New pyrimido[5,4-b]indoles as ligands for α1-adrenoceptor subtypes

Article abstract of DOI:10.1021/jm0307741

A new series of compounds were designed as structural analogues of the α₁-AR ligand RN5 (4), characterized by a tricyclic 5H-pyrimido[5,4-b]indole-(1H,3H)2,4-dione system connected through an alkyl chain to a phenylpiperazine (PP) moiety. These compounds were synthesized and tested in binding assays on human α_1A-AR, α_1B-AR, and α_1D-AR subtypes expressed in HEK293 cells. Several structural modifications were performed on the PP moiety, the tricyclic system, and the connecting alkyl chain. Many of the new molecules showed a preferential affinity for the α_1D-AR subtype. Some compounds, including 39 and 40, displayed substantial α_1D-AR selectivity with respect to α_1A-AR, α_1B-AR, serotonergic 5-HT_1A, 5-HT_1B, 5-HT_2A, and dopaminergic D₁ and D₂ receptors. Two conformationally rigid analogues of 4, useful for studying the architecture of the receptor/ligand complex, were also prepared and tested. A subset of the new compounds was then used to evolve a preliminary pharmacophore model for α_1D-AR antagonists, based on a more generalized model we had developed for α₁-AR antagonists. This new model rationalized the relationships between structural properties and biological data of the pyrimido[5,4-b]indole compounds, as well as other compounds.

Full text of DOI:10.1021/jm0307741

J . Med. Chem. 2003, 46, 2877-2894
2877
New P yr im id o[5,4-b]in d oles a s Liga n d s for r1-Ad r en ocep tor Su btyp es
Giuseppe Romeo,*,§ Luisa Materia,§ Fabrizio Manetti,| Alfredo Cagnotto,† Tiziana Mennini,†
Ferdinando Nicoletti,# Maurizio Botta,| Filippo Russo,§ and Kenneth P. Minneman‡
Dipartimento di Scienze Farmaceutiche, Universita` di Catania, Viale A. Doria 6, 95125 Catania, Italy, Dipartimento di
Scienze Farmacologiche, Universita` “La Sapienza”, Roma, Italy, Istituto di Ricerche Farmacologiche “Mario Negri”, Via
Eritrea 62, 20157, Milano, Italy, Dipartimento Farmaco Chimico Tecnologico, Universita` degli Studi di Siena, Via A. Moro,
53100 Siena, Italy, and Department of Pharmacology, Emory University, 1510 Clifton Road, Atlanta, Georgia 30322
Received J anuary 9, 2003
A new series of compounds were designed as structural analogues of the R1-AR ligand RN5
(4), characterized by a tricyclic 5H-pyrimido[5,4-b]indole-(1H,3H)2,4-dione system connected
through an alkyl chain to a phenylpiperazine (PP) moiety. These compounds were synthesized
and tested in binding assays on human R1A-AR, R1B-AR, and R1D-AR subtypes expressed in
HEK293 cells. Several structural modifications were performed on the PP moiety, the tricyclic
system, and the connecting alkyl chain. Many of the new molecules showed a preferential
affinity for the R1D-AR subtype. Some compounds, including 39 and 40, displayed substantial
R1D-AR selectivity with respect to R1A-AR, R1B-AR, serotonergic 5-HT1A, 5-HT1B, 5-HT2A, and
dopaminergic D1 and D2 receptors. Two conformationally rigid analogues of 4, useful for studying
the architecture of the receptor/ligand complex, were also prepared and tested. A subset of the
new compounds was then used to evolve a preliminary pharmacophore model for R1D-AR
antagonists, based on a more generalized model we had developed for R1-AR antagonists. This
new model rationalized the relationships between structural properties and biological data of
the pyrimido[5,4-b]indole compounds, as well as other compounds.
In tr od u ction
AR selective antagonists in symptomatic therapy of
BPH. To date, several highly selective R1A-AR antago-
nists have been synthesized, some of which are in
preclinical and clinical development.13,14
R1-Adrenergic receptors (R1-AR) belong to the seven-
transmembrane-domain (7-TM) receptor superfamily
and mediate many physiological effects of the catechola-
mines epinephrine and norephineprine.1 Three different
native R1-AR subtypes have been cloned and are re-
ferred to as R1A-AR, R1B-AR, and R1D-AR.2-5 These
subtypes are expressed in many human tissues includ-
ing liver, brain, myocardium, and vascular smooth
muscle. Each R1-AR subtype has a distinct pharmacol-
ogy and shows a discrete tissue distribution.6 Unlike the
On the other hand, the search for R1B-AR or R1D-AR
selective ligands has been less fruitful, with only a few
compounds showing even modest selectivity for these
subtypes being described in the literature. Examples of
such ligands are provided by (+)-cyclazosin (1), L-765,314
(2), and BMY 7378 (3) (Chart 1). (+)Cyclazosin, struc-
turally related to prazosin, is the most selective R1B
-
AR antagonist reported to date, although its R1B-AR
selectivity is disputed.15,16 L-765,314 is another prazosin
analogue that has shown R1B-AR selectivity.17 On the
other hand, BMY 7378 has become the reference ligand
for characterization of R1D-ARs.18 However, the useful-
ness of these molecules is restricted by their limited
selectivity. Moreover, BMY 7378 (3) also binds with high
affinity to serotonergic 5-HT1A receptors, where it has
partial agonist activity.19 Hence, the discovery of new,
highly selective ligands for R1B-ARs and R1D-ARs might
represent a major advance, providing useful probes to
define the functional roles of each subtype that might
possibly possess unique therapeutic applications.
R
1B-AR and R1D-AR subtypes, several splice variants of
R
1A-ARs have been isolated from human heart, prostate,
and hippocampus.7,8 Four of these present a typical
7-TM receptor structure, differing only in their distal
C-terminal sequences, and show apparently identical
pharmacological profiles. Other isoforms appear to be
prematurely truncated and are unable to bind ligand
or activate functional responses. At present, however,
the physiological and pathological roles of full-length
and truncated R1-ARs remain to be clarified.
In the past decades, several non-subtype-selective R1-
AR antagonists, such as prazosin and doxazosin, have
been used effectively for treatment of hypertension and
benign prostatic hyperplasia (BPH), a urologic disorder
prevalent in elderly males.9,10 Increasing evidence for
the role of R1A-ARs in bladder outlet obstruction in
Recently, we were involved in developing new selec-
tive R1-AR ligands characterized by a tricyclic 5H-
pyrimido[5,4-b]indole-(1H,3H)2,4-dione system coupled,
by means of an alkyl chain, to a phenylpiperazine (PP)
moiety.20-22 Among them, RN5 (4) (Chart 1) emerged
as one of the most interesting, showing high affinity and
selectivity for R1-ARs on rat cortical membranes over
R2-AR, â2-AR, and 5-HT1A receptors.20 Several analogues
of 4 were obtained, focusing particularly on structural
variations in the PP moiety. When tested on cloned R1-
patients with BPH11,12 has encouraged the use of R1A
-
* To whom correspondence should be addressed. Phone: +39 095
7384027. Fax: +39 095 222239. E-mail: gromeo@mbox.unict.it.
§ Universita` di Catania.
#
Universita` “La Sapienza”.
† Istituto di Ricerche Farmacologiche “Mario Negri”.
| Universita` degli Studi di Siena.
‡ Emory University.
10.1021/jm0307741 CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/07/2003

2878 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
Ch a r t 1
Sch em e 1a
a
Conditions: (a) RC(OC2H5)3, reflux; (b) H2N(CH2)nOH, reflux;
(c) SOCl2, toluene, reflux.
prepared two conformationally restricted analogues of
4 in which the N-(2-methoxyphenyl)piperazine moiety
was substituted by the strictly related 1,2,3,4,4a,5-
hexahydropyrazino[2,1-c][1,4]benzoxazine. The latter
system can be regarded as a 1-(2-methoxyphenyl)-
piperazine in which the methoxy group is connected
with a carbon atom of the piperazine ring through a
single bond.
Finally, because the new pyrimido[5,4-b]indoles showed
a preferential affinity for R1D-ARs, we used the program
Catalyst to develop a pharmacophore model for R1D-AR
antagonists. The calculated model consists of a positive
ionizable portion, three hydrophobic features, and two
hydrogen bond acceptor groups. This model showed a
good statistical significance with a correlation coefficient
of 0.91 and successfully predicts the affinities of the
molecules of, and external to, the training set.
AR subtypes, derivatives bearing a substituent in the
4-position of the PP aromatic ring showed decreased
affinity but were able to discriminate among R1-AR
subtypes with a preference for R1D-ARs.22 This suggested
that R1D-ARs might, unlike the other two subtypes,
tolerate the steric bulk of a substituent in the 4-position.
The tricyclic 5H-pyrimido[5,4-b]indole-(1H,3H)2,4-dione
moiety of 4 was less thoroughly investigated, and only
a few structural modifications were made.
We now report the synthesis and binding properties
of a new series of pyrimido[5,4-b]indole derivatives
structurally related to 4. The new molecules present
variations in the PP moiety, in the length of the
connecting alkyl chain and in the tricyclic system as
follows. (i) Substituents of increasing steric bulk in the
4-position or alkoxy group in the 2-position were in-
serted on the aryl ring of the PP moiety. (ii) The
ethylene chain in 4 was elongated to three methylene
units in some molecules. (iii) The pyrimido[5,4-b]indole-
2,4-dione moiety was modified by deletion of the CdO
group in the 2-position or substituted by the bioisosteric
benzothieno[3,2-d]pyrimidine-2,4-dione system.
Ch em istr y
The synthetic pathway to the final 3-[2-[4-(2-substi-
tutedphenyl)piperazin-1-yl]alkyl-5H-pyrimido[5,4-b]in-
dole-(3H)4-one derivatives 23-31 is shown in Schemes
1 and 2.
The starting aminoester 5, which was prepared ac-
cording to the Unangst’s method,23 was reacted with
triethylortho esters of formic, acetic, or propionic acids
to afford the corresponding iminoethers 6-8. Products
6-8 were reacted with 2-ethanolamine or 3-amino-1-
propanol to obtain tricyclic alcohols 9-14. By reaction
with SOCl2 in toluene, alcohols 9-14 were converted
to the corresponding alkyl chloride hydrochlorides 15-
20 (Scheme 1). Reaction of chloro derivatives 15-20
with 1-(2-methoxyphenyl)piperazine (21) or 1-(2-ethoxy-
phenyl)piperazine (22) at 140 °C afforded final products
23-31 (Scheme 2).
In addition, the flexibility of the [4-(2-methoxy-
phenyl)piperazin-1-yl]ethyl moiety in 4 allows the mol-
ecule to adopt a variety of low-energy conformations.
To obtain more rigid R1-AR ligands for studying the
architecture of the receptor/ligand complex, we also
The preparation of final compounds 39-46 and 48 is
presented in Scheme 3. Indolylurea 3220 was reacted
with 1-(4-substitutedphenyl)piperazines 33-38 to afford
final products 39-44. Phenylpiperazines 33-38 were
obtained by reaction of the suitable 4-substituted aniline

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2879
Sch em e 2a
reagents (J ones’ reagent as in Gupta’s method, chromic
anhydride in pyridine as in Baxter’s procedure, anhy-
drous DMSO in acetic anhydride, pyridinium chloro-
chromate), Dess-Martin’s reagent,30 as modified by
Geiss,31 was preferred because it gave the best yield
(87%) in the oxidized product 55. It was then converted
into the bicyclic derivate 56 through a catalytic reduc-
tion with H2 on 10% Pd/C at atmospheric pressure, and
after hydrolysis of the phthalimide system with hydra-
zine hydrate, amine 57 was obtained. After the protec-
tion of the primary amine group with benzyl chlorofor-
mate to give 58, 58 was reacted with chloroacetyl
chloride to give amide 59. Cyclization of 59 to the
tricyclic derivate 60 with potassium carbonate in DMSO,
reduction of the amide carbonyl with the borane-THF
complex, and successive deprotection of secondary amine
afforded amine 62.
Finally, reaction of 62 with indolylurea 32 or with
alkyl chloride 16 gave target compounds 63 and 64,
respectively (Scheme 6).
P h a r m a cology
Tricyclic compounds 23-28, 30, 39-46, 48-52, 63,
and 64 along with RN5 (4), its chloro analogue 65,20 and
the des-methyl analogue 6632 (Chart 2) were tested in
binding assays on human R1A-AR, R1B-AR, and R1D-AR
subtypes stably expressed in HEK293 cells using [125I]-
BE 2254 as radioligand. Their affinity values were
expressed as pKi or, for compounds with low water
solubility, as a percentage of inhibition of specific
binding at the highest concentration tested (1 µM).
Affinities of compounds 39 and 40 for some other
receptor classes, such as 5-HT1A, 5-HT1B, 5-HT2A, D1,
and D2 receptors, were also measured.
a
Conditions: (a) 140 °C.
and bis(2-chloroethyl)amine hydrochloride in 2-butoxy-
ethanol at reflux and in the presence of potassium
carbonate.
Furthermore, compound 32, under the same experi-
mental conditions used for the preparation of 39-44,
was reacted with 1-(2-ethoxyphenyl)piperazine (22) or
with 4-(2-methoxyphenyl)piperidine24 to obtain 45 or 46,
respectively.
Analogously, benzothienyl urea 4725 was reacted with
4-[4-(1-methylethyl)phenyl]piperazine (33) to give the
final derivative 48.
Moreover, to evaluate the effects of compounds 40 and
RN5 (4) on the signal transduction pathway coupled to
R1-ARs, we measured their ability to block norepineph-
rine-induced stimulation of inositol phospholipid hy-
drolysis in rat hippocampal slices.
Resu lts a n d Discu ssion
The radioligand binding data on human cloned R1-
AR subtypes of selected new tricyclic derivatives, com-
pounds 4 (RN5) and 65 (whose synthesis and affinities
on rat R1-AR had been already reported)20 and 66 are
shown in Table 1.
Scheme 4 shows the synthetic pathway for the prepa-
ration of the final compounds 49-52, which structurally
bear a (4-substitutedphenyl)piperazine moiety and lack
the carbonylic group in the 2-position of the tricyclic
system. Synthesis was performed by coupling alkyl chlo-
rides 15 or 16 (Scheme 1) with 4-[4-(1-methylethyl)-
phenyl]piperazine (33) or 4-[4-(1,1-dimethylethyl)phenyl]-
piperazine (34).
The synthetic pathway to tricyclic amine 62, required
to prepare final compounds 63 and 64, is shown in
Scheme 5. The preparation of 62 had been previously
reported by Gupta.26 Recently, Baxter described the
synthesis of this amine, even if any spectroscopic data
for intermediates and 62 were reported.27,28 Since we
had some difficulties in reproducing Gupta’s procedure,
Baxter’s method was followed with some modifications.
The starting compound was oxirane 53, which was
prepared by reaction of 2-nitrophenol with epichloro-
hydrin.29 Compound 53 was reacted with phthalimide
to afford the secondary alcohol 54, which was then
oxidized to the corresponding ketone 55. This step was
quite critical. After some attempts with different oxidant
As a general trend, many of tested compounds showed
affinities for the three receptor subtypes in the order
R
1D-AR g R1A-AR > R1B-AR, and some displayed some
selectivity for R1D-ARs. As expected, RN5 (4) maintained
the same high affinity for the human cloned R1-AR
subtypes as was previously observed with rat cortical
R1-AR. RN5 (4) displayed a slight preference for R1A
-
ARs and R1D-ARs (pKi ) 9.57 and 9.44, respectively)
with respect to R1B-ARs (pKi ) 8.74). Its chloro analogue
65, with 10-fold lower affinity values, was still a good
ligand showing no selectivity among subtypes.
Compounds 45 and 66, in which an ethoxy and a
hydroxy group respectively replaces the methoxy group
of 4, showed a decreased affinity for all three subtypes;
however, the reduction was more pronounced for R1A
-
ARs and R1B-ARs. Therefore, 45 and 66 showed a slight
selectivity for R1D-ARs. A similar but less pronounced
reduction in affinity was observed with 46, in which a

2880 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
Sch em e 3a
a
Conditions: (a) 140 °C; (b) 22 or 4-(2-methoxyphenyl)piperidine, 140 °C.
Sch em e 4a
Ta ble 1. Binding Properties of 5H-Pyrimido[5,4-b]indole
Derivatives
pKi (M) or [% of inhibition at 1 µM]a
1A-AR 1B-AR
1D-ARb
compd
R
R
R
23c
24d
25c
26c
27
8.85 ( 0.08
8.52 ( 0.03
7.31 ( 0.13
7.78 ( 0.17
8.31 ( 0.14
7.97 ( 0.11
8.01 ( 0.03
4.89 ( 0.18
[8 ( 3]
7.86 ( 0.11
7.68 ( 0.05
6.61 ( 0.05
6.79 ( 0.04
7.57 ( 0.08
7.56 ( 0.12
7.21 ( 0.07
5.02 ( 0.13
[2 ( 4]
9.14 ( 0.08 (9.46)
9.39 ( 0.19 (9.38)
8.16 ( 0.10 (8.43)
7.79 ( 0.12 (7.74)
8.59 ( 0.16
7.90 ( 0.11 (7.82)
8.22 ( 0.12 (7.70)
7.45 ( 0.14 (5.36)
7.70 ( 0.08 (6.77)
7.22 ( 0.014 (7.43)
6.61 ( 0.07 (6.96)
[3 ( 2] (6.74)
8.00 ( 0.03 (7.92)
8.80 ( 0.10 (8.14)
9.03 ( 0.22
6.90 ( 0.08 (7.13)
<5
5.77 ( 0.11 (6.47)
<5
28c
29d
39d
40c
41c
42c
43c
44c
45c
46
[26 ( 4]
[4 ( 3]
[19 ( 6]
[6 ( 5]
[7 ( 2]
[18 ( 5]
[48 ( 5]
[55 ( 8]
8.14 ( 0.07
8.81 ( 0.15
[18 ( 7]
7.84 ( 0.03
8.54 ( 0.14
[4 ( 4]
48c
49
<5
<5
50c
51
5.45 ( 0.15
<5
5.39 ( 0.23
<5
52c
63c
64d
65c
66d
4c
6.22 ( 0.03
5.10 ( 0.08
<5
5.69 ( 0.08
5.11 ( 0.10
<5
7.01 ( 0.06 (7.33)
6.90 ( 0.10 (7.06)
<5 (5.91)
8.44 ( 0.16 (8.82)
9.20 ( 0.27 (8.32)
9.44 ( 0.11 (9.48)
8.33 ( 0.10
8.05 ( 0.07
9.57 ( 0.14
8.59 ( 0.10
7.93 ( 0.10
8.74 ( 0.14
a
Each value is the mean ( SE for data from three different
b
experiments conducted in duplicate. Estimated and predicted
a
Conditions: (a) 140 °C.
affinity values calculated by Catalyst for the training set and test
set, respectively, are reported in parentheses. c Compound used
to build the training set. Compound used to build the test set.
4-(2-methoxyphenyl)piperidine moiety replaced the N-(2-
methoxyphenyl)piperazine in 4. This indicates that the
piperazine nitrogen vicinal to the aryl ring is not
essential for binding, although it contributes to an
increased affinity.
d
the 4-(2-methoxyphenyl)piperazine moiety from the 2-
to the 4-position invariably produced a notable drop in
affinity.21 However, in some derivatives, it gave rise to
the appearance of selectivity for R1D-ARs.22 In this study,
we synthesized compounds 39-44 to analyze the influ-
In previously reported series of pyrimido[5,4-b]in-
doles, the shift of the substituent on the phenyl ring of

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2881
Sch em e 5a
a
Conditions: (a) phthalimide, butanol, pyridine, reflux; (b) Dess-Martin reagent, CH2Cl2, 60 °C; (c) H2, Pd/C, EtOH; (d) hydrazine
hydrate, EtOH, reflux; (e) ClCbz, THF, -78 °C; (f) ClCOCH2Cl, CH2Cl2, -78 °C; (g) K2CO3, DMSO; (h) BH3-THF, THF, 0 °C; (i) Pd/C,
1,4-cyclohexadiene, CF3COOH, MeOH/H2O.
ence of bulky alkyl substituents in the 4-position of the
phenyl ring on affinity and selectivity for R1-AR sub-
types. Unfortunately, many of them showed a very low
water solubility, precluding their testing at concentra-
tions higher than 1 µM in binding assays. In these cases,
the percentage inhibition of specific binding at the
highest concentration tested (1 µM) was measured and
is reported in Table 1. As expected, compounds 39-44
showed very low affinities, particularly for R1A-AR and
preparing compounds 23-30, in which CdO group was
replaced by a hydrogen, methyl, or ethyl group. The
overall effect of this modification was a reduction in
affinity, particularly for R1A-AR and R1B-AR subtypes.
With respect to R1D-ARs, derivatives with a methyl
group in the 2-position were the most interesting. In
fact, among compounds 23-25, which present an eth-
ylene chain connecting the tricyclic system with the PP
moiety, 24 showed the highest affinity (pKi ) 9.39) and,
as opposed to 4, a slight selectivity for the R1D-AR. The
elongation of the connecting alkyl chain in compounds
26-28 to three methylene units led to a further
decrease in affinity. Again, the methyl derivative 27 was
the best of the three compounds.
Compound 30 bears a 2-ethoxy group on the phenyl
ring of the PP moiety. As already noted for 45, it showed
lower affinity values for all three subtypes compared
to its methoxy analogue 24, indicating that the in-
creased bulk of the ethoxy group is not well tolerated
in R1-AR binding sites.
R
1B-AR subtypes. While the 4-cyclohexyl derivative 43
had no measurable affinity for any of the three subtypes,
the 4-(1-methylethyl) derivative 39 and the 4-(1,1-
dimethylethyl) derivative 40 displayed moderate affini-
ties for R1D-ARs (pKi ) 7.45 and 7.70, respectively) with
no measurable affinity for R1A-AR or R1B-AR subtypes.
In addition, when tested in radioligand assays on
serotonergic 5-HT1A, 5-HT1B, 5-HT2A and dopaminergic
D1 and D2 receptors, they showed no affinity (pKi e 5).
Thus, 39 and 40 represent two interesting molecules
with good selectivity for R1D-ARs.
Some other structural variations were made on the
pyrimido[5,4-b]indole-2,4-dione moiety of RN5 (4). We
had already reported that the replacement of the indole
nucleus of 4 with a benzothieno system leads to R1-AR
ligands retaining high affinity.21 We applied the same
strategy to compound 48, which can be considered the
benzothienyl analogue of 39. However, this variation
was detrimental for selectivity, since 48 showed a lower
affinity for R1D-ARs than 39.
Further structural modifications led to compounds
49-52, which present a 4-(1-methylethyl) or a 4-(1,1-
dimethylethyl) substituent on the phenyl ring in the PP
moiety and lack the CdO group (replaced with a
hydrogen or methyl group) in the 2-position of the
pyrimido[5,4-b]indole system. Almost all these com-
pounds showed no significant affinity or selectivity for
R1-AR subtypes, although methyl derivatives 50 and 52
displayed a minimal degree of affinity, as expected,
whereas the unsubstituted analogues 49 and 51 were
completely inactive.
The role of the carbonyl group in the 2-position of the
pyrimido[5,4-b]indole-2,4-dione system was examined by

2882 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
Sch em e 6a
Ch a r t 2
a
Conditions: (a) 32, 140 °C; (b) 16, 140 °C.
Within this series, RN5 (4) and 24 are the compounds
showing the highest affinity for R1D-ARs. Both com-
pounds bear the flexible [4-(2-methoxyphenyl)piperazin-
1-yl]ethyl moiety, which allows them to adopt a variety
of low-energy conformations. Derivatives 63 and 64 can
be regarded as conformationally restricted analogues of
4 and 24, respectively, bearing a rigid 1,2,3,4,4a,5-
hexahydropyrazino[2,1-c][1,4]benzoxazine in place of the
N-(2-methoxyphenyl)piperazine moiety. When tested in
the binding assay, both 63 and 64 showed a dramatic
loss in affinity for all the R1-AR subtypes. The loss of
affinity was at least 3 and, in some cases, over 4 orders
of magnitude (see 4 vs 63 for R1A-ARs). The only
exception was 63 and R1D-ARs, where there was a
smaller (350-fold) decrease. These results indicate
that the rigid 1,2,3,4,4a,5-hexahydropyrazino[2,1-c][1,4]-
benzoxazine is not recognized by R1-AR binding sites,
and its planar conformation is probably not adopted by
the flexible N-(2-methoxyphenyl)piperazine moiety of 4
and 24 in binding to the receptor.
Ta ble 2. Stimulation of [3H]Inositol Phosphate Formation by
Norepinephrine (100 µM) in the Presence of 4 or 40 (1 µM) in
Rat Hippocampal Slices
[3H]inositol monophosphate,a dpm/mg of protein
compd
control
100 µM norepinephrine
none
4
40
3656 ( 281
3471 ( 167
3968 ( 62
12484 ( 375
3987 ( 96b
10390 ( 343b
a
Values are the mean ( SEM of at least four determinations.
P < 0.01 when compared with value obtained with norepineph-
b
rine in the absence of test compounds.
for R1D-ARs than 4, and the 1 µM concentration tested
may not be sufficient to competitively antagonize the
effects of 100 µM norepinephrine.
RN5 (4) and the selective R1D-AR compound 40 were
also tested to evaluate their effects on R1-AR-coupled
transduction pathways. Compound 4 was able to com-
pletely antagonize norepinephrine-stimulated inositol
phospholipid hydrolysis in rat hippocampal slices at
concentrations of 1 µM, thus showing full antagonistic
properties (Table 2). On the other hand, 40 only
partially reduced (25%) norepinephrine-stimulated inos-
itol phosholipid hydrolysis. One explanation for this
partial reduction is that 40 is R1D-AR selective and
therefore blocks only some of the hippocampal R1-ARs.
Among the three R1-AR subtypes, the R1D-AR is the least
efficiently coupled to inositol phospholipid hydrolysis.33
It should also be pointed out that 40 has a lower affinity
There have been many recent efforts to rationalize
the antagonist selectivity of R1- and R2-ARs at the
molecular level. In this context, some of us have
developed a pharmacophore hypothesis for R1-AR an-
tagonists34 (hereafter referred to as the “old model”)
by means of the software Catalyst.35 The old model
consists of five features (three hydrophobic, a hydrogen
bond acceptor, and a positive ionizable group; Figure
1A) and is in good agreement with a previous model for
R1-AR antagonists reported by De Marinis36 and other
groups.37

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2883
Ta ble 3. Actual and Predicted Binding Affinities for
Compounds Collected from the Literature Used To Validate the
Proposed Pharmacophoric Model for R1D-AR Antagonists
Ki (nM)
a
compd
R1A
251
294
R1B
R1D
R1A/R1D ref
BMY 7378, 3
SNAP 8719, 67
discretamine, 68 616
631
191
360
6.3 (1.9)
39.8
44
38
38
37
37
45
1.6 (6.2)
183.7
24.6
25 (14000)c
315.8 (1200)
69
70
1.4
36
31.1 1.5 (1.6)
0.9
SKF 104856, 71
23
1.6 (86000)b
22.5
a
Predicted affinity values calculated by Catalyst for the test
b
set compounds are reported in parentheses. This compound was
able to map only three pharmacophoric features. c This compound
was able to map only five pharmacophoric features.
of the common chemical features shared by the mol-
ecules and their relative alignment to the common
feature set without considering biological data. On the
other hand, the HypoGen method is usually applied to
develop three-dimensional pharmacophore models from
molecules with a wide range of diversity in both
structure and activity, with the latter spanning at least
4 orders of magnitude. Thus, during common feature
hypothesis generation, affinity data were not taken into
consideration and 67, the most selective R1D-AR com-
pound so far discovered,39 was considered as the refer-
ence structure (see Experimental Section).
F igu r e 1. Comparison between the five-feature pharmaco-
phore model for R1-AR antagonists (A, referred to as the “old
model” in the text) and the HipHop-generated pharmacophore
model for R1D-AR antagonists (B) shows four common features
(i.e. HY1, PI, HBA1, and HY3). Pharmacophore features are
color-coded as follows: blue for hydrophobics (HY); red for
positive ionizable groups (PI); green for hydrogen bond accep-
tors (HBA).
The results showed that all hypotheses generated by
the HipHop program consisted of five features: four of
them (one hydrophobic (HY), a positive ionizable (PI)
group, and two hydrogen bond acceptors (HBA)) were
always found, while the sole difference between hypoth-
eses consisted in an aromatic ring versus an additional
hydrophobic group. This last pharmacophore hypothesis
(two HYs, two HBAs, and one PI, hereafter referred to
as the HipHop model; Figure 1B) was evaluated further.
In fact, a comparison between the HipHop model and
our old model of R1-AR (Figure 1A) shows that four
features of each model (HY1, PI, HBA1, and HY3) are
located at almost identical spatial positions. Only HY2
of the old model and HBA2 of the new model lie in
unoccupied regions of space. Thus, the process of
identifying common chemical features shared by selec-
tive R1D-AR antagonists suggests that an additional
feature (HBA2 of the HipHop model) should be added
to account for the affinity and selectivity of R1D-AR-
selective antagonists and to rationalize the relationships
between structure and biological data of such com-
pounds.
However, it should be emphasized that ligands that
selectively recognize only one of several closely related
subtypes could represent very useful pharmacologic
tools. Consequently, the search for subtype selective R1-
AR ligands has dramatically increased. Similarly, both
ligand- and receptor-based approaches have been used
to decipher the molecular features responsible for af-
finity and selectivity among R1-AR subtypes.
The biological data obtained with the title compounds
indicate that with the exception of 4, 28, and 65, all
compounds showed preferential affinity for the R1D
subtype. Assuming that all these compounds interact
with the similar binding sites at each R1-AR subtype, a
selection of the new pyrimido[5,4-b]indole compounds
was submitted to a computational protocol to improve
the old pharmacophore model for R1-AR antagonists and
to identify a new model specifically for R1D-AR antago-
nists.
The first pharmacophore reported for R1D-ARs was
generated by Bremner and co-workers38 by application
of the Catalyst HypoGen routine to three compounds
with high affinity for R1D-ARs and an R1A/R1D selectivity
of at least 10-fold. This model, consisting of a hydrogen
bond acceptor, a positive ionizable group, and a hydro-
phobic moiety, unfortunately showed no correlation (r
) -0.14) between biological data and structural proper-
ties of the studied compounds.38 Alternatively, we used
the Catalyst HipHop routine (also referred to as com-
mon feature hypothesis generation) to analyze these
three compounds: BMY 7378 (3, Chart 1), SNAP 8719
(67), and discretamine (68) (Chart 2). Because of the
limited number of compounds and their small affinity
range (Table 3), the HipHop method was preferred over
HypoGen. HipHop generates models by identification
To test this hypothesis, we generated another model
based on a selected set of pyrimido[5,4-b]indole deriva-
tives and their R1D-AR affinities. Sixteen molecules of
the new compounds (Table 1) were used to build a
training set following the Catalyst guidelines. pKi (M)
values of these compounds ranged from 9.44 (4) to 5.77
(50), spanning about 3.5 orders of magnitude. Com-
pound 43, which was inactive, was also included with a
pKi value arbitrarily set at 5.70. Conformational models
within a range of 20 kcal/mol with respect to the global
minimum were generated for all compounds by means
of a molecular mechanics approach based on the use of
the CHARMm force field40 and the poling algorithm.41
The training set comprising biological data and confor-
mational models was submitted to the HypoGen routine,

2884 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
F igu r e 2. Pharmacophore hypotheses 1-3 (A) and 4 (B) generated by means of HypoGen from a training set of 16 R1D-AR
antagonists. (C) The final pharmacophore model for R1D-AR antagonists obtained by merging hypotheses 1 and 4 into a new
six-feature pharmacophore hypothesis. Features are color-coded as follows: blue for hydrophobics (HY); red for positive ionizable
groups (PI); green for hydrogen bond acceptors (HBA).
forcing the program to find pharmacophore hypotheses
characterized by at least five features.
On the basis of the summary of the computational
run and a hierarchical cluster analysis of the hypotheses
generated, a selection was made among the 10 phar-
macophores generated. In particular, hypotheses 1-3
showed high similarity in 3D spatial shape and were
therefore considered to be equivalent (Figure 2A). In
contrast, hypothesis 4, belonging to a different cluster,
showed a diverse composition in terms of chemical
features compared to hypotheses 1-3 (Figure 2B).
Moreover, it was shown for many compounds of the
training set that the same conformer of each compound
mapped both hypotheses 1 and 4 as the best fit.
Superposition of these hypotheses led to identification
of four pairs of chemical features located at the same
spatial positions. Both hypotheses showed two hydro-
phobics (HY1 and HY2) accommodating the ortho-
substituted phenyl ring bound to the piperazine and a
positive ionizable feature (PI) able to fit the N1 nitrogen
atom of the same heteroring. Also, in the two hypoth-
eses, the hydrogen bond acceptor group HBA2 is located
almost exactly at the same coordinates. The fixed and
null costs of this run were found to be 68.6 and 123.3,
respectively. For the first four five-feature hypothesis
models, the total costs were close to the fixed cost and
ranged from 70.9 (hypothesis 1) to 76.0 (hypothesis 4).
Finally, the two complementary five-feature hypoth-
eses were merged into a new six-feature model (referred
to as the final pharmacophore, Figure 2C). This model
is depicted in Figure 3 (with compound 4 superposed
as a representative example), while its geometric pa-
rameters (namely, distances and angles between phar-
macophoric features) are reported in Table 4. Although
the final model has an additional feature (HBA2)
compared to the old model, in agreement with the
HipHop calculations, its features are all matched by the
chemical groups of compound 4. Thus, the o-methoxy-
phenyl moiety maps both HY1 and HY2; the N1
nitrogen atom of the piperazine ring corresponds to the
F igu r e 3. Final pharmacophore model for R1D-AR antagonists
with compound 4 bound. Pharmacophore features are color-
coded as follows: blue for hydrophobics (HY); red for positive
ionizable groups (PI); green for hydrogen bond acceptors
(HBA).
Ta ble 4. Geometric Parameters (Distances and Angles
between Features) of the Final Pharmacophoric Model for
R
1D-AR Antagonists
feature distance, Å
feature
angle, deg
HY1-HY2
HY1-PI
HY1-HBA2
HY1-HBA1
HY1-HY3
HY2-PI
HY2-HBA2
HY2-HBA1
HY2-HY3
PI-HBA2
PI-HBA1
PI-HY3
HBA2-HBA1
HBA2-HY3
HBA1-HY3
4.1
7.5
11.0
12.6
16.5
5.9
HY1-HY2-PI
94.6
97.8
69.8
62.3
124.4
117.4
HY2-PI-HBA2
PI-HBA2-HBA1
HBA2-HBA1-HY3
HY2-PI-HY3
HY2-HBA2-HBA1
8.0
10.8
13.7
4.6
5.2
9.4
4.5
5.7
5.5
PI group; the two carbonyl groups are the HBAs; and
the condensed phenyl ring of the terminal heterocyclic
portion of the molecule maps HY3. This model also
accurately estimates the pKi value for 4 (9.48 versus
an experimental value of 9.44) and was submitted to
the Regress Hypothesis routine of Catalyst to allow it

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2885
to be used to estimate affinities of the training set, as
well as to predict affinities of compounds external to the
training set. In fact, a good correlation between esti-
mated and measured affinity values of the entire
training set was obtained, with a correlation coefficient
of 0.91.
This decreased affinity is a consequence of partially
obscuring the N4 atom (corresponding to HBA1) by the
ethyl substituent on the heteroring, with the program
predicting that the nitrogen atom is not sufficiently
uncrowded to have a perfect fit with HBA1.
Fourth, the length of the polymethylene chain linking
the arylpiperazine moiety and the terminal heterocyclic
fragment influenced receptor affinity, with an ethyl
spacer being the optimal length to bring the chemical
features of compounds to the appropriate distance for
interaction with the pharmacophore elements. Com-
pounds with a propyl spacer (26-28) showed lower fit
values for the model, accounting for their decreased
affinity relative to their ethyl counterparts (Table 1).
In addition, the new model accounts for major struc-
ture-activity relationships of the compounds in several
ways. The first is that conformationally constrained
compounds, where the (2-methoxyphenyl)piperazine
moiety was transformed into a tricyclic system, were
predicted to have low affinity due to the conformational
rearrangement of the phenyl ring relative to the pip-
erazine. In fact, our previous work in this field,42 in
agreement with literature reports, demonstrated that
twisted or orthogonal conformations of the piperazine
relative to the phenyl ring of the arylpiperazinyl moiety
of the ligands are important for R1-AR antagonism. As
a consequence, affinities of compounds 63 and 64 were
estimated and predicted to be 7.06 and 5.91, respec-
tively, mainly due to their lack of ability to map one of
the HY1 and HY2 features.
However, the affinity of compounds bearing an N-(4-
substituted phenyl)piperazine moiety was difficult to
rationalize on the basis of our pharmacophore model,
which allowed for a classification of such compounds in
two subclasses based on the structural properties of the
para substituent. As examples, compounds 41 and 42
with alkyl chains of four carbon atoms, similar to 4,
could match HY2 with their phenyl ring, while a
C-shaped conformation of the alkyl chain allowed the
terminal methyl moiety to correspond to HY1. Such an
orientation was possible on the basis of the conforma-
tional freedom of the piperazinylalkyl moiety, which
underwent a conformational rearrangement, while the
most basic nitrogen atom of the piperazine ring re-
mained fixed in three-dimensional space. Affinity values
of these compounds were estimated to be 7.43 and 6.96,
in good agreement with the experimentally observed
values of 7.20 and 6.61, respectively. On the other hand,
compounds bearing bulky para substituents, such as 40
(p-tert-butyl-substituted), or branched and relatively
short alkyl chains, such as 39 (p-isopropyl-substituted),
were underestimated by the pharmacophore model.
Calculated affinities for 39 and 40 were 5.36 and 6.77,
while experimental values of 7.45 and 7.70 were ob-
served, respectively. Such a discrepancy may be due to
a partial fit of the p-substituted phenyl ring into the
HY1/HY2 system, suggesting that while the current
pharmacophore model accounts for a large majority of
the SARs of our new compounds, it is yet unable to
rationalize variation in affinity due to different substit-
uents at the para position.
Second, transformation of the methoxy substituent of
4 into larger alkoxy groups, as well as hydroxy or chloro,
led to a decreased affinity accounted by the model as a
partial fit into the HY1-HY2 system. In fact, lengthen-
ing the 2-methoxy substituent of the arylpiperazinyl
moiety to an ethoxy group led to decreased affinity for
compounds 45 and 30 (calculated to be 8.14 and 7.70,
respectively) on the basis of a partial match between
the ethyl group and the HY1 feature of the model,
compared to a perfect fit of the methoxy substituent.
Similarly, compound 65 with a chlorine atom at the
ortho position instead of the methoxy group of 4 was
predicted to have an affinity of 8.82 and an experimental
value of 8.44, mainly due to the inability of the o-
chlorophenyl moiety to fit well both HY1 and HY2.
Finally, a slightly decreased affinity of 66 relative to 4
was found experimentally (9.20 versus 9.44, respec-
tively). Although the hydrophilic hydroxy group of 66
lies within the sphere representing HY1, this interaction
is not considered profitable by the program because of
the opposing hydrophobic/hydrophilic character of the
feature and the substituent. Therefore, HY1 is viewed
as a missing feature, leading to a predicted affinity of
8.32 for 66 versus the observed value of 9.20. These
findings confirm that the methoxy substituent at the
ortho position of the phenylpiperazine moiety optimizes
interaction with R1D-ARs, as demonstrated generally for
all R1-ARs.
Third, transformation of a pyrimido[5,4-b]indole-
(1H,3H)2,4-dione to a pyrimido[5,4-b]indole-(3H)4-one
system did not affect receptor affinity, as evidenced for
compounds 23 and 24 with respect to 4. Such com-
pounds showed an orientation in the model very similar
to that of 4, with a good fit into all the pharmacophore
features. In fact, the carbonyl moiety and the unsub-
stituted heterocyclic nitrogen atom of 23 and 24 cor-
respond to the HBA1 and HBA2 features of the model,
respectively. Calculated affinities for 23 and 24 were
9.46 and 9.38 versus experimental values of 9.14 and
9.39, respectively. Compound 25 showed similar inter-
actions with the pharmacophore and an estimated
affinity of 8.43 versus an experimental affinity of 8.16.
To further assess the validity and predictive power
of the pharmacophore hypothesis, we used molecules
outside the training set along with their reported
affinities for R1D-ARs. In addition to several pyrimido-
[5,4-b]indoles described above, compounds of this “test
set” were taken from the literature. Their predicted
affinities, calculated by the program Catalyst on the
basis of the final six-feature pharmacophore model for
R
1D-AR antagonists, are reported in Table 3.
In particular, compound 70 (Chart 2) showed molec-
ular fragments fulfilling all features of the pharma-
cophore model. In fact, the 3-methyl group, the nitrogen
atom at the 5-position, and the 7-carbonyl group of the
isoxazolo[3,4-d]pyridazin-7(6H)-one moiety matched HY3,
HBA1, and HBA2, respectively. Moreover, the PI group
was represented by the N1 nitrogen atom of the pip-
erazine ring, and the 2-methoxyphenyl substituent of
the molecule corresponded to the HY1 and HY2 features
of the model. In this orientation, the affinity of 70 for

2886 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
the R1D-AR, expressed as Ki, was predicted to be 1.6 nM
and was in good agreement with the experimental value
of 1.5 nM.37 The shortening of the ethyl chain to a
methylene spacer led to a partial superposition of 69
into the pharmacophoric model, with the methyl group
of the isoxazole ring being unable to reach HY3. Thus,
the predicted affinity of 69 was 1200 nM versus a
measured value of 316 nM.37
Finally, the model was able to predict Ki values of
BMY 7378 (3) and SNAP 8719 (67) (1.9 and 6.2 nM) in
good agreement with their experimentally observed
values of 6.3 and 1.6 nM, respectively.
antagonists identified structural features important for
affinity and selectivity of the new compounds. In
particular, (i) transformation of the (2-methoxyphenyl)-
piperazine moiety into a tricyclic system led to a
decrease in affinity predicted by the model to be due to
a conformational rearrangement of the phenyl ring
relative to the piperazine nucleus. (ii) Two hydrogen
bond acceptor groups are required for R1D-AR binding
properties. Such substituents, represented by the car-
bonyl moieties of the pyrimidindione ring or by both the
carbonyl and the unsubstituted nitrogen atom of the
pyrimidinone ring, matched HBA1 and HBA2 of the
pharmacophore model. (iii) The distance between the
phenylpiperazine and the terminal heterocyclic frag-
ment is crucial for R1D-AR affinity. An ethyl spacer is
the optimal chain to bring domains within the appropri-
ate distance to interact with the features of the phar-
macophore model. (iv) Although the pharmacophore
model is characterized by a good correlation coefficient
(r ) 0.91) and is able to rationalize the major SARs of
the new pyrimido[5,4-b]indole compounds, it was unable
to accurately account for structure-affinity relation-
ships based on different para substituents on the phenyl
ring bound to the piperazine. Consequently, the phar-
macophore hypothesis for R1D-AR antagonists reported
here should be considered as a preliminary model. We
are currently carrying out additional studies to further
refine this model and to better define the influence of
the substituents and substitution pattern on the phenyl
ring linked to the piperazine nucleus.
In contrast, both SKF 104856 (71) and discretamine
(68) showed a much lower predicted affinity than that
reported in the literature. This may be due to the
reduced overall size of these compounds compared to
the others studied, resulting in their inability to reach
all the predicted pharmacophore features. For example,
discretamine mapped only five features in its best
orientation into the model. While ring A and its methoxy
substituent corresponded to HY1 and HY2, the two
oxygen atoms at positions 9 and 10 (ring D) were HBA1
and HBA2 of the model, respectively. However, while
the PI feature was filled by the nitrogen atom, HY3
was completely absent, causing a predicted affinity
(14 000 nM) much lower than the experimental value
(25 nM). Such an orientation was in partial agreement
with a theoretical model of R1D-ARs recently reported
by Carotti and co-workers43 describing a hydrophobic
interaction between aromatic ring A of discretamine and
Phe358, a weak hydrogen bond (or a polar interaction)
between the hydroxy group at C10 and Cys240, and
finally, a salt bridge involving the nitrogen atom of
Moreover, it can be anticipated that the improvement
of this model is ongoing by incorporation of other R1D
-
AR antagonists into the training set to develop a
common pharmacophoric model for all the structural
classes of compounds that proved to be R1D-AR antago-
nists.
discretamine and Asp170
.
In a similar manner, SKF 104856 (71) mapped HY1,
HY2, and PI of the model with its unsaturated side
chain, thiophene ring, and basic nitrogen atom, respec-
tively. Both the hydrogen bond acceptors (HBA1 and
HBA2) and HY3 features of the model were omitted,
however, leading to a predicted affinity of 86 000 nM
versus an experimental value of 1.6 nM.
In summary, computational results from HipHop and
HypoGen calculations on compounds with widely vary-
ing affinities and selectivities toward R1D-ARs led to the
development of a pharmacophore model characterized
by a three-feature system accommodating the substi-
tuted phenylpiperazine moiety, as well as an additional
three-feature system interacting with the terminal
heterocyclic moiety of these compounds. Moreover, the
polymethylene chain appeared to serve only as a spacer,
bringing the two molecular domains in the correct
spatial orientation to profitably interact with these
receptors.
Finally, additional efforts have been planned to
highlight which pharmacophoric features are peculiar
for the R1D-AR recognition with respect to the other R1-
AR subtypes.
Exp er im en ta l Section
Ch em istr y. Melting points were determined in a Gallen-
kamp apparatus with a digital thermometer MFB-595 in glass
capillary tubes and are uncorrected. Infrared spectra were
recorded on a Perkin-Elmer FTIR 1600 spectrometer in KBr
disks. Elemental analyses for C, H, N, and S were within
(0.4% of theoretical values and were performed on a Carlo
Erba elemental analyzer model 1108 apparatus. 1H NMR
spectra were recorded on a Varian Inova Unity 200 spectrom-
eter (200 MHz for 1H NMR and 50 MHz for 13C NMR) in
DMSO-d6 solution. Chemical shifts are given in δ values (ppm),
using tetramethylsilane as the internal standard; coupling
constants (J ) are given in hertz. Signal multiplicities are
characterized as s (singlet), d (doublet), t (triplet), q (quartet),
sp (septet), m (multiplet), br (broad signal). All the synthesized
compounds were tested for purity on TLC (aluminum sheet
coated with silica gel 60 F254, Merck) and visualized by UV (λ
) 254 and 366 nm). All chemicals and solvents were reagent
grade and were purchased from commercial vendors.
Eth yl 3-[(Eth oxym eth ylid en )a m in o]-1H-in d ole-2-ca r -
boxyla te (6). 2-Ethoxycarbonyl-3-aminoindole 5 (2.0 g, 9.79
mmol) was dissolved in 10 mL of triethyl orthoformate, and
the reaction mixture was heated under reflux for 2 days. After
the mixture was cooled, the precipitate was filtered, washed
with cyclohexane, and dried. Recrystallization from cyclohex-
ane gave 6 (1.2 g, 47%) as a white powder: mp 146-148 °C;
IR (KBr) cm-1 3324 (NH), 1677 (CdO); 1H NMR (DMSO-d6) δ
Con clu sion s
A new series of pyrimido[5,4-b]indole derivatives was
prepared and tested in binding assays on the three
human cloned R1-AR subtypes. Most of the new com-
pounds showed a preferential affinity for the R1D-ARs
and some of them, such as 39 and 40, displayed a good
R
1D-AR selectivity with respect to the other two R1-AR
subtypes, as well as some other serotonergic and
dopaminergic receptors.
A structure-affinity relationship analysis based on
fitting a preliminary pharmacophore model for R1D-AR

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2887
1.30 (t, J ) 7.0 Ηz, 3 H, CH3), 1.37 (t, J ) 7.2 Hz, 3 H, CH3),
4.26 (q, J ) 7.0 Hz, 2 H, CH2), 4.37 (q, J ) 7.2 Hz, 2 H, CH2),
6.98-7.07 (m, 1 H, indole), 7.22-7.31 (m, 1 H, indole), 7.34-
7.41 (m, 1 H, indole), 7.48-7.54 (m, 1 H, indole), 8.03 (s, 1 H,
NdCH), 11.32 (br s, 1 H, NH which exchanges with D2O).
Anal. (C14H16N2O3) C, H, N.
Eth yl 3-[(Eth oxyeth ylid en )a m in o]-1H-in d ole-2-ca r box-
yla te (7). The same procedure, as described for the synthesis
of 6, was followed using triethyl orthoacetate. Recrystallization
from cyclohexane afforded 7 (77%) as a white powder: mp
138-139 °C; IR (KBr) cm-1 3318 (NH), 1666 (CdO); 1H NMR
(DMSO-d6) δ 1.28 (t, J ) 7.2 Hz, 3 H, CH2CH3), 1.34 (t, J )
7.2 Hz, 3 H, CH2CH3), 1.74 (s, 3 H, CCH3), 4.24 (q, J ) 7.2
Hz, 2 H, CH2CH3), 4.30 (q, J ) 7.2 Hz, 2 H, CH2CH3), 6.93-
7.04 (m, 1 H, indole), 7.20-7.40 (m, 3 H, indole), 11.25 (br s,
1 H, NH which exchanges with D2O). Anal. (C15H18N2O3) C,
H, N.
Eth yl 3-[(Eth oxyp r op ylid en )a m in o]-1H-in d ole-2-ca r -
boxyla te (8). The same procedure, as described for the
synthesis of 6, was followed using triethyl orthopropionate.
Recrystallization from cyclohexane afforded 8 (94%) as a white
powder: mp 162-164 °C; IR (KBr) cm-1 3310 (NH), 1666 (Cd
O); 1H NMR (DMSO-d6) δ 0.92 (t, J ) 7.6 Ηz, 3 H, CCH2CH3),
1.27 (t, J ) 7.0 Hz, 3 H, OCH2CH3), 1.34 (t, J ) 7.0 Hz, 3 H,
OCH2CH3), 2.06 (q, J ) 7.6 Hz, 2 H, CCH2CH3), 4.23 (q, J )
7.0 Hz, 2 H, OCH2CH3), 4.31 (q, J ) 7.0 Hz, 2 H, OCH2CH3),
6.95-7.03 (m, 1 H, indole), 7.20-7.39 (m, 3 H, indole), 11.24
(br s, 1H, NH which exchanges with D2O). Anal. (C16H20N2O3)
C, H, N.
3-(3-H yd r oxyp r op yl)-2-m et h yl-5H -p yr im id o[5,4-b]in -
d ole-(3H)4-on e (13). The same procedure, as described for
the synthesis of 9, was followed starting from 7 and 3-amino-
1-propanol. Recrystallization from toluene afforded 13 (80%):
mp 220-222 °C; IR (KBr) cm-1 3440 (broad, OH), 3189 (NH),
1677 (CdO); 1H NMR (DMSO-d6) δ 1.77-1.93 (m, 2 H,
CH2CH2CH2), 2.70 (s, 3 H, CH3), 3.36-3.56 (br m, 2 H, CH2-
OH), 4.21 (t, J ) 7.2 Hz, 2 H, NCH2), 4.71 (br s, 1 H, OH which
exchanges with D2O), 7.15-7.25 (m, 1 H, indole), 7.38-7.55
(m, 2 H, indole), 7.95-8.01 (m, 1 H, indole), 11.93 (br s, 1 H,
NH which exchanges with D2O). Anal. (C14H15N3O2) C, H, N.
3-(3-Hydr oxyp r op yl)-2-eth yl-5H-p yr im ido[5,4-b]in d ole-
(3H)4-on e (14). The same procedure, as described for the
synthesis of 9, was followed starting from 8 and 3-amino-1-
propanol. Recrystallization from toluene afforded 14 (77%): mp
183-184 °C; IR (KBr) cm-1 3151 (broad, OH + NH), 1671 (Cd
O); 1H NMR (DMSO-d6) δ 1.36 (t, J ) 7.6 Hz, 3 H, CH3), 1.75-
1.93 (m, 2 H, CH2CH2CH2), 2.98 (q, J ) 7.6 Hz, 2 H, CH2CH3),
3.47-3.55 (m, 2 H, CH2OH), 4.22 (t, J ) 7.6 Hz, 2 H, NCH2),
4.70 (t, J ) 7.2 Hz, 1 H, OH which exchanges with D2O), 7.15-
7.25 (m, 1 H, indole), 7.38-7.54 (m, 2 H, indole), 7.95-8.02
(m, 1 H, indole), 11.89 (br s, 1 H, NH which exchanges with
D2O). Anal. (C15H17N3O2) C, H, N.
3-(2-Ch lor oe t h yl)-5H -p yr im id o[5,4-b]in d ole -(3H )4-
on e Hyd r och lor id e (15). Alcohol 9 (0.7 g, 3.00 mmol) was
dissolved in 20 mL of toluene, and SOCl2 (0.73 g, 6.1 mmol)
was added. The reaction mixture was heated under reflux for
2 h. Then it was concentrated under reduced pressure.
Cyclohexane (20 mL) was added to the residue, and the
mixture was stirred for 30 min. The solid was filtered off,
washed with cyclohexane, and dried to give 15 (0.8 g, 89%).
This crude product was used without further purification: mp
240-242 °C; IR (KBr) cm-1 3198 (NH), 1674 (CdO); 1H NMR
(DMSO-d6) δ 4.02 (t, J ) 5.8 Hz, 2 H, CH2Cl), 4.46 (t, J ) 5.8
Hz, 2 H, NCH2), 7.20-7.32 (m, 1 H, indole), 7.40-7.58 (m, 2
H, indole), 7.98-8.10 (m, 1 H, indole), 8.33 (s, 1 H, NdCH),
3-(2-H yd r oxyet h yl)-5H -p yr im id o[5,4-b]in d ole-(3H )4-
on e (9). Compound 6 (1.2 g, 4.6 mmol) was dissolved in 5 mL
of 2-ethanolamine, and the reaction mixture was heated under
reflux for 7 h. After cooling, the reaction mixture was poured
into water (50 mL). The solid was filtered off, washed with
water, and dried. Recrystallization from EtOH gave 9 (0.8 g,
80%): mp 296-298 °C; IR (KBr) cm-1 3437 (OH), 3315 (NH),
1
12.21 (br s, 1 H, NH which exchanges with D2O). Anal. (C12H11
Cl2N3O) C, H, N.
-
1669 (CdO); H NMR (DMSO-d6) δ 3.70 (t, J ) 5.4 Hz, 2 H,
CH2OH), 4.16 (t, J ) 5.4 Hz, 2 H, NCH2), 4.95 (br s, 1 H, OH
which exchanges with D2O), 7.18-7.28 (m, 1 H, indole), 7.41-
7.57 (m, 2 H, indole), 7.97-8.04 (m, 1 H, indole), 8.21 (s, 1 H,
NdCH), 12.11 (br s, 1 H, NH which exchanges with D2O).
Anal. (C12H11N3O2) C, H, N.
3-(2-Ch lor oeth yl)-2-m eth yl-5H-p yr im id o[5,4-b]in d ole-
(3H)4-on e Hyd r och lor id e (16). The same procedure, as
described for the synthesis of 15, was followed starting from
alcohol 10. Crude 16 (88%) was used without further purifica-
1
tion: mp 224 °C; IR (KBr) cm-1 3226 (NH), 1710 (CdO); H
3-(2-Hydr oxyeth yl)-2-m eth yl-5H-pyr im ido[5,4-b]in dole-
(3H)4-on e (10). The same procedure, as described for the
synthesis of 9, was followed starting from 7. Recrystallization
from EtOH afforded 10 (72%): mp 251-252 °C; IR (KBr) cm-1
NMR (DMSO-d6) δ 2.86 (s, 3 H, CH3), 3.98 (t, J ) 6.6 Hz, 2 H,
CH2Cl), 4.51 (t, J ) 6.6 Hz, 2 H, NCH2), 7.22-7.32 (m, 1 H,
indole), 7.45-7.59 (m, 2 H, indole), 8.15-8.21 (m, 1 H, indole),
1
12.37 (br s, 1 H, NH which exchanges with D2O). Anal. (C13H13
-
3158 (broad, OH + NH), 1671 (CdO); H NMR (DMSO-d6) δ
Cl2N3O‚0.5H2O) C, H, N.
2.72 (s, 3 H, CH3), 3.65-3.77 (m, 2 H, CH2OH), 4.22 (t, J )
5.6 Hz, 2 H, NCH2), 5.02 (t, J ) 5.8 Hz, 1 H, OH which
exchanges with D2O), 7.14-7.24 (m, 1 H, indole), 7.37-7.54
(m, 2 H, indole), 7.94-8.00 (m, 1 H, indole), 11.89 (br s, 1 H,
NH which exchanges with D2O). Anal. (C13H13N3O2) C, H, N.
3-(2-Ch lor oet h yl)-2-et h yl-5H -p yr im id o[5,4-b]in d ole-
(3H)4-on e Hyd r och lor id e (17). The same procedure, as
described for the synthesis of 15, was followed starting from
alcohol 11. Crude 17 (92%) was used without further purifica-
tion: mp 243-245 °C; IR (KBr) cm-1 3158 (NH), 1728 (CdO);
1H NMR (DMSO-d6) δ 1.37 (t, J ) 7.6 Hz, 3 H, CH3), 3.17 (q,
J ) 7.6 Hz, 2 H, CH2CH3), 3.98 (t, J ) 6.8 Hz, 2 H, CH2Cl),
4.53 (t, J ) 6.8 Hz, 2 H, NCH2), 7.22-7.32 (m, 1 H, indole),
7.45-7.60 (m, 2 H, indole), 8.20-8.27 (m, 1 H, indole), 12.34
3-(2-Hyd r oxyeth yl)-2-eth yl-5H-p yr im id o[5,4-b]in d ole-
(3H)4-on e (11). The same procedure, as described for the
synthesis of 9, was followed starting from 8. Recrystallization
from toluene afforded 11 (77%): mp 208-209 °C; IR (KBr) cm-
1
1
3152 (broad, OH + NH), 1660 (CdO); H NMR (DMSO-d6) δ
(br s, 1 H, NH which exchanges with D2O). Anal. (C14H15
Cl2N3O) C, H, N.
-
1.33 (t, J ) 7.4 Hz, 3 H, CH3), 3.05 (q, J ) 7.4 Hz, 2 H, CH2-
CH3), 3.66-3.73 (m, 2 H, CH2OH), 4.23 (t, J ) 6.2 Hz, 2 H,
NCH2), 5.02 (t, J ) 5.6 Hz, 1 H, OH which exchanges with
D2O), 7.15-7.25 (m, 1 H, indole), 7.38-7.54 (m, 2 H, indole),
7.95-8.02 (m, 1 H, indole), 11.93 (br s, 1 H, NH which
exchanges with D2O). Anal. (C14H15N3O2) C, H, N.
3-(3-Hyd r oxyp r op yl)-5H-p yr im id o[5,4-b]in d ole-(3H)4-
on e (12). The same procedure, as described for the synthesis
of 9, was followed starting from 6 and 3-amino-1-propanol.
Recrystallization from EtOH afforded 12 (50%): mp 233-234
°C; IR (KBr) cm-1 3444 (broad, OH), 1660 (CdO); 1H NMR
(DMSO-d6) δ 1.78-2.02 (m, 2 H, CH2CH2CH2), 3.47 (t, J )
6.2 Hz, 2 H, CH2OH), 4.17 (t, J ) 7.2 Hz, 2 H, NCH2), 4.68 (br
s, 1 H, OH which exchanges with D2O), 7.18-7.28 (m, 1 H,
indole), 7.42-7.59 (m, 2 H, indole), 7.98-8.04 (m, 1 H, indole),
8.30 (s, 1 H, NdCH), 11.85 (br s, 1 H, NH which exchanges
with D2O). Anal. (C13H13N3O2) C, H, N.
3-(3-Ch lor op r op yl)-5H -p yr im id o[5,4-b]in d ole-(3H )4-
on e Hyd r och lor id e (18). The same procedure, as described
for the synthesis of 15, was followed starting from alcohol 12.
Crude 18 (85%) was used without further purification: mp
219-222 °C; IR (KBr) cm-1 3218 (NH), 1692 (CdO); 1H NMR
(DMSO-d6) δ 2.18-2.36 (m, 2 H, CH2CH2CH2), 3.75 (t, J )
6.4 Hz, 2 H, CH2Cl), 4.28 (t, J ) 7.2 Hz, 2 H, NCH2), 7.24-
7.34 (m, 1 H, indole), 7.46-7.62 (m, 2 H, indole), 8.10-8.18
(m, 1 H, indole), 8.76 (s, 1 H, NdCH), 12.46 (br s, 1 H, NH
which exchanges with D2O). Anal. (C13H13Cl2N3O) C, H, N.
3-(3-Ch lor opr opyl)-2-m eth yl-5H-pyr im ido[5,4-b]in dole-
(3H)4-on e Hyd r och lor id e (19). The same procedure, as
described for the synthesis of 15, was followed starting from
alcohol 13. Crude 19 (92%) was used without further purifica-
tion: mp 251-254 °C; IR (KBr) cm-1 3221 (NH), 1692 (CdO);

2888 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
1H NMR (DMSO-d6) δ 2.13-2.29 (m, 2 H, CH2CH2CH2), 2.96
(s, 3 H, CH3), 3.83 (t, J ) 6.4 Hz, 2 H, CH2Cl), 4.31 (t, J ) 7.2
Hz, 2 H, NCH2), 7.25-7.36 (m, 1 H, indole), 7.50-7.64 (m, 2
H, indole), 8.38-8.45 (m, 1 H, indole), 12.66 (br s, 1 H, NH
which exchanges with D2O). Anal. (C14H15Cl2N3O) C, H, N.
3-(3-Ch lor op r op yl)-2-eth yl-5H-p yr im id o[5,4-b]in d ole-
(3H)4-on e Hyd r och lor id e (20). The same procedure, as
described for the synthesis of 15, was followed starting from
alcohol 14. Crude 20 (92%) was used without further purifica-
tion: mp 220-225 °C; IR (KBr) cm-1 3212 (NH), 1689 (CdO);
1H NMR (DMSO-d6) δ 1.40 (t, J ) 7.2 Hz, 3 H, CH3), 2.13-
2.28 (m, 2 H, CH2CH2CH2), 3.18 (q, J ) 7.2 Hz, 2 H, CH2-
CH3), 3.84 (t, J ) 6.2 Hz, 2 H, CH2Cl), 4.31 (t, J ) 7.6 Hz, 2
H, NCH2), 7.24-7.34 (m, 1 H, indole), 7.47-7.62 (m, 2 H,
indole), 8.30-8.38 (m, 1 H, indole), 12.50 (br s, 1 H, NH which
exchanges with D2O). Anal. (C15H17Cl2N3O) C, H, N.
3-[2-[4-(2-Meth oxyp h en yl)p ip er a zin -1-yl]eth yl]-5H-p y-
r im id o[5,4-b]in d ole-(3H)4-on e (23). A mixture of compound
15 (0.5 g, 1.76 mmol) and 1-(2-methoxyphenyl)piperazine (21)
(1.7 g, 8.80 mmol) in a 10 mL flask was heated in an oil bath
at 140 °C for 30 min. After the mixture was cooled, the solid
mass was suspended in EtOH (10 mL) and then water (5 mL)
was added. The solid was filtered off, washed with water, and
dried. Recrystallization from EtOH gave 23 (0.5 g, 71%) as a
white powder: mp 219-220 °C; IR (KBr) cm-1 3154 (NH), 1665
(CdO); 1H NMR (DMSO-d6) δ 2.51-2.74 (m, 6 H, NCH2), 2.58-
3.01 (m, 4 H, NCH2), 3.76 (s, 3 H, OCH3), 4.25 (t, J ) 5.6 Hz,
2 H, CONCH2), 6.84-6.98 (m, 4 H, aromatic), 7.19-7.29 (m,
1 H, indole), 7.41-7.58 (m, 2 H, indole), 8.00-8.04 (m, 1 H,
indole), 8.28 (s, 1 H, NCH), 12.13 (br s, 1 H, NH which
exchanges with D2O). Anal. (C23H25N5O2) C, H, N.
3-[2-[4-(2-Meth oxyph en yl)piper azin -1-yl]eth yl]-2-m eth -
yl-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (24). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 16. Recrystallization from toluene afforded 24
(48%) as a white powder: mp 234-236 °C; IR (KBr) cm-1 3182
(NH), 1669 (CdO); 1H NMR (DMSO-d6) δ 2.63-2.76 (m, 6 H,
NCH2), 2.74 (s, 3 H, CH3), 2.95-2.97 (m, 4 H, NCH2), 3.72 (s,
3 H, OCH3), 4.29 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.85-6.96 (m,
4 H, aromatic), 7.15-7.25 (m, 1 H, indole), 7.39-7.55 (m, 2 H,
indole), 7.95-8.00 (m, 1 H, indole), 11.94 (br s, 1 H, NH which
exchanges with D2O). Anal. (C24H27N5O2) C, H, N.
(t, J ) 7.0 Hz, 2 H, CONCH2), 6.81-6.94 (m, 4 H, aromatic),
7.15-7.24 (m, 1 H, indole), 7.39-7.51 (m, 2 H, indole), 7.93-
8.00 (m, 1 H, indole), 11.91 (br s, 1 H, NH which exchanges
with D2O). Anal. (C25H29N5O2) C, H, N.
3-[3-[4-(2-Meth oxyph en yl)piper azin -1-yl]pr opyl]-2-eth yl-
5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (28). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 20. Recrystallization from MeOH/water afforded
28 (65%) as a white powder: mp 164-165 °C; IR (KBr) cm-1
3149 (NH), 1664 (CdO); 1H NMR (DMSO-d6) δ 1.37 (t, J )
7.2 Hz, 3 H, CH2CH3), 1.81-2.02 (m, 2 H, CH2CH2CH2), 2.36-
2.65 (m, 6 H, NCH2), 2.76-3.14 (m + q, 4 H + 2 H, ArNCH2
+ CH2CH3), 3.77 (s, 3 H, OCH3), 4.22 (t, J ) 7.0 Hz, 2 H,
CONCH2), 6.81-6.93 (m, 4 H, aromatic), 7.16-7.25 (m, 1 H,
indole), 7.40-7.55 (m, 2 H, indole), 7.97-8.04 (m, 1 H, indole),
11.93 (br s, 1 H, NH which exchanges with D2O). Anal.
(C26H31N5O2) C, H, N.
3-[2-[4-(2-E t h oxyp h en yl)p ip er a zin -1-yl]et h yl]-5H -p y-
r im id o[5,4-b]in d ole-(3H)4-on e (29). The same procedure, as
described for the synthesis of 23, was followed starting from
15 and 1-(2-ethoxyphenyl)piperazine 22. Recrystallization from
EtOH/water afforded 29 (40%) as a white powder: mp 174-
175 °C; IR (KBr) cm-1 3150 (NH), 1662 (CdO); 1H NMR
(DMSO-d6) δ 1.33 (t, J ) 7.2 Hz, 3 H, OCH2CH3), 2.50-2.77
(m, 6 H, NCH2), 2.82-3.12 (m, 4 H, ArNCH2), 4.00 (q, J ) 7.2
Hz, 2 H, OCH2CH3), 4.22 (t, J ) 6.0 Hz, 2 H, CONCH2), 6.75-
6.97 (m, 4 H, aromatic), 7.18-7.28 (m, 1 H, indole), 7.40-7.58
(m, 2 H, indole), 7.97-8.05 (m, 1 H, indole), 8.28 (s, 1 H, Nd
CH), 12.12 (br s, 1 H, NH which exchanges with D2O). Anal.
(C24H27N5O2) C, H, N.
3-[2-[4-(2-Eth oxyp h en yl)p ip er a zin -1-yl]eth yl]-2-m eth -
yl-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (30). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 16 and 22. Recrystallization from dimethylfor-
mamide afforded 30 (60%) as a white powder: mp 246-248
1
°C; IR (KBr) cm-1 3155 (NH), 1668 (CdO); H NMR (DMSO-
d6) δ 1.34 (t, J ) 7.0 Hz, 3 H, OCH2CH3), 2.50-2.82 (m + s, 6
H + 3 H, NCH2 + NdCCH3), 2.86-3.15 (m, 4 H, ArNCH2),
4.01 (q, J ) 7.0 Hz, 2 H, OCH2CH3), 4.29 (t, J ) 6.2 Hz, 2 H,
CONCH2), 6.75-7.01 (m, 4 H, aromatic), 7.13-7.27 (m, 1 H,
indole), 7.37-7.58 (m, 2 H, indole), 7.92-8.02 (m, 1 H, indole),
11.94 (br s, 1 H, NH which exchanges with D2O). Anal.
(C25H29N5O2) C, H, N.
3-[2-[4-(2-Eth oxyp h en yl)p ip er a zin -1-yl]eth yl]-2-eth yl-
5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (31). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 17 and 22. Recrystallization from dimethylfor-
mamide/water afforded 31 (60%) as a white powder: mp 218-
220 °C; IR (KBr) cm-1 3187 (NH), 1663 (CdO); 1H NMR
(DMSO-d6) δ 1.34 (t, J ) 7.2 Hz, 3 H, NdCCH2CH3), 1.38 (t,
J ) 6.8 Hz, 3 H, OCH2CH3), 2.55-2.82 (m, 6 H, NCH2), 2.86-
3.17 (m + q, 4 H + 2 H, ArNCH2 + NdCCH2CH3), 4.00 (q, J
) 6.8 Hz, 2 H, OCH2CH3), 4.30 (t, J ) 6.2 Hz, 2 H, CONCH2),
6.76-7.00 (m, 4 H, aromatic), 7.13-7.25 (m, 1 H, indole), 7.37-
7.57 (m, 2 H, indole), 7.92-8.03 (m, 1 H, indole), 11.94 (br s,
1 H, NH which exchanges with D2O). Anal. (C26H31N5O2) C,
H, N.
3-[2-[4-[4-(1-Meth yleth yl)p h en yl]p ip er a zin -1-yl]eth yl]-
5H-p yr im id o[5,4-b]in d ole-(1H,3H)2,4-d ion e (39). This pro-
cedure is presented as an example for the synthesis of
compounds 39-44. 4-(1-Methylethyl)aniline (13.5 g, 100 mmol),
bis(2-chloroethyl)amine hydrochloride (17.8 g, 100 mmol), and
potassium carbonate (13.8 g, 100 mmol) were added to 2-bu-
toxyethanol (60 mL), and the mixture was vigorously stirred
and refluxed for 30 h. After the mixture was cooled, water (100
mL) was added and the aqueous layer was separated from the
organic one. Ethyl acetate (100 mL) was added to the latter,
and the resulting solution was washed with 4 M aqueous
NaOH (100 mL) and then with brine (3 × 70 mL). Successively,
the organic layer was dried over sodium sulfate and the
solvents were evaporated in vacuo. The resulting crude oil,
4-[4-(1-methylethyl)phenyl]piperazine (33) (18.7 g), was suc-
cessively used for the synthesis of 39 without further purifica-
tion. However, a sample of 33 was converted to fumarate salt
3-[2-[4-(2-Meth oxyph en yl)piper azin -1-yl]eth yl]-2-eth yl-
5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (25). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 17. Recrystallization from MeOH/water afforded
25 (64%) as a white powder: mp 195-196 °C; IR (KBr) cm-1
3184 (NH), 1664 (CdO); 1H NMR (DMSO-d6) δ 1.37 (t, J )
7.4 Hz, 3 H, CH2CH3), 2.55-2.78 (m, 6 H, NCH2), 2.95-3.10
(m + q, 4 H + 2 H, ArNCH2 + CH2CH3), 3.77 (s, 3 H, OCH3),
4.30 (t, J ) 7.0 Hz, 2 H, CONCH2), 6.87-6.94 (m, 4 H,
aromatic), 7.16-7.25 (m, 1 H, indole), 7.44-7.55 (m, 2 H,
indole), 7.97-8.04 (m, 1 H, indole), 11.94 (br s, 1 H, NH which
exchanges with D2O). Anal. (C25H29N5O2) C, H, N.
3-[3-[4-(2-Met h oxyp h en yl)p ip er a zin -1-yl]p r op yl]-5H -
p yr im id o[5,4-b]in d ole-(3H)4-on e (26). The same procedure,
as described for the synthesis of 23, was followed starting from
18. Recrystallization from EtOH/water afforded 26 (36%) as
a white powder: mp 148-150 °C; IR (KBr) cm-1 3180 (broad,
1
NH), 1673 (CdO); H NMR (DMSO-d6) δ 1.81-2.03 (m, 2 H,
CH2CH2CH2), 2.25-2.60 (m, 6 H, NCH2), 2.72-3.03 (m, 4 H,
ArNCH2), 3.75 (s, 3 H, OCH3), 4.15 (t, J ) 7.0 Hz, 2 H,
CONCH2), 6.72-6.97 (m, 4 H, aromatic), 7.18-7.28 (m, 1 H,
indole), 7.39-7.56 (m, 2 H, indole), 7.97-8.04 (m, 1 H, indole),
8.32 (s, 1 H, NdCH), 12.10 (br s, 1 H, NH which exchanges
with D2O). Anal. (C24H27N5O2) C, H, N.
3-[3-[4-(2-Meth oxyph en yl)piper azin -1-yl]pr opyl]-2-m eth -
yl-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (27). The same pro-
cedure, as described for the synthesis of 23, was followed
starting from 19. Recrystallization from MeOH/water afforded
27 (40%) as a white powder: mp 195-197 °C; IR (KBr) cm-1
3183 (NH), 1654 (CdO); 1H NMR (DMSO-d6) δ 1.78-2.01 (m,
2 H, CH2CH2CH2), 2.35-2.63 (m, 6 H, NCH2), 2.71 (s, 3 H,
CCH3), 2.79-3.04 (m, 4 H, ArNCH2), 3.76 (s, 3 H, OCH3), 4.20

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2889
and characterized. A small portion (1.0 g) of the crude oil was
added to a saturated solution of fumaric acid in 2-propanol
(20 mL). A white solid precipitated, and after collection by
filtration, it was washed with cold 2-propanol (5 mL) and dried.
The precipitated salt was recrystallized from EtOH to give 33‚
0.5(fumaric acid) as white crystals (0.9 g): mp 203-205 °C
dec; IR (KBr) cm-1 3100, 2965, 2860, 2710 (broad bands, NH2+),
1.42 (m, 2 H, CH2CH2CH3), 2.29-2.59 (m, 8 H, CH2CH2CH2-
CH3 + NCH2), 2.90-3.00 (m, 4 H, NCH2), 4.00 (t, J ) 6.8 Hz,
2 H, CONCH2), 6.65-6.74 (m, 2 H, aromatic), 6.87-6.93 (m,
2 H, aromatic), 7.00-7.06 (m, 1 H, indole), 7.20-7.35 (m, 2 H,
indole), 7.80-7.87 (m, 1 H, indole), 11.67 (br s, 1 H, NH which
exchanges with D2O), 11.87 (br s, 1 H, NH which exchanges
with D2O). Anal. (C26H31N5O2) C, H, N.
1
1640, 1620 (CdO); H NMR (D2O) δ 1.02 (d, J ) 6.9 Hz, 6 H,
3-[2-[4-(4-Cycloh exylp h en yl)p ip er a zin -1-yl]eth yl]-5H-
p yr im id o[5,4-b]in d ole-(1H,3H)2,4-d ion e (43). Compound
43 was prepared according to the procedure presented for
compound 39. Starting 4-cyclohexylaniline and bis(2-chloro-
ethyl)amine hydrochloride were reacted to obtain intermediate
1-(4-cyclohexylphenyl)piperazine (37). Reaction between 32
and 37 gave a solid that was recrystallized from dimethylfor-
mamide to afford 43 as a pure product (53%): mp >300 °C;
CH3), 2.71 (sp, J ) 6.9 Hz, 1 H, CHCH3), 3.15-3.31 (m, 8 H,
CH2), 6.33 (s, 1 H, fumarate CH), 6.88-7.20 (m, 4 H, aromatic).
Anal. (C13H20N2‚0.5C4H4O4) C, H, N.
A mixture of N′-(2-chloroethyl)-N′-[3-(2-ethoxycarbonyl)-
indolyl]urea 32 (1.0 g, 3.23 mmol) and 3.3 g of the crude oil
33 previously obtained was heated in an oil bath at 140 °C for
2 h. After being cooled, the reaction mixture was treated with
warm EtOH (20 mL). The crude solid was filtered off, washed
with EtOH and successively with water, and dried. Recrys-
tallization from dimethylformamide afforded 39 as a white
powder (0.8 g, 57%): mp >300 °C; IR (KBr) cm-1 3160 (NH),
1705, 1625 (CdO); 1H NMR (DMSO-d6) δ 1.13 (d, J ) 6.8 Hz,
6 H, CHCH3), 2.50-2.68 (m, 6 H, NCH2), 2.75 (sp, J ) 6.8 Hz,
1 H, CHCH3), 2.93-3.16 (m, 4 H, ArNCH2), 4.09 (t, J ) 6.5
Hz, 2 H, CONCH2), 6.72-6.90 (m, 2 H, aromatic), 6.94-7.18
(m, 1 H + 2 H, indole + aromatic), 7.25-750 (m, 2 H, indole),
7.85-7.95 (m, 1 H, indole), 11.75 (br s, 1 H, NH which
exchanges with D2O), 11.90 (br s, 1 H, NH which exchanges
with D2O); 13C NMR (DMSO-d6) δ 24.10, 32.46, 37.37, 48.62,
52.89, 55.28, 112.75, 113.46, 114.74, 115.49, 119.48, 120.54,
125.87, 126.58, 126.93, 138.06, 138.77, 149.20, 150.97, 156.46.
Anal. (C25H29N5O2) C, H, N.
1
IR (KBr) cm-1 3160, 3100 (NH), 1717, 1624 (CdO); H NMR
(DMSO-d6) δ 1.00-1.51 (m, 5 H, cyclohexane), 1.55-1.89 (m,
5 H, cyclohexane), 2.26-2.45 (m, 1 H, cyclohexane), 2.59-2.61
(m, 6 H, NCH2), 3.03-3.15 (m, 4 H, NCH2), 4.14 (t, J ) 8.0
Hz, 2 H, CONCH2), 6.79-6.85 (m, 2 H, aromatic), 6.99-7.16
(m, 2 H + 1 H, aromatic + indole), 7.32-7.45 (m, 2 H, indole),
7.90-7.97 (m, 1 H, indole), 11.58 (br s, 1 H, NH which
exchanges with D2O), 11.95 (br s, 1 H, NH which exchanges
with D2O). Anal. (C28H33N5O2) C, H, N.
3-[2-[4-[4-(Cya n om eth yl)p h en yl]p ip er a zin -1-yl]eth yl]-
5H -p yr im id o[5,4-b]in d ole-(1H ,3H )2,4-d ion e (44). Com-
pound 44 was prepared according to the procedure presented
for compound 39. Starting 4-aminobenzyl cyanide and bis(2-
chloroethyl)amine hydrochloride were reacted to obtain inter-
mediate 1-[4-(cyanomethyl)phenyl]piperazine (38) as a crude
oil. Reaction between 32 and 38 gave a solid that was
recrystallized from dimethylformamide/water to afford 44 as
a pure product (58%): mp >300 °C; IR (KBr) cm-1 3160, 3102
(NH), 2250 (CN), 1714, 1622 (CdO); 1H NMR (DMSO-d6) δ
2.47-2.53 (m, 6 H, NCH2), 3.05-3.15 (m, 4 H, NCH2), 3.87 (s,
2 H, CH2CN), 4.11 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.82-6.88
(m, 2 H, aromatic), 6.89-7.19 (m, 2 H + 1 H, aromatic +
indole), 7.31-7.44 (m, 2 H, indole), 7.90-7.97 (m, 1 H, indole),
11.76 (br s, 1 H, NH which exchanges with D2O), 11.95 (br s,
1 H, NH which exchanges with D2O). Anal. (C24H24N6O2) C,
H, N.
3-[2-[4-[4-(1,1-Dim et h ylet h yl)p h en yl]p ip er a zin -1-yl]-
et h yl]-5H -p yr im id o[5,4-b]in d ole-(1H ,3H )2,4-d ion e (40).
Compound 40 was prepared according to the procedure
presented for compound 39. Starting 4-(1,1-dimethylethyl)-
aniline and bis(2-chloroethyl)amine hydrochloride were reacted
to obtain intermediate 1-[4-(1,1-dimethylethyl)phenyl]pipera-
zine (34) as a crude oil. Reaction between 32 and 34 gave a
solid that was recrystallized from dioxane to afford 40 as a
pure product (76%): mp >300 °C; IR (KBr) cm-1 3163, 3102
1
(NH), 1707, 1634 (CdO); H NMR (DMSO-d6) δ 1.22 (s, 9 H,-
CCH3), 2.48-2.52 (m, 6 H, NCH2), 3.00-3.10 (m, 4 H, NCH2),
4.11 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.80-6.86 (m, 2 H,
aromatic), 7.06-7.23 (m, 2 H + 1 H, aromatic + indole), 7.35-
7.46 (m, 2 H, indole), 7.91-7.97 (m, 1 H, indole), 11.76 (br s,
1 H, NH which exchanges with D2O), 11.95 (br s, 1 H, NH
which exchanges with D2O). Anal. (C26H31N5O2) C, H, N.
3-[2-[4-(2-E t h oxyp h en yl)p ip er a zin -1-yl]et h yl]-5H -p y-
r im id o[5,4-b]in d ole-(1H,3H)2,4-d ion e (45). A mixture of
compound 32 (0.4 g, 1.3 mmol) and 1-(2-ethoxyphenyl)-
piperazine (1.3 g, 6.50 mmol) 22 in a 10 mL flask was heated
in an oil bath at 140 °C for 30 min. After the mixture was
cooled, the solid mass was suspended in EtOH (5 mL). The
solid was filtered off, washed with water, and dried. Recrys-
tallization from dimethylformamide/water gave 45 (0.2 g, 36%)
as a white powder: mp 288-290 °C; IR (KBr) cm-1 3156 (NH),
3-[2-[4-[4-(1-Meth ylpr opyl)ph en yl]piper azin -1-yl]eth yl]-
5H -p yr im id o[5,4-b]in d ole-(1H ,3H )2,4-d ion e (41). Com-
pound 41 was prepared according to the procedure presented
for compound 39. Starting 4-(1-methylpropyl)aniline and bis-
(2-chloroethyl)amine hydrochloride were reacted to obtain
intermediate 1-[4-(1-methylpropyl)phenyl]piperazine (35) as
a crude oil. Reaction between 32 and 35 gave a solid that was
recrystallized from dioxane to afford 41 as a pure product
(69%): mp >300 °C; IR (KBr) cm-1 3162, 3102 (NH), 1709,
1
1713, 1627 (CdO); H NMR (DMSO-d6) δ 1.34 (t, J ) 7.0 Hz,
3 H, OCH2CH3), 2.50-2.78 (m, 6 H, NCH2), 2.81-3.15 (m, 4
H, ArNCH2), 4.00 (q, J ) 7.0 Hz, 2 H, OCH2CH3), 4.10 (t, J )
6.8 Hz, 2H, CONCH2), 6.75-6.98 (m, 4 H, aromatic), 7.03-
7.18 (m, 1 H, indole), 7.28-7.50 (m, 2 H, indole), 7.90-8.01
(m, 1 H, indole), 11.77 (br s, 1 H, NH which exchanges with
D2O), 11.97 (br s, 1 H, NH which exchanges with D2O). Anal.
(C24H27N5O3) C, H, N.
1
1624 (CdO); H NMR (DMSO-d6) δ 0.61(t, J ) 7.6 Hz, 3 H,
CH2CH3), 0.99 (d, J ) 7.0 Hz, 3 H, CHCH3), 1.27-1.44 (m, 2
H, CH2CH3), 2.26-2.48 (m, 7 H, NCH2 + CH), 2.75-3.10 (m,
4 H, NCH2 piperazine), 3.98 (t, J ) 6.8 Hz, 2 H, CONCH2),
6.63-6.75 (m, 2 H, aromatic), 6.80-6.91 (m, 2 H, aromatic),
6.93-7.03 (m, 1 H, indole), 7.18-7.33 (m, 2 H, indole), 7.75-
7.85 (m, 1 H, indole), 11.64 (br s, 1 H, NH which exchanges
with D2O), 11.84 (br s, 1 H, NH which exchanges with D2O).
Anal. (C26H31N5O2) C, H, N.
3-[2-[4-(2-Meth oxyp h en yl)p ip er id in -1-yl]eth yl]-5H-p y-
r im id o[5,4-b]in d ole-(1H,3H)2,4-d ion e (46). A mixture of
compound 32 (0.3 g, 0.87 mmol) and 4-(2-methoxyphenyl)-
piperidine (0.83 g, 4.35 mmol) in a 10 mL flask was heated in
an oil bath at 140 °C for 30 min. After the mixture was cooled,
the solid mass was suspended in EtOH (5 mL). The solid was
filtered off, washed with water, and dried. Recrystallization
from dimethylformamide/water gave 46 (0.21 g, 52%) as a
white powder: mp 278-280 °C; IR (KBr) cm-1 3157, 3103
(NH), 1711, 1625 (CdO); 1H NMR (DMSO-d6) δ 1.50-1.74 (m,
4 H, ArCHCH2), 1.98-2.20 (m, 2 H, NCH2), 2.57 (t, J ) 7.0
Hz, 2 H, CONCH2CH2), 2.75-2.96 (m, 1 H, ArCH), 2.98-3.16
(m, 2 H, NCH2), 3.77 (s, 1 H, OCH3), 4.09 (t, J ) 7.0 Hz, 2H,
CONCH2CH2), 6.82-6.98 (m, 2 H, aromatic), 7.06-7.22 (m, 2
H + 1 H, aromatic + indole), 7.33-7.48 (m, 2 H, indole), 7.89-
3-[2-[4-(4-Bu tylp h en yl)p ip er a zin -1-yl]eth yl]-5H-p yr im -
id o[5,4-b]in d ole-(1H,3H)2,4-d ion e (42). Compound 42 was
prepared according to the procedure presented for compound
39. Starting 4-butylaniline and bis(2-chloroethyl)amine hydro-
chloride were reacted to obtain intermediate 1-(4-butylphenyl)-
piperazine (36) as a crude oil. Reaction between 32 and 36
gave a solid that was recrystallized from dioxane to afford 42
-1
as a pure product (83%): mp >300 °C; IR (KBr) cm
3162,
1
3103 (NH), 1715, 1622 (CdO); H NMR (DMSO-d6) δ 0.77 (t,
J ) 6.8 Hz, 3 H, CH3), 1.12-1.20 (m, 2 H,CH2CH3), 1.32-

2890 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
8.99 (m, 1 H, indole), 11.75 (br s, 1 H, NH which exchanges
with D2O). Anal. (C24H26N4O3) C, H, N.
tallization from EtOH afforded 52 as a pure product (0.3 g,
67%): mp 275-277 °C; IR (KBr) cm-1 3187 (NH), 1665 (Cd
1
O); H NMR (DMSO-d6) δ 1.23 (s, 9 H, CCH3), 2.63-2.75 (m,
3-[2-[4-[4-(1-Meth yleth yl)p h en yl]p yp er a zin -1-yl]eth yl]-
ben zoth ien o[3,2-d ]p yr im id o-(1H,3H)2,4-d ion e (48). A mix-
ture of N′-(2-chloroethyl)-N′-[3-(2-ethoxycarbonyl)benzothienyl]-
urea 47 (0.5 g, 1.53 mmol) and 1.6 g of the crude oil 33,
prepared in the synthesis of compound 39, was heated in an
oil bath at 140 °C for 2 h. After being cooled, the reaction
mixture was treated with warm EtOH (20 mL). The crude solid
was filtered off, washed with EtOH and successively with
water, and dried. Recrystallization from dimethylformamide/
water afforded 48 as a pure product (0.6 g, 88%): mp >300
°C; IR (KBr) cm-1 3196 (NH), 1709, 1637 (CdO); 1H NMR
(DMSO-d6) δ 1.13 (d, J ) 7.2 Hz, 6 H, CH3), 2.51-2.65 (m, 6
H, NCH2), 2.76 (sp, J ) 7.2 Hz, 1 H, CHCH3), 2.95-3.15 (m,
4 H, NCH2), 4.08 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.78-6.85
(m, 2 H, aromatic), 7.01-7.08 (m, 2 H, aromatic), 7.40-7.67
(m, 2 H, aromatic), 8.06-8.12 (m, 1 H, aromatic), 8.34-8.40
(m, 1 H, aromatic), 12.54 (br s, 1 H, NH which exchanges with
D2O). Anal. (C25H28N4O2S) C, H, N, S.
3-[2-[4-[4-(1-Meth yleth yl)p h en yl]p ip er a zin -1-yl]eth yl]-
5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (49). A mixture of 15
(0.3 g, 1.21 mmol) and 1.5 g of the crude oil 33, prepared in
the synthesis of compound 39, was heated in an oil bath at
140 °C for 30 min. After being cooled, the reaction mixture
was treated with EtOH (10 mL). The crude solid was filtered
off, washed with EtOH, and dried. Recrystallization from
EtOH afforded 49 as a pure product (0.3 g, 60%): mp 254-
257 °C; IR (KBr) cm-1 3180 (NH), 1662 (CdO); 1H NMR
(DMSO-d6) δ 1.15 (d, J ) 7.0 Hz, 6 H, CHCH3), 2.07-2.82 (m
+ sp, 6 H + 1 H, NCH2 + CHCH3), 3.02-3.15 (m, 4 H, NCH2),
4.25 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.80-6.87 (m, 2 H,
aromatic), 7.03-7.10 (m, 2 H, aromatic), 7.22-7.28 (m, 1 H,
indole), 7.45-7.60 (m, 2 H, indole), 7.99-8.03 (m, 1 H, indole),
8.29 (s, 1 H, NCH), 12.13 (br s, 1 H, NH which exchanges with
D2O). Anal. (C25H29N5O) C, H, N.
3-[2-[4-[4-(1-Meth yleth yl)p h en yl]p ip er a zin -1-yl]eth yl]-
2-m eth yl-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (50). A mix-
ture of 16 (0.5 g, 1.68 mmol) and 1.7 g of the crude oil 33,
prepared in the synthesis of compound 39, was heated in an
oil bath at 140 °C for 30 min. After being cooled, the reaction
mixture was treated with EtOH (10 mL). The crude solid was
filtered off, washed with EtOH, and dried. Recrystallization
from toluene afforded 50 as a pure product (0.3 g, 42%): mp
259-261 °C; IR (KBr) cm-1 3152 (NH), 1662 (CdO); 1H NMR
(DMSO-d6) δ 1.14 (d, J ) 6.8 Hz, 6 H, CHCH3), 2.59-2.81 (m
+ s + sp, 6 H + 3 H + 1 H, NCH2 + CH3 + CHCH3), 3.01-
3.10 (m, 4 H, NCH2), 4.29 (t, J ) 6.6 Hz, 2 H, CONCH2), 6.81-
6.87 (m, 2 H, aromatic), 7.03-7.10 (m, 2 H, aromatic), 7.14-
7.24 (m, 1 H, indole), 7.38-7.53 (m, 2 H, indole), 7.93-7.99
(m, 1 H, indole), 11.92 (br s, 1 H, NH which exchanges with
D2O). Anal. (C26H31N5O) C, H, N.
3-[2-[4-[4-(1,1-Dim et h ylet h yl)p h en yl]p ip er a zin -1-yl]-
eth yl]-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (51). A mixture
of 15 (0.4 g, 1.61 mmol) and 1.7 g of the crude oil 34, prepared
in the synthesis of compound 40, was heated in an oil bath at
140 °C for 30 min. After being cooled, the reaction mixture
was treated with EtOH (10 mL). The crude solid was filtered
off, washed with EtOH, and dried. Recrystallization from
EtOH afforded 51 as a pure product (0.4 g, 46%): mp 262-
264 °C; IR (KBr) cm-1 3181 (NH), 1673 (CdO); 1H NMR
(DMSO-d6) δ 1.23 (s, 9 H, CCH3), 2.55-2.72 (m, 6 H, NCH2),
3.03-3.07 (m, 4 H, NCH2), 4.26 (t, J ) 6.6 Hz, 2 H, CONCH2),
6.81-6.87 (m, 2 H, aromatic), 7.18-7.28 (m, 2 H + 1 H,
aromatic + indole), 7.45-7.57 (m, 2 H, indole), 7.99-8.03 (m,
1 H, indole), 8.29 (s, 1 H, NCH), 12.12 (br s, 1 H, NH which
exchanges with D2O). Anal. (C26H31N5O) C, H, N.
6 H, NCH2), 2.74 (s, 3 H, CH3), 3.05-3.09 (m, 4 H, NCH2),
4.31 (t, J ) 6.6 Hz, 2 H, COCH2), 6.83-6.90 (m, 2 H, aromatic),
7.19-7.25 (m, 2 H + 1 H, aromatic + indole), 7.39-7.54 (m, 2
H, indole), 7.94-8.00 (m, 1 H, indole), 11.93 (br s, 1 H, NH
which exchanges with D2O). Anal. (C27H33N5O) C, H, N.
2-[2-Hyd r oxy-3-(2-n itr op h en oxy)p r op yl]isoin d ole-1,3-
d ion e (54). A mixture of 2-(2-nitrophenoxymethyl)oxirane (53)
(24.3 g, 0.12 mol), phthalimide (18.2 g, 0.12 mol), and pyridine
(1.2 mL) in 1-butanol (75 mL) was heated under reflux for 16
h. After being cooled, the suspension was decanted and the
solid was suspended in EtOH (70 mL) and stirred for 1 h. Then
the residue was filtered, washed with EtOH, and dried.
Recrystallization from EtOH afforded 54 as a pure product
(16.4 g, 39%): mp 128-129 °C; IR (KBr) cm-1 3529 (OH), 1710
(CdO); 1H NMR (DMSO-d6) δ 3.65-3.87 (m, 2 H, NCH2), 4.10-
4.31 (m, 2 H + 1 H, OCH2CHOH), 5.45 (d, J ) 5.4 Hz, 1 H,
CHOH which exchanges with D2O), 7.05-7.15 (m, 1 H,
aromatic), 7.30-7.39 (m, 1 H, aromatic), 7.58-7.73 (m, 1 H,
aromatic), 7.75-7.93 (m, 4 H + 1 H, isoindole + aromatic).
Anal. (C17H14N2O6) C, H, N.
2-[3-(2-Nit r op h en oxy)-2-oxyp r op yl]isoin d ole-1,3-d i-
on e (55). A solution of alcohol 54 (9.6 g, 28.04 mmol) in
dichloromethane (390 mL) was added, dropwise at 60 °C, to a
solution of Dess-Martin periodinane (27.2 g, 64.13 mmol) in
anhydrous DMSO (20 mL). After 3 h at 60 °C, the reaction
mixture was stirred at room temperature overnight. Then a
solution of sodium thiosulfate (72.3 g) in 5% aqueous NaHCO3
(900 mL) and chloroform (900 mL) were added. The organic
layer was separated, washed with NaHCO3 (2 × 600 mL) and
water (600 mL), and dried over anhydrous sodium sulfate. The
solvents were eliminated in vacuo and the residue was
recrystallized from EtOH to afford 55 as a pure product (8.3
g, 87%): mp 160-161 °C; IR (KBr) cm-1 1722 (CdO); 1H NMR
(DMSO-d6) δ 4.81 (s, 2 H, NCH2), 5.34 (s, 2 H, OCH2), 7.07-
7.22 (m, 1 H, aromatic), 7.23-7.37 (m, 1 H, aromatic), 7.58-
7.75 (m, 1 H, aromatic), 7.80-8.05 (m, 4 H + 1 H, isoindole +
aromatic). Anal. (C17H12N2O6) C, H, N.
2-(3,4-Dih yd r o-2H-ben zo[1,4]oxa zin -1-ylm eth yl)isoin -
d ole-1,3-d ion e (56). To a well-stirred suspension of ketone
55 (8.3 g, 24.33 mmol) in absolute EtOH (1200 mL) under N2,
10% Pd/C (2.4 g) was added carefully. Then H2 was fluxed in
the reaction mixture, which was stirred at room temperature
for 24 h. Successively, the catalyst was filtered off and the
solvent was evaporated in vacuo. The light-yellow residue was
suspended in cyclohexane (80 mL) and stirred for 12 h. Then
the solid was filtered off and recrystallized from EtOH to give
56 as a pure product (5.8 g, 81%): mp 137-139 °C; IR (KBr)
cm-1 3348 (NH), 1713 (CdO); 1H NMR (DMSO-d6) δ 3.55-
3.80 (m, 2 H + 1 H, NCH2CHNH), 3.93-4.15 (m, 2 H, OCH2-
CH), 6.10 (br s, 1 H, NH which exchanges with D2O), 6.40-
6.57 (m, 2 H, aromatic), 6.59-7.75 (m, 2 H, aromatic), 7.75-
7.97 (m, 4 H, isoindole). Anal. (C17H14N2O3) C, H, N.
C-(3,4-Dih yd r o-2H -b e n zo[1,4]oxa zin -3-yl)m e t h a n a -
m in e (57). Compound 56 (5.8 g, 19.88 mmol) was dissolved
in warm EtOH (150 mL). Then hydrate hydrazine (3.1 g, 61.63
mmol) was added and the reaction mixture was heated under
reflux and stirred for 3 h. After the mixture was cooled, the
solvent was evaporated in vacuo and the solid residue was
suspended in chloroform and stirred overnight. Then the
suspension was filtered and the chloroform was evaporated
to afford a light-yellow oil, which slowly solidified. This residue
(3.2 g, 98%) was used for the succesive step without further
purification: mp 79-82 °C; IR (KBr) cm-1 3363 (broad, NH2,
NH); 1H NMR (DMSO-d6) δ 2.60 (d, J ) 6.2 Hz, 2 H, CHCH2-
NH2), 3.09-3.26 (m, 1 H, CHCH2NH2), 3.30 (br s, 2 H, NH2
3-[2-[4-[4-(1,1-Dim et h ylet h yl)p h en yl]p ip er a zin -1-yl]-
eth yl]-2-m eth yl-5H-p yr im id o[5,4-b]in d ole-(3H)4-on e (52).
A mixture of 16 (0.3 g, 1.00 mmol) and 1.1 g of the crude oil
34, prepared in the synthesis of compound 40, was heated in
an oil bath at 140 °C for 30 min. After being cooled, the
reaction mixture was treated with EtOH (10 mL). The crude
solid was filtered off, washed with EtOH, and dried. Recrys-
2
3
which exchanges with D2O), 3.81 (dd, J ) 10.6 Hz, J ) 7.0
Hz, 1 H, OCHAHBCHCH2NH2), 4.15 (dd, 2J ) 10.6 Hz, 3J )
2.4 Hz, 1H, OCHAHBCHCH2NH2), 5.78 (br s, 1 H, NH which
exchanges with D2O), 6.36-6.50 (m, 1 H, aromatic), 6.53-6.72
(m, 3 H, aromatic); 1H NMR (CDCl3) δ 1.51 (br s, 3 H, NH2
+
2
3
NH which exchanges with D2O), 2.70 (dd, J ) 12.6 Hz, J )

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2891
2
8.0 Hz, 1 H, OCHAHBCHCHAHBNH2), 2.90 (dd, J ) 12.6 Hz,
over anhydrous sodium sulfate and evaporated in vacuo. The
oily residue 61 (2.6 g, 93%) slowly solidified and was used for
the successive step without further purification: mp 88-90
3J ) 4.8 Hz, 1 H, OCHAHBCHCHAHBNH2), 3.28-3.42 (m, 1
H, OCHAHBCHCHAHBNH2), 3.95 (dd, J ) 10.8 Hz, J ) 6.6
Hz, 1 H, OCHAHBCHCHAHBNH2), 4.19 (dd, J ) 10.8 Hz, J
) 2.8 Hz, 1 H, OCHAHBCHCHAHBNH2), 6.56-6.68 (m, 2 H,
aromatic), 6.70-6.81 (m, 2 H, aromatic). Anal. (C9H12N2O) C,
H, N.
Ben zyl (3,4-Dih yd r o-2H-ben zo[1,4]oxa zin -3-ylm eth yl)-
ca r ba m a te (58). A solution of benzyl chloroformate (4.0 g,
23.45 mmol) in anhydrous THF (100 mL) was added, under
N2, dropwise at -78 °C to a well-stirred solution of amine 57
(3.2 g, 19.49 mmol) and triethylamine (2.9 g, 29.23 mmol) in
anhydrous THF (100 mL). Then the reaction mixture was
stirred at room temperature for 3 days. Successively, a
saturated solution of aqueous NaHCO3 (150 mL) and the
mixture was extracted with chloroform (4 × 100 mL). The
organic layer was dried over anhydrous sodium sulfate and
evaporated in vacuo to afford 58 as a brown oil (5.8 g, 99%),
which was used for the successive step without further
purification: IR (KBr) cm-1 3347 ((NH), 1702 (CdO); 1H NMR
(DMSO-d6) δ 2.85-3.22 (m, 2 H, OCHAHBCHCH2NHCO),
3.25-3.43 (m, 1 H, OCHAHBCHCH2NHCO), 3.84 (dd, 2J ) 10.8
2
3
2
3
1
°C; IR (KBr) cm-1 1701 (CdO); H NMR (DMSO-d6) δ 2.55-
2.81 (m, 2 H), 2.88-3.17 (m, 2 H), 3.75-4.16 (m, 4 H), 4.25-
4.40 (m, 1 H), 5.12 (s, 2 H, COOCH2), 6.57-6.98 (m, 4 H,
aromatic), 7.23-7.47 (m, 5 H, aromatic). Anal. (C19H20N2O3)
C, H, N.
1,2,3,4,4a ,5-H exa h yd r o-1H -p yr a zin o[2,1-c][1,4]b en z-
oxa zin e (62). A mixture of 10% Pd/C (2.6 g), 1,4-cyclohexa-
diene (6.4 g, 80.24 mmol), and trifluoroacetic acid (2.7 g, 24.03
mmol) was added under N2 to a cold (at 0 °C) suspension of
carbamate 61 (2.6 g, 8.01 mmol) in a MeOH/water (9:1, v/v)
(300 mL) mixture. The reaction mixture was slowly heated at
room temperature and stirred under N2 for 24 h. After, the
catalyst was filtered off and the solvents were evaporated in
vacuo. Then 10% aqueous sodium carbonate (150 mL) was
added to the residue and the mixture was extracted with
chloroform (5 × 60 mL). The organic layer was dried over
anhydrous sodium sulfate and concentrated in vacuo. The oily
residue 62 (1.1 g, 73%) slowly solidified and was used for the
successive steps without further purification: mp 182-185 °C;
IR (KBr) cm-1 3307 (NH); 1H NMR (DMSO-d6) δ 2.15-2.36
(m, 1 H, OCHAHBCHCHAHBNH), 2.38-2.54 (m, 1 H, ArN-
CHAHBCHAHBNH), 2.55-2.78 (m + br s, 1 H + 1 H, ArN-
CHAHBCHAHBNH + ArNCHAHBCHAHBNH which exchanges
with D2O), 2.80-3.06 (m, 1 H + 1 H + 1 H, ArNCHAHBCHAHB-
NH + OCHAHBCHCHAHBNH), 3.50-3.66 (m, 1 H, ArNCHAHB-
CHAHBNH), 3.83 (dd, 2J ) 10.6 Hz, 3J ) 8.8 Hz, 1 H,
3
2
Hz, J ) 5.4 Hz, 1 H, OCHAHBCHCH2NHCO), 4.05 (dd, J )
10.8 Hz, J ) 2.6 Hz, 1 H, OCHAHBCHCH2NHCO), 5.03 (s, 2
H, COOCH2), 5.86 (br s, 1 H, ArNH which exchanges with
D2O), 6.39-6.51 (m, 1 H, aromatic), 6.54-6.73 (m, 2 H,
aromatic), 7.22-7.43 (m, 5 H, aromatic), 7.48 (br t, J ) 5.6
Hz, 1 H, CH2NHCO which exchanges with D2O). Anal.
(C17H18N2O3) C, H, N.
3
2
3
OCHAHBCHCHAHBNH), 4.18 (dd, J ) 10.6 Hz, J ) 2.8 Hz,
1 H, OCHAHBCHCHAHBNH), 6.55-6.91 (m, 4 H, aromatic);
13C NMR (DMSO-d6) δ 45.07 (CH2), 45.96 (CH2), 46.28 (CH2),
52.46 (NCH), 67.14 (OCH2), 112.99 (CH), 115.83 (CH), 118.88
(CH), 121.13 (CH), 135.92 (C), 144.62 (C). Anal. (C11H14N2O)
C, H, N.
Ben zyl [4-(2-Ch lor oa cetyl)-3,4-d ih yd r o-2H-ben zo[1,4]-
oxa zin -3-m eth yl]ca r ba m a te (59). A solution of chloroacetyl
chloride (2.6 g, 23.02 mmol) in dichloromethane (100 mL) was
added dropwise at -78 °C, under N2, to a solution of the amine
58 (5.8 g, 19.44 mmol) and triethylamine (3.3 g, 32.61 mmol)
in dichloromethane (300 mL), and the reaction mixture was
stirred at room temperature for 24 h. Then water (500 mL)
was added and the organic layer was washed with 1 M HCl (2
× 100 mL), dried over anhydrous sodium sulfate, and evapo-
rated in vacuo. Recrystallization of the brown solid residue
from toluene afforded 59 as a pure product (6.6 g, 90%): mp
157-158 °C; IR (KBr) cm-1 3351 (NH), 1697, 1652 (CdO); 1H
NMR (DMSO-d6) δ 2.95-3.30 (m, 2 H), 4.05-4.17 (m, 1 H),
4.27-4.77 (m, 4 H), 4.97 (s, 2 H, COOCH2), 6.82-6.98 (m, 2
H, aromatic), 7.00-7.15 (m, 1 H, aromatic), 7.22-7.43 (m, 5
H, aromatic), 7.62 (br s, 1 H, CH2NHCO which exchanges with
D2O), 7.75-7.96 (m, 1 H, aromatic). Anal. (C19H19ClN2O4) C,
H, N.
3-[2-[1,2,3,4,4a ,5-H exa h yd r o-1H -p yr a zin o[2,1-c][1,4]-
ben zoxazin -3-yl]eth yl]-5H-pyr im ido[5,4-b]in dole-(1H,3H)-
2,4-d ion e (63). A mixture of 32 (0.2 g, 0.80 mmol) and amine
62 (0.5 g, 2.89 mmol) was heated in an oil bath at 140 °C for
15 min. After being cooled, the reaction mixture was treated
with EtOH (15 mL). The crude solid was filtered off, washed
with water, and dried. Recrystallization from dimethylforma-
mide/water afforded 63 as a pure product (0.1 g, 45%): mp
>300 °C; IR (KBr) cm-1 3151 (NH), 1701, 1636 (CdO); 1H NMR
(DMSO-d6) δ 1.70-1.92 (m, 1 H, OCHAHBCHCHAHBN), 2.28-
2.33 (m, 1 H, ArNCHAHBCHAHBN), 2.45-2.72 (m, 1 H + 2 H,
ArNCHAHBCHAHBN + NCH2CH2NCO), 2.86-3.21 (m, 1 H +
1 H + 1 H, ArNCHAHBCHAHBN + OCHAHBCHCHAHBN),
3.58-3.77 (m, 1 H, ArNCHAHBCHAHBN), 3.79-3.98 (m, 1 H,
OCHAHBCHCHAHBNH), 4.04-4.33 (m, 1 H + 2 H, OCHAHB-
CHCHAHBN + NCH2CH2NCO), 6.55-6.91 (m, 4 H, aromatic),
6.99-7.21 (m, 1 H, indole), 7.27-7.50 (m, 2 H, indole), 7.86-
8.01 (m, 1 H, indole), 11.78 (br s, 1 H, NH which exchanges
with D2O), 11.97 (br s, 1 H, NH which exchanges with D2O);
13C NMR (DMSO-d6) δ 37.27 (CH2), 45.46 (CH2), 51.70 (NCH),
52.28 (CH2), 53.48 (CH2), 55.17 (CH2), 66.94 (CH2), 112.77
(CH), 113.10 (CH), 113.47 (C), 114.74 (C), 115.73 (CH), 118.91
(CH), 119.52 (CH), 120.55 (CH), 121.03 (CH), 125.89 (C),
126.97 (CH), 135.28 (C), 138.06 (C), 144.44 (C), 150.98 (CO),
156.48 (CO). Anal. (C23H23N5O3) C, H, N.
Ben zyl 1,2,3,4,4a ,5-Hexa h yd r o-1H-p yr a zin o[2,1-c][1,4]-
ben zoxa zin e-1-on e-3-ca r boxyla te (60). A suspension of 59
(6.6 g, 17.71 mmol) and anhydrous potassium carbonate (7.7
g, 55.71 mmol) in anhydrous DMSO (400 mL) was stirred
under N2 for 2 days at room temperature. Then water (1000
mL) was added and the mixture was extracted with ethyl
acetate (4 × 100 mL). The organic layer was dried over
anhydrous sodium sulfate and evaporated in vacuo. The brown
oil residue was purified by flash chromatography using ethyl
acetate/cyclohexane (1:1, v/v) as eluent. The homogeneous
fractions were evaporated in vacuo to afford 60 as a pale-yellow
solid product (2.9 g, 49%): mp 111-113 °C; IR (KBr) cm-1
1
1691, 1668 (CdO); H NMR (DMSO-d6) δ 3.20-3.42 (m, 1 H,
OCH2CHCH2NCOO), 3.82-4.25 (m, 4 H), 4.27-4.52 (m, 2 H),
5.13 (s, 2 H, COOCH2), 6.85-6.98 (m, 2 H, aromatic), 7.00-
7.16 (m, 1 H, aromatic), 7.23-7.47 (m, 5 H, aromatic), 8.10-
8.21 (m, 1 H, aromatic). Anal. (C19H18N2O4) C, H, N.
3-[2-[1,2,3,4,4a ,5-H exa h yd r o-1H -p ir a zin o[2,1-c][1,4]-
ben zoxazin -3-yl]eth yl]-2-m eth yl-5H-pyr im ido[5,4-b]in dole-
(1H,3H)2,4-d ion e (64). A mixture of 16 (0.1 g, 0.32 mmol)
and amine 62 (0.2 g, 1.17 mmol) was heated in an oil bath at
140 °C for 15 min. After being cooled, the reaction mixture
was treated with EtOH (2 mL). The crude solid was filtered
off, washed with EtOH, and dried. Recrystallization from
dimethylformamide/water afforded 64 as a powder (0.04 g,
40%): mp 278-280 °C dec; IR (KBr) cm-1 3181 (NH), 1664
Ben zyl 1,2,3,4,4a ,5-Hexa h yd r o-1H-p yr a zin o[2,1-c][1,4]-
ben zoxa zin e-3-ca r boxyla te (61). Borane-THF complex (1
M solution in THF, 42.9 mL) was added dropwise at 0 °C and
under N2 to a solution of lactam 60 (2.9 g, 8.63 mmol) in
anhydrous THF (150 mL). Then the reaction mixture was
stirred at room temperature for 24 h. Successively, the reaction
mixture was cautiously poured in ice/water (500 mL), acidified
to pH 3.5 with 1 M HCl, and stirred for 30 min. The suspension
was basified to pH 10 with potassium carbonate and extracted
with chloroform (4 × 100 mL). The organic layer was dried
1
(CdO); H NMR (DMSO-d6) δ 1.79-2.01 (m, 1 H, OCHAHB-
CHCHAHBN), 2.19-2.35 (m, 1 H, ArNCHAHBCHAHBN), 2.47-
2.71 (m, 1 H + 2 H, ArNCHAHBCHAHBN + NCH2CH2NCO),
2.74 (s, 3 H, CH3), 2.94-3.20 (m, 1 H + 1 H + 1 H, ArNCHAHB-
CHAHBN + OCHAHBCHCHAHBN), 3.62-3.80 (m, 1 H, ArN-

2892 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
CHAHBCHAHBN), 3.81-3.96 (m, 1 H, OCHAHBCHCHAHBNH),
4.12-4.40 (m, 1 H + 2 H, OCHAHBCHCHAHBN + NCH2CH2-
NCO), 6.58-6.91 (m, 4 H, aromatic), 7.13-7.28 (m, 1 H,
indole), 7.36-7.58 (m, 2 H, indole), 7.92-8.04 (m, 1 H, indole),
11.93 (br s, 1 H, NH which exchanges with D2O). Anal.
(C24H25N5O2) C, H, N.
spectrometer. Dose-inhibition curves were analyzed by the
“Allfit”49 program to obtain the concentration of unlabeled
drugs that inhibited ligand binding by 50%. The Ki values were
derived from the IC50 values.50
Estim a tion of In ositol P h osp h olip id Hyd r olysis. Spra-
gue-Dawley rats (about 200 g) were killed by decapitation.
The brains were rapidly removed and dissected on ice. Hip-
pocampi were sliced (350 µm × 350 µm) with a McIlwain tissue
chopper, and the slices were immediately suspended in Krebs-
Hensleit buffer (118 mM NaCl, 4.7 mM KCl, 1.3 mM CaCl2,
1.2 mM K2HPO4, 1.2 mM MgSO4, 25 mM NaHCO3, 11.7 mM
glucose, equilibrated with 95% O2/5% CO2 to raise the pH to
7.4) and incubated at 37 °C for 30 min with three intermediate
changes of the buffer. Forty microliters of gravity-packed slices
were then transferred to 3 mL vials containing 0.3 µM myo-
[2-3H]inositol (New England Nuclear, specific activity of 16.5
Ci/mmol) in a final volume of 275 µL. After 60 min of
incubation, LiCl (7 mM) was added, followed by 100 µM
norepinephrine 10 min later. When present, test compounds
were added 5 min prior to norepinephrine. After 60 min, the
slices were washed three times with buffer and the reaction
(cleavage of inositol phosphates from membrane phospholipids)
was stopped by addition of 0.9 mL of chloroform/methanol (1/
2, v/v). The [3H]inositol monophosphates present in the aque-
ous phase were extracted by anion exchange chromatography
and measured as described previously.51
Bin d in g Exp er im en ts on Hu m a n Clon ed R1AR Su b-
typ es. Tr a n sfection a n d Cell Cu ltu r e. HEK293 cells were
transfected with the constitutively active pRSVICAT vectors
containing the human R1AAR,4
R
1BAR,3 or R1DAR5 cDNA by
calcium phosphate transfection.46 Cells were propagated for
several weeks in the presence of 400 µg/mL gentamycin, and
subclones were screened by radioligand binding for high
receptor expression. Transfected HEK293 cells were propa-
gated in 75 cm2 flasks at 37 °C in a humified 5% CO2 incubator
in Dulbecco’s modified Eagle’s medium containing 4.5 g/L
glucose, 1.4% glutamine, 20 mM HEPES, 100 mg/L strepto-
mycin, 105 units/L penicillin, and 10% calf serum. The cells
were detached by trypsinization and subcultured at a ratio of
1:4 upon reaching confluency.
Ra d ioliga n d Bin d in g. Confluent 100 mm plates were
washed with phosphate-buffered saline (20 mM NaPO4, 154
mM NaCl, pH 7.6) and harvested by scraping. Cells were
collected by centrifugation and homogenized with a Polytron.
Cell membranes were collected by centrifugation at 30000g
for 10 min and resuspended by homogenization. Receptor
density was determined by saturation analysis of the R1-AR
specific antagonist radioligand [125I]BE 2254 (20-800 pM).33
For analysis of competition by selective drugs, 50 pM radio-
ligand was used. Curves were analyzed by nonlinear regression
analysis using GraphPad Prism.47 Nonspecific binding was
determined in the presence of 10 µM phentolamine.
Bin d in g E xp er im en t s on 5-HT1A, 5-HT1B, 5-HT2A, D1,
a n d D2 Recep tor s. Binding assays were performed on male
CRL:CD(SD)BR-COBS rats weighing about 150 g. The animals
were killed by decapitation, and their brains were rapidly
dissected (hippocampus for 5-HT1A; striatum for 5-HT1B, D1,
D2; cortex for 5-HT2A), frozen, and stored at -80 °C until the
day of assay.
Com p u ta tion a l Meth od s. Calculations and graphic ma-
nipulations were performed on a Silicon Graphics Octane
workstation by means of the software Catalyst 4.6.
Compounds reported in the paper were built using the two-
and three-dimensional sketcher of the program. A representa-
tive family of conformations was generated for each molecule
using the poling algorithm and the “best quality conforma-
tional analysis” method, based on the CHARMm force field.
Conformations were collected that fell within a 20 kcal/mol
range above the lowest-energy conformation that was found.
BMY 7378 (3), SNAP 8719 (67), and discretamine (68) with
their associated conformational models were submitted to
common feature hypothesis generation (Catalyst HipHop) with
the aim of producing pharmacophore models by generating
alignments of common chemical features. SNAP 8719 (67), the
most active R1D-AR ligand and selective R1A/R1D antagonist of
this series, was considered as the “reference compound” (except
for this classification, biological data in the analysis were not
used) specifying a “Principal” value of 2 and a “MaxOmitFeat”
value of 0 during hypotheses generation. HipHop uses these
values to determine which molecule should be considered to
build hypothesis space and how many features in the final
hypotheses must map the chemical feature in each compound,
respectively. If Principal is set to 2, the chemical feature space
of the conformers of such a compound is used to define the
initial set of potential hypotheses, while a MaxOmitFeat value
of 0 associated with the reference compound forces it to map
all the features of each pharmacophore hypothesis generated.
On the other hand, 16 pyrimido[5,4-b]indole derivatives
(namely, 4, 23, 25, 26, 28, 40-45, 48, 50, 52, 63, and 65), with
their associated conformational models and R1D-AR affinity
values, were submitted to Catalyst HypoGen with the aim of
building a potential pharmacophore model for this R1-AR
subtype.
Tissue was homogenized in about 50 volumes of ice-cold 50
mM Tris-HCl buffer (pH 7.4) using an Ultra Turrax TP-180
(2 × 20 s) and centrifuged at 50000g for 10 min (Beckman
model J -21B refrigerated centrifuge). The pellet was resus-
pended in the same volume of fresh buffer, incubated at 37 °C
for 10 min, and centrifuged again at 50000g for 10 min. The
pellet was then washed once by resuspension in fresh buffer
and centrifuged as before. The pellet was then resuspended
in the appropriate incubation buffer: 50 mM Tris-HCl (pH 7.7)
for 5-HT2A receptors; same buffer with the addition of 10 µM
pargyline for the other receptors; with 4 mM CaCl2 for 5-HT1A
receptors; with 4 mM CaCl2 and 0.1% ascorbic acid for 5-HT1B
and 5-HT2C receptors; 50 mM Tris-HCl, pH 7.1, containing 10
µM pargyline, 120 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM
MgCl2, and 0.1% ascorbic acid for D1 and D2 receptors.
Binding assays were done as described previously.48 Briefly,
the following incubation conditions were used: for 5-HT1A
,
[3H]-8-OH-DPAT (specific activity of 157 Ci/mmol, NEN) final
concentration of 1 nM, 30 min at 25 °C (nonspecific binding,
5-HT 10 µM); for 5-HT1B, [3H]-5-HT (specific activity 14.6 Ci/
mmol, NEN) final concentration of 2 nM, 30 min at 25 °C
(nonspecific binding, 5-HT 10 µM); for 5-HT2A, [3H]ketanserin
(specific activity of 60 Ci/mmol, Amersham) final concentration
of 0.7 nM, 15 min at 37 °C (nonspecific binding, methysergide
1 mM); for D1, [3H]SCH23390 (specific activity of 71 Ci/mmol,
NEN) final concentration of 0.4 nM, 15 min at 37 °C (nonspe-
cific binding, (-)-cis-flupentixol 10 µM); for D2, [3H]spiperone
(specific activity of 19 Ci/mmol, NEN) final concentration of
0.2 nM, 15 min at 37 °C (nonspecific binding, (-)-sulpiride 100
µM).
The chemical functions included in both HipHop and Hy-
poGen calculations were the hydrogen bond acceptor lipid
(HBA), hydrogen bond donor (HBD), positive ionizable (PI),
aromatic ring (RA), and hydrophobic (HY) features.
The program was forced to keep only hypotheses with at
least five features and to include a positive ionizable group
(reported to be a critical key for R1-AR antagonistic activity)
in the composition of hypotheses generated by both HipHop
and HypoGen.
Incubations were stopped by rapid filtration under vacuum
through GF/B filters which were then washed with 12 mL (4
× 3 times) of ice-cold 50 mM Tris-HCl buffer (pH 7.4) using a
Brandel M-48R apparatus and counted in 4 mL of Filter Count
(Packard) in a LKB 1214 RACKBETA liquid scintillation
Ack n ow led gm en t. The authors thank Dr. Ellen W.
Baxter (The R. W. J ohnson Pharmaceutical Research
Institute, Spring House, PA) for helpful suggestions in
the synthesis of intermediate 61. We also thank Vanitha

Pyrimido[5,4-b]indoles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14 2893
(16) Giardina`, D.; Crucianelli, M.; Romanelli, R.; Leonardi, A.;
Poggesi, E.; Melchiorre, C. Synthesis and biological profile of
the enantiomers of [4-(4-amino-6,7-dimethoxyquinazolin-2-yl)-
cis-octahydroquinoxalin-1-yl]furan-2-ylmethanone (Cyclazosin),
a potent competitive R1B-adrenoceptor antagonist. J . Med. Chem.
1996, 39, 4602-4607.
(17) Patane, M. A.; Scott, A. L.; Broten, T. P.; Chang, R. S. L.;
Ransom, R. W.; DiSalvo, J .; Forray, C.; Bock, M. G. 4-Amino-
2-[4-[1-(benzyloxycarbonyl)-2(S)-[[(1,1-dimethylethyl)amino]car-
bonyl]piperazinyl]-6,7-dimethoxyquinazoline (L-765,314): a po-
tent and selective R1b adrenergic antagonist. J . Med. Chem. 1998,
41, 1205-1208.
(18) Goetz, A. S.; King, H. K.; Ward, S. D. C.; True, T. A.; Rimele, T.
J .; Saussy, D. L., J r. BMY 7378 is a selective antagonist of the
D subtype of R1-adrenoceptor. Eur. J . Pharmacol. 1995, 272, R5-
R6.
(19) Yocca, F. D.; Hyslop, D. K.; Smith, D. W.; Maayani, S. BMY 7378,
a buspirone analogue with high affinity, selectivity and low
intrinsic activity at the 5-HT1A receptor in rat and guinea pig
hippocampal membranes. Eur. J . Pharmacol. 1987, 137, 293-
294.
Subramanian for valuable technical assistance in the
radioligand binding assays. This work was in part
supported by a grant from MIUR (Project “Progettazi-
one, Sintesi e Valutazione Biologica di Nuovi Farmaci
Cardiovascolari”). Financial support provided by the
Italian Ministero dell’Istruzione, dell’Universita` e della
Ricerca Scientifica (Project “Progettazione e Sintesi di
Agenti Neuroprotettivi”), Italian Research National
Council (CNR) “Progetto Finalizzato Biotecnologie”
(CNR Target Project on “Biotechnology”), and the
National Institutes of Health is gratefully acknowl-
edged. M.B. thanks the Merck Research Laboratories
for the 2002 Academic Development Program (ADP)
Chemistry Award.
(20) Russo, F.; Romeo, G.; Guccione, S.; De Blasi, A. Pyrimido[5,4.b]-
indole derivatives. 1. A new class of potent and selective
R1-adrenoceptor ligands. J . Med. Chem. 1991, 34, 1850-1854.
(21) Romeo, G.; Russo, F.; Guccione, S.; Barbarulo, D.; De Blasi, A.
Heterocyclic systems containing the pyrimido-2,4-dione ring as
selective ligands for the R1-adrenoceptors. Farmaco 1995, 50,
471-477.
(22) Romeo, G.; Russo, F.; De Blasi, A. Synthesis of novel 5H-
pyrimido[5,4.b]indole-(1H,3H)2,4-diones as potential ligands for
the cloned R1-adrenoceptor subtypes. J . Heterocycl. Chem. 2001,
38, 391-395.
(23) Unangst, P. C. Pyrimido[1′,2′:1,2]pyrimido[5,4-b]indoles. A new
heterocyclic ring system. J . Heterocycl. Chem. 1983, 20, 495-
499.
(24) Parcell, R. F. 4-(o-Substitutedphenyl)-1-hydroxyalkylpiperidines.
U.S. Patent 2,891,066, 1959.
(25) Romeo, G.; Ambrosini, G.; Guccione, S.; De Blasi, A.; Russo, F.
Pyrimido[5,4-b]benzofuran and pyrimido[5,4-b]benzothiophene
derivatives. Ligands for R1- and 5-HT1A receptor. Eur. J . Med.
Chem. 1993, 24, 499-504.
(26) Gupta, S. P.; Chatterjee, S. S.; Bindra, J . S.; J ain, P. C.;
Anand, N. Compounds acting on CNS: Part XXIIsSynthesis
of 2,3,4,4a,5,6-hexahydro-1(H)-pyrazino[2,1-c]-1,4-benzoxazines.
Indian J . Chem. 1975, 13, 462-467.
Refer en ces
(1) Graham, R. M.; Perez, D. M.; Hwa, J .; Piascik, M. T.; Alpha1-
adrenergic receptor subtypes. Molecular structure, function, and
signaling. Circ. Res. 1996, 78, 737-749.
(2) Hieble, J . P.; Bylund, D. B.; Clarke, D. E.; Eikenburg, D. C.;
Langer, S. Z.; Lefkowitz, R. J .; Minneman, K. P.; Ruffolo, R. R.,
J r. International Union of Pharmacology. X. Recommendation
for nomenclature of alpha1-adrenoceptors: consensus update.
Pharmacol. Rev. 1995, 47, 267-270.
(3) Ramarao, C. S.; Denker, J . M.; Perez, D. M.; Gaivin, R. J .; Riek,
R. P.; Graham, R. M.; Genomic organization and expression of
the human alpha 1B-adrenergic receptor. J . Biol. Chem. 1992,
267, 21936-21945.
(4) Hirasawa, A.; Horie, K.; Tanaka, T.; Takagaki, K.; Murai, M.;
Yano, J .; Tsujimoto, G. Cloning, functional expression and tissue
distribution of human cDNA for the alpha 1C-adrenergic recep-
tor. Biochem. Biophys. Res. Commun. 1993, 195, 902-909.
(5) Esbenshade, T. A.; Hirasawa, A.; Tsujimoto, G.; Tanaka, T.;
Yano, J .; Minneman, K. P.; Murphy, T. J .; Cloning of the human
R1D-adrenergic receptor and inducible expression of three human
subtypes in SKNMC cells. Mol. Pharmacol. 1995, 47, 591-598.
(6) Price, D. T.; Lefkowitz, R. J .; Caron, M. G.; Berkowitz, D.;
Schwinn, D. A. Localization of mRNA for three distinct alpha
1-adrenergic receptor subtypes in human tissues; implication for
human alpha-adrenergic physiology. Mol. Pharmacol. 1994, 45,
171-175.
(27) Baxter, E. W.; Reitz, A. B. Tricyclic compounds having affinity
for the 5-HT1A receptor. U.S. Patent 5,512,566, April 30, 1996.
(28) Baxter, E. W.; Reitz, A. B. Hindered rotation congeneres of
mazapertine: high affinity ligands for the 5-HT1A receptors.
Bioorg. Med. Chem. Lett. 1997, 7, 763-768.
(7) Chang, D. J .; Chang, T. K.; Yamanishi, S. S.; Rick Salazar, F.
H.; Kosaka, A. H.; Khare, R.; Bhakta, S.; J asper, J . R.; Shieh,
I.; Lesnick, J . D.; Ford, A. P. D. W.; Daniels, D. V.; Eglen, R. M.;
Clarke, D. E.; Bach, C.; Chan, H. W. Molecular cloning, genomic
characterization and expression of novel human R1A-adrenocep-
tor isoforms. FEBS Lett. 1998, 422, 279-283.
(8) Coge, F.; Guenin, S.; Renuard-Try, A.; Rique, H.; Ouvry, C.;
Fabry, N.; Beauverger, P.; Nicolas, J .; Galizzi, J .; Boutin, J . A.;
Canet, E. Truncated isoforms inhibit [3H]prazosin binding and
cellular trafficking of native human R1A-adrenoceptors. Biochem.
J . 1999, 343, 231-239.
(9) Timmermans, B. M. W. M.; van Zwieten, P. A. R-Adrenoceptor
agonists and antagonists. Drugs Future 1984, 9, 41-55.
(10) Holmes, J . B.; Christensen, M. M.; Rasmussen, P. C.; J acobsen,
F.; Nielsen, J .; Norgaard, J . P.; Olesen, S.; Noev, I.; Wolf, H.;
Husted, S. E. 29-Week doxazosin treatment in patients with
symptomatic benign prostatic hyperplasia. Scand. J . Urol.
Nephrol. 1994, 28, 77-82.
(11) Forray, C.; Bard, J . A.; Wetzel, J . M.; Chiu, G.; Shapiro, E.; Tang,
R.; Lepor, H.; Hartig, P. R.; Weinshank, R. L.; Branchek, T. A.
The alpha 1-adrenergic receptor that mediates smooth muscle
contraction in human prostate has the pharmacological proper-
ties of the cloned alpha 1c subtype. Mol. Pharmacol. 1994, 45,
703-708.
(12) Moriyama, N.; Kurimoto, S.; Horie, S.; Nasu, K.; Tanaka, T.;
Yano, K.; Hirano, H.; Tsujimoto, G.; Kawabe, K. Detection of
R1-AR subtypes in human hypertrophied prostate by in situ
hybridization. Histochem. J . 1996, 28, 283-288.
(13) Sorbera, L. A.; Silvestre, J .; Castaner, J . KMD-3213. Treatment
of BPH. R1-Adrenoceptor antagonist. Drugs Future 2001, 26,
553-550.
(14) Lagu, B. Identification of R1A-adrenoceptor selective antagonists
for the treatment of benign prostatic hyperplasia. Drugs Future
2001, 26, 757-765.
(15) Stam, W. B.; Van der Graaf, P. N.; Saxena, P. R. Functional
characterization of the pharmacological profile of the putative
alpha1B-adrenoceptor antagonist, (+)-cyclazosin. Eur. J . Phar-
macol. 1998, 361, 79-83.
(29) Predvoditeleva, G. S.; Shchukina, M. N. Synthesis of benzomor-
pholine derivatives. I. Zh. Obshch. Khim. 1963, 33, 145-150;
Chem. Abstr. 1963, 59, 609a.
(30) Dess, D. B.; Martin, J . C. Readily accessible 12-I-51 oxidant for
the conversion of primary and secondary alcohols to aldehydes
and ketones. J . Org. Chem. 1983, 48, 4155-4156.
(31) Geiss, W. B. Technical Report; Albany Molecular Research,
Inc.: Albany, NY, 1999; Vol. 1, Issue 11..
(32) Matarrese, M.; Moresco, R. M.; Romeo, G.; Turolla, E. A.;
Simonelli, P.; Todde, S.; Belloli, S.; Carpinelli, A.; Magni, F.;
Russo, F.; Galli Kienle, M.; Fazio, F. [11C]RN5: A New Promising
Agent for the in Vivo Imaging of Myocardial R1-Adrenoceptor.
Eur. J . Pharmacol. 2002, 453 (2, 3), 231-238.
(33) Theroux, T. L.; Esbenshade, T. A.; Peavy, R. D.; Minneman, K.
P. Coupling Efficiencies of Human R1-Adrenergic Receptor
Subtypes: Titration of Receptor Density and Responsiveness
with Inducible and Repressible Expression Vectors. Mol. Phar-
macol. 1996, 50, 1376-1387.
(34) Barbaro, R.; Betti, L.; Botta, M.; Corelli, F.; Giannaccini, G.;
Maccari, L.; Manetti, F.; Strappaghetti, G.; Corsano, S. Synthe-
sis, biological evaluation, and pharmacophore generation of new
pyridazinone derivatives with affinity toward R1- and R2-
adrenoceptors. J . Med. Chem. 2001, 44, 2118-2132.
(35) Catalyst, version 4.6; Accelerys, Inc.: San Diego, CA, 2000.
(36) De Marinis, R. M.; Wise, M.; Hieble, J . P.; Ruffolo, R. R., J r.
Structure-Activity Relationships for Alpha-1 Adrenergic Recep-
tor Agonists and Antagonists. In The Alpha-1 Adrenergic Recep-
tor; Ruffolo, R. R., J r., Ed.; Humana Press: Clifton, NJ , 1987;
pp 211-265.
(37) Montesano, F.; Barlocco, D.; Dal Piaz, V.; Leonardi, A.; Poggesi,
E.; Fanelli, F.; De Benedetti, P. G. Isoxazolo-[3,4-d]-pyridazin-
7-(6H)-ones and Their Corresponding 4,5-Disubstituted-3-(2H)-
pyridazinone Analogues as New Substrates for R1-Adrenoceptor
Selective Antagonists: Synthesis, Modeling, and Binding Stud-
ies. Bioorg. Med. Chem. 1998, 6, 925-935.
(38) Bremner, J . B.; Coban, B.; Griffith, R.; Groenewoud, K. M.; Yates,
B. F. Ligand Design for R1 Adrenoceptor Subtype Selective
Antagonists. Bioorg. Med. Chem. 2000, 8, 201-214.

2894 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 14
Romeo et al.
(39) Konkel, M. J .; Wetzel, J . M.; Cahir, M.; Craig, D.; Noble, S. A.;
Gluchowski, C. Discovery of Antagonists Selective for the Alpha-
1D-Adrenoceptor. Presented at the 216th National Meeting of
the American Chemical Society, Boston, August 23-27, 1998;
Paper MEDI 129.
(40) (a) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J .;
Swaminathan, S.; Karplus, M. CHARMm: A Program for
Macromolecular Energy, Minimization, and Dynamics Calcula-
tions. J . Comput. Chem. 1983, 4, 187, 217. (b) Catalyst force
field and potential energy functions are described in the file
REFenergy.doc.html provided by Accelerys along with the
program Catalyst 4.6.
(41) (a) Smellie, A.; Teig, S. L.; Towbin, P. Poling: Promoting
Conformational Coverage. J . Comput. Chem. 1995, 16, 171-187.
(b) Smellie, A.; Kahn, S. D.; Teig, S. L. An Analysis of Confor-
mational Coverage 1. Validation and Estimation of Coverage.
J . Chem. Inf. Comput. Sci. 1995, 35, 285-294. (c) Smellie, A.;
Kahn, S. D.; Teig, S. L. An Analysis of Conformational Coverage
2. Applications of Conformational Models. J . Chem. Inf. Comput.
Sci. 1995, 35, 295-304.
process for the R1d-adrenergic receptor activation: a molecular
dynamics and docking study. Proteins 2001, 43, 382-394.
(44) Kenny, B.; Ballard, S.; Blagg, J .; Fox, D. Pharmacological options
in the treatment of benign prostatic hyperplasia. J . Med. Chem.
1997, 40, 1293-1315.
(45) Ruffolo, R. R., J r.; Bondinell, W.; Ku, T.; Naselsky, D. P.; Hieble,
J . P. Alpha1-adrenoceptors: pharmacological classification and
newer therapeutic applications. Proc. West. Pharmacol. Soc.
1995, 38, 121-126.
(46) Minneman, K. P.; Theroux, T. L.; Hollinger, S.; Han, C.;
Esbenshade, T. A. Selectivity of agonist for cloned alpha1-
adrenergic receptor subtypes. Mol. Pharmacol. 1994, 46, 929-
936.
(47) GraphPad Prism; GraphPad Software Inc., San Diego, CA, 2000.
(48) Caccia, S.; Confalonieri, S.; Guiso, G.; Bernasconi, P.; Cagnotto,
A.; Skorupska, M.; Mennini, T. Brain uptake and distribution
of the potential memory enhancer CL 275,838 and its main
metabolites in rats: relationship between brain concentrations
and in vitro potencies on neurotransmitter mechanisms. Psycho-
pharmacology 1994, 115, 502-508.
(42) (a) Betti, L.; Botta, M.; Corelli, F.; Floridi, M.; Fossa, P.;
Giannaccini, G.; Manetti, F.; Strappaghetti, G.; Corsano, S. R1-
Adrenoceptor Antagonists. Rational Design, Synthesis and
Biological Evaluation of New Trazodone-like Compounds. Bioorg.
Med. Chem. Lett. 2002, 12, 437-440. (b) Barbaro, R.; Betti, L.;
Botta, M.; Corelli, F.; Giannaccini, G.; Maccari, L.; Manetti, F.;
Strappaghetti, G.; Corsano, S. Synthesis and Biological Activity
of New 1,4-Benzodioxan-Arylpiperazine Derivatives. Further
(49) Allfit; National Institutes of Health: Bethesda, MD, 1988.
(50) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (IC50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(51) Nicoletti, F.; Iadarola, M. J .; Wroblewski, J . T.; Costa, E.
Excitatory amino acid recognition sites couplet with inositol
phospholipid metabolism: Developmental changes and interac-
tion with R1-adrenoceptors. Proc. Natl. Acad. Sci. U.S.A. 1986,
83, 1931-1935.
Validation of
a Pharmacophore Model for R1-Adrenoceptor
Antagonists. Bioorg. Med. Chem. 2002, 10, 361-369.
(43) Carrieri, A.; Centeno, N. B.; Rodrigo, J .; Sanz, F.; Carotti, A.
Theoretical evidence of a salt bridge disruption as the initiating
J M0307741