Synthesis and evaluation of phosphoramidate and phosphorothioamidate analogues of amiprophos methyl as potential antimalarial agents

Article abstract of DOI:10.1016/j.bmcl.2011.07.088

A series of phosphoramidate and phosphorothioamidate compounds based on the lead antitubulin herbicidal agents amiprophos methyl (APM) and butamifos were synthesised and evaluated for antimalarial activity. Of these compounds, phosphorothioamidates were m

Full text of DOI:10.1016/j.bmcl.2011.07.088

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6180–6183
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of phosphoramidate and phosphorothioamidate
analogues of amiprophos methyl as potential antimalarial agents
Christine Mara a, Enda Dempsey b, Angus Bell b, James W. Barlow a,
⇑
a Department of Pharmaceutical & Medicinal Chemistry, Royal College of Surgeons in Ireland, Stephens Green, Dublin 2, Ireland
b Department of Microbiology, School of Genetics & Microbiology, Moyne Institute of Preventive Medicine, Trinity College Dublin, Dublin 2, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of phosphoramidate and phosphorothioamidate compounds based on the lead antitubulin her-
bicidal agents amiprophos methyl (APM) and butamifos were synthesised and evaluated for antimalarial
activity. Of these compounds, phosphorothioamidates were more active than their oxo congeners and the
nature of both aryl and amido substituents influenced the desired activity. The most active compound
was 46, O-ethyl-O-(2-methyl-4-nitrophenyl)-N-cyclopentyl phosphorothioamidate, which was more
effective than the lead compound.
Received 16 March 2011
Revised 22 July 2011
Accepted 23 July 2011
Available online 29 July 2011
Keywords:
Ó 2011 Elsevier Ltd. All rights reserved.
Phosphoramidate
Phosphorothioamidate
Antimalarial
Tubulin
Plasmodium falciparum
Malaria is a devastating parasitic disease caused by apicom-
plexan species of the genus Plasmodium, notably the most virulent
species Plasmodium falciparum, and is responsible for about
800,000 deaths per annum, many of these in children and pregnant
women, and most of these in Africa.1 Most current therapies for
malaria target the blood stage in the life cycle of the Plasmodium par-
asite.2 The prototypic agent quinine and its derivatives, the 4-
aminoquinolines such as chloroquine, target primarily the asexual,
erythrocytic stages of the disease, while the 8-aminoquinolines such
as primaquine target hepatic stages. Other agents include anti-fo-
lates (pyrimethamine–sulfadoxine), inhibitors of mitochondrial
electron transport (atovaquone) and artemisinins, whose mecha-
nism of action is controversial. While this last class of drugs has
undoubtedly revolutionised the treatment of malaria and are the
agents of choice in malarial treatment protocols as artemisinin
based combination therapies, the ever present spectre of resistance
is a constant concern, especially in areas where sub-optimal dose re-
gimes are employed. Current best-practice favours the use of more
than one agent to reduce the chance of spontaneous mutation and
to prevent resistance. Despite these measures, worryingly, reports
of artemisinin resistance have emerged.3 The need for novel thera-
peutics to combat this insidious pathogen is thus evident, including
agents targeting novel parasitic targets.4 One potential class of anti-
malarial agents may be inhibitors of the protein tubulin, and the
activity of anti-tubulin compounds as antiparasitic agents has been
recently reviewed.5,6 Although a ubiquitous protein, there is signif-
icant amino acid sequence heterogeneity between mammalian and
parasite tubulins, prompting the hope that selective antitubulin
agents may prove useful.7 Typical antitubulin agents such as taxanes
and Vinca alkaloids, while potent against Plasmodium, are non-
selective. In contrast, herbicidal agents of the dinitroaniline class
such as trifluralin 1 have shown activity against P. falciparum with
minimal cytotoxicity.8 The site of action of dinitroaniline herbicides
in plant and protozoal species is controversial and as yet unproven,
but is likely to be located on a
-tubulin.9 Electron microscopic auto-
radiography showed radioactive trifluralin associated with microtu-
bule fragments.10 In an in vivo model of mouse malaria, trifluralin
was inactive against Plasmodium berghei despite submicromolar
activity in vitro, an effect ascribed to poor absorption.11 Trifluralin
has poor aqueous solubility, with a reported c log P of 5.322,12 and
as well as affecting its pharmacokinetics this may cause it to accu-
mulate in membranous compartments of the parasite and its host
erythrocyte, away from its site of action in the cell.13 A further disad-
vantage regarding the use of dinitroanilines such as trifluralin as
lead compounds is their potential carcinogenicity.14 Despite these
drawbacks, many dinitroanilines have been synthesised and tested
against both apicomplexan and kinetoplastid parasites. Of the many
derivatives evaluated, some of the more interesting compounds
with activity against apicomplexans are shown in Figure 1.
Armson has reported IC50 values for trifluralin and the sulfon-
amide dinitroaniline oryzalin 2 against Cryptosporidium of 750
and 800 nM, respectively.15 Oryzalin and its analogues have also
been investigated for activity against Toxoplasma gondii. While
Abbreviations: APM, amiprophos methyl.
⇑
Corresponding author. Tel.: +353 1 402 8520; fax: +353 1 402 2765.
oryzalin itself had an IC50 value of 0.25 lM, benzenediamine 3
E-mail address: jambarlow@rcsi.ie (J.W. Barlow).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.088

C. Mara et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6180–6183
6181
Figure 1. Antitubulin herbicidal compounds and derivatives.
had an IC50 value of 36 nM.16 Regioisomeric dinitroanilines exhib-
ited interesting activity against several parasites, with 4 and 5 the
date compounds, our synthetic strategy employed the commer-
cially available O-alkylated reagent ethyl phosphorodichloridate.
Sequential displacement of the chlorine atoms in this compound
was accomplished easily with equimolar quantities of phenol and
amine synthons, as shown in Scheme 1, and the desired phospho-
ramidates 8–23 were obtained in 17–73% yields (Table 1). Analo-
gous phosphorothioamidates, including 6, were prepared by the
sequential reaction of ethyl phosphorothiodichloridate with
phenolic and amine synthons.
Having varied the phenolic substituents in phosphoramidates
8–23, we then wished to investigate the effect of varying the nat-
ure of the amide functionality, by introduction of primary, second-
ary and tertiary nitrogen species, and also branched and cyclic
systems. These compounds, 25–49, were synthesised using analo-
gous methodology to that used in Scheme 1, and the resultant
phosphoro(thio)amidates are shown in Table 2. Structures for the
proposed target compounds were confirmed with the aid of 1H,
13C and 31P NMR spectroscopy and mass spectrometry analysis,
while the purity of the compounds was assessed by elemental
analysis. Synthetic procedures, spectral data and testing
methodology are provided in the Supplementary section.
Tables 1 and 2 show the anti-malarial activity of the phospho-
roamidate and phosphorothioamidate compounds tested. In our
experiments, APM had an IC50 value of 4 lM. All compounds in
Table 1 reflected the intention of preserving the isopropylamido
group of 6 intact while varying the nature of the aromatic moiety.
Phosphoramidate regioisomers 9–14 were devoid of activity or al-
most so, and while replacement of the 2-nitro group by cyano- or
methoxy- substituents did not improve activity, replacement by a
halogen, as in compounds 16 and 18, reduced the IC50 value to
most potent, both with an IC50 value of 0.44 lM against P.
falciparum.17
Phosphorothioamidate herbicides, typified by APM 6 and butam-
ifos 7 (Fig. 1), are considered to bind tubulin in the same way as the
dinitroanilines. APM was shown competitively to inhibit the binding
of 14C oryzalin to tobacco tubulin, indicating the formation of a mod-
erate affinity tubulin/APM complex that may interact with the ends
of microtubules. APM concentrations that inhibited plant cell
growth were within the threshold range of APM concentrations that
depolymerised cellular microtubules, suggesting that growth inhi-
bition is caused by microtubule depolymerisation.18 Molecular
modelling studies have shown that the electrostatic surfaces of the
phosphorothioamidates are very similar to those of the dinitroani-
lines, with the electronegative domains of the ortho nitro (-phenyl)
group and the phosphorothio group matching both the shape and
spacing of the equivalent regions of the dinitroanilines.19 In addi-
tion, plant species with resistance to dinitroanilines also show cross
resistance to phosphorothioamidates.20,21 The observationthat APM
and the dinitroanilines 1 and 2 prevented P. falciparum erythrocytic
shizogony, blocked mitosis and resulted in accumulation of ab
normal microtubular structures22 suggested that the phospho-
rothioamidate class of herbicides were also worthy of investigation
as potential antimalarial lead compounds. In that study, the IC50 va-
lue of APM against P. falciparum was 3.5
trifluralin of 2.9 M, while the N-phenylcarbamate herbicide chlo-
ropropham lacked any inhibitory effect up to 128 M. In addition,
neither trifluralinnor APM showed any inhibitoryeffect on mamma-
lian (Vero monkey kidney) cells at concentrations up to 64 M.
lM, comparable to that for
l
l
l
Phosphorothioamidates also lack the potentially toxophoric dinitro-
aniline motif. Cognisant of these facts, we decided to synthesise and
test a range of phosphoramidate and phosphorothioamidate ana-
logues of APM and butamifos and test these compounds for activity
against P. falciparum.
Original syntheses for pesticidal phosphorothioamidates in-
volve sequential substitution of a pentavalent phosphorus species.
One available approach is the reaction of an anhydrous alkali metal
phenolate and an alkali metal monohydric alkoxide with an
N-substituted dichlorothiophosphoramide in an excess of the alco-
hol used to prepare the alkoxide.23 The order of substitution is not
critical, and the order of addition of the phenolate and alkoxide
may be reversed. Indeed, it is also possible to react either as
starting material an O-alkyl dichlorothiophosphate or an O-phenyl
dichlorothiophosphate sequentially with the remaining two
synthons. As we first wished to generate a series of phosphorami-
39 lM. The mono-substituted trifluoromethylphenol derivatives
19–21 demonstrated potential, with the meta and para compounds
20 and 21 the better of those tested, both with IC50 values of
50 lM. Bulkier naphthalene groups as in 22 and 23 failed to im-
prove activity. The compounds listed in Table 2 show the effect
of varying the amido substituent within both APM-like and p-
trifluoro or halogen series. Within APM/butamifos-like compounds
25–31, the most potent was the n-butyl 26. Within the trifluoro
series 32–44, elongation of the amido alkyl moiety to C5 enhanced
activity whether acyclic or cyclic; this is exemplified by N-pentyl
36, at 4.5
Replacement of the oxygen atom by sulfur raised activity three-
fold, as shown by a comparison of 37 and 38 (26 vs 8.6 M). Inter-
estingly, compounds 45 and 46, both 2-methyl-4-nitro substituted
phosphorothioamidates, had activities of 6.9 and 1.6 M, respec-
tively. To compare the effect of the X group on phosphorus further,
lM and its N-cyclopentyl analogue 38, at 8.6 lM.
l
l

6182
C. Mara et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6180–6183
Scheme 1. Synthesis of substituted phosphoramidates and phosphorothioamidates (8–49). Reagents and conditions: (a) C2H5XPCl2; Et3N; THF (series A) or PhCH3 (series B);
0 °C; N2; (b) R2NH2; Et3N; THF (series A) or PhCH3 (series B); 0 °C; N2.
Table 1
Anti-malarial activity of phosphoroamidates 8–23 as assessed by inhibition of pLDH
at 72 h
Known SAR for herbicidal activity24 predicts that deactivating
aromatic substituents such as a nitro group may enhance activity.
With regard to antimalarial activity, Table 1 shows that while this
may be a shared effect, as suggested by the activity of trifluoro-
methyl derivative 21, it is difficult to ascertain within the phosphor-
amidate series, due to the fact that many compounds (e.g.,
nitro-substituted compounds 10–15) were devoid or almost devoid
of activity. This may suggest an underlying requirement for the
phosphorothio rather than the phosphoro moiety. Compared to
APM, its O-ethyl phosphoramidate analogue has a 20-fold higher
IC50 value. This suggests the sulfur analogues may be a great deal
more potent than their oxygen congeners. In vivo, however, oxida-
tion of thiophosphate compounds to their phosphate analogues
occurs via cytochrome-P450-dependent desulfuration in the liver.25
Interestingly, trifluoro compound 21 has been previously reported;
Compound
R1
R2
X
IC50
4
>128
126
128
>128
128
79
128
128
39
128
39
(
l
M)
c log Pa
6 (APM)
8
9
4-CH3-2-NO2
5(H)
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
—
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
i-Propyl
S
3.50
2.68
3.11
3.11
3.11
3.11
3.11
3.11
2.97
3.94
2.90
3.67
6.23
3.57
3.57
3.57
3.69
3.64
3.84
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
S
4-CH3-2-NO2
2-CH3-4-NO2
5-CH3-2-NO2
2-CH3-5-NO2
3-CH3-4-NO2
2-CH3-3-NO2
2-CN-4-CH3
2-Br-4-CH3
2-CH3O-4-CH3
2-Cl-4-CH3
2-Cl-4-CH3
2-CF3
3-CF3
4-CF3
2-Naphthol
1-NO2-2-Naphthol
4-CH3-2-NO2
10
11
12
13
14
15
16
17
18
18b
19
20
21
22
23
24
102
87
50
50
72
a
German patent describes the use of 4-trifluoromethyl-
phenyl(thio)phosphoric acid amides as pesticides.26
87
—
Variation of the amido functionality on the nitrogen however
had an obvious effect on activity. Within the amido homologues
25–30, this effect was not as dramatic as within the para-trifluoro
series 32–44, as increasing the chain length and branching did im-
prove activity up to C5, as evidenced by the IC50 values of com-
a
c log P values calculated using MarvinSketch 5.1.4 from ChemAxon.
pounds 36 and 38, at 4.5 and 8.6 lM, respectively. This indicated
Table 2
Anti-malarial activity of compounds 25–49 as assessed by inhibition of pLDH at 72 h
that pentyl or cyclopentylamino substituents were optimal, as fur-
ther enhancing the lipophilic nature of the side chain resulted in
reduced activity, as shown by the n-heptyl compound 40. Analysis
of the predicted c log P values as shown in Tables 1 and 2 revealed
a modest correlation with activity for all compounds (R2 = 0.504).
Interestingly, the degree of substitution on nitrogen also appeared
to affect activity within the para-trifluoro phosphoramidate series.
The primary amido 32 was not very potent at 79 lM, yet tertiary
amides 41–43 displayed some activity, comparable to the
secondary phosphoramides among 33–40.
The present study has allowed us to conclude that an electron-
withdrawing group on the phenolic ring may enhance activity
within the phosphoroamidate series, particularly in the para-posi-
tion. Additional aromatic alkyl substitution may enhance activity,
although it does not appear to be crucial. For most compounds
tested, phosphorothioamidates were more active than their oxo
congeners. Within these, an alkylamido C5 residue seems to be
optimal. Of all compounds prepared, 36 and 46 are most worthy
of further investigation.
Compound
R1
R2
X
IC50
4
>128
28
75
51
(lM)
c log Pa
6 (APM)
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
4-CH3-2-NO2
4-CH3-2-NO2
4-CH3-2-NO2
4-CH3-2-NO2
4-CH3-2-NO2
4-CH3-2-NO2
5-CH3-2-NO2
5-CH3-2-NO2
4-CF3
i-Propyl
n-Propyl
n-Butyl
i-Butyl
n-Pentyl
Cyclopentyl
n-Propyl
i-Butyl
NH2
S
3.50
3.16
3.56
3.56
3.95
3.54
3.16
3.56
2.56
4.02
4.04
3.60
5.15
4.00
4.73
4.40
5.21
3.74
3.78
3.38
2.72
4.69
4.27
3.91
4.64
4.63
O
O
O
O
O
O
O
O
O
O
O
S
47
102
>128
79
4-CF3
n-Butyl
32
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
4-CF3
2-CH3-4-NO2
2-CH3-4-NO2
4-Br
4-Br
4-Br
sec-Butyl
Cyclobutyl
n-Pentyl
40
45
4.5
26
8.6
43
44
48
84
56
Cyclopentyl
Cyclopentyl
Cyclohexyl
n-Heptyl
N,N-Diethyl
Piperidino
Pyrrolidino
Morpholino
n-Pentyl
Cyclopentyl
Cyclopentyl
Cyclopentyl
Cyclopentyl
O
S
O
O
O
O
O
O
S
S
O
S
98
6.9
1.6
17
23
128
Acknowledgment
We acknowledge Enterprise Ireland (PC/2005/070) for financial
support.
—
a
c log P values calculated using MarvinSketch 5.1.4 from ChemAxon.
Supplementary data
we prepared 47–49. However, unusually, oxo and thio congeners
Supplementary data (experimental procedures and spectral
data) associated with this article can be found, in the online ver-
sion, at doi:10.1016/j.bmcl.2011.07.088.
were similarly active at 17 and 23
tricoordinate species was inactive.
lM, respectively, while the

C. Mara et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6180–6183
6183
13. Naughton, J. A.; Hughes, R.; Bray, P.; Bell, A. Biochem. Pharmacol., 2008. 1580.
14. Saghir, S. A.; Charles, G. D.; Bartels, M. J.; Kan, L. H. L.; Dryzga, M. D.; Brzak, K.
A.; Clark, A. J. Toxicol. Lett. 2008, 180, 38.
15. Armson, A.; Sargent, K.; MacDonald, L. M.; Finn, M. P.; Thompson, R. C.;
Reynoldson, J. A. FEMS Immunol. Med. Mic. 1999, 26, 109.
16. Endeshaw, M. M.; Li, C.; de Leon, J.; Yao, N.; Latibeaudiere, K.; Premalatha, K.;
Morrissette, N.; Werbovetz, K. Bioorg. Med. Chem. Lett. 2010, 20, 5179.
17. Best, W. M.; Sims, C. G.; Thompson, R. C. A.; Armson, A.; Reynoldson, J. A.
Antiparasitic compounds. WO/2006/108224.
References and notes
1. WHO. World Malaria Report, 2010.
2. Schlitzer, M. Arch. Pharm. Chem. Life Sci. 2008, 341, 149.
3. 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.; Lindegardh, N.; Socheat, D.; White, N. J. N. Engl. J. Med. 2009, 361,
455. Erratum in: N. Engl. J. Med. 2009, 361, 1714..
4. Sahu, N. K.; Sahu, S.; Kohli, D. V. Chem. Biol. Drug Des. 2008, 71, 287.
5. Kappes, B.; Rohrbach, P. Future Microbiol. 2007, 2, 409.
6. Fennell, B. J.; Naughton, J. A.; Barlow, J.; Brennan, G.; Fairweather, I.; Hoey, E.;
McFerran, N.; Trudgett, A.; Bell, A. Expert Opin. Drug Discov. 2008, 3, 501.
7. Bell, A. Parasitol. Today 1998, 14, 234.
8. Nath, J.; Schneider, J. Clin. Res. 1992, 40, 331A.
9. Nyporko, Y.; Yemets, A.; Brytsun, V.; Lozinsky, M.; Blume, Y. Tsitol. Genet+ 2009,
43, 56.
10. Kaidoh, T.; Nath, J.; Fujioka, H.; Okoye, V.; Aikawa, M. J. Eukaryot. Microbiol.
1995, 42, 61.
11. Dow, G. S.; Armson, A.; Boddy, M. R.; Itenge, T.; McCarthy, D.; Parkin, J. E.;
Thompson, J. C. A.; Reynoldson, J. A. Exp. Parasitol. 2002, 100, 155.
12. Mitra, A.; Sept, D. J. Med. Chem. 2006, 49, 5226.
18. Murthy, J. V.; Kim, H. H.; Hanesworth, V. R.; Hugdahl, J. D.; Morejohn, L. C. Plant
Physiol. 1994, 105, 309.
19. Ellis, J. R.; Taylor, R.; Hussey, P. J. Plant Physiol. 1994, 105, 15–18.
20. Vaughn, K. C.; Marks, M. D.; Weeks, D. P. Plant Physiol. 1987, 83, 956.
21. James, S. W.; Silflow, C. D.; Stroom, P.; Lefebvre, P. A. J. Cell Sci. 1993, 106, 209.
22. Fennell, B.; Naughton, J.; Dempsey, E.; Bell, A. Mol. Biochem. Parasitol. 2006, 145,
226.
23. Moyle, C. L.; Kenaga, E. E. U.S. Patent 2,252,574; Moyle, C. L. U.S. Patent
2,252,576.
24. Blume, Ya.; Nyporko, A.; Yemets, A.; Baird, W. Cell Biol. Int. 2003, 27, 171.
25. Tang, J.; Chambers, J. J. Biochem. Mol. Toxic. 1999, 13, 261.
26. Buerstinghaus, R.; Varwig, J.; Kiehs, K.; Adolphi, H. Ger. Offen. 1984, DE
3,231,191.