Kojic acid derived hydroxypyridinone-chloroquine hybrids: Synthesis, crystal structure, antiplasmodial activity and β-haematin inhibition

Article abstract of DOI:10.1016/j.bmcl.2014.06.012

Aminochloroquinoline-kojic acid hybrids were synthesized and evaluated for β-haematin inhibition and antiplasmodial activity against drug resistant (K1) and sensitive (3D7) strains of Plasmodium falciparum. Compound 7j was the most potent compound in both strains (IC₅₀^3D7 = 0.004 μM; IC₅₀^K1 = 0.03 μM) and had the best β-haematin inhibition activity (0.07 IC₅₀ equiv vs 1.91 IC ₅₀ equiv for chloroquine). One compound 8c was found to be equipotent in both strains (IC₅₀ = 0.04 μM).

Full text of DOI:10.1016/j.bmcl.2014.06.012

Bioorganic & Medicinal Chemistry Letters 24 (2014) 3263–3267
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Kojic acid derived hydroxypyridinone–chloroquine hybrids:
Synthesis, crystal structure, antiplasmodial activity and b-haematin
inhibition
Warren Andrew Andayi a, Timothy J. Egan a, Kelly Chibale a,b,
⇑
a Department of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
b South African Medical Research Council Drug Discovery & Development Research Unit, Institute of Infectious Diseases and Molecular Medicine, University of Cape Town,
Rondebosch 7701, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Aminochloroquinoline–kojic acid hybrids were synthesized and evaluated for b-haematin inhibition and
Received 8 May 2014
Revised 2 June 2014
Accepted 5 June 2014
Available online 14 June 2014
antiplasmodial activity against drug resistant (K1) and sensitive (3D7) strains of Plasmodium falciparum.
K1
Compound 7j was the most potent compound in both strains (IC350D7 = 0.004
lM; IC = 0.03 lM) and had
50
the best b-haematin inhibition activity (0.07 IC50 equiv vs 1.91 IC50 equiv for chloroquine). One com-
pound 8c was found to be equipotent in both strains (IC50 = 0.04 M).
l
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Hydroxypyridinone
Kojic acid
Haematin inhibition
Antiplasmodial
Hybrids
Malaria is a lethal human parasitic disease with over 200 mil-
lion annual cases reported worldwide in 2010.1 The World Health
Organization estimated 660,000 deaths due to malaria worldwide
of which 90% were in the African region and the remainder in
the South East Asia and eastern Mediterranean regions.1
Thus antiplasmodial iron chelators have such a potential.
Iron chelators especially 3,4-hydroxypyridinones (3,4-HPOs)
have been proposed as potential antimalarial agents because of
the central role of iron for the rapid proliferation of the malaria
parasite and the arrest of parasite growth by iron chelators
in vivo and in vitro. An example is deferiprone, an orally bioavail-
able 3,4-HPO that is used for the treatment of iron overload or
b-thallasaemia.10–14 On the other hand 4-aminoquinoline based
compounds continue to be utilized as starting parts for new anti-
malarial agents capable of circumventing aminoquinoline drug
resistance.2
The rationale for pharmacophore hybridization of various
bioactive agents is to increase their therapeutic potential.15–17 This
may be achieved through improvement in treatment efficacy in
part through improvement in the pharmacokinetic profile of
individual moieties through the resulting single molecular entity,
reduction in the emergence and spread of drug resistance due to
structural novelty of the hybrid as well as decrease in dose-
dependent toxicity as the hybrid may be used at lower doses due
to improved efficacy over individual components. In addition to
these aforementioned potential benefits, hybridization provides
synthetic advantages as it leads to structural novelty and
diversity.17
The problem of endemic malaria has been exacerbated by the
emergence of widespread resistance to the once effective first line
treatment drugs such as chloroquine (CQ) and sulphadoxine-
pyrimethamine,2 and more recently early indications of resistance
to artemisinin monotherapy.3,4 Artemisinin based combination
therapies, ACTs, have replaced the failed therapies and are the
recommended first line treatments of falciparum malaria in all
endemic countries. With increasing resistance to other available
agents, intensive drug discovery efforts aimed at developing new
antimalarial drugs or modifying existing ones are on-going.5
It has been reported that an elevated host iron level is a serious
risk factor for human malaria.6 This has been observed in pregnant
women in the 3rd trimester of gestation and in non-pregnant per-
sons on iron supplements.6–9 Antimalarial drugs that can address
the excess physiologic iron load problem may be beneficial to
infected pregnant mothers and persons with excess physiologic
iron.
Recently we demonstrated that the incorporation of the maltol-
derived 3,4-HPO moiety into aminochloroquinolines by molecular
⇑
Corresponding author. Tel.: +27 21 650 2553; fax: +27 21 689 7499.
E-mail address: Kelly.Chibale@uct.ac.za (K. Chibale).
http://dx.doi.org/10.1016/j.bmcl.2014.06.012
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

3264
W. A. Andayi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3263–3267
hybridization led to enhanced activity against drug resistant
Plasmodium falciparum.18 In a bid to optimize the antiplasmodial
activity of these hybrid compounds, basic amino groups were
introduced in the side chain adjacent to the 3,4-HPO moiety. This
modification was intended to enhance accumulation of the drug
in the parasitic food vacuole, lysosomes and other acidic intracellu-
lar vacuoles.19,20
preparative HPLC. Deprotection was achieved with palladium-
catalysed hydrogenolysis or by acid-catalysed hydrolysis. The
target compounds (7a–7j and 8a–8j) were characterised by NMR,
HR-MS, elemental analysis and mp. The elemental analyses indi-
cated the molecules to be hydrated and this was further corrobo-
rated by the proton NMR spectra and the X-ray crystallographic
data of one of the compounds; 7c (Figs. 1 and 2).
The antiplasmodial effect of structural modification of these
3,4-HPO-chloroquine hybrids by incorporation of a basic amino
group in the side chain is further explored in this communication
using kojic acid. Herein we report the synthesis and mechanistic
investigation of kojic acid-derived 3,4-HPO-chloroquine hybrid
compounds and investigation of their mode of antiplasmodial
activity with respect to inhibition of haemozoin formation.21 Kojic
acid was chosen as a precursor to the 3,4-HPO scaffold due to its
synthetic accessibility and ease of synthetic manipulation.22
The Synthesis of the target compounds 1, 2, 3, 4, 5 and
N-(7-chloro-4-quinolinyl)diaminoalkanes 6a–6e, as illustrated in
Scheme 1 is based on documented information with some modifi-
cations.19,23–26 Chemoselective O-benzylation of the 5-hydroxyl
group in the presence of the 2-hydroxymethyl substituent was
observed. The Michael addition of methylamine or cyclopropyl-
amine gave the pyridinone 2 or 3. Chlorination of the pyridinones
using neat thionyl chloride afforded the alkylhalide intermediates
4 or 5. These intermediates were easily isolated as precipitates or
crystals in good yields. The final step to the benzyl protected
The X-Ray crystallographic analysis of 7c indicated that one
methanol and three water molecules were incorporated for each
molecule of 7c (MÁ3H2OÁCH3OH). The methanol molecule is disor-
dered by alternating of the positions of C and O, which were refined
isotropically due to their large thermal motions. Two of the water
molecules are disordered with oxygen over two positions. The
hydrogen atoms on these disordered methanol and water mole-
cules were excluded from the final structure model. For the main
molecule, all non-hydrogen atoms were refined anisotropically
and all hydrogen atoms on carbons were positioned geometrically
with C–H distances ranging from 0.95 Å to 1.00 Å and refined as rid-
ing on their parent atoms, with Uiso (H) = 1.2–1.5 Ueq (C). The posi-
tion of amine hydrogen H2N (on N-2) was located in the difference
electron density maps and refined with simple bond length con-
straints. The structure was refined successfully with a R factor of
0.0579.
The parameters for crystal data collection and structure refine-
ments are presented in the supplementary data. The data was also
deposited at the the Cambridge Crystallographic Data Centre. The
deposit number is CCDC 1006985. Further analysis showed that
the heterocyclic ring is not regular as it has two C–C [1.436 (3),
1.419(3)], two C@C [1.364(3), 1.343(2)] and two C–N [1.371(3),
1.353(3)], bonds, respectively. The C(17)–O(1) bond (1.276 Å) is
significantly longer than a pure ketone C@O bond (1.210 Å),25 this
provides O(1) with a partial negative charge that is used to form
strong hydrogen bonds. The C(17)–O(1) bond length is similar to
hybrids involved reaction of
4 or 5 with the appropriate
N-(7-chloro-4-quinolinyl) diaminoalkane (6a–6e). The nucleo-
philic substitution reaction was successful under both microwave
and reflux conditions. A combination of NaHCO3 and triethylamine
were used as bases to optimize the neutralization of the HCl
byproduct. Purification of the benzylated conjugates was achieved
by a combination of column chromatography, crystallization and
Scheme 1. (i) BnCl (1.1 equiv), NaOH (1.1 equiv), EtOH, reflux, 24 h; (ii) CH3NH2 or cyclopropylamine (1.8 equiv), NaOH (pH >10), 50% aq EtOH, reflux, 12 h; (iii) SOCl2
(8 equiv), À7 °C; stir, 2–12 h, to ambient temperature; (iv) N-(7-chloro-4-quinolinyl) diaminoalkane (1.2 equiv), DMF, Et3N (1.2 equiv), Na2CO3 (1.2 equiv), reflux 2-24 h or N-
(7-chloro-4-quinolinyl) diaminoalkane (1 equiv), AcN, NaOH (s) (1.1 equiv), microwave, 18 min., 100 °C, 250W, 249 bars; (v) 2 M ethanolic HCl, reflux 74 °C or H2O/EtOH/
concd HCl 1:2:3, H2, Pd/C, 4 atm., 4–6 h. Note: structural details of the target compounds are in Table 2 and in the Supplementary information.

W. A. Andayi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3263–3267
3265
Figure 1. Crystal structure of 7c.
Figure 2. Projection viewed along b. Note the disordered solvent molecules (methanol and water) in the crystal of 7c.
the corresponding bond length in the pyrone precursor, kojic acid
(1.244 Å) and deferiprone (1.278 Å) closely related pyridi-
The most lipophilic compound 7j, as determined from clogP,29
exhibited the best b-haematin inhibition activity (Table 2). How-
ever, no trend or correlation was observed between b-haematin
inhibition and lipophilicity in both deprotected and benzylated
a
none.27,28 The ketone oxygen O(1) is not protonated because pro-
tonation would have resulted in the extension of the C(17)–O(1)
bond length to ꢀ1.346 Å, the typical length of a protonated ketonic
oxygen in a hydroxypyridinone.25 Protonation is not expected
since the compound was obtained as a free base and no acid was
used in the recrystallization process. The water molecules are
shown to participate in intramolecular hydrogen bonding along
with the ketonic oxygen O(1) and the quinoline nitrogens N(1)
and N(2) (Table 1).
analogues. Generally, deprotection resulted in
a decrease in
b-haematin inhibition activity of these compounds except for 7a.
This observation implies the involvement of the benzyl group in
the inhibition of b-haematin formation. The b-haematin inhibition
activity of all compounds was superior or comparable to that of CQ
(1.9 IC50 equiv) except for some deprotected analogues; 8c (6.12
IC50 equiv), 8h (5.24 IC50 equiv) 8g (2.62 IC50 equiv) and 8i (2.57
IC50 equiv).
The introduction of the tertiary amino group resulted in a
decrease in b-haematin inhibition activity in both the benzylated
and deprotected analogues, for example, 7b (IC50 equiv = 1.3) com-
pared to 7c (IC50 equiv = 1.98); 7g (IC50 equiv = 0.22) compared to
7h (IC50 equiv = 1.87); 8g (IC50 equiv = 2.62) compared to 8h (IC50
equiv = 5.24). The only exception was the deprotected analogue,
8b (6.12 IC50 equiv) when compared to 8c (1.95 IC50 equiv).
Table 1
Hydrogen bonds for 7c [A and deg.]
D-H. . .A
d(D–H)
d(H. . .A)
d(D. . .A)
<(DHA)
N(2)–H(2N). . .O(1)#1
O(1W)–H(1W1). . .O(1)
O(1W)–H(1W2). . .N(1)#2
0.956(10)
0.966(10)
0.961(10)
2.011(16)
1.813(14)
1.779(11)
2.908(2)
2.767(3)
2.740(3)
155(3)
169(4)
178(4)

3266
W. A. Andayi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3263–3267
Table 2
In vitro antiplasmodial and b-haematin inhibition activities of target compoundsa
Compd
n
R1
R2
R5
cLogP
bHIAb IC50 (equiv)
3D7 IC50
(l
M)
K1 IC50
(
lM)
RIc
7a
8a
7b
8b
7c
8c
7d
8d
7e
8e
7f
1
1
2
2
2
2
3
3
5
5
1
1
2
2
2
2
3
3
5
5
Me
Me
Me
Me
Me
Me
Me
Me
H
H
H
H
Me
Me
H
H
H
H
H
H
H
Bn
H
Bn
H
Bn
H
Bn
H
Bn
H
Bn
H
Bn
H
Bn
H
2.7
1.0
3.2
1.5
3.7
2.1
3.7
2.0
4.7
3.0
3.4
1.6
3.9
2.0
4.4
2.6
4.3
2.6
5.3
3.6
1.27
0.63
1.30
6.12
1.98
1.95
0.42
ND
0.51
0.70
0.56
1.41
0.22
2.62
1.87
5.24
0.82
2.57
0.071
1.39
0.54
0.58
0.30
0.12
0.39
0.04
0.24
0.67
0.04
0.19
0.21
0.23
0.19
0.20
0.15
0.30
0.16
0.17
0.004
0.03
0.009
0.016
2.56
1.75
2.08
0.27
0.15
0.04
1.71
0.07
0.32
0.32
1.45
0.36
0.27
0.72
0.97
0.46
0.85
0.36
0.03
0.08
0.004
0.20
4.7
3.0
6.9
2.2
0.4
1.0
7.1
0.1
8
1.7
7.2
1.6
1.4
3.6
6.5
1.5
5.3
2.1
7.5
2.7
0.4
4.7
Me
Me
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
Cyclopropyl
8f
7g
8g
7h
8h
7i
8i
7j
8j
ART
CQ
H
Me
Me
H
H
H
Bn
H
Bn
H
H
5.3
1.9
a
b
c
SE 6 7%, n = 3.
bHIA, b-haematin inhibition activity.
RI, resistance index calculated as [IC50(K1)/IC50(3D7)], ND, not determined. cLogP, calculated partition coefficient.
The b-haematin inhibition activity was generally observed to
series in the resistant strain. This is clear when 7c (IC50 = 0.15
is compared to 7b (IC50 = 2.08 M) and 8c (IC50 = 0.04 M) com-
pared to 8b (IC50 = 0.12 M).
lM)
improve on replacement of the pyridinone N-methyl with a
N-cyclopropyl group with one exceptional pair; 7i and 7d.
Antiplasmodial activity against the 3D7 strain was observed to
increase with increase in lipophilicity. Only compound 7j
l
l
l
Compound 8c was identified as the only hybrid with a tertiary
amine group that had its b-haematin inhibition potency correlate
with to antiplasmodial potency in both the sensitive and resis-
tant strain. All the N-methylpyridinones (except 8b) were stron-
ger b-haematin inhibitors than 8c but the latter had superior
antiplasmodial activity. Surprisingly the most potent b-haematin
inhibitor 7d among the N-methylpyridinones was not the most
potent antiplasmodial. It may be assumed that factors, other than
b-haematin inhibition are responsible for the antiplasmodial
activity of 8c. Presumably accumulation of this compound in
the parasitic food vacuole is enhanced by the presence of the ter-
tiary amino group.30 Furthermore compounds 7c and 8c exhib-
ited significantly higher antiplasmodial activity in the K1 strain
than related analogues that lack the tertiary amine group (7b
and 8b).
(IC50 = 0.004
lM) was more active than chloroquine (IC50 = 0.016
l
M) against the sensitive strain 3D7. A decrease or an increase in
antiplasmodial activity against the 3D7 strain upon deprotection
was found to be compound specific.
The decreased b-haematin inhibition did not translate to lower
antiplasmodial activity in all the deprotected analogues. For
instance significant differences in b-haematin inhibition activities
of compounds with the cyclopropyl group were observed but their
antiplasmodial activities in the 3D7 strain appeared to be similar
for most of them (except for compounds 7j and 8j). This may imply
that inhibition of haemozoin formation may not be the only mode
of action responsible for their antiplasmodial activity or that differ-
ences in cellular uptake are involved.
Among the benzylated analogues, antiplasmodial activity in the
resistant K1 strain was observed to increase with increase in lipo-
philicity (length of alkyl side chain). However, no trend could be
discerned for the deprotected analogues. The antiplasmodial activ-
ity was found to improve on deprotection for most compounds
implying susceptibility of the resistant strain to, presumably, the
hydroxypyridinone moiety. This is further corroborated by the
low resistance indices observed for deprotected analogues when
compared to the benzylated ones (Table 2). In general many com-
pounds showed cross resistance with chloroquine in being less
active in the resistant strain than in the sensitive strain. However,
five compounds, 8c, 7c, 7j, 8d and 8j were more potent in the resis-
tant strain than chloroquine.
All the deprotected compounds in the series had lower resis-
tance indices (RIs) than chloroquine suggesting that they are likely
to be active against resistant parasites. Generally the deprotected
analogues exhibited lower RI values than their benzylated ana-
logues and the introduction of the tertiary amine group improved
the resistance indices of hybrid compounds with the N-cyclopropyl
group.
Acknowledgments
This work was supported by the South African Malaria initiative
SAMI (T.J.E), the South African Medical Research Council and the
South African Research Chairs Initiative of the Department of
Science and Technology administered by the South African
National Research Foundation (K.C.).
In contrast to what was observed in the sensitive strain, the
introduction of a tertiary amine group enhanced activity of the

W. A. Andayi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 3263–3267
3267
12. Hershko, C.; Link, G.; Pinson, A.; Peter, H. H.; Dobbin, P.; Hider, R. C. Blood 1991,
77, 2049.
Supplementary data
13. Heppner, D. G.; Hallaway, P. E.; Kontoghiorghes, G. J.; Eaton, J. W. Blood 1988,
72, 358.
14. Raventos-Suarez, C.; Pollack, S.; Nagel, R. L. Am. J. Trop. Med. Hyg. 1982, 31, 919.
15. Njogu, P. M.; Chibale, K. Curr. Med. Chem. 2013, 20, 1715.
16. Decker, M. Curr. Med. Chem. 2011, 18, 1464.
17. Vandekerckhove, S.; D’hooghe, M. Bioorg. Med. Chem. 2013, 21, 3643.
18. Andayi, W. A.; Egan, T. J.; Gut, J.; Rosenthal, P. J.; Chibale, K. ACS. Med. Chem.
Lett. 2013, 4, 642.
19. Dehkordi, L. S.; Liu, Z. D.; Hider, R. C. Eur. J. Med. Chem. 2008, 43, 1035.
20. Egan, T. J.; Kaschula, C. H. Curr. Opin. Infect. Dis. 2007, 20, 598.
21. Pirselova, K.; Balaz, S.; Sturdik, E.; Ujhelyova, R.; Veverka, M.; Uher, M.; Brtko, J.
Quant. Struct. Act. Relat. 1997, 16, 283.
22. Ncokazi, K. K.; Egan, T. J. Anal. Biochem. 2005, 338, 306.
23. Storr, T.; Mitchell, D.; Buglyo, P.; Thompson, K. H.; Yuen, G. V.; McNeill, J. H.;
Ovig, C. Bioconjugate Chem. 2003, 14, 212.
24. Ma, Y.; Luo, W.; Quinn, P. J.; Liu, Z.; Hider, R. C. J. Med. Chem. 2004, 47, 6349.
25. Dobbin, P.; Hider, R. C.; Hall, A. D.; Taylor, P. D.; Sarpong, P.; Porter, J. B.; Xiao,
G.; Helm, D. J. Med. Chem. 1993, 36, 2448.
26. Solomon, V. R.; Haq, W.; Srivastava, K.; Puri, S. K.; Katti, S. B. J. Med. Chem. 2007,
50, 394.
27. Lokaj, J.; Kozisek, J.; Koren, B.; Uher, M.; Vrabel, V. Acta Crystallogr. 1991, C47,
193.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.06.
012. These data include MOL files and InChiKeys of the most
important compounds described in this article.
References and notes
1. World Health Organization. World Malaria Report. 2012, Geneva, Switzerland.
2. Wells, T. N. C.; Alonso, P. L.; Gutteridge, W. E. Nat. Rev. Drug Disc. 2009, 8, 879.
3. Dondorp, A. M.; Norsten, F.; Yi, P.; Das, D.; Phyo, A. P.; Tarning, J.; Lwin, K. M.;
Ariey, F.; Hanipithithakpong, W.; Lee, S.; Ringwald, P.; Silamut, K.; Imwong, M.;
Chotivanich, K.; Lim, P.; Herdman, T.; An, S. S.; Yeung, S.; Singhasivanon, P.;
Day, N. P. J.; Lindegardh, N.; Socheat, D.; White, N. J. N. Engl. J. Med. 2009, 361,
455.
4. Wiwanitkit, V. Int. J. Gen. Med. 2010, 3, 327.
5. Burrows, J. N.; Chibale, K.; Wells, T. N. Curr. Top. Med. Chem. 2011, 11, 1226.
6. Weinberg, E. D.; Moon, J. Drug Metab. Rev. 2009, 41, 644.
7. Barrett, J. F. R.; Whittaker, P. G.; Williams, J. G.; Lind, T. Br. J. Med. 1994, 309, 79.
8. Berger, J.; Dyek, J. L.; Galou, P.; Aplogan, A.; Schneider, D.; Traissac, P. Eur. J. Clin.
Med. 2000, 54, 29.
9. Murray, M. J.; Murray, M. B.; Murray, C. J. Br. J. Med. 1978, 2, 1113.
10. Cabantchik, Z. I.; Glickstein, H.; Golenser, J.; Loyevsky, M.; Tsafack, A. Acta
Haematol. 1996, 95, 70.
28. Clarke, E. T.; Martell, A. E.; Reibenspies, J. Inorg. Chim. Acta 1992, 196, 177.
29. MoKa software from www.Moldiscovery.com.
30. Egan, T. J. Mini-Rev. Med. Chem. 2001, 1, 113.
11. Hershko, C.; Theanacho, E. N.; Spira, D. T.; Peter, H. H.; Dobbin, P.; Hider, R. C.
Blood 1991, 77, 637.