Novel non-classical C9-methyl-5-substituted-2,4-diaminopyrrolo[2,3-d]pyrimidines as potential inhibitors of dihydrofolate reductase and as anti-opportunistic agents

Article abstract of DOI:10.1016/j.bmc.2006.09.008

Six novel C9-methyl-5-substituted-2,4-diaminopyrrolo[2,3-d]pyrimidines 18-23 were synthesized as potential inhibitors of dihydrofolate reductase (DHFR) and as anti-opportunistic agents. These compounds represent the only examples of 9-methyl substitution in the carbon-carbon bridge of 2,4-diaminopyrrolo[2,3-d]pyrimidines. The analogs 18-23 were synthesized in a concise eight-step procedure starting from the appropriate commercially available aromatic methyl ketones. The key step involved a Michael addition reaction of 2,4,6-triaminopyrimidine to the appropriate 1-nitroalkene, followed by ring closure of the nitro adducts via a Nef reaction. The compounds were evaluated as inhibitors of DHFR from Pneumocystis carinii (pc), Toxoplasma gondii (tg), Mycobacterium avium (ma) and rat liver (rl). The biological result indicated that some of these analogs are potent inhibitors of DHFR and some have selectivity for pathogen DHFR. Compound 23 was a two digit nanomolar inhibitor of tgDHFR with 9.6-fold selectivity for tgDHFR.

Full text of DOI:10.1016/j.bmc.2006.09.008

Bioorganic & Medicinal Chemistry 14 (2006) 8341–8351
Novel non-classical C9-methyl-5-substituted-2,4-diaminopyrrolo-
[2,3-d]pyrimidines as potential inhibitors of dihydrofolate
reductase and as anti-opportunistic agentsq
Aleem Gangjee,a,* Jie Yanga and Sherry F. Queenerb
aDivision of Medicinal Chemistry, Graduate School of Pharmaceutical Sciences, 600 Forbes Avenue,
Duquesne University, Pittsburgh, PA 15282, USA
bDepartment of Pharmacology and Toxicology, School of Medicine, Indiana University, Indianapolis, IN 46202, USA
Received 13 July 2006; revised 28 August 2006; accepted 7 September 2006
Available online 28 September 2006
Abstract—Six novel C9-methyl-5-substituted-2,4-diaminopyrrolo[2,3-d]pyrimidines 18–23 were synthesized as potential inhibitors of
dihydrofolate reductase (DHFR) and as anti-opportunistic agents. These compounds represent the only examples of 9-methyl sub-
stitution in the carbon–carbon bridge of 2,4-diaminopyrrolo[2,3-d]pyrimidines. The analogs 18–23 were synthesized in a concise
eight-step procedure starting from the appropriate commercially available aromatic methyl ketones. The key step involved a
Michael addition reaction of 2,4,6-triaminopyrimidine to the appropriate 1-nitroalkene, followed by ring closure of the nitro
adducts via a Nef reaction. The compounds were evaluated as inhibitors of DHFR from Pneumocystis carinii (pc), Toxoplasma gon-
dii (tg), Mycobacterium avium (ma) and rat liver (rl). The biological result indicated that some of these analogs are potent inhibitors
of DHFR and some have selectivity for pathogen DHFR. Compound 23 was a two digit nanomolar inhibitor of tgDHFR with 9.6-
fold selectivity for tgDHFR.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
and piretrexim (PTX, 4) (Fig. 1) are potent inhibitors
of both pathogenic and human DHFR which results
in high toxicity to host cells. Hence 3 and 4 are used
in combination with leucovorin (N5-formyltetrahydro-
folate) which is necessary to rescue the host cells.4 Thus
the development of agents that display both high poten-
cy and selectivity for pathogenic DHFR remains a desir-
able goal for the treatment of opportunistic infections in
AIDS patients.
Opportunistic infections caused by pathogenic organ-
isms such as Pneumocystis, Toxoplasma gondii (tg),
and Mycobacterium avium (ma) are the principal cause
of morbidity and mortality in AIDS patients.2 Inhibitors
of dihydrofolate reductase (DHFR) from Pneumocystis
and Toxoplasma are useful anti-opportunistic agents
but DHFR inhibitors have yet to be clinically applied
for M. avium infections. DHFR inhibitors that display
potent activity and high selectivity for these pathogenic
organisms versus human DHFR remain elusive for the
treatment of opportunistic infection in AIDS patients.
Infections caused by Pneumocystis or Toxoplasma are
currently treated with weak but selective DHFR inhibi-
tors such as trimethoprim (TMP 1) and pyrimethamine
(PYR 2) (Fig. 1), and their action is augmented with sul-
fonamides. The combination therapy however often re-
sults in severe side effects to sulfa drugs and leads to
discontinuation of treatment.3 Trimetrexate (TMQ, 3)
Gangjee et al.5–10 reported two-atom bridged classical
and non-classical, 6–5 bicyclic analogs, as potential
DHFR inhibitors (Fig. 2). The structure–activity rela-
tionship (SAR) studies revealed that a N9- or C9-methyl
group in the two-atom bridged side chain of the 6–5 sys-
tem significantly contributes to the DHFR inhibitory
activity and selectivity. Thus, in the furo[2,3-d]pyrimi-
dine analogs (5–10) (Fig. 2), the N9- or C9-methyl com-
pounds were more potent than the corresponding N9- or
C9-desmethyl compounds. Molecular modeling using
Sybyl 6.710 suggested that the 9-methyl moiety in
furo[2,3-d]pyrimidines interacts with Val115 in human
DHFR which may account for their increased potency
against human DHFR. In addition, molecular modeling
also indicated that the 9-methyl group restricts the rota-
q
See Ref. 1.
Corresponding author. Tel.: +1 412 396 6070; fax: +1 412 396
5593; e-mail: gangjee@duq.edu
*
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.09.008

8342
A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
NH2
NH2
OMe
Cl
N
N
H2N
N
OMe
H2N
N
OMe
2 Pyrimethamine
1 Trimethoprim (TMP)
OMe
NH2 CH3
OMe
OMe
OMe
OMe
NH2 CH3
N
N
N
H
H2N
N
N
H2N
N
3 Trimetrexate (TMQ)
Figure 1. Known inhibitors of pathogenic DHFR.
4 Piritrexim (PTX)
8
8
9
9
NH2
N
NH2
CH3
Ar
Y
8 9
6
10
NH2
N
5
5
Ar
Ar
5
N
N
6
6
R
N
X
N
H2N
H2N
N
7
7
H
N
H2N
7
H
5, X = O, Y = NH, Ar = p-benzoyl-L-glutamate
6, X = O, Y = NCH3, Ar = p-benzoyl-L-glutamate
7, X = O, Y = CH2, Ar = p-benzoyl-L-glutamate
8, X = O, Y = CHCH3, Ar = p-benzoyl-L-glutamate
9, X = O, Y = NH, Ar = 2-Naphthyl
Ar = p-benzoyl-L-glutamate
15, R = H,
16, R = CH3,
Ar
18 Ph
19 2-Naphthyl
20 2-Fluorenyl
21 4-biPh
22 2,5-diOMePh
23 3,4,5-triOMePh
NH2
10, X = O, Y = NCH3, Ar = 2-Naphthyl
OMe
OMe
N
11, X = NH, Y = NH, Ar = p-benzoyl-L-glutamate
12, X = NH, Y = NCH3, Ar = p-benzoyl-L-glutamate
13, X = NH, Y = NH, Ar = 1-Naphthyl
N
H2N
N
H
MeO
14, X = NH, Y = NCH3, Ar = 1-Naphthyl
17
Figure 2. Two atom bridged furo- and pyrrolo[2,3-d] pyrimidines.
tion of the C8–C9 bond (compounds 7 and 8) and leads
to a decreased number of accessible low energy confor-
mations conducive for binding to human DHFR.10
Thus the classical 9-methyl substituted compound 8
was twice as potent against human DHFR and dis-
played 10-fold greater activity in cell culture as com-
pared to the corresponding desmethyl analog 7.10
contribute to the DHFR inhibitory activity.11 Miwa
et al.12 reported the classical pyrrolo[2,3-d]pyrimidine
anti-folates 15 and 16 as DHFR inhibitors. These
compounds possess a three-carbon atom bridge. The
C10-methylated compound 16 was similar in activity
to the desmethylated compound 15. Rosowsky
et al.13 reported the two-carbon atom bridged non-
classical pyrrolo[2,3-d]pyrimidine analog 17 as a po-
tent inhibitor of DHFR. It was therefore of interest
to introduce a methyl group at the C9-position of
the bridge in pyrrolo[2,3-d]pyrimidines to determine
the effect on inhibitory activity and selectivity against
pathogenic DHFR. Thus compounds 18–23 (Fig. 2),
were synthesized as potential inhibitors of DHFR
and as potential anti-opportunistic agents. X-ray crys-
tal structure of similar furo[2,3-d]pyrimidine analogs
with pcDHFR indicated that the side chain phenyl
substituent has a hydrophobic interaction with Phe69
in pcDHFR;9 similar interactions were anticipated
with tgDHFR (Phe91) and maDHFR (Val58). These
hydrophobic interactions would be absent with mam-
malian DHFR where an Asn64 replaces a Phe or
Val. Thus large hydrophobic aromatic rings, such as
naphthyl, fluorenyl, biphenyl and substituted phenyls
were also chosen for compounds 19–21 to enhance
the hydrophobic interactions with the pathogenic
DHFR.
In the two-carbon atom bridged pyrrolo[2,3-d]pyrimi-
dine series 11–14 (Fig. 2), N9-methylation of the classi-
cal compound 11 maintained the potency but increased
the selectivity for tgDHFR. Similarly, the non-classical
analog 14 was more potent than the corresponding
desmethyl compound 13.5,6
Classical anti-folates such as 5–8, 11, 12, 15 and 16
(Fig. 2) have a polar L-glutamate side chain, and hence
utilize carrier-mediated active transport mechanisms for
uptake into cells. However, those pathogenic organisms
that lack active transport mechanisms are not suscepti-
ble to classical anti-folates. Non-classical anti-folates
without the polar L-glutamate side chain such as
TMP, TMQ and PTX, are lipophilic and have been used
to target these pathogenic microorganisms.
In the two-atom bridged 6–6 bicyclic system, a methyl
group in the side chain of the bridge was reported to

A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
8343
2. Results and discussion
Our alternate method was to generate the a,b-unsaturat-
ed aldehyde 30 via a Wittig reaction from 27, followed
by reduction of the double bond to afford the desired
aldehydes (Scheme 3). Unfortunately the initial Wittig
reaction was not successful. At this stage it was decided
to synthesize the esters 32–36,18 which could be reduced
by LiAlH4 to the primary alcohols,19 followed by oxida-
tion with PCC to aldehydes.20 The total yield for this
two step procedure was reasonably good (40%) for 46,
but purification of the compounds remained a problem,
and the use of impure aldehyde for the subsequent step
was not feasible. In view of the fact that the ester could
be reduced to the corresponding aldehyde with DIBAL-
H, an alterate synthetic approach was adopted in which
the double bond of the a,b-unsaturated esters 32–36
were reduced by catalytic hydrogenation,21 followed
by reduction with DIBAL-H to afford the aldehydes
42–46 in very high yield (two steps >80%, Scheme 3).22
Gangjee et al. reported7–10 the synthesis of two-carbon
atom bridged furo[2,3-d]pyrimidines 5–10 by utilizing a
Wittig reaction. It was anticipated that a similar Wittig
reaction (Scheme 1) of compound I and substituted (1-
chloro-ethyl)-benzene should afford intermediate III,
which on hydrogenation would give the desired target
compounds. However the Wittig reaction of I and (1-
chloro-ethyl)-benzene under a variety of different condi-
tions of time, base, temperature and solvent failed to
afford the desired product III. A second synthetic strat-
egy was devised in which compound I was to be convert-
ed to II by first reduction to the 5-hydroxymethyl analog
followed by bromination. This was to be followed by a
Wittig reaction with appropriately substituted ketones.
However the reduction of I to the corresponding 5-hy-
droxymethyl analog could not be accomplished despite
several attempts with a variety of reducing agents. Thus
the proposed synthetic methodology in Scheme 1 was
unsuccessful.
The Henry reaction of aldehydes 42–47 with nitrometh-
ane gave the expected nitro alcohol intermediates 48–53
(Scheme 4). This reaction was carried out under a vari-
ety of different reaction conditions using different bases
(K2CO3, NaOEt, NaOMe, NaOH and KF), different
solvents (MeOH, EtOH, THF, CHCl3 and i-PrOH),
and different temperatures (from room temperature to
heating to reflux).23–26 The optimized conditions in our
hands involved using i-PrOH as the solvent, KF as a
weak base and stirring at room temperature overnight
to afford compounds 48–53 in >85% yield.27 Under
these conditions the product was readily purified via col-
umn chromatography.
The failure of the Wittig reaction, as well as the conver-
sion of I to its 5-hydroxymethyl analog, prompted the
search for an alternate synthetic procedure. Taylor and
Liu14,15 had reported the synthesis of 2,4-diamino-pyr-
rolo[2,3-d]pyrimidines via a Nef reaction of a nitro inter-
mediate. The key nitro intermediates were in turn
obtained by a Michael addition of 5-unsubstituted-
2,4,6-triaminopyrimidine and the appropriate 1-nitro-
alkenes. The 1-nitroalkenes were synthesized from
appropriately substituted aldehydes and nitromethane
by a Henry reaction (nitroaldol condensation).16 For
the synthesis of target compounds 18–23 the appropri-
ately substituted aldehydes were not commercially avail-
able. The synthesis of these aldehydes was first
attempted using a Heck coupling (Scheme 2),17 however
low yields (<10%) along with problems in purification of
the desired product prompted the search for alternate
synthetic strategies.
The formation of 1-nitroalkenes 54–59 was carried out
under mild reaction conditions with MsCl and Et3N at
0 ꢁC and stirring for one half hour to afford 54–59 in
>90% yield.28,29 The Michael addition of 2,4,6-triamino-
pyrimidine to the 1-nitroalkenes 54–59 afforded the ni-
tro adducts 60–65.30 The transformation of the nitro
adduct to the carbonyl intermediate was the key step
NH2
CHO
Ph
Cl
CH3
NH2
N
N
Ph
Wittig reaction
H2N
N
N
H
target compound
[H]
I
NH2
N
N
III
Br
H2N
Ph
H
O
N
Wittig reaction
N
H2N
N
H
II
Scheme 1. Attempted synthesis of target compounds.
CH3
CHO
I
CH3
Pd(OAc)2
+
OH
CH3
CH3
24
25
26
Scheme 2. Heck coupling for the synthesis of 26.

8344
A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
CH3
CH3
[H2]
Ph3PCHCHO
CHO
CHO
Ar
Ar
28
30
CH3
OH
CH3
(EtO)2P(O)CH2COOEt
CH3
LiAlH4
29
COOEt
Ar
NaH/THF, 0 oC - r.t
2 days, 75-85%
Ar
Ar
O
31
Ar
27
R=3,4,5-triOMephenyl
32 2-Naphthyl
33 2-Fluorenyl
34 4-biPh
H2
35 2,5-diOMePh
36 3,4,5-triOMePh
PCC
5% Pd/C
Ar
Ar
37 2-Naphthyl
38 2-Fluorenyl
39 4-biPh
40 2,5-diOMePh
41 3,4,5-triOMePh
42 2-Naphthyl
43 2-Fluorenyl
44 4-biPh
DIBAL-H
two steps:
80-85%
CH3
CH3
45 2,5-diOMePh
COOEt
CHO
46 3,4,5-triOMePh
Ar
Ar
Scheme 3. Synthesis of aldehydes 42–46.
CH3NO2
iPrOH/KF
MsCl/Et3N
CH2Cl2
0 oC, 30 min
90-92%
CH3
CH3 OH
CHO
NO2
Ar
Ar
47 Ph
Ar
r.t. overnight
85-90%
Ar
48 Ph
42 2-Naphthyl
43 2-Fluorenyl
44 4-biPhl
49 2-naphthyl
50 2-Fluorenyl
51 4-biPh
45 2,5-diOMePh
52 2,5-diOMePh
53 3,4,5-triOMePh
46 3,4,5-triOMePh
NH2
N
CH3
N
NH2
N
1. NaOH,
2hr, r.t.
CH3
Ar
H2N
NH2
18-23
NO2
2. H2SO4,
1hr, r.t.
Ar
EtOAc/H2O
50-55 oC,
16 h, 80-84 %
NO2
H2N
NH2
Ar
60 Ph
N
20-35%
Ar
54 Ph
55 2-Naphthyl
56 2-Fluorenyl
57 4-biPh
58 2,5-diOMePh
59 3,4,5-triOMePh
61 2-Naphthyl
62 2-Fluorenyl
63 4-biPh
64 2,5-diOMePh
65 3,4,5-triOMePh
Scheme 4. Synthesis of 18–23.
in forming the desired pyrrolo[2,3-d]pyrimidines. There
are numerous Nef reaction conditions available for con-
verting nitro adducts to carbonyl intermediates.31 How-
ever the one-pot conditions reported by Taylor and
Liu.14,15 were particularly appealing. Thus the nitro ad-
ducts were first treated with aqueous NaOH solution
followed by aqueous H2SO4 solution to afford the final
products 18–23 (Scheme 4). The compounds were evalu-
ated as inhibitors of pcDHFR, tgDHFR and maDHFR.
Rat liver (rl) DHFR served as the mammalian standard.
(23) did not improve activity compared to the desmethyl
analog 17. In fact, the addition of a 9-CH3 to 17 caused
a 3-fold decrease in activity. This result was unexpected
since the 9-CH3 was expected to interact hydropho-
bically or via van der Walls interaction with the close
Ile123 of pcDHFR.32 Perhaps the decrease in activity
is due to an inability to accommodate both a bulky sub-
stitution at the C5-position along with a C9-CH3.
Toxoplasma gondii DHFR (Table 1): the compounds
were significantly more potent and selective against
tgDHFR compared to pcDHFR. The most potent com-
pound was again the 2-naphthyl analog 19 which had an
IC50 value of 15 nM. Compounds 20, 21 and 23 were al-
most as active, with IC50 values between 37 and 72 nM,
indicating that bulk on the phenyl ring was highly con-
ducive to tgDHFR inhibition. The unsubstituted phenyl
analog 18 and the 2,5-dimethyl analog 22 were the least
active compounds. Once again 9-CH3 substitution as
in 23 did not significantly increase potency against
Pneumocystis carinii DHFR (Table 1): against pcDHFR
the most potent analog bore a 2-naphthyl group (19).
Increasing the bulk of the naphthyl to a fluorenyl group
(20) or a biphenyl (21) caused a drop in activity as did
the 3,4,5-triOMe substitution (23). However, the most
detrimental substitution was the 2,5-diOMe (22), which
was about 10-fold less potent than 19. The unsubstituted
phenyl analog 18 was the least active. The compounds
were not selective against pcDHFR and C9-methylation

A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
8345
Table 1. Inhibitory concentration (IC50 lM) and selectivity ratios against rlDHFR, pcDHFR, tgDHFR and maDHFR versus rlDHFR by
compounds 18–23a
Compound
pc
0.77
17.1
0.859
rl
rl/pc
tg
rl/tg
ma
rl/ma
17b
18
0.20
0.26
0.04
0.05
0.03
0.07
0.06
0.2
0.037
0.138
0.0148
0.0483
0.0717
0.16
5.4
5
0.077
5.34
2.6
0.13
0.696
0.0399
0.0567
0.142
0.516
0.396
19
20
2.7
1.2
2
0.109
0.246
0.0747
1.86
0.37
0.23
1.9
2.09
1.99
8.06
2.22
21
22
3.2
9.6
65
0.28
1.23
23
0.0414
2.8
0.321
0.30
TMP
PTX
TMQ
12
180.0
0.0033
0.003
14
600
0.013
0.042
0.26
0.07
0.0043
0.01
0.76
0.3
0.00061
0.0015
5.3
2.0
a All enzymes were assayed spectroscopically in a solution containing 90 micromolar dihydrofolate, 119 lM NADPH, 41 mM Na phosphate buffer,
and 8.9 mM 2-mercaptoethanol at pH 7.4 and 37 C. Rat liver and Toxoplasma DHFR are assayed in the presence of 150 mM KCl. Reactions are
initiated with an amount of enzyme yielding a change in OD at 340 nM of 0.035/min.
b The data for 17 was taken from Ref. 13.
tgDHFR compared to the desmethyl analog 17. Selec-
tivity, though improved from that toward pcDHFR,
was not significant, with the best compound 23 being
9-fold more selective compared to rlDHFR. This was
not very different from the desmethyl analog.
pared separately with the mammalian standard. In this
limited series N9-methylation and C5-bulky substitu-
tions does not afford both potent and selective inhibitors
for any of the pathogenic DHFR evaluated. However,
the 2 digit nanomolar potency of compounds 19 and
20 against tgDHFR and 21 against tgDHFR and maD-
HFR, indicates that potency can be maintained or im-
proved with some bulky substitutions at C5 compared
to the phenyl analog 18.
Mycobacterium avium DHFR (Table 1): against maD-
HFR the most potent analog was 23. All analogs were
more potent against the bacterial maDHFR than
against pcDHFR. Comparing 17 with 23 indicates that
9-CH3 substitution has a detrimental effect on maD-
HFR inhibition. There was no selectivity for maDHFR
for this series of analogs. The most potent analogs were
17 and 21 indicating that neither N9-methylation nor
bulk at C5 contributes significantly to activity against
maDHFR.
3. Experimental
Melting points were determined on a Mel-Temp II melt-
ing point apparatus with FLUKE 51 K/J electronic ther-
mometer and were uncorrected. Nuclear magnetic
resonance spectra for proton (1H) were recorded on a
Bruker WH-300 (300 MHz) spectrometer. The chemical
shift values were expressed in ppm (parts per million)
relative to tetramethylsilane as internal standard;
s = singlet, d = double, t = triplet, q = quartet, m = mul-
tiplet, br s = broad singlet. The relative integrals of peak
areas agreed with those expected for the assigned struc-
tures. High-resolution mass spectra (HRMS), using elec-
tron impact (EI), were recorded on a VG AUTOSPEC
(Fisons Instruments) micromass (EBE Geometry) dou-
ble focusing mass spectrometer. Thin-layer chromatog-
raphy (TLC) was performed on POLYGRAM Sil G/
UV254 silica gel plates with fluorescent indicator, and
the spots were visualized under 254 and 366 nm illumi-
nation. Proportions of solvents used for TLC were by
volume. Column chromatography was performed on
230–400 mesh silica gel purchased from Aldrich Chemi-
cal Co., Milwaukee, WI. All evaporations were carried
out in vacuo with a rotary evaporator. Analytical sam-
ples were dried in vacuo (0.2 mmHg) in an Abderhalden
drying apparatus over P2O5 at 75–110 ꢁC. Elemental
analysis was performed by Altlantic Microlabs, Nor-
cross, GA. Element compositions are within 0.4% of
calculated values. Fractional moles of water or organic
solvents frequently found in some analytical samples
could not be prevented despite 24–48 h of drying in vac-
uo and were confirmed where possible by their presence
in the 1H NMR spectra. All solvents and chemicals were
Rat liver DHFR (Table 1): compounds 19 and 20 were
potent nanomolar inhibitors of mammalian DHFR.
These two analogs have bulky aromatic substitutions
and were not expected to interact with rlDHFR which
contain an Asn in place of Phe69 of pcDHFR. It is pos-
sible that the compounds bind to tgDHFR and to
rlDHFR in different orientations that perhaps allow po-
tent inhibitory activity for both analogs against both en-
zymes. Again there is no advantage of the 9-CH3
substitution over the 9-desmethyl compound.
In summary, the phenyl unsubstituted analog 18 and the
2,5-diOMe substituted analog 22 were the least potent
DHFR inhibitors tested against the four enzymes. The
most potent compound against rat liver DHFR and
tgDHFR was the 2-naphthyl analog 19; compounds 17
and 19 were similar in potency against pcDHFR. The
other analogs did not show any particular trend with re-
spect to all DHFR, however differences in activities of
the compounds against specific DHFR were noted.
Direct comparison of the N9-CH3 analog 23 with the
corresponding N9-desmethyl analog 17 indicates that
the N9-CH3 group is detrimental to activity. Bulky sub-
stitutions on the C5-position did not afford significant
increase in potency (except against maDHFR) or selec-
tivity. Thus the SAR is different against different DHFR
and the SAR of each enzyme target needs to be com-

8346
A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
purchased from Aldrich Chemical Co. and Fisher Scien-
tific and were used as received.
OCH3), 4.18 (q, 2H, CH2), 5.90 (s, 1H, CH), 6.72–6.89
(m, 3H, Ar-H · 3).
3.1. General procedure for the synthesis of E-ethyl 3-
(substituted-phenyl)but-2-enoate (32–36)
3.6. E-Ethyl 3-(3,4,5-trimethoxyphenyl)but-2-enoate (36)
Using the general procedure described above, com-
pound 36 was obtained from 3,4,5-trimethoxyacetoph-
enone as an off-white solid in 63% yield. Mp: 57-59 ꢁC;
(lit.34 55–57 ꢁC); Rf 0.50 (hexane/EtOAc/3:1). 1H
NMR (CDCl3): d 1.30 (t, 3H, CH3), 2.55 (s, 3H,
CH3), 3.86 (s, 3H, OCH3), 3.89 (s, 6H, 2 · OCH3),
4.20 (q, 2H, CH2), 6.09 (s, 1H, CH), 6.75 (s, 2H,
Ar-H · 2).
To a suspension of NaH (290 mg, 11.9 mmol) in 30 mL
of dry THF was added triethyl phosphonoacetate
(2.67 g, 11.9 mmol). The mixture was stirred for 1 h at
room temperature and then the appropriate commer-
cially available arylmethylketone (11.9 mmol) in 30 mL
THF was added dropwise. After stirring at room tem-
perature for 3 days, the reaction mixture was concen-
trated by evaporation of the solvent and then diluted
with 300 mL of water. The resulting aqueous solution
was extracted with Et2O (100 mL · 3) and the organic
layers combined and dried over Na2SO4. The solvent
was removed under reduced pressure to give 2–3 mL
of a liquid residue, which was loaded on a silica gel col-
umn (15 · 150 mm) and eluted with hexane, 5% EtOAc
in hexane and then 10% EtOAc in hexane. The fractions
containing the product (TLC) were pooled and evapo-
rated to afford 32–36. Here only the E-isomer of the
product was isolated and characterized. The Z-isomer
was difficult to isolate from the mixture (E/Z), and
was not characterized.
3.7. General procedure for synthesis of ( )-ethyl 3-
(substituted-phenyl)butanoate (37-41)
Ethyl E-3-(substituted-phenyl)but-2-enoate (32–36)
(11.7 mmol) was dissolved in 150 mL of EtOAc in a
500 mL flask, and 5% Pd/C (2.5 g) was added. The sus-
pension was stirred under 1 atm H2 until the disappear-
ance of the starting material (TLC). The catalyst was
filtered and the solvent evaporated under reduced pres-
sure to afford 37–41, which was used directly for the
next step without further purification.
3.8. ( )-Ethyl 3-(2-naphthyl)butanoate (37)
3.2. E-Ethyl 3-(2-naphthyl)but-2-enoate (32)
Using the general procedure described above, com-
pound 37 was obtained from E-ethyl 3-(2-naph-
thyl)but-2-enoate 32 as a colorless oil in 95% yield. Rf
Using the general procedure described above, com-
pound 32 was obtained from 2-acetonaphthone as a col-
1
orless oil in 75% yield. Rf 0.53 (hexane/EtOAc/3:1). H
1
0.65 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.17 (t,
NMR (CDCl3): d 1.34 (t, 3H, CH3), 2.69 (s, 3H, CH3),
4.23 (q, 2H, CH2), 6.29 (s, 1H, CH), 7.47–7.48 (m, 7H,
Ar-H · 7).
3H, CH3), 1.39 (d, 3H, CH3), 2.65–2.72 (m, 2H, CH2),
3.46–3.48 (m, 1H, CH), 4.07 (q, 2H, CH2), 7.37–7.45
(m, 3H, Ar-H · 3), 7.79–7.81 (m, 4H, Ar-H · 4).
3.3. E-Ethyl 3-(2-fluorenyl)but-2-enoate (33)
3.9. ( )-Ethyl 3-(2-fluorenyl)butanoate (38)
Using the general procedure described above, com-
pound 33 was obtained from 2-acetylfluorene as an
off-yellow solid in 61% yield. Mp: 87–89 ꢁC; Rf 0.49
Using the general procedure described above, com-
pound 38 was obtained from E-ethyl 3-(2-fluore-
nyl)but-2-enoate 33 as a colorless oil in 97% yield. Rf
1
1
(hexane/EtOAc/3:1). H NMR (CDCl3): d 1.33 (t, 3H,
0.71 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.18 (t,
CH3), 2.64 (s, 3H, CH3), 3.91 (s, 2H, CH2), 4.23 (q,
2H, CH2), 6.21 (s, 1H, CH), 7.30–7.81 (m, 7H, Ar-
H · 7).
3H, CH3), 1.34 (d, 3H, CH3), 2.55–2.70 (m, 2H, CH2),
3.32–3.40 (m, 1H, CH), 3.80 (s, 2H, CH2), 4.07 (q, 2H,
CH2), 7.22–7.65 (m, 4H, Ar-H · 4), 7.70–7.76 (m, 3H,
Ar-H · 3).
3.4. E-Ethyl 3-(4-biphenyl)but-2-enoate (34)
3.10. ( )-Ethyl 3-(4-biphenyl)butanoate (39)
Using the general procedure described above, com-
pound 34 was obtained from 4-acetyl-biphenyl as an
off-white solid in 80% yield. Mp: 63–65 ꢁC (lit. 63–
64 ꢁC); (lit.33 63–64ꢁ C); Rf 0.55 (hexane/EtOAc/3:1).
1H NMR (CDCl3): d 1.33 (t, 3H, CH3), 2.64 (s, 3H,
CH3), 4.26 (q, 2H, CH2), 6.21 (s, 1H, CH), 7.34–7.71
(m, 9H, Ar-H · 9).
Using the general procedure described above, com-
pound 39 was obtained from E-ethyl 3-(4-biphe-
nyl)but-2-enoate 34 as a colorless oil in 98% yield. Rf
1
0.70 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.19 (t,
3H, CH3), 1.35 (d, 3H, CH3), 2.65–2.70 (m, 2H, CH2),
3.34–3.41 (m, 1H, CH), 4.10 (q, 2H, CH2), 7.27–7.72
(m, 9H, Ar-H · 9).
3.5. E-Ethyl 3-(2,5-dimethoxyphenyl)but-2-enoate (35)
3.11. ( )-Ethyl 3-(2,5-dimethoxyphenyl)butanoate (40)
Using the general procedure described above, com-
pound 35 was obtained from 2,5-dimethoxyacetophe-
none as a colorless oil in 74% yield. Rf 0.49 (hexane/
EtOAc/3:1). H NMR (CDCl3): d 1.28 (t, 3H, CH3),
2.56 (s, 3H, CH3), 3.76 (s, 3H, OCH3), 3.78 (s, 3H,
Using the general procedure described above, compound
40 was obtained from E-ethyl 3-(2,5-dimethoxyphe-
nyl)but-2-enoate 35 as a colorless oil in 98% yield. Rf
1
1
0.60 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.18 (t,

A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
8347
3H, CH3), 1.25 (d, 3H, CH3), 2.43–2.70 (m, 2H, CH2),
3.57–3.71 (m, 1H, CH), 3.72 (s, 3H, OCH3), 3.75 (s, 3H,
3.17. ( )-3-(2,5-Dimethoxyphenyl)butrylaldehyde (45)
OCH3), 4.08 (q, 2H, CH2), 6.67–6.79 (m, 3H, Ar-H · 3).
Using the general procedure described above, com-
pound 45 was obtained from ( )-ethyl 3-(2,5-dimeth-
oxyphenyl)butanoate 40 as a colorless oil in 85% yield.
3.12. ( )-Ethyl 3-(3,4,5-trimethoxyphenyl)butanoate (41)
1
Rf 0.55 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.29
Using the general procedure described above, com-
pound 41 was obtained from E-ethyl 3-(3,4,5-trimeth-
oxyphenyl)but-2-enoate 36 as a colorless oil in 100%
(d, 3H, CH3), 2.54–2.74 (m, 2H, CH2), 3.60–3.65 (m,
1H, CH), 3.83 (s, 3H, OCH3), 3.86 (s, 3H, OCH3),
6.69–6.80 (m, 3H, Ar-H · 3).
1
yield. Rf 0.60 (hexane/EtOAc/3:1). H NMR (CDCl3):
d 1.18 (t, 3H, CH3), 1.28 (d, 3H, CH3), 2.47–2.63 (m,
2H, CH2), 3.18–3.23 (m, 1H, CH), 3.81 (s, 3H,
OCH3), 3.85 (s, 6H, 2 · OCH3), 4.08 (q, 2H, CH2),
6.43 (s, 2H, Ar-H · 2).
3.18. ( )-3-(3,4,5-Trimethoxyphenyl)butrylaldehyde (46)
Using the general procedure described above, com-
pound 46 was obtained from ( )-ethyl 3-(3,4,5-trimeth-
oxyphenyl)butanoate 41 as a colorless oil in 81% yield.
1
3.13. General procedure for synthesis of ( )-3-(substitut-
ed-phenyl)butylaldehyde (42–46)
Rf 0.38 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.28
(d, 3H, CH3), 2.65–2.72 (m, 2H, CH2), 3.28–3.33 (m,
1H, CH), 3.82 (s, 3H, OCH3), 3.85 (s, 6H, 2 · OCH3),
6.42 (s, 2H, Ar-H · 2), 9.71 (s, 1H, CHO).
To a cooled (À78 ꢁC) solution of ( )-ethyl 3-(substitut-
ed-phenyl)butanoate (37-41) (10 mmol) in 15 mL of dry
CH2Cl2 was added dropwise a solution of DIBAL-H
(10 mL, 10 mmol, 1.0 M in hexane solution). After the
reaction mixture was stirred at À78 ꢁC for 30 min, meth-
anol (6 mL) was added dropwise and then the reaction
mixture was allowed to warm to room temperature.
To this was added a solution of saturated NH4Cl
(10 mL) and the mixture was extracted with CHCl3
(25 mL · 3). The organic layers were combined and
washed with 1 N HCl (5 mL) and brine (5 mL) and dried
(MgSO4). The solvent was removed under reduced pres-
sure to give 2–3 mL of an oil, which was transferred to a
silica gel column (15 · 150 mm), and eluted with 5%
EtOAc in hexane and 10% EtOAc in hexane. Fractions
containing the product (TLC) were pooled and evapo-
rated to afford 42–46.
3.19. General procedure for the synthesis of 1-nitro-4-
(substituted-phenyl)-2-pentanol (48–53)
To a solution of ( )-3-(substituted-phenyl)-butrylalde-
hyde (42–47) (10 mmol) in 10 mL of iso-propanol was
added KF (29 mg, 0.5 mmol) and nitromethane
(2.5 mL, 40 mmol). The reaction mixture was stirred at
room temperature overnight, and the solid KF was fil-
tered and the solvent and excess nitromethane were
evaporated under reduced pressure. The residue was dis-
solved in ethyl acetate (50 mL) and washed with water
(2·10 mL), and dried (MgSO4). After evaporating the
solvent, the residue was dissolved in 2 mL of CHCl3
and was loaded on a 15 · 150 mm silica gel column
and eluted with hexane, hexane/EtOAc/10:1 and hex-
ane/EtOAc/7:1. Fractions containing the product
(TLC) were pooled and evaporated to afford 48–53.
3.14. ( )-3-(2-Naphthyl)butylaldehyde (42)
Using the general procedure described above, com-
pound 42 was obtained from ( )-ethyl 3-(2-naph-
thyl)butanoate 37 as a colorless oil in 85% yield. Rf
0.60 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.42 (d,
3H, CH3), 2.65–2.95 (m, 2H, CH2), 3.45–3.70 (m, 1H,
3.20. 1-Nitro-4-phenyl-2-pentanol (48)
Using the general procedure described above, com-
pound 48 was obtained from ( )-3-phenyl-butrylalde-
hyde 47 as a colorless oil in 70% yield. Rf 0.43
1
1
CH), 7.35–7.82 (m, 7H, Ar-H · 7), 9.75 (s, 1H, CHO).
(CHCl3). H NMR (DMSO-d6): d 1.22 (d, 3H, CH3),
1.50–1.90 (m, 2H, CH2), 2.75–3.10 (m, 1H, CH), 3.65–
4.15 (m, 1H, CH), 4.25–4.80 (m, 2H, CH2), 5.36–5.43
(m, 1H, OH), 7.19–7.32 (m, 5H, Ar-H · 5).
3.15. ( )-3-(2-Fluorenyl)butylaldehyde (43)
Using the general procedure described above, com-
pound 43 was obtained from ( )-ethyl 3-(2-fluore-
nyl)butanoate 38 as a colorless oil in 80% yield. Rf
3.21. 1-Nitro-4-(2-naphthyl)-2-pentanol (49)
1
0.62 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.39 (d,
Using the general procedure described above, com-
pound 49 was obtained from ( )-3-(2-naphthyl)-butryl-
aldehyde 42 as a colorless oil in 80% yield. Rf 0.40
3H, CH3), 2.65–2.95 (m, 2H, CH2), 3.30–3.60 (m, 1H,
CH), 3.88 (s, 2H, CH2), 7.23–7.76 (m, 7H, Ar-H · 7),
9.74 (s, 1H, CHO).
1
(hexane/EtOAc/3:1). H NMR (CDCl3): d 1.38 (d, 3H,
CH3), 1.65–2.20 (m, 2H, CH2), 2.54 (d, 1H, OH),
3.05–3.25 (m, 1H, CH), 3.90–4.15 (m, 1H, CH), 4.20–
4.60 (m, 2H, CH2), 7.34–7.83 (m, 7H, Ar-H · 7).
3.16. ( )-3-(4-Biphenyl)butylaldehyde (44)
Using the general procedure described above, com-
pound 44 was obtained from ( )-ethyl 3-(4-biphe-
nyl)butanoate 39 as a colorless oil in 75% yield. Rf
3.22. 1-Nitro-4-(2-fluorenyl)-2-pentanol (50)
1
0.61 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.35 (d,
3H, CH3), 2.55–2.90 (m, 2H, CH2), 3.37–3.50 (m, 1H,
Using the general procedure described above, com-
pound 50 was obtained from ( )-3-(2-fluorenyl)-butryl-
aldehyde 43 as a colorless oil in 67% yield. Rf 0.45
CH), 7.23–7.58 (m, 9H, Ar-H · 9), 9.23 (s, 1H, CHO).

8348
A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
1
(hexane/EtOAc/3:1). H NMR (CDCl3): d 1.34 (d, 3H,
CH3), 1.60–1.95 (m, 2H, CH2), 2.52 (d, 1H, OH),
3.00–3.30 (m, 1H, CH), 3.89 (s, 2H, CH2), 3.90–4.10
(m, 1H, CH), 4.15–4.50 (m, 2 H, CH2), 7.20–7.95 (m,
7H, Ar-H · 7).
CH), 6.88 (d, 1H, CH), 7.11–7.35 (m, 6H, CH and Ar-
H · 5).
3.28. ( )-1-Nitro-4-(2-naphthyl)-l-pentene (55)
Using the general procedure described above, com-
pound 55 was obtained from ( )-1-nitro-4-(2-naph-
thyl)-2-pentanol 49 as a light yellow oil in 91% yield.
3.23. 1-Nitro-4-(4-biphenyl)-2-pentanol (51)
1
Using the general procedure described above, com-
pound 51 was obtained from ( )-3-(4-biphenyl)-butryl-
aldehyde 44 as a colorless oil in 72% yield. Rf 0.51
(hexane/EtOAc/3:1). 1H NMR (CDCl3): d 1.35 (d,
3H, CH3), 1.60–2.00 (m, 2H, CH2), 2.53 (d, 1H,
OH), 2.95–3.20 (m, 1H, CH), 4.00–4.20 (m, 1H,
CH), 4.20–4.50 (m, 2H, CH2), 7.25–7.74 (m, 9H, Ar-
H · 9).
Rf 0.73 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.41
(d, 3H, CH3), 2.50–2.80 (m, 2H, CH2), 3.00–3.30 (m,
1H, CH), 6.87 (d, 1H, CH), 7.15–7.90 (m, 8H, CH and
Ar-H · 7).
3.29. ( )-1-Nitro-4-(2-fluorenyl)-l-pentene (56)
Using the general procedure described above, com-
pound 56 was obtained from ( )-1-nitro-4-(2-fluore-
nyl)-2-pentanol 50 as a light yellow oil in 90% yield.
3.24. 1-Nitro-4-(2,5-dimethoxyphenyl)-2-pentanol (52)
1
Rf 0.69 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.30
Using the general procedure described above, com-
pound 52 was obtained from ( )-3-(2,5-dimethoxyphe-
nyl)-butrylaldehyde 45 as a colorless oil in 75% yield.
(d, 3H, CH3), 2.33–2.60 (m, 2H, CH2), 2.75–3.00 (m,
1H, CH), 3.79 (s, 2H, CH2), 6.78 (d, 1H, CH), 7.00–
7.90 (m, 8H, CH and Ar-H · 7).
1
Rf 0.37 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.29
(d, 3H, CH3), 1.40–1.90 (m, 2H, CH2), 2.54 (d, 1H,
OH), 3.32–3.51 (m, 1H, CH), 3.78 (s, 3H, OCH3), 3.82
(s, 3H, OCH3), 3.85–4.05 (m, 1H, CH), 4.15–4.40 (m,
2H, CH2), 6.72–6.86 (m, 3H, Ar-H · 3).
3.30. ( )-1-Nitro-4-(4-biphenyl)-l-pentene (57)
Using the general procedure described above, com-
pound 57 was obtained from ( )-1-nitro-4-(4-biphe-
nyl)-2-pentanol 51 as a light yellow oil in 92% yield.
1
3.25. 1-Nitro-4-(3,4,5-trimethoxyphenyl)-2-pentanol (53)
Rf 0.69 (hexane/EtOAc/3:1). H NMR (CDCl3): d 1.38
(d, 3H, CH3), 2.49–2.65 (m, 2H, CH2), 2.98–3.15 (m,
1H, CH), 6.91 (d, 1H, CH), 7.14–7.70 (m, 10H, CH
and Ar-H · 9).
Using the general procedure described above, com-
pound 53 was obtained from ( )-3-(3,4,5-trimethoxy-
phenyl)-butrylaldehyde 46 as a colorless oil in 78%
1
yield. Rf 0.45 (CHCl3). H NMR (CDCl3): d 1.43 (d,
3.31. ( )-1-Nitro-4-(2,5-dimethoxyphenyl)-l-pentene (58)
3H, CH3), 1.60–1.88 (m, 2H, CH2), 2.56 (d, 1H, OH),
2.89–3.02 (m, 1H, CH), 3.83 (s, 3H, OCH3), 3.86 (s,
6H, 2 · OCH3), 3.94–4.04 (m, 1H, CH), 4.29–4.43 (m,
2H, CH2), 6.41 (s, 2H, Ar-H · 2).
Using the general procedure described above, com-
pound 58 was obtained from ( )-1-nitro-4-(2,5-dimeth-
oxyphenyl)-2-pentanol 52 as a light yellow oil in 90%
1
yield. Rf 0.60 (hexane/EtOAc/3:1). H NMR (CDCl3):
3.26. General procedure for the synthesis of ( )-1-nitro-4-
(substituted-phenyl)-1-pentene (54–59)
d 1.26 (d, 3H, CH3), 2.46–2.25 (m, 2H, CH2), 3.30–
3.50 (m, 1H, CH), 3.76 (s, 3H, OCH3), 3.77 (s, 3H,
OCH3), 6.70–6.80 (m, 3H, Ar-H · 3), 6.87 (d, 1H,
CH), 7.16–7.25 (m, 1H, CH).
To a solution of 1-nitro-4-(substituted-phenyl)-2-penta-
nol (48–53) (10 mmol) in 15 mL of dry CH2Cl2 at
0 ꢁC, was added methanesulfonyl chloride (1.18 g,
0.8 mL, 10 mmol), followed by Et3N (2.2 g, 2.8 mL,
20 mmol). The mixture was stirred at 0 ꢁC for 20 min,
and then poured into 30 mL of water and the aqueous
mixture extracted with CH2Cl2 (50 mL · 3). The organic
layers were combined and dried (MgSO4). After evapo-
rating the solvent, the residue was dissolved in 1 mL of
CH2Cl2 and was placed on a 15 · 150 mm silica gel col-
umn and eluted with hexane, hexane/EtOAc/50:1. Frac-
tions containing the product (TLC) were pooled and
evaporated to afford 54–59.
3.32. ( )-1-Nitro-4-(3,4,5-trimethoxyphenyl)-l-pentene
(59)
Using the general procedure described above, com-
pound 59 was obtained from ( )-1-nitro-4-(3,4,5-tri-
methoxyphenyl)-2-pentanol 53 as a light yellow oil in
91% yield. Rf 0.40 (hexane/EtOAc/3:1). 1H NMR
(CDCl3): d 1.32 (d, 3H, CH3), 2.46–2.52 (m, 2H,
CH2), 3.30–3.14 (m, 1H, CH), 3.84 (s, 3H, OCH3),
3.86 (s, 3H, OCH3), 6.41 (s, 2H, Ar-H · 2), 6.91 (d,
1H, CH), 7.10–7.19 (m, 1H, CH).
3.27. ( )-1-Nitro-4-phenyl-l-pentene (54)
3.33. General procedure for the synthesis of 5-[1-
(nitromethyl)-3-substituted arylbutyl]pyrimidine-
2,4,6-triamine (60–65)
Using the general procedure described above, com-
pound 54 was obtained from ( )-1-nitro-4-phenyl-2-
pentanol 48 as a light yellow oil in 90% yield. Rf 0.63
A mixture of ( )-1-nitro-4-(substituted-aryl)-pentene
(54–59) (5 mmol) and (5 mmol) of 2,4,6-triaminopyrim-
idine in 20 mL water and 20 mL ethyl acetate was stirred
1
(hexane/EtOAc/3:1). H NMR (CDCl3): d 1.35 (d, 3H,
CH3), 2.48–2.62 (m, 2H, CH2), 2.91–3.00 (m, 1H,

A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
8349
at 50–55 ꢁC for 24 h. The solid dissolved on heating. The
3.38. 5-[3-(2,5-Dimethoxyphenyl)-1-(nitromethyl)bu-
tyl]pyrimidine-2,4,6-triamine (64)
reaction mixture was poured into 200 mL ethyl acetate,
washed with water (2·40 mL), dried (Na2SO4) and puri-
fied by silica gel column chromatography with 5%
CH3OH in CHCl3. Fractions containing the product
(TLC) were pooled and evaporated to afford 60–65.
Using the general procedure described above, com-
pound 64 was obtained from ( )-1-nitro-4-(2,5-dimeth-
oxyphenyl)-1-pentene 58 as a yellow solid in 80%
yield. Mp: 80–83 ꢁC; Rf 0.41 (CHCl3:CH3OH/5:1). H
NMR (DMSO-d6): d 1.05–1.16 (m, 3H, CH3), 1.65–
2.15 (m, 2H, CH2), 2.85–3.10 (m, 1H, CH), 3.40–3.60
(m, 1 H, CH), 3.66 (s, 3H, OCH3), 3.68 (s, 3H,
OCH3). 4.65–4.90 (m, 2H, CH2), 5.20–5.70 (m, 6H,
3 · NH2), 6.72–6.85 (m, 3H, Ar-H · 3). Anal. Calcd
for C17H24N6O4Æ0.4H2O: C, 53.23; H, 6.52; N, 21.91.
Found: C, 52.89; H, 6.38; N, 21.70.
1
3.34. 5-[(1-Nitromethyl)-3-phenylbutyl]pyrimidine-2,4,6-
triamine (60)
Using the general procedure described above, com-
pound 60 was obtained from ( )-1-nitro-4-phenyl-1-
pentene 54 as a yellow solid in 82% yield. Mp: 70–
73 ꢁC; Rf 0.30 (CHCl3:CH3OH/5:1). 1H NMR
(DMSO-d6): d 1.13–1.20 (m, 3H, CH3), 1.60–2.30 (m,
2H, CH2), 2.50–2.65 (m, 1H, CH), 2.97–3.54 (m, 1H,
CH), 4.65–4.80 (m, 2H, CH2), 5.05–5.35 (m, 4H,
2 · NH2), 5.40–5.65 (m, 2H, NH2), 7.10–7.90 (m, 5H,
Ar-H · 5). Anal. Calcd for C15H20N6O2Æ0.2CH3OH: C,
56.56; H, 6.50; N, 26.04. Found: C, 56.73; H, 6.40; N,
25.87.
3.39. 5-[3-(3,4,5-Trimethoxyphenyl)1-(nitromethyl)bu-
tyl]pyrimidine-2,4,6-triamine (65)
Using the general procedure described above, com-
pound 65 was obtained from ( )-1-nitro-4-(3,4,5-tri-
methoxyphenyl)-1-pentene 59 as a yellow solid in 79%
yield. Mp: 84–87 ꢁC; Rf 0.41 (CHCl3:CH3OH/5:1). H
NMR (DMSO-d6): d 1.19 (d, 3H, CH3), 1.62–2.20 (m,
2H, CH2), 2.85–3.10 (m, 1H, CH), 3.40–3.60 (m, 1H,
CH), 3.74 (s, 3H, OCH3), 3.76 (s, 6H, 2 · OCH3),
4.69–4.78 (m, 2H, CH2), 5.10–5.60 (m, 6H, 3 · NH2),
6.44 (s, 2H, Ar-H · 2). Anal. Calcd for C15H20N6O2Æ0.6-
H2O: C, 51.81; H, 6.57; N, 20.14. Found: C, 52.06; H,
6.43; N, 19.75.
1
3.35. 5-[3-(2-Naphthyl)-1-(nitromethyl)butyl]pyrimidine-
2,4,6-triamine (61)
Using the general procedure described above, com-
pound 61 was obtained from ( )-1-nitro-4-(2-naph-
thyl)-1-pentene 55 as a yellow solid in 81% yield. Mp:
75–77 ꢁC; Rf 0.50 (CHCl3:CH3OH/5:1). 1H NMR
(DMSO-d6): d 1.23–1.30 (m, 3H, CH3), 1.65–2.30 (m,
2H, CH2), 2.60–2.90 (m, 1H, CH), 3.55–3.75 (m, 1H,
CH), 4.65-4.95 (m, 2H, CH2), 5.09–5.53 (m, 6H,
3 · NH2), 7.36–7.83 (m, 7H, Ar-H · 7). Anal. Calcd
for C19H22N6O2: C, 62.28; H, 6.05; N, 22.94. Found:
C, 62.44; H, 6.03; N, 22.60.
3.40. General procedure for the synthesis of ( )-2,4-
diamino-5-[2-methyl-2-(substituted-pheny)]ethyl-pyr-
rolo[2,3-d]pyrimidine (18–23)
To a solution of NaOH (0.24 g, 6.0 mmol) in 3.0 mL
of water was added (1.0 mmol) of ( )-1-nitro-2-
(2,4,6-triaminopyrimidin-5-yl)-4-(substituted-phenyl)-
pentane (60–65) and the mixture was stirred at room
temperature for 2 h. This mixture was added dropwise
3.36. 5-[3-(2-Fluorenyl)-1-(nitromethyl)butyl]pyrimidine-
2,4,6-triamine (62)
Using the general procedure described above, com-
pound 62 was obtained from ( )-1-nitro-4-(2-fluore-
nyl)-1-pentene 56 as a yellow solid in 75% yield. Mp:
78–80 ꢁC; Rf 0.53 (CHCl3:CH3OH/5:1). 1H NMR
(DMSO-d6): d 1.19–1.26 (m, 3H, CH3), 1.15–2.25 (m,
2H, CH2), 2.50–2.80 (m, 1H, CH), 3.90–3.65 (m, 1H,
CH), 3.86 (s, 2H, CH2), 4.65–4.90 (m, 2H, CH2), 5.03–
5.60 (m, 6H, 3 · NH2), 7.18–7.86 (m, 7H, Ar-H · 7).
Anal. Calcd for C22H24N6O2: C, 65.33; H, 5.98; N,
20.78. Found: C, 65.14; H, 5.99; N, 20.66.
to a solution of 98% H2SO4 (0.55 mL, 0.98 g,
10.0 mmol) in 4.0 mL of water at 0 ꢁC. After stirring
for an additional 3 h, the pH of the mixture was
adjusted to 7 with 2 N NaOH at 0 ꢁC. The mixture
was stirred at room temperature for another 1 h and
then acidified with acetic acid (0.5 mL). The precipi-
tated solid was collected by filtration, washed with
water, followed by ethyl acetate, and dried in vacuum
to afford an off-white solid. This solid was dissolved in
5 mL of methanol and 100 mg of silica gel was added
and the solvent was removed to form a plug, which
was loaded on a silica gel column (15 · 150 mm),
and eluted with 4% methanol in chloroform. Fractions
containing the product (TLC) were pooled and evapo-
rated to afford 18–23.
3.37. 5-[3-(4-Biphenyl)-1-(nitromethyl)butyl]pyrimidine-
2,4,6-triamine (63)
Using the general procedure described above, com-
pound 63 was obtained from ( )-1-nitro-4-(4-biphe-
nyl)-1-pentene 57 as a yellow solid in 78% yield. Mp:
75–76 ꢁC; Rf 0.45 (CHCl3:CH3OH/5:1). 1H NMR
(DMSO-d6): d 1.22 (d, 3H, CH3), 1.65-2.25 (m, 2H,
CH2), 2.30–2.55 (m, 1H, CH), 3.70–3.95 (m, 1H, CH),
4.65–4.90 (m, 2H, CH2), 5.30–5.51 (m, 6H, 3 · NH2),
7.20–7.70 (m, 9H, Ar-H · 9). Anal. Calcd for
C21H24N6O2: C, 64.27; H, 6.16; N, 21.41. Found: C,
64.22; H, 6.17; N, 21.48.
3.41. ( )-2,4-Diamino-5-(2-methyl-2-phenyl)ethyl-pyr-
rolo[2,3-d]pyrimidine (18)
Using the general procedure described above, com-
pound 18 was obtained from ( )-1-nitro-2-(2,4,6-triami-
nopyrimidin-5-yl)-4-phenyl-pentane 60 as an off-white
solid in 49% yield. Mp: >250 ꢁC; Rf 0.65
1
(CHCl3:CH3OH/5:1). H NMR (DMSO-d6): d 1.18 (d,

8350
A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
3H, CH3), 2.89–2.98 (m, 3H, CH and CH2), 5.42 (s, 2H,
NH2), 5.98 (s, 2H, NH2), 6.23 (s, 1H, CH), 7.14–7.23 (m,
5H, Ar-H · 5), 10.35 (s, 1H, NH). Anal. Calcd for
C15H17N5: C, 67.39; H, 6.41; N, 26.20. Found: C,
67.25; H, 6.41; N, 26.08.
3.46. ( )-2,4-Diamino-5-[2-methyl-2-(3,4,5-triOMephe-
ny)]ethyl-pyrrolo[2,3-d]pyrimidine(23)
Using the general procedure described above, com-
pound 23 was obtained from ( )-1-nitro-2-(2,4,6-triami-
nopyrimidin-5-yl)-4-(3,4,5-trimethoxyphenyl)-pentane
65 as an off-white solid in 49% yield. Mp: >250 ꢁC; Rf
0.53 (CHCl3:CH3OH/5:1). 1H NMR (DMSO-d6): d
1.18 (d, 3H, CH3), 2.84–2.94 (m, 3H, CH and CH2),
3.60 (s, 3H, OCH3), 3.65 (s, 6H, 2 · OCH3), 5.36 (s,
2H, NH2), 5.92 (s, 2H, NH2), 6.31 (s, 1H, CH), 6.51
(s, 2H, Ar-H · 2), 10.34 (s, 1H, NH). Anal. Calcd
C18H23N5O3Æ0.1H2O: C, 60.19; H, 6.51; N, 19.50.
Found: C, 59.86; H, 6.62; N, 19.24.
3.42. ( )-2,4-Diamino-5-[2-methyl-2-(2-naphthyl)]ethyl-
pyrrolo[2,3-d]pyrimidine (19)
Using the general procedure described above, com-
pound 19 was obtained from ( )-1-nitro-2-(2,4,6-triami-
nopyrimidin-5-yl)-4-(2-naphthyl)-pentane 61 as an
off-white solid in 53% yield. Mp: >200 ꢁC; Rf 0.64
1
(CHCl3:CH3OH/5:1). H NMR (DMSO-d6): d 1.31 (d,
3H, CH3), 2.95–3.20 (m, 3H, CH and CH2), 5.35 (s,
2H, NH2), 5.96 (s, 2H, NH2), 6.23 (s, 1H, CH), 7.46–
7.92 (m, 7H, Ar-H · 7), 10.27 (s, 1H, NH). Anal. Calcd
for C15H17N5Æ0.6H2O: C, 69.53; H, 6.20; N, 21.34.
Found: C, 69.61; H, 6.25; N, 21.02.
3.47. Sources of DHFR enzymes
Dihydrofolate reductase from P. carinii was produced as
the recombinant enzyme expressed in Escherichia coli.35
The sequence of the protein was identical to that pre-
dicted for the previously reported gene sequence.36
3.43. ( )-2,4-Diamino-5-[2-methyl-2-(2-fluorenyl)]ethyl-
pyrrolo[2,3-d]pyrimidine (20)
Dihydrofolate reductase from T. gondii was isolated
directly from the RH strain of T. gondii grown in culture
on mutant Chinese hamster ovary cells lacking dihydro-
folate reductase (CHO/dhfrÀ, American Type Culture
Collection 3952 CL).36 The organisms are introduced
into a confluent monolayer of the cells and harvested
when they have lysed the mammalian cells. The
100,000g supernate is stored in liquid nitrogen.
Using the general procedure described above, com-
pound 20 was obtained from ( )-1-nitro-2-(2,4,6-tria-
minopyrimidin-5-yl)-4-(2-fluorenyl)-pentane 62 as an
off-white solid in 48% yield. Mp: >200 ꢁC; Rf 0.68
(CHCl3:CH3OH/5:1). 1H NMR (DMSO-d6): d 1.27
(d, 3H, CH3), 2.85–3.15 (m, 3H, CH and CH2), 3.87
(s, 2H, CH2), 5.37 (s, 2H, NH2), 5.96 (s, 2H, NH2),
6.28 (s, 1H, CH), 7.15–7.95 (m, 7H, Ar-H · 7), 10.32
(s, 1H, NH). Anal. Calcd for C22H21N5Æ0.8H2O: C,
71.44; H, 6.16; N, 18.94. Found: C, 71.28; H, 6.10;
N, 18.63.
Mycobacterium avium used in these studies was a clinical
isolate from Indiana University School of Medicine,
Department of Pathology; it was identified as a serovar
4. The strain was maintained on Lowenstein-Jensen
slants (Baxter Scientific) grown at room temperature.
To produce enzyme, the organism was grown in Middle-
brook 7H-9 liquid medium at 37 ꢁC to an OD660 of 0.5–
0.7, which took several weeks. At harvest, the bacteria
were sedimented by centrifugation, sonicated, and the
100,000g supernate was stored under liquid nitrogen un-
til assay. These supernates contained both dihydrofolate
reductase and dihydroopteroate synthetase activity.
3.44. ( )-2,4-Diamino-5-[2-methyl-2-(4-biphenyl)]ethyl-
pyrrolo[2,3-d]pyrimidine (21)
Using the general procedure described above, com-
pound 21 was obtained from ( )-1-nitro-2-(2,4,6-triami-
nopyrimidin-5-yl)-4-(4-biphenyl)-pentane 63 as an
off-white solid in 45% yield. Mp: >230 ꢁC; Rf 0.66
1
(CHCl3:CH3OH/5:1). H NMR (DMSO-d6): d 1.22 (d,
3H, CH3), 2.85–3.15 (m, 3H, CH and CH2), 5.42 (s,
2H, NH2), 5.99 (s, 2H, NH2), 6.29 (s, 1H, CH), 7.33–
7.64 (m, 9H, Ar-H · 9), 10.36 (s, 1H, NH). Anal. Calcd
for C21H21N5Æ1.2H2O: C, 69.09; H, 6.46; N, 19.18.
Found: C, 69.33; H, 6.07; N, 18.83.
Rat liver dihydrofolate reductase was prepared from liv-
ers of adult female Sprague–Dawley rats. The 100,000g
supernate prepared from crude homogenates was par-
tially purified by ammonium sulfate precipitation; the
50–90% precipitate was redissolved and stored in liquid
nitrogen.
3.45. ( )-2,4-Diamino-5-[2-methyl-2-(2,5-dimethoxyphe-
nyl)]ethyl-pyrrolo[2,3-d]pyrimidine (22)
3.48. DHFR assay
Using the general procedure described above, com-
pound 22 was obtained from ( )-1-nitro-2-(2,4,6-triami-
nopyrimidin-5-yl)-4-(2,5-dimethoxyphenyl)-pentane 64
as an off-white solid in 50% yield. Mp: >250 ꢁC; Rf
0.57 (CHCl3:CH3OH/5:1). 1H NMR (DMSO-d6): d
1.08 (d, 3H, CH3), 2.36–3.30 (m, 3H, CH and CH2),
3.70 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 5.38 (s, 2H,
NH2), 6.20 (s, 2H, NH2), 6.39 (s, 1H, CH), 6.72–6.90
(m, 3H, Ar-H · 3), 10.39 (s, 1H, NH). Anal. Calcd for
C17H21N5O2: C, 62.37; H, 6.47; N, 21.39. Found: C,
62.21; H, 6.59; N, 21.30.
The spectrophotometric assay for dihydrofolate reduc-
tase was modified to optimize for temperature, substrate
concentration, and cofactor concentration for each
enzyme form assayed. The standard assay contained
sodium phosphate-buffer, pH 7.4 (40.7 mM), 2-mercap-
toethanol (8.9 mM), NADPH (0.117 mM), dihydrofolic
acid (0.092 mM), and sufficient enzyme to produce a
change in OD340 of 0.035/min. KCL (150 mM) was add-
ed to the reaction for the T. gondii, M. avium, and rat
liver enzymes. All assays were performed at 37 ꢁC. The

A. Gangjee et al. / Bioorg. Med. Chem. 14 (2006) 8341–8351
8351
9. Gangjee, A.; Guo, X.; Queener, S. F.; Cody, V.; Galitsky,
N.; Luft, J. R.; Pangborn, W. J. Med. Chem. 1998, 41,
1263.
10. Gangjee, A.; Zeng, Y.; McGuire, J. J.; Kisliuk, R. L.
J. Med. Chem. 2000, 43, 3125.
11. Gangjee, A.; Adair, O.; Queener, S. F. J. Med. Chem.
1999, 42, 2447.
12. Miwa, T.; Hitaka, T.; Akimoto, H.; Nomura, H. J. Med.
Chem. 1991, 34, 555.
reaction was followed for 3 min with continuous record-
ing. Activity under these conditions was linear with en-
zyme concentration over at least a 4-fold range.
Background activity in the absence of dihydrofolic acid
was near zero and was subtracted from all rates.
3.49. Determination of IC50 values
Dihydrofolate reductase was assayed in the presence of
a series of concentrations of inhibitor to produce a range
of rates from 1% to 90% of the uninhibited rate. At least
three concentrations were required for calculation; most
curves contained five or more concentrations. Semiloga-
rithmic plots of the data yielded normal sigmoidal
curves for most inhibitors. The concentration yielding
50% inhibition (IC50) was calculated using Prism 4.0
(GraphPad).
13. Rosowsky, A.; Chen, H.; Fu, H.; Queener, S. F. Bioorg.
Med. Chem. 2003, 11, 59.
14. Taylor, E. C.; Liu, B. Tetrahedron Lett. 1999, 40, 4023.
15. Taylor, E. C.; Liu, B. Tetrahedron Lett. 1999, 40, 4027.
16. Luzzio, F. A. Tetrahedron 2001, 57, 915.
17. Melpolder, J. B.; Heck, R. F. J. Org. Chem. 1976, 41, 265.
18. Galemmo, R. A., Jr.; Johnson, W. H., Jr.; Learn, K. S.;
Lee, T. D. Y.; Huang, F. C.; Campbell, H. F.; Youssefyeh,
R.; O’Rourke, S. V.; Schuessler, G. J. Med. Chem. 1990,
33, 2828.
19. Albrecht, M.; Riether, C. Synthesis 1997, 957.
20. Crombie, L.; Mistry, J. Tetrahedron Lett. 1975, 31, 2647.
21. Durman, J.; Elliott, J.; McElroy, A. B.; Warren, S.
J. Chem. Soc., Perkin Trans. 1 1985, 1237.
Acknowledgments
22. Asaoka, M.; Shima, K.; Fujii, N.; Takei, H. Tetrahedron
1988, 44, 4757.
23. Ambros, R.; Schneider, M. R.; Von Angerer, S. J. Med.
Chem. 1990, 33, 153.
This work was supported, in part, by a grant from the
NIH, National Institutes of Allergy and Infections Dis-
ease AI 47759 (AG).
24. Chen, Q.; Huggins, M. T.; Lightner, D. A.; Norona, W.;
McDonagh, A. F. J. Am. Chem. Soc. 1999, 121, 9253.
25. Sinhababu, A. K.; Borchardt, R. T. Tetrahedron Lett.
1983, 24, 227.
References and notes
26. Gairaud, C. B.; Lappin, G. R. J. Org. Chem. 1953, 15, 1.
27. Kambe, S. Chem. Soc. Bull. JP 1968, 41, 1444.
28. Urban, F. J.; Breitenbach, R.; Murtiashaw, C. W.;
Vanderplas, B. C. Tetrahedron: Asymmetry 1995, 6, 321.
29. Melton, J.; McMurry, J. E. J. Org. Chem. 1975, 40, 2138.
30. Taylor, E. C.; Liu, B. J. Org. Chem. 2003, 68, 9938.
31. Pinnick, H. W. Org. React. 1990, 38, 655.
32. Cody, V.; Galitsky, N.; Luft, J. R.; Pangborn, W.;
Queener, S. F.; Gangjee, A. Acta Cryst. D: Biol. Cryst.
2002, D58, 1393.
33. Anderson, P. L.; Brittian, D. A. Ger. Offen. DE 2439294,
1975; Chem. Abstr. 1975, 82, 170341.
34. Moffett, R. B. J. Med. Chem. 1964, 7, 319.
35. Bartlett, M. S.; Shaw, M.; Navaran, P.; Smith, J. W.;
Queener, S. F. Antimicrob. Agents Chemother. 1995, 39,
2441.
1. Presented in part at 225th American Chemical Society
National Meeting. New Orleans, LA, March 2003;
Amercian Chemical Society: Washington, DC, 2003;
MEDI 134.
2. Gangjee, A.; Elzein, E.; Kothare, M.; Vasudevan, A. Curr.
Pharm. Des. 1996, 2, 263.
3. Klepser, M. E.; Klepser, T. B. Drugs 1997, 53, 40.
4. Masur, H.; Polis, M. A.; Tuazon, C. U.; Ogata-Arakaki,
D.; Kovacs, J. A.; Katz, D.; Hilt, D.; Simmons, T.;
Feuerstein, I.; Lundgren, B.; Lane, H. C.; Chabner, B. A.;
Allegra, C. J. J. Infect. Dis. 1992, 167, 1422.
5. Gangjee, A.; Mavandadi, F.; Queener, S. F.; McGuire, J.
J. J. Med. Chem. 1995, 38, 2158.
6. Gangjee, A.; Mavandadi, F.; Queener, S. F. J. Med.
Chem. 1997, 40, 1173.
7. Gangjee, A.; Devraj, R.; McGuire, J. J.; Kisliuk, R. L.;
Queener, S.; Barrows, L. R. J. Med. Chem. 1994, 37, 1169.
8. Gangjee, A.; Devraj, R.; McGuire, J. J.; Kisliuk, R. L.
J. Med. Chem. 1995, 38, 3798.
36. Edman, J. C.; Edman, U.; Cao, M.; Lundgren, B.;
Kovacs, J. A.; Santi, D. V. Proc. Natl. Acad. Sci. U.S.A.
1989, 85, 8625.