Solid-phase synthesis of highly diverse purine-hydroxyquinolinone bisheterocycles

Article abstract of DOI:10.1021/cc100132z

Solid-phase synthesis of bisheterocyclic compounds that contain purine and the 3-hydroxyquinolin-4(1H)-one skeleton connected with an aliphatic spacer of a different length/structure is described. The reaction sequence started from the primary amines immobilized on aminomethylated polystyrene resin equipped with an acid-labile linker (4-(4-formyl-3-methoxyphenoxy)butyric acid). After the arylation of amines with 2,6-dichloropurine via its C⁶, purine N ⁹ was alkylated and subsequently the chlorine at purine C² was substituted with aliphatic diamines. The resulting terminal amino group was used as the starting point for the synthesis of 3-hydroxyquinolin-4(1H)-one precursors based on the acylation with 3-amino-4-(methoxycarbonyl)benzoic acid followed by the saponification of the methyl ester and esterification of the resulting carboxylic acid with various haloketones. The intermediates were cleaved from the resin, and their cyclization to the target purine- hydroxyquinolinone bisheterocycles was accomplished by heating in acetic or trifluoroacetic acid.

Full text of DOI:10.1021/cc100132z

890
J. Comb. Chem. 2010, 12, 890–894
Solid-Phase Synthesis of Highly Diverse Purine-Hydroxyquinolinone
Bisheterocycles
Barbora Vanˇkova´, Jan Hlava´cˇ, and Miroslav Soural*
Department of Organic Chemistry, Faculty of Science, Institute of Molecular and Translational
Medicine, UniVersity of Palacky´, 771 46 Olomouc, Czech Republic
ReceiVed July 14, 2010
Solid-phase synthesis of bisheterocyclic compounds that contain purine and the 3-hydroxyquinolin-4(1H)-
one skeleton connected with an aliphatic spacer of a different length/structure is described. The reaction
sequence started from the primary amines immobilized on aminomethylated polystyrene resin equipped
with an acid-labile linker (4-(4-formyl-3-methoxyphenoxy)butyric acid). After the arylation of amines with
2,6-dichloropurine via its C6, purine N9 was alkylated and subsequently the chlorine at purine C2 was
substituted with aliphatic diamines. The resulting terminal amino group was used as the starting point for
the synthesis of 3-hydroxyquinolin-4(1H)-one precursors based on the acylation with 3-amino-4-
(methoxycarbonyl)benzoic acid followed by the saponification of the methyl ester and esterification of the
resulting carboxylic acid with various haloketones. The intermediates were cleaved from the resin, and
their cyclization to the target purine-hydroxyquinolinone bisheterocycles was accomplished by heating in
acetic or trifluoroacetic acid.
Introduction
and enzymes acting in their biological transformations,
quinolinones are widely used as inhibitors of topoisomerase,
enzyme responsible for a cleavage and recombination of
DNA. Hydroxyquinolinones were also described as reverse
transcriptase inhibitors,15 providing further evidence of their
ability to interact with nucleic acids. In this article our
intention was to develop a methodology for the construction
of a chemical library consisting of purine-hydroxyquinoli-
none bisheterocycles with five diversity positions: Two
positions at purine (C6 and N9 substitution), two positions
at hydroxyquinolinone (C2 and the carboxamide C7 N-
substitution), and the fifth diversity position resulting from
the presence of a spacer between both heterocyclic systems
(Figure 1).
Although a number of potential combinations in terms of
the spacer location on each heterocycle exist, this initial study
is focused on the connection of both heterocycles via
hydroxyquinolinone C7 and purine C2 position. The synthetic
route partially takes advantage of the previously described
solid-phase synthesis of hydroxyquinolinone-7-carboxam-
ides.10 According to our latest experiments, selected hydroxy-
Synthetic derivatives of 2-substituted-3-hydroxyquinolin-
4(1H)-ones (termed “hydroxyquinolinones” hereafter) rep-
resent quite a new class of biologically promising sub-
stances.1 Among their biological effects which have been
observed belong, for instance anticancer2-4 or immunosup-
pressive activity.3 In the literature, different routes for the
preparation of hydroxyquinolinones from various starting
materials have been described;5-8 however, an acidic thermal
cylization of anthranilates is considered as the most advanta-
geous method.9-11 Although hydroxyquinolinones have been
subjected to relatively intense biological research in a past
decade, only little attention has been paid to their combina-
tion with other condensed heterocyclic systems. The only
paper focusing on the preparation and study of these
compounds described hydroxyquinolinone-oxazole deriva-
tives such as inosine 5′-monophosphate dehydrogenase
inhibitors.12 In one of our previous articles we introduced a
general methodology for the preparation of bisheterocyclic
compounds using solid-phase synthesis.13 Our most recent
goal was to apply the described strategy for the synthesis of
bisheterocyclic derivatives in which the hydroxyquinolinone
skeleton is accompanied by another heterocyclic moiety. The
second heterocycle was selected with respect to a known
biological potential of single heterocycles and their deriva-
tives. Concerning biological properties, nucleic bases (i.e.,
purines and pyrimidines) are in a privileged position among
the nitrogenous heterocycles as they are being included in
large number of active compounds.14 Purine derivatives are
well-known for their ability to interact with both nucleic acids
Figure 1. Suggested general structure of the target bishetero-
cycles.
* To whom correspondence should be addressed. E-mail: soural@orgchem.
upol.cz. Phone: +420 585634418. Fax: +420 585634465.
10.1021/cc100132z  2010 American Chemical Society
Published on Web 09/17/2010

Synthesis of Purine-Hydroxyquinolinone Bisheterocycles
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 6 891
Scheme 1. Preparation of Purine-hydroxyquinolinone Bisheterocycles (8)a
a Reagents: (i) amine, 10% AcOH/DMF, overnight, NaBH(OAc)3, 4 h; (ii) 2,6-dichloropurine, DIEA, THF, 50 °C, overnight; (iii) alkyl iodide, DBU,
DMSO, 50 °C, overnight, (repeated once in THF); (iv) diamine, diethylene glycol diethyl ether, 150 °C, 24 h; (v) aminopropanol, diethylene glycol diethyl
ether, 150 °C, 24 h; (vi) mesylchloride, pyridine, room temp., 2 h, then propylamine, DMSO, room temp., overnight; (vii) 3-amino-4-(methoxycarbonyl)benzoic
acid, DIC, BtOH, DCM, DMF, room temp., overnight; (viii) TMSOK, THF, room temp., 24 h; (ix) haloketones, DIEA, DMF, room temp., 2 h; (x) TFA,
DCM, room temp., 30 min.; (xi) AcOH, reflux, 3 h or TFA, reflux, 2 h.
quinolinone-7-carboxamides exhibit strong anticancer activity
in vitro (IC50 < 500 nmol/L) which strongly predisposes such
compounds for further structure modification and biological
research.
alkylated. Alkylation of purine derivatives using solution-
phase synthesis is usually performed with alkyliodides in
the presence of a suitable base whereas the N9 modification
of solid-supported purine derivatives takes advantage of the
Mitsunobu protocol which allows alkylation with various
alcohols under mild conditions.17 In our case, when Mit-
sunobu alkylation of the intermediates (2) with ethanol or
isopropanol was tested, reaction yields were unsatisfactory
although various conditions and reagents were tested. As a
result we used triphenylphosphine or tributylphosphine in
various concentrations, different solvents (dry THF or NMP),
and varied reaction times as well as repeating the reaction
step, however, without success in achieving quantitative
Results and Discussion
The synthesis was carried out on aminomethylated poly-
styrene resin equipped with an acid-labile linker (4-(4-formyl-
3-methoxyphenoxy)butyric acid) according to Scheme 1.
After the immobilization of the primary amines using
reductive amination16 the resulting secondary amines (1)
were regioselectively arylated with 2,6-dichloropurine to give
the intermediates (2). In a next step, the purine N9 had to be

892 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 6
Vanˇkova´ et al.
Figure 2. Building blocks successfully used for the evaluation of the synthetic route.
conversion. Additionally, the purity of the products was
diminished by the appearance of number of unknown
impurities when the alkylation step was repeated 2 or 3 times.
Discouraged by such problems we turned our attention to
the solution-phase method of alkylation with alkyl iodides,
and were finally successful in transferring this strategy to
the solid-phase. The reaction was performed in dimethyl-
sulfoxide (DMSO) at elevated temperature (50 °C). Various
bases were tested (such as DIEA, NaH, Na2CO3, Cs2CO3)
but the best results (70-80% of conversion, HPLC-UV
traces) were obtained with diazabicyclo[5.4.0]undec-7-ene
(DBU). For the quantitative yield, the reaction was repeated
in tetrahydrofuran (THF) giving intermediates (3) of high
purity. If the reaction was repeated in DMSO instead of in
THF, the yield did not exceed 90%. The developed alkylation
method was shown to be limited by the character of the alkyl
iodide used. It did not work for methyl iodide and cyclohexyl
iodide but it was successfully tested for ethyl iodide,
isopropyl iodide, t-butyl iodide, and cyclopentyl iodide.
Among all used amines and alkyl iodides (see Figure 2) only
one exception was detected: The alkylation of the 6-N-
benzylamine intermediate (2) with t-butyl iodide furnished
only 60% of conversion.
The subsequent substitution of the chlorine atom at purine
C2 required a high temperature (150 °C) to proceed; it was
tested for various aliphatic diamines. First, the reaction was
performed in solvents typically used for a nucleophilic
substitution such as DMSO, DMF or NMP but the interme-
diates (4) were obtained in limited purity. In each case,
formation of unknown side-products (20-30%, HPLC-UV
traces) was observed which was probably caused by the
reaction of intermediates (3) with the solvent under the harsh
conditions required. The selection of the suitable solvent was
crucial at this step - when the reaction was carried out in
inert diethylene glycol diethyl ether, the intermediates (4a)
were obtained in excellent purity. Surprisingly, the method
was successfully tested for propylenediamine, butane-1,4-
diamine and 3-oxapentane-1,5-diamine but the same reaction
with ethylenediamine afforded only a mixture of unknown
compounds. In this step we also attempted to increase the
diversity of the target bisheterocycles. Instead of diamines,
the intermediates (3) were reacted with an aminoalcohol
(aminopropanol in our case) and the derivatives (4a) were
obtained. Our intention was to modify the terminal hydroxyl
group using mesylation and subsequent reaction with primary
amines (propylamine in our case) resulting in possible
formation of the secondary aminoderivatives (4c). Unfortu-
nately, this modification was unsuccessful and afforded only
mixtures of compounds whose major products corresponded
to the formation of alkene derivatives (4b). Compounds 4
were acylated with 3-amino-4-(methoxycarbonyl)benzoic
acid to give intermediates (5). If aminopropanol was used
for the reductive amination (step i), its hydroxy group was
also partially acylated (20%, HPLC-UV traces); however,
this side product was removed in the next step, when
saponification of the intermediates (5) was performed with
potassium trimethylsilanolate leading to the carboxylic acids
(6). Surprisingly, the saponification took place very slowly,
and a longer reaction time (at least 24 h) was necessary for
the quantitative hydrolysis. After the esterification of the
carboxylic acids (6) with haloketones of aliphatic, aromatic,
or heterocyclic character, the precursors (7) were obtained
in excellent purity above 90% (HPLC-UV traces). The final
cyclization to bisheterocycles (8) was performed after the
acid-mediated cleavage of the precursors (7) from the resin.
Two cyclization methods were tested. When trifluoroacetic
acid (method A) was used as the cyclizing agent, the target
products were usually obtained in excellent purity; in several
cases, however, the partial acidic hydrolysis of the esters
(8) resulting in a contamination of the products with
derivatives (6) was observed. Thus, we also tested the
cyclization in acetic acid (method B) which did not cause
the competitive hydrolysis and afforded the target products
in very good purity, typically over 90% (HPLC-UV traces).
Nevertheless, when the method B was used for the cycliza-
tion of the hydroxygroup-containing intermediates (obtained

Synthesis of Purine-Hydroxyquinolinone Bisheterocycles
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 6 893
Table 1. Summary of the Prepared Products
Table 2. Purity and Yield of the Crude Productsa
by the reductive amination with aminopropanol, step i), the
final compounds were not obtained as the hydroxyl deriva-
tives but the corresponding O-acetyl derivatives were
isolated. Generally, TFA cyclization can be considered as
the more versatile method; however, the cyclization of
derivatives (7) bearing a nitro group needed to be performed
using method B, otherwise the cyclization was not observed.
The purification of the crude products was accomplished
by simple sonification in diethylether and subsequent filtra-
tion of the precipitated material.
compound purity (%) yield(%) compound purity (%) yield (%)
8(1,1,3,2)
8(1,2,3,3)
8(1,2,3,5)
8(1,2,1,1)
8(1,3,3,6)
8(1,4,1,1)
8(2,2,3,6)
8(2,2,2,2)
8(3,1,3,3)
8(3,1,3,2)
90
93
98
85
90
82
95
98
86
80
80
80
43
82
79
65
67
14
52
81
8(3,1,1,1)
8(2,2,2,1)
8(3,2,1,2)
8(3,2,1,4)
8(3,2,3,2)
8(4,1,1,2)
8(4,1,3,2)
8(4,2,1,4)
8(4,3,3,6)
8(4,3,1,5)
88
90
93
98
95
95
99
95
65
86
89
12
45
42
97
98
95
25
98
95
a Purity was determined on basis of HPLC-UV traces.
The building blocks used for the synthesis were selected
with respect to the highest skeletal diversity of the target
substances as well as the commercial availability of the
synthones. As can be seen in Figure 2, ligands with aliphatic,
aromatic, and heterocyclic character were introduced for R1
and R2, also substituted phenyls for R3 were successfully
tested (with both electron-donating and electron-withdrawing
functional groups). Three different aliphatic spacers of
various length/constitution were selected. Other spacers such
as aromatic skeleton containing ligands have been not tested
but could be of interest in the future.
In conclusion, we have developed a method of solid-phase
synthesis of novel purine-hydroxyquinolinone bisheterocycles
with four diversity positions. The versatility of the method
was demonstrated by the preparation, isolation and full
characterization of 20 compounds (see Table 1). After careful
optimization of the reaction sequence the target compounds
could be obtained in very good crude purity (typically over
90%) and good yield (about 60% on average) (Table 2).
Although the time needed for the entire reaction sequence
is approximately 10 days, the methodology is robust,
variable, and applicable for the preparation of a large
chemical library from commercially available synthones. The
target molecules consist of two heterocyclic systems of high
biological potential. Hence they represent excellent substrates
for biological research.
Acknowledgment. This project was supported in part by
the Ministry of Education, Youth, and Sport of the Czech
Republic (Grants MSM6198959216 and ME09057) and FM
EEA/Norska (A/CZ0046/1/0022). The infrastructural part of
this project (Institute of Molecular and Translational Medi-
cine) was supported from the Operational Program Research
and Development for Innovations (project CZ.1.05/2.1.00/
01.0030).
Supporting Information Available. Details of experi-
mental procedures and spectroscopic data for synthesized
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.

894 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 6
Vanˇkova´ et al.
(9) Hradil, P.; Jirman, J. Collect. Czech. Chem. Commun. 1995,
60, 1357.
(10) Soural, M.; Krchnˇa´k, V. J. Comb. Chem. 2007, 9, 793–796.
(11) Krupkova, S.; Soural, M.; Hlavac, J.; Hradil, P. J. Comb.
Chem. 2009, 6, 951–955.
(12) Iwanowicz, J. E.; Watterson, S. H.; Murali, T. G.; Pitts, W. J.;
Gu, H. H. Patent WO 2001081340.
(13) Soural, M.; Bouillon, I.; Krchnˇa´k, V. J. Comb. Chem. 2008,
References and Notes
(1) Hradil, P.; Hlava´cˇ, J.; Soural, M.; Hajdu´ch, M.; Kola´rˇ, M.;
Vecˇeˇrova´, R. Mini-ReV. Med. Chem. 2009, 9, 696–702.
(2) Hradil, P.; Krejcˇ´ı, P.; Hlava´cˇ, J.; Wiedermannova´, I.; Lycˇka,
A.; Bertolasi, V. J. Heterocycl. Chem. 2004, 41, 375–379.
(3) Krejcˇ´ı, P.; Hradil, P.; Hlava´cˇ, J.; Hajdu´ch, M. Patent WO
2008028427.
(4) Soural, M.; Hlava´cˇ, J.; Hradil, P.; Frysova, I.; Hajdu´ch, M.;
Bertolasi, V.; Malonˇ, M. Eur. J. Med. Chem. 2006, 41, 467–
474.
(5) Velezheva, V. S.; Mel’man, A. I.; Pols´hakov, V. I.; Anisimova,
O. S. Khimiya Geterotsiklicheskikh Soedinenii 1992, 2, 279–
280.
(6) Goldsworthy, J.; Ross, W. J.; Verge, J. P. Eur. Pat. Appl. EP
198255068.
10, 923–933.
(14) Legraveren, M.; Grierson, D. S. Bioorg. Med. Chem. Lett.
2006, 14, 3987–4006.
(15) Loya, S.; Rudi, A.; Tal, R.; Kashman, Y.; Loya, Y.; Hizi, A.
Arch. Biochem. Biophys. 1994, 309, 315–22.
(16) Krchnak, V.; Smith, J.; Vagner, J. Collect. Czech. Chem.
Commun. 2001, 66, 1078–1106.
(17) Dorff, P. H.; Garigipati, R. S. Tetrahedron. Lett. 2001, 42,
2771–2773.
(7) Spence, T. W. M.; Tennant, G. J. Chem. Soc. 1971, 22, 3712–
3719.
(8) Sword, I. P. J. Chem. Soc. 1971, 5, 820–823.
CC100132Z