High-affinity inhibitors of dihydrofolate reductase: Antimicrobial and anticancer activities of 7,8-dialkyl-1,3-diaminopyrrolo[3,2-f]quinazolines with small molecular size

Article abstract of DOI:10.1021/jm9505122

A series of 7,8-dialkylpyrrolo[3,2-f]quinazolines were prepared as inhibitors of dihydrofolate reductase (DHFR). On the basis of an apparent inverse relationship between compound size and antifungal activity, the compounds were designed to be relatively small and compact. Inhibitor design was aided by a GRID analysis of the three-dimensional structure of Candida albicans DHFR, which suggested that relatively small, branched alkyl groups at the 7- and 8-positions of the pyrroloquinazoline ring system would provide optimal interactions with a hydrophobic region of the protein. The compounds were potent inhibitors of fungal and human DHFR, with K(i) values as low as 7.1 and 0.1 pM, respectively, and were highly active against C. albicans and an array of tumor cell lines. In contrast to known lipophilic inhibitors of DHFR such as trimetrexate and piritrexim, members of this series of pyrroloquinazolines were not susceptible to P-glycoprotein-mediated multidrug resistance and also showed significant distribution into lung and brain tissue. The compounds were active in lung and brain tumor models and displayed in vivo activity against Pneumocystis carinii and C. albicans.

Full text of DOI:10.1021/jm9505122

892
J . Med. Chem. 1996, 39, 892-903
High -Affin ity In h ibitor s of Dih yd r ofola te Red u cta se: An tim icr obia l a n d
An tica n cer Activities of 7,8-Dia lk yl-1,3-d ia m in op yr r olo[3,2-f]qu in a zolin es w ith
Sm a ll Molecu la r Size
Lee F. Kuyper,*,† David P. Baccanari,† Michael L. J ones,† Robert N. Hunter,† Robert L. Tansik,†
Suzanne S. J oyner,† Christine M. Boytos,† Sharon K. Rudolph,† Vince Knick,† H. Robert Wilson,†
J . Marc Caddell,† Henry S. Friedman,‡ J ohn C. W. Comley,§ and J eremy N. Stables§
Wellcome Research Laboratories, Research Triangle Park, North Carolina 27709, Wellcome Research Laboratories,
Beckenham, Kent BR3 3BS, U.K., and Duke University Medical Center, Durham, North Carolina 27710
Received J uly 13, 1995X
A series of 7,8-dialkylpyrrolo[3,2-f]quinazolines were prepared as inhibitors of dihydrofolate
reductase (DHFR). On the basis of an apparent inverse relationship between compound size
and antifungal activity, the compounds were designed to be relatively small and compact.
Inhibitor design was aided by a GRID analysis of the three-dimensional structure of Candida
albicans DHFR, which suggested that relatively small, branched alkyl groups at the 7- and
8-positions of the pyrroloquinazoline ring system would provide optimal interactions with a
hydrophobic region of the protein. The compounds were potent inhibitors of fungal and human
DHFR, with Ki values as low as 7.1 and 0.1 pM, respectively, and were highly active against
C. albicans and an array of tumor cell lines. In contrast to known lipophilic inhibitors of DHFR
such as trimetrexate and piritrexim, members of this series of pyrroloquinazolines were not
susceptible to P-glycoprotein-mediated multidrug resistance and also showed significant
distribution into lung and brain tissue. The compounds were active in lung and brain tumor
models and displayed in vivo activity against Pneumocystis carinii and C. albicans.
In tr od u ction
The enzyme dihydrofolate reductase (DHFR) is a
known target for drug action.1,2 Inhibitors of DHFR
have proven useful in the treatment of cancer,3-6
bacterial infections,7 malaria,8 and Pneumocystis carinii
pneumonia (PCP).9,10 Efforts to develop new therapies
based on DHFR inhibition continue. One area of recent
focus is the search for inhibitors of DHFR from various
The lipophilic inhibitors metoprine (DDMP, 2), piri-
trexim (PTX, 3), and trimetrexate (TMX, 4) were
designed to avoid the limitations of MTX. DDMP, for
example, concentrates in the brain relative to plasma
and was evaluated as a potential treatment for brain
tumors.22 Both PTX and TMX are active against
transport-impaired MTX-resistant cell lines and show
tissue distribution profiles significantly different from
that of MTX.23,24 In addition, compounds PTX and TMX
also exhibit clinical activity against P. carinii, an
opportunistic fungal organism afflicting many AIDS
patients.25-28 As in many other fungal and bacterial
species, folates are synthesized de novo by P. carinii,
and the organism lacks the ability to take up folates
from its environment. Thus, PTX and TMX can be
given in combination with folinic acid (leucovorin),
which is taken up only by the host cells and counteracts
the antifolate toxicity of the drugs.9,10
opportunistic organisms such as the protozoan parasite
11-13
Toxoplasma gondii
and the fungus Candida albi-
cans.14 The importance of opportunistic diseases has
risen considerably in recent years, owing to the large
increase in the population of immunocompromised
patients associated with the AIDS (acquired immune
deficiency syndrome) epidemic, organ transplantation,
and cancer chemotherapy.15-17 Therapeutic interven-
tion is limited by the dearth of safe and effective
antiparasitic and antifungal agents.17
Another area of recent interest is the development of
novel lipophilic DHFR inhibitors for neoplastic disease.4
Although methotrexate (MTX, 1) is effective against
acute lymphocytic leukemia, non-Hodgkin’s lymphoma,
and osteosarcoma,3,18 the hydrophilic nature of MTX
restricts its distribution to various body tissues, such
as lung and brain,18 and prevents entry into cells by
diffusion. Thus, the antitumor activity of MTX relies
on the active transport of the drug into cancer cells,19,20
and resistance to MTX can arise from impairment of
this transport mechanism.21
* To whom correspondence should be addressed. Current address:
Glaxo Wellcome Inc., Five Moore Dr., Research Triangle Park, NC
27709.
† Wellcome Research Laboratories, NC.
‡ Duke University Medical Center.
§ Wellcome Research Laboratories, U.K.
X Abstract published in Advance ACS Abstracts, J anuary 15, 1996.
0022-2623/96/1839-0892$12.00/0 © 1996 American Chemical Society

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 893
Sch em e 1a
Sch em e 2a
a
Reagents and conditions: (a) NaN(CN)2, 1-octanol, reflux, 13
h; (b) isopentyl tosylate, NaH, DMF.
This paper describes a series of lipophilic DHFR
inhibitors that showed antimicrobial and anticancer
activities and potentially advantageous tissue distribu-
tion properties. The compounds arose from an effort to
identify DHFR inhibitors with useful activity against
the fungus C. albicans. Initial attempts to identify lead
compounds from our corporate collection of DHFR
inhibitors were disappointing. Although a number of
compounds showed significant inhibition of C. albicans
DHFR (e.g., PTX and TMX), few displayed activity
against C. albicans in vitro. Inconsistencies between
the level of enzyme inhibition and antimicrobial activity
were observed with other classes of inhibitors as well.
As one working hypothesis, we speculated that antimi-
crobial activity might be inversely related to the mo-
lecular size of the inhibitor. That assumption led us to
design a series of 7,8-dialkylpyrrolo[3,2-f]quinazolines
5 that were relatively small and compact but showed
high affinity for DHFR. Inhibitor design was based on
the three-dimensional structure of C. albicans DHFR
and aided by the computer program GRID.29 The
resulting compounds were potent inhibitors of bacterial,
fungal, protozoal, and human cells. Compared to PTX,
TMX, and MTX, the new compounds displayed (1)
substantially greater activity against human and fungal
DHFR, (2) more effective growth inhibition of a number
of tumor and fungal cell lines, (3) activity against cells
with P-glycoprotein-mediated multidrug resistance, (4)
superior tissue distribution properties for lung and brain
tumor applications, (5) activity in murine models for
brain and lung tumors, and (6) superior in vivo activity
against P. carinii and C. albicans.
a
Reagents and conditions: (a) H2NR, DMSO, 50 °C; (b) ketone,
NaBH3CN, HCl, CH3OH; (c) ICl; (d) alkyne, CuI, Et3N, DMF,
Pd(PPh3)2Cl2; (e) CuI, DMF, reflux; (f) ROTs, NaH, DMF; (g)
HNO3, H2SO4; (h) H2, Pd/C; (i) NaN(CN)2, DMF; (j) refluxing
diglyme or BF3‚etherate, DME.
11 using sodium hydride in DMF and the desired alkyl
tosylate, followed by regioselective nitration, or (2)
copper-catalyzed cyclization of 2-alkynyl-4-nitroanilines
8 (Table 2). The latter compounds were derived from
the 4-nitro-N-alkylanilines 6 (Table 2), which were
prepared from 4-nitroaniline via reductive alkylation32,33
or from 4-fluoronitrobenzene by aromatic nucleophilic
substitution with the appropriate alkylamine.34,35 Io-
dination of the N-alkylanilines 6 was effected with ICl
in methanol.36,37 Introduction of the alkynyl moiety
employed palladium catalysis,38 and cyclization of the
alkynylanilines 8 to the indoles 9 was performed with
copper iodide.39,40 Conversion of the nitroindoles 9 to
the aminoindoles 13 was performed in high yield using
catalytic hydrogenation.
Construction of the diaminopyrimidine ring followed
our previously reported procedures for the preparation
of the parent ring system.31 The aminoindoles 13 were
condensed with dicyanamide to give the cyanoguanidine
derivatives 14 (Table 3), and ring closure to the target
compounds 5 (Table 4) was carried out thermally or with
boron trifluoride etherate. Synthesis of tert-butyl-
substituted analogues 5f,l,p ,r was incompatible with
the procedure employing BF3‚etherate but was effected
successfully using thermal conditions.
Ch em istr y
The preparation of 1,3-diamino-7H-pyrrolo[3,2-f]-
quinazolines was originally reported by Ledig30 and
involved a thermally induced condensation of 5-ami-
noindole and dicyanamide. 7-Substituted analogues
were prepared by alkylation of the parent pyrrolo-
quinazoline or the intermediate indole. Our initial
preparation of compound 5i, for example, employed
analogous procedures, alkylating with isopentyl tosylate
as illustrated in Scheme 1. However, the low yields of
the alkylation reactions and the need to explore a
variety of 7,8-dialkylated analogues prompted us to
develop alternative procedures. We also devised im-
proved conditions for effecting the cyclization using
boron trifluoride etherate.31
Resu lts a n d Discu ssion
In h ibitor Design . Hundreds of inhibitors of DHFR
have been reported during the past several decades,41
and most of the compounds that show high affinity for
DHFR appear to make use of the large hydrophobic cleft
observed in X-ray crystal structures of the enzyme from
various species.2,42-44 For example, PTX, a potential
treatment for PCP,10 binds to P. carinii DHFR with its
dimethoxybenzyl group positioned within the large
hydrophobic cavity of that enzyme.45 As shown in
Figure 1, the benzyl group of the inhibitor is surrounded
by the side chains of Leu-25, Ile-33, Phe-36, Ile-65, Pro-
66, Phe-69, and Leu-72.
Scheme 2 outlines procedures employed in our im-
proved synthesis of compounds 5. The key 2,3-dialkyl-
5-nitroindole intermediates 9 (Table 1) were obtained
by one of two methods: (1) N-alkylation of 2-alkylindoles
Of particular pertinence to the work described here
is a series of 7-substituted-pyrrolo[3,2-f]quinazolines

894 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Ta ble 1. Substituted Indolesa
Kuyper et al.
first reported by Ledig.30 Those compounds are inhibi-
tors of mammalian and bacterial DHFR46 and show in
vitro activity against the C. albicans organism.47 Mo-
lecular modeling suggested that this class of inhibitor
binds to DHFR with the 7-substituent occupying the
large hydrophobic cleft in a manner similar to the
dimethoxybenzyl group of PTX; those compounds with
larger 7-substituents generally displayed higher affinity
for C. albicans DHFR. For example, compound 5u was
comparable to PTX and TMX in its inhibition of the
fungal enzyme (see Table 5). However, the inhibitors
with the larger substituents were relatively ineffective
inhibitors of C. albicans growth. Thus the challenge
was to reduce inhibitor size and yet retain tight enzyme
binding, and the design problem was to identify small
substituents for the pyrroloquinazoline heterocycle that
would impart high affinity for DHFR. The relatively
potent activity reported for 7-(cyclopropylmethyl)-2,4-
diaminopyrrolo[3,2-f]quinazoline suggested that small
alkyl substituents can impart high affinity for DHFR46
and effective inhibition of C. albicans cell growth.47
Further exploration of such small alkyl substituents was
aided by the use of GRID, a computer program that
evaluates the interaction between a protein active site
and a variety of probe functions.29 A crystal structure
of the C. albicans holoenzyme was furnished by Whitlow
and co-workers,48 and a methyl group probe was used
in GRID to analyze the binding site. As illustrated in
Figure 2, regions of favorable interaction energy for a
methyl group, formed by Met-25, Ile-33, Phe-36, Met-
54, Ile-62, and Leu-69, were adjacent to the 7- and
8-positions of enzyme-bound pyrroloquinazoline. The
area next to the 8-position was relatively small and
centered in the plane of the inhibitor’s ring system.
Small, branched alkyl groups at the 8-position appeared
to take good advantage of that site. On the other hand,
the 7-position of the pyrroloquinazoline was positioned
at the edge of the large hydrophobic cavity of the
protein, and favorable binding locations for the methyl
probe were found close to the parent inhibitor but above
and below the plane of the pyrroloquinazoline ring
system. Substituents such as the 3-pentyl group that
would adopt conformations that are perpendicular to the
pyrroloquinazoline ring system, as shown in Figure 2,
were required to position methyl groups into regions
yieldb
compd
X
R′
R
mp (°C)
(%)
formula
9c
9d
9e
9f
NO2 Me
NO2 Et
NO2 i-Pr
NO2 t-Bu
NO2 Et
Me
Me
H
H
Et
128-129
127-128
151-152
134-136
84-85
79 C10H10N2O2
38 C11H12N2O2
73 C11H12N2O2
86 C12H14N2O2
42 C12H14N2O2‚
0.1H2O
9g
9h
9j
9k
9l
NO2 Me
NO2 i-Pr Et
NO2 n-Pr Et
i-Pr
156-158
70-72
63 C12H14N2O2
45 C13H16N2O2
65 C13H16N2O2
66 C14H18N2O2
74 C14H18N2O2
81 C14H18N2O2
31 C14H18N2O2
95 C15H20N2O2
92 C15H20N2O‚
0.1H2O
67-68
NO2 Et
t-Bu
121-123
9m NO2 Me
C(Me)2Et 66-68
CH(Me)Et 63-64
CHEt2
9n
9o
9p
9q
NO2 Et
NO2 Me
NO2 t-Bu i-Pr
NO2 Et
69-71
125-126
oil
CHEt2
9r
9s
9t
NO2 n-Bu t-Bu
NO2 n-Pr CHEt2
NO2 i-Pr CHEt2
90-91
55-56
oil
71 C16H22N2O2
98 C16H22N2O2
92 C16H22N2O2
11e
11k
12d
12g
H
H
H
H
i-Pr
n-Pr
Et
H
H
Me
Et
72-73
oil
oil
69 C11H13
N
N
N
81
60
46
C
C
C
11H13
11H13
12H15N‚
Et
oil
0.2hexane
12h
12j
H
H
H
H
Me
i-Pr Et
n-Pr Et
i-Pr
oil
oil
oil
oil
26
78
90
C
C
C
12H15
13H17
13H17
N
N
N
N
12k
12o
13c
13d
13f
13g
13h
Me
CHEt2
50 C14H19
92 C10H12N2‚HCl
NH2 Me
NH2 Et
NH2 t-Bu
NH2 Et
NH2 Me
Me
Me
H
Et
i-Pr
>250
220-221 100 C11H14N2‚HCl
>250
175-176
100 C12H16N2‚HCl
93 C12H16N2‚HCl
230-231 100 C12H16N2‚
0.25EtOH‚
0.2HCl
13j
13k
13o
NH2 i-Pr Et
NH2 n-Pr Et
230-231
185-186
175-176
96 C13H18N2‚HCl
98 C13H18N2‚HCl
96 C14H20N2‚HCl
NH2 Me
CHEt2
a
All compounds were characterized by 1H NMR, mass spectra,
and elemental analyses. Other compounds in these series were
not purified for analytical purposes and were characterized only
b
by 1H NMR spectra. Yields were not optimized.
F igu r e 1. Stereoview of the active site region of the X-ray crystal structure of the P. carinii DHFR-NADPH-piritrexim complex.
The protein backbone is represented as a dark pink tube, and and atoms of selected residues are shown. Atoms of piritrexim and
NADPH are colored by atom type: carbon, white; nitrogen, cyan; oxygen, red; and phosphorus, orange.

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 895
Ta ble 2. Substituted Anilinesa
compd
X
Y
R
mp (°C)
yieldb (%)
formula
C9H12N2O2
6h
6l
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
H
H
H
H
H
H
I
I
I
I
I
i-Pr
t-Bu
82-84
68-69
oil
oil
oil
52-53
115-116
oil
oil
oil
131-132
oil
oil
oil
oil
oil
oil
oil
92
84
67
89
83
67
71
81
76
71
89
95
98
85
99
90
94
91
C10H14N2O2
C11H16N2O2
C10H14N2O2
C11H16N2O2
C9H11N2O2I
C10H13N2O2I
C11H15N2O2I
C10H13N2O2I
C11H15N2O2I
C14H18N2O2
C14H18N2O2
C14H18N2O2
C14H18N2O2‚0.2H2O
C15H20N2O2
C16H22N2O2
6m
6n
6o
7h
7l
7m
7n
7o
8l
8m
8n
8o
8q
8r
C(Me)2Et
CH(Me)Et
CHEt2
i-Pr
t-Bu
C(Me)2Et
CH(Me)Et
CHEt2
t-Bu
C(Me)2Et
CH(Me)Et
CHEt2
CHEt2
t-Bu
-CtC-Et
-CtC-Me
-CtC-Et
-CtC-Me
-CtC-Et
-CtC-Bu
-CtC-i-Pr
-CtC-n-Pr
10e
10k
H
H
C11H13
11H13
N
N
H
C
a
All compounds were characterized by 1H NMR, mass spectra, and elemental analyses. Other compounds in these series were not
purified for analytical purposes and were characterized only by 1H NMR spectra. Yields were not optimized.
b
Ta ble 3. N-Cyano-N′-(1,2-dialkylindol-5-yl)guanidinesa
substantially with additional alkyl substitution at either
the 7- or 8-position. The compound with the lowest
measured Ki value (5m , Ki 7.1 pM) was 3200-fold more
active than its parent compound 5a and more than 260-
fold more active than PTX and TMX. Alkyl substitution
at the 8-position of the pyrroloquinazoline ring system
enhanced activity significantly. An 8-methyl group
provided a 12-fold increase in affinity for the fungal
enzyme (5c versus 5b), and isopropyl substitution
increased activity more than 140-fold (5e versus 5a ).
Tertiary butyl and n-propyl groups also fit into the
fungal enzyme binding pocket (compounds 5f,k ,p ,s), but,
as suggested by the molecular modeling experiments,
the binding pocket imposed a substituent size limit at
the 8-position. The n-butyl group appeared to encounter
those limits; compound 5r was significantly less active
than the corresponding ethyl-substituted analogue 5l.
The enzyme active site imposed fewer restrictions on
size for substituents at the 7-position of the pyrrolo-
quinazoline ring system, as exemplified by the tri-
methoxybenzyl group of compound 5u (and also the
large substituents of PTX and TMX). However, the
GRID analysis suggested that smaller 7-substituents
with appropriate shape could contribute significantly to
binding. This was verified by compound 5i; its 3-pentyl
group provided a contribution to enzyme affinity (Ki 0.22
nM) equivalent to that of the much larger trimethoxy-
benzyl group of compound 5u (Ki 0.23 nM). A combina-
tion of small, compact alkyl substituents at the 7- and
8-positions provided inhibitors with exceptional affinity
for the fungal enzyme (compounds 5l-q,s,t).
compd
R′
R
mp (°C)
yield (%)b
formula
14c
14i
Me
H
n-Pr
Et
Me
Et
n-Bu t-Bu
Me
215-216
68-69
195-196
89
47
86
57
97
37
67
C12H13N5
C15H19N5
C15H19N5
C16H21N5
C16H21N5
C17H23N5
C18H25N5
CHEt2
Et
CH(Me)Et 135-137
CHEt2
CHEt2
14k
14n
14o
14q
14r
151-152
145-147
127-128
a
1
All compounds were characterized by elemental analyses, H
NMR, and mass spectra. Other compounds in this series were
not purified for analytical purposes and were characterized only
by 1H NMR spectra. Yield was not optimized.
b
indicated by the GRID analysis. On the basis of this
simple analysis, the series of 7,8-dialkylpyrrolo[3,2-f]-
quinazolines listed in Table 4 were prepared.
In Vitr o Biologica l Activities. The series of 7,8-
dialkylpyrrolo[3,2-f]quinazolines 5a -u were evaluated
as inhibitors of C. albicans and human DHFR and for
growth inhibition of fungal and human cells. As shown
in Table 5, several members of the series displayed
exceptional affinity for DHFR and potent inhibition of
cell growth. DHFR Ki values were as low as 7.1 pM
against C. albicans DHFR and 0.1 pM for the human
enzyme. Compound concentrations effective against
whole cells ranged as low as 1 ng/mL for C. albicans
and 0.57 nM for the human colon tumor cell line HCT-
8. Potent activity was also observed against bacteria.
For example, compound 5o inhibited Escherichia coli
with a minimum inhibitory concentration (MIC) of 0.1
ng/mL.
The effects of 7- and 8-alkyl substituents on affinity
for human DHFR were similar to those observed for the
fungal enzyme. The most active pyrroloquinazoline
inhibitors, compounds 5l-q,s,t, displayed Ki values in
the sub-picomolar range and were significantly more
active than PTX and TMX. For reference purposes, the
most potent pyrroloquinazolines were also 1 order of
magnitude more active than MTX (Ki 1.3 pM), the
bench-mark hydrophilic anticancer agent.3 No evidence
for substitution patterns that might selectively favor
In this series of pyrroloquinazolines, only the parent
compound 5a and its 7-methyl derivative 5b were less
active than PTX and TMX as inhibitors of C. albicans
DHFR. Activity against the fungal enzyme increased

896 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Ta ble 4. 7,8-Disubstituted-1,3-diaminopyrrolo[3,2-f]quinazolinesa
Kuyper et al.
compd
R′
R
mp (°C)
yield (%)b
formula
5a c
5bc
5c
5d
5e
5f
5g
5h
5i
5j
5k
5l
5m
5n
5o
5p
5q
5r
H
H
H
Me
Me
Me
H
H
Et
i-Pr
264-266
248-250
303-305
64
18
38
57
20
21
36
42
47
20
41
65
28
37
33
65
60
23
64
30
46
C10H9N5
C11H11N5‚0.6H2O
Me
Et
i-Pr
t-Bu
Et
Me
H
i-Pr
n-Pr
Et
Me
Et
Me
t-Bu
Et
n-Bu
n-Pr
i-Pr
H
C
C
12H13N5‚0.5H2O
13H15N5
>250
>250
>250
227-229
C13H15N5‚0.2EtOAc‚0.3H2O
C
C
C
14H17N5‚0.5H2O
14H17N5‚0.95H2O
14H17N5‚0.7H2O
150-160
197-198
217-219
239-241
200-202
254-256
199-200
200-201
286-287
155-157
233-234
121-122
243-244
242-243
CHEt2
Et
Et
C15H19N5‚0.9H2O
15H19N5
C15H19N5‚0.1H2O
16H21N5
C
t-Bu
C(Me)2Et
CH(Me)Et
CHEt2
i-Pr
CHEt2
t-Bu
CHEt2
CHEt2
C
C16H21N5
C16H21N5‚0.7H2O
C16H21H5‚0.25H2O
C
17H23N5‚0.2H2O
C17H23N5‚H2O
C18H25N5‚TFA
5s
C18H25N5
5t
C18H25N5‚0.5H2O
C20H21N5O3‚0.5H2O
5u c
3,4,5-trimethoxybenzyl
a
b
All compounds were characterized by 1H NMR, mass spectra, and elemental analyses. Yield was not optimized. c Originally reported
in ref 30.
Ta ble 5. Biological Data for 7,8-Disubstituted-1,3-diaminopyrrolo[3,2-f]quinazolines
C. albicans
human
C. albicans
MIC (µg/mL)
compd
R′
R
MW
DHFR Ki (nM)a
DHFR Ki (pM)a
HCT-8 IC50 (nM)
PTX
TMX
5a
5b
5c
5d
5e
5f
5g
5h
5i
5j
5k
5l
5m
5n
5o
5p
5q
5r
5s
5t
5u
325
369
199
213
227
241
241
255
255
255
269
269
269
283
283
283
283
297
297
311
311
311
379
1.9
1.9
23
25
2.0
1.3
0.16
0.12*
0.33
0.22
25*
1.4*
1000
230
87
29
10
<2.0
10
6.4
4.5
<2.0
24
0.1*
0.4*
0.1*
0.3*
0.2*
0.2*
30
0.4*
0.1*
6.0*
>10
>50
0.8
68
1.8
1450
580
85
H
H
H
Me
Me
Me
H
H
Et
i-Pr
1.6
0.1
0.1
0.05
0.05
0.2
0.025
0.001
0.1
Me
Et
i-Pr
t-Bu
Et
Me
H
i-Pr
n-Pr
Et
Me
Et
Me
t-Bu
Et
n-Bu
n-Pr
i-Pr
H
8.0
3.3
0.59
7.4
6.0
1.4
3.1
1.3
2.1
0.82
0.74
1.3
0.57
52
CHEt2
Et
Et
0.22
0.33
0.65
0.1
t-Bu
C(Me)2Et
CH(Me)Et
CHEt2
i-Pr
CHEt2
t-Bu
CHEt2
CHEt2
<0.05
0.0071*
<0.05
0.030*
<0.06
<0.06
0.38
<0.11
<0.05
0.23
>0.1
0.025
0.1
0.025
0.05
0.1
>0.1
>0.1
>0.1
5.0
4.4
0.75
3,4,5-trimethoxybenzyl
a
Values marked with an asterisk were measured directly. Otherwise, values were calculated from the IC50 value.
inhibition of the fungal enzyme versus human DHFR
emerged from this series of pyrroloquinazolines.
which exhibited a fungal enzyme Ki essentially identical
with that of analogue 5u , was 5000-fold more active
than 5u against the organism in vitro. The majority of
pyrroloquinazolines showed C. albicans MICs within a
narrow range (0.025-0.1 µg/mL), but those compounds
had fungal DHFR Ki values that differed up to 280-fold.
The working hypothesis that antifungal activity might
be related to the molecular size of the inhibitor was not
A number of the alkyl-substituted pyrroloquinazolines
were potent inhibitors of C. albicans growth. However,
the level of growth inhibition did not correlate with
DHFR Ki. For example, the trimethoxybenzyl-substi-
tuted pyrroloquinazoline 5u weakly inhibited C. albi-
cans growth (MIC 5 µg/mL). In contrast, compound 5i,

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 897
F igu r e 2. Stereoview of the active site region of the X-ray crystal structure of the C. albicans DHFR-NADPH-compound 5i
complex. The green mesh represents the -1.5 kcal/mol isoenergy contour surface from GRID. The surface was calculated using
coordinates of the holoenzyme and a methyl group probe. Molecule representation is as described in Figure 1.
Ta ble 6. Cytotoxicities of Compounds 5i,o Compared to MTX,
Ta ble 7. Comparative IC50 Values against Normal and
PTX, and TMX (IC50, nM)
Multidrug-Resistant Cell Lines
cell line
5i
5o
MTX
PTX
TMX
relative IC50 for compounda
P388D1
L cells
D54
4.5
2.2
0.12
0.6
4.6
0.8
2.1
0.4
0.5
2.3
2.3
1.0
4.8
8.3
16
8.8
22
21
76
230
2.7
4.7
60
cell line
VBL
ADR PTX TMX 5i 5o
KB3-1
KBV-1
1
1
61
1
16
1
20
1
1.1
1
1
2000
143B(TK-)
U87MG
A549
KBV-1 + reserpine
1.3
1
17500
1.6
1.5
1.5 1.1 0.8
8
10
48
24
260
70
55
30
11
5.3
MCF7
1
1
1
1
1
31
MCF7/ADR
MCF7/ADR + reserpine
200
1.5
1
6.8
0.8
1
8.6 0.9 0.9
1.2 0.7 0.6
H460
Daoy
U373MG
Vero
9.5
9.0
12
0.5
1
19
0.2
80
12
9.6
16
P388D1
1
1
P388/ADR
P388/ADR + reserpine
28
3.7
3.7
0.3
8.9
0.3
0.8
0.9
9.2
a
The IC50 values were normalized to that observed for the
parent sensitive cell line for each compound.
supported by our observations; analysis of data from
almost 100 pyrroloquinazolines did not reveal any
distinct relationships between activity and inhibitor size
and/or lipophilicity.
inhibitor MTX is not.53 In contrast to PTX and TMX,
compounds 5i,o were equally effective against both the
parent and MDR cell lines, and their cytotoxicity was
not affected by reserpine (Table 7). Since acquired MDR
and normal expression of P-glycoprotein in tumors
derived from tissues such as colon, small intestine,
kidney, and liver are important problems in cancer
therapy, a compound such as 5o could have broad
therapeutic applications.
Unlike the poor correlation between C. albicans
DHFR Ki and C. albicans MIC, the activities of the
compounds in Table 5 against HCT-8 tumor cells were
highly correlated with human DHFR Ki values (r2
)
0.97). The most potent pyrroloquinazoline inhibitor of
HCT-8 cells, compound 5q, was 3-fold more active than
TMX.
The small molecular size of 5i,o may play a role in
their lack of susceptibility to P-glycoprotein-mediated
MDR. The degree of susceptibility to MDR was quan-
titatively related to molecular weight for a series of
unrelated anticancer agents by Selassie et al.54 More-
over, the high activity of compounds 5o,i against C.
albicans may be linked to their poor activity as sub-
strates for the mammalian P-glycoprotein efflux pump.
A relative of C. albicans, Saccharomyces cerevisiae,
expresses a protein with high homology to human
P-glycoprotein55 that mediates the transmembrane
transport of an endogenous peptide.56 The presence of
a P-glycoprotein in a fungal organism suggests the
possibility that fungi might also use such proteins for
protection against cytotoxic agents. The MDR suscep-
tibilities and activities against C. albicans of PTX, TMX,
and compounds 5i,o were consistent with this idea.
Another possible benefit for compounds not affected
by MDR involves tissue distribution. P-Glycoprotein is
found in normal tissues such as liver, kidney, and colon
and also is expressed in endothelial cells at the blood-
Compounds 5i,o were subjected to further in vitro and
in vivo studies. In comparison to PTX, TMX, and MTX,
compound 5o showed superior activity against an array
of human and murine cell lines, as shown in Table 6.
The average cytotoxicity of compound 5o was ap-
proximately 10-fold greater than that of MTX and TMX
and 50-fold greater than that of PTX.
Su scep tibility to Mu ltip le Dr u g Resista n ce. Cell
lines KBV-1, MCF7/ADR, and P388/ADR, which express
the multiple drug resistance phenotype (MDR), are
resistant to adriamycin (ADR), vinblastine (VBL), and
a number of structurally unrelated antitumor agents.
The resistance stems from the action of a membrane-
bound glycoprotein (P-glycoprotein) that actively ef-
fluxes cytotoxic agents by an energy-dependent mech-
anism.49,50 Reserpine is a reversing agent that blocks
the efflux activity of P-glycoprotein and restores sensi-
tivity of MDR cells to ADR and VBL.51 The lipophilic
DHFR inhibitors PTX and TMX also are susceptible to
the MDR phenotype,52 whereas the hydrophilic DHFR

898 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Kuyper et al.
Ta ble 8. Tissue Distribution of Compounds 5i,o Compared to
MTX, DDMP, PTX, and TMX
6), the reason for inactivity in these brain tumor models
is not known.
relative level compared to plasma
In vivo studies with compound 5o were extended to
lung tumor models. Therapy with compound 5o (10 mg/
kg, sc) reduced the number of lung tumor nodules by
68% with no significant toxicity in mice with intratho-
racic (it) implants of human A549 non-small-cell lung
carcinoma (NSCLC). The activity of compound 5o
approached that of mitomycin C (4 mg/kg, nodule
reduction 92%), whereas MTX (2 mg/kg) was inactive
in this model. However, compound 5o did not increase
the life span of mice with it implants of A549 cells or
another NSCLC line (NCI H460).
compd
brain
10
10
0.006
0.65
0.1
lung
5ia
57
50
0.05
5.5
5oa
MTXb
PTXa
TMXa
DDMPc
6.5
31
a
Studies were performed in mice. At time of tissue collection,
plasma drug concentration of 5i was 0.4 µg/mL, compound 5o was
b
0.4 µg/mL, PTX was 3.5 µg/mL, and TMX was 9.4 µg/mL. Rabbits
received high-dose MTX (70 mg/kg) by infusion; steady-state
plasma MTX concentration was 160 µg/mL.79 c Rats received 1
mg/kg DDMP po; tissues were collected 5 h postdose; plasma
DDMP concentration was 0.6 µg/mL.22
An tifu n ga l a n d An tip r otozoa l Stu d ies. Com-
pound 5o was a potent, though nonselective, inhibitor
of DHFR from T. gondii and P. carinii. The inhibitory
potency of compound 5o against T. gondii DHFR (Ki 0.9
nM) was 300-fold greater than that of pyrimethamine
(Ki 300 nM). Pyrimethamine, in combination with
sulfonamides, is clinically effective in the treatment of
toxoplasmosis.61,62 In an in vitro T. gondii plaque
reduction assay, compound 5o (MIC 0.00025 µg/mL) was
2000-fold more potent than pyrimethamine (MIC 0.5 µg/
mL) (R. Berens, University of Colorado Health Sciences
Center). In comparison, the MIC value of TMX was 0.01
µg/mL. Leucovorin (25 µM) did not reverse the in vitro
activity of compound 5o against T. gondii.
brain barrier.57 Lack of drug distribution into these
tissues, especially the brain, is a common limitation for
treatment of tumors and fungal infections, and drug
distribution properties may, in part, be related to
susceptibility to the P-glycoprotein efflux pump. PTX
amd TMX do not distribute significantly into the brain.58
However, smaller DHFR inhibitors with similar lipo-
philicity, such as pyrimethamine and metoprine, do
achieve substantial concentrations in brain tissue and
are not substrates for P-glycoprotein.58 Tissue distribu-
tion properties of compounds 5i,o are described below.
Tissu e Distr ibu tion Stu d ies. Therapy with MTX,
PTX, and TMX is limited by the tissue distribution
properties of these compounds. They do not readily
cross the blood-brain barrier and do not attain high
lung concentrations relative to plasma (Table 8). Com-
pounds 5i,o exhibited greatly improved brain/plasma
and lung/plasma concentration ratios and, thus, may
be useful against lung and brain tumors. DDMP also
concentrated in the brain and lung. However, clinical
trials with DDMP revealed central nervous system
(CNS), cutaneous, and gut toxicities59 that may be
Compound 5o was a potent inhibitor of P. carinii
DHFR (IC50 14 nM), comparable to PTX (IC50 32 nM)
and TMX (IC50 66 nM) and considerably more active
than trimethoprim (IC50 24 µM),63 three agents that are
clinically active against PCP.9,10,64,65 PTX and TMX, like
compound 5o, are not selective for P. carinii DHFR, and
leucovorin must be coadministered to prevent antifolate-
related toxicity in patients. We compared the activity
of compound 5o/leucovorin to that of PTX/leucovorin and
trimethoprim/sulfamethoxazole (TMP/SMX) against PCP
in SCID mice (Table 10). PTX/leucovorin had no effect
on the extent of the infection, whereas significant
reductions in the intensity of the infection were seen
with both TMP/SMX and compound 5o/leucovorin. The
superior activity of compound 5o compared to PTX may
reflect the enhanced distribution of compound 5o to
lungs.
related to inhibition of histamine metabolism.60
A
therapeutic dose of compound 5o (10 mg/kg, sc) did not
elevate brain histamine above control levels (20 ng/mL)
in mice, whereas DDMP (10 mg/kg, ip) elevated brain
histamine to 100 ng/mL for 15 h postdose.
In Vivo An titu m or Stu d ies. The high brain/plasma
and lung/plasma concentration ratios of compounds 5i,o
and their in vitro activity against MDR-expressing cell
lines suggested the compounds may be active against
brain, lung, and MDR tumors in vivo.
Several therapy experiments were performed with
P388 leukemia cells in mice. MTX, PTX, ADR, and
compounds 5i,o produced significant increases in life
span in mice with ip implants of P388 cells (Table 9).
Compound 5o, when administered ip, showed antitumor
activity in mice with intraperitoneal implants of MDR
cell line P388/ADR comparable to that in mice with
implants of the parent P388 tumor line. The sensitivity
of the p388/ADR implants to compound 5o was reduced
in a second experiment in which the compound was
given sc.
Compounds 5i,o were more active than MTX or PTX
against intracranially (ic) implanted P388. However,
compound 5o was inactive when tested against ic
implants of other tumor lines, including D54 glioma,
U87MG glioma, and 143B(TK-) osteosarcoma cells.
Since these tumor cell lines were sensitive to inhibition
by compound 5o in in vitro cytotoxicity assays (Table
Compounds 5i,o were potent nonselective inhibitors
of C. albicans DHFR and the C. albicans organism in
vitro, with MICs of 0.001 and 0.025 µg/mL, respectively
(Table 5). These compounds were evaluated in murine
models of Candida nephritis with concomitant menin-
gitis. The infection level in the brain 48 h after
inoculation with C. albicans was ca. 105 colony forming
units (cfu)/g of tissue in vehicle- or SMX-treated control
mice (Table 11). Treatment with compound 5o reduced
brain infection levels 10-fold, consistent with the ob-
servation that 5o distributed to brain. Compound 5o
also was efficacious against candidal nephritis. In
vehicle- or SMX-treated mice, greater than 107 cfu/g
were recovered from kidney tissue. Treatment with
compound 5o alone led to a 7200-fold reduction of fungi
in the kidney, and in mice treated with the combination
of compound 5o and SMX, infection levels were 44 000-
fold lower than those of control mice. Compound 5i also
was active in this model but to a lesser degree.
In conclusion, the relatively small, lipophilic DHFR
inhibitors described here showed a variety of interesting
properties in comparison to known inhibitors. In con-

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 899
Ta ble 9. Effect of Compounds 5i,o and MTX, PTX, and ADR on Life Span of Mice with Intraperitoneally and Intracranially
Implanted P388 Leukemic Cells
increase in life span (%)
compd
P388 (ipa)
P388/ADR (ipa)
P388 (ica)
5i (40 mg/kg, sc)
5o (10 mg/kg, sc)
5o (12.5 mg/kg, sc)
5o (10 mg/kg, ip)
ADR
80 (n ) 8)
40-55b (n ) 22)
27-50c (n ) 32)
60 (n ) 8)
80 (n ) 8)
90 (n ) 8)
20 (n ) 8)
30 (n ) 8)
38 (n ) 5)
50 (n ) 5)
69-130d (n ) 13)
40 (n ) 8)
8-30c (n ) 13)
-5 (n ) 8)
130 (n ) 8)
PTX
MTX
0 (n ) 10)
115 (n ) 8)
0-15d (n ) 14)
a
b
Implant site: ip, intraperitoneal; ic, intracranial. Results from three separate experiments. c Results from four separate experiments.
d
Results from two separate experiments.
Ta ble 10. Comparison of Compound 5o, PTX, and
Trimethoprin against P. carinii Pneumonia in SCID Mice
(triphenylphosphine)palladium(II) chloride, and 35 mg (0.18
mmol) of CuI in 250 mL of triethylamine was added 7.0 mL
(70 mmol) of 1-pentyne. The mixture was stirred under
nitrogen at room temperature for 48 h. Solvent was removed
in vacuo, and the residue was taken up in 400 mL of diethyl
ether. The mixture was filtered through Celite, and the filtrate
was dried over MgSO4. Removal of solvent on a rotary
evaporator left a dark residue, which was subjected to flash
chromatography (silica gel, 95:5 hexanes:ethyl acetate) to give
5.8 g (91%) of compound 10k as an oil: 1H NMR (CDCl3) 7.25
(m, 1H), 7.10 (m, 1H), 6.65 (m, 2H), 3.90 (br s, 2H), 2.45 (t,
2H, J ) 7 Hz), 1.65 (dt, 2H, J ) 7 Hz), 1.05 (t, 3H, J ) 7 Hz);
MS (CI) M + 1, 160 (100). Anal. (C11H13N) C, H, N.
2-P r op ylin d ole (11k ). A mixture of 3.8 g (24 mmol) of
compound 10k , 15 mg (7.9 mmol) of CuI, and 100 mL of DMF
was kept at reflux under nitrogen for 4 h. The mixture was
stirred at room temperature for 16 h and concentrated to
dryness. The residue was taken up in diethyl ether and
filtered through Celite. The filtrate was subjected to flash
chromatography (silica gel, 49:1 hexanes:ethyl acetate) to
furnish 3.0 g (81%) of 11k as an amber oil: 1H NMR (CDCl3)
7.80 (br s, 1H), 7.55 (m, 1H), 7.30 (m, 1H), 7.3 (m, 1H), 7.2-
7.0 (m, 2H), 6.25 (br s, 1H), 2.7 (t, J ) 7 Hz, 2H), 1.75 (dt, J
) 7, 7 Hz, 2H), 1.05 (t, J ) 7 Hz, 3H); MS (CI) M + 1, 160
(100). Anal. (C11H13N) C, H, N.
N-Eth yl-2-p r op ylin d ole (12k ). To a mixture of 2.4 g (0.1
mol) of NaH (97%) in 100 mL of DMF was added 8.0 g (0.050
mmol) of compound 11k . The mixture was stirred at room
temperature for 0.5 h, and 8.0 mL (0.11 mol) of bromoethane
was added dropwise. The mixture was stirred for 1 h and
concentrated on a rotary evaporator. Methanol was added
cautiously to the residue, and the mixture was concentrated
to dryness. The residue was partitioned between diethyl ether
and water. The ether layer was washed with water and dried
over MgSO4. Removal of solvent on a rotary evaporator left
8.49 g (90%) of 12k as an amber oil: 1H NMR (CDCl3) 7.60 (d,
J ) 7 Hz, 1H), 7.35 (d, J ) 7 Hz, 1H), 7.25-7.10 (m, 2H), 6.30
(s, 1H), 4.20 (q, J ) 6 Hz, 2H), 2.75 (t, J ) 6 Hz, 2H), 1.85 (m,
2H), 1.40 (t, J ) 6 Hz, 3H), 1.1 (t, J ) 6 Hz, 3H); MS (CI) M
+ 1, 188 (100). Anal. (C13H17N) C, H, N.
infection scorea
mean SEMb infected/totalc significanced
no. of mice
statistical
treatment
D5W ip and sc (control) 3.0
0.13
0.26
0.11
0.13
0.17
11/11
11/11
2/11
11/11
8/10
leucovorin
3.0
5o + leucovorin
PTX + leucovorin
TMP + SMX
0.18
3.27
0.9
B, C
A
B
a
Infection scores: 0 ) no infection evident; 1 ) very weak
infection; 2 ) mild infection; 3 ) moderate infection; 4 ) heavy
b
infection. Standard error of mean. c Total number of mice at end
d
of experiment. A, not significantly different (P < 0.05) from
results for the controls and leucovorin alone; B, significantly
different (P < 0.05) from results for the controls; C, significantly
different (P < 0.05) from results for TMP + SMX.
Ta ble 11. Effect of Compounds 5i,o on C. albicans Nephritis
and Meningitis in Immunosuppressed Mice
log cfu/g of tissuea
compd
brain
kidney
vehicle control
5i
5.1 ( 0.2
nd
7.1 ( 0.1
6.0
5o
SMX
5i + SMX
5o + SMX
3.6 ( 0.3
3.4
7.3 ( 0.1
4.6
4.7
nd
3.4
2.5 ( 0.1
a
Values are averages of duplicate experiments, with the range
indicated. Those values without a range are from single experi-
ments.
trast to PTX and TMX, the new compounds were potent
inhibitors of C. albicans cell growth, were not suscep-
tible to P-glycoprotein-mediated MDR, distributed sig-
nificantly into lung and brain tissue, were active in
murine models for lung and brain tumors, and showed
superior in vivo activity against P. carinii and C.
albicans.
N-Eth yl-2-p r op yl-5-n itr oin d ole (9k ). To an ice bath-
cooled solution of 8.0 g (43 mmol) of 12k in 450 mL of H2SO4
was added dropwise a solution of 4.0 g (47 mmol) of NaNO2 in
50 mL of H2SO4. The solution was stirred at 0 °C for 5 h and
at room temperature for 12 h. The mixture was poured onto
ice and stirred for 0.75 h. The resulting precipitate was
collected by filtration and dissolved in ethyl acetate. The
organic solution was washed with saturated NaHCO3 and
water and dried over MgSO4. Solvent was removed in vacuo,
and the residue was subjected to flash chromatography (silica
gel, 9:1 hexanes:ethyl acetate) to provide 6.40 g (65%) of 9k
as a white solid: mp 67-68 °C; 1H NMR (DMSO-d6) 8.4 (s,
1H), 7.95 (dd, J ) 8, 3 Hz, 1H), 7.6 (d, J ) 8 Hz, 1H), 6.5 (s,
1H), 4.2 (q, J ) 7 Hz, 2H), 2.7 (t, J ) 7 Hz, 2H), 1.7 (dt, J )
7, 7 Hz, 2H), 1.25 (t, J ) 7 Hz, 3H), 1.0 (t, J ) 7 Hz, 3H); MS
M + 1, 233 (100). Anal. (C13H16N2O3). C, H, N.
Exp er im en ta l Section
Ch em istr y. 1H NMR spectra were recorded on Varian XL-
200 and XL-300 spectrometers. Chemical shifts are in parts
per million (δ), relative to the observed solvent resonance
(DMSO, 2.50). Mass spectra were determined by Oneida
Research Services (Whitesboro, NY) on a Finnegan 4500
instrument. Analytical samples of intermediates moved as
single spots on TLC (Whatman MK6F silica gel plates).
Column chromatography was carried out on silica gel 60 (E.
Merck, Darmstadt, Germany). Melting points were deter-
mined with a Thomas-Hoover melting point apparatus and are
uncorrected. Analyses were performed by Atlantic Microlab,
Inc., Atlanta, GA, and all values are within 0.4% of theory.
Octanol/water partition coefficients were measured by Mid-
western Research Institute. The pKa of compound 5o was
determined using ultraviolet methods as described for related
compounds.66 A sample of TMX was generously provided by
the National Cancer Institute.
1,3-Diam in o-7-(1-eth ylpr opyl)-7H-pyr r olo[3,2-f]qu in azo-
lin e (5i). A suspension of 180 g (2.04 mol) of 3-pentyl alcohol,
1 L of methylene chloride, 600 mL of pyridine, and 307 g (1.62
mol) of p-toluenesulfonyl chloride was stirred for 24 h.
A
2-(1-P en tyn yl)a n ilin e (10k ). To a mixture of 8.8 g (40
mmol) of 2-iodoaniline (Aldrich), 150 mg (0.21 mmol) of bis-
solution of 300 mL of concentrated hydrochloric acid in 1 L of
water was cautiously added to the reaction mixture. The

900 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Kuyper et al.
mixture was stirred for 2 h, and the methylene chloride layer
was separated. The water layer was extracted with diethyl
ether, and the extractions were combined with the methylene
chloride layer. The volume of the organic solution was reduced
to 500 mL on a rotary evaporator and washed with 0.1 N
hydrochloric acid until the aqueous layer remained acidic. The
organic layer was then washed thoroughly with water, dried
(MgSO4), and filtered, and the solvent was removed in vacuo
to give 230 g (60%) of 1-ethylpropyl 4-toluenesulfonate as a
white solid: mp 44-45 °C; HPLC (Nova Pak C18, 70% MeOH/
H2O/0.1% Et3N/0.1% TFA) K′ 2.67; 1H NMR (CDCl3) 7.8 (d, J
) 7 Hz, 2H), 7.3 (d, J ) 7 Hz, 2H), 4.45 (p, J ) 5 Hz, 1H), 2.4
(s, 3H), 1.6 (m, 4H), 0.8 (t, J ) 7 Hz, 6H). Anal. (C12H18O3S)
C, H.
solution of 4.0 g (12 mmol) of compound 7o in 150 mL of
triethylamine in a Parr pressure reaction vessel. To this
solution was added 0.030 g (0.16 mmol) of CuI and 0.10 g (0.14
mmol) of bis(triphenylphosphine)palladium(II) chloride, and
the pressure vessel was then sealed. The mixture was stirred
at room temperature for 20 h. The solvent was removed in
vacuo, and the residue was suspended on silica gel. The silica
gel suspension of the residue was placed on a silica gel column
and eluted with 10% ethyl acetate in hexanes to provide 2.5 g
(85%) of compound 8o as a yellow oil: 1H NMR (DMSO-d6)
7.98 (s, 1H), 7.90 (s, 1H), 6.75 (d, J ) 9 Hz, 1H), 5.8 (d, J ) 9
Hz, 1H), 3.5 (m, 1H), 2.1 (s, 3H), 1.5 (m, 4H), 0.8 (t, J ) 7 Hz,
6H). Anal. (C14H18N2O2‚0.2H2O) C, H, N.
5-Nitr o-2-m eth yl-1-(1-eth ylp r op yl)in d ole (9o). A sus-
pension of 2.5 g (10 mmol) of compound 8o and 0.20 g (0.11
mmol) of copper iodide in 200 mL of DMF was refluxed for 5
h. Solvent was removed in vacuo to leave a dark residue,
which was taken up in methanol. Silica gel was added, and
solvent was removed in vacuo. The silica gel suspension of
the residue was placed on a silica gel column and eluted with
5% ethyl acetate in hexane to furnish 2.25 g (90%) of compound
9o as an oil: 1H NMR (DMSO-d6) 8.4 (m, 1H), 7.9 (br d, 1H),
7.7 (d, 1H), 6.5 (br s, 1H), 4.2 (m, 1H), 1.8-2.2 (m, 4H), 0.6 (t,
6H). Anal. (C14H18N2O2) C, H, N.
5-Am in o-2-m eth yl-1-(1-eth ylp r op yl)in d ole Hyd r och lo-
r id e (13o). A mixture of compound 9o (2.5 g, 10 mmol) in
100 mL of methanol and 10% palladium on carbon (0.3 g) was
subjected to hydrogenation on a Parr apparatus (40 psi).
When uptake of hydrogen ceased (ca. 2 h), the reaction mixture
was filtered through Celite and 12 N HCl (3 mL) was added
to the filtrate. The filtrate was concentrated to dryness. The
residue was taken up in ethanol, and the solution was again
concentrated to dryness. This sequence was repeated several
times to remove excess HCl and left 2.5 g (97%) of compound
13o as a white solid: mp 175 °C dec; HPLC (Nova Pak C18,
60% MeOH/H2O, 0.1% Et3N/0.1% TFA) K′ 2.50; 1H NMR
(DMSO-d6) 10.1 (s, 3H), 7.6 (d, J ) 9 Hz, 1H), 7.4 (d, J ) 2
Hz, 1H), 6.9 (dd, J ) 2, 9 Hz, 1H), 6.3 (s, 1H), 4.1 (br s, 1H),
2.4 (s, 3H), 1.8-2.2 (m, 4H), 0.62 (t, J ) 7 Hz, 6H). Anal.
(C14H20N2‚HCl) C, H, N, Cl.
N-Cyan o-N′-(2-m eth yl-1-(1-eth ylpr opyl)in dol-5-yl)gu an i-
d in e (14o). A suspension of compound 13o (2.4 g, 9.5 mmol)
and sodium dicyanamide (2.5 g, 28 mmol) in 125 mL of DMF
was heated to 50 °C for 4 h. Solvent was removed in vacuo,
water (ca. 250 mL) was added, and the mixture was stirred
for 1 h. Filtration furnished 2.7 g (100%) of compound 14o as
a tan solid: mp 151-152 °C; 1H NMR (DMSO-d6) 8.85 (s, 1H),
7.44 (d, J ) 9 Hz, 1H), 7.34 (s, 1H), 6.84 (d, J ) 9 Hz, 1H),
6.80 (s, 2H), 6.17 (s, 1H), 4.05 (m, 1H), 2.38 (s, 3H), 1.79-2.19
(m, 4H), 0.64 (t, 6H). Anal. (C16H21N5) C, H, N.
1,3-Dia m in o-7-(1-et h ylp r op yl)-8-m et h yl-7H -p yr r olo-
[3,2-f]qu in a zolin e (5o). Meth od A. A mixture of 278 g (0.98
mol) of compound 14o and 3.5 L of diglyme was kept at reflux
for 20.5 h. The mixture was cooled slowly to 30 °C and filtered.
The filter cake was washed with warm diglyme, and the
combined filtrates were subjected to flash chromatography on
silica gel, eluting with ethyl acetate:ethanol, 9:1. Fractions
containing product were combined and concentrated to a thick
slurry which was stirred with cooling (ice/water bath) for 1 h.
The slurry was filtered, and the off-white filter cake was
washed with cold ethanol and dried in vacuo at 45 °C
overnight. A slurry of the solid in 1.2 L of 1 N NaOH was
stirred overnight, and the solid was isolated by filtration.
Recrystallization from DMSO, followed by a thorough water
wash, provided 46.2 g (17%) of compound 5o as a white solid:
mp 198-199 °C; 1H NMR (DMSO-d6) 7.76 (d, J ) 9.2 Hz, 1H),
6.90 (d, J ) 9.2 Hz, 1H), 6.78 (s, 1H), 6.58 (s, 2H), 5.64 (s,
2H), 4.10 (m, 1H), 2.42 (s, 3H), 2.10 (m, 2H), 1.88 (m, 2H),
0.62 (t, J ) 7.2 Hz, 6H); MS (CI) M + 1 (100). Anal
(C16H21N5‚0.7H2O) C, H, N.
A suspension of 18.0 g (89.8 mmol) of compound 5a 31 in 800
mL of dry DMF was stirred under nitrogen while 4.0 g (170
mmol) of ca. 97% sodium hydride was added. The mixture
was stirred for 1 h, and a solution of 24 g (99 mmol) of
1-ethylpropyl 4-toluenesulfonate in 80 mL of dry DMF was
added dropwise over
a period of 30 min. Stirring was
continued for 8 h, and solvent was removed in vacuo. The dark
residue was dissolved in 300 mL boiling methanol. The
solution was treated with charcoal and filtered through Celite.
The resulting dark solution was boiled down to 200 mL and
allowed to cool. The resulting crystalline solid was isolated
by filtration and washed with two 50-mL portions of cold
methanol, sonicated with 200 mL of 0.1 N sodium hydroxide,
and filtered. The solid was then washed with two 50-mL
portions of water and dried in vacuo to afford 12.5 g (52%) of
compound 5i as an off-white solid: mp 197-198 °C; 1H NMR
(DMSO-d6) 7.80 (d, J ) 9 Hz, 1H), 7.51 (d, J ) 3 Hz, 1H), 7.07
(d, J ) 3 Hz, 1H), 7.02 (d, J ) 9 Hz, 1H), 6.62 (br s, 2H), 5.62
(s, 2H), 4.32 (m, 1H), 1.85 (m, 4H), 0.64 (t, J ) 7 Hz, 6H); λmax
(0.1 N HCl) 342 (ꢀ 11 208), 230 (ꢀ 26 874), 257 sh (ꢀ 18 339)
nm; λmin 284 (ꢀ 846), 219 (ꢀ 22 396) nm; HPLC (Nova Pak C18,
30% MeOH/H2O with 0.1% TEA, 0.1% TFA) tR 5.38 min. Anal.
(C15H19N5‚H2O) C, H, N.
N-(1-Eth ylp r op yl)-4-n itr oa n ilin e (6o). A solution of
4-nitroaniline (13.8 g, 0.099 mol) and 12 N HCl (10 mL) in
200 mL of methanol was treated with a solution of 3-pentanone
(14.5 mL, 0.144 mol) in methanol (15 mL) and stirred for 1 h
at room temperature. The reaction mixture was then cooled
in an ice bath, and a solution of sodium cyanoborohydride (8.7
g, 0.138 mol) in methanol (35 mL) was added at such a rate
that the reaction temperature remained below 20 °C. The
reaction mixture was stirred at room temperature for 2 h and
basified with 10% NaOH (70 mL). The mixture was partially
concentrated in vacuo to remove most of the methanol and
diluted to 350 mL total volume with water. The aqueous
solution was extracted with two 250-mL portions of ether, and
the combined extracts were dried (MgSO4) and concentrated
to give an oil. Chromatography on silica gel (40% hexane/CH2-
Cl2) provided 17.3 g (84%) of compound 6o as an orange oil:
1H NMR (CDCl3) 8.06 (d, J ) 9 Hz, 2H), 6.50 (d, J ) 9 Hz,
2H), 4.20-4.30 (br s, 1H), 3.30-3.40 (m, 1H), 1.40-1.70 (m,
4H), 0.94 (t, J ) 7 Hz, 6H). Anal. (C11H16N2O2) C, H, N.
N-(1-Eth ylp r op yl)-2-iod o-4-n itr oa n ilin e (7o). A mixture
of compound 6o (10.4 g, 0.0499 mol) and 20 mL of concentrated
HCl in 150 mL of H2O was heated to 50 °C and treated
dropwise over 1 h with a solution of ICl (17 g, 0.1047 mol) in
concentrated HCl (30 mL). The mixture was stirred at 50 °C
for 2.5 h, cooled in an ice bath, and treated with Na2SO4 (10
g). The solution was stirred for 10 min at room temperature,
diluted with 125 mL of diethyl ether, and stirred an additional
20 min. The aqueous layer was extracted with ether (100 mL),
and the combined organic extracts were washed with saturated
NaHCO3 solution (100 mL), dried (MgSO4), and concentrated
in vacuo to leave an oil. The oil was subjected to flash
chromatography on silica gel (33% hexane/CH2Cl2) to give
15.59 g of an orange oil. Distillation (bp0.25 ) 170-172 °C)
provided 11.86 g (71%) of compound 7o as a yellow oil: 1H
HMR (CDCl3) 8.57 (d, J ) 3 Hz, 1H), 8.10 (dd, J ) 3, 9 Hz,
1H), 6.47 (d, J ) 9 Hz, 1H), 4.80 (br d, 1H), 3.40 (m, 1H), 1.50-
1.75 (m, 4H), 0.96 (t, J ) 7 Hz, 6H). Anal. (C11H15N2O2I) C,
H, N.
Meth od B. A suspension of compound 14o (2.6 g, 9.2 mmol)
in 300 mL of methanol was placed in a Parr pressure reactor
and heated to 150 °C for 35 h. The reaction mixture was
cooled, and solvent was removed in vacuo. The residue was
subjected to chromatography on silica gel, eluting with 10%
EtOH in CH2Cl2. The resulting solid was sonicated with 200
mL of 0.1 N NaOH for 1 h. The mixture was filtered, and the
N-(1-Eth ylpr opyl)-2-pr opyn yl-4-n itr oan ilin e (8o). 1-Pro-
pyne (2.7 mL, 48 mmol) was condensed and added to a cold

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 901
solid was dried in vacuo to yield 0.87 g (33%) of compound 5o
as an off-white solid: mp 200-201 °C. Anal. (C16H21N5‚0.25
H2O) C, H, N.
In experiments measuring reversal of multiple drug resistance,
5 mM reserpine was used as a reversing agent. Cytotoxicity
measurements were carried out in 96-well microtiter plates
using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT) dye reduction assay72 for suspension cultures
or the sulforhodamine B assay73 for monolayer cultures. Cell
lines A549 (human non-small-cell lung carcinoma), Daoy
(human medulloblastoma), U87MG and U373MG (human
glioblastomas), HCT-8 (human ileocecal adenocarcinoma),
143B(TK-) (thymidine kinase-deficient human osteosarcoma),
Vero (African green monkey kidney), and P388D1 (mouse
lymphoid neoplasm) and mouse L cells were obtained from
ATCC. Human large cell lung carcinoma NCI H460 was from
the Southern Research Institute. P388/ADR was obtained
from D. W. Fry at Parke-Davis Warner-Lambert. The vin-
blastine-resistant KBV-1 and its parent line KB3-1 (human
epidermoid carcinoma) were obtained from the laboratory of
M. Gottesman, National Institute of Health (via A. C. King,
Wellcome Research Laboratories). The adriamycin-resistant
MCF7/ADR10 was from the laboratory of R. Fine (Veterans
Administration Medical Center, Durham, NC); the sensitive
MCF7 cell line, conditioned to folate-free media, was from D.
Duch (Wellcome Research Laboratories).
Meth od C. To a mixture of 510 g (1.80 mol) of compound
14o and 2.5 L of 1,2-dimethoxyethane under nitrogen was
added 664 mL (766 g, 5.4 mol) of BF3‚Et2O while the temper-
ature was maintained at e45 °C (ice/water bath). The mixture
was stirred overnight at room temperature and then concen-
trated on a rotary evaporator. The residual dark oil was
dissolved in a minimum amount of methanol and added to 5
L of NH4OH. The resulting slurry was stirred vigorously
overnight. The solid was isolated by filtration, washed with
water, and dried in vacuo at 40 °C overnight. This crude
material was subjected to flash chromatography on silica gel,
eluting with 9:1 ethyl acetate:ethanol containing 1% NH4OH.
The purified material was slurried overnight in 1 N NaOH
(1.5 L) and filtered. The filter cake was washed with water
and dried in vacuo at 50 °C. Recrystallization from DMSO,
followed by extensive washing with water, provided 184 g
(36%) of compound 5o as an off-white solid: mp 188-189 °C.
Anal. (C16H21N5‚0.3H2O‚0.1DMSO) C, H, N, S. The measured
log P (octanol/water) of compound 5o was 3.0. The pKa of
compound 5o was 8.1, and its log D measured at pH 7.4 was
2.4.
The methanesulfonate salt of compound 5o, which had
significantly greater water solubility than the free base, was
prepared as follows. A solution of 3.0 g (10 mmol) of compound
5o in 100 mL of methanol was added dropwise to a solution of
0.74 mL (1.1 g, 11 mmol) of methanesulfonic acid in 100 mL
of methanol. The solution was stirred at room temperature
for 1 h, and solvent was removed in vacuo. The light brown
residue was triturated with diethyl ether and recrystallized
twice from 5:1 diethyl ether:methanol to furnish 2.85 g (73%)
of the methanesulfonate salt of compound 5o as a tan solid:
mp 157-158 °C; 1H NMR (DMSO-d6) 12.2 (s, 1H), 8.8 (s, 1H),
8.1 (d, J ) 8.8 Hz, 1H), 7.6 (br s, 1H), 7.5 (br s, 2H), 7.1 (m,
2H), 4.2 (m, 1H), 2.4 (s, 3H), 1.8-2.2 (m, 4H), 0.6 (t, J ) 7.5
Hz, 6H). Anal. (C17H25N5SO3) C, H, N, S.
Biology. Dihydrofolate reductase from C. albicans was
prepared as previously described14 and assayed in 0.1 M
imidazole chloride buffer (pH 6.4), with 70 µM NADPH and
45 µM dihydrofolate in a final volume of 1 mL at 30 °C.
Recombinant human DHFR (from Anatrace, Maumee, OH)
was assayed in 50 mM sodium phosphate, pH 7.0. For
determinations of IC50 values (the concentration of inhibitor
necessary to inhibit enzymatic activity by 50%), the enzyme,
NADPH, and varying concentrations of inhibitor were prein-
cubated for 2 min, and the reaction was initiated by dihydro-
folate. Steady-state velocities were measured, and plots of
logarithm of inhibitor concentration versus percent inhibition
were used to estimate IC50 values; coefficients of variation for
IC50 values were <10%. For weak-binding compounds, Ki
values were calculated from IC50 using Cha’s equation for
competitive inhibitors.67 The Henderson method68 as described
by Baccanari and J oyner69 was used for determining Ki values
for tight-binding inhibitors; coefficients of variation between
replicate assays were <20%. Dihydrofolate Km values for C.
albicans and human DHFR are 2.7 and 0.036 µM, respec-
tively.14,70 Compounds were tested as inhibitors of dihydro-
folate reductase from P. carinii as previously described.63 T.
gondii DHFR inhibition was measured under the conditions
used for P. carinii DHFR, except that substrate concentrations
were 52 µM dihydrofolate and 73 µM NADPH.
T. gondii cell-free extracts were prepared from frozen T.
gondii tachyzoites (stripped of human fibroblasts) supplied by
Dr. R. Berens of the University of Colorado.71 Briefly, 2.1 ×
109 organisms were suspended in 1.5 mL of 50 mM PIPES
(pH 6.8) and 0.1 M KCl at 4 °C and lysed with six 10-s bursts
from a Vibracell sonifier (Sonics & Materials, Inc.). After
centrifugation for 60 min at 39000g at 4 °C, the supernatant
was collected and loaded onto a Sephadex G-25 column (1 × 8
cm). Fractions containing DHFR activity (eluting near the
void volume of the column) were pooled and stored at -70 °C.
For growth inhibition assays, mammalian cells were adapted
to and maintained in folic acid-free RPMI-1640 (GIBCO BRL)
containing 10 nM (6R,S)-5-formyltetrahydrofolic acid and 10%
fetal bovine serum (HyClone) that had been charcoal-dialyzed.
The in vitro T. gondii plaque reduction assay was performed
as described by Ou-Yang et al.74
Intrathoracic tumor cell implants were performed in nude
mice (NCr-nu/nu or Swiss) essentially as described by McLem-
ore and co-workers.75 Approximately 1 × 106 cells/mouse were
implanted in anesthetized mice on day 0. Mice (10/group) were
treated with compound 5o (10 mg/kg) dissolved in 5% dextrose/
water and administered sc daily on days 1-10, the maximum
tolerated dose of MTX (2 mg/kg) dissolved in 2% NaHCO3 and
administered ip daily on days 1-9, or mitomycin C (4 mg/kg)
dissolved in saline and administered ip on days 1, 5, and 9. In
experiments with A549 NSCLC, mice were sacrificed on day
40, and antitumor activity was assessed by comparing the
number of tumor foci on the lungs and thoracic cavity of control
mice (36 and 48 foci/mouse in two separate experiments) to
those of treated mice. In experiments with intrathoracic
implants of other tumor cell lines, the median survival of drug-
treated animals was compared to that of saline-treated
controls.
For the D54 glioma, U87MG glioma, and 143B(TK-) os-
teosarcoma brain tumor models, 1-2 × 106 cells (5-10 µL)
were implanted intracranially essentially as described by
Wang et al.76 Mice were treated intraperitoneally with 5%
dextrose in distilled water (D5W) or compound 5o (10-15 mg/
kg) in D5W qd × 10. Efficacy was measured by increase in
life span, compared to controls.
The ic and ip studies with P388 utilized viral antibody-free
CDF-1 female mice (18-20 g, 5 or 8/group) from Charles River
(Portage, MI). P388, a murine leukemia cell line, was main-
tained in vivo by limited serial passage. Tumor cells were
implanted intraperitoneally (1 × 106 cells) or intracranially
(1 × 105 cells), and treatment was begun 24 h later. Com-
pounds 5i (40 mg/kg) and 5o (12.5 mg/kg) were suspended in
0.5% carboxymethyl cellulose plus 1% Tween 80 and admin-
istered sc or ip qd × 10. MTX (4 mg/kg) was given ip qd × 9,
PTX (40 mg/kg) ip q7h × 2 on days 1-5; and ADR (5 mg/kg)
ip on days 1, 5, and 9.
Full details of the PCP model in SCID mice were recently
reported.77,78 Briefly, each drug or control group consisted of
10-11 mice. Drugs were evaluated for prophylaxis by admin-
istration once a day from day 1 postinfection for a total of 42
doses. Compound 5o (20 mg/kg) and PTX (12 mg/kg, the
estimated maximum tolerated dose) were dissolved in D5W and
administered ip followed 30 min later by leucovorin (in D5W)
dosed sc 12 mg/kg. Controls received D5W ip and sc. Trime-
thoprim (50 mg/kg) and sulfamethoxazole (250 mg/kg) suspen-
sion was given orally. Mice received 0.1 mL of drug suspension/
20 g of body weight. Twenty-four hours after the last drug
dose, the presence of P. carinii in lung impression smears was
rapidly and unambiguously identified by immunofluorescence.
The intensity of the PCP was graded by scanning the impres-
sion smears and assigning, on a semilogarithmic basis, an
infection score described in a footnote to Table 10. Results
are presented as the calculated mean infection scores ((stan-

902 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4
Kuyper et al.
(10) Falloon, J .; Allegra, C.; Kovacs, J .; O’Neill, D.; Ogata-Arakaki,
D.; Feuerstein, I.; Polis, M.; Davey, R.; Lane, H. C.; LaFon, S.;
Rogers, M.; Zunich, K.; Zurlo, J .; Tuazon, C.; Parenti, D.; Simon,
G.; Masur, H. Piritrexim with leucovorin for the treatment of
pneumocystis pneumonia (PCP) in AIDS patients. Clin. Res.
1990, 38, 361A.
(11) Rosowsky, A.; Hynes, J . B.; Queener, S. F. Structure-activity
and structure-selectivity studies on diaminoquinazolines and
other inhibitors of Pneumocystis carinii and Toxoplasma gondii
dihydrofolate reductase. Antimicrob. Agents Chemother. 1995,
39, 79-86.
(12) Kovacs, J . A.; Allegra, C. J .; Masur, H. Characterization of
dihydrofolate reductase of Pneumocystis carinii and Toxoplasma
gondii. Exp. Parasitol. 1990, 71, 60-68.
(13) Derouin, F.; Chastang, C. In vitro effects of folate inhibitors on
Toxoplasma gondii. Antimicrob. Agents Chemother. 1989, 33,
1753-1759.
(14) Baccanari, D. P.; Tansik, R. L.; J oyner, S. S.; Fling, M. E.; Smith,
P. L.; Freishheim, J . H. Characterization of Candida albicans
dihydrofolate reductase. J . Biol. Chem. 1989, 264, 1100-1107.
(15) Edwards, J . E. Invasive Candida infections. Evolution of a fungal
pathogen. N. Engl. J . Med. 1991, 324, 1060-1062.
(16) Leoung, G., Mills, J ., Eds. Opportunistic Infections in Patients
with the Acquired Immunodeficiency Syndrome; Marcel Dek-
ker: New York, 1989.
dard error of mean, SEM) and the ratio of the number of mice
infected with P. carinii over the total number of mice remain-
ing in each group at the end of the experiment. Since the data
did not follow a pattern of normal distribution, a nonpara-
metric test (Mann-Whitney U test) was used to compare the
infection score between selected individual groups in the same
drug study.
In vivo antifungal activity was assayed in a model of
Candida nephritis. Nephritis was established in immunosup-
pressed CD-1 female mice (10/group) by inoculation of 5 × 104
cfu/mouse into the lateral tail vein. Immunosuppression was
maintained by ip injection of 160 and 40 mg/kg cyclophospha-
mide on days -4 and -1, prior to inoculation, respectively.
Treatment with compounds 5i,o was by subcutaneous injection
of 20 or 50 mg/kg, respectively, in 0.1 mL of sesame oil at 3, 7,
24, and 30 h, postinoculation. When used, sulfamethoxazole
(200 mg/kg, ip, in D5W) was administered 1 h prior to each
compound dose. Efficacy was measured by culture of the
kidneys and brain at 48 h postinoculation. After the tissues
were ground, diluted in saline, plated, and incubated for 48
h, colonies were counted and the cfu/g of tissue was deter-
mined.
(17) Sternberg, S. The emerging fungal threat. Science 1994, 266,
1632-1634.
Tissue distribution of DHFR inhibitors was determined in
male CD-1 mice (5/group). Compounds 5i,o (60 mg/kg) were
administered sc, and tissues were harvested at 3 and 1 h
postdose, respectively. PTX (200 mg/kg) and TMX (100 mg/
kg) were administered ip, and tissues were harvested 40 min
postdose. At time of sacrifice, blood was drawn, the plasma
was separated and frozen, and brains and lungs were removed
and stored at -20 °C. Tissues were homogenized in 0.1 N HCl
(in methanol) and centrifuged at 12000g for 30 min at 4 °C.
The supernatants and plasma samples were prepared for
analysis by solid phase extraction on C2 BondElut columns
(Analytichem International). Drug concentrations were de-
termined by HPLC on a Waters C18 µBondaPak column in
0.1 M ammonium acetate buffer at pH 4.0 with a 0-60%
acetonitrile gradient. Brain histamine levels were measured
with a histamine radioimmunoassay kit from Immunotech
International.
(18) J olivet, J .; Cowan, K. H.; Curt, G. A.; Clendeninn, N. J .; Chabner,
B. A. The pharmacology and clinical use of methotrexate. N.
Engl. J . Med. 1983, 309, 1094-1104.
(19) Dembo, M.; Sirotnak, F. M. Membrane transport of folate
compounds in mammalian cells. In Folate Antagonists as
Therapeutic Agents; Sirotnak, J . M., Burchall, J . J ., Ensminger,
W. B., Montgomery, J . A., Eds.; Academic Press: New York,
1984; Vol. 1, pp 173-217.
(20) Henderson, G. B. Transport of folate compounds into cells. In
Nutritional, Pharmacologic and Physiologic Aspects of Folates
and Pterins; Blakeley, R. I., Whitehead, M. V., Eds.; Wiley: New
York, 1986; pp 207-250.
(21) Sirotnak, F. M. Determinants of resistance to antifolates:
Biochemical phenotypes, their frequency of occurence and
circumvention. Natl. Cancer Inst. Monogr. 1987, 5, 27-35.
(22) Cavallito, J . C.; Nichol, C. A.; Brenckman, W. D., J r.; Deangelis,
R. L.; Stickney, D. R.; Simmons, W. S.; Sigel, C. W. Lipid-soluble
inhibitors of dihydrofolate reductase I. Kinetics, tissue distribu-
tion, and extent of metabolism of pyrimethamine, metoprine, and
etoprine in the rat, dog, and man. Drug Metab. Dispos. 1978, 6,
329-337.
(23) Sigel, C. W.; Macklin, A. W.; Woolley, J . L., J r.; J ohnson, N. W.;
Collier, M. A.; Blum, M. R.; Clendenin, N. J .; Everitt, B. J . M.;
Grebe, G.; Mackars, A.; Foss, R. G.; Duch, D. S.; Bowers, S. W.;
Nichol, C. A. Preclinical biochemical pharmacology and toxicol-
ogy of piritrexim, a lipophilic inhibitor of dihydrofolate reductase.
Natl. Cancer Instit. Monogr. 1987, 5, 111-120.
(24) Lin, J . T.; Bertino, J . R. Update on trimetrexate, a folate
antagonist with antineoplastic and antiprotozoal properties.
Cancer Invest. 1991, 9, 159-172.
(25) Fulton, B.; Wagstaff, A. J .; McTavish, D. Trimetrexate. A review
of its pharmacodynamic and pharmacokinetic properties and
therapeutic potential in the treatment of Pneumocystis carinii
pneumonia. Drugs 1995, 49, 563-576.
(26) Kovacs, J . A.; Allegra, C. J .; Swan, J . C.; Drake, J . C.; Parrillo,
J . E.; Chabner, B. A.; Masur, H. Potent antipneumocystis and
antitoxoplasma activities of piritrexim, a lipid-soluble antifolate.
Antimicrob. Agents Chemother. 1988, 32, 430-433.
(27) Queener, S. F.; Bartlett, M. S.; J ay, M. A.; Durkin, M. M.; Smith,
J . W. Activity of lipid-soluble inhibitors of dihydrofolate reduc-
tase against Pneumocystis carinii in culture and in a rat model
of infection. Antimicrob. Agents Chemother. 1987, 31, 1323-
1327.
(28) Sattler, F. R.; Allegra, C. J .; Verdegem, T. D.; Akil, B.; Tuazon,
C. U.; Hughlett, C.; Ogata-Arakaki, D.; Feinberg, J .; Shelhamer,
J .; Lane, H. C.; Davis, R.; Boylen, C. T.; Leedom, J . M.; Masur,
H. Trimetrexate-leucovorin dosage evaluation study for treat-
ment of Pneumocyctis carinii pneumonia. J . Infect. Dis. 1990,
161, 91-96.
(29) Goodford, P. J . A computational procedure for determining
energetically favorable binding sites on biologically important
macromolecules. J . Med. Chem. 1985, 28, 849-857.
(30) Ledig, K. W. 7-(Substituted)-7H-pyrrolo[3,2-f]quinazoline-1,3-
diamines. U.S. Patent 4,118,561, 1978.
(31) J ones, M. L.; Kuyper, L. F.; Styles, V. L.; Caddell, J . M. Lewis
acid assisted cyclization of arylcyanoguanidines to 2,4-diamino-
quinazolines. J . Heterocycl. Chem. 1994, 31, 1681-1683.
(32) Parr, W. J . E. Synthesis of selenium-containing para-phenylene-
diamines: Novel antidegradants for natural rubber. J . Chem.
Soc., Perkin Trans. I 1981, 3002-3007.
(33) Verardo, G.; Giumanini, A. G.; Strazzolini, P.; Poiana, M.
Reductive N-monoalkylation of primary aromatic amines. Syn-
thesis 1993, 121-125.
Ack n ow led gm en t. We thank J . Champness for
crystal structure coordinates of the P. carinii DHFR-
NADPH-piritrexim complex. Determinations of pKa
were performed by D. Minick. We thank G. Martin, L.
Taylor, and their respective staffs for NMR and mass
spectrometric measurements and L. Elwell for the
determination of antibacterial activities.
Refer en ces
(1) Schweitzer, B. I.; Dicker, A. P.; Bertino, J . R. Dihydrofolate
reductase as a therapeutic target. FASEB J . 1990, 4, 2441-
2452.
(2) Kuyper, L. F. Inhibitors of dihydrofolate reductase. In Computer-
Aided Drug Design; Perun, T. J ., Propst, C. L., Eds.; Marcel
Dekker Inc.: New York, 1989; pp 327-369.
(3) Bertino, J . R. Karnofsky Memorial Lecture: Ode to methotrex-
ate. J . Clin. Oncol. 1993, 11, 5-14.
(4) Fleming, G. F.; Schilsky, R. L. Antifolates: The next generation.
Semin. Oncol. 1992, 19, 707-719.
(5) Sirotnak, F. M., Burchall, J . J ., Ensminger, W. B., Montgomery,
J . A., Eds. Folate Antagonists as Therapeutic Agents; Academic
Press: New York, 1984.
(6) Roth, B.; Cheng, C. C. Recent progress in the medicinal
chemistry of 2,4-diaminopyrimidines. In Progress in Medicinal
Chemistry; Ellis, G. P., West, G. B., Eds.; Elsevier Biomedical
Press: Amsterdam, 1982; Vol. 19, pp 269-331.
(7) Salter, A. J . Trimethoprim-sulfamethoxazole: An assessment
of more than 12 years of use. Rev. Infect. Dis. 1982, 4, 196-236.
(8) Hitchings, G. H. The metabolism of plasmodia and the chemo-
therapy of malarial infections. In Tropical Medicine from
Romance to Reality; Wood, C., Ed.; Academic Press: London,
1978; pp 79-98.
(9) Allegra, C. J .; Chabner, B. A.; Tuazon, C. U.; Ogata-Arakaki,
D.; Baird, B.; Drake, J . C.; Simmons, J . T.; Lack, E. E.;
Shelhamer, J . H.; Balis, F.; Walker, R.; Kovacs, J . A.; Lane, H.
C.; Masur, H. Trimetrexate for the treatment of Pneumocystis
carinii pneumonia in patients with the acquired immunodefi-
ciency syndrome. N. Engl. J . Med. 1987, 317, 978-985.

High-Affinity Inhibitors of Dihydrofolate Reductase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 4 903
(34) Kotsuki, H.; Kobayashi, S.; Matsumoto, K.; Suenaga, H.; Nish-
izawa, H. High Pressure Organic Chemistry; XII. A Convenient
Synthesis of Aromatic Amines from Activated Aromatic Fluo-
rides. Synthesis 1990, 1147-1148.
(58) Baccanari, D. P.; Tansik, R. L. Multiple drug resistance and its
relationship to brain penetration by lipophilic dihydrofolate
reductase inhibitors. Proc. Am. Assoc. Cancer Res. 1990, 31, 339.
(59) Hill, B. T.; Price, L. A. DDMP (2,4-diamino-5-(3′,4′-dichloro-
phenyl)-6-methylpyrimidine). Cancer Treat. Rev. 1980, 7, 95-
112.
(60) Duch, D. S.; Edelstein, M. P.; Nichol, C. A. Inhibition of
histamine-metabolizing enzymes and elevation of histamine
levels in tissues by lipid-soluble anticancer folate antagonists.
Mol. Pharmacol. 1980, 18, 100-104.
(61) Luft, B. J .; Remington, J . S. AIDS commentary; toxoplasmic
encephalitis. J . Infect. Dis. 1988, 157, 1-6.
(62) Frenkel, J . K.; Hitchings, G. H. Relative reversal by vitamins
(p-aminobenzoic, folic, and folinic acids) of the effects of sulfa-
diazine and pyrimethamine on Toxoplasma, mouse and man.
Antibiot. Chemother. 1957, 7, 630-638.
(63) Delves, C. J .; Ballantine, S. P.; Tansik, R. L.; Baccanari, D. P.;
Stammers, D. K. Refolding of recombinant Pneumocystis carinii
dihydrofolate reductase and characterization of the enzyme.
Protein Express. Purif. 1993, 4, 16-23.
(64) Hughes, W. T. Comparison of dosages, intervals, and drugs in
prevention of Pneumocystis carinii pneumonia. Antimicrob.
Agents Chemother. 1988, 32, 623-625.
(65) Ruskin, J .; LaRiviere, M. Low-dose co-trimoxazole for prevention
of Pneumocystis carinii pneumonia in human immunodeficiency
virus disease. Lancet 1991, 337, 468-471.
(66) Roth, B.; Strelitz, J . Z. The protonation of 2,4-diaminopyrim-
idines. I. Dissociation constants and substituent effects. J . Org.
Chem. 1969, 34, 821-836.
(35) Suhr, H. Nucleophilic substitution. V. Reaction of 4-nitrofluo-
robenzene with primary amines. J ustus Liebigs Ann. Chem.
1965, 687, 175-182.
(36) Peters, A. T.; Soboyejo, N. Iodo-substituted 4-aminoazobenzenes
and 4-phenylazonaphthylamines - dyes for synthetic-polymer
fibres. J . Soc. Dyers Colour. 1988, 104, 486-491.
(37) Bui, N. M.; Gillet, R.; Dumont, P. An improved synthesis of
5-iodo-2′-deoxyuridine-I131 and 5-iodouracil-I131. Int. J . Appl.
Radiat. Isot. 1965, 16, 337-339.
(38) Heck, R. F. Palladium Reagents in Organic Syntheses; Academic
Press: London, 1985.
(39) Sakamoto, T.; Kondo, Y.; Iwashita, S.; Nagano, T.; Yamanaka,
H. Condensed heteroaromatic ring systems. XIII. One-step
synthesis of 2-substituted 1-methylsulfonylindoles from N-(2-
halophenyl)methanesulfonamides. Chem. Pharm. Bull. 1988, 36,
1305-1308.
(40) Arcadi, A.; Cacchi, S.; Marinelli, F. Palladium-catalysed coupling
of aryl and vinyl triflates or halides with 2-ethynylaniline: An
efficient route to functionalized 2-substituted indoles. Tetrahe-
dron Lett. 1989, 30, 2581-2584.
(41) Blaney, J . M.; Hansch, C.; Silipo, C.; Vittoria, A. Structure-
activity relationships of dihydrofolate reductase inhibitors.
Chem. Rev. 1984, 84, 333-407.
(42) Champness, J . N.; Kuyper, L. F.; Beddell, C. R. Interaction
between dihydrofolate reductase and certain inhibitors. In Topics
in Molecular Pharmacology; Burgen, A. S. V., Roberts, G. C. K.,
Tute, M. S., Eds.; Elsevier: New York, 1986; Vol. 3, pp 335-
362.
(43) Kraut, J .; Matthews, D. A. Dihydrofolate reductase. In Biological
Macromolecules and Assemblies; J urnak, F. A., McPherson, A.,
Eds.; Wiley: New York, 1987; Vol. 3, pp 1-71.
(44) Freisheim, J . H.; Matthews, D. A. The comparative biochemistry
of dihydrofolate reductase. In Folate Antagonists as Therapeutic
Agents; Sirotnak, F. M., Burchall, J . J ., Ensminger, W. D.,
Montgomery, J . A., Eds.; Academic Press: New York, 1984; Vol.
1, pp 69-131.
(45) Champness, J . N.; Achari, A.; Ballantine, S. P.; Bryant, P. K.;
Delves, C. J .; Stammers, D. K. The structure of Pneumocystis
carinii dihydrofolate reductase to 1.9 A resolution. Structure
1994, 2, 915-924.
(67) Cha, S. Tight-binding inhibitors - 1: Kinetic behavior. Biochem.
Pharmacol. 1975, 24, 2177-2185.
(68) Henderson, P. J . F. A linear equation that describes the steady-
state kinetics of enzymes and subcellular particles interacting
with tightly bound inhibitors. Biochemistry 1972, 127, 321-323.
(69) Baccanari, D. B.; J oyner, S. S. Dihydrofolate reductase hysteresis
and its effect on inhibitor binding analyses. Biochemistry 1981,
20, 1710-1716.
(70) Delcamp, T. J .; Susten, S. S.; Blankenship, D. T.; Freisheim, J .
H. Purification and characterization of dihydrofolate reductase
from methotrexate-resistant human lymphoblastoid cells. Bio-
chemistry 1983, 22, 633-639.
(71) Krug, E. C.; Marr, J . J .; Berens, R. L. Purine metabolism in
Toxoplasma gondii. J . Biol. Chem. 1989, 264, 10601-10607.
(72) Carmichael, J .; Mitchell, J . B.; DeGraff, W. G.; Gamson, J .;
Gazdar, A. F.; J ohnson, B. E.; Glatstein, E.; Minna, J . D.
Chemosensitivity testing of human lung cancer cell lines using
the MTT assay. Br. J . Cancer 1988, 57, 540-547.
(73) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; MacMahon,
J .; Vistica, D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M.
R. New colorimetric cytotoxicity assay for anticancer-drug
screening. J . Natl. Cancer Inst. 1990, 82, 1107-1112.
(74) Ou-Yang, K.; Krug, E. C.; Marr, J . J .; Berens, R. L. Inhibition
of growth of Toxoplasma gondii by qinghasosu and derivatives.
Antimicrob. Agents Chemother. 1990, 34, 1961-1965.
(75) McLemore, T. L.; Eggleston, J . C.; Shoemaker, R. H.; Abbott, B.
J .; Bohlman, M. E.; Liu, M. C.; Fine, D. L.; Mayo, J . G.; Boyd,
M. R. Comparison of intrapulmonary, percutaneous intratho-
racic, and subcutaneous models for the propagation of human
pulmonary and nonpulmonary cancer cell lines in athymic nude
mice. Cancer Res. 1988, 48, 2880-2886.
(76) Wang, A. M.; Elion, G. B.; Friedman, H. S.; Bodell, W. J .; Bigner,
D. D.; Schold, S. C. Positive therapeutic interaction between
thiopurines and alkylating drugs in human glioma xenografts.
Cancer Chemother. Pharmacol. 1991, 27, 278-284.
(77) Comley, J . C. W.; Sterling, A. M. Artificial infections of Pneu-
mocystis carinii in the SCID mouse and their use in the in vivo
evaluation of antipneumocystis drugs. J . Eukaryot. Microbiol.
1994, 41, 540-546.
(78) Comley, J . C. W.; Sterling, A. M. Effect of atovaquone and
atovaquone drug combinations on prophylaxis of Pneumocystis
carinii Pneumonia in SCID mice. Antimicrob. Agents Chemother.
1995, 39, 806-811.
(79) Foss, R. G.; Sigel, C. W. Lipid-soluble inhibitors of DHFR. III:
Quantitative thin-layer and high performance liquid chromato-
graphic methods of the measurement of plasma concentrations
of the antifolate, 2,4-diamino-6-(2,5-dimethoxybenzyl)-5-meth-
ylpyrido[2,3-d]pyrimidine. J . Pharm. Sci. 1982, 71, 1176-1178.
(46) McCormack, J . J .; Allen, B. A.; Ledig, K. W.; Hillcoat, B. L.
Inhibition of dihydrofolate reductases by derivatives of 2,4-
diaminopyrroloquinazoline. Biochem. Pharmacol. 1979, 28, 3227-
3229.
(47) Castaldo, R. A.; Gump, D. W.; McCormack, J . J . Activity of 2,4-
diaminoquinazoline compounds against Candida species. Anti-
microb. Agents Chemother. 1979, 15, 81-86.
(48) Whitlow, M.; Howard, A. J .; Stewart, D.; Hardman, K. D.;
Kuyper, L. F.; Baccanari, D. P.; Fling, M.; Tansik, R. Unpub-
lished results.
(49) van der Bliek, A. M.; Borst, P. Multidrug resistance. Adv. Cancer
Res. 1989, 52, 165-203.
(50) Gottesman, M. M.; Pastan, I. The multidrug transporter, a
double-edged sword. J . Biol. Chem. 1988, 263, 12163-12166.
(51) Zamora, J . M.; Pearce, H. L.; Beck, W. T. Physical-chemical
properties shared by compounds that modulate multidrug
resistance in human leukemic cells. Mol. Pharmacol. 1988, 33,
454-462.
(52) Klohs, W. D.; Steinkampf, R. W.; Besserer, J . A.; Fry, D. W. Cross
resistance of pleiotropically drug resistant P388 leukemia cells
to the lipophilic antifolates trimetrexate and BW 301U. Cancer
Lett. 1986, 31, 253-260.
(53) Assaraf, Y. G.; Molina, A.; Schimke, R. T. Cross-resistance to
the lipid-soluble antifolate trimetrexate in human carcinoma
cells with the multidrug-resistant phenotype. J . Natl. Cancer
Inst. 1989, 81, 290-294.
(54) Selassie, C. D.; Hansch, C.; Khwaja, T. A. Structure-activity
relationships of antineoplastic agents in multidrug resistance.
J . Med. Chem. 1990, 33, 1914-1919.
(55) McGrath, J . P.; Varshavsky, A. The yeast STE6 gene encodes a
homologue of the mammalian multidrug resistance P-glycopro-
tein. Nature 1989, 340, 400-404.
(56) Raymond, M.; Gros, P.; Whiteway, M.; Thomas, D. Y. Functional
complementation of yeast ste6 by a mammalian multidrug
resistance mdr gene. Science 1992, 256, 232-234.
(57) Cordon-Cardo, C.; O’Brien, J . P.; Casals, D.; Rittman-Grauer,
L.; Biedler, J . L.; Melamed, M. R.; Bertino, J . R. Multidrug-
resistance gene (P-glycoprotein) is expressed by endothelial cells
at blood-brain barrier sites. Proc. Natl. Acad. Sci. U.S.A. 1989,
86, 695-698.
J M9505122