Structural studies on bioactive compounds. Part 36: Design, synthesis and biological evaluation of pyrimethamine-based antifolates against Pneumocystis carinii

Article abstract of DOI:10.1016/S0968-0896(02)00128-1

As part of a research effort to improve the quality of current chemotherapy of Pneumocystis carinii pneumonia, we report a structure-based design project to optimise activity, species selectivity and pharmaceutical properties of the triazenyl-pyrimethamine TAB (4) (IC₅₀=0.17 μM; rat liver DHFR IC₅₀/P. carinii DHFR IC₅₀=114). This has led us to design, synthesise and evaluate four new series of pyrimethamine derivatives bearing triazole, triazolium, triazinium and amino moieties at the 3′-position of the p-chlorophenyl ring. Such stabilised `triazene' derivatives address the potentially compromised pharmaceutical profile of TAB and the 3′-amine substituted agents afford conformationally flexible substitutes. The benzylamino-pyrimethamine derivative (24a) (IC₅₀=0.12 μM, rat liver DHFR IC₅₀/P. carinii DHFR IC₅₀: 5.26) was the most potent and the only P. carinii-selective antifolate of the new series.

Full text of DOI:10.1016/S0968-0896(02)00128-1

Bioorganic & Medicinal Chemistry 10 (2002) 3001–3010
Structural Studies on Bioactive Compounds.
Part 36: Design, Synthesis and Biological Evaluation of
Pyrimethamine-Based Antifolates Against
Pneumocystis carinii §
David C. M. Chan,a Charles A. Laughton,a
Sherry F. Queenerb and Malcolm F. G. Stevensa,*
aCancer Research Laboratories, School of Pharmaceutical Sciences, University of Nottingham, Nottingham NG7 2RD, UK
bIndiana University School of Medicine, Indianapolis, IN 46202, USA
Received 22 February 2002; accepted 2 April 2002
Abstract—As part of a research effort to improve the quality of current chemotherapy of Pneumocystis carinii pneumonia, we
report a structure-based design project to optimise activity, species selectivity and pharmaceutical properties of the triazenyl-pyri-
methamine TAB (4) (IC50=0.17 mM; rat liver DHFR IC50/P. carinii DHFR IC50=114). This has led us to design, synthesise and
evaluate four new series of pyrimethamine derivatives bearing triazole, triazolium, triazinium and amino moieties at the 30-position
of the p-chlorophenyl ring. Such stabilised ‘triazene’ derivatives address the potentially compromised pharmaceutical profile of
TAB and the 30-amine substituted agents afford conformationally flexible substitutes. The benzylamino-pyrimethamine derivative
(24a) (IC50=0.12 mM, rat liver DHFR IC50/P. carinii DHFR IC50: 5.26) was the most potent and the only P. carinii-selective
antifolate of the new series. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
selectivity and potency remains unachieved despite the
efforts of a number of research groups.1
The fungus Pneumocystis carinii is one of the most pre-
valent opportunistic microorganisms afflicting indivi-
duals with compromised immune systems. HIV positive
patients are particularly vulnerable and a substantial
number will develop P. carinii pneumonia (PCP) which
gives rise to significant morbidity and mortality. Dihydro-
folate reductase (DHFR) inhibitors such as trimetho-
prim (1) and trimetrexate (2) (Fig. 1) have been drugs of
first choice to treat PCP, but they have limitations.
Maximal clinical efficacy of trimethoprim requires com-
bination with sulphamethoxazole which leads to an
increase in the frequency of adverse drug reactions.
Although trimetrexate does not require the synergistic
effect of sulphonamides, its lack of species selectivity
renders the concomitant use of leucovorin mandatory to
ameliorate severe antifolate toxicity towards the host.
The goal of identifying antifolates with high species
Using pyrimethamine (3) as a starting point Stevens et
al.2 reported that the acetoxyethyl-derivative of a benzyl-
triazenyl-substituted pyrimethamine (4; TAB) has a pro-
mising species selectivity for P. carinii DHFR over that
of the mammalian enzyme. This prompted a new search
for robust, potent and species selective antifolates based
on this pharmacophore.3 The increase in species selec-
tivity of TAB has been attributed to favourable hydro-
phobic contact of the benzyl group with the unique
P. carinii DHFR Phe69 residue.4 (Asn64 is the equiva-
lent residue in the human enzyme.) However, 1-aryl-3,3-
dialkyltriazenes (e.g., 5) as a class are compromised as
therapeutic agents, particularly in the setting where long-
term administration might be required. Firstly, they are
only moderately stable under physiological conditions;5,6
moreover, they are potential substrates for metabolic
dealkylation to mutagenic monoalkyltriazenes (6),7À9
which, following heterolytic degradation, liberate an aryl-
amine (7) and an alkylating alkanediazonium species (8).
The biological sequelae of these events include the
alkylation of nucleic acids (9) (Scheme 1).9
§For Part 35, See ref 30.
*Corresponding author. Tel.: +44-115-951-3414; e-mail: malcolm.
stevens@nottingham.ac.uk
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00128-1

3002
D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
Figure 1. Structures of inhibitors of P. carinii DHFR.
Scheme 1. Metabolism of 3,3-dialkyl-1-aryltriazenes.
Sustaining further interest in triazene-based agents
would depend on a radical redesign to remove these
undesirable features whilst optimising inhibitory activity
against the P. carinii DHFR enzyme target. To provide
greater chemical stability and to remove opportunities
for bio-conversion to monoalkyltriazene species for the
putative second generation diaminopyrimidines, we have
adopted two approaches: (i) to constrain the triazene NNN
linkage within a triazole or triazine ring which removes
the ‘masked diazonium’ character implicit in a triazene
structure;5,6 and (ii) to replace the triazene linker with a
benzanilide or benzylamine moiety. In order to provide a
qualitative assessment of the binding potential of these
novel analogues to P. carinii DHFR, molecular modelling
studies utilising manual docking and molecular dynamics
were initiated with the prototypical benzyltriazolyl-sub-
stituted pyrimethamine (11a; AZO) which was chosen as a
mimic of TAB (4) devoid of the acetoxyethyl substituent
but retaining the critical benzyl function (Fig. 1).
extended benzyl sidechain occupying the pocket above
the plane of the pyrimidine ring, where it could make
extensive hydrophobic contacts with residues such as
Phe36, Ile65, Phe69 and Leu72. Aromatic–aromatic
interactions of the p-stacking type with a centroid
separation of 4 A between the benzyl moiety of AZO
and Phe69 were possible; the benzyl moiety could also
establish a weaker contact of 5.7 A with Phe36 in a face-
edge manner. In the second orientation the torsion
angle between the pyrimidine ring and the phenyl ring
led to the benzyltriazole branch being located in the
opening of the active site cleft, where there were sig-
nificantly fewer possible favourable hydrophobic con-
tacts. In this orientation the N(2) triazole nitrogen was
4.1 A away from the centre of the nicotinamide ring of
NADPH. Only one other significant hydrophobic contact
appeared possible, between the phenyl ring and Leu25. On
the basis of these manual docking studies, it was con-
cluded that this second orientation for AZO within the
active site should be considerably less favoured.
These predictions were then tested by application of the
molecular dynamics/simulated annealing method we
have previously described for performing a potentially
less biased prediction of binding geometries.4 Six MD
runs were performed, each of 100 ps. The resulting final
conformations of AZO are shown in Figure 2, and the
total molecular mechanics energies of the P. carinii
DHFR–AZO–NADPH complex are given in Table 1.
From Figure 2it can be seen that the MD runs led to
the sampling of both classes of AZO conformations
predicted from the manual docking. Runs c, d, and f led
Molecular Modelling
When this study was initiated the only structural infor-
mation available was from the X-ray crystal structure of
the P. carinii DHFR-NADPH-trimethoprim ternary
complex determined by Champness et al.10 Confor-
mational analysis of AZO within the active site of the
DHFR–NADPH complex was performed first through
a manual docking study. Essentially, two orientations
were predicted to be likely. In the first the torsion angle
between the pyrimidine and phenyl rings led to the

D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
3003
to structures where the phenyl sidechain occupied the
upper pocket (illustrated for c in Fig. 3), while a, b, and
e resulted in structures where it sampled the lower
pocket (illustrated for b in Fig. 4). In contrast to the
expectations from the inspection of the manually
docked structures, we see that the lowest energy inter-
action geometry actually corresponds to AZO being
oriented so that the phenyl sidechain occupies the lower
pocket. It is then distant from the residues, particularly
Phe69, that are predicted to be important for providing
species selectivity. Although these results did not augur
well for the activities of these classes of compounds, the
qualitative nature of the modelling study demanded its
full validation through the synthesis of a range of com-
pounds from all structural classes.
yield in the absence of solvent was 71%. The proportion
of each cycloaddition product in the crude reaction
mixture was estimated as 2:1 in favour of the least
sterically crowded 4-benzyltriazole regioisomer (11a),
based on integrating signals due to the a-methylene
protons of the 6-ethyl group and the methylene protons
1
of the benzyl function in the H NMR spectra of the
mixed products. Structure 11a for the major isomer was
1
confirmed by H–1H proximity determinations: irradia-
tion of the triazole 5-H singlet at d 8.27 caused NOE
enhancements in two locations: at d 4.11 (3.7%) and d
7.20 (3.1%) which correspond to the benzyl methylene
Chemistry
1,2,3-Triazole derivatives
Interaction of m-azidopyrimethamine (10), prepared by
the method of Bliss et al.,11 with an excess (5 mol equiv)
of benzylacetylene in DMF at 100 ꢀC for 48 h resulted
in the isolation of two regioisomeric cycloaddition
products (11a,b) in 77% total yield (Scheme 2); overall
Figure 3. Orientation of AZO (11a) in the active site of P. carinii
DHFR resulting from MD run c (green). The corresponding structure
predicted by manual docking is also shown (grey).
Figure 2. Binding conformations of AZO (11a) at P. carinii DHFR
active site after six MD simulations. The views are relative to an identical
orientation of the protein in each case, but this is not shown, for clarity.
Table 1. Summary of the total molecular mechanics energies of the
P. carinii DHFR–AZO–NADPH complex after 100 ps simulated
annealing run at P. carinii DHFR active site
Simulation
Energy (kcal molÀ1
)
a
b
c
d
e
f
À341.0
À366.1
À350.0
À355.4
À335.4
À311.3
Figure 4. Orientation of AZO (11a) in the active site of P. carinii
DHFR resulting from MD run b (green). The corresponding structure
predicted by manual docking is also shown (grey).

3004
D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
Scheme 2. Reagents: (a) R1CꢁCR2 in DMF, 100 ꢀC; (b) TBAF in THF.
protons and H-20 in the chlorophenyl ring, respectively.
In the reaction of azide (10) with phenylacetylene the
ratio of the phenyltriazoles (11c) and (11d) was 3:2
(79% overall yield) in favour of the 4-substituted isomer
(11c). Irradiation of the 5-H singlet of 11c at d 9.0 gave
rise to two NOE enhancements (0.5 and 2.4%) corre-
sponding to the two 20-aromatic protons of the chloro-
phenyl and phenyl rings, respectively.
salts.19,20 In 1968, Shealy recorded this reaction follow-
ing his synthesis of the experimental anti-leukemic agent
5-[3,3-bis(2-chloroethyl)triazen-1-yl]imidazole-4-carb-
oxamide (12) which cyclised in the solution and solid
states to the biologically inactive ionic triazolium salt
(13) (Scheme 3).21À23 Hansen et al. have also reported
the one-pot cyclisation of hydroxyalkyl-triazenes (14)
with mesyl chloride to the dihydrotriazolium salts (16):
the intermediate methanesulfonyl esters (15) could not
be isolated.24
The lack of efficient chromatographic methods to sepa-
rate the regioisomers (11a,b) and (11c,d) thwarted the
efficient isolation of pure products and led us to seek
alternative routes to substituted triazoles. Reaction
between azide (10) and the sterically-demanding (tri-
methylsilyl)acetylene was regiospecific, the 4-(tri-
methylsilyl)triazole (11e) being formed in 85% yield:
irradiation of the triazole H-5 singlet at d 8.52induced
the expected NOE enhancement (3.7%) at H-20 in the
chlorophenyl ring. Other cycloaddition reactions invol-
ving (trimethylsilyl)acetylene have been reported to
proceed regioselectively.12À14 Sakamoto et al. have
reported an efficient two-step synthetic synthesis of 1,4-
diphenyl-1,2,3-triazole15 from the regioselective addi-
tion reaction of (trimethylstannyl)acetylene with phenyl
azide to give exclusively the 4-trimethylstannyl inter-
mediate followed by a palladium-catalysed phenylation
reaction to yield the diphenyltriazole. The air-sensitive
(trimethylstannyl)acetylene is not commercially available
and its synthesis and storage are problematic.16À18 The
commercially available (tributylstannyl)acetylene (10 mol
equiv) was chosen to replace (trimethylstannyl)acetylene
and was reacted with 10 at 110 ꢀC for 4 days in an-
hydrous DMF. None of the expected (tributylstannyl)-
triazole (11f) was formed and the only product isolated
was the 1-substituted triazole (11g) (39%). The latter
triazole could be prepared more efficiently (74%) by
removal of the TMS group of 11e with a slight excess of
TBAF (1.1 mol equiv) (Scheme 2).
In the present work, triazenes (19a,b) were formed by
coupling diazonium salts derived from 2-chloroaniline
(17; R=H) with benzylalkanolamines (18a,b) in aqu-
eous sodium carbonate. These were then converted with
thionyl chloride (as solvent and reactant) at 0 ꢀC, via the
intermediate chloroalkyltriazenes (20a,b), to the model
triazolium (21a) and triazinium (21b) products, respec-
tively (Scheme 4). The diazonium chloride (17; R=2,4-
diamino-6-ethylpyrimidin-5-yl)11 was similarly coupled
with substituted benzylalkanolamines (18c–f) to yield
hydroxyalkyl-triazenes (19c–f) which were converted
with thionyl chloride to the required diaminopyri-
midinyl-substituted salts (21c–f) which were precipitated
from the reaction mixtures in yields >75% by addition
of excess ether. The products were freely water-soluble
1
consistent with their ionic nature, and their H NMR
spectra did not show any sign of peak broadening of the
signals of the methylene protons in the dihydro-
triazolium and tetrahydrotriazinium rings. Ring-opened
triazenyl-pyrimidines related in structure to intermediate
chloroalkyl-triazenes (20c–f), which were not isolated,
do show this characteristic feature because of restricted
rotation around the N¼N bond.3
Benzanilide and benzylamine derivatives
Acylation of 5-(3-amino-4-chlorophenyl)-2,4-diamino-
6-ethylpyrimidine (22)11 with the appropriate acid chloride
in hot pyridine furnished the anilides (23a,b). Benzyl-
amines (24a,b) were obtained by alkylation of 22 with
the requisite benzyl bromide in refluxing isopropanol
(Scheme 5). Subsequently, reductive methylation reac-
tions were carried out on the two secondary amines
1,2,3-Triazolium and 1,2,3-triazinium derivatives
The facile cyclisation of aryl-[3,3-bis(2-chloroethyl)]-
triazenes by intramolecular displacement of a chloro group
has provided access to a number of 1,2,3-triazolium

D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
3005
using 37% formaldehyde and sodium cyanoboro-
hydride in an acidic medium to afford the methylated
analogues (24c,d) in >90% yield.
of four triazolyl analogues (11a,c,e,g) were evaluated as
inhibitors of P. carinii, T. gondii, M. avium and rat liver
DHFR. The two most potent compounds against the P.
carinii enzyme were those with a benzyl (11a) or phenyl
group (11c) which had IC50 values of 5.18 and 3.53 mM,
respectively. By comparison, the TMS analogue (11e)
and the unsubstituted triazole (11g) were less active
(IC50 values 10.5 and 24.8 mM, respectively). This class
of compounds successfully addressed the problematic
Biological Results
The assay methods used were described in previous
reports25À27 and results are given in Table 2. The activity
Scheme 3. (a) Spontaneous cyclisation; (b) mesyl chloride.
Scheme 4. (a) Aqueous Na2CO3, 0 ꢀC; (b) SOCl2 in Et2O, 0 ꢀC; (c) spontaneous cyclisation.

3006
D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
i
Scheme 5. Reagents: (a) RCOCl, pyridine, reflux; (b) R1CH2Br, PrOH, reflux; (c) HCHO/HCl/NaBH3CN in MeCN.
Table 2. Structure–activity relationship of pyrimethamine based 2,4-
diaminopyrimidines towards P. carinii (pc), rat liver (rl) T. gondii (tx),
and M. avium (ma) DHFR
positively charge species in a hydrophobic pocket would
severely impede binding was not realised. Four com-
pounds with a nitrogen bridge (benzanilides and benzy-
lamines) were tested for activities. Compounds in the
benzanilide series (23a,b) are less potent than the ben-
zylamines (24a,c). In the former series, addition of tri-
methoxy groups (23b) resulted in a near 10-fold increase
in activity over that of the unsubstituted benzanilide
(23a). With an IC50 value of 0.12 mM compound (24a) is
the most potent member of the group as an inhibitor of
P. carinii DHFR and is also the only agent with species
selectivity (5.26 vs rat liver DHFR). Interestingly,
methylation of the amino bridge (24c) resulted a 10-fold
reduction of activity towards the P. carinii enzyme and
nearly a 7-fold gain of potency towards the rat liver
DHFR.
Compd
IC50 (mM) vs DHFR
Selectivity ratio
P. carinii rat liver T. gondii M. avium rl/pc rl/tx rl/ma
2a
0.0420.003
0.01
—
0.07 0.30
114 28
0.27 2.65
0.04 0.41
0.32 1.83 1.34
0.02 1.36 0.12
0.24 2.39
0.38 8.04
0.77 12.5
59.2.0 —
—
4a
0.17
5.18
3.53
19.4
0.69
0.52
0.37
1.83
0.28
1.38
0.51
0.08
—
—
—
2.5
3.18
—
—
—
6
—
—
—
11a
11c
11e
11g
21c
23a
23b
24a
24c
1.38
0.15
3.34
0.38
3.3
10.5
24.8
13.5
10.8
1.3
—
—
—
4.1
0.99
0.120.63
1.07
0.07
—
0.09
0.02—
0.08 4.50 —
aData reported previously, see refs 2and 3.
Overall, the compounds listed in Table 2were con-
sistently more potent against the T. gondii enzyme than
the P. carinii enzyme. The most potent compounds were
the benzylamines (24a,c) but the trimethoxybenzoyl
derivative (23b) displayed the greatest selectivity (selec-
tivity ratio of 12.5 vs rat liver DHFR).
stability issue associated with the triazene TAB (4) but
did not produce superior activity nor comparable spe-
cies selectivity to the lead structure. This result was
predicted from modelling studies which suggested that
aromatic interaction with Phe69 alone cannot dictate
significant species selectivity. Nevertheless, it is apparent
from the structure–activity relationships that substitu-
tion at the 4-position of the triazole ring gives rise to
compounds with improved activity compared with pre-
viously reported triazenyl-pyrimethamine derivatives,3
presumably through an increasingly optimal interaction
with the enzyme active site residues. Only one analogue
with a triazolium moiety (21c) was evaluated for activity
(IC50 13.5 mM against P. carinii DHFR; 3.3 mM against
rat DHFR). Although the DHFR inhibitory activity is
not substantial, our intuitive concern that positioning a
Conclusion
The outcome of the manual docking study of the
benzyltriazolyl-substituted pyrimethamine (11a; AZO)
correlates well with the result of the dynamics simu-
lations. Although the ligand orientations generated
from dynamics simulations are expected to be only
qualitatively correct, these results have shown that
automated docking using simulated annealing can pro-
duce an effective search of conformational space.
Deriving useful quantitative data such as binding

D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
3007
constants of the designed compounds was beyond the
scope of the present molecular modelling study, because
high quality ternary X-ray crystal data of the P. carinii
DHFR–TAB–NADPH ternary complex was not avail-
able at the beginning of this project. Four series of
structurally diverse analogues with varying chemical
and physical properties have been designed, synthesised
and assayed against DHFR enzymes from different
species. Although compound (11a) was thought to be a
promising mimic of TAB by molecular modelling ana-
lysis, it did not display any substantial potency or spe-
cies selectivity towards P. carinii DHFR. Compounds
belonging to the triazole (11), triazolium and triazinium
(21) series showed activities consistently weaker than
the benzylamine series (24). Although the benzylamine
(24a) (IC50 0.12 mM; rat liver DHFR IC50/P. carinii
DHFR IC50=5.26) was the most active (and the only
selective) antifolate towards P. carinii within the four
series of compounds, it was not as potent or selective as
the original lead compound, (acetoxyethyl)benzyl-
triazenyl-pyrimethamine (4; TAB).
ethylacetate mixture. TLC spots were visualised using
UV light. Flash column chromatography was per-
formed using silica gel C60H.
2,4-Diamino-5-[3-(4-benzyl-1,2,3-triazol-1-yl)-4-chloro-
phenyl]-6-ethylpyrimidine (11a; AZO) and regioisomer
(11b). To m-azidopyrimethamine11 (10; 1.5 g, 5.2
mmol) in anhydrous DMF (10 mL) was added benzyl-
acetylene (3.0 g, 25.9 mmol, 5 mol equiv). The suspen-
sion was heated at 100 ꢀC for 48 h and the brown
solution was then cooled to 25 ꢀC and was shaken with
hexane (3Â50 mL) to remove the excess benzyla-
cetylene. Ice cold water was added to the mixture to
precipitate a mixture of 11a and 11b in the ratio 2:1
(total yield 77%). A pure sample of 11a (120 mg) could
be isolated from the regioisomeric mixture (600 mg) by
flash column chromatography in silica with chloroform/
methanol/ammonium acetate mixture (400:10:1) as elu-
ent. Crystallisation from ethanol gave creamy needles,
mp 122–123 ꢀC; IR (KBr) ꢀmax/cmÀ1 3381, 3149 (NH),
2972, 1601, 1551, 1431, 1040, 721; 1H NMR d 0.97 (3H,
t, J=7.5 Hz, CH3), 2.15 (2H, q, J=7.5 Hz, CH2), 4.11
(2H, s, CH2Ph), 5.80 (2H, br s, NH2), 5.97 (2H, br s,
NH2), 7.20–7.74 (7H, m, Ar-H), 7.76 (1H, d, J=8.2,
Experimental
Ar-H), 8.27 (1H, s, CH); 13C NMR d 13.2(CH ), 27.7
3
Molecular modelling
(CH2), 31.2(CH ), 104.4 (C), 124.8 (CH), 126.4 (CH),
2
127.0 (C), 128.7 (C), 128.8 (C), 130.3 (CH), 131.1
(CH), 133.9 (CH), 135.1 (C), 136.9 (C), 139.5 (C),
146.2 (C), 162.1 (C), 162.5 (C), 166.7 (C); FAB+-HRMS
Comprehensive details have been given elsewhere.4
Briefly, the docking study was performed using the
DISCOVER module of the molecular modelling pack-
age Insight II28 as an interface to the AMBER 3.5 force
field while the stand alone AMBER 4.0 package29 was
used for the molecular dynamics simulations.
calcd for C21H21N7Cl: 406.154697, 408.151747
[M+H]+; found: 406.154751, 408.153182[M+H]
+
.
The regioisomeric mixture could be prepared by heating
azidopyrimethamine (10) (50 mg, 0.17 mmol) in benzyl-
acetylene (0.5 mL) at 100 ꢀC for 18 h in a sealed tube.
After heating for ca. 2h the off-white suspension went
dark and subsequently formed a deep brown solution.
The reaction mixture was then cooled to room tempera-
ture and washed with several portions of hexane (20 mL)
until precipitation occurred. The yellow solid was fil-
tered and characterised by 1H NMR. The spectrum
corresponds to a sample of that of the regioisomeric
mixture of 11a and 11b in the ratio 2:1 (total yield
71%).
Synthetic chemistry
Chemicals were purchased from Aldrich Chemical Co.
and Fisher Scientific and other commercial sources and
were used without further purification. Melting points
were determined in open capillary tubes on a Gallen-
kamp melting point apparatus and are uncorrected. IR
spectra were measured in KBr on a Mattson 2020
Galaxy Series FT-IR spectrometer. The optical mea-
surements were made in tubes with length 20 cm on a
Bellingham & Stanley ADP 220 polarimeter. 1H and 13
C
2,4-Diamino-5-[4-chloro-3-(4-phenyl-1,2,3-triazol-1-yl)-
phenyl]-6-ethylpyrimidine (11c) and regioisomer (11d).
Similarly prepared, from m-azidopyrimethamine11 (10;
2.0 g, 6.9 mmol) in anhydrous DMF (15 mL) and phe-
nylacetylene (3.53 g, 34.5 mmol; 5 mol equiv) at 100 ꢀC
for 48 h was a mixture of 11c and 11d (79%) in a ratio
3:2(by 1H NMR). A pure sample (190 mg) of 11c could
be isolated from the regioisomeric mixture (1.0 g) by
flash column chromatography in silica with chloroform/
methanol/ammonium acetate mixture (400:10:1) as
eluent. Crystallisation of 11c from acetone gave white
needles, mp 174–176 ꢀC; IR (KBr) ꢀmax/cmÀ1 3395,
3324, 1601, 1553, 1433, 1239, 1099, 764; 1H NMR d 1.01
(3H, t, J=7.5 Hz, CH3), 2.20 (2H, q, J=7.5 Hz, CH2),
5.85 (2H, br s, NH2), 5.99 (2H, br s, NH2), 7.35–7.53
(5H, m, Ar–H), 7.55 (1H, d, J=2.0, Ar–H), 7.80 (1H, d,
J=8.3, Ar–H5), 7.80 (1H, d, J=7.0, Ar–H), 8.99 (1H, s,
NMR spectra were recorded on a Bruker spectrometer
in DMSO-d6 solutions at 250.1 and 62.9 MHz, respec-
tively. Chemical shifts (d) are reported in parts per mil-
lion (ppm) with tetramethysilane as an internal
standard. 13C assignments for secondary and qua-
ternary carbons were based on DEPT135 experiments.
MS were recorded on an ANI MS-902, a VG Micromass
7070E or a VG platform spectrometer. Elemental ana-
lyses were performed on either a Perkin-Elmer PE240B
Elemental Analyser or an Exeter Analytical CE-440
Elemental Analyser by the Microanalysis service at the
Department of Chemistry, University of Nottingham,
UK. Samples were dried in vacuo overnight in the pre-
sence of phosphorous pentoxide at room temperature
prior to the submission for elemental analysis. Pre-
coated silica gel 60F254 plates were used as the absor-
bent for thin layer chromatography, with the developing
solvent being either chloroform–methanol or hexane–
CH); 13C NMR d 13.2(CH ), 27.6 (CH2), 104.4 (C),
3
123.6 (CH), 125.6 (CH), 127.2 (C), 128.4 (CH), 129.2

3008
D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
(CH), 130.3 (C), 130.4 (CH), 131.2(CH), 134.2(C),
135.0 (C), 136.9 (C), 146.6 (C), 162.1 (C), 162.2 (C),
166.5 (C); ES+-MS m/z 392, 394 [M+H]+. Anal. calcd
for C20H18N7Cl 0.65H2O: C, 59.53; H, 4.82; N, 24.29;
found: C, 59.87; H, 4.58; N, 23.89%.
to room temperature slowly whereupon a solution was
formed. The reaction mixture was stirred at room tem-
perature overnight before the addition of anhydrous
diethyl ether (20 mL). An orange coloured suspension
was formed. The reaction mixture was stirred for a fur-
ther 12h and the orange solid was collected and washed
well with diethyl ether under a stream of nitrogen.
Compounds were dissolved in isopropanol and re-pre-
cipitated by addition of anhydrous cyclohexane. The
cream solid triazolium and triazinium chloride salts
were collected under an inert atmosphere. The following
compounds were prepared.
2,4-Diamino-5-[4-chloro-3-(4-trimethylsilyl-1,2,3-triazol-
1-yl)phenyl]-6-ethylpyrimidine (11e). Similarly prepared,
from m-azidopyrimethamine11 (10; 2g, 6.9 mmol) in
anhydrous DMF (25 mL) and (trimethylsilyl)acetylene
(2.03 g, 20.7 mmol, 3 mol equiv) in an inert atmosphere,
the chalky pink solid was purified using flash column
chromatography on silica with chloroform/methanol
mixture as eluent (9:1). Crystallisation from ethanol
afforded colourless needles (85%), mp 223–225 ꢀC; IR
(KBr) ꢀmax/cmÀ1 3397, 3327, 3167 (NH), 1601, 1561,
1431, 1250, 843; 1H NMR d 0.33 (9H, s, Si(CH3)3), 1.02
(3H, t, J=7.5 Hz, CH3), 2.17 (2H, q, J=7.5 Hz, CH2),
5.78 (2H, br s, NH2), 5.94 (2H, br s, NH2), 7.39–7.43
(2H, m, Ar–H), 7.78 (1H, d, J=8.2Hz, Ar–H), 8.52
1-Benzyl-3-(2-chlorophenyl)-4,5-dihydro-1,2,3-triazolium
chloride hydrochloride (21a). From 3-benzyl-1-(2-chloro-
phenyl)-3-(2-hydroxyethyl)triazene (19a), yield (88%);
IR (NaCl) ꢀmax/cmÀ1 3407, 1633, 1510, 1485, 1256, 860,
1
764, 704; NMR d 4.63 (2H, t, J=12.3 Hz, NCH2),
4.87 (2H, t, J=12.3 Hz, NCH2), 4.47 (2H, s, CH2Ph),
7.46–7.84 (9H, m, Ar–H); 13C NMR d 54.9 (CH2), 55.0
(CH2), 56.7 (CH2), 127.0 (CH), 127.6 (C), 129.0 (CH),
129.3 (CH), 129.5 (CH), 129.9 (CH), 131.4 (C), 131.5
(CH), 132.2 (CH), 134.0 (C); EI+-HRMS m/z
272.09593 (M); calcd for C15H15N3Cl, 272.09546.
(1H, s, CH); 13C NMR d À0.8 (CH3), 13.2(CH ), 27.7
3
(CH2), 79.4 (CH), 104.4 (CH), 127.0 (C), 130.3 (CH),
131.2(CH), 132.5 (CH), 133.8 (CH), 135.1 (C), 136.8
(C), 145.1 (C), 162.1 (C), 162.5 (C), 166.8 (C); ES+-MS
m/z 388, 390 [M+H]+. Anal. calcd for C17H22N7ClSi
0.2H2O 0.1 EtOH: C, 52.15; H, 5.85; N, 24.75; found:
C, 51.84; H, 5.64; N, 24.76%.
1-Benzyl-3-(2-chlorophenyl)-3,4,5,6-tetrahydro-1,2,3-tri-
azinium chloride hydrochloride (21b). From 3-benzyl-1-
(2-chlorophenyl)-3-(3-hydroxypropyl)triazene (19b), yield
(73%); 1H NMR d 2.28 (2H, quintet, J=5.5 Hz,
CH2CH2CH2), 3.95 (2H, t, J=5.5 Hz, NCH2), 4.16
(2H, t, J=5.5 Hz, NCH2), 5.32(2H, s, CH 2Ph), 7.31–
7.90 (9H, m, Ar–H); 13C NMR 14.3 (CH2), d 46.9
(CH2), 49.0 (CH2), 62.2 (CH2), 128.5 (CH), 128.8 (C),
129.1 (CH), 129.3 (CH), 129.4 (CH), 130.4 (CH), 131.0
(C), 132.1 (CH), 132.5 (CH), 139.8 (C); FAB+-HRMS
matrix NBA m/z 286.111458 (M); calcd for
C16H17N3Cl, 286.111100.
2,4-Diamino-5-[4-chloro-3-(1,2,3-triazol-1-yl)phenyl]-6-
ethylpyrimidine (11g). (i) 2,4-Diamino-5-[4-chloro-3-(4-
trimethylsilyl-1,2,3-triazol-1-yl)phenyl]-6-ethylpyrimidine
(11e; 450 mg, 1.15 mmol) was dissolved in THF (5 mL)
and tetrabutylammonium fluoride (TBAF; 0.47 mL
75% w/v, 1.1 mol equiv) was added dropwise. A fine
precipitate was observed after ca. 3 h of stirring and the
reaction mixture was stirred at 25 ꢀC overnight. Excess
solvent was removed under reduced pressure and the
off-white solid obtained was suspended in boiling water
and recovered by filtration. The triazole was crystallised
from ethanol to give an off-white solid (74%), mp 243–
244 ꢀC; IR (KBr) ꢀmax/cmÀ1 3426, 3320, 3163 (NH),
1-Benzyl-3-[2-chloro-5-(2,4-diamino-6-ethylpyrimidin-5-
yl)]phenyl-4,5-dihydro-1,2,3-triazolium chloride hydro-
chloride (21c). From 2,4-diamino-5-{3-[3-benzyl-3-(2-
hydroxyethyl)triazen-1-yl]-4-chlorophenyl}-6-ethylpyri-
midine (19c),2 yield 82%; mp 178 ꢀC (efferv); IR (KBr)
1
1634, 1557, 1439, 1235, 1099; H NMR d 0.99 (3H, t,
J=7.5 Hz, CH3), 2.16 (2H, q, J=7.5 Hz, CH2), 5.80
(2H, br s, NH2), 5.96 (2H, br s, NH2), 7.39–7.45 (2H,
m, Ar–H), 7.78 (1H, d, J=8.1 Hz, Ar–H), 7.98 (1H, s,
ꢀ
max/cmÀ1 3324, 3156, 1655, 1508, 1456, 1252), 1045,
1
704; H NMR d 1.09 (3H, t, CH3), 2.26 (2H, q, CH2),
4.60 (2H, t, J=12.5 Hz, NCH2), 4.77 (2H, t, J=12.5
Hz, NCH2) 5.43 (2H, s, CH2Ph), 7.09 (1H, s, NH, exch
D2O), 7.46–7.88 (m, 8H, Ar–H), 8.31 (1H, br s, NH,
exch D2O), 13.22 (1H, br s, NH, exch D2O); 13C NMR
CH), 8.51 (1H, s, CH); 13C NMR d 13.2(CH ), 27.7
3
(CH2), 104.3 (C), 127.2 (CH), 130.4 (CH), 131.1 (CH),
133.6 (CH), 134.0 (CH), 135.0 (C), 137.0 (C), 162.1
(C), 162.5 (C), 166.8 (C); ES+-MS m/z 316, 318
[M+H]+. Anal. calcd for C14H14N7Cl 0.25H2O: C,
52.50; H, 4.56; N, 30.61; found: C, 52.72; H, 4.44; N,
30.92%.
d 12.6 (CH3), 24.0 (CH2), 55.1 (CH2), 55.2(CH ), 56.9
2
(CH2), 106.1 (C), 128.2 (C), 128.8 (CH), 129.3 (CH),
129.6 (CH), 130.0 (CH), 130.6 (CH), 131.3 (C), 132.1
(C), 132.5 (CH), 134.3 (CH), 134.9 (C), 154.7 (C), 155.4
(C), 163.9 (C); ES+-MS m/z 408, 410 [MÀH]+; FAB+-
HRMS calcd for C21H24N7Cl: 408.170347, 410.167397
[MÀH]+; found: 408.171020, 410.169142 [MÀH]+.
(ii) The same triazole (39%) was isolated when m-azido-
pyrimidine was reacted with (tributylstannyl)acetylene
(10 mol equiv) in DMF under an inert atmosphere at
110 ꢀC for 4 days.
1-Benzyl-3-[2-chloro-5-(2,4-diamino-6-ethylpyrimidine-5-
yl)]phenyl-3,4,5,6-tetrahydro-1,2,3-triazinium chloride
(21d). From 2,4-diamino-5-{3-[3-benzyl-3-(3-hydroxy-
General procedure for the synthesis of compound (21a–
f). To ice cold thionyl chloride (5 mL) was added the
hydroxyalkyltriazene (19; 2.5 g, 5.9 mmol) portionwise
over a period of 5 min. An amber suspension was
formed and the reaction mixture was allowed to warm
propyl)-triazene-1-yl]-4-chlorophenyl}-6-ethylpyrimidine
(19d),3 yield 77%; mp 178 ꢀC (decomp); IR (KBr) ꢀmax
/
cmÀ1 3310, 3146 (N–H), 1657, 1535, 1481, 1290, 1194,

D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
3009
700; 1H NMR d 1.09 (3H, t, J=7.5 Hz, CH3), 2.28 (4H,
m, CH2CH3 and CH2CH2CH2), 3.95 (2H, br s, NCH2),
4.25 (2H, br s, NCH2), 7.15 (1H, br s, NH2), 7.44–7.94
(m, 8H, Ar–H), 8.27 (1H, br s, NH, exch D2O), 13.3
(1H, br s, NH); 13C NMR d 12.3 (CH3), 14.3 (CH2),
24.0 (CH2), 47.2(CH ), 49.2(CH ), 64.3 (CH2), 106.0
(C), 128.7 (CH), 129.2 (C), 129.3 (CH), 129.4 (CH),
129.5 (CH), 130.5 (CH), 130.6 (CH), 132.0 (CH), 132.1
(C), 132.2 (C), 134.6 (CH), 140.4(C), 154.5(C), 155.4(C),
163.9 (C); FAB+-HRMS calcd for C22H25N7Cl:
422.185997, 424.183047 [MÀH]+; found: 422.187354,
424.184595 [MÀH]+.
(23a), 3.53 g (95%); mp 164–165 ꢀC; IR (KBr) ꢀmax/cmÀ1
3385, 3169 (NH), 1647 (C¼O), 1574, 1439, 1406, 1283,
706; 1H NMR d 0.99 (3H, t, J=7.5 Hz, CH3), 2.15 (2H,
q, J=7.5 Hz, CH2), 5.64 (2H, br s, NH2), 5.96 (2H, br s,
NH2), 7.10 (1H, dd, J=2.5, 7.5 Hz, Ar–H), 7.47 (1H, d,
J=2.5 Hz, Ar–H), 7.51–7.68 (4H, m, Ar–H), 7.99 (2H,
d, J=7.5 Hz, Ar–H), 10.11 (1H, br s, NH); 13C NMR d
13.4 (CH3), 27.7 (CH2), 105.3 (CH), 127.7 (C), 128.7
(CH), 129.6 (CH), 129.8 (CH), 130.3 (CH), 132.1 (CH),
134.2 (C), 135.5 (C), 135.6 (C), 162.1 (C), 162.3 (C),
165.6 (C), 166.8 (C); ES+-MS m/z 368, 370 [M+H]+.
Anal. calcd for C19H18N5OCl 0.3 EtOH 0.2H2O: C,
61.11; H, 5.28; N, 18.18; found: C, 60.94; H, 4.99; N,
18.11%.
2
2
S-(+)-1-(ꢀ-Methylbenzyl)-3-[2-chloro-5-(2,4-diamino-6-
ethylpyrimidin-5-yl)]phenyl-3,4-dihydro-1,2,3-triazolium
chloride hydrochloride (21e). From S-(+)-2,4-diamino-
5-{3-[3-(2-hydroxyethyl)-3-(a-methylbenzyl)triazen-1-yl]-
4-chlorophenyl}-6-ethylpyrimidine (19e),3 yield 93%; mp
2,4 - Diamino - 5 - [3 - (3,4,5 - trimethoxybenzoyl)amino - 4 -
chlorophenyl]-6-ethylpyrimidine (23b). Prepared from 22
and 3,4,5-trimethoxybenzoyl chloride in pyridine (35
mL) as above, the trimethoxybenzoylamino-pyrimidine
(14%) had mp 265–267 ꢀC (decomp) (from ethanol); IR
(KBr) ꢀmax/cmÀ1 3434, 3179 (N–H), 1630, 1574, 1503,
1335, 1120, 997; 1H NMR d 0.95 (3H, t, J 7.5 Hz, CH3),
2.18 (2H, q, J=7.5 Hz, CH2), 3.73 (3H, s, OCH3), 3.86
(6H, s, 2ÂOCH3), 5.60 (2H, br s, NH2), 5.95 (2H, br s,
NH2), 7.11 (1H, dd, J=2.5, 7.5 Hz, Ar–H), 7.33 (2H, s,
Ar–H), 7.37 (1H, d, J=2.5 Hz, Ar–H), 7.64 (1H, d,
J=7.5 Hz, Ar–H), 10.03 (1H, s, NH); 13C NMR d 13.3
28:0
ꢀ
132 ꢀC (efferv); ½ꢁꢂD 56.0 (c 0.804 mg mLÀ1, EtOH);
IR (KBr)ꢀmax/cmÀ1 3341, 3169, 1655, 1485, 1452, 1240,
702; H NMR d 1.11 (3H, t, J=7.5 Hz, CH2CH3), 1.83
1
(3H, d, J=7.5 Hz, CHCH3), 2.28 (2H, q, J=7.5 Hz,
CH2CH3), 4.40–4.88 (4H, m, 2ÂCH2), 5.63 (1H, q,
J=7.5 Hz, CHCH3), 7.11 (1H, br s, NH, exch D2O),
7.45–7.90 (m, 8H, Ar–H), 8.32(1H, br s, NH, exch
D2O), 13.36 (1H, br s, NH, exch D2O); 13C NMR d 12.5
(CH3), 19.8 (CH3), 24.0 (CH2), 54.2(CH ), 54.9 (CH2),
2
63.6 (CH), 106.0 (C), 127.8 (CH), 128.9 (C), 129.4
(2ÂCH), 132.0 (C), 132.5 (CH), 134.0 (CH), 134.9 (C),
136.6 (C), 154.5 (C), 155.4 (C), 163.8 (C); FAB+-
HRMS calcd for C22H25N7Cl: 422.185997, 424.183047
[M–H]+; Found: 422.187316, 424.184664 [M–H]+.
(CH3), 27.7 (CH2), 56.2(CH ), 60.3 (CH3), 105.2(C),
3
105.5 (CH), 128.1 (C), 129.3 (C), 129.8 (CH), 130.3
(CH), 135.5 (C), 135.6 (C), 140.7 (C), 152.7 (C), 162.1
(C), 162.4 (C), 165.0 (C), 166.8 (C); ES+-MS m/z 458,
460 [M+H]+. Anal. calcd for C22H24N5O4Cl 0.2H2O
0.2EtOH: C, 57.15; H, 5.48; N, 14.88; found: C, 57.15;
H, 5.16; N, 14.85%.
R-(À)-3-(ꢀ-Methylbenzyl)-1-[2-chloro-5-(2,4-diamino-6-
ethylpyrimidin-5-yl)]phenyl-3,4-dihydro-1,2,3-triazolium
chloride hydrochloride (21f). From R-(À)-2,4-diamino-
5-{3-[3-(2-hydroxyethyl)-3-(a-methylbenzyl)triazen-1-yl]-
4-chloro-phenyl}-6-ethylpyrimidine (19f),3 yield 89%; mp
2,4-Diamino-5-{4-chloro-3-[N-(benzyl)amino]phenyl}-6-
ethylpyrimidine (24a). (i) Aminopyrimethamine (22)
(1.32g, 5 mmol) was suspended in isopropanol (30 mL)
and benzyl bromide (1.71 g, 10 mmol) was added. The
mixture was refluxed for 36 h and upon cooling, an off-
white precipitate formed. The product was collected,
stirred with aqueous ammonia to furnish the free base.
Flash column chromatography (dichloromethane/
methanol 10:1) followed by crystallisation from acetone
gave the benzylamine (24a) (66%); Rf (chloroform/
methanol 5:1) 0.71; mp 180–182 ꢀC; IR (KBr)ꢀmax/cmÀ1
3454, 3318, 3162(NH), 1630, 1570, 1433, 1030, 690; 1H
NMR d 0.72(3H, t, J=7.5 Hz, CH3), 1.90 (2H, q,
J=7.5 Hz, CH2CH3), 4.42(2H, dd, J=6.6 Hz, CH2Ph),
5.61 (2H, br s, NH2), 5.90 (2H, br s, NH2), 6.19 (1H, t,
J=6.1 Hz, NH), 6.25 (1H, s, Ar–H), 6.34 (1H, dd,
J=1.3, 6.8 Hz, Ar–H), 7.15–7.32(6H, m, Ar–H); 13C
NMR d 13.2(CH ), 27.4 (CH2), 46.2(CH ), 106.4 (C),
27:8
122ꢀC (efferv); ½ꢁꢂD À53.4ꢀ (c 0.562mg mL À1, EtOH);
IR (KBr) ꢀmax/cmÀ1 3324, 3167, 1657, 1485, 1452, 1246,
1
704; H NMR d 1.11 (3H, t, J=7.5 Hz CH2CH3), 1.83
(3H, d, J=7.5 Hz, CHCH3), 2.31 (2H, q, J=7.5 Hz,
CH2CH3), 4.39–4.90 (m, 4H, 2ÂCH2), 5.68 (1H, q,
J=7.5 Hz, CHCH3), 7.13 (1H, br s, NH, exch D2O),
7.44–7.92(m, 8H, Ar–H), 8.82 (1H, br s, NH, exch
D2O), 13.36 (1H, br s, NH, exch D2O); 13C NMR d 12.5
(CH3), 19.8 (CH3), 24.0 (CH2), 54.3 (CH2), 54.8 (CH2),
63.7 (CH), 106.0 (C), 127.8 (CH), 128.8 (CH), 129.4
(CH), 129.5 (CH), 132.0 (C), 132.5 (CH), 134.0 (CH),
134.9 (C), 136.5 (C), 154.5 (C), 155.4 (C), 163.8 (C);
FAB+-HRMS calcd for C22H25N7Cl: 422.185997,
424.183047 [M–H]+; found: 422.186106, 424.184254
[M–H]+.
2
2
144.2(CH), 117.0 (C), 118.7 (CH), 126.8 (2 ÂCH), 128.5
(CH), 129.5 (CH), 135.7 (C), 39.7 (C), 144.2 (C), 162.0
(C), 166.5 (C); ES+-MS m/z 354, 356 [M+H]+; EI+-
HRMS calcd for C19H21N5Cl: 353.14072; found:
353.14047 [M+H]+.
2,4-Diamino-5-(3-benzoylamino-4-chlorophenyl)-6-ethyl-
pyrimidine (23a). A mixture of aminopyrimethamine
(22)11 (2.64 g, 10 mmol), benzoyl chloride (2 mL, 17
mmol), and pyridine (15 mL) was heated at 95 ꢀC for 3
h and then cooled to room temperature. The reaction
mixture was quenched with ice-ammonia and the resul-
tant precipitate was collected. The recovered solid was
washed successively with hexane and water and crystal-
lised from ethanol gave the benzoylamino-pyrimidine
(ii) The benzylamine (24a) could also be prepared (59%)
by treating aminopyrimethamine (0.66 g, 2.5 mmol)
with benzyl chloride (0.86 mL, 3 mol equiv) in ethanol
(4 mL) for 24 h.

3010
D. C. M. Chan et al. / Bioorg. Med. Chem. 10 (2002) 3001–3010
2,4-Diamino-5-{4-chloro-3-[N-(3,4,5-trimethoxybenzyl)-
amino]phenyl}-6-ethylpyrim-idine (24b). Similarly pre-
pared (above) from 22 (2.64 g, 10 mmol) and 3,4,5-
trimethoxybenzyl bromide (3.2g, 20 mmol) in refluxing
isopropanol (60 mL) (5 days). Flash column chroma-
tography (dichloromethane/methanol 10:1) followed by
crystallisation from acetone gave the pyrimidine as a
white solid (12%); mp 93–94 ꢀC; Rf (chloroform/
methanol 5:1) 0.49; IR (KBr) ꢀmax/cmÀ1 3395, 3160
60.0 (CH3), 105.3 (CH), 105.7 (C), 124.5 (CH), 125.9
(CH), 126.7 (C), 130.8 (CH), 133.8 (C), 135.9 (C), 136.5
(C), 149.2 (C), 152.9 (C), 162.0 (C), 162.2 (C), 166.7 (C);
ES+-MS m/z 458, 460 [M+H]+. Anal. calcd for
C23H28N5O3Cl (CH3)2CO: C, 60.52; H, 6.64; N, 13.57;
found: C, 60.11; H, 6.42; N, 13.81.
References and Notes
1
(NH), 2936, 1593, 1460, 1414, 1120, 810; H NMR d
0.73 (3H, t, J=7.5 Hz, CH3), 1.90 (2H, q, J=7.5 Hz,
CH2), 3.59 (3H, s, OCH3), 3.68 (6H, s, 2ÂOCH3), 4.26–
4.37 (2H, m, CH2Ph), 5.49 (2H, br s, NH2), 5.82(2H, br
s, NH2), 6.11 (1H, t, J=6.3 Hz, NH), 6.32–6.38 (2H, m,
Ar–H), 6.62 (2H, s, Ar–H), 7.28 (1H, d, J=7.8 Hz, Ar–
H); 13C NMR d 13.1 (CH3), 27.4 (CH2), 46.4 (CH2),
55.9 (CH3), 60.0 (CH3), 104.0 (CH), 106.5 (C), 114.4
(CH), 117.0 (C), 118.8 (CH), 129.5 (CH), 135.4 (C),
135.7 (C), 136.3 (C), 144.3 (C), 153.1 (C), 162.0 (C),
162.1 (C), 166.3 (C); ES+-MS m/z 444, 446 [M+H]+.
1. Queener, S. F. J. Med. Chem. 1995, 38, 4739.
2. Stevens, M. F. G.; Phillip, K. S.; Rathbone, D. L.; O’Shea,
D. M.; Queener, S. F.; Schwalbe, C. H.; Lambert, P. A.
J. Med. Chem. 1997, 40, 1886.
3. Chan, D. C. M.; Laughton, C. A.; Queener, S. F.; Stevens,
M. F. G. J. Med. Chem. 2001, 44, 2555.
4. Cody, V.; Chan, D.; Galitsky, N.; Rak, D.; Luft, J. R.;
Pangborn, W.; Queener, S. F.; Laughton, C. A.; Stevens,
M. F. G. Biochemistry 2000, 39, 3556.
5. Vaughan, K.; Stevens, M. F. G. Chem. Soc. Rev. 1978, 7,
377.
6. Gescher, A.; Threadgill, M. D. Pharma. Ther. 1987, 32, 191.
7. Connors, T. A.; Goddard, P. M.; Merai, K.; Ross, W. C. J.;
Wilman, D. E. V. Biochem. Pharmacol. 1976, 25, 241.
8. Jones, C. C.; Kelly, M. A.; Sinnott, M. L.; Smith, P. J.;
Tzotzos, G. T. J. Chem. Soc., Perkin Trans. 2 1982, 1655.
9. Clark, A. S.; Deans, B.; Stevens, M. F. G.; Tisdale, M. J.;
Wheelhouse, R. T.; Denny, B. J.; Hartley, J. A. J. Med. Chem.
1995, 38, 1493.
10. Champness, J. N.; Achari, A.; Ballantine, S. P.; Bryant,
P. K.; Delves, C. J.; Stammers, D. K. Structure 1994, 2, 915.
11. Bliss, E. A.; Griffin, R. J.; Stevens, M. F. G. J. Chem.
Soc., Perkin Trans. 1 1987, 2217.
12. Padwa, A.; Wannamaker, M. W. Tetrahedron 1990, 46, 1145.
13. Hlasta, D. J.; Ackerman, J. H. J. Org. Chem. 1994, 59,
6184.
14. Hagan, D. J.; Chan, D.; Schwalbe, C. H.; Stevens, M. F. G.
J. Chem. Soc., Perkin Trans. 1 1998, 915.
.
.
Anal. calcd for C22H26N5O3Cl 0.1(CH2)3CO 0.4H2O: C,
58.62; H, 6.04; N, 15.32; found: C, 58.94; H, 5.80; N, 15.40.
2,4-Diamino-5-{3-[N-(benzyl)-N-methylamino]-4-chloro-
phenyl}-6-ethylpyrimidine (24c). To benzylamino-pyri-
methamine (24a) (500 mg, 1.4 mmol) suspended in
acetonitrile (100 mL) was added formaldehyde (47 mg,
1.54 mmol, 1.1 mol equiv) with constant stirring.
Sodium cyanoborohydride (150 mg, 2.4 mmol) was
added followed by the dropwise addition of con-
centrated hydrochloric acid until the pH of the reaction
mixture reached 2–3. After 30 min solvents were
removed by vacuum evaporation. The residue was basi-
fied with aqueous ammonia solution, collected and
crystallised from ethanol to yield 24c (90%); mp 121–
122 ꢀC; IR (KBr) ꢀmax/cmÀ1 3453, 3310, 3152(NH),
15. Sakamoto, T.; Uchiyama, D.; Kondo, Y.; Yamanaka, H.
Heterocycles 1993, 35, 1273.
1
1624, 1557, 1435, 1107, 698; H NMR d 0.90 (3H, t,
J=7.5 Hz, CH2CH3), 2.02 (2H, q, J=7.5 Hz,
CH2CH3), 2.63 (3H, s, NCH3), 4.22 (2H, m, CH2Ph),
5.73 (2H, br s, NH2), 6.04 (2H, br s, NH2), 6.81 (2H, m,
Ar–H), 7.20–7.48 (6H, m, Ar–H); 13C NMR d 13.2
(CH3), 27.2 (CH2), 40.1 (CH3), 58.8 (CH2), 105.9 (CH),
124.5 (CH), 125.9 (CH), 126.8 (C), 127.2 (CH), 128.4
(2ÂCH), 131.0 (CH), 135.4 (C), 138.0 (C), 149.3 (C),
161.5 (C), 162.3 (C), 165.5 (C); ES+-MS m/z 368, 370
[M+H]+. Anal. calcd for C20H22N5OCl 0.3 EtOH
0.2H2O: C, 64.06; H, 6.27; N, 18.31; found: C, 64.00; H,
6.12; N, 18.18%.
16. West, R.; Kraianzel, C. S. Inorg. Chem. 1962, 1, 967.
17. le Quan, M.; Cadiot, P. Bull. Soc. Chim. Fr. 1965, 35.
18. Zavgorodnii, V. S.; Sharanina, L. G.; Petrov, A. A. J. Gen.
Chem. USSR 1967, 1469.
19. Kornev, K. A.; Khomenkova, K. K. Ukrain. Kim. Zhur.
1959, 25, 484.
20. Papanastassiou, Z. B.; Bruni, R. J.; White, E. V.; Levins,
P. L. J. Med. Chem. 1966, 9, 725.
21. Shealy, Y. F.; Krauth, C. A. Nature 1966, 210, 208.
22. Shealy, Y. F.; Krauth, C. A.; Holum, L. B.; Fitzgibbon,
W. E. J. Pharm. Sci. 1968, 57, 83.
23. Shealy, Y. F. J. Pharm. Sci. 1970, 59, 1533.
24. Hansen, H.; Hunig, S.; Kishi, K.-I. Chem. Ber. 1979, 112,
¨
2,4-Diamino-5-{4-chloro-3-[N-(3,4,5-trimethoxybenzyl)-
N-methylamino]phenyl}-6-ethylpyrimidine (24d). Simi-
larly prepared (above) from 24b, formaldehyde, sodium
cyanoborohydride in acetonitrile. The product was
crystallised from acetone to give 24d (93%); mp 161–
162 ꢀC; IR (KBr) ꢀmax/cmÀ1 3455, 3169 (NH), 1632,
445.
25. Broughton, M. C.; Queener, S. F. Antimicrob. Agents
Chemother. 1991, 35, 1348.
26. Chio, L.-C.; Queener, S. F. Antimicrob. Agents Che-
mother. 1993, 37, 1914.
27. Chio, L.-C.; Boylard, L. A.; Nasr, M.; Queener, S. F.
Antimicrob. Agents Chemother. 1996, 40, 727.
28. Insight II; Biosym: 9685 Scranton Road, San Diego, CA
9221, USA, 1995.
29. Cornell, W. D.; Cipplak, P.; Bayly, C. I.; Gould, I. R.;
Merz, K. M.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.;
Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995, 117,
5179.
1
1553, 1447, 1362, 1233, 1125; H NMR d 0.87 (3H, t,
J=7.5 Hz, CH2CH3), 1.98 (2H, q, J=7.5 Hz,
CH2CH3), 2.67 (3H, s, NCH3), 3.61 (3H, s, OCH3), 3.70
(6H, s, 2ÂOCH3), 4.17 (2H, d, CH2Ph), 5.54 (2H, br s,
NH2), 5.87 (2H, br s, NH2), 6.64 (2H, s, Ar–H), 6.82
(1H, dd, J 2.0 6.1 Hz, Ar–H), 6.87 (1H, d, J=1.9 Hz,
Ar–H), 7.46 (1H, d, J=4.0 Hz, Ar–H); 13C NMR d 13.1
(CH3), 27.5 (CH2), 30.8 (CH3), 55.8 (CH2), 58.7 (CH3),
30. Croughton, K. A.; Matthews, C. S.; Griffin, R. J.; Stevens,
M. F. G. Anti-Cancer Drug Des. 2002, 16, 119.