Synthesis of solution-phase combinatorial library of 4,6-diamino-1,2-dihydro-1,3,5-triazine and identification of new leads against A16V+S108T mutant dihydrofolate reductase of Plasmodium falciparum.

Article abstract of DOI:10.1016/S0968-0896(02)00344-9

An efficient method to synthesize solution-phase combinatorial library of 1-aryl-4,6-diamino-1,2-dihydro-1,3,5-triazine was developed. The strategy involved an acid-catalyzed cyclocondensation between arylbiguanide hydrochlorides and carbonyl compounds in the presence of triethyl orthoacetate as water scavenger. A 96-membered combinatorial library was constructed from 6 aryl biguanides and 16 carbonyl compounds. Screening of the library by iterative deconvolution method revealed two candidate leads which are equally active against wild-type Plasmodium falciparum dihydrofolate reductase, but are about 100-fold more effective against the A16V+S108T mutant enzyme as compared to cycloguanil.

Full text of DOI:10.1016/S0968-0896(02)00344-9

Bioorganic & Medicinal Chemistry 11 (2003) 217–224
Synthesis of Solution-Phase Combinatorial Library of
4,6-Diamino-1,2-dihydro-1,3,5-triazine and Identification of
New Leads Against A16V+S108T Mutant Dihydrofolate
Reductase of Plasmodium falciparum
Tirayut Vilaivan,a,* Neungruthai Saesaengseerung,a Deanpen Jarprung,b
Sumalee Kamchonwongpaisan,b Worachart Sirawarapornc and Yongyuth Yuthavongb
aOrganic Synthesis Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Phayathai Road,
Patumwan, Bangkok 10330, Thailand
bNational Center for Genetic Engineering and Biotechnology, National Science and Technology Development Agency,
Rama VI Road, Bangkok 10400, Thailand
cDepartment of Biochemistry, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailand
Received 21 January 2002; accepted 29 July 2002
Abstract—An efficient method to synthesize solution-phase combinatorial library of 1-aryl-4,6-diamino-1,2-dihydro-1,3,5-triazine
was developed. The strategy involved an acid-catalyzed cyclocondensation between arylbiguanide hydrochlorides and carbonyl
compounds in the presence of triethyl orthoacetate as water scavenger. A 96-membered combinatorial library was constructed from
6 aryl biguanides and 16 carbonyl compounds. Screening of the library by iterative deconvolution method revealed two candidate
leads which are equally active against wild-type Plasmodium falciparum dihydrofolate reductase, but are about 100-fold more
effective against the A16V+S108T mutant enzyme as compared to cycloguanil.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
108 and 164 are associated with Pyr- and Cyc-cross
The dihydrofolate reductase (DHFR) domain of the
bifunctional dihydrofolate reductase-thymidylate syn-
thase (DHFR-TS) enzyme of Plasmodium falciparum is
the target of pyrimethamine (Pyr) and cycloguanil (Cyc)
(Scheme 1), two historically important therapeutic
drugs commonly used for the treatment of malaria
infection.1,2 However, in recent years, resistance to
antifolate antimalarials in P. falciparum has become
widespread, and hence reducing the effectiveness and
the clinical utilities of the drugs. With the limited spec-
trum of the currently available antimalarials, there is an
immediate need to search for new drugs and/or identify
new targets in order to combat the resistant malaria.
Resistance to Pyr and Cyc in P. falciparum has been
shown to be associated with point mutations of the
DHFRdomain.While mutations at amino acids 51, 59,
resistance, mutation at residue 16 has been shown to be
responsible for Cyc resistance.3ꢀ8 The mutations were
found to affect both the binding affinity to inhibitors
and kinetic properties of the enzymes.9 The lack of three
dimensional structure of P. falciparum DHFRhas pre-
cluded rational inhibitor design and development of
more effective inhibitors, therefore molecular modeling
approach has been exploited as an alternative strat-
egy.10,11 Our previous works have led to the findings that
steric clash imposed by A16V mutation effects binding of
Cyc, and not Pyr, to the active site of the A16V+S108T
mutant enzyme.11 The hypothesis was verified by syn-
thesizing and measuring the inhibition constants (Ki) of
various Cyc derivatives (4,6-diamino-1,2-dihydro-1,3,5-
triazine compounds) devoid of one or both of C-2 sub-
stituents, from which promising leads were identified.12
Recent advancement in the field of combinatorial
chemistry has attracted much interest in drug discovery
research due to the possibility of rapid generation of
*Corresponding author. Tel.: +66-2-2187627; fax: +66-2-2187598;
e-mail: vtirayut@chula.ac.th
0968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00344-9

218
T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
Our objective was to develop a technique by which a
library of compounds comprising of pools of statistical
distribution of the desired member, irrespective of the
reactivities of starting materials and solubilities of the
products, could be produced. The original Modest’s
three component14 as well as two-component16 dihy-
drotriazine syntheses (Scheme 2) gave satisfactory results
with only limited range of substrates therefore a more
efficient method is required in order to achieve our goal.
We have recently noted that addition of a water sca-
venger such as triethyl orthoacetate have dramatic effect
to the formation of dihydrotriazine by 2-component
synthesis reaction.21 Reactions between a number of
phenylbiguanide hydrochloride and excess of carbonyl
compounds (5 equiv) proceeded smoothly at room tem-
perature (30 ꢁC) in the presence of triethyl orthoacetate
and catalytic amounts of concentrated HCl. Evaporation
of the solvents and trituration of the residue with diethyl
ether afforded the dihydrotriazines (I) in good yield and
purity (>90%). The insolubility of the dihydrotriazines
in ether provides the advantage of solid phase synthesis
in such a way that excess reagents and ether-soluble by-
products may be removed by exhaustive washing with-
out the need for expensive solid supports and attach-
ment-cleavage steps. To investigate the range of suitable
substrates to be included in the library, the rates of
reactions of a representative set of carbonyl compounds
(4-heptanone, 3-octanone, benzaldehyde, acetone, 4-
methyl-3-pentanone, 3-methyl-2-butanone) and aryl
biguanides (phenyl, 3,4-dichlorophenyl, 4-methylphenyl,
2,4-dichlorophenyl, 4-chlorophenylbiguanides) were
examined by HPLC analyses. It was revealed that most
substrates reacted completely within 12–72 h under the
conditions described above. Only the reactions involving
2,4-dichlorophenylbiguanide was slow and incomplete
even on prolonged stirring. Incidentally, according to the
structure–activity relationship data available,17 com-
pounds (I) bearing ortho-substituent on the N-1 aryl
group are unlikely to exhibit significant inhibitory
activity against DHFRs, therefore 2,4-dichlor-
ophenylbiguanide and other ortho-substituted bigua-
nides were not included in the library.
Scheme 1.
many structurally related compounds from which, in
conjunction with a suitable screening technique, poten-
tial drug candidates can be rapidly identified.13 In this
paper, we wish to demonstrate the use of combinatorial
technique to discover lead compounds with significantly
more potent than the parent cycloguanil against cyclo-
guanil-resistant DHFRfrom P. falciparum by synthe-
sizing a small and well-defined combinatorial library of
dihydrotriazines in solution phase. We also present the
combinatorial library screening results and the identifi-
cation of two potential drug candidates which are effec-
tive against both wild-type and A16V+S108T mutant
DHFRof P. falciparum.
Results and Discussion
Chemistry
Synthesis of 1-aryl-4,6-diamino-1,2-dihydro-1,3,5-tria-
zine (I) by the acid catalyzed three-component cyclo-
condensation between an aromatic amine, a carbonyl
compound and dicyandiamide is well-documented
(Scheme 2).14 A synthesis of 72-membered combinator-
ial mixture of such library employing this method has
been recently reported.15 It was reported that only 64
out of 72 compounds were detected in the library. Fur-
thermore, it is doubtful whether all the compounds
successfully formed are present in equimolar quantities
due to the difference in reactivities and/or the different
crystallization behavior of each compound.
Having established the optimal synthetic conditions for
individual compounds, the libraries were then synthe-
sized as mixture libraries by the split-and-mix metho-
dology18 for the purpose of screening by iterative
deconvolution approach.19,20 Chemical structures of the
constituents of the library involved are given in Scheme
Scheme 2.

T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
219
Scheme 3.
3, and the strategy for creating a 96-membered combi-
natorial library from these compounds is illustrated in
Scheme 4. According to this strategy, sixteen portions of
an equimolar mixture of six aryl biguanides in the form
of hydrochloride salts are separately treated with excess
of 16 carbonyl compounds to give 16 sub-libraries,
I-1(a–f)– I-16(a–f), each of which had the C-2 sub-
stituents (derived from the carbonyl compounds 1–16)
fixed, and contained 6 compounds bearing 6 different
N-1-phenyl substituents (derived from the biguanides a–f).
As a result, a total number of 96 compounds, excluding
those enantiomeric pairs whose C-2 of the dihydrotriazine
is asymmetrically substituted, are synthesized.
an off-white solid. Reverse phase HPLC analysis of the
sublibrary I-11(a–f) revealed the presence of six major
peaks with lmax at 249 nm, characteristic for dihydro-
triazines. The areas under each peak from the HPLC
analysis profile (Fig. 1) were comparable and were in
good agreement with the expected equimolar mixture of
the six compounds (assuming the molar extinction
coefficient of the triazine chromophore is independent
of substituents). In addition, mass spectrometric analy-
sis (ESI) revealed (M-Cl+) peaks at m/z=273, 287, 302,
307, 341 and 351, corresponding to the six desired pro-
ducts as anticipated. The remaining 15 sublibraries were
synthesized and analyzed in the same way. All sub-
libraries I-1(a–f)–I-16(a–f) yielded six major peaks in
approximately equimolar proportion upon HPLC ana-
lysis, and gave mass spectrometric data which corre-
sponded to the desired products.
A representative sublibrary I-11(a–f) was synthesized by
treating the arylbiguanide mixture with 5 equivalents of
4-heptanone in the presence of triethyl orthoacetate and
concentrated HCl in methanol at room temperature.
After the reaction was completed, as judged by HPLC
analysis, the solvent was removed and the product was
triturated with diethyl ether to obtain the sublibrary as
We have also explored the possibilities of creating com-
binatorial libraries in a complementary manner by
reacting a biguanide with a mixture of carbonyl com-

220
T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
P. falciparum for the inhibitory activities (Table 1).
Three sublibraries were found to exhibit reasonably
good inhibition against both enzymes. These included
sublibraries of acetaldehyde (I-1(a–f): Ki wt 2.5ꢂ0.2
nM; Ki mut 35.5ꢂ7.3 nM), benzaldehyde (I-16(a–f): Ki
wt 6.4ꢂ1.0 nM; Ki mut 146.3ꢂ44.5 nM) and 2-octa-
none (I-12(a–f): Ki wt 1.6ꢂ0.2 nM; Ki mut 315.9ꢂ60.4
nm). While the sublibraries I-1(a–f), I-12(a–f) and
I-16(a–f) were about as effective as Cyc against the wild-
type P. falciparum DHFR, the libraries were sig-
nificantly more active than Cyc against the
A16V+S108T mutant enzyme. To assess the effective-
ness of the compounds in the sublibraries, the ratios of
the Ki values of the inhibitors for the A16V+S108T
mutant enzyme and the wild-type enzyme (Ki-mut/Ki-
wt) were calculated. As shown in Table 1, the I-1(a–f),
I-12(a–f) and I-16(a–f) sublibraries were about 38-, 4-
and 84-fold more effective than Cyc, respectively,
against the A16V+S108T mutant enzyme.
Scheme 4.
pounds. Since the easily removable carbonyl com-
pounds must be used in excess in order to drive the
reaction to completion, complication arose due to the
difference of reactivities of each carbonyl compound.
Thus treatment of 4-chlorophenylbiguanide with excess
equimolar mixture of 5-methylhexan-2-one, 2-penta-
none, 2-octanone, 3-octanone, 2-hexanone and 2-hep-
tanone gave variable amounts of products. Such
difficulties in preparing isokinetic mixture of the car-
bonyl compounds made this approach less attractive
and was not further investigated.
To further investigate and identify the active compounds
from the pool of each sublibrary, all 18 members from the
three sublibraries were individually synthesized, and the
Ki values against both wild-type and A16V+S108T
mutant enzymes were determined. As shown in Table 2,
two most potent lead inhibitors I-1f and I-16f were
identified from the sublibraries I-1(a–f) and I-16(a–f)
respectively (Scheme 5). While the activities of the two
compounds against the wild-type DHFRare compar-
able to that of Cyc (I-1f, Ki wt=1.4ꢂ0.2 nM; I-16f Ki
wt=1.6ꢂ0.2 nM; Cyc Ki wt=1.5ꢂ0.3 nM), they are
much more effective against the A16V+S108T mutant
(I-1f, Ki mut=17.8ꢂ0.8 nM; I-16f Ki mut=11.0ꢂ1.8
nM; Cyc Ki mut=1,313.7ꢂ164.5 nM).
Enzyme assays
Primary screening of the 16 sublibraries was carried out
with wild-type and A16V+S108T mutant pfDHFRof
Figure 1. HPLC data of I-11(a–f). Column: HypersilTM 5 mm BDS C18 (4.6ꢃ250 mm); Mobile phase: 0.1 M triethylammonium acetate (pH
7.0):acetonitrile gradient from 15–60% over a period of 15 min with an equilibration period of 5 min; UV-detector at 254 nm.

T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
221
Table 1. Ki values of 4,6-diamino-1,2-dihydro-1,3,5-triazine sublibraries for the wild-type and A16V+S108T mutant DHFRs of P. falciparum
Sublibrary ID
Ar
R1
R2
Ki (wt)a (nM)
Ki (mut)b (nM)
Ki (mut)/Ki (wt)
I-1(a–f)
I-2(a–f)
I-3(a–f)
I-4(a–f)
I-5(a–f)
I-6(a–f)
I-7(a–f)
I-8(a–f)
I-9(a–f)
I-10(a–f)
I-11(a–f)
I-12(a–f)
I-13(a–f)
I-14(a–f)
I-15(a–f)
I-16(a–f)
Cyc
-H
-Me
-Me
-Me
-Me
-Me
-Et
-Me
-Me
-Me
-nC3H7
-Me
-Et
-Me
-Me
-Et
-iC3H7
-nC3H7
2.5ꢂ0.2
2.4ꢂ0.0
4.0ꢂ0.5
36.4ꢂ1.4
6.1ꢂ0.6
5.0ꢂ0.3
11.4ꢂ1.7
5.5ꢂ0.3
3.9ꢂ0.3
4.9ꢂ0.4
470ꢂ32
1.6ꢂ0.2
43.0ꢂ2.6
19.4ꢂ3.2
414ꢂ58
6.4ꢂ1.0
1.5ꢂ0.3
35.5ꢂ7.3
1428ꢂ286
1939ꢂ225
27,658ꢂ2096
3499ꢂ601
4301ꢂ246
5664ꢂ352
2016ꢂ120
1104ꢂ0.3
1528ꢂ181
39,546ꢂ10,153
316ꢂ60.4
12,233ꢂ438
668ꢂ99
14
595
485
760
574
860
497
367
283
312
83
198
284
34
14
23
mixture of
4-H-C6H4-
4-ClC6H4-
4-BrC6H4-
4-MeC6H4-
4-EtC6H4-
3,4-Cl2
-CH2CH(Me)2
-Et
-nC4H9
-(CH2)2CHMe2
-nC5H11
-nC3H7
-nC6H13
-nC5H11
C6H4-
-(CH2)4-
-(CH2)5-
5594ꢂ867
146ꢂ45
-H
-Me
-Ph
-Me
4-ClC6H4-
1314ꢂ165
876
aWild-type pfDHFR.
bA16V+S108T mutant pfDHFR.
Table 2. Ki values of each member of the sublibraries I-1(a–f), I-12(a–f) and I-16(a–f) for the wild-type and A16V+S108T mutant DHFRs of P.
falciparum
Compound ID
R1
R2
Ar
Ki (wt)a (nM)
Ki (mut)b (nM)
Ki (mut)/Ki (wt)
I-1a
I-1b
I-1c
I-1d
I-1e
I-1f
I-12a
I-12b
I-12c
I-12d
I-12e
I-12f
I-16a
I-16b
I-16c
I-16d
I-16e
I-16f
-H
-H
-H
-H
-H
-H
-Me
-Me
-Me
-Me
-Me
-Me
-H
-H
-H
-H
-H
-Me
-Me
-Me
-Me
-Me
-Me
-nC6H13
-nC6H13
-nC6H13
-nC6H13
-nC6H13
-nC6H13
-Ph
-Ph
-Ph
-Ph
-Ph
C6H5-
4-ClC6H4-
4-BrC6H4-
4-MeC6H4-
4-EtC6H4-
3,4-Cl2C6H3-
C6H5-
4-ClC6H4-
4-BrC6H4-
4-MeC6H4-
4-EtC6H4-
3,4-Cl2C6H3-
C6H5-
113ꢂ15
4.1ꢂ0.0c
5.7ꢂ0.5c
23.4ꢂ1.9c
11.0ꢂ3.0
1.4ꢂ0.2c
1.5ꢂ0.8
0.7ꢂ0.5
3.0ꢂ1.8
0.8ꢂ0.6
1.1ꢂ0.4
1.4ꢂ0.2
49.1ꢂ3.5
4.5ꢂ0.2c
2.9ꢂ1.2c
7.7ꢂ2.2c
12.1ꢂ2.3
1.6ꢂ0.2c
456ꢂ0.3
127ꢂ13.6c
202ꢂ17c
4.0
31c
35c
7.9c
29
13c
120
806
729
1029
502
76
186ꢂ22.4c
322ꢂ56.4
17.8ꢂ0.8c
180ꢂ12.6
564ꢂ151
2186ꢂ338
823ꢂ155
552ꢂ39.4
107ꢂ32.2
61.8ꢂ15.6
49.3ꢂ3.3c
90.3ꢂ11.4c
170ꢂ13.8c
86.6ꢂ10.8
11.0ꢂ1.8c
1.3
4-ClC6H4-
4-BrC6H4-
4-MeC6H4-
4-EtC6H4-
3,4-Cl2C6H3-
11c
31c
22c
7.2
-H
-Ph
6.9c
aWild-type pfDHFR.
bA16V+S108T mutant pfDHFR.
cData from Yuthavong et al. (ref 12).
Scheme 5.

222
T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
It is interesting to note that both compounds share two
common features: (i) they possess only one substituent
at C-2 position and (ii) they possess an extra chloro
substituent at the 30 position of the N-1 aromatic ring.
The result confirmed that the presence of one hydrogen
atom, or rather the absence of one substituent at the C-2
position of dihydrotriazine ring is an essential element
of good binding with the A16V+S108T mutant
enzyme as evident by the Ki-mut/Ki-wt ratio. This is in
agreement with our previous model11 which suggested
that the steric effect resulted from interaction between
one of the methyl group in cycloguanil at C-2 position
with Val-16 in A16V+S108T pfDHFRis responsible
for its poor activity. It also further emphasizes the con-
tribution of the additional substituent at the 30 position
on the 1-aryl groups to good binding to the
A16V+S108T pfDHFR, the fact that has been pre-
viously observed.12 Although the two compounds had
been previously identified as highly potent A16V+
S108T pfDHFRinhibitors by conventional screening
method,12 the present study has demonstrated that the
principle of combinatorial chemistry can be applied to
the same problem to give the same outcome in a short
time.
equimolar mixture of the arylbiguanide hydrochloride
(1 mmol) and the ketone or aldehyde (5 mmol) in abso-
lute MeOH (5 mL) was added triethyl orthoacetate
(0.75 mL) and concentrated HCl (0.025 mL). The reac-
tion mixture was stirred at room temperature until the
reaction is judged complete by HPLC analysis or by
biguanide test (24–120 h). The residue from evaporation
of the solvent was triturated and washed exhaustively
with ether. The solid was collected by filtration and air
dried.
Synthesis of dihydrotriazine libraries. Sixteen portions of
a mixture of 6 biguanides were reacted with 16 different
carbonyl compounds (Scheme 3) as described above (1
mmol scale reaction each) to give 16 sublibraries of
dihydrotriazines: m/z (MALDI-TOF) I-1(a–f) 204.0,
218.0, 237.9, 281.9, 232.0, 271.9; I-2(a–f) 218.0, 232.0,
251.9, 296.0, 246.0, 287.0; I-3(a–f) 232.2, 246.2, 266.2,
310.2, 260.3, 301.1; I-4(a–f) 246.0, 260.0, 279.2, 324.0,
274.0, 314.0; I-5(a–f) 246.0, 260.4, 280.0, 324.0, 274.4,
314.0; I-6(a–f) 260.1, 274.1, 294.0, 338.1, 288.1, 328.0;
I-7(a–f) 245.0, 259.0, 280.0, 324.0, 274.0, 314.0; I-8(a–f)
259.0, 273.0, 293.0, 338.0, 287.0, 327.0; I-9(a–f) 274.4,
288.4, 308.4, 352.0, 302.4, 342.0; I-10(a–f) 273.9, 287.9,
308.1, 352.1, 301.9, 342.1; I-11(a–f) 273.0, 287.0, 307.0,
351.0, 302.0, 341.0; I-12(a–f) 288.3, 302.3, 322.3, 366.2,
316.3, 356.3; I-13(a–f) 287.9, 301.8, 320.9, 365.9, 315.9,
355.9; I-14(a–f) 243.9, 257.9, 278.0, 322.0, 271.9, 312.0;
I-15(a–f) 257.0, 271.0, 291.0, 335.0, 286.0, 325.0; I-16(a–
f) 266.3, 280.4, 300.3, 344.3, 294.4, 334.3.
Conclusion
We have developed an efficient method to synthesize a
96-membered solution-phase combinatorial mixture
library of 4,6-diamino-1,2-dihydro-1,3,5-triazines. Two
highly potent lead inhibitors, one is over 100-folds more
active against A16V+S108T mutant DHFRthan
cycloguanil, were identified from the library by iterative
deconvolution technique.
HPLC analysis of dihydrotriazine libraries. Samples for
reverse phase HPLC analyses were dissolved in a sui-
table aqueous solvent and filtered through a Teflon filter
(0.47 m pore size). The HPLC analyses were performed
on a Waters 600 system equipped with a Waters 996
photodiode array detector. HPLC conditions—Column:
HypersilTM 5 mm BDS C18 HPLC column (4.6ꢃ250
mm); Mobile phase: 0.1 M triethylammonium acetate
(pH 7.0):acetonitrile gradient from 15–60% over a per-
iod of 15 min with an equilibration period of 5 min.
Retention time (tR): I-1(a–f) 3.7, 4.8, 5.3, 5.8, 7.0, 8.0; I-
2(a–f) 4.4, 6.6, 7.4, 8.4, 10.2, 10.8; I-3(a–f) 3.7, 4.3, 4.5,
4.8, 5.4, 5.8; I-4(a–f) 4.3, 5.5, 6.0, 6.6, 8.0, 8.7; I-5(a–f)
4.4, 5.7, 6.3, 6.7, 7.7, 8.2; I-6(a–f) 4.7, 6.4, 7.1, 7.7, 9.3,
9.7; I-7(a–f) 3.7, 4.2, 4.5, 4.8, 5.5, 5.9; I-8(a–f) 5.1, 6.5,
7.1, 7.7, 9.2, 9.8; I-9(a–f) 6.0, 8.9, 10.0, 10.8, 12.5, 12.9;
I-10(a–f) 6.4, 8.9, 9.6, 10.3, 11.9, 12.4; I-11(a–f) 5.5, 7.5,
8.4, 9.8, 11.2, 12.3; I-12(a–f) 11.1, 12.5, 12.8, 13.8, 14.7,
15.1; I-13(a–f) 10.7, 12.8, 13.4, 13.8, 14.8, 15.2; I-14(a–f)
4.5, 6.4, 7.1, 7.7, 9.1, 9.7; I-15(a–f) 4.7, 5.9, 6.3, 6.8, 7.6,
8.0; I-16(a–f) 4.0, 4.8, 5.2, 5.5, 6.0, 6.7 min.
Experimental
Materials and methods
Elemental analyses were performed by Miss A. Ungpa-
kornkaew on a Perkin–Elmer CHN analyzer model
PE2400 series II (Chulalongkorn Research Equipment
Centre, Bangkok). 1H NMRspectra were recorded on a
Bruker ACF 200 (Chulalongkorn University, Bangkok)
operating at 200 MHz. Chemical shifts are quoted in
ppm downfield relative to tetramethylsilane. MALDI-
TOF mass spectrum was recorded by Miss N. Panchan
on a Bruker Biflex mass spectrometer (The Institute of
Genetic Engineering and Biotechnology, Chula-
longkorn University, Bangkok) using a-cyanocinnamic
acid (CCA) matrix. ESI mass spectra were recorded on
a Fisons Instruments Mass Spectrometer model Trio
2000 in ESI mode. Masses are quoted as m/z unless
otherwise stated, only molecular ions and major frag-
ments being quoted. Absolute methanol and other che-
micals from standard suppliers were used without
further purification. All samples were dried in vacuo
before analyses and biological evaluation.
The order of elution is determined by comparison with
standards. This was found to be dependent on polarity
of N-1 substituents: C6H5>4-MeC6H4>4-ClC6H4>4-
BrC6H4>4-EtC6H4>3,4-Cl2C6H3.
Synthesis of derivatives of 4,6-diamino-1,2-dihydro-1,3,5-
triazines as individual compounds. The individual 4,6-
diamino-1,2-dihydro-1,3,5-triazine derivatives were syn-
thesized by the original Modest two-component synth-
esis (Method A)14 or by the modified method employing
General procedure for synthesis of individual 4,6-dia-
mino-1,2-dihydro-1,3,5-triazine sublibraries by a modi-
fied two-component synthesis. To a suspension of an

T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
223
triethyl orthoacetate as water scavenger (Method B).21
Samples for biological assays and microanalyses were
recrystallized from methanol or ethanol-ether. All
(3H, t, J=7.2 Hz, CH3(CH2)5-2), 1.05–1.70 (5ꢃ2H, m,
CH3(CH2)5-2), 1.17 (3H, s, CH3-2), 2.20 (3H, s, CH3-40)
7.05 (2H, m, aromatic C-H), 7.20 (2H, d, J=8.0 Hz,
aromatic C-H); m/z (MALDI-TOF) 303.7 (M-Cl)+.
1
cycloguanil derivatives were characterized by H NMR
and mass spectra (ESI or MALDI-TOF) and also by
elemental analyses (CHN) for new compounds. Spec-
troscopic and microanalytical data of I-1b, I-1c, I-1d,
I-1f, I-16b, I-16c, I-16d, I-16f have been previously
reported.12
4,6-Diamino-1-(40-ethylphenyl)-2-hexyl-1,2-dihydro-2-methyl-
1,3,5-triazine hydrochloride (I-12e). Method B, 84%
yield; colorless prisms (MeOH–Et2O); found C, 61.4; H,
8.6; N, 19.9% (C18H30ClN5 requires C, 61.4; H, 8.6;
N,19.9%); dH (D2O, 200MHz); 0.67 (3H, t, J=7.2 Hz,
CH3(CH2)5-2), 1.05 (3H, t J=7.0 Hz, CH3CH2-40),
1.10–1.70 (5ꢃ2H, m, CH3(CH2)5-2), 1.20 (3H, s, CH3-
2), 2.02 (2H, q, J=7.0 Hz, CH3CH2-40), 7.12 (2H, m,
aromatic C-H) 7.25 (2H, d, J=8.0 Hz, aromatic C-H);
m/z (MALDI-TOF) 361.5 (M–Cl)+.
4,6-Diamino-1,2-dihydro-2-methyl-1-phenyl-1,3,5-triazine
hydrochloride (I-1a). Method A; 51% yield; colorless
needles (MeOH–Et2O); found: C, 47.0; H, 6.2; N,
27.7%. (C10H14ClN5+H2O requires C, 46.6; H, 6.2; N,
27.2%); dH (D2O) 1.16 (3H, d, J=6.4 Hz, CH3-2), 4.96
(1H, q, J=6.0 Hz, H-2), 7.20 and 7.36 (5H, 2ꢃm, aro-
matic C-H); m/z (MALDI-TOF) 204 (M–C1)+.
4,6-Diamino-1-(30,40-dichlorophenyl)-2-hexyl-1,2-dihydro-
2-methyl-1,3,5-triazine hydrochloride (I-12f). Method B;
92% yield; colorless plates (MeOH–Et2O); found C,
48.9; H, 6.1; N, 17.9% (C16H24Cl3N5 requires C, 48.9;
H, 6.2; N, 17.8%); dH (D2O, 200 MHz); 0.63 (3H, t,
J=7.2 Hz, CH3(CH2)5-2), 1.02–1.70 (5ꢃ2H, m,
CH3(CH2)5-2), 1.68 (3H, s, CH3-2), 7.13 and 7.47
(2ꢃ2H, 2ꢃm, aromatic C-H); m/z (MALDI-TOF) 356.4
(M–C1)+.
4,6-Diamino-(40-ethylphenyl)-1,2-dihydro--2-methyl-1,3,5-
triazine hydrochloride (I–1e). Method A; 43% yield;
colorless needles (EtOH–Et2O); found: C, 53.8; H, 6.6,
N, 26.1% (C12H18ClN5 requires C, 53.8; H, 6.7; N,
26.1%); dH (D2O) 1.01 (3H, t, J=7.0 Hz, CH3CH2-40),
1.16 (3H, d, J=5.9 Hz, CH3-2), 2.05 (2H, q, J=7.4 Hz,
CH3CH2-40), 4.93 (1H, q, J=6.4 Hz, H-2), 7.13 and
7.24 (2ꢃ2H, AB doublet, J=8.2 Hz, aromatic C-H); m/
z (MALDI-TOF) 232 (M.H+).
4,6-Diamino-1,2-dihydro-1,2-diphenyl-1,3,5-triazine
hydrochloride (I–16a). Method A, 78% yield, white solid
(MeOH–Et2O); found: C, 50.7; H, 4.6; N, 19.2%.
(C15H16ClN5+1.5HCl requires C, 50.5; H, 4.9; N,
19.6%); dH (D2O) 5.98 (1H, s, H-2), 7.02 (2H, dd,
J=6.4, 4.0 Hz, aromatic C–H) and 7.22 (8H, m, aro-
matic C–H); m/z (MALDI-TOF) 266 (M–Cl)+.
4,6-Diamino-2-hexyl-1,2-dihydro-2-methyl-1-phenyl-1,3,5-
triazine hydrochloride (I-12a). Method B; 89% yield;
light brown crystalline solid (MeOH–Et2O); found C,
57.40; H, 8.39; N, 21.64% (C16H26ClN5+0.5H2O
requires C, 57.7; H, 8.8; N, 21.0%); dH (D2O,
200 MHz); 0.63 (3H, t, J=7.2 Hz, CH3(CH2)5-2), 1.05–
1.73 (5ꢃ2H, m, CH3(CH2)5-2), 1.15 (3H, s, CH3-2), 7.18
and 7.38 (5H, 2ꢃm, aromatic C-H); m/z (MALDI-
TOF) 288.5 (M–Cl)+.
4,6-Diamino-1-(40-ethylphenyl)-1,2-dihydro-2-phenyl-1,3,5-
triazine hydrochloride (I-16e). Method A, 78% yield;
colorless needles (EtOH–Et2O); found: C, 61.8; H, 6.2,
N, 21.2% (C17H20ClN5 requires C, 61.9; H, 6.1; N,
21.2%); dH (D2O) 0.91 (3H, t, J=6.5 Hz, CH3CH2-40),
2.36 (2H, q, J=7.3 Hz, CH3CH2-40), 5.95 (1H, s, H-2),
6.87 and 7.03 (2ꢃ2H, AB doublet, J=8.0 Hz, aromatic
C–H), 7.17 (5H, m, aromatic C–H); m/z (MALDI-TOF)
294 (M–Cl)+.
4,6-Diamino-1-(40-chlorophenyl)-2-hexyl-1,2-dihydro-2-
methyl-1,3,5-triazine hydrochloride (I-12b). Method B;
85% yield; colorless prisms (MeOH–Et2O); found C,
53.6; H, 7.0; N, 19.6% (C16H25Cl2N5 requires C, 53.6
;H, 7.0 ;N, 19.6%); dH (D2O, 200 MHz); 0.65 (3H, t,
J=7.2 Hz, CH3(CH2)5-2), 1.08–1.70 (5ꢃ2H, 5ꢃm,
CH3(CH2)5-2), 1.18 (3H, s, CH3-2), 7.20 (2H, m, aro-
matic C-H), 7.40 (2H, d, J=8.0 Hz, aromatic C-H); m/z
(MALDI-TOF) 322.5 (M–Cl)+.
Enzyme assays and inhibition by Cyc analogues. The
wild-type and A16V+S108T mutant pfDHFRs were
prepared and their activities were determined spectro-
photometrically according to the method previously
described.22 The reaction (200 mL) contained 1x DHFR
buffer (50 mM N-[tris(hydroxymethyl)methyl-2-ami-
noethanesulfonic acid, pH 7.0, 75 mM b-mercaptoetha-
nol, 1 mg/mL bovine serum albumin), 100 mM each of
the substrate H2folate and cofactor NADPH, and
appropriate amount (0.001–0.005 units) of the affinity-
purified enzymes. The inhibition of the enzymes by Cyc
analogues and combinatorial libraries was determined
in triplicate in a 96 well plate employing 200 mL samples
from the above mixtures in the presence of antifolate.
The reaction kinetic was followed on a microplate reader
(Labsystems, Finland) and the Ki values of inhibitors
were determined by the equation IC50=Ki (1+([S]/
Km)),23 where IC50 is the concentration of inhibitor
which inhibits 50% of the enzyme activity under the
4,6-Diamino-1-(40-bromophenyl)-2-hexyl-1,2-dihydro-2-methyl-
1,3,5-triazine hydrochloride (I-12c). Method B; 86%
yield; colorless needles (MeOH–Et2O); Found C, 47.6;
H, 6.3; N, 17.3% (C16H25BrC1N5 requires C, 47.7 ; H,
6.3 ;N, 17.4%); dH (D2O, 200 MHz); 0.67 (3H, t, J=7.2
Hz, CH3(CH2)5-2), 1.07–1.72 (5ꢃ2H, m, CH3(CH2)5-2),
1.17 (3H, s, CH3-2), 7.15 (2H, m, aromatic C-H), 7.55
(2H, d, J=8.0 Hz, aromatic C-H); m/z (MALDI-TOF)
366.1 and 368.1 (M–Cl)+.
4,6-Diamino-2-hexyl-1,2-dihydro-2-methyl-1-(40-methyl-
phenyl)-1,3,5-triazine hydrochloride (I-12d). Method B;
90% yield; pale yellow crystalline solid (MeOH–Et2O);
Found C, 60.2; H, 8.7; N, 20.9% (C17H24ClN5 requires
C, 60.4; H, 8.4; N, 20.7%); dH (D2O, 200 MHz); 0.64

224
T. Vilaivan et al. / Bioorg. Med. Chem. 11 (2003) 217–224
standard assay conditions and Km is the Michaelis con-
stant for the substrate dihydrofolate. The Ki values were
expressed in nanomolar unit (nM). For Ki of mixture
libraries, the average molecular weights of library
members were used to calculate the molar concentration
of the inhibitors.
7. Peterson, D. S.; Milhous, W. K.; Wellems, T. E. Proc. Natl.
Acad. Sci. U.S.A. 1990, 87, 3018.
8. Basco, L. K.; Eldin de Pecoulas, P. H.; Wilson, C. M.; Le
Bras, J. Mol. Biochem. Parasitol. 1995, 69, 135.
9. Sirawaraporn, W.; Yuthavong, Y. Mol. Biochem. Parasitol.
1984, 10, 355.
10. Warhurst, D. C. Drug Discov. Today 1998, 3, 538.
11. Rastelli, G.; Sirawaraporn, W.; Sompornpisut, P.; Vilai-
van, T.; Kamchonwongpaisan, S.; Quarrell, R.; Lowe, G.;
Thebtaranonth, Y.; Yuthavong, Y. Bioorg. Med. Chem. 2000,
8, 1117.
Acknowledgements
We acknowledge financial support from the Graduate
School and Department of Chemistry, Chulalongkom
University (to N.S.), the European Union (INCO-DC
IC18CT970223) to Y.Y., the Thailand Research Fund
(PDF/57/2540) to T.V., and the World Health Organi-
zation (TDR) to W.S. We also thank Professor Yod-
hathai Thebtaranonth (Mahidol University) for his
helpful comments and guidance, Dr. Aroonsiri Shi-
tangkoon (Chulalongkorn University) for HPLC
method development and Miss Netnapa Char-
oensetakul for technical assistance.
12. Yuthavong, Y.; Vilaivan, T.; Chareonsethakul, N.; Kam-
chonwongpaisan, S.; Sirawaraporn, W.; Quarrell, R.; Lowe,
G. J. Med. Chem. 2000, 43, 2738.
13. For monographs see: Wilson, S. R.; Czarnik, A. W., Eds.;
Combinatorial Chemistry: Synthesis and Application; John
Wiley and Sons; New York, 1997. Terrett, N. K. Combinator-
ial Chemistry; Oxford University Press: Oxford, 1998.
14. Modest, E. J. J. Org. Chem. 1956, 21, 1.
15. Lee, H. K.; Chui, W. K. Bioorg. Med. Chem. 1999, 7,
1255.
16. Modest, E. J.; Levine, P. J. Org. Chem. 1956, 21, 14.
17. Blaney, J. M.; Hansch, C.; Silipo, C.; Vittoria, A. Chem.
Rev. 1984, 84, 333, and references therein.
18. Furka, A.; Sebestyen, F.; Asgedom, M.; Dibo, G. Int. J.
Pept. Protein Res. 1991, 38, 344.
19. Erb, E.; Janda, K. D. Proc. Natl. Acad. Sci. U.S.A. 1994,
91, 11422.
20. Freier, S. M.; Konings, D. A. M.; Wyatt, J. R.; Ecker,
D. J. J. Med. Chem. 1995, 38, 344.
21. Saesaengseerung, N.; Thebtaranonth, Y. ; Vilaivan, T.
Synth. Commun. 2002, 32, 2089.
22. Sirawaraporn, W.; Prapunwattana, P.; Sirawaraporn,
R.; Yuthavong, Y.; Santi, D. V. J. Biol. Chem. 1993,
268, 21637.
23. Segal, I. H. Behavior and Analysis of Steady-State and
Rapid Equilibrium Enzyme Systems. In Enzyme Kinetics;
References and Notes
1. Olliaro, P. L.; Yuthavong, Y. Pharmacol. Ther. 1999, 81, 91.
2. Yuthavong, Y. J. Sci. Soc. Thailand 1996, 22, 181.
3. Peterson, D. S.; Walliker, D.; Wellems, T. E. Proc. Natl.
Acad. Sci. U.S.A. 1988, 85, 9114.
4. Thaithong, S.; Chan, S. W.; Songsomboon, S.; Wilairat, P.;
Seesod, N.; Sueblinwong, T.; Goman, M.; Ridley, R.; Beale,
G. Mol. Biochem. Parasitol. 1992, 52, 149.
5. Hyde, J. E. Parasitol. Today 1989, 5, 252.
6. Foote, S. J.; Galatis, D.; Cowman, A. F. Proc. Natl. Acad.
Sci. U.S.A. 1990, 87, 3014.