Potent antimalarial 4-pyridones with improved physico-chemical properties

Article abstract of DOI:10.1016/j.bmcl.2011.07.044

Antimalarial 4-pyridones are a novel class of inhibitors of the plasmodial mitochondrial electron transport chain targeting Cytochrome bc1 (complex III). In general, the most potent 4-pyridones are lipophilic molecules with poor solubility in aqueous medi

Full text of DOI:10.1016/j.bmcl.2011.07.044

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5214–5218
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Potent antimalarial 4-pyridones with improved physico-chemical properties
⇑
José M. Bueno , Pilar Manzano, María C. García, Jesús Chicharro, Margarita Puente, Milagros Lorenzo,
Adolfo García, Santiago Ferrer, Rubén M. Gómez, María T. Fraile, José L. Lavandera, José M. Fiandor,
Jaume Vidal, Esperanza Herreros, Domingo Gargallo-Viola
Tres Cantos Medicines Development Campus, Diseases of the Developing World, GlaxoSmihKline, C/Severo Ochoa, 2, 28760-Tres Cantos, Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Antimalarial 4-pyridones are a novel class of inhibitors of the plasmodial mitochondrial electron trans-
port chain targeting Cytochrome bc1 (complex III). In general, the most potent 4-pyridones are lipophilic
molecules with poor solubility in aqueous media and low oral bioavailability in pre-clinical species from
the solid dosage form. The strategy of introducing polar hydroxymethyl groups has enabled us to main-
tain the high levels of antimalarial potency observed for other more lipophilic analogues whilst
improving the solubility and the oral bioavailability in pre-clinical species.
Received 15 June 2011
Revised 11 July 2011
Accepted 12 July 2011
Available online 23 July 2011
Keywords:
Malaria
Ó 2011 Elsevier Ltd. All rights reserved.
Plasmodium falciparum
4-Pyridones
Solubility
Pharmacokinetic
Malaria is still one of the major causes of death in the planet,
with over half of the world’s population exposed to the risk of
infection. The World Health Organisation (WHO) has estimated
the burden of malaria to be around 225 million clinical cases, lead-
ing to nearly 781,000 deaths every year, most of which are
pregnant women and children under the age of five.1
were BCS class II compounds,11 with very low solubility in aqueous
media, good oral bioavailability in rodents from solution formula-
tions at low doses (>50% at dose <1 mg/kg),12 but low oral bioavail-
ability at higher doses administered in suspension (<20%). The main
issue for the development of these molecules is the lack of linearity
Chloroquine and multidrug resistant Plasmodium falciparum
strains have spread globally,2 and there are a limited number of
efficacious drug combinations available.3 On top of this, the signs
of emerging resistance to the artemisinins,4 the basic component
of the current gold standard combinations for the treatment of ma-
laria, has led to renewed efforts to find novel antimalarial drugs.5,6
The mitochondrial respiratory chain of P. falciparum makes an
attractive target for chemotherapy, and is validated by atovaquone.
Furthermore, it differs from the analogous mammalian system in a
number of ways, suggesting that parasite specific agents should be
achievable.7–9
In our efforts to find novel and affordable antimalarials, scientists
at the former Wellcome Laboratories and more recently at Glaxo-
SmithKline have developed a Medicinal Chemistry programme
based on 4-pyridones related to the anti-coccidial drug clopidol
(Fig. 1). As a result, a first series of potent diaryl-ether substituted
2,6-dimethyl-4-pyridones, have been reported.10 In general, the
most potent pyridones, exemplified by GW844520 and GW308678
O
O
O
R
3
5
X
Cl
Cl
2
6
N
H
N
H
clopidol
GW844520, X= Cl, R= 4-OCF3, Pf IC50= 0.007 µM
GW308678, X= Cl, R= 3-CF3, Pf IC50= 0.003 µM
O
O
R
Cl
2
6
Y
N
H
Z
I: Z= CH3; Y=CH2OH, CH2NR1R2,
II: Y=CH3; Z= CH2OH, CH2NR1R2; COOH, CON(CH3)2
R=4-OCF3, 3-CF3
⇑
Corresponding author. Tel.: +34 91 807 0616; fax: +34 91 807 0550.
E-mail address: jose.m.bueno@gsk.com (J.M. Bueno).
Figure 1.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.044

J. M. Bueno et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5214–5218
5215
O
O
O
O
O
O
OH
OTf
OH
OH
OH
a
b
c
d
HO
HO
HO
HO
O
O
1
O
3
Kojic Acid
2
O
O
O
O
O
O
R
R
R
f
Cl
e
HO
HO
HO
5a,b
N
H
N
H
6a,b
4a,b
O
O
O
O
Cl
OCF3
g
Cl
OCF3
O
O
N
H
h
R1R2N
R= 4-OCF3
N
H
H
8a-c
7
O
OH
I
j, k
i
R
R
R
+
HO
B
OH
Br
Br
10a: R= 4-OCF3
10b: R= 3-CF3
9a: R= 4-OCF3
9b: R= 3-CF3
Scheme 1. Reagents and conditions: (a) 37% aq CH2O, aq NaOH, rt 50%; (b) Zn, cc HCl, rt 53%; (c) Tf2NPh, DMF, K2CO3, rt 50%; (d) 10a,b, PdCl2(PPh3)2, Toluene/EtOH, aq
Na2CO3, 80 °C, 80% (4a), 90% (4b); (e) aq NH3, EtOH, 140 °C, 50% (5a), 60% (5b); (f) trichloroisocyanuric acid, MeOH/DCM, 0 °C to rt 77% (6a), 75% (6b); (g) SO3.Py, DMSO, TEA,
DCM, 0 °C to rt 90%; (h) R1R2NH, NaBH(AcO)3, AcOH, DCE, rt 60–70%; (i) CuCl, Cs2CO3, tetramethyl-heptanedione, NMP, 100 °C, 70% (9a), 90% (9b); (j) (PriO)3B, nBuLi, THF,
À78 °C; (k) 6N HCl, rt 80% (10a), 70% (10b).
O
O
O
O
OTf
OTBDMS
OH
OTBDMS
OBn
OTBDMS
OBn
OH
c
d
b
a
O
14
O
13
O
12
O
11
O
O
O
O
O
O
R
R
R
Cl
OTBDMS
OH
e
f
OH
O
N
H
N
H
15a,b
16a,b
17a,b
O
O
O
O
Cl
Cl
g
OCF3
OCF3
O
h
NR1R2
O
R= 4-OCF3
N
H
N
H
19a-c
H
18
i
O
O
O
Cl
Cl
OCF3
OCF3
O
N(CH3)2
j
N
H
N
H
OH
O
20
21
Scheme 2. Reagents and conditions: (a) TBDMSCl, DMF, imidazole, rt 96%; (b) H2, 10% Pd(C), EtOAc; (c) Tf2NPh, DMF, K2CO3, rt 90% (overall yield steps b + c); (d) 10a,b,
PdCl2(PPh3)2, Toluene/EtOH, aq Na2CO3, 80 °C, 81% (15a), 75% (15b); (e) aq NH3, EtOH, 140 °C, 60% (16a), 50% (16b); (f) trichloroisocyanuric acid, MeOH/DCM, 0 °C to rt 80%
(17a), 78% (17b); (g) SO3.Py, DMSO, TEA, DCM, 0 °C to rt 84%; (h) R1R2NH, NaBH(AcO)3, AcOH, DCE, rt, 60–70%; (i) NaClO2, NH2SO3H, acetone/water, rt, 82%; (j) (CH3)2NH,
TOTU, TEA, DCE, 0 °C to rt 72%.
between doses and oral exposures that hinders the safety assess-
ment studies in pre-clinical species.
GW844520 was progressed to pre-clinical studies and a back-
up programme aimed at designing novel 4-pyridone derivatives
with improved solubility and PK profile was started. In previous
studies we have found that the introduction of tertiary amines,10
esters or polar hydroxyl groups13 at position C3 had deleterious ef-
fects on the antimalarial activity (Pf IC50 >0.25 lM). Herein we

5216
J. M. Bueno et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5214–5218
Table 1
synthetic route depicted in Scheme 1 (see Refs. 10 and 15 for
experimental details). The introduction of the key phenyl-oxy-phe-
nyl lipophilic moiety was carried out by Suzuki coupling between
triflate 3 and the appropriate boronic acids 10a,b10 to afford pyr-
ones 4a,b which were subjected to ammonolysis and further chlo-
rination by reaction with trichloroisocyanuric acid.16 In general,
the use of trichloroisocyanuric acid was found to provide a conve-
nient substitute for N-chlorosuccinimide used previously in the
chlorination of pyridones,10 with advantages in shorter reaction
times, lower reaction temperatures, cleaner reaction crudes and
higher yields of final materials.
Antimalarial activity of compounds of general Formulae
I and II (P. falciparum
3H-hypoxanthine uptake)
O
O
R
Cl
Y
N
H
Z
Compound
Y
Z
R
Pf IC50 (mM)
The synthesis of compounds of general Formula II (Scheme 2),
has been carried out from Kojic acid derivative 1117 and involves
the preparation of key intermediate 13, in which the primary hy-
droxyl group is protected as TBDMS ether in order to prevent the
migration of the reactive triflate group from C3–OH to the vicinal
C2–CH2OH group.18 The introduction of the phenyl-oxy-phenyl
lipophilic moiety and the preparation of the 4-pyridone nucleus
have been accomplished in a similar way as described in Scheme 1.
Table 1 shows the in vitro antimalarial activity of the 4-pyri-
done derivatives prepared in terms of IC50 in the P. falciparum
3H-hypoxanthine uptake assay, as described by Desjardins et al.19
As shown, only compounds 17a,b, with the hydroxymethyl group
attached at position C6 of the 4-pyridone ring maintained the high
level of antimalarial activity observed for their non hydroxylated
counterparts GW844520 and GW308678, respectively. It is note-
worthy that compounds 6a,b, bearing the hydroxymethyl group
at position C2 of the 4-pyridone ring are significantly less potent
than their corresponding isomers 17a,b. Concerning other polar
or ionisable moieties, all the compounds having tertiary amines (
8a–c and 19a–c), carboxylic acid (20) or dimethyl-amide (21)
6a
6b
8a
CH2OH
CH2OH
Me2NCH2
CH3
CH3
CH3
4-OCF3
3-CF3
4-OCF3 >0.25
0.13
0.12
8b
CH3
CH3
4-OCF3 >0.25
N
CH2
O
8c
4-OCF3 >0.25
N
CH2
17a
17b
19a
CH3
CH3
CH3
CH2OH
CH2OH
Me2NCH2
4-OCF3
3-CF3
4-OCF3 >0.25
0.002
0.003
19b
CH3
4-OCF3 >0.25
N
CH2
CH2
O
19c
CH3
4-OCF3 >0.25
N
20
21
CH3
CH3
CO2H
CO2N(Me)2
CH3
4-OCF3 >0.25
4-OCF3 >0.25
4-OCF3
3-CF3
GW844520 CH3
GW308678 CH3
0.007
0.003
CH3
showed no activity in the assay (>0.25
position in the 4-pyridone ring.
lM), irrespective of their
describe the synthesis and biological activity of a series of 4-pyri-
dones bearing polar or ionizable moieties at positions C2 and C6 of
the 4-pyridone ring (Fig. 1, Formulae I and II).
A 3D theoretical model of the P. falciparum Cytochrome b:Rieske
protein complex has been built by using the crystal structures of
the bovine, chicken and yeast Cytochrome b as templates.20–22
Compounds were docked within the active site and the conforma-
tions with best docking scores were then minimised. Figure 2
shows the best conformation identified for compound 17a. From
the analysis of these results we can conclude that whilst most of
the interactions are hydrophobic, the most interesting one comes
from a hydrogen bond formed between the C6–CH2OH group and
the carboxylic acid of Glu-261. This is a new interaction identified
for 17a which could explain the increase in potency seen for this
The classical approach for the construction of the 2,6-dimethyl-
4-pyridone scaffold involves the condensation of the appropriate 2-
propanone derivatives with acetic anhydride and polyphosphoric
acid10 or Eaton’s reagent14 followed by ammonolysis. However,
the presence of substituents different from methyl at position C2
(Formula I) or at position C6 (Formula II), precludes the use of acetic
anhydride, thus making the synthesis particularly challenging.
The synthesis of compounds of general Formula I was carried
out from commercially available Kojic acid according to the
Figure 2. 3D-Theoretical model of the P. falciparum Cytbc1 active site showing the best conformation identified for compound 17a. The program MOE (2008.09) was used for
homology protein modelling and molecular calculations. Ligand docking studies were performed with program GOLD (v4.01) (Refs. 20 and 21).

J. M. Bueno et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5214–5218
5217
Table 2
Physicochemical properties of compounds 17a and GW844520
a
Compound
CHI log D (pH)
log Pa
mp (°C)
PSAb
pKa
Solubility, mg/ml (pH)
c
2
7.4
10.5
S0
SGF (1.2)
FeSSIF (5.0)
FaSSIF (6.8)
PBS (7.4)
17a
GW844520
2.43
2.75
2.42
2.79
2.2
2.77
4.4
4.8
272-274
296-298
71.55
51.32
9.73
10.1
4.6
1.4
372
287
2.34
0.7
0.69
0.5
<0.1
<0.1
a
b
c
Potentiometric titration measurements taken using a Gemini Profiler from pION.
Calculated data (Adamantis software).
Obtained from log P (Ref. 23).
from experimental log P values.24 As shown in Table 2 and Figure
3, the intrinsic solubility S0 obtained for compound 17a is approx-
imately three times higher than that for GW844520 over a wide
range of pH.
The equilibrium solubility experiments were performed in PBS
and bio-relevant media25 (SGF, FaSSIF and FeSSIF) in order to gain
insight about the potential absorption behaviour of these com-
pounds after oral administration. Compound 17a was more soluble
than GW844520 in all the media tested except PBS (pH 7.4), where
the equilibrium solubility measured for both compounds was ex-
tremely low. As expected, the best solubility was found in SGF at
pH = 1.2, where the nitrogen atom of the pyridone ring is proton-
ated (pKa <1.5).
Melting points are indicative of the crystal-lattice energy, which
for crystalline solids contribute significantly to solubility.26
Although both 17a and GW844520 are chemically stable solids
with high melting points, the lower value measured for 17a indi-
cates a lower crystal-lattice energy, thus contributing to increase
the solubility in comparison to the non hydroxylated analogue
GW844520.
Figure 3. pH-Solubility profile obtained by using the Henderson–Hasselbach
approach. Solubility profiles were calculated fitting curve to potentiometric
titration data in PBS, having previously measured pKa and log P to estimate S0
a
values (Ref. 23).
In order to evaluate the impact of the physico-chemical param-
eters measured on the in vivo behaviour, the pharmacokinetic pro-
file of 17a in comparison to GW844520 in CD1 mice and Beagle
dogs was assessed (Table 3).
compound in comparison with those found for its isomer 6a and
GW844520 in similar docking studies. On the other hand, the drop
of potency observed for compound 6a is also in agreement with
molecular modelling calculations, due to the hydrophobic nature
of the aminoacids surrounding the position C2 of the pyridone ring
(Ile-258, Val259 and Met-133).
Furthermore, the steric constrains around both positions C6,
and C2 may explain why compounds with larger substituents such
as di-alkyl-amines or amides were not active in the assay.
In order to study the influence of the 6-hydroxymethyl group
present in compound 17a on its physicochemical properties, in
Table 2 are reported the chromatographic hydrophobicity indexes
(CHI log D)23 measured at pH 2.0, 7.4 and 10.5, log P, melting point,
calculated polar surface area, pKa, intrinsic solubility S0, pH-solu-
bility profile24 and equilibrium solubility in different media, in
comparison with GW844520.
As shown, the increase in the solubility of 17a in comparison
with its nonhydroxylated analogue GW844520, particularly in
SGF and FeSSIF, translated into a significant enhancement in the
oral bioavailability from the solid dosage form (suspension in 1%
methylcellulose) both in mice (50% vs 20%) and particularly in dogs
(16% vs 4.4%). On the other hand, compound 17a seems to be more
sensitive to metabolic clearance, as demonstrated by the signifi-
cant reduction in the half-lives in both species, possibly due to
the oxidation of the hydroxyl group to carboxylic acid or by direct
elimination via conjugation. In spite of its shorter half-life, com-
pound 17a has displayed excellent in vivo antimalarial efficacy in
our murine model of P. falciparum malaria27 (ED50 = 0.6 mg/kg).
In conclusion, the data reported suggest that it is possible to im-
prove the physicochemical properties and pharmacokinetic profile
of 4-pyridones by performing certain chemical transformations at
the 4-pyridone ring. Although the presence of ionisable tertiary
amines, carboxylic acid or amide groups are not allowed, hydroxy-
methyl groups show a very different behaviour, depending on their
As shown, compound 17a is less lipophilic and, in general, it dis-
plays better physicochemical properties than its non hydroxylated
analogue GW844520. Both molecules remained unionised at pH 2–
10 as demonstrated by the CHI log D data and the pH-solubility
curve (Fig. 3), obtained by using the Henderson–Hasselbach ap-
proach that uses the pKa and the intrinsic solubility S0 obtained
position. Thus, whilst the CH2OH group induces
a
drop in
Table 3
Pharmacokinetic parameters of compounds 17a and GW844520 in CD1 mice and Beagle dogs.
Compound
Mouse
Dog
po 10 mg/kga
Cmax (mg/ml)
iv 0.2 mg/kgb
po 2 mg/kga
Cmax (mg/ml)
iv 0.05 mg/kgb
Cl (ml/min/kg)
%F
Vd (L/kg)
Cl (ml/min/kg)
t1/2 (h)
%F
Vd (L/kg)
t1/2 (h)
17a
GW844520
1.1
0.9
50
20
1
1.2
3
0.6
3.8
24.1
0.3
0.03
16
4.4
2.8
2.9
0.8
0.2
42
143
a
Oral gavage. Suspension in 1% methyl cellulose.
Solution in 1% DMSO/7.5% PEG400/20% encapsin/saline, pH 6.
b

5218
J. M. Bueno et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5214–5218
10. Yeates, C. L.; Batchelor, J. F.; Capon, E. C.; Cheesman, N. J.; Fry, M.; Hudson, A. T.;
Pudney, M.; Trimming, H.; Woolven, J.; Bueno, J. M.; Chicharro, J.; Fernández,
E.; Fiandor, J. M.; Gargallo-Viola, D.; Gómez de las Heras, F.; Herreros, E.; León,
M. L. J. Med. Chem. 2008, 51, 2845.
11. Waiver of in vivo bioavailability and bioequivalence studies for immediate
release solid oral dosage form based on a biopharmaceutics classification
system. http://www.fda.gov/AboutFDA/CentersOffices/cder/ucm128219.htm.
12. Xiang, H.; Surdy-Freed, J.; Moorthy, G. S.; Hugger, E.; Bambal, R.; Han, C.; Ferrer,
S.; Gargallo-Viola, D.; Davis, C. B. J. Pharm. Sci. 2006, 95, 2657.
13. Bueno, J. M. GlaxoSmithKline Laboratories, Unpublished Results.
14. Eaton, P. E.; Carlson, G. R.; Lee, J. T. J. Org. Chem. 1973, 38, 4071.
15. For experimental details: Bueno, J. M.; Chicharro, J.; Lorenzo, M.; Manzano, P.
WO2007138048, in PCT Int Appl. 2007.
antimalarial activity in vitro when it is located at position C2 in
comparison with its non hydroxylated counterpart, it is possible
to obtain potent derivatives with improved solubility and pharma-
cokinetic properties by introducing the CH2OH group at position
C6 of the 4-pyridone ring. This behaviour offers new opportunities
in the design of novel potent, more soluble and bioavailable anti-
malarial 4-pyridones.
Acknowledgement
16. Tilstam, U.; Winmann, H. Org. Proc. Res. Dev. 2002, 6, 384.
17. Liu, D. Z.; Piyamongkol, S.; Liu, D. Y.; Khodr, H. H.; Lu, S. L.; Hider, R. C. Bioorg.
Med. Chem. 2001, 9, 563.
We wish to thank MMV (Medicines for Malaria Venture) for
financial support.
18. Pankiewicz, K. W.; Nawrot, B. C.; Watanaba, K. A. J. Org. Chem. 1986, 51, 1525.
19. Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D. Antimicrob. Agents
Chemother. 1979, 16, 710.
References and notes
20. Lavandera, J. L. GlaxoSmithKline Laboratories, unpublished work.
21. Chemical Computing Group, 1010 Sherbrooke St., West, Suite 910, Montreal
Quebec, Canada H3A 2R7.
22. CCDC, The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ, UK.
23. Valko, K.; Du, C. M.; Bevan, Ch.; Reynolds, D. P.; Abraham, M. H. Curr. Med.
Chem. 2001, 8, 1137.
24. Avdeef, A. Curr. Top. Med. Chem. 2001, 1, 277.
25. Dressman, J. B.; Reppas, Ch. Eur. J. Pharm. Sci. 2000, 11, S73.
26. Raevsky, O. A. Mini-Rev. Med. Chem. 2004, 4, 1041.
27. Jiménez-Díaz, M. B.; Mulet, T.; Viera, S.; Gómez, V.; Garuti, H.; Ibáñez, J.;
Álvarez-Doval, A.; Shultz, L. D.; Martínez, A.; Gargallo-Viola, D.; Angulo-
Barturen, I. Antimicrob. Agents Chemother. 2009, 53, 4533.
1. World Malaria Report 2010. WHO Geneva: Switzerland. http://www.who.int/
malaria/world_malaria_report_2010/en/index.html.
2. Wells, T. N.; Poll, E. M. Discov. Med. 2010, 9, 389.
3. Olliaro, P.; Wells, T. N. Clin. Pharmacol. Ther. 2009, 85, 584.
4. Dondorp, A. M.; Yeung, S.; White, L.; Nguon, C.; Day, N. P.; Socheat, D.; von
Sidlein, L. Nat. Rev. Microbiol. 2010, 8, 272.
5. Slichtzer, M.; Ortmann, R. ChemMedChem 2010, 5, 1837.
6. Burrows, J. N.; Waterson, D. Discovery New Medicines to Control and Eradicate
Malaria In Trop. Med. Chem. doi:10.1007/7355_2011_14, Springer-Verlag:
Berlin-Heidelberg; 2011.
7. Rodrigues, T.; Lopes, F.; Moreira, R. Curr. Med. Chem. 2010, 17, 929.
8. Mather, M. W.; Henry, K. W.; Vaidya, A. B. Curr. Drug Targets 2007, 8, 49.
9. Painter, H. J.; Morrisey, J. M.; Mather, M. W.; Vaidya, A. B. Nature 2007, 446, 88.