Synthesis and evaluation of phosphonated N-heteroarylcarboxamides as DOXP-reductoisomerase (DXR) inhibitors

Article abstract of DOI:10.1016/j.bmc.2010.11.062

The diethyl esters and disodium salts of a range of heteroarylcarbamoylphosphonic acids have been prepared and evaluated as analogues of the highly active DOXP-reductoisomerase (DXR) inhibitor, fosmidomycin. Computer-simulated docking studies, Saturation Transfer Difference (STD) NMR analysis and enzyme inhibition assays have been used to explore enzyme-binding and -inhibition potential, while in silico analysis of the DXR active site has highlighted the importance of including a well-parameterised metal co-factor in docking studies and has revealed the availability of an additional binding pocket to guide future drug design.

Full text of DOI:10.1016/j.bmc.2010.11.062

Bioorganic & Medicinal Chemistry 19 (2011) 1321–1327
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of phosphonated N-heteroarylcarboxamides
as DOXP-reductoisomerase (DXR) inhibitors
Taryn Bodill , Anne C. Conibear , Gregory L. Blatch b,c, Kevin A. Lobb a,c, Perry T. Kaye a,c,
c
a
⇑
a
Department of Chemistry, Rhodes University, Grahamstown 6140, South Africa
Biomedical Biotechnology Research Unit and Department of Biochemistry, Rhodes University, Grahamstown 6140, South Africa
Centre for Chemico- and Biomedicinal Research, Rhodes University, Grahamstown 6140, South Africa
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Thediethylestersanddisodiumsaltsofarangeofheteroarylcarbamoylphosphonicacidshavebeenprepared
and evaluated as analogues of the highly active DOXP-reductoisomerase (DXR) inhibitor, fosmidomycin.
Computer-simulated docking studies, Saturation Transfer Difference (STD) NMR analysis and enzyme inhi-
bition assays have been used to explore enzyme-binding and -inhibition potential, while in silico analysis of
the DXR active site has highlighted the importance of including a well-parameterised metal co-factor in
docking studies and has revealed the availability of an additional binding pocket to guide future drug design.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 22 September 2010
Revised 22 November 2010
Accepted 25 November 2010
Available online 2 December 2010
Keywords:
Anti-malarial
Phosphonates
DOXP-reductoisomerase
Enzyme inhibitors
In silico docking
Saturation Transfer Difference NMR
7
1
. Introduction
Malaria continues to be an enormous health-threat in the devel-
development of novel and effective antimalarials. Although few of
the DXR inhibitors described in the literature have been more effec-
tive than fosmidomycin, they provide useful structure–activity rela-
tionships to guide the design of new inhibitors. Deng et al. have
highlighted the importance of a metal-coordinating group, espe-
cially nitrogen- and oxygen-containing functionalities which are
able to bind to thehard Mg ion. Thephosphonate moietyoffosm-
idomycin 2 is more biologically stable than the corresponding phos-
phate of DOXP 1 and the importance of the double negative charge
on the phosphonate group has also been demonstrated.
In our laboratories, attention has been given to designing and
synthesising fosmidomycin 2 analogues as potential DXR inhibi-
oping world. Treatment is compounded by the phenomenon of drug
resistance, and the development of novel therapeutics has become
a research priority.
responsible for the most dangerous form of human malaria, and
the enzyme, 1-deoxy-1-D-xylulose 5-phosphate (1; Fig. 1) reducto-
isomerase (PfDXR) has recently been validated as a target for the de-
sign of potential antimalarial drugs.4 This enzyme is involved in a
parasite-specific, isoprenoid biosynthetic DOXP/MEP pathway, for
1
–3
Plasmodium falciparum (Pf) is the parasite
2
+
10
,5
11,12
6,7
which the natural product, fosmidomycin 2, and its acetyl ana-
logue, FR900098 3, have been shown to act as inhibitors. Fosmido-
6
,7
9
tors, and we have recently reported the synthesis of a series of
mycin 2 was isolated from Streptomyces lavendulae in the 1970s and
phosphonated N-phenylcarboxamides 4. We now describe: (i) the
preparation of novel N-heteroarylamino-2-oxoethylphosphonate
esters 8a–e and the disodium salts 11a–e of the corresponding
phosphonic acids 9a–e (Scheme 1); (ii) the use of EcDXR in STD
NMR ligand-binding and enzyme inhibition studies; (iii) the results
of in silico ligand-docking studies; and (iv) progress in the topolog-
ical analysis of the enzyme active site.
found to have antibiotic and herbicidal properties.5
,6,8,9
The DOXP/
MEP pathway has also been found in several pathogenicspecies such
as Helicobacter pylori, Mycobacterium tuberculosis, Escherichia coli
(
Ec) and P. falciparum but, since it is not found in humans, it has
8
obvious potential as a target for chemotherapy. In humans, fosmi-
domycin 2 has been shown to be active against uncomplicated
P. falciparum malaria, but its efficacy as a clinical drug is compro-
mised by poor absorption, short plasma half-life and the high
observed rate of recrudescence.7,8 Nevertheless, the high levels of
DXR inhibition exhibited by fosmidomycin 2 and FR900098 3
have made these compounds particularly promising leads for the
2
2
. Results and discussion
.1. Chemistry
Deng et al.10 have highlighted the role of metal-coordinating
groups (especially nitrogen- or oxygen-containing functionalities)
⇑
Corresponding author. Tel.: +27 46 6037030; fax: +27 46 6225109.
E-mail address: P.Kaye@ru.ac.za (P.T. Kaye).
0
968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.11.062

1
322
T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
OH
OH
N
OH
OH
N
H
PO H
3
2
O
OH
O
O P O-
O-
O
O
P
O
NH
-
O
-
1
2
3
4
O
O
P OEt
OEt
Figure 1. Structures of DOXP 1 and known and putative DXR inhibitors 2–4.
O
O
O
O
O
O
OEt
iii
X
OH
Cl
i
Cl
ii
P
P
RNH2
a-e
+
Cl
RHN
RHN
O
RHN
RHN
OEt
OH
5
6
7a-e
8a-e
9a-e
iv
O
P
O
O
OSiMe3
OSiMe3
v
ONa
P
RHN
ONa
10a-e
11a-e
NH2
N
N
S
RNH2
=
;
;
;
NH
2
;
NH
2
N
NH2
N
O
S
O
NH2
O
a
b
c
d
e
Scheme 1. Synthesis of phosphonate esters 8a–e and phosphonic acid salts 11a–e. Reagents and conditions: (i) NaH, chloroacetyl chloride 6, THF, 6 h, rt; (ii) triethyl
phosphite, 9 h, reflux; (iii) TMSBr (2 equiv), CH CN, microwave; (iv) TMSBr, DCM, overnight, rt; (v) 0.4 M-NaOH.
3
in binding to the hard Mg2+ cation in the DXR active site, and the
heterocyclic moieties in compounds 8a–e and 9a–e were expected
serve this function. Ester pro-drugs of fosmidomycin analogues
have been shown to exhibit increased lipophilicity and absorp-
intermediates with aqueous sodium hydroxide afforded the diso-
dium salts 11a–e, two of which (11a and 11c) required purification
by reverse-phase semi-preparative HPLC.
1
1–13
tion,
and the phosphonate esters 8a–e might well provide
2.2. Computer modelling
similar advantages. Access to these compounds and the disodium
salts 11a–e of the corresponding phosphonic acids 9a–e is outlined
in Scheme 1.
Thus, acylation of the respective primary amines 5a–e with
chloroacetyl chloride 6 afforded the corresponding chloroaceta-
mides 7a–e in yields in excess of 70% and, generally, in sufficient
purity to be used without further purification. The only exception
was the 2-aminopyridine derivative 7a, which was chromato-
graphed prior to use. Phosphonation of the chloroacetamides
Simulated docking studies were undertaken to explore the
binding affinity of the ligands 8a–e and 9a–e and the topology of
the binding site. An in vacuo global minimum was located for each
of the ligands following a systematic conformational search using
Cerius . Solvent-corrected, energy-minimised structures (includ-
ing frequency calculations) were then obtained at the DFT level
2
21
2
2
23
using Gaussian 03. AutoDock version 4.0 was used for the dock-
2
4
ing studies, Discovery Studio Visualiser to visualise the docked
2
5
7
a–e was effected via the Arbuzov reaction using triethyl phos-
conformations and UCSF Chimera to explore the active site. Pre-
liminary in silico docking of the phosphonate esters 8a–e, the
phosphonic acids 9a–e and the corresponding phosphonate dia-
nions was explored using the active site (from which fosmidomy-
cin 2 had been removed) in the crystal structure 1Q0L of EcDXR
(which includes bound fosmidomycin 2 and NADPH). In order to
phite, and the diethyl phosphonates 8a–e were readily isolated
in moderate yields (48–67%). In each case, 1H NMR analysis re-
vealed disappearance of the chloromethylene singlet at ca.
2
6
4
3
.2 ppm and the appearance of a methylene proton doublet at ca.
ppm, reflecting the distinctive coupling (J = ca. 21 Hz) to the adja-
3
1
2+
cent P nucleus. Since hydrolysis of the phosphonate esters 8a–e
simulate coordination to the divalent metal cation, Mn was in-
1
2,14–
to the corresponding acids 9a–e is expected to occur in vivo,
cluded in the model and positioned according to its location rela-
1
6
the latter compounds were also required for in vitro enzyme-
inhibition assays. Microwave-assisted hydrolysis of the esters in
tive to fosmidomycin 2 and the coordinating ligands Asp150,
Glu231 and Glu152 in the EcDXR crystal structure 1ONP.27
A
1
7
acetonitrile —a method successfully used on similar com-
pounds —proved unsuccessful here. Consequently, the esters were
charge of +2 was manually assigned to the manganese ion and
the fosmidomycin was removed from the active site, as were the
solvating water molecules. The active-site residues Ser186,
Asn227, Lys228 and Glu231 were defined as flexible residues and
simulated dockings were carried out with each of the ligands
8a–e and 9a–e as well as with DOXP 1, fosmidomycin 2,
FR900098 3 and the biosynthetic product, 2-C-methyl-D-erythritol
4-phosphate (MEP). Examination of docked conformations of each
9
simply stirred overnight with bromotrimethylsilane, followed by
hydrolysis of the TMS-substituted intermediates 10a–e, as de-
scribed by McKenna et al.18 This approach proved successful
19
although extension of the reported reaction time was necessary.
3
1
(
A
P NMR-based kinetics study has provided useful insights into
20
the silylation mechanism. ) Titration of the TMS-substituted

T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
1323
of the ligands relative to fosmidomycin 2 and the proximal amino
acid residues in the active site of the EcDXR crystal structure
1
tion distances to be determined. While the conformations of the
majority of the docked ligands appeared to differ little from their
minimum energy conformers, the solvent-corrected energies cal-
ion, close to NADPH, while the phosphonate group occupies a
2
6,28
well-defined phosphonate-binding pocket
—an orientation
Q0L permitted potential hydrogen-bonding and Mn2+-coordina-
2
6
which is essential for the conversion of DOXP to MEP. However,
this does not preclude competitive ‘reverse’ binding of the natural
substrate (DOXP) nor would it necessarily prevent a ‘reverse’-
bound ligand from acting as an inhibitor! Although the ‘reverse’-
binding mode is chemically feasible, Cheng and Oldfield have
pointed out that the resultant four-membered ring would be more
strained than the five-membered ring formed by ‘normal’ coordi-
nation of the hydroxamate moiety in fosmidomycin 2 to the metal
culated for the docked ligands were all higher (by 63 ± 15 kcal -
À1
mol
) than their undocked conformers—the conformational
energy cost of binding (Table 1).
In most cases, the ligands exhibited, unexpectedly, docking ori-
entations opposite to that of fosmidomycin 2 (as illustrated in
Fig. 2). In fact, in all of the crystal structures which include bound
fosmidomycin, the hydroxamate moiety coordinates to the metal
2
9
cation.
It was observed that a region adjacent to the phosphonate-bind-
ing pocket and lined by the residues, His257, Thr184, Glu152,
Table 1
a
Conformational binding cost, and STD NMR binding, enzyme inhibition and IC50 data for ligands 8a–e and 11a–e using recombinant EcDXR
a
À1
Enzyme inhibitionb,c/%
Compound
Conformational binding cost /kcal mol
STD result
IC50
472
1171
730
/lM
8a
31.92
54.45
67.10
81.01
66.08
+
+
+
+
+
+
+
+
+
+
58.5
11.7
28.9
67.7
21.6
8b
8c
8d
408
8e
1286
d
e
e
1
1a
1b
1c
1d
1e
84.47
À
À
1
1
1
1
45.02d
12.2
7.4
>1500
>1500
1027
d
44.79
47.06
d
30.0
16.6
41.60d
>1500
a
b
c
Energy between solvent-corrected, energy-minimised and docked conformers.
At ligand concentration of 500 M. At 0.3 M, fosmidomycin 2 exhibits 99.3% inhibition of EcDXR.
Activity in the absence of inhibitor set at 100%.
For the corresponding phosphonic acids 9a–e.
Anomalous result suggesting enzyme activation.
l
l
d
e
Figure 2. Docked conformation of ligand 9b in the EcDXR active site (1Q0L) with Mn2+ added, showing ‘reverse’ binding of the ligand relative to fosmidomycin 2. Protein
active-site residues are shown in wireframe coloured by atom type, NADPH as green sticks, Mn2 as a purple sphere, and the ligand as sticks coloured by atom type. The
crystal structure conformation of fosmidomycin 2 is shown in yellow ball-and-stick format. Hydrogen atoms have been omitted for clarity and hydrogen-bonding distances
+
(
in Å) are shown as yellow dashed lines.

1
324
T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
Trp212 and the backbone atoms of Ser254 and Gly185, was regu-
larly occupied by the heterocyclic moieties of the docked ligands
suggests that attention should be given to including a well-param-
eterised and accurately charged metal cation when carrying out
docking studies on DXR.
(
as illustrated in Fig. 3). Interestingly, some of the docking results
for fosmidomycin 2 also indicate occupation of this region—an
observation which could prove useful in the design of new inhibi-
tors. In fact, this region may well be occupied by the highly active
2.3. Saturation Transfer Difference (STD) NMR experiments
30
a
al.
-aryl phosphonates reported by Haemers et al. and by Kurz et
STD NMR techniques have proved to be useful in qualitatively
differentiating binding and non-binding ligands simultaneously
3
1
Little attention appears to have been given to placing the M2+
in the presence of the enzyme. Saturation of the protein is trans-
37
cation in the protein model32 and to assigning a realistic charge
to the metal centre. Typically, in docking studies which actually in-
clude the metal cation, Mg2 has been used and has either been as-
signed a full +2 charge or the docking programme has been
allowed to calculate a charge.
Perruchon et al. state that Mg is better parameterised than
ferred to bound ligands resulting in an increase in their signal
intensities; non-binding ligands are unaffected and subtraction of
the reference spectrum from the saturation spectrum results in po-
sitive signals for the binding ligands alone. STD experiments were
+
2
9,31
33
37
Although Silber et al. and
conducted, as described by Mayer and Meyer, to screen the esters
3
4
2+
8a–e and the disodium salts 11a–e for binding to EcDXR (Table 1).
The ligands all showed binding to EcDXR, with minor, apparent
2
+
35
Mn in AutoDock, a Gasteiger charge of 0.00 is automatically as-
signed to either metal ion in AutoDock 4.0. Neither a charge of +2
nor 0 is likely to model the metal co-factor realistically in the pres-
ence of the surrounding, negatively charged, coordinating ligands.
Consequently, in the present study, the Mg2+ cation, its coordinat-
ing residues (Asp149, Glu151 and Glu230), a water molecule and
fosmidomycin 2, were removed as a group from the crystal struc-
1
anomalies being attributed to differences in T relaxation times
3
8
or distance-dependent saturation transfer. Although the STD data
provide an indication of protein–ligand-binding, the possibility of
non-specific binding in regions other than the active site cannot
3
9
be excluded.
2
8
ture 2EGH,
Hartree–Fock calculation.22 This calculation afforded a charge of
1.058 on the Mg cation—a value consistent with significant dissi-
and the group was energy-minimised, using a
2.4. Enzyme-inhibition bioassays
+
Preliminary screening of the ability of the ligands 8a–e and
11a–e to inhibit EcDXR was conducted using an enzyme assay
based on the spectrophotometric measurement of the conversion
of NADPH to NADP which occurs when DOXP is converted to MEP
pation the formal +2 charge by the coordinating ligands. The calcu-
lated Mulliken charges on all of the atoms in the energy-minimised
group were transferred to the input files used by AutoDock and,
when fosmidomycin 2 (with its Hartree–Fock calculated charges)
was docked into the modified protein model, an increase in the
number of correctly oriented dockings was observed. The impor-
3
4,40,42
by DXR.
The ligands were tested at concentrations between
M and the 500 M data (Table 1) was used to as-
10 M and 500
l
l
l
sess their inhibition potential. While the % enzyme inhibition lev-
els are relatively low and the IC50 concentrations correspondingly
1
0,36
tance of substrate or inhibitor binding to the metal cation
Figure 3. Docked conformation of ligand 9d in EcDXR (1Q0L) showing pocket in front of phosphonate-binding site occupied by the furan ring. The protein surface at 6.5 Å
from fosmidomycin 2 is shown with 40% transparency, coloured to indicate hydrophilicity (blue) or hydrophobicity (orange) and clipped in front to expose the open pocket.
Protein residues are shown in wireframe, coloured by atom type, the crystal structure fosmidomycin 2 in ball-and-stick format coloured by atom type and ligand 9d and
NADPH as sticks coloured by atom type. Hydrogen atoms have been omitted for clarity.

T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
1325
high, several of the ligands exhibit IC50 values in the micromolar
range, the most active of these being the furan derivative 8d
decanted and stirring and decanting was repeated with three fur-
ther aliquots of hexane. Evaporation of residual hexane and triethyl
phosphite in vacuo afforded the desired ester 8.
(
67.7% enzyme inhibition at 500 lM and IC50 = 408 lM). The data
in Table 1 appear to reflect several structure–activity relationship
patterns.
4.1.1.1. Diethyl 2-[(pyridin-2-yl)carbamoyl]ethylphosphonate
+
8
a. As a dark brown oil (0.078 g, 12%) (Found: M , 272.091731.
(
i) The phosphonate esters 8 appear to exhibit significantly
lower IC50 values than the corresponding phosphonic acid
salts 11—due, perhaps, to their more hydrophobic character
and, hence, their relative preference for an enzymic rather
than the aqueous environment. Perruchon et al.3 explored
the effect of extending the phosphonate ester groups and
suggested that the negative effect of decreasing the charge
on the phosphonate moiety is partly counteracted by the
occupation of a nearby hydrophobic pocket by the alkyl ester
C
11
H
17
N
2
O
4
P requires M, 272.092596); d
), 3.04 (2H, d, JH–P = 21.0 Hz, CH
), 7.04 (1H, m, ArH), 7.69 (1H, m, ArH), 8.13
(1H, d, J = 8.4 Hz, ArH), 8.28 (1H, d, J = 4.8 Hz, ArH), 8.96 (1H, s,
NH); d (150 MHz; CDCl ), 36.9 (d, JC–P = 129.0 Hz,
) 16.4 (2 Â CH
CH ), 114.0 (ArCH), 120.0 (5-ArCH),
H
(600 MHz; CDCl
3
) 1.35
(6H, t, J = 6.6 Hz, 2 Â CH
3
2
P), 4.20
(4H, m, 2 Â OCH
2
4
C
3
3
2
P), 63.0 (d, JC–P = 6 Hz, 2 Â OCH
2
+
138.3 (ArCH), 148.0 (ArCH), 151.0 (C-2), 162.7 (C@O); m/z 272 (M ,
48%) and 94 (100).
3
4
groups. This explanation could account for the higher inhi-
bition shown by the phosphonate esters 8 compared to the
phosphonic acid salts 11.
4.1.1.2. Diethyl 2-[(isoxazol-3-yl)carbamoyl]ethylphosphonate
8b. As a dark brown oil which crystallised on standing (0.800 g,
+
48%), mp 60–66 °C (Found: M , 262.071312. C
9
H
15
2
N O
5
P requires
) 1.35 (6H, t, J = 7.2 Hz,
P), 4.20 (4H, m, 2 Â OCH ),
6.98 (1H, d, J = 1.8 Hz, ArCH), 8.25 (1H, d, J = 1.8 Hz, 5-ArCH), 9.95
(1H, s, NH); (150 MHz; CDCl ), 36.2 (d,
16.3 (2 Â CH
C–P = 130.5 Hz, CH ), 99.3 (ArCH),
(
ii) Compound 8e, which contains the bulkiest heteroaryl moi-
ety, showed no inhibition at 500 M, suggesting size con-
straints in the vicinity of the active site.
M, 262.071860); d
H 3
(600 MHz; CDCl
l
2 Â CH ), 3.10 (2H, d, JH–P = 21.0 Hz, CH
3
2
2
(
iii) The fact that the furan derivatives 8d and 11d appear to be
the most active ligands in their respective series—the former
most active overall—may be due the flexibility provided by
the furfuryl methylene group, which could increase the like-
lihood of the furan ring occupying the ‘front’ pocket. (In the
other ligands, delocalisation of the amide nitrogen lone-pair
into the heteroaryl ring would restrict their flexibility.) In
fact, this possibility is supported by the observation that
ligand 9d docks most consistently (albeit in reversed orien-
tation) with the furan ring occupying the ‘front’ pocket as
illustrated in Figure 3; in this orientation the furan oxygen
is suitably located to hydrogen-bond to the Ser151 or
Glu152 residues.
d
C
3
)
3
J
2
P), 63.2 (d, JC–P = 6 Hz, 2 Â OCH
2
+
157.1 (C-3), 158.7 (ArCH), 162.5 (C@O); m/z 262 (M , 67%) and
179 (100).
4.1.1.3. Diethyl 2-[(1,3-thiazol-2-yl)carbamoyl]ethylphospho-
nate 8c. As a brown oil (1.04 g, 59%) (Found: M , 278.047217.
+
C
9
H
15
N
2
O
4
PS requires M, 278.049017); d
(6H, t, J = 7.1 Hz, 2 Â CH ), 3.17 (2H, d, JH–P = 21.6 Hz, CH
(4H, m, 2 Â OCH ), 6.96 (1H, d, J = 3.0 Hz, ArCH), 7.47 (1H, d,
J = 3.0 Hz, ArCH); d (150 MHz; CDCl ), 35.6 (d,
) 16.2 (2 Â CH
C–P = 131.3 Hz, CH ), 113.6 (ArCH),
P), 63.1 (d, JC–P = 6 Hz, 2 Â OCH
137.2 (ArCH), 158.5 (C-2), 162.4 (C@O); m/z 278 (M , 50%) and 179
100).
H
(600 MHz; CDCl
3
) 1.32
3
2
P), 4.18
2
C
3
3
J
2
2
+
(
4
.1.1.4. Diethyl 2-[(furan-2-ylmethyl)carbamoyl]ethylphospho-
3
. Conclusions
+
nate 8d. As a red-brown oil (1.09 g, 62%) (Found: M , 275.090706.
P requires M, 275.092261); d (600 MHz; CDCl ) 1.29
6H, t, J = 7.2 Hz, 2 Â CH ) 2.85 (2H, d, JH–P = 20.7 Hz, CH P), 4.90
4H, m, 2 Â OCH ), 4.42 (2H, d, J = 5.7 Hz, CH N), 6.22 (1H, d,
C
(
(
11
H18NO
5
H
3
The phosphonate esters 8a–e and the corresponding phos-
phonic acid salts 11a–e have been successfully synthesised and
subjected to preliminary screening as DXR inhibitors using a com-
bination of computer-modelling, STD NMR and enzyme-inhibition
techniques. A number of the synthetic ligands have been shown to
exhibit both enzyme-binding and enzyme-inhibition activity,
while detailed investigation of the EcDXR active site has permitted
identification of an available pocket adjacent to the phosphonate-
binding region, which is now being explored in the de novo design
and development of a new series of potential ligands.
3
2
2
2
J = 3.0 Hz, 3-CH), 6.28 (1H, dd, J = 3.0 and 1.7 Hz, 4-CH), 7.17 (1H,
s, NH), 7.31 (1H, d, J = 1.7 Hz, 5-CH); d (150 MHz; CDCl ) 16.2
2 Â CH ), 35.0 (d, JC–P = 129.0 Hz, CH P), 36.7 (CH NH), 62.7 (d,
), 107.3 (ArCH), 110.3 (ArCH), 142.1 (ArCH),
C
3
(
3
2
2
J
1
C–P = 6 Hz, 2 Â OCH
2
+
51.0 (C-2), 163.8 (C@O); m/z 275 (M , 22%) and 96 (100).
4
.1.1.5. Diethyl 2-[(5-acetyl-4-methyl-1,3-thiazol-2-yl)carbam-
oyl]ethyl-phosphonate 8e. As a yellow-brown solid (1.44 g, 67%),
+
mp 151–154 °C [Found: (MÀ1) , 333.0667. C12
MÀ1, 333.0674); d (600 MHz; CDCl ) 1.38 (6H, t, J = 7.2 Hz,
 CH ), 2.42 (3H, s, 4-CH ), 2.49 (3H, s, COÁCH ), 3.19 (2H, d,
H–P = 22.2 Hz, CH ); d (150 MHz; CDCl
P), 4.27 (4H, m, 2 Â OCH
6.3 (2 Â CH CH ), 18.2 (CO.CH ), 30.2 (CH 35.7 (d,
C–P = 129.0 Hz, CH ), 125.5
P), 63.6 (d, JC–P = 7.5 Hz, 2 Â OCH
ArC), 155.4 (ArC), 158.4 (ArC), 162.6 (NHC@O), 190.6 (CH C@O);
19 2 5
H N O PS requires
4
4
. Experimental
.1. Synthesis
H
3
2
J
1
J
3
3
3
2
2
C
3
)
3
)
The a-chloroamides 7a–e, which are known, were prepared fol-
3
2
3
41
lowing a reported procedure. The phosphonate esters 8a–e and
the corresponding phosphonic acid salts 11a–e, which, with the
exception of compound 11a, are all new and were obtained as
follows.
2
2
(
3
m/z 333 (MÀ1, 100%).
4
.1.2. General procedure for the hydrolysis of phosphonate
1
9
4
.1.1. General procedure for the synthesis of the phosphonate
esters 8a–e
The -chloroamide starting material (7a–e) was placed in a dry,
round-bottomed flask under N and triethyl phosphite (5 equiv)
esters to afford the phosphonic acid salts 11a–e
The phosphonate ester 8 was dissolved in dry CH Cl2 (3.5 mL)
2
a
under N2 and TMSBr (4 equiv) was added dropwise. The mixture
was stirred overnight at room temperature and, after removal of
2
was added through a septum. The mixture was refluxed (oil bath
at 110 °C) for ca. 9 h. After cooling, hexane (20 mL) was added
and the mixture stirred for ca. 20 min. The hexane was then
the volatile material, in vacuo, the residue was dissolved in H O
(2 mL) and the solution titrated with 0.40 M-NaOH to a pH of ca.
8.0. The aqueous solution was added to MeOH (ca. 200 mL) and
2

1
326
T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
the resulting solution evaporated to dryness in vacuo. The crude
product was purified by reverse-phase column chromatography
ligands in the enzyme solution were 11b (748
11d (969 M) and 11e (782 M), and sodium 3-(trimethylsilyl)tet-
radeuteriopropanoate (TSP; 813 M) was included as a standard.
lM), 11c (748 lM),
l
l
on C18 cellulose, eluting with 1% NH
4
OH in MeOH.
l
The STD assay was optimised in terms of the number of pre-
saturation Gaussian pulses and the power level of the saturation
pulse. Saturation of the protein was achieved using a train of 4
Gaussian bell-shaped pulses separated by a 1 ms delay at a power
level of 40 dB. The frequency of the saturating on-resonance pulse
was at 0.73 ppm and the off-resonance pulse at 20 ppm. Phase
cycling between the on- and off-resonances was used to minimise
the effects of slight changes in temperature or magnetic field homo-
geneity. No spin-lock filter was used. A 3-9-19 water suppression
pulse was applied and 7168 scans were acquired. Spectra were sub-
tracted and processed using Topspin software from Bruker.
4
1
.1.2.1. Disodium 2-[(pyridin-2-yl)carbamoyl]ethylphosphonate
1a. As a red-brown solid, which was further purified by HPLC
18 column, 250 Â 10.00 mm; elution with H O),
3 mg, 2.3% from HPLC) and which decomposes above 300 °C
(
(
[
(
LUNA 10
l
C
2
+
Found: (MHÀNa
2
) , 215.0221. C
7
8
H N
2
O
4
P requires 215.0222]; d
H
600 MHz; D O) 2.79 (2H, d, JH–P = 18.6 Hz, CH
2
2
P), 7.23 (1H, t,
J = 6.0 Hz, ArH), 7.87 (2H, overlapping m, 2 Â ArH), 8.31 (1H, d,
J = 4.8 Hz, ArH); d (150 MHz; D O) 40.1 (d, JC–P = 109.5 Hz, CH P),
15.8 (d, ArCH), 120.8 (d, ArCH), 139.6 (d, ArCH), 147.7 (d, ArCH),
C
2
2
1
1
2
+
50.4 (s, 2-C), 171.9 (s, C@O); m/z 215 [(MHÀNa
2
) , 25%] and
18 (100).
4
.3. Expression and purification of EcDXR
4
.1.2.2. Disodium 2-[(isoxazol-3-yl)carbamoyl]ethylphospho-
nate 11b. As a dark brown solid (0.438 g, quantitative yield),
EcDXR was expressed and purified according to standard proce-
+
40,42
mp 283–285 °C (Found: MH , 250.9806. C
M+1, 250.9810); d (600 MHz; D O) 2.79 (2H, d, JH–P = 18.6 Hz,
P), 6.80 (1H, d, J = 1.8 Hz, ArH), 8.51 (1H, d, J = 1.8 Hz, ArCH);
5
6
H N
2
O
5
PNa
2
requires
dures.
In brief, XL-1 Blue competent cells were transformed
H
2
with EcDXR plasmid DNA. IPTG was used to induce expression of
the recombinant EcDXR gene. EcDXR was then purified by a combi-
CH
2
2
+
d
1
C
(150 MHz; D
2
O) 39.5 (d, JC–P = 111.0 Hz, CH
2
P), 99.1 (d, ArCH),
nation of Ni affinity- and size exclusion-chromatography. The
+
57.2 (s, ArC), 160.0 (d, ArCH), 170.8 (s, C@O); m/z 251 (MH ,
protein was stored at À20 °C in a sodium phosphate buffer of pH 7.
8
7%) and 217 (100).
4
.4. DXR inhibition assay34,40,42
4
.1.2.3. Disodium 2-[(1,3-thiazol-2-yl)carbamoyl]ethylphospho-
nate 11c. As a light brown solid (0.385 g, quantitative yield)
which decomposes above 290 °C and which was further purified
by reverse-phase chromatography [on C18 cellulose; elution with
Assays were conducted in
100 mM Tris–HCl (pH 7.5), 1 mM MgCl
0.3 mM DOXP in a total volume of 100
a
reaction mixture containing
, 0.3 mM NADPH and
L. Equal volumes of
EcDXR and ligand were incubated at 37 °C for 5 min; 100 L of
this enzyme–ligand mixture was then added to the rest of the
2
l
MeOH-0.1% aq NH
4
OH] followed by HPLC (LUNA 10
l
C
18 column,
l
+
2
C
2
7
50 Â 10.00 mm; elution with H
SPNa requires M+1, 266.9581); d
.80 (2H, d, JH–P = 18.6 Hz, CH P), 7.14 (1H, d, J = 3.6 Hz, ArH),
.43 (1H, d, J = 3.6 Hz, ArH); (150 MHz; O) 39.0 (d,
P), 114.0 (d, ArCH), 137.0 (d, ArC), 159.0 (s,
2
O) (Found: MH , 266.9582.
H
5 6
N
2
O
4
2
H
(600 MHz; D O)
2
assay components to make a total of 200
lL with a final EcDXR
À1
2
concentration of
5
l
g mL
.
The decrease in absorbance at
d
C
D
2
NADPH
340 nm due to the decreasing concentration of NADPH (e =
3
À1
À1
JC–P = 108.0 Hz, CH
2
6.3 Â 10 L mol cm ) was followed for 10 min at 37 °C, rela-
tive to a blank lacking the DOXP substrate. For each ligand,
enzyme activity in the absence of inhibitor was deemed to be
100% (i.e., 0% inhibition) and the % relative inhibition was deter-
mined in triplicate. Under these conditions, fosmidomycin 2 at
+
ArC), 170.4 (s, C@O); m/z 267 (MH , 22%) and 223 (100).
4
.1.2.4. Disodium 2-[(furan-2-ylmethyl)carbamoyl]ethylphosph-
onate 11d. As a brown solid (0.567 g, quantitative yield) which
+
decomposes above 250 °C (Found: MH , 264.0010. C
requires M, 264.0014); (600 MHz; O) 2.63 (2H, d,
H–P = 19.2 Hz, CH P), 4.40 (2H, d, CH N), 6.36 (1H, d, J = 3.6 Hz, 3-
CH), 6.44 (1H, dd, J = 3.6 and 1.8 Hz, 4-CH), 7.48 (1H, d, J = 1.8 Hz,
-CH); d (150 MHz; D O) 36.4 (t, CH N), 38.2 (d, JC–P = 114.0 Hz,
CH P), 107.1 (d, ArCH), 110.5 (d, ArCH), 142.5 (d, ArCH), 151.3 (s,
H
7 9
5
NO PNa
2
0.3 lM exhibited 99.3% inhibition.
d
H
D
2
J
2
2
Acknowledgements
5
C
2
2
The authors thank the Beit Trust for a bursary (to A.C.C),
Rhodes University, LIFElab and the South African Medical
Research Council (MRC) for generous financial support, Jessica
Goble for technical assistance with the expression and purifica-
tion of EcDXR and Professor Hassan Jomaa (Institute of Biochem-
istry, Academic Hospital Centre, Justus-Liebig-University) for the
gift of the DXR plasmid.
2
+
2
-C), 172.1 (s, C@O); m/z 264 (MH , 3%) and 394 (100).
4
.1.2.5. Disodium 2-[(5-acetyl-4-methyl-1,3-thiazol-2-yl)car-
bamoyl]ethylphos-phonate 11e. As a light brown solid (0.197 g,
+
quantitative yield) which decomposes above 290 °C (Found: MH ,
3
22.9856. C
MHz; D O) 2.59 (3H, s, CH
H–P = 18.6 Hz, CH P); d (150 MHz; D
), 38.9 (d, JC–P = 109.5 Hz, CH
s, ArC), 161.1 (s, ArC), 170.5 (s, NHC@O), 195.7 (s, CH
8
H
10
N
2
O
5
Na
2
PS requires M+1, 322.9843); d
), 2.61 (3H, s, CO.CH ), 2.90 (2H, d,
O) 17.4 (q, CH ) 29.3
P), 125.7 (s, ArC), 156.3
C@O); m/z
H
(600
2
3
3
Supplementary data
J
2
C
2
3
(
(
q, COÁCH
3
2
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.11.062.
3
+
3
23 (MH , 100%).
References and notes
4
.2. Saturation Transfer Difference (STD) experiments37
1
2
.
.
Areqawi, M.; Cibulskis, R.; Otten, M.; Williams, R.; Dye, C. World Malaria Report;
World Health Organisation: Geneva, 2008; pp 1–125.
Müller, O.; Sié, A.; Meissner, P.; Schirmer, R. H.; Kouyaté, B. Lancet 2009, 374,
EcDXR was extracted and purified as described below. From a
fraction containing 17.6
buffer pH 7.0, 1.1 mL was removed and freeze-dried and then re-
suspended in an equal volume of D O. An enzyme activity assay
was performed on 100 L of the re-suspended enzyme and the
lM of EcDXR in a 50 mM sodium phosphate
1419.
3. Trigg, P. I.; Kondrachine, A. V. The Current Global Malaria Situation. In Malaria,
Parasite Biology, Pathogenesis, and Protection; Sherman, I. W., Ed.; American
Society for Microbiology: Washington, DC, 1998; pp 11–22.
2
l
4.
Rohmer, M.; Seemann, M.; Horbach, S.; Bringer-Meyer, S.; Sahm, H. J. Am. Chem.
Soc. 1996, 118, 2564–2566.
ligands were dissolved in theremaining1 mL to give an approximate
ratio of protein to ligand of 1:40. The actual concentrations of the 4
5. Odom, A. R.; Van Voorhis, W. C. Mol. Biochem. Parasitol. 2010, 170, 108–111.

T. Bodill et al. / Bioorg. Med. Chem. 19 (2011) 1321–1327
1327
6
.
.
Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.; Weidemeyer, C.; Hintz,
M.; Türbachova, I.; Eberl, M.; Zeidler, J.; Lichtenthaler, H. K.; Soldati, D.; Beck, E.
Science 1999, 285, 1573–1576.
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 03, Revision E.01; Gaussian, Inc.: Wallingford, CT, 2004.
23. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comput. Chem. 1998, 19, 1639–1662.
7
Singh, N.; Cheve, G.; Avery, M. A.; McCurdy, C. R. Curr. Pharm. Des. 2007, 13,
1
161–1177.
8
9
.
.
Wiesner, J.; Borrmann, S.; Jomaa, H. Parasitol. Res. 2003, 90, S71–S76.
Mutorwa, M.; Salisu, S.; Blatch, G. L.; Kenyon, C.; Kaye, P. T. Synth. Commun.
2
009, 39, 2723–2736.
0. Deng, L.; Sundriyal, S.; Rubio, V.; Shi, Z.; Song, Y. J. Med. Chem. 2009, 52, 6539–
542.
24. Discovery Studio Visualizer, Release 2.0; Accelrys Software Inc.: San Diego, 2007.
25. Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.;
Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004, 25, 1605–1612.
26. Mac Sweeney, A.; Lange, R.; Fernandes, R. P. M.; Schulz, H.; Dale, G. E.;
Douangamath, A.; Proteau, P. J.; Oefner, C. J. Mol. Biol. 2005, 345, 115–127.
27. Steinbacher, S.; Kaiser, J.; Eisenreich, W.; Huber, R.; Bacher, A.; Rohdich, F. J.
Biol. Chem. 2003, 278, 18401.
1
1
1
6
1. Ortmann, R.; Wiesner, J.; Reichenberg, A.; Henschker, D.; Beck, E.; Jomaa, H.;
Schlitzer, M. Bioorg. Med. Chem. Lett. 2003, 13, 2163–2166.
2. Reichenberg, A.; Wiesner, J.; Weidemeyer, C.; Dreiseidler, E.; Sanderbrand, S.;
Altincicek, B.; Beck, E.; Schlitzer, M.; Jomaa, H. Bioorg. Med. Chem. Lett. 2001, 11,
8
33–835.
28. Yajima, S.; Hara, K.; Iino, D.; Sasaki, Y.; Kuzuyama, T.; Ohsawa, K.; Seto, H. Acta
Crystallogr., Sect. F: Struct. Biol. Cryst. Comm. 2007, 63, 466–470.
29. Cheng, F.; Oldfield, E. J. Med. Chem. 2004, 47, 5149–5158.
30. Haemers, T.; Wiesner, J.; Busson, R.; Jomaa, H.; Van Calenbergh, S. Eur. J. Org.
Chem. 2006, 17, 3856–3863; Haemers, T.; Wiesner, J.; Van Poecke, S.; Goeman,
J.; Henschker, D.; Beck, E.; Jomaa, H.; Van Calenbergh, S. Bioorg. Med. Chem. Lett.
2006, 16, 1888–1891.
1
1
3. Kurz, T.; Schlüter, K.; Kaula, U.; Bergmann, B.; Walter, R. D.; Geffken, D. Bioorg.
Med. Chem. 2006, 14, 5121–5135.
4. Wiesner, J.; Ortmann, R.; Jomaa, H.; Schlitzer, M. Arch. Pharm. (Weinheim) 2007,
340, 667–669.
1
1
5. Krise, J. P.; Stella, V. J. Adv. Drug Delivery Rev. 1996, 19, 287–310.
6. Devreux, V.; Wiesner, J.; Jomaa, H.; Van der Eycken, J.; Van Calenbergh, S.
Bioorg. Med. Chem. Lett. 2007, 17, 4920–4923.
31. Kurz, T.; Schlueter, K.; Kaula, U.; Bergmann, B.; Walter, R. D.; Geffken, D. Bioorg.
Med. Chem. 2006, 14, 5121–5135.
1
1
1
7. Kumar, K. G. D.; Saenz, D.; Lokesh, G. L.; Natarajan, A. Tetrahedron Lett. 2006, 47,
6
281–6284.
8. McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna, M. Tetrahedron Lett. 1977,
8, 155–158.
32. Singh, N.; Cheve, G.; Avery, M. A.; McCurdy, C. R. J. Chem. Inf. Model. 2006, 46,
1360–1370.
33. Silber, K.; Heidler, P.; Kurz, T.; Klebe, G. J. Med. Chem. 2005, 48, 3547–3563.
34. Perruchon, J.; Ortmann, R.; Altenkämper, M.; Silber, K.; Wiesner, J.; Jomaa, H.;
Klebe, G.; Schlitzer, M. ChemMedChem 2008, 3, 1232–1241.
35. Gasteiger, J.; Marsili, M. Tetrahedron 1980, 36, 3219–3228.
36. Argyrou, A.; Blanchard, J. S. Biochemistry 2004, 43, 4375–4384.
37. Mayer, M.; Meyer, B. Angew. Chem., Int. Ed. 1999, 38, 1784–1788.
38. Yan, J.; Kline, A. D.; Mo, H.; Shapiro, M. J.; Zartler, E. R. J. Magn. Reson. 2003, 163,
270–276.
39. Ji, Z.; Yao, Z.; Liu, M. Anal. Biochem. 2009, 385, 380–382.
40. Kuzuyama, T.; Shimizu, T.; Takahashi, S.; Seto, H. Tetrahedron Lett. 1998, 39,
7913–7916.
1
9. Giessmann, D.; Heidler, P.; Haemers, T.; Van Calenbergh, S.; Reichenberg, A.;
Jomaa, H.; Weidemeyer, C.; Sanderbrand, S.; Wiesner, J.; Link, A. Chem.
Biodiversity 2008, 5, 643–656.
2
2
2
0. Conibear, A. C.; Lobb, K. A.; Kaye, P. T. Tetrahedron 2010, 66, 8446–8449.
1. Cerius , Version 4.10 L revision 05.0708; Accelrys Inc.: Taipei, 1997.
2
2. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr., Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
41. Salisu, S. T. ATP Mimics as Glutamine Synthetase Inhibitors—an Exploratory
Synthetic Study. PhD Thesis, Rhodes University, Grahamstown, 2008.
42. Kuntz, L.; Tritsch, D.; Grosdemange-Billiard, C.; Hemmerlin, A.; Willem, A.;
Bach, T. J.; Rohmer, M. Biochem. J. 2005, 386, 127–135.