Synthesis and biological evaluation of nonclassical 2,4-diamino-5- methylpyrido[2,3-d]pyrimidines with novel side chain substituents as potential inhibitors of dihydrofolate reductases

Article abstract of DOI:10.1021/jm960734f

Nine novel 2,4-diamino-5-methyl-6-substituted-pyrido[2,3-d]pyrimidines, 2-10, were synthesized as potential inhibitors of Pneumocystis carinii dihydrofolate reductase (pcDHFR) and Toxoplasma gondii dihydrofolate reductase (tgDHFR). Compounds 2-5 were designed as conformationally restricted analogues of trimetrexate (TMQ), in which rotation around τ₃ was constrained by incorporation of the side chain nitrogen as part of an indoline or an indole ring. Analogue 6, which has an extra atom between the side chain nitrogen and the phenyl ring, has its nitrogen as part of a tetrahydroisoquinoline ring. Analogues 7-9 are epiroprim (Ro 11-8958) analogues and contain a pyrrole ring as part of the side chain substitution on the phenyl ring similar to epiroprim. These analogues were designed to investigate the role of the pyrrole substitution on the phenyl ring of 2,4- diamino-5-methyl-6-(anilinomethyl)pyrido[2,3-d]pyrimidines. Molecular modeling indicated that a pyrrole substituent in the ortho position of the side chain phenyl ring was most likely to interact with pcDHFR in a manner similar to the pyrrole moiety of epiroprim. Analogue 10, in which a phenyl ring replaced a methoxy group, was synthesized to determine the contribution of a phenyl ring on selectivity, lipophilicity, and cell penetration. The synthesis of analogues 2-4 was achieved via reductive amination of 2,4- diamino-5-methyl 6-carboxaldehyde with the appropriately substituted indolines. The indolines were obtained from the corresponding indoles via NaCNBH₃ reductions. Analogues 5-10 were synthesized by nucleophilic displacement of 2,4-diamino-5-methyl-6-(bromemethyl)-pyrido[2,3-d]pyrimidine with the 5-methoxyindolyl anion, 6,7-dimethoxytetrahydroisoquinoline, the appropriately substituted pyrroloaniline or 2-methoxy-5-phenylaniline. The pyrroloanilines were synthesized in two steps by treating the substituted nitroanilines with 2,5-dimethoxy-tetrahydrofuran to afford the nitropyrrole intermediates, followed by reduction of the nitro group with Raney Ni. The analogues were more potent than trimethoprim and epiroprim and more selective than TMQ and piritrexim against pcDHFR and tgDHFR. Compounds 5 and 10 had IC₅₀ values of 1 and 0.64 μM, respectively, for the inhibition of the growth of T. gondii cells in culture, and showed excellent culture IC₅₀/enzyme IC₅₀ ratios, which were correlated with their calculated log P values, indicating a direct relationship between calculated lipephilicity and cell penetration.

Full text of DOI:10.1021/jm960734f

+
+
J . Med. Chem. 1997, 40, 479-485
479
Syn th esis a n d Biologica l Eva lu a tion of Non cla ssica l
2,4-Dia m in o-5-m eth ylp yr id o[2,3-d ]p yr im id in es w ith Novel Sid e Ch a in
Su bstitu en ts a s P oten tia l In h ibitor s of Dih yd r ofola te Red u cta ses1
Aleem Gangjee,*,† Anil Vasudevan,† and Sherry F. Queener‡
Division of Medicinal Chemistry, Graduate School of Pharmaceutical Sciences, Duquesne University,
Pittsburgh, Pennsylvania 15282, and Department of Pharmacology and Toxicology, Indiana University School of Medicine,
Indianapolis, Indiana 46202
Received October 18, 1996X
Nine novel 2,4-diamino-5-methyl-6-substituted-pyrido[2,3-d]pyrimidines, 2-10, were synthe-
sized as potential inhibitors of Pneumocystis carinii dihydrofolate reductase (pcDHFR) and
Toxoplasma gondii dihydrofolate reductase (tgDHFR). Compounds 2-5 were designed as
conformationally restricted analogues of trimetrexate (TMQ), in which rotation around τ3 was
constrained by incorporation of the side chain nitrogen as part of an indoline or an indole ring.
Analogue 6, which has an extra atom between the side chain nitrogen and the phenyl ring,
has its nitrogen as part of a tetrahydroisoquinoline ring. Analogues 7-9 are epiroprim (Ro
11-8958) analogues and contain a pyrrole ring as part of the side chain substitution on the
phenyl ring similar to epiroprim. These analogues were designed to investigate the role of the
pyrrole substitution on the phenyl ring of 2,4-diamino-5-methyl-6-(anilinomethyl)pyrido[2,3-
d]pyrimidines. Molecular modeling indicated that a pyrrole substituent in the ortho position
of the side chain phenyl ring was most likely to interact with pcDHFR in a manner similar to
the pyrrole moiety of epiroprim. Analogue 10, in which a phenyl ring replaced a methoxy
group, was synthesized to determine the contribution of a phenyl ring on selectivity, lipophilicity,
and cell penetration. The synthesis of analogues 2-4 was achieved via reductive amination
of 2,4-diamino-5-methyl 6-carboxaldehyde with the appropriately substituted indolines. The
indolines were obtained from the corresponding indoles via NaCNBH3 reductions. Analogues
5-10 were synthesized by nucleophilic displacement of 2,4-diamino-5-methyl-6-(bromomethyl)-
pyrido[2,3-d]pyrimidine with the 5-methoxyindolyl anion, 6,7-dimethoxytetrahydroisoquinoline,
the appropriately substituted pyrroloaniline or 2-methoxy-5-phenylaniline. The pyrroloanilines
were synthesized in two steps by treating the substituted nitroanilines with 2,5-dimethoxy-
tetrahydrofuran to afford the nitropyrrole intermediates, followed by reduction of the nitro
group with Raney Ni. The analogues were more potent than trimethoprim and epiroprim and
more selective than TMQ and piritrexim against pcDHFR and tgDHFR. Compounds 5 and 10
had IC50 values of 1 and 0.64 µM, respectively, for the inhibition of the growth of T. gondii
cells in culture, and showed excellent culture IC50/enzyme IC50 ratios, which were correlated
with their calculated log P values, indicating a direct relationship between calculated
lipophilicity and cell penetration.
The principal causes of morbidity and mortality in
patients with AIDS are opportunistic infections with
Pneumocystis carinii and Toxoplasma gondii.2 One of
the ways of tackling P. carinii infections is by prophy-
laxis including the dihydrofolate reductase (DHFR)
inhibitor, trimethoprim (TMP). TMP is a weak inhibitor
of P. carinii (pc) DHFR and T. gondii (tg) DHFR and
hence is used along with sulfamethoxazole (SMX) to
augment its inhibition. Other drugs such as atova-
quone3 and pentamidine4 have been evaluated for the
treatment of these infections, but TMP/SMX is consid-
ered the most effective first-line treatment for P. carinii
pneumonia (PCP). The combination of pyrimethamine
and sulfadiazine is currently approved for the treatment
of T. gondii infections. Both TMP/SMX and py-
rimethamine/sulfadiazine suffer from disadvantages
which often force discontinuation of therapy.5 Drugs
suggested as substitutes include trimetrexate (TMQ)
and piritrexim (PTX), two potent lipophilic, nonclassical
inhibitors of pcDHFR and tgDHFR.6-8 These drugs are
able to penetrate P. carinii and T. gondii cells by passive
diffusion, thus circumventing the requirement for the
carrier-mediated active transport system(s) needed for
the uptake of classical antifolates such as methotrexate
(MTX), but both TMQ and PTX lack selectivity for either
pcDHFR or tgDHFR, and their use as a single agent is
associated with considerable host toxicity. As a result,
TMQ has to be coadministered with leucovorin (N5
formyl tetrahydrofolate) which is selectively taken up
by host cells, thus providing a rescue of host cells and
decreasing toxicity.
With the recent report of the crystal structure of
pcDHFR,9 the search for potent and selective inhibitors
of pcDHFR has gained momentum. Gangjee et al.10-12
recently reported the synthesis and biological activities
of series of 2,4-diamino-6-substituted-pyrido[2,3-d]py-
rimidines with a variety of side chain substituents as
semirigid analogues of TMQ, in an attempt to increase
the selectivity and/or potency of TMQ. In general, these
analogues were more selective than TMQ against
pcDHFR and tgDHFR. Some analogues displayed
1000-1400-fold greater selectivity than TMQ against
tgDHFR and pcDHFR, respectively. It was observed
† Duquesne University.
‡ Indiana University School of Medicine.
X Abstract published in Advance ACS Abstracts, J anuary 15, 1997.
S0022-2623(96)00734-0 CCC: $14.00 © 1997 American Chemical Society

+
+
480 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4
Gangjee et al.
that the presence of a methyl moiety on the side chain
bridge nitrogen resulted in increased potency and, in
some instances, improved selectivity against pcDHFR
and tgDHFR. This suggested that conformational
restriction about τ2 and τ3 induced by the N-methyl
moiety may be conducive to potency and selectivity, in
addition to the hydrophobic interactions of the methyl
group with the enzyme. The ability of these analogues
to penetrate T. gondii cells in culture, an important
attribute of a clinically viable agent, was also studied.
Gangjee et al.11 have reported that incorporation of
bulky naphthalene or 4-methoxynaphthalene rings in
the side chain resulted in excellent cell penetration.
Further, dimethoxyphenyl-substituted analogues dis-
played greater cell penetration compared to trimethoxy-
phenyl-substituted analogues. As part of our continuing
effort to synthesize potent and selective inhibitors of
pcDHFR and tgDHFR, we synthesized compounds
2-10. Analogues 2-5 were designed as conformation-
serve to further increase the potency and/or selectivity
of these analogues by orienting the side chain in a more
favorable conformation for interaction with tgDHFR. In
addition, in a series of nonclassical tetrahydroquinazo-
line analogues recently reported by Gangjee et al.,14
an
indoline substitution resulted in a selectivity ratio of 6
for tgDHFR (vs rlDHFR). Thus, it was of interest to
evaluate compounds 2-5 as inhibitors of pcDHFR and
tgDHFR. Comparison of the pcDHFR and tgDHFR
inhibitory activities as well as the in vitro P. carinii and
T. gondii cell culture inhibition results of an indoline
and indole analogue would provide an insight into the
role of a flat, heteroaromatic system (such as an indole)
vs a bulky substitution (such as an indoline) on the side
chain for inhibition of pcDHFR and tgDHFR as well as
for cell penetration. This aspect was of special interest,
since the hydrophobic interactions of the indole and the
indoline moieties with the enzyme as well as their
contribution to the overall lipophilicity (as reflected in
their calculated log P values15) of the compound, and
hence cell penetration was anticipated to be different.
In their study of trimethoprim analogues, Rauckmann
et al.,16 had reported significant differences in DHFR
inhibition and selectivities for E. coli (ec) DHFR with
the quinoline and tetrahydroquinoline moieties, lending
further credence to the idea that the indoline and indole
analogues were worth investigating as inhibitors of
pcDHFR and tgDHFR as well as inhibitors of P. carinii
and T. gondii cells in culture.
Analogue 6 was designed to study the effect of
extending the bridge length between the pyrido[2,3-d]-
pyrimidine moiety and the side chain phenyl ring by
an extra atom, thus probing any additional contacts that
the extended side chain phenyl ring might provide with
either pcDHFR and/or tgDHFR. Such an extension of
the bridge between the 2,4-diamino-substituted hetero-
cyclic moiety and the side chain phenyl ring from two
atoms to three atoms has been recently reported by
Gangjee et al.17
Then et al.18 reported a series of trimethoprim ana-
logues, one of which, epiroprim (Ro 11-8958), had a
reported 200-fold selectivity for pcDHFR vs human (h)
DHFR. This is the most selective inhibitor of pcDHFR
known to date. In addition, epiroprim was also an
extremely selective (3 × 105-fold) and potent inhibitor
of ecDHFR. The ability of epiroprim to inhibit PCP in
combination with other antifolates has been studied,
and some synergy has been observed with SMX and
with dapsone.19 The analogues of epiroprim which
displayed appreciable selectivity (>20-fold) for pcDHFR
(vs hDHFR) had a pyrrole ring as part of the side chain,
suggesting a selective interaction of the pyrrole ring
with pcDHFR which was absent with hDHFR. Re-
cently, epiroprim has been evaluated against tgDHFR20
and was found to be 650-fold more selective for tgDHFR
(vs human DHFR) and was 86-fold more selective
compared to pyrimethamine. When administered in-
traperitoneally in combination with sulfadiazine at the
relatively high doses of 300 and 375 mg/kg, respectively,
100% survival was seen in mice acutely infected with
the recombinant human strain of T. gondii. Either drug
alone was ineffective at the same doses, establishing
effective synergy in this model. The potency of epiroprim,
however, is in the micromolar range. In an attempt to
ally restricted analogues of TMQ in which torsion angle
τ3 was restricted by incorporating the side chain nitro-
gen within an indoline or indole ring. In our previous
report,10 we found that 2,4-diamino-5-methyl-6-[(3′,4′,5′-
trimethoxy-N-methylanilino)methyl]pyrido[2,3-d]pyri-
midine (1, R1 ) CH3; R2 ) 3,4,5-OCH3) was both a
selective (IC50 rlDHFR/IC50 tgDHFR ) 8.94) and potent
(IC50 ) 0.85 nM) inhibitor of tgDHFR. Further, the N10
formyl analogue was as selective against tgDHFR as the
N10 methyl analogue. Molecular modeling using Sybyl
6.013 indicated that restriction about torsion angles τ2
and τ3, imposed by the N10 substituent, could be
responsible for the selective inhibition of tgDHFR, in
addition to a possible selective interaction of the N10
substituent with the enzyme. Compounds 2-5 are
conformationally restricted analogues of the N10 methyl
analogue (1, R1 ) CH3; R2 ) 3,4,5-OCH3), which could

+
+
Nonclassical 2,4-Diamino-5-methylpyrido[2,3-d]pyrimidines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 481
Sch em e 1a
Sch em e 2a
a
Reagents: (a) Raney Ni or Ni Al; (b) NaBH4 or LiEtBH3; (c)
PPh3Br2 or anhd. HBr/dioxane or 30% HBr/AcOH; (d) NaH,
DMAC; (e) K2CO3 or CsCO3.
incorporate the selectivity of epiroprim into potent
pyrido[2,3-d]pyrimidines, we synthesized the pyrrole-
substituted analogues 7-9. The 2′,5′-dimethoxy-sub-
stituted analogue (1, R1 ) H; R2 ) 2,5-OCH3) displayed
a 3-fold selectivity for pcDHFR and an 8-fold selectivity
for tgDHFR (vs rlDHFR).11 Comparison of the biological
activity of this analogue with analogue 7 would permit
an assessment of the contribution to the activity and
selectivity of a pyrrole moiety in the pyrido[2,3-d]-
pyrimidine series. Molecular modeling using Sybyl 6.0
indicated that a pyrrole substituent in the ortho position
of the side chain phenyl ring was most likely to interact
with pcDHFR in the same manner as the pyrrole moiety
of epiroprim. Further, the substitution of a pyrrole
moiety at the ortho position as in 8 might also serve to
increase the conformational restriction about τ3, a
feature which could provide increased selectivity and/
or potency against pcDHFR and/or tgDHFR. Compound
9 was synthesized to significantly increase the lipid
solubility of 7 by replacing the 2′,5′-dimethoxy moieties
with methyl moieties and further adding two methyl
groups at the 3′ and 6′ positions. Analogue 10, in which
a methoxy group has been replaced with a phenyl ring,
would provide an assessment of the contribution of a
hydrophobic phenyl moiety in place of a methoxy group,
in particular with respect to P. carinii and T. gondii cell
penetration, since this analogue is anticipated to possess
improved lipid solubility compared to 1 (R1 ) H; R2 )
2,5-diOCH3).
a
Reagents: (a) 2,5-dimethoxytetrahydrofuran, CH3COOH, re-
flux; (b) Raney Ni, EtOH, 35 psi; (c) NaCNBH3, CH3COOH.
to the two pairs of methylene protons, confirmed the
product. Reductive amination of the 6-carboxaldehyde
with these indolines was attempted in methanol.14
Despite variation in reaction time and temperature, the
yield of the desired analogues, as evidenced from TLC
analysis, was poor (<5%). One of the reasons for the
poor yield was the insolubility of the aldehyde in
methanol. To improve the solubility of the aldehyde, it
was converted to its acetate salt. The reductive ami-
nation was then attempted using the more soluble
acetate salt of the aldehyde and the appropriate indoline
using NaCNBH3. HCl (6 N) in methanol was added to
the reaction mixture to maintain the pH at 6. The use
of freshly activated molecular sieves23 afforded ana-
logues 2-4 in 30-35% yields.
For the synthesis of the indole analogue 5, the
6-aldehyde 12 was reduced to the 6-alcohol 13 using
NaBH4 in methanol.11 The alcohol 13 was brominated
using one of three methods reported in the literature:
(a) triphenylphosphine dibromide,21 (b) anhydrous HBr
in dioxane,24 and (c) 30% HBr in acetic acid.25 The
bromination was most efficient using 30% HBr/AcOH
and afforded the 6-bromomethyl analogue 14 as a
hydrobromide salt in 90% yield. Initial attempts at
alkylating the sodium salt of the indole nitrogen with
the bromomethyl compound did not afford the desired
analogue due to rapid quenching of the anion with the
hydrogen bromide associated with the bromomethyl
compound. Hence, the bromomethyl compound 14 was
first treated with triethylamine under an atmosphere
of argon, and then the sodium salt of the indole was
added dropwise at room temperature to afford the
desired analogue in 30% yield. The regioselectivity of
Ch em istr y
The synthetic route adopted for these analogues was
a modification of that reported earlier.11 This involved
the synthesis of 2,4-diamino-5-methylpyrido[2,3-d]py-
rimidine-6-carbonitrile, 11, in improved yields11 com-
pared to the literature procedure21 (74%; lit. 54%). This
nitrile was then reduced to the 6-carboxaldehyde 12
(Scheme 1) using either Raney Ni or a Ni-Al alloy in
formic acid. Due to the instability of the aldehyde, it
was not purified but used directly for further transfor-
mations.
1
the alkylation (N1 vs C3) was confirmed from the H
NMR which indicated the presence of the two doublets
for C2-H and C3-H of the indole moiety and the absence
of the N1-H of the indole.
The synthesis of the target compounds 7-9 required
the synthesis of the appropriate pyrrole-substituted
anilines. These were synthesized in two steps, starting
from the appropriately substituted nitroanilines (Scheme
2). Refluxing these anilines with 2,5-dimethoxytetrahy-
drofuran in acetic acid afforded the pyrrolonitroben-
zenes in quantitative yield. For the tetramethylnitro-
aniline, it was necessary to reflux the reaction for 10 h
to allow for complete formation of the pyrrole (TLC),
presumably due to steric hindrance of the amino group.
Selective reduction of the nitro group using Raney nickel
We have previously reported14 the reductive amina-
tion of 2,4-(bisacetylamino)-5,6,7,8-tetrahydroquinazo-
line-6-carboxaldehyde with various secondary amines
including indoline. We adopted a similar strategy for
the synthesis of compounds 2-4 (Scheme 1). The
indoline precursors were obtained in quantitative yield
via the reduction of the corresponding methoxy-substi-
tuted indole analogues using NaCNBH3 in acetic acid
(Scheme 2).22 The presence of two triplets in the 1H
NMR spectra in deuterated chloroform, corresponding

+
+
482 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4
Gangjee et al.
Ta ble 1. Inhibitory Concentrations (IC50, µM) and Selectivity
Ratios against P. carinii (pc), T. gondii (tg), and Rat Liver (rl)
DHFRa
the effect of a substituted indoline and an indole on all
three DHFRs is quite similar, although a 2-fold increase
in selectivity with the indole analogue 5, for tgDHFR,
was obtained.
pcDHFR rlDHFR rl/pc tgDHFR rl/tg
1 (R1 ) H;
0.046b
0.13b
2.8b
0.016b
8.0b
Incorporation of the side chain nitrogen in a tetrahy-
droisoquinoline ring as in analogue 6 resulted in sub-
micromolar inhibitory activity against all three DHFRs.
The weak activity of this analogue against pcDHFR and
tgDHFR could be attributed to the lack of sufficient
space in the active site of the enzyme to accommodate
the bulky isoquinoline moiety, or to the presence of a
basic, protonated nitrogen in the side chain, or both. The
active site of DHFR has been shown by Piper et al.25 to
be sufficiently large to accommodate bulky substituents
on the side chain. Gangjee et al.17 have speculated on
the role of a basic, protonated nitrogen in the side chain
as being one of the reasons for the lack of significant
DHFR inhibitory activity of classical furo[2,3-d]pyrimi-
dine analogues. Results obtained with analogue 6
corroborate that initial17 assumption. However, further
studies are required to assess the exact role of a basic
nitrogen in the side chain of DHFR inhibitors toward
inhibition of the enzyme.
R2 ) 2,5OCH3)
2
3
4
5
6
7
8
9
0.29
0.25
0.41
0.57
0.85
0.35
1.8
0.62
0.64
0.042
0.038
0.15
0.17
0.23
0.47
0.13
0.23
3.5
0.17
0.44
0.003
0.0015
0.52
0.68
0.56
0.83
0.2
0.7
1.9
0.3
0.7
0.07
0.04
11.1
12.8
0.048
0.057
0.049
0.077
0.11
3.1
3.0
4.7
6.1
1.2
7.0
5.8
2.3
6.5
0.3
0.14
0.033
0.6
0.075
0.068
0.010
0.011
2.7
10
TMQ
PTX
TMP
epiroprim
12
2.6
133
33.2
49
70.6
0.48
a
These assays were carried out at 37 °C, under conditions of
substrate (90 µM dihydrofolic acid) and cofactor (119 µM NaDPH)
in the presence of 150 mM KCl and 2-mercaptoethanol (8.9 mM),
b
in a sodium phosphate buffer (pH 7.4).26,27 Data from ref 11.
Ta ble 2. T. gondii Cell Culture Inhibition (IC50, µM) of
Selected Analoguesa
IC50
Analogues 7-9 (which were designed to investigate
the importance and/or interaction of the pyrrole ring)
were more active than epiroprim against pcDHFR and
rlDHFR; 7 and 9 were more active than epiroprim
against tgDHFR, while 8 was equipotent with epiroprim.
Substitution of a pyrrole ring at the para position of the
phenyl ring as in analogue 7 resulted in a decrease in
activity against pcDHFR and tgDHFR as well as
rlDHFR compared to the pyrrole-unsubstituted ana-
logue (1, R1 ) H; R2 ) 2,5-OCH3). Compound 7 retained
the selectivity for tgDHFR displayed by compound 1.
Further, it was 15-fold more potent than epiroprim
against tgDHFR. Analogue 8, with an ortho pyrrolo-
4,5-dimethoxy substituent, displayed IC50 values for all
three DHFRs in the micromolar range. This result
indicates that substitution of a pyrrole ring in the ortho
position is detrimental to potency against pcDHFR and
tgDHFR; nevertheless, this analogue displayed a selec-
tivity ratio of 1.9 for pcDHFR (vs rlDHFR) and was the
most pcDHFR selective analogue of this series. The
analogue also had a 6-fold selectivity for tgDHFR.
Analogue 10, in which one of the methoxy groups was
replaced with a phenyl ring, provided interesting re-
sults. Comparison of its activity with that of the 2′,5′-
dimethoxyphenyl-substituted analogue (1, R1 ) H; R2
) 2,5-OCH3)11 indicates the importance of the substitu-
tion at the 5′-position for inhibition of pcDHFR. Re-
placement of the 5′-methoxy moiety with a phenyl ring
results in a decrease in potency against all of these
DHFR, but pcDHFR is most strongly affected with a
14-fold decrease in potency compared to the parent
analogue (1, R1 ) H, R2 ) 2,5-OCH3).11 These results
once again underscore the point that the active site of
these DHFR are quite different and that a separate
structure-activity/selectivity relationship needs to be
established for each enzyme.
tgDHFR
culture
culture/enz
log P
4
5
10
0.049
0.077
0.068
25.10
1.0
0.64
512
13
10
2.3343
3.0668
4.7690
a
T. gondii cell culture inhibition was assessed by measuring
the incorporation of [3H]uracil by T. gondii cells.26
in ethanol at 35 psi afforded the substituted pyrrolo-
anilines. Alkylation (Scheme 1) of 6,7-dimethoxytet-
rahydroisoquinoline, the pyrrole-substituted anilines
and 2-methoxy-5-phenylaniline, with the 6-bromomethyl
compound 14 using either potassium carbonate or
cesium carbonate afforded analogues 6-10 in 65-70%
yield after purification via column chromatography.
Biologica l Eva lu a tion a n d Discu ssion
All the analogues were evaluated against P. carinii
(pcDHFR), T. gondii (tgDHFR), and rat liver DHFR
(rlDHFR), and the results are indicated in Table 1.
Selectivity ratios are expressed as ratios of rlDHFR IC50/
pcDHFR IC50 and rlDHFR IC50/tgDHFR IC50, and are
also indicated in Table 1. Selected analogues were
evaluated for the inhibition of the growth of T. gondii
cells in culture, and the results are reported in Table 2.
All of the analogues were more selective than TMQ and
PTX against both pcDHFR and tgDHFR. Comparison
of the biological activities of analogues 2-4 indicates
the effect of conformational restriction around τ3 in the
form of an indoline ring. The results show that ana-
logues 2-4 have similar inhibitory effects against
pcDHFR, tgDHFR, and rlDHFR, suggesting that con-
formational restriction via an indoline ring orients the
side chain of these three analogues in similar positions,
enabling these analogues to interact with all three
DHFRs to the same extent, and that moving the position
of the methoxy group in the phenyl ring does not afford
any differential inhibition. However, analogues 2-4
were more selective than TMQ and PTX against both
pcDHFR and tgDHFR. Analogue 4, in particular was
16-fold more selective than TMQ and 33-fold more
selective than PTX against tgDHFR. Comparison of the
DHFR inhibitory concentrations of 3 and 5 indicate that
Three of the analogues (4, 5, and 10) were evaluated
as inhibitors of the growth of T. gondii cells in culture,26
and the results are shown in Table 2. The correspond-
ing calculated log P values of these analogues are also
listed. Thus, while the indoline and indole analogues
(4 and 5, respectively) displayed similar inhibition of

+
+
Nonclassical 2,4-Diamino-5-methylpyrido[2,3-d]pyrimidines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 483
4-Meth oxyin d olin e (15). This was synthesized as a
colorless oil from 4-methoxyindole (0.30 g, 1.84 mmol), and
NaCNBH3 (0.35 g, 5.52 mmol) in 20 mL of acetic acid (0.30 g,
98%): TLC (hexane:EtOAc, 1:1) Rf ) 0.45; 1H NMR (CDCl3) δ
3.04 (m, 2 H, NHCH2CH2), 3.65 (t, 2 H, CH2CH2), 3.81 (s, 3 H,
4-OCH3), 6.45 (d, 1 H, 7-H), 6.53 (d, 1 H, 5-H), 7.06 (t, 1 H,
6-H).
isolated tgDHFR, in T. gondii cell culture studies, the
indole analogue was 25-fold superior. One of the
reasons for this enhanced inhibition of the indole
analogue could be a differential transport across the cell
membrane, as is reflected in the higher calculated log
P value of 5 as compared to that of 4. This assumption
is corroborated by the lower IC50 ratio of culture/enzyme
of 10, which has a higher calculated log P value than
that of 5. It should be noted that an increase in
calculated log P value beyond a certain limit will result
in decreased cell penetration. This is substantiated in
Table 2, where there is a leveling off of the IC50 ratio of
culture/enzyme with an increase in log P, which attests
to the fact that a suitable balance between lipophilicity
and hydrophilicity is needed for superior cell penetra-
tion.
In summary, conformational restriction in the form
of an indoline or an indole ring, while providing selective
analogues, did not provide analogues which were better
in potency or selectivity than the open-chain analogues
synthesized previously.11 Further, incorporation of a
pyrrole moiety in the side chain of these molecules did
not provide any beneficial selectivity as compared to
epiroprim, suggesting that whatever interactions the
side chain of epiroprim undergoes with pcDHFR could
not be exploited with analogues 7-9. Despite the fact
that selective inhibitors of pcDHFR and tgDHFR were
attained in this study, along with other analogues with
excellent cell penetration properties, the challenge
remains to design analogues which combine both of
these attributes in a single molecule for the treatment
of P. carinii and T. gondii infections.
Anal. (C9H11NO‚0.2CH3COOH) C, H, N.
5-Meth oxyin d olin e (16). This was synthesized as an oil
from 5-methoxyindole (0.40 g, 2.45 mmol), and NaCNBH3 (0.46
g, 7.36 mmol) in 25 mL of acetic acid (0.38 g, 94%): TLC
1
(hexane:EtOAc, 1:1) Rf ) 0.46; H NMR (CDCl3) δ 2.92 (t, 2
H, NHCH2CH2), 3.34 (t, 2 H, CH2CH2), 3.73 (s, 3 H, 5-OCH3),
6.41 (d, 1H, 7-H), 6.64 (m, 1 H, 6-H), 6.72 (s, 1 H, 4-H).
Anal. (C9H11NO‚0.6CH3COOH) C, H, N.
5,6-Dim eth oxyin d olin e (17). This was synthesized as a
colorless oil from 5,6-dimethoxyindole (0.35 g, 1.81 mmol), and
NaCNBH3 (0.34 g, 5.44 mmol) in 20 mL of acetic acid (0.32 g,
92%): TLC (hexane:EtOAc, 1:1) Rf ) 0.54; 1H NMR (CDCl3) δ
3.41 (m, 2 H, NHCH2CH2), 3.67 (t, 2 H, CH2CH2), 3.75 (s, 3 H,
5-OCH3), 3.79 (s, 3 H, 6-OCH3), 6.68 (s, 1 H, 7-H), 6.75 (s, 1
H, 4-H).
Anal. (C10H13NO2‚0.5CH3COOH) C, H, N.
2,4-Diam in o-5-m eth yl-6-[(4′-m eth oxyin dolin yl)m eth yl]-
p yr id o[2,3-d ]p yr im id in e (2). To a solution of the acetate
salt of 2,4-diamino-5-methylpyrido[2,3-d]pyrimidine-6-carbox-
aldehyde (12, 0.25 g, 1.23 mmol) in 25 mL of absolute methanol
was added 4 Å molecular sieves (previously activated by
heating at 100 °C), followed by 4-methoxyindoline (0.27 g, 1.85
mmol). The reaction mixture was stirred under nitrogen for
6 h, after which NaCNBH3 (0.19 g, 3.05 mmol) was added. HCl
(6 N) in methanol was then added dropwise periodically, to
maintain the pH at 6. After 44 h, 2 mL of water was added to
the reaction mixture which was then filtered, 1.0 g of silica
gel was added to the filtrate, and the solvent was evaporated
under vacuum. The resulting dry plug was chromatographed
on a silica gel column (1.05 in. × 23 in.) using CHCl3:CH3OH
(10:1) as the eluant. Fractions containing the desired product
(Rf ) 0.61) were pooled and evaporated to afford compound 2
as a brown solid (0.27 g, 66%): mp 277 °C; IR (Nujol) 3320,
Exp er im en ta l Section
All evaporations were carried out in vacuo with a rotary
evaporator. Analytical samples were dried in vacuo (0.2
mmHg) in an Abderhalden drying apparatus over phosphorous
pentoxide and refluxing ethanol or toluene. Thin layer chro-
matography was performed on Sigma Aldrich silica gel chro-
matogram plates with fluorescent indicator. Spots were
visualized by UV light (254 and 350 nm). Proportions of
solvents used are by volume. All analytical samples were
homogeneous in at least three different solvent systems.
Melting points were determined on a Fisher J ohns melting
point apparatus or a Meltemp and are uncorrected. Purifica-
tion by gravity column chromatography and flash chromatog-
raphy were carried out using Merck silica gel 60 (230-400
mesh). Infrared (IR) spectra were determined neat or in Nujol
mulls on a Perkin-Elmer 1430 ratio recording infrared spec-
trophotometer and are reported in reciprocal centimeters. The
1H NMR spectra were recorded on a Bruker WH-300 (300
MHz) spectrometer. The chemical shift (δ) values are ex-
pressed in ppm (parts per million) relative to tetramethylsilane
as internal standard: s ) singlet, d ) doublet, t ) triplet, m
) multiplet, exch ) protons exchangeable on addition of
deuterium oxide. Elemental analyses were performed at
Atlantic Microlabs and are within +0.4 of the calculated
values. All solvents and chemicals were purchased from
Aldrich Chemical Co. and Fisher Scientific and were used as
received.
Gen er a l P r oced u r e for th e Syn th esis of th e Meth oxy-
in d olin es. To a solution of the methoxyindole (1 equiv) in
acetic acid was added NaCNBH3 (3 equiv) in portions (0.5
equiv each time) over a 10 min period (ca u tion : exothermic
reaction). The reaction was stirred at room temperature for
one-half hour after the last addition. TLC analysis (CHCl3:
Et2O, 5:1) indicated the presence of a new spot corresponding
to the product. A small amount of water (∼2-3 mL) was
added to the reaction, the acetic acid evaporated under reduced
pressure (0.1 mm, 45 °C bath tempeature) and the residue
cooled (0 to -5 °C) to afford the desired indoline.
1
3140 (NH2) cm-1; H NMR (DMSO-d6) δ 2.79 (s, 3 H, 5-CH3),
3.10 (t, 2 H, CH2CH2), 3.61 (t, 2 H, CH2CH2), 3.71 (s, 3 H,
OCH3), 4.34 (s, 2 H, CH2), 6.87 (d, 1 H, 7′-H), 7.02-7.07 (m, 2
H, 5′-H and 6′-H), 8.6 (s, 1 H, H-7).
Anal. (C18H20N6O‚1.0H2O) C, H, N.
2,4-Diam in o-5-m eth yl-6-[(5′-m eth oxyin dolin yl)m eth yl]-
p yr id o[2,3-d ]p yr im id in e (3). To a solution of the acetate
salt of 12 (0.30 g, 1.48 mmol) in 30 mL of absolute methanol
(charged with freshly activated 4 Å molecular sieves) was
added 5-methoxyindoline (0.33 g, 2.20 mmol) and the reaction
mixture was stirred under an atmosphere of nitrogen. After
5 h, NaCNBH3 (0.23 g, 3.7 mmol) was added, followed by the
periodic addition of 6 N HCl in absolute methanol to maintain
the pH at 6. After 48 h, 2 mL of water was added, the reaction
mixture filtered, and the filtrate purified as described for 2 to
afford pure 3 as a brown solid (0.31 g, 62%): mp 266 °C; IR
(Nujol) 3330, 3140 (NH2) cm-1; 1H NMR (DMSO-d6) δ 2.72 (s,
3 H, 5-CH3), 2.98 (t, 2 H, CH2CH2), 3.47 (t, 2 H, CH2CH2), 3.78
(s, 3 H, OCH3), 4.49 (s, 2 H, CH2), 6.47 (d, 1 H, 7′-H), 6.63 (d,
1 H, 6′-H), 6.78 (br s, 2 H, 2-NH2), 7.10 (d, 1 H, 4′-H), 7.63 (s,
2 H, 4-NH2), 8.41 (s, 1 H, H-7).
Anal. (C18H20N6O‚0.6H2O) C, H, N.
2,4-Dia m in o-5-m et h yl-6-[(5′,6′-d im et h oxyin d olin yl)-
m eth yl]p yr id o[2,3-d ]p yr im id in e (4). To a solution of the
acetate salt of 12 (0.23 g, 1.13 mmol) in 25 mL of absolute
methanol was added 5,6-dimethoxyindoline (0.29 g, 1.65
mmol), followed by 4 Å molecular sieves. The reaction mixture
was stirred under an atmosphere of nitrogen for 6 h, following
which NaCNBH3 (0.26 g, 4.13 mmol) was added followed by 6
N HCl in methanol to maintain the pH at 6. The reaction
mixture was then stirred for 22 h and filtered, 1.0 g of silica
gel was added to the filtrate, and the filtrate was evaporated
to dryness to afford a dry plug. This plug was loaded on the
surface of a silica gel column (1.05 in. × 23 in.) and eluted
using CHCl3:CH3OH (10:1). Fractions containing the desired
product were pooled and evaporated to afford a yellow solid

+
+
484 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4
Gangjee et al.
(0.28 g, 67%): mp 267 °C; IR (Nujol) 3330, 3140 (NH2) cm-1
;
85%): TLC (hexane:EtOAc, 1:1) Rf ) 0.59; 1H NMR (CDCl3) δ
3.94 (s, 3 H, 5-OCH3), 3.96 (s, 3 H, 4-OCH3), 6.34 (t, 2 H, 3′,4′-
H), 6.75 (d, 2 H, 2′,5′-H), 6.84 (s, 1H, 3-H), 7.50 (s, 1 H, 6-H).
Anal. (C12H12N2O4‚0.3CH3COOH) C, H, N.
1H NMR (DMSO-d6) δ 2.52 (s, 3 H, CH3), 3.01 (t, 2 H, CH2),
3.62 (t, 2 H, CH2), 3.76 (s, 6 H, OCH3), 4.35 (s, 2 H, CH2), 6.45
(s, 1 H, 7′-H), 6.72 (s, 2 H, 2-NH2), 6.91 (s, 1 H, 4′-H), 7.62 (s,
2 H, 4-NH2) 8.48 (s, 1 H, H-7).
2,3,5,6-Tetr am eth yl-4-pyr r olo-1-n itr oben zen e (20). This
was obtained as a black solid from 2,3,5,6-tetramethyl-4-
amino-1-nitrobenzene (0.50 g, 2.58 mmol) and 2,5-dimethoxy-
tetrahydrofuran (0.34 g, 2.58 mmol) in 25 mL of acetic acid
Anal. (C19H22N6O2‚1.0H2O) C, H, N.
2,4-Dia m in o-5-m eth yl-6-[(5′-m eth oxyin d olyl)m eth yl]-
p yr id o[2,3-d ]p yr im id in e (5). To a solution of 2,4-diamino-
5-methyl-6-(bromomethyl)pyrido[2,3-d]pyrimidine (14, (0.20 g,
0.75 mmol), in 20 mL of anhydrous N,N-dimethylacetamide
(DMAC) was added triethylamine dropwise until a pH of 8.
The mixture was rapidly filtered and the filtrate stirred under
nitrogen for 10 min. In another flask, 5-methoxyindole (0.11
g, 0.75 mmol) was dissolved in 15 mL of DMAC and NaH (0.03
g, 1.13 mmol), as a 60% dispersion in mineral oil was added.
The reaction mixture was stirred for 15 min and then was
added dropwise via a double-tipped canula to the solution of
the 6-bromomethyl compound under an atmosphere of nitro-
gen. The reaction mixture was stirred for 24 h while being
protected from light, at the end of which 5 mL of water was
added to the reaction mixture and the solvent evaporated
under reduced pressure. The residue was then dissolved in
100 mL of methanol, 1.0 g of silica gel was added, and the
methanol was evaporated to afford a dry plug. This plug was
eluted with CHCl3:CH3OH (15:1), and fractions containing
pure product were pooled together and evaporated to afford
pure 5 as a brown solid (0.16 g, 63%): mp 300 °C dec; TLC
(CHCl3:CH3OH, 5:1 Rf ) 0.64) IR (Nujol) 3290, 3130 (NH2)
cm-1; 1H NMR (DMSO-d6) δ 2.71 (s, 3 H, 5-CH3), 3.81 (s, 3 H,
OCH3), 4.49 (s, 2 H, CH2), 6.51 (s, 2 H, 2-NH2), 6.87 (d, 1 H,
7′-H), 7.02 (m, 2 H, 4′-H and 6′-H), 7.09 (d, 1 H, 3′-H), 7.16 (d,
1 H, 2′-H), 7.62 (s, 2 H, 4-NH2), 8.6 (s, 1 H, H-7).
1
(0.42 g, 66%): TLC (hexane:EtOAc, 1:1) Rf ) 0.76; H NMR
(CDCl3) δ 1.88 (s, 6 H, 2- and 6-CH3), 2.19 (s, 6 H, 3- and
5-CH3), 6.33 (t, 2 H, 3′,4′-H), 6.54 (d, 2 H, 2′,5′-H).
Anal. (C14H16N2O2‚0.7CH3COOH) C, H, N.
2,5-Dim eth oxy-4-p yr r oloa n ilin e (21). This was obtained
as a brown solid by hydrogenation, at 35 psi, of 2,5-dimethoxy-
4-pyrrolo-1-nitrobenzene (18) (0.20 g, 0.81 mmol) and Raney
Ni (1.0 g) in 10 mL of absolute ethanol for 2 h. The solvent
was evaporated to afford pure 21 (0.12 g, 85%): TLC (hexane:
EtOAc, 1:1) Rf ) 0.63; 1H NMR (DMSO-d6) δ 3.69 (s, 3 H,
2-OCH3), 3.82 (s, 2 H, 5-OCH3), 6.16 (m, 3 H, NH2 and 6-H),
6.99 (m, 5 H, 3-H and 2′,3′,5′,6′-H).
Anal. (C12H14N2O2‚0.2CH3COOH) C, H, N.
4,5-Dim eth oxy-2-p yr r oloa n ilin e (22). This was obtained
as a brown solid from 4,5-dimethoxy-2-pyrrolo-1-nitrobenzene
19 (0.31 g, 1.25 mmol) and Raney Ni (1.5 g) (0.18 g, 81%): TLC
1
(hexane:EtOAc, 1:1) Rf ) 0.69; H NMR (CDCl3) δ 3.85 (s, 3
H, 5-OCH3), 3.94 (s, 3 H, 4-OCH3), 6.21 (m, 3 H, NH2 and 6-H),
6.52 (s, 1 H, 3-H), 6.98 (m, 4 H, 2′,3′,4′,5′-H).
Anal. (C12H14N2O2‚0.8CH3COOH) C, H, N.
2,3,5,6-Tetr a m eth yl-4-p yr r oloa n ilin e (23). This was
obtained as a black solid from 2,3,5,6-tetramethyl-4-pyrrolo-
1-nitrobenzene (20, 0.50 g, 2.05 mmol) and Raney Ni (2.0 g)
1
(0.29 g, 66%): TLC (hexane:EtOAc, 1:1) Rf ) 0.83; H NMR
Anal. (C18H18N6O‚1.0H2O) C, H, N.
(CDCl3) δ 1.81 (s, 6 H, CH3), 2.21 (s, 6 H, CH3), 6.14-6.21 (m,
4 H, 2′,3′,4′,5′-H), 6.35 (s, 2 H, NH2). Anal. (C14H18N2‚0.3CH3-
COOH) C, H, N.
2,4-Diam in o-5-m eth yl-6-[(6′,7′-dim eth oxytetr ah ydr oiso-
qu in olin yl)m eth yl]p yr id o[2,3-d ]p yr im id in e (6). To a so-
lution of 2,4-diamino-5-methyl-6-(bromomethyl)pyrido[2,3-d]-
pyrimidine (14) (0.23 g, 0.86 mmol) in 25 mL of anhydrous
DMAC was added K2CO3 (0.24 g, 1.72 mmol) followed by 6,7-
dimethoxytetrahydroisoquinoline (0.17 g, 0.86 mmol), and the
reaction mixture was stirred under an atmosphere of nitrogen
for 24 h. TLC analyses (CHCl3:CH3OH, 5:1) indicated the
presence of the desired analogue (Rf ) 0.6) along with some
baseline contamination. The reaction mixture was filtered,
1.0 g of silica gel was added, and the filtrate was evaporated
to dryness. This plug was eluted on a silica gel column (1.05
in. × 23 in.) with CHCl3:CH3OH (100:1 to 100:20), and
fractions containing pure product were pooled and evaporated
2,4-Dia m in o-5-m et h yl-6-[(2′,5′-d im et h oxy-4′-p yr r olo-
a n ilin o)m eth yl]p yr id o[2,3-d ]p yr im id in e (7). To a solution
of 14 (0.20 g, 0.75 mmol) in 20 mL of DMAC was added 2,5-
dimethoxy-4-pyrroloaniline (0.21 g, 1.13 mmol) followed by
potassium carbonate (0.16 g, 1.13 mmol). The reaction mix-
ture was stirred at room temperature under an atmosphere
of nitrogen for 48 h. TLC analyses (CHCl3:CH3OH, 5:1) at the
end of this period indicated the presence of a new spot
corresponding to the product (Rf ) 0.6), along with some
unreacted starting material. The reaction mixture was fil-
tered, 1.0 g of silica gel added to the filtrate, and the mixture
evaporated to dryness under reduced pressure to afford a dry
plug. This plug was loaded on the surface of a 1.05 in. × 23
in. silica gel column and eluted with CHCl3:CH3OH using
gradient elution (99:1 to 80:20). Fractions containing the
desired product were pooled and evaporated to afford pure 7
1
to afford 6 (0.22 g, 68%): mp 276 °C; H NMR (DMSO-d6) δ
2.61 (s, 3 H, CH3), 2.79 (t, 2 H, 3′-CH2), 3.18 (t, 2 H, 4′-CH2),
4.01 (s, 2 H, CH2), 4.40 (s, 2 H, CH2), 6.48 (br s, 2 H, 2-NH2),
6.66 (overlapping s, 2 H, 5′-H and 8′-H), 7.46 (s, 2 H, 4-NH2),
8.48 (s, 1 H, H-7).
1
as a yellow solid (0.20 g, 69%): mp 277 °C; H NMR (DMSO-
Anal. (C20H24N6O2‚0.6H2O) C, H, N.
d6) δ 2.61 (s, 3 H, 5-CH3), 3.67 (3, 3 H, 2′-OCH3), 3.77 (s, 3 H,
5′-OCH3), 4.42 (d, 2 H, CH2), 6.20 (m, 2 H, NH and 6′-H), 6.64
(s, 2 H, 2-NH2), 6.98 (m, 5 H, 3′-H and 2′′-, 3′′-, 5′′-, and 6′′-H),
7.47 (s, 2 H, 4-NH2), 8.49 (s, 1 H, H-7).
Gen er a l P r oced u r e for th e Syn th esis of th e P yr -
r olon it r ob en zen es fr om t h e Cor r esp on d in g Nit r o-
a n ilin es. To a solution of the commercially available nitro-
aniline (1.0 equiv) in 30 mL of glacial acetic acid, at reflux,
was added 2,5-dimethoxytetrahydrofuran (1.0 equiv), and the
mixture was heated for 10 min. (Note: For the tetrameth-
ylnitroaniline, the reaction mixture had to be refluxed for 1
h). TLC analyses (hexanes:EtOAc, 1:1) indicated the presence
of a new spot, corresponding to the product. The acetic acid
was evaporated and the resulting residue sonicated in diethyl
ether for 30 min to afford the desired pyrrole as a solid.
2,5-Dim eth oxy-4-p yr r olo-1-n itr oben zen e (18). This was
obtained as a brown solid from 2,5-dimethoxy-4-amino-1-
nitrobenzene (0.44 g, 2.21 mmol) and 2,5-dimethoxytetrahy-
drofuran (0.29 g, 2.21 mmol) in 20 mL of acetic acid (0.50 g,
91%): TLC (hexane:EtOAc, 1:1) Rf ) 0.54; 1H NMR (CDCl3) δ
3.86 (s, 3 H, 5-OCH3), 3.96 (s, 3 H, 2-OCH3), 6.37 (t, 2 H, 3′,4′-
H), 7.02 (s, 1 H, 3-H), 7.09 (d, 2 H, 2′,5′-H), 7.65 (s, 1 H, 6-H).
Anal. (C12H12N2O4‚0.6CH3COOH) C, H, N.
Anal. (C20H23N7O2‚0.9H2O) C, H, N.
2,4-Dia m in o-5-m et h yl-6-[(4′,5′-d im et h oxy-2′-p yr r olo-
a n ilin o)m eth yl]p yr id o[2,3-d ]p yr im id in e (8). To a solution
of the 6-bromomethyl compound 14 (0.20 g, 0.75 mmol) in 30
mL of DMAC was added 4,5-dimethoxy-2-pyrroloaniline (0.21
g, 1.13 mmol), followed by potassium carbonate (0.16 g, 1.13
mmol), and the mixture was stirred for 48 h under an
atmosphere of nitrogen. At the end of this period, the mixture
was filtered and the solvent evaporated under reduced pres-
sure. The residue was dissolved in methanol and 1.0 g of silica
gel added to the solution and the methanol evaporated to afford
a dry plug. This plug was eluted on a column using the same
procedure as described for 7. Fractions containing the desired
product were pooled and evaporated to afford pure 8 as a
brown solid (0.19 g, 66%): TLC (CHCl3:CH3OH, 6:1) Rf ) 0.65;
mp 270 °C; IR (Nujol) 3330, 3140 (NH2) cm-1; 1H NMR (DMSO-
d6) δ 2.63 (s, 3 H, 5-CH3), 3.69 (s, 3 H, 4′-OCH3), 3.77 (s, 3 H,
5′-OCH3), 4.47 (d, 2 H, CH2), 6.21 (m, 2 H, NH and 6′-H), 6.71
(s, 2 H, 2-NH2), 6.99 (s, 1 H, 3′-H), 7.01 (m, 4 H, 2′′,3′′,4′′,5′′-
H), 7.81 (s, 2 H, 4-NH2), 8.51 (s, 1 H, H-7).
4,5-Dim eth oxy-2-p yr r olo-1-n itr oben zen e (19). This was
obtained as a brown solid from 4,5-dimethoxy-2-amino-1-
nitrobenzene (0.53 g, 2.66 mmol) and 2,5-dimethoxytetrahy-
drofuran (0.35 g, 2.66 mmol) in 25 mL of acetic acid (0.56 g,

+
+
Nonclassical 2,4-Diamino-5-methylpyrido[2,3-d]pyrimidines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 4 485
(10) Gangjee, A.; Shi, J .; Queener, S. F.; Barrows, L. R.; Kisliuk, R.
L. Synthesis of 5-Methyl-5-deaza Nonclassical Antifolates as
Inhibitors of Dihydrofolate Reductases and As Potential Anti-
Pneumocystis, Anti-Toxoplasma and Antitumor Agents. J . Med.
Chem. 1993, 36, 3437-3443.
(11) Gangjee, A.; Vasudevan, A.; Queener, S. F.; Kisliuk, R. L.
6-Substituted 2,4-Diamino-5-Methylpyrido[2,3-d]pyrimidines as
Inhibitors of Dihydrofolate Reductases From Pneumocystis cari-
nii and Toxoplasma gondii and as Antitumor Agents. J . Med.
Chem. 1995, 38, 1778-1785.
(12) Gangjee, A.; Vasudevan, A.; Queener, S. F.; Kisliuk, R. L. 2,4-
Diamino-6-Substituted Pyrido[2,3-d]pyrimidine Antifolates as
Potent and Selective Nonclassical Inhibitors of Dihydrofolate
Reductases. J . Med. Chem. 1996, 39, 1448-1456.
(13) Tripos Associates, Inc., 1699 S. Hanley Rd., Suite 303, St. Louis,
MO 63144.
Anal. (C20H23N7O2‚0.5H2O) C, H, N.
2,4-Dia m in o-5-m eth yl-6-[(2′,3′,5′,6′-tetr a m eth yl-4′-p yr -
r oloa n ilin o)m eth yl]p yr id o[2,3-d ]p yr im id in e (9). To a
solution of the 6-bromomethyl compound 14 (0.24 g, 0.89
mmol) in 25 mL of anhydrous DMAC was added 2,3,5,6-
tetramethyl-4-pyrroloaniline (0.29 g, 1.34 mmol) followed by
CsCO3 (0.50 g), and the mixture was stirred under an
atmosphere of nitrogen for 72 h. The reaction mixture was
then filtered, and purification was carried out in the same
manner as described previously for 7 to afford pure 9 as a
brown solid (0.25 g, 69%): TLC (CHCl3:CH3OH, 5:1) Rf ) 0.78;
mp 299 °C dec; IR (Nujol) 3290, 3080 (NH2) cm-1 1H NMR
;
(DMSO-d6) δ 1.81 (s, 6 H, 2′,6′-CH3), 2.20 (s, 6 H, 3′,5′-CH3),
2.68 (s, 3 H, CH3), 4.31 (d, 2 H, CH2), 5.31 (t, 1 H, NH), 6.14-
6.22 (m, 6-H, 2′′,3′′,4′′,5′′-H and 2-NH2), 7.01 (s, 2 H, 4-NH2),
8.48 (s, 1 H, H-7).
(14) Gangjee, A.; Zaveri, N.; Kothare, M.; Queener, S. F. Nonclassical
2,4-Diamino-6-(aminomethyl)-5,6,7,8-tetrahydroquinazoline An-
tifolates: Synthesis and Biological Activities. J . Med. Chem.
1995, 38, 3660-3668.
Anal. (C23H27N7‚1.0 H2O) C, H, N.
2,4-Diam in o-5-m eth yl-6-[(2′-m eth oxy-5′-ph en ylan ilin o)-
m eth yl]p yr id o[2,3-d ]p yr im id in e (10). To a solution of 14
(0.29 g, 1.09 mmol) in 25 mL of anhydrous DMAC was added
2-methoxy-5-phenylaniline (0.33 g, 1.63 mmol) followed by K2-
CO3 (0.22 g, 1.63 mmol). The reaction mixture was stirred
under nitrogen for 48 h, filtered, and purified as described
previously for 7 to afford pure 10 as a yellow solid (0.29 g,
71%): TLC (CHCl3:CH3OH, 5:1) Rf ) 0.84; mp 265 °C dec; IR
(Nujol) 3290, 3080 (NH2) cm-1; 1H NMR (DMSO-d6) δ 2.67 (s,
3 H, CH3), 3.71 (s, 3 H, 2′-OCH3), 4.41 (d, 2 H, CH2), 5.22 (t,
1 H, NH), 6.12 (m, 2 H, 3′-H and 6′-H), 6.22 (s, 2 H, 2-NH2),
6.29 (d, 1 H, 4′-H), 7.12 (s, 2 H, 4-NH2), 7.36 (m, 5 H), 8.47 (s,
1 H, H-7).
(15) Calculated using Log Kow - Estimation of Log Octanol/Water
Partition Coefficient. Version 1.03. Syracuse Research Corpora-
tion.
(16) Rauckman, B. S.; Tidwell, M. Y.; J ohnson, J . V.; Roth, B. 2,4-
Diamino-5-Benzyl pyrimidines and Analogues as Antibacterial
Agents. 10. 2,4-Diamino-5-(6-quinolylmethyl)- and -[(tetrahydro-
6-quinolyl)methyl)]pyrimidine Derivatives. Further Specificity
Studies. J . Med. Chem. 1989, 32, 1927-1935.
(17) Gangjee, A.; Devraj, R.; McGuire, J . J .; Kisliuk, R. L. Effect of
Bridge Region Variation on Antifolate and Antitumor Activity
of Classical 5-Substituted 2,4-Diamino furo[2,3-d]pyrimidines.
J . Med. Chem. 1995, 38, 3798-3805.
(18) Then, R. L.; Hartman, P. G.; Kompis, I.; Santi, D. Selective
Inhibition of Dihydrofolate Reductase from Problem Human
Pathogens. Adv. Exp. Med. Biol. 1993, 38, 533-536
(19) (a) Walzer, P. D.; Foy, J .; Steale, P.; White, M. Synergistic
Combinations of Ro 11-8958 and Other Dihydrofolate Reductase
Inhibitors with Sulfamethoxazole and Dapsone for Therapy of
Experimental Pneumocystis. Antimicrob. Agents Chemother.
1993, 37, 1436-1443. (b) Pascaud, B. M.; Chau, F.; Garry, L.;
J acobus, D.; Derouin, F.; Girard, P. M. Combination of PS-15,
Epiroprim, or Pyrimethamine with Dapsone in Prophylaxis of
Toxoplasma gondii and Pneumocystis carinii Dual Infection in
a Rat Model. Antimicrob. Agents Chemother. 1996, 40, 2067-
2070.
Anal. (C22H22N6O‚0.5H2O) C, H, N.
Ack n ow led gm en t. This work was supported in part
by a grant from the National Institutes of Health, NIH
GM 40998 (A.G.), and NIH Contracts N01-AI-87240
(S.F.Q.) and N01-AI-35171 (S.F.Q.)
Refer en ces
(1) (a) Taken in part from the thesis submitted by A.V. to the
Graduate School of Pharmaceutical Sciences, Duquesne Uni-
versity, in partial fulfillment of the requirements for the degree
of Doctor of Philosophy, J une 1996. (b) Presented in part at the
208th National American Chemical Society Meeting, Chicago,
IL, Aug 23-28, 1993; MEDI 112.
(20) Martinez, A.; Allegra, C. J .; Kovacs, J . A. Efficacy of Epiroprim
(Ro11-8958), a New Dihydrofolate Reductase Inhibitor in the
Treatment of Acute Toxoplasma Infections in Mice. Am. J . Trop.
Med. Hygiene 1996, 54 (3), 249-252.
(21) Piper, J . R.; McCaleb, G. S.; Montgomery, J . A.; Kisliuk, R. L.;
Gaumont, Y.; Sirotnak, F. M. Synthesis and Anitfolate Activity
of 5-Methyl-5-deaza Analogues of Aminopterin, Methotrexate,
Folic Acid and N10-Methyl Folic Acid. J . Med. Chem. 1986, 29,
1080-1087.
(22) Gribble, G. W.; Hoffman, J . H. Reactions of Sodium Borohydride
in Acidic Media;VI. Reduction of Indoles with Cyanoborohydride
in Acetic Acid. Synthesis 1977, 859-860.
(23) Bigham, E. C.; Hodson, S. J .; Mallory, W. R.; Wilson, D.; Duch,
D. S.; Smith, G. K.; Ferone, R. Synthesis and Biological Activity
of Open Chain Analogues of 5,6,7,8-Tetrahydrofolic Acid-
Potential Antitumor Agents. J . Med. Chem. 1992, 35, 1399-
1410.
(2) Mills, J .; Masur, H. AIDS-Related infections. Sci. Am. 1991,
August, 50-57.
(3) Hughes,W. T.; Gray, W. L.; Gutteridge, W. E.; Latter, V. S.;
Pudney, M. Efficacy of a Hydroxynaphthoquinone 566C80, In
Experimental P. carinii Pneumonitis. Antimicrob. Agents.
Chemother. 1990, 34, 225-228.
(4) Sattler, F. R.; Cowan, R.; Nielsen,, D. M.; Ruskin, J . TMP-
Sulfamethoxazole compared with Pentamidine for the Treatment
of P. carinii pneumonia in the Acquired Immunodeficiency
Syndrome; A Prospective, Non-crossover Study. Ann. Intern.
Med. 1988, 109, 280-287.
(5) Dannemann, B.; McCutchan, J . A.; Israelski, D. Treatment of
Toxoplasmic Encephalitis in Patients with AIDS : A Random-
ized Trial Comparing Pyrimethamine Plus Clindamycin to
Pyrimethamine Plus Sulfadiazine. Ann. Intern. Med. 1992, 116,
33-43.
(24) Su, T. L.; Huang, J . T.; Burchenal, J . H.; Watanabe, K. A.; Fox,
J . J . Synthesis and Biological Activities of 5-Deaza Analogues
of Aminopterin and Folic Acid. J . Med. Chem. 1986, 29, 709-
715.
(25) Piper, J . R.; J ohnson, C. A.; Maddry, J . A.; Malik, N. D.; McGuire,
J . J .; Otter, G. M.; Sirotnak, F. M. Studies on Analogues of
Classical Antifolates Bearing the Naphthoyl Group in Place of
Benzoyl in the Side Chain. J . Med. Chem. 1993, 36, 4161-4171.
(26) Chio, L.; Queener, S. F. Identification of Highly Potent and
Selective Inhibitors of Toxoplasma gondii Dihydrofolate Reduc-
tase. Antimicrob. Agents. Chemother. 1993, 37, 1914-1923.
(27) Broughton, L.; Queener, S. F. Pneumocystis carinii Dihydrofolate
Reductase Used to Screen Potential Antipneumocystis Drugs.
Antimicrob. Agents Chemother. 1991, 35, 1348-1355.
(6) Allegra, C. J .; Kovacs, J . A.; Drake, J . C.; Swan, J . C.; Chabner,
B. A.; Masur, H. Potent In Vitro and In Vivo Antitoxoplasma
Activity of the Lipid Soluble Antifolate Trimetrexate. J . Clin.
Invest. 1987, 79, 478-482.
(7) Kovacs, J . A.; Allegra, C. J .; Swan, J . C.; Drake, J . C.; Parrilllo,
J . E.; Chabner, B. A.; Masur, H. Potent Antipneumocystis and
Antitoxoplasma Activities of Piritrexim, A Lipid Soluble Anti-
folate. Antimicrob. Agents Chemother. 1988, 32, 430-433.
(8) News. FDA Approves Trimetrexate As A Second Line Therapy
For Pneumocystis carinii Pneumonia. Am. J . Hosp. Pharm. 1994,
51, 591-592.
(9) Champness, J . N.; Achari, A.; Ballantine, S. P.; Bryant, P. K.;
Delves, C. J .; Stammers, D. K. The Structure of P. carinii DHFR
to 1.9 Å Resolution. Structure 1994, 2, 915-924.
J M960734F