmcl.2005.12.082

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1888–1891
Synthesis of a-substituted fosmidomycin analogues as highly potent
Plasmodium falciparum growth inhibitors
Timothy Haemers,a Jochen Wiesner,b Sara Van Poecke,a Jan Goeman,c
Dajana Henschker,b Edwald Beck,d Hassan Jomaab and Serge Van Calenbergha,*
aLaboratory for Medicinal Chemistry (FFW), Ghent University, Harelbekestraat 72, 9000 Gent, Belgium
bUniversita¨tsklinkum Giessen und Marburg, Institut fu¨r Klinische Chemie und Pathobiochemie,
Gaffkystrasse 11, 35392 Giessen, Germany
cLaboratory of Organic and Bioorganic Chemistry, Department of Organic Chemistry, Faculty of Sciences,
Ghent University, Krijgslaan 281 (S4), 9000 Gent, Belgium
dBiochemisches Institut, Justus-Liebig-Universita¨t Giessen, Friedrichstrasse 24, 35392 Giessen, Germany
Received 28 November 2005; revised 23 December 2005; accepted 28 December 2005
Available online 24 January 2006
Abstract—In view of the promising antimalarial activity of fosmidomycin or its N-acetyl homologue FR900098, the objective of this
work was to investigate the influence of aromatic substituents in the a-position of the phosphonate moiety. The envisaged analogues
were prepared using a linear route involving a 3-aryl-3-phosphoryl propanal intermediate. The activities of all compounds were eval-
uated on Eschericia coli 1-deoxy-D-xylulose 5-phosphate reductoisomerase and against two Plasmodium falciparum strains. Com-
pared with fosmidomycin, several analogues displayed enhanced activity towards the P. falciparum strains. Compound 1e with a
3,4-dichlorophenyl substitution in the a-position of fosmidomycin emerged as the most potent analogue of this series. It is approx-
imately three times more potent in inhibiting the growth of P. falciparum than FR900098, the most potent representative of this class
reported so far.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Malaria is estimated to kill more than one million peo-
ple annually and possibly as many as three million, with
most of the deaths among children below age six living
in sub-Saharan Africa. As a result of the resurgence of
malaria and increased resistance to prevalent anti-mala-
rials such as chloroquine, there is an urgent need for new
efficient chemotherapeutics against this disease.1 Drug
development has shifted towards targeting specific pro-
teins that are unique and critical for cellular growth
and survival of the parasite.2,3 Recently, the presence
of a mevalonate-independent pathway for isoprenoid
biosynthesis in Plasmodium falciparum was discov-
ered.4,5 1-Deoxy-D-xylulose 5-phosphate (DOXP)
reductoisomerase plays an essential role in this non-
mevalonate pathway, which is absent in humans. Previ-
ous studies demonstrated that fosmidomycin exerts
potent antimalarial activity by inhibition of DOXP
reductoisomerase (DXR), the second enzyme in the
reaction cascade.6,7 In recent clinical trials conducted
in Gabon and Thailand, fosmidomycin proved to be effi-
cient in the treatment of patients suffering from acute,
uncomplicated P. falciparum malaria.8,9 Fosmidomycin
has the advantage to be remarkably nontoxic and to
exhibit activity against multiresistant parasite strains.
Limitations are the short plasma half-life and the mod-
erate resorption rate.1 The acetyl derivative of fosmido-
mycin, FR900098, was shown to be approximately twice
as active against P. falciparum in vitro as well as in a
P. vinckei mouse model (Fig. 1).
Fosmidomycin also represents a valuable lead for fur-
ther modification. In order to study the structure–activ-
ity relationships, hydroxamic moiety modifications,
including benzoxazolone and oxazolopyridinone func-
tionaltities, have been reported.10 Also, the phosphonate
moiety has been altered to produce prodrugs with im-
proved oral bioavailability.11–13
Keywords: Fosmidomycin; 2C-Methyl-D-erithrytol 4-phosphate path-
way; 1-Deoxy-D-xylulose-5-phosphate Reductoisomerase.
Interestingly, modifications addressing the three carbon
spacer are scarce. The objective of this work was to
*
Corresponding author. Tel.: +32 (0)9 264 81 24; fax: +32 (0)9 264 81
46; e-mail: serge.vancalenbergh@ugent.be
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.12.082

T. Haemers et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1888–1891
1889
R1
R1
OH
N
O
P
OH
OH
N
O
P
OH
H
H3C
OH
OH
(a)
O
O
O
O
Fosmidomycin
FR900098
(b)
(d)
P
P
EtO
EtO
OEt
OEt
EtO
EtO
O
P
O
3a,b
4a,b
OH
N
O
P
1: R = H
2: R = CH3
H
R
OH
R1
OH
O
R2
EtO
EtO
O
O
a: R1 = R2 = H
P
R2
b: R1 = Me; R2 = H
c: R1 = OMe; R2 = H
d: R1 = Cl; R2 = H
e: R1 = R2 = Cl
7a-e
R1
O
a: R1 = H; R2 = H
R1
b: R1 = Me; R2 = H
c: R1 = OMe; R2 = H
d: R1 = Cl; R2 = H
e: R1 = Cl; R2 = Cl
R2
6c-e
(c)
Figure 1. Structures of fosmidomycin, FR900098 and analogues under
study.
O
H
R1
synthesize a series of fosmidomycin or FR900098
analogues containing a phenyl moiety in the a-position.
To sort out the influence of lipophilicity and electronic
properties of this phenyl moiety, substituents were intro-
duced according to Topliss’ methodology.14 Briefly, in
this methodology an operational scheme is used to
quickly identify the optimum substitution on a benzene
ring for maximizing drug potency by virtue of resulting
changes in hydrophobic, electronic and steric effects.
R2
5c-e
Scheme 2. Reagents and conditions: (a) i—n-BuLi, THF, À50 to
À70 ꢁC, ii—allyl bromide, À70 ꢁC; (b) i—OsO4, 4-methyl-morpholine
N-oxide, dioxane, ii—NaIO4; (c) triethyl phosphite, phenol, 100 ꢁC;
(d) 2 N HCl, rt.
The remainder of the synthesis is depicted in Scheme 3.
Treatment of 7a–e with O-benzylhydroxylamine yielded
(67–92%) oximes 8a–e, which were reduced with sodium
cyanoborohydride to produce benzyloxyamines 9a–e in
91–96% yield. Subsequent acylation afforded 10c,d,e
and 11a–e in good yield. Benzyl deprotection by catalyt-
ic hydrogenation gave 12c,e and 13a–e, which occurred
as mixtures of two hydroxamic acid isomers (syn and
anti). Generally, hydrogenolysis of the benzyl group
was accompanied by partial deoxygenation, while
attempted deprotection of compound 10d using hydrog-
enolysis or BCl3 led to loss of the formyl moiety. Com-
pounds 12c,e and 13a–e were finally deprotected with
TMSBr in CH2Cl2 to afford pure 1c,e and 2a–e after
purification by reversed-phase HPLC.18
Retrosynthetic analysis towards the synthesis of the de-
sired analogues is depicted in Scheme 1. 3-Aryl-substi-
tuted 3-phosphoryl propanal was anticipated to be a
convenient precursor.
Two synthetic pathways towards this aldehyde synthon
were explored (Scheme 2). The first route started from
allyl bromide, which upon reaction with the appropriate
diethyl benzylphosphonate in the presence of n-BuLi,
afforded 4a,b in 97 and 33% yield.15 Oxidation of 4a,b
to the vicinal cis-diol, followed by sodium periodate
cleavage, gave aldehydes 7a,b.
When the desired benzylphosphonate was not commer-
cially available, an alternative strategy to prepare the de-
sired aldehydes was followed.16 A 1,4-addition of
triethyl phosphite to the appropriately substituted cin-
namaldehyde in the presence of phenol gave the acetals
6c–e in 70–85% yield. Subsequent deprotection afforded
in 76–83% yield the corresponding aldehydes, which
were stable enough to be purified by flash chromatogra-
phy. If necessary, substituted cinnamaldehydes were
synthesized. Several procedures are described in the lit-
erature. In our hands, a palladium-catalyzed synthesis
from acrolein diethyl acetal and the corresponding aryl
iodide was very efficient.17
All final compounds were tested for inhibition of recom-
binant Escherichia coli DXR. The assay used was based
on the photometric measurement of the NADPH turn-
over associated with the DXR catalysed reaction.7 In
addition, the in vitro antimalarial activity of the com-
pounds was determined. Intraerythrocytic stages of the
P. falciparum strains D2d or 3D7 were incubated with
serial dilutions of the drugs and the viability of the par-
asites assessed by their ability to incorporate [3H]hypo-
xanthine into DNA.19 Fosmidomycin and FR900098
were included as reference compounds.
Compared with fosmidomycin, all investigated
analogues were weaker inhibitors of E. coli DXR
(Table 1). Analysis of the order of DXR inhibitory
HO
HO
Ar
EtO
EtO
Ar
O
O
P
R
P
N
O
NH
potency
(2d ꢀ 2e > 2a > 2b > 2c) indicates that the activity is
mainly +r-controlled
in
the
N-acetyl
series
2a–e
OH
OBn
1 or 2
(4-Cl ꢀ 3,4-diCl > H >
EtO
EtO
EtO
Ar
OBn
4-Me > 4-MeO). Although generally weaker than fosm-
idomycin in inhibiting E. coli DXR, all a-substituted
analogues studied surpass the activity of fosmidomycin
to inhibit the parasite growth. Remarkably, the formyl
O
O
P
N
P
O
EtO
Ar
H
Scheme 1. Retrosynthetic route toward analogues 1 and 2.

1890
T. Haemers et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1888–1891
EtO
EtO
O
OBn
H
EtO
EtO
O
P
N
P
OBn
(a)
(b)
N
7a-e
H
R1
R1
R2
R2
8a-e
9a-e
(c)
HO
HO
O
EtO
O
EtO
EtO
O
P
R
P
R
P
R
EtO
(e)
(d)
N
O
N
O
N
O
OH
OH
OBn
R1
R1
R1
R2
R2
R2
1c,e (R = H)
2a-e (R = CH3)
12c,e (R = H)
13a-e (R = CH3)
10c,d,e (R = H)
11a-e (R = CH3)
Scheme 3. Reagents and conditions: (a) O-benzylhydroxylamine, pyridine, EtOH, rt; (b) NaCNBH3, MeOH, HCl, rt; (c) acetyl chloride, CH2Cl2,
EtN3, 0 ꢁC or carbonyldiimidazole, HCOOH, CH2Cl2, rt; (d) H2, Pd/C, MeOH, rt; (e) TMSBr, CH2Cl2, rt.
Table 1. IC50 values against recombinant Escherichia coli DXR and Dd2 and 3D7 Plasmodium falciparum strains
HO
HO
O
P
R
N
O
OH
R1
R2
Compound
R
R1
R2
IC50 (lM) E. coli DXRa
IC50 (lM) Dd2
IC50 (lM) 3D7
2a
2b
CH3
CH3
H
H
H
H
H
H
H
Cl
Cl
0.311 0.120
0.396 0.061
0.156 0.043
0.459 0.109
0.099 0.026
0.059 0.020
0.119 0.019
0.030 0.008
0.030 0.008
0.35
0.22
0.20
0.27
0.095
0.028
0.090
0.36
0.18
0.55
0.95
0.36
0.85
0.35
0.090
0.25
1.1
Me
OMe
OMe
Cl
1c
2c
CH3
CH3
H
2d
1e
Cl
Cl
2e
Fosmidomycin
FR900098
CH3
0.32
a Mean values standard deviation of 3–5 independent measurements.
analogues 1c and 1e consistently outperformed the ace-
tyl derivatives 2c and 2e, both in the enzyme and the
parasite growth inhibition assay. An opposite trend
was observed for the fosmidomycin/FR900098 couple
in the parasite growth inhibition assay, in accordance
with previous studies.4 Interestingly, a worthwhile cor-
relation was observed between the IC50 values of 1c, 1e
and 2a–e for DXR and their IC50 for the malaria
strains, indicating that E. coli DXR inhibition is a use-
ful predictor to estimate the in vitro antimalarial activ-
ity. However, this correlation only seems to hold when
considering a set of closely related analogues. Indeed,
while fosmidomycin is clearly superior to the a-substi-
tuted analogues as DXR inhibitor, it is less active in
inhibiting P. falciparum growth.
In summary, a practical method for the synthesis of
a-aryl-substituted fosmidomycin analogues was devel-
oped. Several analogues were superior to fosmidomycin
in inhibiting the growth of malaria parasites.
Acknowledgments
This study was supported by grants from the European
Commission (QLK2-CT-2002-00887) and INTAS
(03-51-4077).
References and notes
1. Wiesner, J.; Ortmann, R.; Jomaa, H.; Schlitzer, M. Angew.
Chem., Int. Ed. 2003, 42, 5274.
2. Breman, J. G. Am. J. Trop. Med. Hyg. 2001, 64, 1.
3. Winstanley, P. A.; Ward, S. A.; Snow, R. W. Microbes
Infect. 2002, 4, 157.
4. Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.;
Weidemeyer, C.; Hintz, M.; Turbachova, I.; Eberl, M.;
Zeidler, J.; Lichtenthaler, H. K.; Soldati, D.; Beck, E.
Science 1999, 285, 1573.
5. Wiesner, J.; Hintz, M.; Altincicek, B.; Sanderbrand, S.;
Weidemeyer, C.; Beck, E.; Jomaa, H. Exp. Parasitol 2000,
96, 182.
Compound 1e emerged as the most promising analogue
of this series. Its in vitro antimalarial activity indicated
that it is 12-fold more active than fosmidomycin and
also exceeds the activity of FR900098, the most potent
analogue known to date. Apparently, the lipophilic
and electronegative properties of the 3,4-dichloro-sub-
stitution pattern selectively favour the interaction with
the P. falciparum DXR homologue. Alternatively, the
aromatic ring in the a-position may improve the capac-
ity of the compounds to enter the parasite cells.

T. Haemers et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1888–1891
1891
6. Zeidler, J.; Schwender, J.; Muller, C.; Wiesner, J.; Weide-
meyer, C.; Beck, E.; Jomaa, H. Z. Naturforsch, C: Biosci.
1998, 53, 980.
16. Harvey, R. G. Tetrahedron 1966, 22, 2561.
17. Battistuzzi, G.; Cacchi, S.; Fabrizi, G. Org. Lett. 2003, 5,
777.
7. Kuzuyama, T.; Shimizu, T.; Takahashi, S.; Seto, H.
Tetrahedron Lett. 1998, 39, 7913.
8. Missinou, M. A.; Borrmann, S.; Schindler, A.; Issifou, S.;
Adegnika, A. A.; Matsiegui, P. B.; Binder, R.; Lell, B.;
Wiesner, J.; Baranek, T.; Jomaa, H.; Kremsner, P. G.
Lancet 2002, 360, 1941.
9. Lell, B.; Ruangweerayut, R.; Wiesner, J.; Missinou, M. A.;
Schindler, A.; Baranek, T.; Hintz, M.; Hutchinson, D.;
Jomaa, H.; Kremsner, P. G. Antimicrob. Agents Chemo-
ther. 2003, 47, 735.
10. Courtois,M.;Mincheva,Z.;Andreu,F.;Rideau,M.;Viaud-
Massuard, M. C. J.EnzymeInhib.Med.Chem. 2004,19, 559.
11. Ortmann, R.; Wiesner, J.; Reichenberg, A.; Henschker,
D.; Beck, E.; Jomaa, H.; Schlitzer, M. Arch. Pharm. 2005,
338, 305.
18. All final compounds were purified using a preparative
HPLC system on a C18 column (5 lm; Phenomenex;
Luna; 250 · 21.2 mm) with a linear gradient of aceto-
nitrile in 5 mM NH4OAc solution over 20 min at a
flow rate of 17.5 mL/min. The purity of all target
compounds was assessed by analytical HPLC (5 lm;
Phenomenex; C18(2); 250 · 4.6 mm) using the same
gradient at a flow rate of 1 mL/min. Spectral data for
analogue 1e: 1H NMR (300 MHz; D2O) d = 2.07 (m,
1H, b-CH); 2.31 (m, 1H, b-CH); 2.80 (m, 1H, a-CH);
3.32 (m, 2H, c-CH2); 7.10 (m, 1H, arom. H); 7.36 (m,
2H, arom. H); 7.44 and 8.07 (2· s, 1H, major and
minor HC@O) ppm. 13C NMR (75 MHz; D2O)
1
d = 26.49 (s, b-CH2); 42.82 (d, a-CH, JPC = 129.6 Hz);
48.86
(d,
c-CH2,
3JPC = 17.0 Hz);
128.92
(d,
12. Ortmann, R.; Wiesner, J.; Reichenberg, A.; Henschker,
D.; Beck, E.; Jomaa, H.; Schlitzer, M. Bioorg. Med. Chem.
Lett. 2003, 13, 2163.
13. Reichenberg, A.; Wiesner, J.; Weidemeyer, C.; Dreiseidler,
E.; Sanderbrand, S.; Altincicek, B.; Beck, E.; Schlitzer, M.;
Jomaa, H. Bioorg. Med. Chem. Lett. 2001, 11, 833.
14. Topliss, J. G. J. Med. Chem. 1972, 15, 1006; Topliss, J. G.
J. Med. Chem. 1977, 20, 463.
JPC = 5.8 Hz, @CH); 130.04 (d, JPC = 3.8 Hz, @C) ;
130.55 (d, JPC = 2.6 Hz, @CH); 130.73 (d, JPC = 6.0 Hz,
@CH); 131.88 (d, JPC = 3.2 Hz, @C); 138.80 (d,
JPC = 7.2 Hz, @C); 159.70 and 163.76 (2· s, major
and minor C@O) ppm. 31P NMR (121 MHz; D2O)
d = 21.46 and 21.78 (major and minor isomer); HRMS
(ESI-MS) [MÀH]À found, 325.9745; calcd, 325.9751.
19. Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.;
Chulsay, J. D. Antimicrob. Agents Chemother. 1979, 16,
710.
15. Flitsch, W.; Lubisch, W.; Rosche, J. Liebigs Ann. Chem.
1987, 8, 661.