Ligand-mediated highly effective and selective C - N coupling for synthesizing bioactive N -aryltriazole acyclonucleosides

Article abstract of DOI:10.1021/ol102537p

N-aryltriazole nucleosides are new chemical entities with potential biological activity. The two phosphor ligands, Synphos and Xantphos, had a selective and effective impact on Pd-catalyzed C - N coupling with the 5- and 3-bromotriazole acyclonucleoside i

Full text of DOI:10.1021/ol102537p

ORGANIC
LETTERS
2010
Vol. 12, No. 24
5712-5715
Ligand-Mediated Highly Effective and
Selective C-N Coupling for
Synthesizing Bioactive N-Aryltriazole
Acyclonucleosides
Yuting Fan,† Yi Xia,‡ Jingjie Tang,† Palma Rocchi,§ Fanqi Qu,† Juan Iovanna,§
and Ling Peng‡,*
State Key Laboratory of Virology, College of Chemistry and Molecular Sciences,
Wuhan UniVersity, Wuhan 430072, People’s Republic of China, and CNRS UPR 3118
CINaM and INSERM U624, 163, aVenue de Luminy, 13288 Marseille, France
ling.peng@uniVmed.fr
Received October 19, 2010
ABSTRACT
N-aryltriazole nucleosides are new chemical entities with potential biological activity. The two phosphor ligands, Synphos and Xantphos, had
a selective and effective impact on Pd-catalyzed C-N coupling with the 5- and 3-bromotriazole acyclonucleoside isomers, affording the
corresponding and otherwise difficult to achieve N-aryltriazole nucleosides with good to excellent yields. In addition, two of the synthesized
nucleosides showed superior anticancer activity against drug-resistant pancreatic cancer, compared to the reference drug gemcitabine.
Nucleoside mimics are of considerable importance in the
search for antiviral and anticancer drug candidates.1 Com-
pounds with triazole heterocycles as nucleobases are unusual
nucleoside analogues, some of which exhibit biological
activity involving unique modes of action. One noteworthy
example is ribavirin, the first synthetic triazole nucleoside
antiviral drug discovered 40 years ago, which remains the
only small molecule approved to treat hepatitis C virus
(HCV).2 Recently, ribavirin has been reported to demonstrate
apoptosis-related anticancer effects.3 Consequently, there is
renewed interest in creating new structural entities of triazole
nucleosides with the aim of developing potent therapeutic
† Wuhan University.
(2) (a) Sidwell, R. W.; Huffman, J. H.; Khare, G. P.; Allen, L. B.;
Witkowski, J. T.; Robins, R. K. Science 1972, 177, 705. (b) Feld, J. J.;
Hoofnagle, J. H. Nature 2005, 436, 967.
‡ CNRS UMR 3118, CINaM.
§ INSERM U624.
(1) (a) Herdewijn, P., Ed. Modified nucleosides in biochemistry, bio-
technology and medicine; WILEY-VCH: Weinheim, Germany, 2008. (b)
Claire, S. Nucleoside mimetics: Their chemistry and biological properties;
Gordon and Beach: Amsterdam, 2001.
(3) (a) Assouline, S.; Culjkovic, B.; Cocolakis, E.; Rousseau, C.; Beslu,
N.; Amri, A.; Caplan, S.; Leber, B.; Roy, D.-C.; Miller Jr, W. H.; Borden,
K. L. B. Blood 2009, 114, 257. (b) Borden, K. L. B.; Culjkovic-Kraljacic,
B. Leuk. Lymphoma 2010, 51, 1805.
10.1021/ol102537p  2010 American Chemical Society
Published on Web 11/18/2010

agents with novel mechanisms of action. Over the past few
years, we have developed a series of triazole nucleoside
analogs bearing aromatic moieties on the triazole nucleo-
base4-6 some of which exhibited antiviral and anticancer
activity involving novel modes of action.4-6 Of particular
interest are some N-aryltriazole nucleosides, which showed
potent anticancer activity against drug-resistant pancreatic
cancer.6a However, their synthesis via Cu-mediated N-
arylation was far from satisfactory, with narrow substrate
scope and low product yields,6 thus limiting further biological
evaluation and structure/activity relationship studies.
Pd-catalyzed C-N coupling7 is becoming a powerful
technique in the synthesis of various N-arylated nucleoside
analogs.8 However, no reports exist on Pd-catalyzed C-N
cross coupling with triazole nucleoside precursors. In marked
contrast to simpler aromatic systems, triazole nucleosides
are particularly challenging for Pd-catalyzed cross coupling
reactions due to the low reactivity of the triazole ring, the
multiple coordinating N- and O-atoms and the labile glyco-
sidic bond. Here, we report the synthesis of novel N-
aryltriazole acyclonucleosides via highly reactive Pd-
catalyzed C-N cross-coupling. The phosphor ligands, Synphos
and Xantphos, had a selective and effective impact on the
C-N coupling reactions with the bromotriazole nucleoside
substrates, giving the corresponding products with good to
excellent yields. In addition, some of the synthesized triazole
nucleosides showed potent anticancer activity against drug-
resistant pancreatic cancer, with potentially novel modes of
action.
In this study, we used both 5-bromotriazole acyclonucleo-
side 1 and its structural isomer, 3-bromotriazole nucleoside
29 as the substrates for Pd-catalyzed C-N coupling (Scheme
1). Under the classical condition of 10 mol % Pd(OAc)2, 12
mol % BINAP and 2 equiv of Cs2CO3 in toluene,10 we only
obtained the C-N coupling product with 1 (Table S1, entry
1, Supporting Information), but not with 2. We therefore first
focused our attention on the reaction with 1. After extensive
screening (Table S1, Supporting Information), we defined
the optimized condition as Pd2(dba)3 with the combination
of Synphos as the ligand, K2CO3 as the base and toluene as
the solvent. It is noteworthy that monocoordinating ligands
Scheme 1. Bromotriazole Nucleosides 1 and 2 Were Used As
Starting Materials for the C-N Coupling Reactions in This
Study
turned out to be inefficient in this reaction (Figure S1, Table
S1, entries 4-6, Supporting Information), in concordance
with previous report.11 Although C3-Tunephos yielded
considerable amount of the product 3a (Table S1, entry 2,
Supporting Information), it also led to the formation of
numerous byproduct. Therefore, Synphos emerged as the
most effective ligand in our conditions (Table S1, entry 8,
Supporting Information). Furthermore, the nature and the
strength of bases affected importantly the reaction. Potassium
carbonate was superior to all the other bases tested (Table
S1, entry 13, Supporting Information): strong bases such as
NaOtBu and LiHMDS resulted in substrate decomposition;
whereas weak bases such as Li2CO3 were also ineffective
and failed to promote the reaction, but did not result in
substrate decomposition. The choice of solvent was also
critical, and toluene appeared to be the best choice (Table
S1, entry 13, Supporting Information), whereas the polar
solvent DMF yielded no product, but rather substrate
decomposition. Finally, the catalyst precursors also had a
considerable impact, with Pd2(dba)3 proving to be the most
efficient (Table S1, entries 19-21, Supporting Information)
allowing a high yield even with a catalyst loading of 1 mol
% (Table S1, entries 22, Supporting Information).
We next explored the scope of arylamine substrates (Table
1). The reaction worked extremely well with electron-rich
arylamines, affording the corresponding products with excel-
lent yields (Table 1, entries 1-7). Arylamines with electron-
deficient substituents led to decreased yet very good yields
(Table 1, entries 9-10). The use of even sterically hindered
arylamines offered the products with good to excellent yields
(Table 1, entries 6-7), except for pyrenylamine which gave
considerably reduced yields (Table 1, entry 8) probably due
to the particularly large pyrenyl moiety which would create
steric hindrance during the N-arylation process. A support
for this hypothesis came from the crystal structure of 3f
(Figure 1). As we can see, the glycosidic moiety in 3f is in
such a position that the approach of an exceedingly larger
amine to the reaction center might create steric hindrance
and therefore impede the reaction.
(4) Xia, Y.; Qu, F. Q.; Peng, L. Mini-ReV. Med. Chem. 2010, 10, 806
.
(5) (a) Xia, Y.; Liu, Y.; Wan, J. Q.; Wang, M. H.; Rocchi, P.; Qu, F. Q.;
Neyts, J.; Iovanna, J. L.; Peng, L. J. Med. Chem. 2009, 52, 6083. (b) Wan,
J. Q.; Xia, Y.; Liu, Y.; Wang, M. H.; Rocchi, P.; Yao, J. H.; Qu, F. Q.;
Neyts, J.; Iovanna, J. L.; Peng, L. J. Med. Chem. 2009, 52, 1144. (c) Wang,
M. H.; Xia, Y.; Fan, Y. T.; Rocchi, P.; Qu, F. Q.; Iovanna, J. L.; Peng, L.
Bioorg. Med. Chem. Lett. 2010, 20, 5979
.
(6) (a) Liu, Y.; Xia, Y.; Fan, Y. T.; Maggiani, A.; Rocchi, P.; Qu, F. Q.;
Iovanna, J. L.; Peng, L. Bioorg. Med. Chem. Lett. 2010, 20, 2503. (b) Li,
W.; Fan, Y. T.; Xia, Y.; Rocchi, P.; Zhu, R. Z.; Qu, F. Q.; Neyts, J.; Iovanna,
J. L.; Peng, L. HelV. Chim. Acta 2009, 92, 1503
.
(7) (a) Surry, D.; Buchwald, S. Angew. Chem., Int. Ed. 2008, 47, 6338.
(b) Hartwig, F. Nature 2008, 455, 314.
(8) (a) Thomson, P. F.; Lagisetty, P.; Balzarini, J.; De Clercq, E.;
Lakshman, M. K. AdV. Synth. Catal. 2010, 352, 1728. (b) Lagisetty, P.;
Russon, L. M.; Lakshman, M. K. Angew. Chem., Int. Ed. 2006, 45, 3660.
(c) Ngassa, F. N.; DeKorver, K. A.; Melistas, T. S.; Yeh, E. A.-H.;
Lakshman, M. K. Org. Lett. 2006, 8, 4613. (d) Lakshman, M. K. Curr.
Org. Synth. 2005, 2, 83, and references therein.
(9) Zhu, R. Z.; Qu, F. Q.; Que´le´ver, G.; Peng, L. Tetrahedron Lett. 2007,
48, 2389.
(10) Chakraborti, D.; Colis, L.; Schneider, R.; Basu, A. K. Org. Lett.
2003, 5, 2861–2864.
(11) Lakshman, M. K.; Hilmer, M. J. Q.; Keeler, J. Q.; Dinh, Y. V.;
Ngassa, F. N.; Russon, L. M. J. Am. Chem. Soc. 2001, 123, 7779.
Org. Lett., Vol. 12, No. 24, 2010
5713

toluene emerged as being the most effective for the C-N
coupling of 2 (Table S2, entry 17-19, Supporting Informa-
tion). This condition was itself, however, unsuitable for the
C-N coupling of 1 (data not shown). The explanation for
the observed difference in reactivity and selectivity of the
two catalytic systems for the C-N coupling with 1 and 2 is
complex. First, the reactivity of 1 and 2 is considered to be
different due to their differing electronic and structural
features.12 Second, Xantphos is a large bite-angle ligand,13
and its Pd-complexes may be more cumbersome in space.
Moreover, on the basis of the structure of the C-N coupling
product 4a (Figure 2), we may speculate that the reaction
Table 1. Synthesis of N-Aryltriazole Acyclonucleosides 3 via Pd
Catalyzed C-N Coupling with 5-Bromotriazole Nucleoside 1a
Figure 2. X-ray structure of 4a.
site of 2 for C-N coupling could be exposed to exterior
and easily accessible for Pd-complexes, allowing large-bite
angle ligand Xantphos to participate favorably in the catalytic
process; whereas the structure of 3f (Figure 1) might suggest
that the reaction site of 1 could be highly congested due to
the unfavorable position of the glycosyl moiety, and unable
to accommodate favorably the large bite-angle ligand Xant-
phos, but rather the small bite-angle ligand Synphos.14
Finally, the triazole ring and the phenyl moiety together with
the N-linkage are in almost perfect coplanarity in the product
4a (Figure 2), leading to an enlarged aromatic systems, which
might confer a further thermodynamic advantage and con-
tribute favorably to the C-N bond formation with 2. This
might also explain why a good yield could be obtained even
with a very low catalyst loading of 0.5 mol % (Table S1,
entry 20, Supporting Information).
a Reaction condition is: 1 (0.2 mmol), arylamine (0.4 mmol), Pd2(dba)3
(0.005 mmol), Synphos (0.012 mmol), K2CO3 (0.4 mmol), toluene (2 mL),
100 °C, 16 h.
Under the optimized condition, excellent yields were
obtained for the C-N coupling of 2 with arylamines bearing
a wide range of functional groups including both electron-
donating and electron-withdrawing groups (Table 2). This
may be attributed to the highly reactive Pd-Xantphos
catalytic system. Even the sterically hindered pyrenylamine
gave excellent result (Table 2, entry 8). A hypothesis could
be formulated based on the crystal structure of 4a (Figure
2): with the glycosyl moiety situated far away, the reaction
site of 2 might be openly accessible for the C-N bond
formation and no steric congestion would be created even
with an amine bearing an exceedingly cumbersome aromatic
moiety.
Figure 1. X-ray structure of 3f.
Encouraged by the extremely positive result with 1, we
further investigated the Pd-catalyzed C-N coupling approach
using the 3-bromotriazole nucleoside 2. Unfortunately, the
optimized condition for C-N coupling of 1 was not suitable
for 2 (Table S2, entry 1, Supporting Information), obviously
owing to the different reactivity of 1 and 2 toward the C-N
coupling. We thus further scrutinized the reaction conditions
by examining various Pd-catalysts, ligands, bases and
solvents (Table S2, Supporting Information). The combina-
tion of Pd2(dba)3 and Xantphos in the presence of K2CO3 in
We further deprotected 3 and 4 in NH3/MeOH to give,
respectively, the N-aryltriazole nucleosides 5 and 6 with good
(12) Wu, Q. Y.; Qu, F. Q.; Wan, J. Q.; Zhu, X.; Xia, Y.; Peng, L. HelV.
Chim. Acta 2004, 87, 811.
(13) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M. Organometallics 1995, 14, 3081.
(14) Jeulin, S.; Duprat de Paule, S.; Ratovelomanana-Vidal, V.; Geneˆt,
J.; Champion, N.; Dellis, P. Proc. Natl. Acad. Sci. 2004, 101, 5799.
5714
Org. Lett., Vol. 12, No. 24, 2010

Table 2. Synthesis of N-Aryltriazole Nucleosides 4 via Pd
Catalyzed C-N Bond Formation Starting with 2a
Scheme 2
.
Deprotection of 3 and 4 Led to the Nucleoside
Analogues 5 and 6, Respectively
Figure 3. Compounds 6c and 6h showed superior anticancer activity
to gemcitabine against drug-resistant pancreatic cancer MiaPaCa-2
cells, assessed by MTT assay (A); however, they did not inhibit
DNA synthesis as gemcitabine, measured by [3H]-dTMP incorpora-
tion (B) (Ctrl, control; Gem, gemcitabine).
a Reaction condition is: 2 (0.2 mmol), arylamine (0.4 mmol), Pd2(dba)3
(0.002 mmol), Xantphos (0.0048 mmol), K2CO3 (0.4 mmol), toluene (2
mL), 100 °C, 16 h.
respectively. A large array of N-aryltriazole nucleosides was
synthesized efficiently, and two of them showed remarkable
anticancer activity against drug-resistant pancreatic cancer.
The synthetic methods presented here not only help us to
gain a better knowledge of ligand-impacted Pd-catalyzed
coupling reactions in organic synthesis but also provide us
with the opportunity to prepare various new structural
paradigms of N-aryltriazole nucleosides with ease for our
drug discovery program based on triazole nucleoside chem-
istry.
to excellent yields (Scheme 2, Tables S3 and S4, Supporting
Information). Two of them, 6c and 6h, were revealed to have
superior anticancer activity against drug-resistant pancreatic
cancer MiaPaCa-2 cells, compared to gemcitabine (Figure
3A), the current first-line drug used to treat pancreatic
cancer.15 In addion, unlike gemcitabine, neither 6c nor 6h
inhibited DNA synthesis (Figure 3B), suggestive of poten-
tially novel mechanisms of action underlying their anticancer
activity.
Acknowledgment. We acknowledge financial support
from the National Mega Project on Major Drug Development
(2009ZX09301-014), Wuhan University, CNRS and IN-
SERM. We thank Dr. Sandrine Henri for the help with
radioactivity experiments. Dr. Yi Xia is supported by la
Fondation pour la Recherche Me´dicale.
In summary, we have reported highly reactive and selective
Pd/ligand systems for C-N cross-coupling to synthesize
previously unavailable and otherwise difficult to achieve
N-arylated triazole acyclonucleoside analogs. The two phos-
phor ligands, Synphos and Xantphos, are fully complemen-
tary in terms of selectivity and reactivity for Pd-catalyzed
C-N coupling with the two nucleoside isomers 1 and 2,
Supporting Information Available: Experimental pro-
cedures and spectral data. This material is available free of
charge via the Internet at http://pubs.acs.org.
(15) Berlin, J.; Benson, A. B. Nat. ReV. Clin. Oncol. 2010, 7, 135.
OL102537P
Org. Lett., Vol. 12, No. 24, 2010
5715