Solid-phase synthesis of 5-noranagrelide derivatives

Article abstract of DOI:10.1021/co4001315

Solid-phase synthesis of 1H-benzo[d]imidazo[1,2-a]imidazol-2(3H)-one derivatives employing Fmoc-α-amino acids and nitroaryl fluorides as key building blocks has been developed. The Fmoc-α-amino acids immobilized on Wang resin, equipped with a piperazine carbamate linker, were transformed to o-nitroanilines in two steps. After reduction of the nitro group, the corresponding o-phenylenediamines gave the 2-aminobenzimidazole scaffold by reaction either with cyanogen bromide or with Fmoc-NCS. Cleavage from the polymer support and further cyclization afforded the target compounds. The developed methodology represents a versatile and simple approach for the preparation of various corresponding 1H-benzo[d]imidazo[1,2-a]imidazol-2(3H)- ones from a large number of commercially available building blocks.

Full text of DOI:10.1021/co4001315

Research Article
pubs.acs.org/acscombsci
Solid-Phase Synthesis of 5‑Noranagrelide Derivatives
Ivano Messina,† Igor Popa,† Vítezslav Maier,‡ and Miroslav Soural*,†
̌
†Department of Organic Chemistry, Institute of Molecular and Translational Medicine, Faculty of Science, Palacky University, 771 46
Olomouc, Czech Republic
‡Regional Centre of Advanced Technologies and Materials, Department of Analytical Chemistry, Faculty of Science, Palacky
University, 77146 Olomouc, Czech Republic
S
* Supporting Information
ABSTRACT: Solid-phase synthesis of 1H-benzo[d]imidazo[1,2-
a]imidazol-2(3H)-one derivatives employing Fmoc-α-amino acids
and nitroaryl fluorides as key building blocks has been developed.
The Fmoc-α-amino acids immobilized on Wang resin, equipped with
a piperazine carbamate linker, were transformed to o-nitroanilines in
two steps. After reduction of the nitro group, the corresponding o-
phenylenediamines gave the 2-aminobenzimidazole scaffold by reaction either with cyanogen bromide or with Fmoc-NCS.
Cleavage from the polymer support and further cyclization afforded the target compounds. The developed methodology
represents a versatile and simple approach for the preparation of various corresponding 1H-benzo[d]imidazo[1,2-a]imidazol-
2(3H)-ones from a large number of commercially available building blocks.
KEYWORDS: anagrelide, Fmoc-amino acids, solid-phase synthesis, combinatorial chemistry
times by the use of traditional solution-phase synthesis and 2-
aminobenzimidazole as a key intermediate (Figure 2).6−10
In the past, biological properties of 5-noranagrelide
derivatives have been studied very rarely. Only antifungal
activity has been reported,7 and similar tricyclic compounds
have been patented as CRF receptor antagonists.11 In order to
get a deeper understanding of this lightly explored field, we
focused on the development of the simple solid-phase synthesis
of 5-noranagrelide analogues to access chemical libraries for a
detailed biological assay.
In 2003, Kundu et al. described the high throughput
synthesis of anagrelide derivatives based on solid-phase
synthesis with the use of Fmoc-α-amino acids immobilized
on Wang resin and o-nitrobenzaldehydes (Scheme 1).12
Inspired by Kundu et al.’s strategy, we proposed the use of o-
nitrofluorobenzenes instead of o-nitrobenzaldehydes to access a
high throughput synthesis of 5-noranagrelide derivatives.
However, in such a case, the immobilization of Fmoc-amino
acids via the ester bond used by Kundu et al. was not suitable,
as it has been shown that a spontaneous cyclative cleavage
occurs,13 leading to the undesired release of reaction
intermediates from the resin (Scheme 2).
INTRODUCTION
■
In our long-term research, we have been focused on the
development of simple methodologies suitable for the
preparation of pharmacologically promising substances. For
such purposes, the solid-phase synthesis concept has been
applied typically due to its significant advantages. In particular,
the simple isolation of reaction intermediates allows for a rapid
preparation of various target derivatives and easy access to
combinatorial chemistry.1
Essential thrombocytosis is a chronic blood cancer
characterized by the overproduction of platelets by mega-
karyocytes in the bone marrow. Anagrelide (Agrylid/Xagrid;2,3
Figure 1) was the first platelet reducing agent that acts by
Figure 1. Anagrelide and a general structure of target compounds (5-
noranagrelide scaffold).
To overcome this problem, we suggested the use of a
retarding the maturation of megakaryocytes via the thrombo-
poietin receptor and reducing their chromosome number and
cell diameter.4 It also inhibits platelet aggregation as a result of
the potent inhibition of cyclic adenosine monophosphate
(cAMP) phosphodiesterase III activity.5
̌ ́
piperazine carbamate linker as recently reported by Krchnak
and Neagoie.14 The principle of this strategy centers around the
immobilization of the Fmoc-amino acid via the amide bond,
which does not undergo the intramolecular aminolysis
Along with the preparation of anagrelide scaffold containing
compounds, some of its structural analogues have been
reported in the literature. For instance, synthesis of 5-
noranagrelide derivatives (Figure 1) has been described several
Received: October 10, 2013
Revised: November 28, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX

ACS Combinatorial Science
Research Article
Figure 2. Some retrosynthetic approaches to 5-noranagrelide derivatives.
Scheme 1. Solid-Phase Synthesis of Anagrelide Derivatives by Kundu et al.
Scheme 2. Use of Wang Resin: Unapplicable Approach
Scheme 3. Use of Wang Resin and Carbamate Linker: Theoretically Applicable Approach
spontaneously but requires harsher conditions to proceed
(Scheme 3).
transfer catalyst as recently reported in the literature.15
Reaction of intermediate 5{1,1} with cyanogen bromide gave
the corresponding 2-aminobenzimidazole intermediate 7{1,1}.
After cleavage of intermediate 8{1,1} from the polymer
support, the final cyclization was achieved by heating at 80
°C in DMSO for 3 days. The model compound 9{1,1} was
obtained in good crude purity (62%, calculated from LC-UV
traces) and an overall yield of 58% (quantified after preparative
HPLC purification). Increasing the temperature (up to 150 °C)
in order to shorten the reaction time led to the formation of
multiple side products, lowering the efficiency of the cyclization
step. Also, microwave heating of the cleaved material as well as
microwave-assisted cyclative cleavage of the final compound
from the resin 7{1,1} were tested, but the crude purity was
significantly lower in each case.
RESULTS AND DISCUSSION
■
To test the suggested synthetic approach, Fmoc-L-Ala-OH and
3-fluoro-4-nitrobenzotrifluoride were selected as the represen-
tative building blocks. The general synthetic pathway leading to
target compounds is shown in Scheme 4. Wang resin was
equipped with piperazine carbamate linker 1 which was then
acylated with the Fmoc-amino acid to give intermediate 2.
Cleavage of the Fmoc protective group with piperidine afforded
the intermediate 3{1}. The subsequent reaction with 3-fluoro-
4-nitrobenzotrifluoride led to the formation of nitroaniline
derivative 4{1,1}. The nitro group was further reduced to
amine 5{1,1}, employing sodium dithionite as the reducing
agent and tetrabutylammonium hydrogen sulfate as the phase-
B
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX

ACS Combinatorial Science
Research Article
a
Scheme 4. Synthetic Pathway Leading to Target Compounds
a
Reagents and conditions: (i) (a) CDI, pyridine, DCM, 2 h; (b) piperazine, DCM, overnight. (ii) Fmoc-α-amino acid, DIC, HOBt, DCM/DMF, rt,
overnight. (iii) piperidine, DMF, rt, 30 min. (iv) nitroaryl fluoride, DIPEA, 1,4-dioxane, 80 °C, overnight. (v) sodium dithionite, potassium
carbonate, TBAHS, DCM/water, rt, 2 h. (vi) Only for intermediates 5{R1,3}: Fmoc-NCS, THF, rt, 30 min. (vii) Only for intermediates 6{R1,3}:
DIC, DMF, rt, overnight. (viii) BrCN, DCM, 45 °C, overnight. (ix) TFA, DCM, rt, 30 min. (x) DMSO, 80 °C, 3 days.
Figure 3. Building blocks used for the verification of a synthetic route.
To evaluate limitation and scopes of the developed
methodology, various building blocks with different electronic
and steric properties were tested. For such purposes, we
selected four different nitroaryl fluorides and five Fmoc-amino
acids with different substitutions of the side chain (Figure 3).
When testing the selected nitroaryl fluorides in the reaction
sequence, we found out that intermediates 5{R1,3} prepared
from 3-fluoro-2-nitropyridine did not furnish the desired
compounds 7{R1,3} with use of the cyanogen bromide method.
After the reaction, a mixture of unknown compounds was
detected. For this reason, we synthesized such intermediates in
two steps via the corresponding Fmoc-thioureas 6{R1,3} (see
Scheme 4, steps vi and vii). In the case of Fmoc-amino acids,
the building blocks 1−3 (Figure 3) afforded the final
compounds 9{R1,R2} in good overal crude purity. The only
exception appeared in the case of Fmoc-Ser(OtBu)-OH: The
standard cleavage procedure (TFA/DCM, rt, 30 min) led to
partial cleavage of the tBu-protective group, and two
intermediates, 8{2,R2} and 8{2a,R2}, were obtained. After the
cyclization step, the final compounds 9{2,R2} and 9{2a,R2}
were detected. However, the expected compounds were also
accompanied by compounds 9{2b,R2}, resulting from a thermal
elimination of water (Scheme 5). Only with the use of this
procedure was it possible to isolate some compounds, 10{2,R2},
C
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX

ACS Combinatorial Science
Research Article
a
Scheme 5. Cyclization of Intermediates Prepared from Fmoc-Ser(OtBu)-OH
a
Reagents and conditions: (i) TFA, DCM, rt, 30 min; (ii) TFA, DCM, rt, 120 min; (iii) DMSO, 80 °C, 3 days.
with semipreparative HPLC (in limited yields, see Table 1).
When the longer treatment (2 h) with TFA was used and
we obtained a mixture of compounds 9{2a,1}/9{2b,1} and
9{2a,4}/9{2b,4} in a ratio of 30:70. The separation of
individual compounds was successful in some cases with
semipreparative HPLC. The structure of the methylene
byproduct was finally confirmed by NMR analysis of compound
9{2b,2}.
Table 1. List of Prepared Compounds
Cyclization of intermediates 8{5,R2}, originated from Fmoc-
Gly-OH, totally failed, and only the starting material was
recovered (Scheme 6). Increasing the temperature or the
Scheme 6. Cyclization of Intermediates Prepared from
Fmoc-Gly-OH and Fmoc-β-Ala-OH
crude
purity
(%)
final
purity
(%)
a
b
c
yield
compound
R1
R2
X
(%)
9{1,1}
9{1,2}
9{1,3}
9{1,4}
9{2,1}
9{2,4}
9{2a,1}
9{2a,4}
9{2b,2}
9{3,1}
9{3,2}
9{3,3}
9{3,4}
10{4,1}
10{4,2}
10{4,3}
10{4,4}
Me
Me
Me
Me
7-CF3
H
CH
CH
N
62
73
64
78
52
56
70
58
67
63
65
66
75
61
72
65
58
58
51
54
52
10
14
30
25
30
42
42
65
49
71
57
59
66
92
98
93
91
88
97
87
82
88
98
98
92
93
96
98
93
98
H
6,7-Cl
7-CF3
6,7-Cl
7-CF3
6,7-Cl
H
CH
CH
CH
CH
CH
CH
CH
CH
N
CH2OtBu
CH2OtBu
CH2OH
CH2OH
CH2
reaction time led only to decomposition and the formation of
numerous unknown products. The same results were obtained
for all of the nitroaryl fluorides that were tested. On the other
hand, the use of Fmoc-β-Ala-OH gave the desired derivatives of
3,4-dihydrobenzo[4,5]imidazo[1,2-a]pyrimidin-2(10H)-ones
9{4,R2} for each combination. A list of all prepared derivatives
is displayed in Table 1.
Finally, we have been interested in the resulting stereo-
chemistry of final compounds 9{R1,R2}. To evaluate the effect
of the cyclization step on the single stereocenter leading to
potential racemization, we synthesized two more compounds.
Product 9D{1,1} was prepared from Fmoc-D-Ala-OH, and
compound 9DL{1,1} was synthesized from an equimolar
mixture of Fmoc-D-Ala-OH and Fmoc-L-Ala-OH. The three
compounds 9{1,1}, 9DL{1,1}, and 9D{1,1} were subjected to
capillary electrophoresis analysis by direct chiral separation
employing sulfobuthyl ether β-cyclodextrin as a chiral selector.
The results indicate that the racemization occurred during the
synthesis of 9{1,1} and 9D{1,1}, although in both cases the
pure starting enantiomers were used (see Figure 4).
CH2Ph
7-CF3
H
CH2Ph
CH2Ph
H
CH2Ph
6,7-Cl
7-CF3
H
CH
CH
CH
N
H
6,7-Cl
CH
a
Crude purity calculated from LC-UV traces after the cleavage from
the resin. Overall yields calculated from NMR after preparative
HPLC purification. Calculated from LC-UV traces after preparative
HPLC purification.
b
c
followed by the cyclization step, just two products, 9{2a,R2}
and 9{2b,R2}, were obtained. Their ratio was influenced by the
substitution of the benzene ring (R2): in the case of
unsubstituted and pyridine derivatives, we observed complete
elimination leading to single products 9{2b,2} and 9{2b,3},
while in the case of trifluoromethyl and dichloro substitution,
D
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX

ACS Combinatorial Science
Research Article
Figure 4. Electropherograms of enantioseparation of compounds 9DL{1,1}, 9{1,1}, and 9D{1,1}. Conditions: 100 mM sodium phosphate with 1.0%
of sulfobuthyl ether β-cyclodextrin, U = −15 kV, λ = 280 nm, sample injection 50 mbar/5s.
In conclusion, we have developed an efficient method for the
solid-phase synthesis of 5-noranagrelide derivatives with the use
of Wang resin and the recently described piperazine carbamate
linker. Various building blocks were tested, and 17 model
compounds were prepared and characterized. The target
compounds were isolated in very good overall yields. Despite
observing a few limitations, this method can be applied for the
preparation of chemical libraries with the use of combinatorial
solid-phase organic synthesis due to the simplicity of the
synthetic route and the commercial availability of larger
numbers of building blocks (nitroaryl fluorides, Fmoc-amino
acids).
REFERENCES
■
(1) Hlavac, J.; Soural, M.; Krchnak, V. Practical aspects of
combinatorial solid-phase synthesis. From Solid-Phase Organic Syn-
thesis; Toy, P. H.; Lam, Y., Eds.; Wiley: Hoboken, NJ, 2012; pp 95−
130.
(2) The Merck Index; 11th edition; Merck & Co., Inc.: Whitehouse
Station, NJ; p 98.
(3) Fleming, J. S.; Buyniski, J. P. A potent new inhibitor of platelet
aggregation and experimental thrombosis, anagrelide (BL-4162A).
Thromb. Res. 1979, 15 (3−4), 373−88.
(4) Dingli, D.; Tefferi, A. A critical review of anagrelide therapy in
essential thrombocythemia and related disorders. Leuk. Lymphoma
2005, 46 (5), 641−650.
́
(5) Robak, T.; Trelinski, J.; Hołub, A. Anagrelide−new antiplatelet
drug. Acta Haematol Pol. 1994, 25 (4), 309−15.
(6) Akbar, M.; Foroughifar, N.; Kalhor, M.; Mirabolfathy, M.
Synthesis and antifungal activity of novel 2-benzimidazolylamino-5-
arylidene-4-thiazolidinones. J. Heterocycl. Chem. 2010, 47 (1), 77−80.
(7) Descours, D.; Festal, D. Reactions of N-(ω-chloroalkanoyl)-
carbonimidic dichlorides: a new synthesis of 2-oxo-1,2,3,4-
tetrahydropyrimido[1,2-a]benzimidazoles, 2-oxo-2,3-dihydro-1H-
imidazo[1,2-a]benzimidazoles, and 2-oxo-2,3,5,10-tetrahydro-1H-
imidazo[1,2-b][2,4]benzodiazepines. Synthesis 1983, 12, 1033−1036.
(8) Zarguil, A.; Saadouni, M.; Boukhris, S.; Habbadi, N.; Hassikou,
A.; Souizi, A. Facile and convenient new synthesis of imidazobenzi-
midazol-2-one and pyrimidobenzimidazol-2,3-dione derivatives. Med-
iterr. J. Chem. 2011, 1, 30−37.
ASSOCIATED CONTENT
■
S
* Supporting Information
Details of experimental synthetic and analytical procedures and
spectroscopic data for synthesized compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +420 585634418. Fax: +420 585634465. E-mail:
soural@orgchem.upol.cz.
Notes
(9) Troxler, F.; Weber, H. P. Synthesis of pyrimido[1,2-a]-
benzimidazole by reaction of 2-aminobenzimidazole with dimethyl
acetylenedicarboxylate and their conversion to imidazo[1,2-a]-
benzimidazole. Helv. Chim. Acta 1974, 57 (8), 2356−2364.
(10) Botyanszki, J.; Dickerson, S. H.; Leivers, M. R.; Li, X.;
Mcfadyen, R. B.; Redman, A. M.; Shotwell, J. B.; Xue J. PCT Int. Appl.
2012174312.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are grateful for projects CZ.1.07/2.3.00/20.0009
and CZ.1.07/2.3.00/30.0060 coming from European Social
Fund, from Palacky University (internal grant No.
PrF_2013_027) and the Operational Program Research and
Development for Innovations − European Regional Develop-
ment Fund (project CZ.1.05/2.1.00/03.0058). The infra-
structural part of this project (Institute of Molecular and
Translational Medicine) was supported from the Operational
Program Research and Development for Innovations (project
CZ.1.05/2.1.00/01.0030).
(11) Aso, K.; Hoashi, Y.; Kobayashi, K.; Kojima,T.; Mochizuki, M.;
Takai, T.; Tokumaru, K. U.S. Patent 20090186879.
(12) Srivastava, G. K.; Kesarwani, A. P.; Grover, R. K.; Roy, R.;
Srinivasan, T.; Kundu, B. Solid Phase Synthesis of 2-Aminoquinazo-
line-Based Compounds. J. Comb. Chem. 2003, 5, 769−774.
(13) Ganesan, A. Cyclative Cleavage as the Solid-Phase Strategy. In
Linker Strategies; Scott, P. J. H., Ed.; Wiley: Chichester, West Sussex,
U. K., 2012, p 135−150.
E
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX

ACS Combinatorial Science
Research Article
̌ ́
(14) Neagoie, C.; Krchnak, V. Piperazine Amide Linker for Cyclative
Cleavage from Solid Support: Traceless Synthesis of Dihydroquinox-
alin-2-ones. ACS Comb. Sci. 2012, 14 (7), 399−402.
̌ ́
(15) Kaplanek, R.; Krchnak, V. Fast and effective reduction of
nitroarenes by sodium dithionite under PTC conditions: application in
solid-phase synthesis. Tetrahedron Lett. 2013, 54 (21), 2600−2603.
F
dx.doi.org/10.1021/co4001315 | ACS Comb. Sci. XXXX, XXX, XXX−XXX