Target Guided Synthesis of 5-Benzyl-2,4-diamonopyrimidines: Their Antimalarial Activities and Binding Affinities to Wild Type and Mutant Dihydrofolate Reductases from Plasmodium falciparum

Article abstract of DOI:10.1021/jm0303352

The resistance to pyrimethamine (PYR) of Plasmodium falciparum arising from mutation at position 108 of dihydrofolate reductase (pfDHFR) from serine to asparagine (S108N) is due to steric interaction between the bulky side chain of N108 and C1 atom of the 5-p-C1 aryl group of PYR, which consequently resulted in the reduction in binding affinity between the enzyme and inhibitor. Molecular modeling suggested that the flexible antifolate, such as trimethoprim (TMP) derivatives, could avoid this steric constraint and should be considered as new, potentially effective compounds. The hydrophobic interaction between the side chain of inhibitor and the active site of the enzyme around position 108 was enhanced by the introduction of a longer and more hydrophobic side chain on TMP's 5-benzyl moiety. The prepared compounds, especially those bearing aromatic substituents, exhibited better binding affinities to both wild type and mutant enzymes than the parent compound. Binding affinities of these compounds correlated well with their antimalarial activities against both wild type and resistant parasites. Molecular modeling of the binding of such compounds with pfDHFR also supported the experimental data and clearly showed that aromatic substituents play an important role in enhancing binding affinity. In addition, some compounds with 6-alkyl substituents showed relatively less decrease in binding constants with the mutant enzymes and relatively good antimalarial activities against the parasites bearing the mutant enzymes.

Full text of DOI:10.1021/jm0303352

J . Med. Chem. 2004, 47, 345-354
345
Ta r get Gu id ed Syn th esis of 5-Ben zyl-2,4-d ia m on op yr im id in es: Th eir
An tim a la r ia l Activities a n d Bin d in g Affin ities to Wild Typ e a n d Mu ta n t
Dih yd r ofola te Red u cta ses fr om P la sm od iu m fa lcipa r u m
Chawanee Sirichaiwat,†,‡ Chakapong Intaraudom,† Sumalee Kamchonwongpaisan,† J arunee Vanichtanankul,†
Yodhathai Thebtaranonth,*,†,‡ and Yongyuth Yuthavong†
National Center for Genetic Engineering and Biotechnology, National Science and Technology Development Agency,
113 Thailand Science Park, Pahonyothin Road, Klong 1, Klongluang, Pathumtani 12120, Thailand, and Department of
Chemistry, Faculty of Science, Mahidol Univerity, Rama 6 Road, Bangkok 10400, Thailand
Received J uly 11, 2003
The resistance to pyrimethamine (PYR) of Plasmodium falciparum arising from mutation at
position 108 of dihydrofolate reductase (pfDHFR) from serine to asparagine (S108N) is due to
steric interaction between the bulky side chain of N108 and Cl atom of the 5-p-Cl aryl group
of PYR, which consequently resulted in the reduction in binding affinity between the enzyme
and inhibitor. Molecular modeling suggested that the flexible antifolate, such as trimethoprim
(TMP) derivatives, could avoid this steric constraint and should be considered as new, potentially
effective compounds. The hydrophobic interaction between the side chain of inhibitor and the
active site of the enzyme around position 108 was enhanced by the introduction of a longer
and more hydrophobic side chain on TMP’s 5-benzyl moiety. The prepared compounds, especially
those bearing aromatic substituents, exhibited better binding affinities to both wild type and
mutant enzymes than the parent compound. Binding affinities of these compounds correlated
well with their antimalarial activities against both wild type and resistant parasites. Molecular
modeling of the binding of such compounds with pfDHFR also supported the experimental
data and clearly showed that aromatic substituents play an important role in enhancing binding
affinity. In addition, some compounds with 6-alkyl substituents showed relatively less decrease
in binding constants with the mutant enzymes and relatively good antimalarial activities
against the parasites bearing the mutant enzymes.
In tr od u ction
malaria treatment. Unfortunately, resistance of the
parasite to these drugs occurred rapidly, and hence,
their clinical utility has been drastically affected. The
mechanism of resistance has been shown to be due to
the various point mutations of DHFR, including those
at positions 16, 50, 51, 59, 108, 140, and 164.3-7
Malaria, one of the most widespread tropical diseases,
is caused by parasites of Plasmodium spp., of which
falciparum is the most lethal. The emergence of resis-
tant parasites to various available drugs has become a
very serious problem in malaria control.1,2 Plasmodium
falciparum dihydrofolate reductase (pfDHFR), which is
an essential enzyme in the folate pathway and exists
as a bifunctional enzyme linked to thymidylate synthase
(TS), has long been an important target for malarial
therapy. Pyrimethamine (PYR, 1) and cycloguanil (CYC,
During the past few years, molecular modeling of
pfDHFR has extensively been studied in order to
understand the basis of drug resistance that can lead
to the development of new effective inhibitors.8-11 From
a homology model based on leishmania DHFR, McKie
et al. pointed out that S108N mutation had led to steric
clash with the p-Cl atom of PYR.9 Results from other
molecular modeling studies of various different species
of DHFRs also gave a similar explanation for the
resistance of this mutant.8,10,11
It has been suggested that the flexible antifolate could
avoid this steric constraint and should be an effective
inhibitor.12 This was demonstrated in the case of
WR99210, the triazine derivative with a flexible trichlo-
rophenoxypropyloxy side chain, which is effective against
PYR-resistant parasites.13,14 Trimethoprim (TMP, 3)
also possesses a flexible benzyl side chain, and it was
shown that the binding affinity of TMP analogues to
mutant enzymes was relatively less affected than PYR,
although the relative influence of side chain flexibility
must be weighed against other influences of structural
changes in the various derivatives.15 It was also shown
that derivatives with a benzyloxy substituent on the
5-benzyl group had favorable binding characteristics.
2), the active metabolite of proguanil, are potent DHFR
inhibitors and were effectively used for P. falciparum
* To whom correspondence should be addressed. Phone: 662-
2015135. Fax: 662-2458332. E-mail: scytr@mahidol.ac.th.
† National Science and Technology Development Agency.
‡ Mahidol University.
10.1021/jm0303352 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/16/2003

346 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2
Sirichaiwat et al.
F igu r e 1. Predicted conformation of TMP in the binding pocket of wild type P. falciparum DHFR: (a) overall folding of the
enzyme; (b) conformation of TMP with NADPH; (c) superimposition of TMP (red) and WR99210 (green).
by our group,15 and the synthetic route is shown in
Sch em e 1
Scheme 1. Derivatives of TMP with the 5-benzyl sub-
stituent were synthesized by aldol condensation be-
tween 3-morpholinopropionitrile and substituted benz-
aldehyde, followed by treatment of the intermediates
with guanidine. In the cases of derivatives with 6-alkyl
substituents, an additional methylation step, using
diazomethane, was needed prior to the treatment with
guanidine. For derivatives bearing long ether side
chains, the preparation of starting aldehydes is outlined
in Scheme 2. 4-(3-Hydroxypropoxy)-3-substituted benz-
aldehyde (14) and 3-substituted phenoxy-1-propanol
(17) were synthesized by alkylation of the corresponding
hydroxybenzaldehyde (12) or phenol (16) with 3-chloro-
1-propanol using K2CO3/DMF at 80 °C. Treatment of
compound 14 with p-toluenesulfonyl chloride at 0 °C
gave the tosylated product 15 as a white solid. Synthesis
of compound 18 was achieved by alkylation of compound
17 with tosylated 15 under NaH/DMF conditions.
Molecu la r Mod elin g of p fDH F R Bou n d w it h
TMP . Since the structures of pfDHFRs were solved
during the course of this research, molecular modeling
of trimethoprim with the enzyme was conducted with
these solved structures in order to understand the mode
of binding of the inhibitor in the active site pocket.
Predicted conformations of this compound in the active
site pocket of wild type and quadruple mutant pfDHFRs
are shown in Figures 1 and 2. The benzyl side chain of
the inhibitor was found to be oriented in the “up”
conformation, bending away from NADPH, as observed
in eucaryotes, whereas in bacterial DHFR the benzyl
group was pointing “downward” toward the nicotina-
mide ring.17,18 In this case, the inhibitor was positioned
more closely to NADPH than in the bacteria, and
adjustment to the “down” conformation led to steric
clash with the nicotinamide ring of NADPH. It is noted
These implied that the aromatic group at the 5-benzyl
moiety played an important role in the binding affinity.
To explore the significance of aromatic and other
hydrophobic substituents on the 5-benzyl moiety, the
binding affinity of various TMP derivatives bearing
these substituents with both wild type and mutant
enzymes was extensively studied. Moreover, molecular
modeling of the binding of TMP derivatives with the
active site of the recently solved pfDHFR16 was also
studied.
Resu lts a n d Discu ssion
Ch em ica l Syn th eses. Preparation of TMP analogues
was achieved by the improved method recently reported
Sch em e 2

5-Benzyl-2,4-diamonopyrimidines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2 347
F igu r e 2. TMP in the binding pocket of pfDHFRs with selected amino acid residues: (a) wild type; (b) quadruple mutant.
that the conformation of TMP in the active site pocket
of the enzyme reported in this study is very similar to
the established conformation of WR99210 in the same
active site (Figure 1c).16
to this mutant enzyme when compared to the results of
the wild type.
In h ibition of th e Wild Typ e a n d Mu ta n t p fDH-
F Rs by TMP Der iva tives. From the characteristics
of binding discussed, we propose that aromatic and
other hydrophobic substituents on the 5-benzyl moiety
of TMP would increase the binding affinity, resulting
in enhanced enzyme inhibition and parasite killing. The
trimethoprim derivatives with these substituents were
subjected to screening for inhibitory activity with wild
type, S108N, C59RS108N double mutant, C59RS108-
NI164L triple mutant, and N51IC59RS108NI164L qua-
druple mutant DHFRs. Ki values for these compounds,
binding with wild type and mutant enzymes, are shown
in Tables 2-5 along with Ki values of trimethoprim for
comparison. Effects of substituents on the inhibitors in
their binding to both wild type and mutant DHFRs were
assessed by determination of the ratios of inhibition
constants for the mutants compared to the wild type
enzyme [Ki(mutant)/Ki(wild type)], as well as their Ki
values relative to trimethoprim.
The Ki values of trimethoprim analogues having
various long-chain alkoxy substituents are shown in
Table 2. In the case of compounds 20-22, it was noted
that the increase of carbon atoms at the p-alkoxy side
chain resulted in a decrease of Ki values for both wild
type (wt) and mutant enzymes, compared to the simple
p-methoxy substituent (compound 19). Binding affinities
of derivatives with substituents at positions 3 and 4,
compounds 23-27, in which one of these two substitu-
ents was a long alkoxy chain, improved the binding by
about 5- to 10-fold against both wild type and mutant
enzymes. In Table 3, trimethoprim derivatives with
aromatic substituents showed relatively more favorable
binding to both wild type and mutant enzymes, espe-
cially for compound 40, which is about 30-fold and 60-
to 200-fold more effective than trimethoprim against
wild type and mutant enzymes, respectively, as has been
reported in the previous study.15 In addition, the change
from H to an alkoxy group at the meta position
improved binding affinity to both wild type and mutant
enzymes (e.g., compounds 38-41) while the Ki values
As Figures 1a and 2a show, the inhibitor was bound
in the active site of the enzyme with its pyrimidine ring
occupying the interior of the deep cleft while its benzyl
side chain extended toward the entrance of the hydro-
phobic binding cavity. This inhibitor was held in the
pocket by a combination of van der Waals, hydrogen
bonding, and ionic interactions with the protein. The
key interaction was the hydrogen bonding between
carboxylate side chain of Asp54 and both the N1 and
2-amino groups of the pyrimidine ring. Furthermore, the
backbone carbonyl oxygens of Ile14 and 164 were located
in the plane of the pyrimidine ring, which formed a
hydrogen bond with the 4-amino group of the inhibitor.
Additional major hydrophobic contacts occurred between
the pyrimidine ring and the side chain of Phe58 and
Leu46, located above the ring and adjacent to its
exterior edge, respectively. Three methoxy groups were
in van der Waals contact with the side chain of Met55,
Leu46, Ile164, and the nicotinamide ring of NADPH.
The model showed that interaction between the side
chain of the inhibitor and amino acid at residue 108 and
the π-π interaction caused by the stacking between the
phenyl ring of the inhibitor and the nicotinamide ring
of NADPH, which took place in case of PYR and CYC,8,10
were absent and therefore resulted in poor binding. To
produce optimal interaction, addition of longer and more
hydrophobic substituents was required. However, the
space around methoxy group at position 5 was quite
limited for modification. Consequently, modification was
focused at positions 3 and 4 on the benzyl moiety.
It is shown in Figure 2b that in the case of the
quadruple mutant enzyme there is no interaction be-
tween the benzyl side chain and the nearby amino acids
Met55 and Leu164. In addition, the distance between
the backbone oxygen of Leu164 and the 4-amino group
of the pyrimidine ring increased significantly, resulting
in weaker hydrogen-bonding interaction. All these
results are consistent with poor binding of the inhibitor

348 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2
Ta ble 1. Data of TMP Analogues
Sirichaiwat et al.
compd
R1
R2
R3
R4
yield (%)
mp (°C)
formula
m/z: [M + 1]+
anal.
19
20
21
22
23
24
25
26
27
28
29
30a
31a
32a
33
34
35
36
37a
38a
39a
40a
41a
42
43a
44
45
46
47
48
49
50
51
52
53
54
55a
56a
57
H
H
H
H
OMe
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
68
74
65
68
63
65
71
77
74
65
85
84
65
63
67
76
79
89
89
90
83
65
87
85
80
87
83
77
84
66
65
68
89
90
91
87
92
91
89
206.5-207.5
164-165
C
C
C
C
C
C
C
C
C
C
C
12H14N4O
14H18N4O
16H22N4O
18H26N4O
15H20N4O2
16H22N4O2
17H24N4O2
19H28N4O2
18H26N4O2
14H18N4O2
13H16N4O2
231.1221
259.1569
287.1867
315.2184
289.1656
303.1826
317.1980
345.2297
331.2136
275.1520
261.1349
337.1672
351.1828
413.1983
307.1569
307.1562
293.1408
337.1665
397.1905
427.1985
441.2133
455.2296
469.2447
335.1870
365.1975
379.2135
351.1823
381.1931
395.2084
409.2240
439.2345
453.2506
305.1612
319.1769
351.1816
365.1970
427.2130
441.2290
483.2650
CHN
CHN
CHN
CHN
HNb
OPrn
OPenn
OHepn
OMe
OMe
OMe
165-166
161.5-162.5
151-152
153.5-154.5
161-162
OPrn
OBun
OPenn
OHepn
OEt
CHc
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHd
OMe
138.5-139.5
151.5-152
194.5-195.5
234.5-235
162.5-163
173.5-174
137-137.5
201.5-202
154-155
OPenn
OEt
OMe
OMe
OMe
OEt
OCH2Ph
H
OCH2Ph
OPh
OCH2Ph
H
OMe
OCH2Ph
OCH2Ph
OCH2Ph
OCH2Ph
H
C19H20N4O2
C20H22N4O2
C25H24N4O2
C18H18N4O
C18H18N4O
H
OMe
202-203
C17H16N4O
214.5-215.5
174-174.5
155.5-156
153-154
C19H20N4O2
C21H24N4O4
C22H26N4O5
C23H28N4O5
C24H30N4O5
C25H32N4O5
C23H28N4O5
C21H24N4O2
C22H26N4O2
C20H22N4O2
C21H24N4O3
C22H26N4O3
C23H28N4O3
C24H30N4O4
C25H32N4O4
OCH2-[3,4,5-(OMe)3]Ph
OCH2-[3,4,5-(OMe)3]Ph
OCH2-[3,4,5-(OMe)3]Ph
OCH2-[3,4,5-(OMe)3]Ph
OCH2-[3,4,5-(OMe)3]Ph
OC3H6Ph
OC3H6Ph
OC3H6Ph
OC3H6OPh
OC3H6OPh
OMe
OEt
OPrn
OBun
H
143.5-144.5
154.5-155.5
167-168
134.5-135.5
137-138
151.5-152
154-154.5
141.5-142
108.5-109
131.2-131.8
147.5-148
193-194
218.219
180-181
176-177
153-153.5
114.5-115
161-162
OMe
OEt
H
OMe
OEt
H
OC3H6OPh
OC3H6OC3H6OPh
OC3H6OC3H6OPh
OC3H6OC3H6OPh
OMe
OMe
OCH2Ph
OCH2Ph
OCH2Ph
OCH2Ph
OCH2-[3,4,5-(OMe)3]Ph
OMe
OEt
CHN
CHN
CHN
CHN
HNe
CHN
CHN
CHN
HNf
OMe
OMe
OMe
OMe
OCH2Ph
OCH2Ph
OPrn
OMe
OMe
H
H
H
Me
Et
Me
Et
Me
Et
Et
C
C
15H20N4O3
16H22N4O3
C20H22N4O2
C21H24N4O2
C26H26N4O2
C27H28N4O2
C26H34N4O5
H
H
a
b
d
Data from ref 13. C: calcd, 62.46; found, 61.76. c N: calcd, 18.54; found, 19.20. N: calcd, 13.72; found, 14.59. e C: calcd, 68.55;
found, 67.82. f C: calcd, 64.69; found, 63.85.
Ta ble 2. Ki Values of Trimethoprim Derivatives with Various Long Alkoxy Chain Substituents
rel to
TMP
Ki C59RS108N
rel to
TMP
Ki C59RS108NI164L
rel to
TMP
Ki N51I C59RS108NI164L
rel to
TMP
compd
Ki wt (nM)
(nM)
(nM)
(nM)
TMP
19
20
21
22
23
24
25
26
27
28
29
10.3 ( 0.5
23.5 ( 8.0
15.1 ( 1.5
8.6 ( 0.6
17.8 ( 0.5
1.8 ( 0.1
2.5 ( 0.2
2.2 ( 0.2
3.1 ( 0.1
1.6 ( 0.3
15.4 ( 0.8
11.5 ( 0.6
1.0
2.3
1.5
0.8
1.7
0.2
0.2
0.2
0.3
0.1
1.5
1.1
242.1 ( 40.1
464.1 ( 59.1
211.5 ( 31.5
137.3 ( 18.5
196.3 ( 25.1
52.1 ( 10.0
194.1 ( 24.9
134.5 ( 18.5
103.5 ( 8.2
46.6 ( 5.7
1.0
1.9
0.9
0.6
0.8
0.2
0.8
0.5
0.4
0.2
0.8
1.0
5688.7 ( 557.1
3957.8 ( 1358.4
1091.3 ( 171.0
539.5 ( 103.3
2208.4 ( 229.0
144.4 ( 22.2
1.0
0.7
0.2
0.1
0.4
0.03
0.1
0.04
0.05
0.2
6664.5 ( 1516.3
20491.0 ( 5209.4
1518.0 ( 268.2
914.6 ( 147.4
1380.0 ( 190.8
919.6 ( 176.8
1033.3 ( 48.3
444.2 ( 71.9
1.0
3.1
0.2
0.1
0.2
0.1
0.2
0.1
0.1
0.1
1.8
2.1
771.8 ( 106.9
252.6 ( 1.0
292.4 ( 64.4
445.1 ( 67.2
1162.5 ( 638.1
3541.8 ( 1788.8
2646.8 ( 460.3
901.7 ( 225.5
11817.9 ( 950.6
13858.0 ( 970.7
187.4 ( 20.6
234.8 ( 32.6
0.6
0.5
decreased when the methoxy group (e.g., compounds 30
and 38) at this position was replaced by an ethoxy
functionality (e.g., compounds 31 and 39). These results
support our hypothesis concerning the addition of more
hydrophobic substituents to produce the optimal inter-
action and the importance of the aromatic ring on the
binding affinity.
(compounds 42-44) or ether chain (compounds 45-50)
had no significant effect on the binding affinity with the
enzymes, compared to derivatives having aromatic
substituents without any extension as shown in Table
3. Derivatization at the 6-position of the pyrimidine ring
by replacing an H atom with either a methyl or an ethyl
group did not improve the binding affinity to wild type
enzyme (Table 5) compared to those without the 6-alkyl
substituent (Table 3). However, in the case of compound
An attempt to extend the aromatic side chain (Table
4) by increasing the number of carbon atoms of the alkyl

5-Benzyl-2,4-diamonopyrimidines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2 349
Ta ble 3. Ki Values of Trimethoprim Derivatives with Various Aromatic Substituents
Ki wt
(nM)
rel to
TMP
Ki C59RS108N
rel to
TMP
Ki C59RS108NI164L
rel to
TMP
Ki N51I C59RS108NI164L
rel to
TMP
compd
(nM)
(nM)
(nM)
30
31
32
33
34
35
36
37
38
39
40
41
2.2 ( 0.5
0.8 ( 0.3
1.7 ( 0.2
6.5 ( 0.5
6.3 ( 0.6
5.3 ( 0.5
7.2 ( 0.9
1.3 ( 0.2
0.6 ( 0.0
0.3 ( 0.0
0.4 ( 0.2
0.4 ( 0.2
0.2
0.09
0.2
0.6
0.6
0.5
0.7
0.1
0.06
0.03
0.04
0.04
60.7 ( 8.7
9.7 ( 1.4
0.3
0.04
0.2
0.4
0.9
1974.1 ( 123.3
570.3 ( 70.1
491.2 ( 56.0
994.0 ( 289.2
515.4 ( 127.8
161.2 ( 16.2
629.8 ( 56.9
511.9 ( 62.9
47.6 ( 5.4
0.4
0.1
0.09
0.2
0.09
0.03
0.11
0.09
0.01
0.004
0.02
0.02
1375.7 ( 223.4
431.4 ( 112.4
2454. ( 257.1
4636.7 ( 730.8
455.5 ( 124.8
671.4 ( 281.9
1129.6 ( 55.7
76.5 ( 20.2
0.21
0.06
0.4
0.7
0.07
0.1
60.2 ( 2.6
101.0 ( 22.8
113.5 ( 17.6
81.7 ( 16.9
143.4 ( 17.8
5.2 ( 0.3
0.3
0.6
0.2
0.02
0.04
0.009
0.001
0.02
0.01
0.02
0.02
0.005
0.04
8.7 ( 1.0
124.0 ( 12.9
119.3 ( 22.5
35.8 ( 3.6
2.2 ( 0.1
24.7 ( 2.6
3.5 ( 0.8
88.6 ( 6.2
105.6 ( 8.4
5.6 ( 0.4
258.6 ( 39.3
Ta ble 4. Ki Values of Trimethoprim Derivatives with Various Long-Chain Aromatic and Ether Substituents
Ki wt
(nM)
rel to
TMP
Ki C59RS108N
rel to
TMP
Ki C59RS108NI164L
rel to
TMP
Ki N51I C59RS108NI164L
rel to
TMP
compd
(nM)
(nM)
(nM)
42
43
44
45
46
47
48
49
50
5.9 ( 1.1
1.8 ( 0.4
1.0 ( 0.3
6.5 ( 1.5
2.0 ( 0.6
0.6 ( 0.2
7.0 ( 0.5
1.8 ( 0.3
2.3 ( 0.4
0.57
0.17
0.10
0.63
0.19
0.06
0.68
0.18
0.22
58.1 ( 17.0
44.2 ( 2.7
0.44
0.33
0.08
0.70
0.15
0.2
0.45
0.63
0.19
3044.8 ( 595.7
3100.2 ( 456.9
374.1 ( 21.8
470.6 ( 28.4
256.9 ( 30.6
166.4 ( 7.1
263.6 ( 16.0
424.8 ( 21.9
313.7 ( 68.8
0.54
0.54
0.06
0.08
0.05
0.03
0.05
0.07
0.06
801.4 ( 210.4
1122.8 ( 644.8
525.3 ( 148.3
2261.0
605.8 ( 38.1
770.1
1589.0 ( 33.5
2191.7
937.4 ( 49.9
0.12
0.17
0.08
0.34
0.09
0.12
0.24
0.33
0.14
10.6 ( 1.2
169.9 ( 11.4
37.1 ( 0.8
47.7 ( 6.5
107.7 ( 11.6
153.5 ( 22.9
46.9 ( 2.2
Ta ble 5. Ki Values of Trimethoprim Derivatives with Various 6-Alkyl Substituents
Ki wt
(nM)
rel to
TMP
Ki C59RS108N
rel to
TMP
Ki C59RS108NI164L
rel to
TMP
Ki N51I C59RS108NI164L
rel to
TMP
compd
(nM)
(nM)
(nM)
51
52
53
54
55
56
57
38.2 ( 3.1
4.3 ( 0.6
3.4 ( 0.5
3.5 ( 0.5
8.6 ( 0.4
3.7 ( 0.9
1.8 ( 1.0
3.7
0.4
0.3
0.3
0.8
0.4
0.17
520.3 ( 122.8
116.4 ( 25.9
89.5 ( 12.7
60.1 ( 8.1
80.0 ( 10.3
99.1 ( 17.2
48.7 ( 6.95
2.2
0.5
0.4
0.25
0.3
1364.3 ( 132.1
466.8 ( 37.7
66.3 ( 14.9
3737.7 ( 1497.3
174.6 ( 24.1
273.4 ( 24.7
154.2 ( 42.0
0.24
0.08
0.01
0.66
0.03
0.05
0.03
1634.3 ( 362.6
697.6 ( 41.1
93.4 ( 2.6
732.5 ( 98.1
241.0 ( 36.6
204.7 ( 42.5
693.1 ( 125.6
0.25
0.11
0.01
0.11
0.04
0.03
0.1
0.4
0.20
52, small decreases in Ki values were observed (com-
pared to TMP). These results might be due to steric
clash of the larger substituent with the benzyl group,
resulting in conformational change that affected the
binding affinity to the enzyme. Some of these com-
pounds (i.e., 51-56) showed better binding affinity,
specifically with the mutant enzymes. The results
described above implied that the alkyl group at position
6 on the pyrimidine ring played a significant role in the
binding affinity with mutant enzymes.
In Vitr o An tip la sm od ia l Activity. All compounds
were also subjected to an in vitro malaria screening
system against wild type and resistant Plasmodium
falciparum parasites carrying various mutation en-
zymes, and some of these results are shown in Table 6.
Antiplasmodial activity (IC50 values) of compounds
described in this study ranged in the micromolar
regions. Although not as good as predicted according to
the enzyme inhibition constants, some exhibited IC50
values against both wild type and resistant parasites
at low micromolar levels (e.g., compounds 37-41).
Correlations between antimalarial and inhibitory activi-
ties of these analogues are shown in Figure 3, where
values of log Ki (of wild type, double, triple, and qua-
druple mutant enzymes) are plotted against the corre-
sponding log IC50. Antimalarial activity of almost all
compounds against wild type and resistant strains
correlated linearly with the ability of these compounds
to inhibit wild type and mutant DHFRs. This implied
that the drug target in the cell was the enzyme DHFRs.
However, lack of correlation between antimalarial and
inhibitory activities of some trimethoprim analogues
was also observed. For example, compounds 33-36
showed better Ki values than TMP for wild type enzyme
but exhibited less potency in antimalarial activity
against wild type parasite. The discrepancies from the
expected correlation between the Ki values and the
antimalarial activities can be due to various reasons,
including differences in drug transport into and out of
the parasite, access of the drug to the target enzyme,
and the presence of competing sites for drug binding.
Toxicity tests of these compounds to evaluate the
selective inhibition against three mammalian cell lines,
African green monkey kidney fibroblast (Vero cells),
human epidermoid carcinoma (KB), and human breast
cancer (BC), showed that the compounds are relatively
nontoxic compared to their activity against the malaria
parasites (see data in Supporting Information).
Molecu la r Mod elin g of Bin d in g of Tr im eth op r im
Der iva tives w ith p fDHF R. To understand further the
role of hydrophobic and aromatic substituents on the
binding affinity, the molecular modeling of compound
40 (which showed high binding affinity to both wild type
and mutant enzymes) bound with both wild type and
quadruple mutant enzymes was studied. Predicted
conformations of this compound in the active site pocket

350 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2
Sirichaiwat et al.
Ta ble 6. Antimalarial Activity of Some of Trimethoprim Derivatives, IC50 (µM)
compd
TM4/8.2a
rel to TMP
K1CB1b
rel to TMP
Csl-2c
rel to TMP
Vl/Sd
rel to TMP
30
31
32
33
34
35
36
37
38
39
40
41
51
52
53
54
55
56
4.66 ( 1.71
0.47 ( 0.10
3.12 ( 0.56
15.15 ( 5.10
15.22 ( 5.42
19.08 ( 5.98
17.48 ( 0.74
0.68 ( 0.21
0.31 ( 0.06
0.07 ( 0.02
0.15 ( 0.05
0.33 ( 0.06
22.73 ( 2.77
2.40 ( 0.67
6.84 ( 1.36
5.06 ( 1.45
4.48 ( 0.29
3.57 ( 0.52
0.70
0.07
0.47
2.27
2.28
2.86
2.62
0.10
0.05
0.01
0.02
0.05
3.41
0.36
1.03
0.76
0.67
0.54
>100e
>0.73
0.12
0.11
0.13
0.03
0.03
0.03
0.15
0.13
0.04
0.03
0.07
>0.73
>0.37
0.18
0.12
0.03
0.03
25.61 ( 1.26
14.36 ( 2.90
6.75 ( 4.28
15.26 ( 2.35
3.33 ( 0.19
3.517 ( 1.01
5.73 ( 1.94
19.85 ( 7.49
30.58 ( 7.47
13.15 ( 0.73
7.70 ( 4.34
3.40 ( 0.33
90.81 ( 2.28
0.21
0.12
0.06
0.13
0.03
0.03
0.05
0.17
0.26
0.11
0.06
0.03
0.76
>0.42
0.02
0.05
0.02
0.03
38.02 ( 1.69
16.47 ( 4.63
8.13 ( 1.53
<0.19
<0.08
<0.04
<0.11
<0.06
<0.15
0.25
<0.11
<0.13
<0.10
<0.05
<0.13
0.5
16.59 ( 1.85
14.58 ( 2.33
17.14 ( 6.56
3.52 ( 0.82
3.69 ( 0.51
3.83 ( 0.45
19.88 ( 3.79
27.76 ( 10.63
5.01 ( 1.53
4.29 ( 0.96
10.16 ( 4.23
21.49 ( 2.25
12.75 ( 3.18
29.61 ( 20.21
>50e
21.12 ( 0.09
26.08 ( 2.30
19.45 ( 0.19
9.75 ( 0.92
15.91 ( 5.68
>100e
>50e
25.24 ( 7.02
>100e
>50e
4.82 ( 1.87
>50e
0.25
2.85 ( 0.40
5.37 ( 1.78
2.64 ( 0.91
3.46 ( 0.51
<0.02
<0.06
<0.02
<0.02
16.26 ( 3.71
4.38 ( 1.69
4.18 ( 1.47
11.30 ( 0.82
3.11 ( 0.30
3.02 ( 0.29
a
b
Parasite strain with wild type DHFR. Parasite strain with double mutation (C59R + S108N) DHFR. c Parasite strain with triple
mutation (C59R + S108N + I164L) DHFR. Parasite strain with quadruple mutation (N51I + C59R + S108N + I164L) DHFR. e Maximun
concentration that the inhibitor could be dissolved in DMSO.
d
F igu r e 3. Correlation between (a) log Ki wt and IC50 TM4/8.2, (b) log Ki C59RS108N and IC50 K1CB1, (c) log Ki C59RS108NI164L
and IC50 Csl-2, (d) log Ki N51IC59RS108NI164L and IC50 Vl/S.
of wild type and quadruple mutant DHFRs are shown
in Figure 4.
good binding to both wild type and mutant enzymes.
For the quadruple mutant enzyme (Figure 4b), the
solved structure16 showed significant movement of
residues 48-51 and 164-166, which opened up the gap
between residues 50 and 164 in the active site. Accord-
ingly, the distance between the inhibitor and the nearby
amino acid side chains in this mutant enzyme was
larger than that observed in wild type enzyme. This
leads to a decrease in various interactions, for example,
H-bonding between the side chain of Cys50 and p-
methoxy oxygen of the benzyl substituent, van der
Waals interaction of Met55 and the methoxy group,
favorable stacking between the aromatic ring and the
phenyl side chain of Phe116, and H-bonding between
the backbone oxygen of Leu164 and the 4-amino group
of the pyrimidine ring, which resulted in a reduction of
binding affinity between the inhibitor and the enzyme.
Compounds with 6-alkyl substituents (52-56) showed
relatively less decrease in binding affinities with the
mutant enzymes and are therefore of higher potential
interest. To study the effect of the alkyl group at the
6-position of the pyrimidine ring on the binding affinity,
In Figure 4a, the model showed that the aromatic
substituent of this inhibitor was stacked against the
phenyl side chain of Phe116 and formed a π-π interac-
tion. This interaction led to better binding affinity of
trimethoprim derivatives having aromatic substituents
to both wild type and mutant enzymes than the deriva-
tives with alkoxy substituents. The additional three
methoxy groups on the benzyl substituent were in van
der Waals contact with the amino acid residues Pro113,
Phe116, Cys50, and Met55 near the entrance of the
active site. The p-methoxy oxygen was positioned 3.27
Å from the side chain of Cys50 and could have H-
bonding interaction. Although well-documented hydro-
gen bonds involving methoxy groups were rare, they
were not entirely unknown.19 Moreover, the propoxy
group at the 3′-position was bound closely to the side
chain of Ile112, Val45, and Leu46 with the distances of
3.75, 3.54, and 3.55 Å, respectively, close enough to be
involved in van der Waals interactions. This model
supported the experimental data, which showed very

5-Benzyl-2,4-diamonopyrimidines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2 351
F igu r e 4. Compound 40 in the binding pocket of pfDHFR with selected amino acid residues: (a) wild type; (b) quadruple mutant.
F igu r e 5. Compound 53 in the binding pocket of pfDHFR with selected amino acid residues: (a) wild type; (b) quadruple mutant.
molecular modeling of compound 53 bound to both wild
type and quadruple mutant enzymes was performed,
and results are shown in Figure 5. Because of the steric
interaction between the alkyl group at the 6-position
and the 5-benzyl group, the conformation of the 5-benzyl
moiety was adjusted in order to avoid this steric
constraint and also the aromatic substituent was repo-
sitioned. This phenomenon led to unfavorable stacking
interaction between the aromatic substituent and the
phenyl side chain of Phe116 of the wild type enzyme,
resulting in the reduction of binding affinity of this
inhibitor to the enzyme compared with those without
6-alkyl substituents (compound 30). For quadruple
mutant enzyme (Figure 5b), there is relatively more
favorable stacking interaction between aromatic sub-
stituent and the phenyl side chain of Phe116 after the
conformation adjustment of the 5-benzyl moiety. More-
over, the additional methyl group at position 6 on the
pyrimidine ring was in van der Waals contact with
Leu46 and Met55. All these interactions led to better
binding affinity of this inhibitor to the mutant enzymes.
Con clu sion
The extension of the hydrophobic side chain on the
5-benzyl moiety of 5-benzyl-2,4-diaminopyrimidine led
to better binding affinity to both wild type and mutant
enzymes than that of trimethoprim, especially with
benzyloxy substituents that were about 5- to 30-fold and
60- to 200-fold more effective against wild type and
mutant enzymes, respectively. Some compounds (52-
56) showed relatively less decrease in binding affinities
with the mutant enzymes and higher antimalarial

352 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2
Sirichaiwat et al.
neutralization with diluted aqueous HCl, the reaction mixture
was extracted with CH2Cl2, dried over MgSO4, and evaporated
to dryness. Purification by column chromatography (silica gel,
25% EtOAc/45% hexane as eluent) gave compound 15 as a
white solid. NMR spectra of compounds 15a -c indicated that
they were of high purity; therefore, compounds 15a -c were
used in the next step without further purification.
4-[3-(4-Tolu en esu lfon yl)p r op oxy]b en za ld eh yd e (15a ,
R ) H): white solid (10.4 g, 62%); 1H NMR (400 MHz, CDCl3)
δ 2.17 (2H, quint, J ) 5.4 Hz, CH2CH2CH2), 2.37 (3H, s,
ArCH3), 4.06 (2H, t, J ) 5.4 Hz, ArOCH2CH2), 4.26 (2H, t,
J ) 5.4 Hz, CH2CH2OSO2), 6.88 (2H, d, J ) 8.3 Hz, ArH), 7.25
(2H, d, J ) 7.8 Hz, ArH), 7.76 (2H, d, J ) 7.8 Hz, ArH), 7.81
(2H, d, J ) 8.3 Hz, ArH), 9.90 (1H, s, CHO).
3-Meth oxy-4-[3-(4-tolu en esu lfon yl)p r op oxy]ben za ld e-
h yd e (15b, R ) OMe): white solid (10.7 g, 59%); 1H NMR
(400 MHz, CDCl3) δ 2.23 (2H, tt, J ) 5.9, 6.0 Hz, CH2CH2-
CH2), 2.38 (3H, s, ArCH3), 3.88 (3H, s, OCH3), 4.18 (2H, t, J )
6.0 Hz, ArOCH2CH2), 4.30 (2H, t, J ) 5.9 Hz, CH2CH2OSO2),
6.89 (1H, d, J ) 8.1 Hz, ArH), 7.25 (2H, d, J ) 8.1 Hz, ArH),
7.40 (1H, d, J ) 1.3 Hz, ArH), 7.43 (1H, dd, J ) 8.1, 1.3 Hz,
ArH), 7.77 (2H, d, J ) 8.1 Hz, ArH), 9.87(1H, s, CHO).
3-E t h oxy-4-[3-(4-t olu en esu lfon yl)p r op oxy]b en za ld e-
h yd e (15c, R ) OEt): white solid (12.5 g, 66%); 1H NMR (400
MHz, CDCl3) δ 1.42 (3H, t,J ) 6.9 Hz, CH2CH3), 2.21 (2H,
quint, J ) 5.9 Hz, CH2CH2CH2), 2.37 (3H, s, CH3), 4.08 (4H,
m, ArOCH2CH2, ArOCH2CH3), 4.29 (2H, t, J ) 5.9 Hz,
CH2CH2OSO2), 6.87 (1H, d, J ) 8.1 Hz, ArH), 7.23 (2H, d, J )
8.1 Hz, ArH), 7.37 (1H, s, ArH), 7.39 (1H, d, J ) 8.1 Hz, ArH),
7.75 (2H, d, J ) 8.1 Hz, ArH), 9.85 (1H, s, CHO).
activities and are therefore of higher potential interest.
These compounds also exhibited IC50 values against
wild type and resistant parasites at low micromolar
levels, which correlated well with their binding affinity.
Molecular modeling studies of pfDHFR bound with
trimethoprim derivatives bearing aromatic substituents
showed the stacking of the aromatic substituent with
the phenyl side chain of Phe116. Evidently, this interac-
tion led to better binding affinity to both wild type and
mutant enzymes than those derivatives bearing alkoxy
substituents.
Exp er im en ta l Section
Meth od s a n d Ma ter ia ls. For the synthesis of TMP ana-
logues, solvents (DMSO, ethanol, and dioxane) were dried
according to standard methods. Reagents were purchased from
Fluka, Merck, and Sigma-Aldrich Ltd. and were distilled before
use. For enzyme studies, chemicals were obtained from Sigma-
Aldrich Ltd., Merck, and BDH and were used without further
purification. Melting points were determined by an Electro-
thermal 9100 melting point apparatus and were uncorrected.
Nuclear magnetic resonance (NMR) spectra were recorded in
DMSO-d6 and CDCl3 on a Bruker DRX 400 spectrometer;
chemical shifts are reported in parts per million (ppm) using
TMS (0 ppm) as the internal standard. Mass spectra were
recorded on a Micromass LCT using the electrospray ionization
technique. Elemental analyses were carried out using a
Perkin-Elmer elemental analyzer 2400.
Ch em ica l Syn th eses of TMP An a logu es. TMP analogues
were synthesized by an improved method described in a
previous study,15 as shown in Scheme 1, and the results are
summarized in Table 1. The general procedure for the syn-
theses of long-chain ether-substitued benzaldehyde is outlined
in Scheme 2.
P r ep a r a tion of Com p ou n d 14. To a suspension of K2CO3
(15.2 g, 0.11 mol) in DMF (10 mL) was added a solution of
3-substitued 4-hydroxybenzaldehyde (12, 0.1 mol) in DMF (10
mL). After the reaction mixture was heated at 60 °C for 30
min, a solution of 3-chloro-1-propanol (13, 9.2 g, 0.11 mol) in
DMF (5 mL) was added dropwise and the mixture was left
stirring at 60 °C overnight. The mixture was neutralized with
diluted aqueous HCl and then extracted with CH2Cl2, dried
over MgSO4, and evaporated to dryness under reduced pres-
sure to yield the crude product as a yellow oil. The crude
product was purified by column chromatography (silica gel,
55% EtOAc/45% hexane as developing solvent) to obtain the
desired product as a colorless oil.
4-(3-Hyd r oxyp r op oxy)ben za ld eh yd e (14a , R ) H): col-
orless oil (13.7 g, 76%); 1H NMR (400 MHz, CDCl3) δ 2.02 (2H,
quint, J ) 6.1 Hz, CH2CH2CH2), 3.06 (1H, s, OH), 3.81 (2H, t,
J ) 6.1 Hz, HOCH2CH2), 4.13 (2H, t, J ) 6.1 Hz, ArOCH2-
CH2), 6.94 (2H, d, J ) 8.6 Hz, ArH), 7.74 (2H, d, J ) 8.6 Hz,
ArH), 9.78 (1H, s, CHO).
3-Met h oxy-4-(3-h yd r oxyp r op oxy)b en za ld eh yd e (14b,
R ) OMe): colorless oil (17.0 g, 81%); 1H NMR (400 MHz,
CDCl3) δ 2.15 (2H, tt, J ) 6.0, 4.9 Hz, CH2CH2CH2), 2.40 (1H,
bs, OH), 3.91 (2H, t, J ) 4.9 Hz, HOCH2CH2), 3.94 (3H, s,
OCH3), 4.30 (2H, t, J ) 6.0 Hz, ArOCH2CH2), 7.01 (1H, d, J )
8.0 Hz, ArH), 7.43 (1H, s, ArH), 7.47 (1H, d, J ) 8.0 Hz, ArH),
9.87 (1H, s, CHO).
P r ep a r a tion of Com p ou n d 17. Compound 17 was pre-
pared from substituted phenol (16) and chloropropanol (13)
according to the procedure described for the synthesis of
compound 14 and was purified by column chromatography
(silica gel, 60% EtOAc/40% hexane as eluent). Compounds 17a
and 17b were of high purity (NMR spectra); therefore, they
were used in the next reaction step without additional
purification.
3-P h en oxy-1-p r op a n ol (17a , R′ ) H): colorless oil (10.5
1
g, 69%); H NMR (400 MHz, CDCl3) δ 2.03 (2H, tt, J ) 6.1,
6.2, Hz, CH2CH2CH2), 3.61 (1H, bs, OH), 3.82 (2H, t, J ) 6.2
Hz, CH2CH2OH), 4.07 (2H, t, J ) 6.1 Hz, ArOCH2CH2), 6.97
(3H, m, ArH), 7.32 (2H, m, ArH).
3-(2,4,5-Tr ich lor op h en oxy)-1-p r op a n ol (17b, R′ ) 2,4,5-
tr ich lor o): colorless oil (18.4 g, 72%); 1H NMR (400 MHz,
CDCl3) δ 2.09 (2H, tt, J ) 6.0, 5.8, Hz, CH2CH2CH2), 2.42 (1H,
bs, OH), 3.88 (2H, t, J ) 5.8 Hz, CH2CH2OH), 4.15 (2H, t, J )
6.0 Hz, ArOCH2CH2), 7.0 (1H, d, J ) 4.6 Hz, ArH), 7.42 (1H,
d, J ) 4.6 Hz, ArH).
P r ep a r a tion of Com p ou n d 18. To a suspension of NaH
(0.92 g, 21 mmol) in DMF (2 mL) at 0 °C was added a solution
of alcohol 17 (20 mmol) in DMF (3 mL), and the reaction
mixture was left stirring at 0 °C for 30 min. Then a solution
of aldehyde 15 (21 mmol) in DMF (5 mL) was added. After
continuous stirring at 0 °C for 2 h, the reaction mixture was
carefully neutralized with diluted aqueous HCl solution,
extracted with CH2Cl2, dried over MgSO4, and evaporated to
dryness. Purification by column chromatography (silica gel,
40% EtOAc/75% hexane as eluent) gave pure compound 18 as
a colorless oil.
4-[3-(3-P h en oxyp r op oxy)p r op oxy]ben za ld eh yd e (18a ,
1
R ) R′ ) H): colorless oil (2.4 g, 38%); H NMR (400 MHz,
3-E t h oxy-4-(3-h yd r oxyp r op oxy)b en za ld eh yd e (14c,
R ) OEt): colorless oil (16.1 g, 72%); 1H NMR (400 MHz,
CDCl3) δ 1.47 (3H, t, J ) 7.0 Hz, OCH2CH3), 2.14 (2H, quint,
J ) 5.6 Hz, CH2CH2CH2), 2.730 (1H, t, J ) 5.7 Hz, CH2OH),
3.91 (2H, dt, J ) 5.6, 5.6 Hz, HOCH2CH2), 4.21 (2H, q, J )
7.0 Hz, OCH2CH3), 4.29 (2H, t, J ) 5.7 Hz, ArOCH2CH2), 6.99
(1H, d, J ) 8.1 Hz, ArH), 7.40 (1H, d, J ) 1.6 Hz, ArH), 7.44
(1H, dd, J ) 8.1, 1.6 Hz, ArH), 9.85 (1H, s, CHO).
P r ep a r a tion of Com p ou n d 15. To a solution of compound
14 (0.05 mol) in CH2Cl2 (5 mL) and Et3N (9 mL) was added
p-toluenesulfonyl chloride (14.3 g, 0.075 mol) at 0 °C, and the
reaction mixture was continuously stirred for 2 h. After
CDCl3) δ 2.08 (4H, m, 2 × CH2CH2CH2), 3.65 (4H, t, J ) 6.0
Hz, 2 × OCH2CH2), 4.06 (2H, t, J ) 6.2 Hz, ArOCH2CH2), 4.15
(2H, t, J ) 6.3 Hz, ArOCH2CH2), 6.89 (2H, d, J ) 8.3 Hz, ArH),
6.97 (3H, m, ArH), 7.28 (3H, m, ArH), 7.82 (2H, d, J ) 8.4 Hz,
ArH), 9.89 (1H, s, CHO).
3-Met h oxy-4-[3-(3-p h en oxyp r op oxy)p r op oxy]b en za l-
d eh yd e (18b, R ) OMe, R′ ) H): colorless oil (2.5 g, 36%);
1H NMR (400 MHz, CDCl3) δ 2.01 (2H, quint, J ) 6.4 Hz,
CH2CH2CH2), 2.13 (2H, quint, J ) 6.2 Hz, CH2CH2CH2), 3.62
(4H, m, 2 × OCH2CH2), 3.86 (3H, s, OCH3), 4.02 (2H, t, J )
6.2 Hz, ArOCH2CH2), 4.18 (2H, t, J ) 6.4 Hz, ArOCH2CH2),

5-Benzyl-2,4-diamonopyrimidines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2 353
6.91 (3H, m, ArH), 7.25 (3H, m, ArH), 7.37 (2H, m, ArH), 9.81
(1H, s, CHO).
methamine, of methotrexate and trimethoprim.25,26 The inhibi-
tor was docked into the active site of pfDHFR in which the
protonated amino portion was arranged to initially occupy a
position suitable for interaction with the carboxylate side chain
of aspatate 54 (D54). Five-thousand steps of minimization on
the complex, which included the inhibitor, NADPH, and all
protein residues located within 7 Å from the inhibitor, were
performed on a Silicon Graphics using the Insight II program
and molecular mechanics (MM2) force field.
3-Eth oxy-4-[3-(3-p h en oxyp r op oxy)p r op oxy]ben za ld e-
1
h yd e (18c, R ) OEt, R′ ) H): colorless oil (3.8 g, 41%); H
NMR (400 MHz, CDCl3) δ 1.46 (3H, t, J ) 6.9 Hz, CH2CH3),
2.04 (2H, quint, J ) 6.1 Hz, CH2CH2CH2), 2.13 (2H, quint,
J ) 6.1 Hz, CH2CH2CH2), 3.64 (4H, m, 2 × OCH2CH2), 4.03
(2H, t, J ) 6.1 Hz, ArOCH2CH2), 4.12 (2H, q, J ) 6.9 Hz, OCH2-
CH3), 4.18 (2H, t, J ) 6.1 Hz, ArOCH2CH2), 6.90 (3H, m, ArH),
7.26 (3H, m, ArH), 7.39 (2H, m, ArH), 9.82 (1H, s, CHO).
4-{3-[3-(2,4,5-Tr ich lor oph en oxy)pr opoxy]pr opoxy}ben z-
a ld eh yd e (18d , R ) H, R′ ) 2,4,6-tr ich lor o): colorless oil
(3.9 g, 48%); 1H NMR (400 MHz, CDCl3) δ 2.08 (4H, m,
2xCH2CH2CH2), 3.63 (4H, m, 2 × OCH2CH2), 4.10 (4H, m,
2 × ArOCH2CH2), 6.94 (2H, d, J ) 8.7 Hz, ArH), 7.27 (1H, s,
ArH), 7.39 (1H, s, ArH), 7.80 (2H, d, J ) 8.7 Hz, ArH), 9.88
(1H, s, CHO).
Ack n ow led gm en t. We thank the Bioassay Re-
search Facility of the BIOTEC Center, NSTDA, for
performing cytotoxicity tests. This research was sup-
ported by grants to Y.Y. from MMV, TDR, the European
Union (INCO-DCand INCO-DEV), and the Wellcome
Trust and grants from Thailand-TDR and Biodiversity
Research and Training (BRT) Programs to S.K., from
BIOTEC/NSTDA to Y.T., and from Thailand Graduate
Institute of Science and Technology (TGIST) to C.S.
3-Met h oxy-4-{3-[3-(2,4,5-t r ich lor op h en oxy)p r op oxy]-
p r op oxy}ben za ld eh yd e (18e, R ) OMe, R′ ) 2,4,6-tr i-
ch lor o): colorless oil (4.6 g, 51%); 1H NMR (400 MHz, CDCl3)
δ 2.10 (4H, m, 2 × CH2CH2CH2), 3.65 (4H, m, 2 × OCH2CH2),
3.88 (3H, s, OCH3), 4.14 (4H, m, ArOCH2CH2), 6.89 (1H, m,
ArH), 6.94 (1H, s, ArH), 7.37 (2H, m, ArH), 7.42 (1H, s, ArH),
9.82 (1H, s, CHO).
Su p p or tin g In for m a tion Ava ila ble: Additional experi-
mental data (1H NMR) of all compounds not listed in the
Experimental Section and the results of toxicity tests of TMP
analogues to mammalian cells. This material is available free
of charge via the Internet at http://pubs.acs.org.
3-Eth oxy-4-{3-[3-(2,4,5-tr ich lor op h en oxy)p r op oxy]p r o-
p oxy}ben za ld eh yd e (18f, R ) OEt, R′ ) 2,4,6-tr ich lor o):
1
colorless oil (4.2 g, 45%); H NMR (400 MHz, CDCl3) δ 1.46
(3H, t, J ) 7.0 Hz, CH2CH3), 2.11 (4H, m, 2 × CH2CH2CH2),
3.66 (4H, m, 2 × OCH2CH2), 4.06 (2H, t, J ) 6.1 Hz, ArOCH2-
CH2), 4.13 (2H, q, J ) 7.0 Hz, OCH2CH3), 4.18 (2H, t, J ) 6.1
Hz, ArOCH2CH2), 6.91 (1H, m, ArH), 6.94 (1H, m, ArH), 7.40
(2H, m, ArH), 7.46 (1H, s, ArH), 9.83 (1H, s, CHO).
Refer en ces
(1) White, N. J . Drug resistance in Malaria. Br. Med. Bull. 1998,
54, 703-715.
(2) Morel, C. M. Reaching maturity-25 years of the TDR. Parasitol.
Today 2000, 16, 503-551.
En zym e P r ep a r a tion . The wild type and mutant pfDHFR
enzymeswere expressed in E. coli BL21(DE3)pLysS according
to the procedure previously described.20 After disruption of the
bacterial cells by a French pressure cell at 18 000 psi, the crude
extract in 20 mM potassium phosphate buffer at pH 7.0, 0.1
mM EDTA, 10 mM DTT, 50 mM KCl, and 20% v/v glycerol
was applied to Methotrexate-Sepharose column, and the
pfDHFR was purified according to the procedure described.15,21
E n zym e Assa ys a n d In h ib it ion b y An t ifola t es a n d
Der iva tives. The methods used for determination of pfDHFR
activities and for the study of inhibition by antifolates and
derivatives were described previously.15,20 Calculation of Ki
values was based on the assumption that the inhibitors
compete for substrate binding to the active site of the enzyme.
P a r a site Cu ltu r e a n d An tim a la r ia l Testin g in Vitr o.
Four P. falciparum strains were used in this study. Strains
TM4/8.2 (wild type DHFR) and K1CB1 (C59R + S108N) were
generous gifts from S. Thaithong, Department of Biology,
Faculty of Science, Chulalongkorn University; Csl-2 (C59R +
S108N + I164L) was from A. Cowman WEHI, Australia, and
V1/S (N51I + C59R + S108N + I164L) was from D. Kyle
through MR4. These parasites were maintained continuously
in human erythrocytes at 37 °C under 3% CO2 in RPMI 1640
culture media supplemented with 25 mM HEPES, pH 7.4,
0.2% NaHCO3, 40 µg/mL gentamicin, and 10% human serum.22
(3) Cowman, A. F.; Morry, M. J .; Biggs, B. A.; Cross, G. A.; Foote,
S. J . Amino acid changes linked to pyrimethamine resistance
in the dihydrofolate reductase-thymidylate synthase gene of
Plasmodium falciparum. Proc. Natl. Acad. Sci. U.S.A. 1998, 85,
9109-9113.
(4) Peterson, D. S.; Walliker, D.; Wellems, T. E. Evidence that a
point mutation in dihydrofolate reductase-thymidylate synthase
confers resistance to pyrimethamine in Plasmodium falciparum
malaria. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 9114-9118.
(5) Hyde, J . E. Mechanisms of resistance of Plasmodium falciparum
to antimalarial drugs. Microb. Inf. 2002, 4, 165-174.
(6) Yuthavong, Y. Basis of antifolate action and resistance in
malaria. Microb. Inf. 2002, 4, 175-182.
(7) Quaye, I.; Sibley, C. H. Molecular data on Plasmodium falci-
parum chloroquine and antifolate resistance: a public health tool.
Trends Parasitol. 2002, 18, 184-186.
(8) Lemcke, T.; Christensen, I. T.; J orgensen, F. S. Towards an
understanding of drug resistance in malaria: Three dimensional
structure of Plasmodium falciparum by homology building.
Bioorg. Med. Chem. 1999, 7, 1003-1010.
(9) McKie, J . H.; Douglas, K. T.; Chan, C.; Roser, S. A.; Yates, R.;
Read, M.; Hyde, J . E.; Dascombe, M. J .; Yuthavong, Y.; Sira-
waraporn, W. Rational drug design approach for overcoming
drug resistance: Application to pyrimethamine resistance in
malaria. J . Med. Chem. 1998, 41, 1367-1370.
(10) Rastelli, G.; Sirawaraporn, W.; Sompornpisut, P.; Vilaivan, T.;
Kamchonwongpaisan, S.; Quarrell, R.; Lowe, G.; Thebtaranonth,
Y.; Yuthavong, Y. Interaction of pyrimethamine, cycloguanil,
WR99210 and their analogues with Plasmodium falciparum
dihydrofolate reductase: Structural basis of antifolate resistant.
Bioorg. Med. Chem. 2000, 8, 1117-1128.
(11) Santos-Filho, O. A.; Alencastro, R. B.; Figueroa-Villar, J . D.
Homology modeling of wild type and pyrimethamine/cycloguanil
cross-resistant mutant type Plasmodium falciparum dihydro-
folate reductase. A model for antimalarial chemotherapy resis-
tance. Biophys. Chem. 2001, 91, 305-317.
(12) Warhurst, D. C. Antimalarial drug discovery: Development of
inhibitors of dihydrofolate reductase active in drug resistance.
Drug Discovery Today 1998, 3, 538-546.
(13) Canfield, C. J .; Milhous, W. K.; Ager, A. L.; Rossen, S. N.;
Sweeney, T. R.; Lewis, N. J .; J acobus, D. P. PS-15: a potent,
orally active antimalarial from a new class of folic acid antago-
nists. Am. J . Trop. Med. Hyg. 1993, 49, 121-126.
(14) Hekmat-Nejad, M.; Rathod, P. K. Plasmodium falciparum:
Kinetic interactions of WR99210 with pyrimethamine-sensitive
and pyrimethamine resistant dihydrofolate reductase. Exp.
Parasitol. 1997, 87, 222-228.
(15) Tarnchompoo, B.; Sirichaiwat, C.; Phupong, W.; Intaraudom, C.;
Sirawaraporn, W.; Kamchonwongpaisan, S.; Vanichtanankul, J .;
Thebtaranonth, Y.; Yuthavong, Y. Development of 2,4-diamino-
In vitro antimalarial activity was determined by using [3H]-
23
hypoxanthine incorporation method
as described in detail
elsewhere.15,20
Cytotoxicity tests of selected analogues against African
green monkey kidney fibroblast (Vero cells), human epider-
moid carcinoma (KB) cells, and human breast cancer (BC) cells
were performed at the Bioassay Research Facility of the
BIOTEC Center, NSTDA, according to the protocol described
by Skehan et al. (1990).24
Molecu la r Mod elin g. Since the structures of pfDHFRs
have already been solved, these enzymes were used for the
molecular modeling study. The force field AM1 in the semiem-
pirical method of the program Hyperchem 7 was employed,
and the structure of small molecules (inhibitors) were geo-
metrically optimized to a gradient of 0.001. The inhibitor was
modeled as a protonated state at N1, as evidence by the NMR
studies of both pfDHFR-bound and free states of pyri-

354 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 2
Sirichaiwat et al.
pyrimidines as antimalarials based on inhibition of the S108N
and C59R+S108N mutants of dihydrofolate reductase from
pyrimethamine-resistant Plasmodium falciparum. J . Med. Chem.
2002, 45, 1244-1252.
(21) Sirawaraporn, W.; Prapunwattana, P.; Sirawaraporn, R.;
Yuthavong, Y.; Santi, D. V. The dihydrofolate reductase domain
of Plasmodium falciparum thymidylate synthase-dihydrofolate
reductase. J . Biol. Chem. 1993, 29, 21637-21644.
(16) Yuvaniyama, J .; Chitnumsub, P.; Kamchonwongpaisan, S.;
Vanichtanankul, J .; Sirawaraporn, W.; Taylor, P.; Walkinshaw,
M.; Yuthavong, Y. Insights into antifolate resistance from
malarial DHFR-TS structures. Nat. Struct. Biol. 2003, 10, 357-
365.
(17) Matthews, D. A.; Bolin, J . T.; Burridge, J . M.; Filman, D. T.;
Volz, K. M.; Kaufman, B. T.; Beddell, C. R.; Champness, J . N.;
Stammers, D. K.; Kraut, J . Refined crystal structures of Es-
cherichia coli and chicken liver dihydrofolate reductase contain-
ing bound trimethoprim. J . Biol. Chem. 1985, 260, 381-391.
(18) Matthews, D. A.; Bolin, J . T.; Burridge, J . M.; Filman, D. J .;
Volz, K. M.; Kraut, J . Dihydrofolate reductase: The stereochem-
istry of inhibitor selectivity. J . Biol. Chem. 1985, 260, 392-399.
(19) Domak, M.; Riche, C. Crystal and Molecular Structure of
Polycarpine. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst.
Chem. 1977, 33, 3419-3422.
(22) Trager, N.; J ensen, J . B. Human malarial parasites in continuous
culture. Science 1976, 193, 673-675.
(23) Desjardins, R. E.; Canfield, C. J .; Haynes, J . D.; Chulay, J . D.
Quantitative assessment of antimalarial activity in vitro by a
semiautomated microdilution technique. Antimicrob. Agents
Chemother. 1979, 16, 710-718.
(24) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; Mcmahon, J .;
Vistica, D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M. R.
New colorimetric cytotoxicity assay for anticancer-drug screen-
ing. J . Natl. Cancer. Inst. 1990, 82, 1107-1112.
(25) Cocco, L.; Groff, J . P.; Temple, C., J r.; Montgomery, J . A.; London,
R. E.; Matwiyoff, N. A.; Blakley, R. L. Carbon-13 nuclear
magnetic resonance study of protonation of methotrexate and
aminopterin bound to dihydrofolate reductase. Biochemistry
1981, 20, 3972-3978.
(26) Cocco, L.; Roth, B.; Temple, C., J r.; Montgomery, J . A.; London,
R. E.; Blakley, R. L. Protonated state of methotrexate, tri-
methoprim and pyrimethamine bound to dihydrofolate reduc-
tase. Arch. Biochem. Biophys. 1983, 226, 567-577.
(20) Yuthavong, Y.; Vilaivan, T.; Chareonsethakul, N.; Kamchon-
wongpaisan, S.; Sirawaraporn, W.; Quarrell, R.; Lowe, G.
Development of a lead inhibitor for the A16V+S108T mutant of
dihydrofolate reductase from the cycloguanil-resistant strain (T9/
94) of Plasmodium falciparum. J . Med. Chem. 2000, 43, 2738-
2744.
J M0303352