Click fleximers: A modular approach to purine base-expanded ribonucleoside analogues

Article abstract of DOI:10.1039/c2ob25678a

The synthesis of nucleoside analogues incorporating 4-(5-pyrimidinyl)-1,2, 3-triazole aglycons as expanded purine nucleobase mimics were accessed using the copper-catalyzed azide-alkyne Huisgen cycloaddition between a ribosyl azide and 5-alkynylpyrimidines. Depending on the nature of the alkyne employed, other nucleoside analogues that possess fluorescence or potential metal-binding properties were prepared. Computational studies were undertaken on the purine analogues and indicate that the heterocycles of the unfused nucleobase prefer a coplanar arrangement and the anti-glycosidic conformer is favoured in most instances.

Full text of DOI:10.1039/c2ob25678a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 6521
www.rsc.org/obc
PAPER
Click fleximers: a modular approach to purine base-expanded ribonucleoside
analogues†
André H. St. Amant, Leslie A. Bean, J. Peter Guthrie and Robert H. E. Hudson*
Received 5th April 2012, Accepted 6th June 2012
DOI: 10.1039/c2ob25678a
The synthesis of nucleoside analogues incorporating 4-(5-pyrimidinyl)-1,2,3-triazole aglycons as
expanded purine nucleobase mimics were accessed using the copper-catalyzed azide–alkyne Huisgen
cycloaddition between a ribosyl azide and 5-alkynylpyrimidines. Depending on the nature of the alkyne
employed, other nucleoside analogues that possess fluorescence or potential metal-binding properties
were prepared. Computational studies were undertaken on the purine analogues and indicate that the
heterocycles of the unfused nucleobase prefer a coplanar arrangement and the anti-glycosidic conformer
is favoured in most instances.
Introduction
The synthesis of a new class of expanded nucleobases termed
“fleximers” was first described by Seley in 2001.1,2 The separa-
tion of a purine nucleobase into its imidazole and pyrimidine
components attached via a single C–C bond effectively con-
verted a rigid fused bicyclic system into one that permitted con-
formational freedom.3,4 These compounds were designed to be
flexible bioprobes for enzymes. In this vein, a fleximer inhibitor
of S-adenosyl-L-homocysteine hydrolase has been discovered5
and an azido derivative of a G-fleximer has been used in photo-
affinity labelling of catalytically important amino acids in the
active site of GTP fucose pyrophosphorylase.6
The first generation fleximers possessed a C5-imidazole to
Fig. 1 Comparison of
analogues.
a generic purine nucleoside to fleximer
C6-pyrimidine attachment. The second generation fleximers
were constructed with a C4-imidazole to C5-pyrimidine linkage
which alters the trajectory of the base-pairing moiety relative to
the former attachment (Fig. 1).7,8
The synthesis of both generation fleximers was synthetically
challenging and given the potential utility of fleximers, we were
inspired to examine an alternative route to related analogues.
Some recent research in our group has focused on the derivatiza-
tion of 5-ethynyl-2′-deoxycytidine via the copper catalyzed
azide–alkyne Huisgen cycloaddition (CuAAC),9 and recognizing
the structural similarity between this class of molecule and
fleximers motivated the current synthetic study (Fig. 2). By
Department of Chemistry, The University of Western Ontario, London,
Ontario, Canada N6A 5B7. E-mail: robert.hudson@uwo.ca;
Fax: +01 519 661 3022; Tel: +01 519 661 2166
Fig. 2 Design of nucleoside analogues accessed by click chemistry.
†Electronic supplementary information (ESI) available: Complete experi-
mental details and characterization of compounds 2c, 2d, 3c, 3d, 4a–g,
5a–g and the fluorescence data for 4f, 4g, 5d, 5f, and 5g and details of
the computational studies. See DOI: 10.1039/c2ob25678a
replacing the imidazole in Seley’s second generation fleximers
with a 1,2,3-triazole we envisioned a facile synthesis using the
CuAAC.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6521–6525 | 6521

View Article Online
Of course, the popularity of CuAAC has led to its wide
exploitation in nucleoside, nucleotide and oligonucleotide
chemistries.10 Ribosyl azides have been employed in CuAAC to
produce N-triazolylglycosides predominately with a preponder-
ance of aromatic or alkyl alkynes that do not possess nucleo-
base-like functionality.11–14 In related work, N-1 or N-9
propargyl nucleobases have been reacted with 1-azido-2-deoxy-
ribofuranose under CuAAC conditions to produce tethered bases
which may be viewed as methylene-spaced fleximer homo-
logues.10 To date, there have been no reported synthetic efforts
toward nucleoside fleximers accessed via CuAAC, which we
now report, along with two new nucleoside fluorophores con-
ceptually related to “fluorosides”.15
These nucleobases lack an acidic imide which then dictated an
acidic work up to afford water solubility. The synthesis of the
2-ethynylpyridine (3e) followed literature precedent from
2-bromopyridine.20 The fluorescent alkyne 2-ethynylfluorene
(3f) was synthesized following literature procedure from
fluorene21 and 1-ethynylpyrene (3g) was obtained commercially.
The CuAAC to produce the benzoyl-protected click fleximers
required more forcing conditions than typical click reactions.22
The optimized conditions were refluxing THF and CuI paired
with the metal chelator N,N′-dimethylethylenediamine thus pro-
ducing 4a–g in 75% to quantitative yield (Scheme 2). Com-
pounds 4a–c were discovered to be organogelators while
attempting to characterize the compounds in CDCl3. Previously,
the ability of nucleosides to gelate solvents has been investigated
for use as supramolecular templates or as drug delivery assem-
blies.23 1,2,3-Triazole containing nucleosides have been discov-
ered to be hydro- and organogelators, a property attributed to the
perpendicular nature of their interactions (base pairing and π–π
stacking),24,25 so it was not completely surprising that the
benzoyl-protected nucleoside click fleximers 4a–c were able to
gelate organic solvents. The gelation properties were not investi-
gated fully; however, it was observed that reversible gels were
formed in CHCl3, EtOAc, and DMSO. Despite its structural
similarity to aminopyrimidine 4c, the 2,4-diaminopyrimidine
click fleximer 4d was found to be freely soluble in CHCl3.
Results and discussion
We chose to target the ribonucleosides in order to exploit the
β-directing ability of the 2′-hydroxyl protecting group during
Lewis-acid catalyzed glycosyl azide formation. The peracylated
riboses, acetyl or benzoyl, were synthesized from D-ribose fol-
lowing literature procedures.12,16 Although the preparation of the
peracyl-β-azidoribose is reported as being highly stereo-
selective,16 under a variety of reaction conditions we were
unable to reproduce these results. Consequently, the α-anomer
was removed by column chromatography, which was easier to
achieve for the benzoyl-protected ribose (1).
For the selection of alkynes the easily accessible uracil (3a),
cytosine (3b) and 4-aminopyrimidine (3c) alkynes were chosen
to produce analogues of xanthosine (X), isoguanosine (isoG)
and adenosine (A), respectively. The 2,4-diaminopyrimidine
alkyne (3d) was selected as a mimic of diaminopurine riboside
(DAP). The 2-pyridinyl (3e) alkyne was chosen for its potential
ability as a bidentate ligand for metal chelation.17 The alkynes 3f
(fluorene) and 3g (pyrene) were selected as fluorescent reporter
groups with the potential to form intercalation complexes.
The 5-ethynylpyrimidines of uracil (3a) and cytosine (3b)
were synthesized based on published procedures.18,19 The
5-iodopyrimidine was used in a Sonogashira coupling with
TMS-acetylene, followed by removal of the TMS group under
alkaline conditions. The 4-aminopyrimidine and 2,4-diaminopyri-
midine alkynes (3c and 3d) were synthesized from the 5-iodopyri-
midines in 72% and 76% yield, respectively (Scheme 1).
The pyridine compounds 4e and 5e have been reported pre-
viously but were not examined for metal binding properties
at that time.14 Subsequently, it has been reported that the
Scheme 1 Synthesis of 5-ethynyl-pyrimidines.
Scheme 2 Synthesis of “click fleximers” 5a–g.
6522 | Org. Biomol. Chem., 2012, 10, 6521–6525
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Steady state fluorescent properties of click fleximers
ΦF
Excitation λmax (nm)
Emission λmax (nm)
EtOH
H2O
4f
292
352
311
311
346
317
383
356
331
382
0.16
0.16
0.48
—
—
—
0.05
0.06
0.46
4g
5d
5f
5g
—
Fig. 4 Nucleoside numbering scheme.
Table 2 Summary of relative conformational energies
Glycosidic
conformation
(kcal mol−1
)
Minimum barrier to rotation of
heterobiaryl bonda (kcal mol−1
)
syn-
anti-
X: 5a
IsoG: 5b
A: 5c
0.509
−1.338
1.965
0
0
0
0
11.9
4.8
7.1
6.1
DAP: 5d
2.167
a Estimate of the minimal barrier from the energy vs. torsion profiles for
the N1-methyl base analogues. The transition state energy was found
rigorously only for 5c, and had a relative energy of 7.28 kcal mol−1
.
The click fleximer 5d is strongly emissive in EtOH (ΦF = 0.48)
and much less so in H2O (ΦF = 0.05). This large change in
emissivity may be useful for studying its interaction with
enzymes or as a reporter group if incorporated into oligomers.
The conformations for the flexible nucleoside analogues were
studied computationally using DFT methods (Bb3lyp/6-31+G**)
implemented by Gaussian 09 (see ESI† for details).27 Favoured
glycosidic conformers and the favoured orientation of the
heterobiaryl bond were identified as well as barriers to rotation
about this bond. The nucleosides are described by the numbering
scheme and nomenclature given in Fig. 4. The conventional
numbering for the ribose and pyrimidine are preserved and the
glycosidic linkage is assigned to N1 of the triazole. The anti
glycoside is defined when the triazole H5 is anti to the ribose
(O4′–C1′–N1–C5 torsion angle ∼ −130°, or +230°) whereas
the syn conformer places the triazole H5 over the ribose ring
(O4′–C1′–N1–C5 torsion angle ∼63°).
Fig. 3 Steady state fluorescence spectra for 5d (10−5 M in EtOH):
excitation (blue line) and emission (red line).
2-pyridinyl-4-[1,2,3-triazole] moiety acts as a metal chelator for
several metals including Pt2+, but not in the context as a nucleo-
side analog.17 Preliminary studies in our lab with a 2-pyridinyl-
4-[1,2,3-triazole] analogue have shown the ability to form
a complex with Ni2+, in the context of modified peptide nucleic
acids.26
The fluorescence properties of the selected fleximers were
studied (Table 1). The fluorescent quantum yields (ΦF) were
determined in EtOH, for solubility of the benzoyl-protected click
fleximers, or in water for the free nucleosides. Removal of the
benzoyl protecting groups in compounds 4a–g was achieved by
suspension in methanolic ammonia to yield 5a–g in 47–91%.
Somewhat to our surprise, the diaminopyrimidine fleximer
5d displayed purple fluorescence when examined under long-
wavelength UV irradiation (360 nm) on a TLC plate, whereas
the other nucleobase triazoles were not obviously emissive. The
fluorescence properties of 5d (Fig. 3) were characterized along
with the other fleximers incorporating known luminophores.
The fluorescence emission spectrum of the pyrene fleximer 4g
displayed an elongated tail from 450–550 nm which is not due
to excimer formation as revealed by examination of the concen-
tration dependence of the spectra (see ESI†). Both the fluorene
and pyrene exhibited significant solvatochromism. The ΦF of the
fluorene fleximer 5f increases approximately 3-fold from water
(0.06) to the less polar EtOH, indicating its potential usefulness
an environmentally sensitive fluorescence reporter group. The
pyrene derivatives 5g and 4g displayed the reverse trend, where
the ΦF increases about 3-fold upon moving from EtOH to H2O.
The relaxed structures of the fleximers 5a–5d maintained
3′-endo-like ring pucker that is usual for ribosides. The anti-
glycosidic conformer was the lowest energy structure for
fleximers A (5c, shown below, Fig. 5), X and DAP, whereas the
syn conformation was slightly preferred by isoG (see ESI†),
Table 2. Observed in both the anti- and syn- conformations, the
triazole ring (either N2 or H5) is oriented to nearly bisect the
furanose ring (Fig. 5a). The energy profile for rotation about
the glycosidic bond indicates that the anti-conformer is favoured
by nearly 2 kcal mol−1. The higher energy structures correspond
to eclipsing interactions of N1–C5(triazole) with C1′–C2′(points
21, 22) or O4′–C1′ (point 13, Fig. 5b).
The anti-conformers are potentially stabilized by an O5′–H to
N2 hydrogen bond, which is realized in the calculations for
fleximers A, X and DAP. The nonconventional hydrogen bond
from C5–H(triazole) to O5′, although possible in all the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6521–6525 | 6523

View Article Online
Fig. 6 Energy profile for rotation about the biaryl bond for fleximer A,
wherein the ribose was replaced with a methyl group, see ESI.†
Fig. 5 (a) Molecular representations of fleximer A, 5c, illustrating ring
pucker and orientation of the aglycone of the lowest energy structure #4.
(b) Energy profile for stepwise rotation about the glycosidic bond (O4′–
C1′–N1–C5 torsion angle) in 15° increments.
fleximers, is only preferred for isoG. There are no obvious
structural features noted that are responsible for this change in
glycosidic conformational preference.
The preferred conformation of the biaryl bond was examined
for derivatives wherein the ribose was replaced with a methyl
group to reduce the computational burden. The conformations
are dictated by the presence of an intramolecular H-bond
between the triazole ring and the pyrimidine. Each fleximer
possesses the ability to form a conventional H-bond between the
exocyclic amino group and the triazole N3 (A, DAP, isoG) or
a non-conventional H-bond between the triazole C5–H and
carbonyl O4 (X) which favours a coplanar arrangement of the
rings (Fig. 6).
For example, the lowest energy structure corresponding to a
torsion of ∼0° (N3–C4–C5–C4), for fleximer A (5c), (Fig. 6), is
given along with representations of the relaxed structures of the
other fleximers (Fig. 7, details for the calculations are found in
the ESI†).
Fig. 7 Representations of the relaxed conformations of the fleximers.
approach to new fleximer analogues, of which four have been
synthesized. These analogues possess subtly different confor-
mational properties than those already known and may usefully
augment studies on enzymatic nucleoside substrate acceptance.
Importantly, the expanded DAP analogue was found to be lumi-
nescent, a property that may be useful for studying its interaction
with enzymes or upon incorporation into oligonucleotides.
Additionally, it was demonstrated that new ribo-“fluorosides”
could be accessed by the same chemical route.
Calculations showed the heterocycles were coplanar which is
favoured due to the bonding interactions as shown in Fig. 7.
Whereas calculations on Seley’s fleximers showed that the aryl
groups were not coplanar, the triazole fleximers appear to have a
preference for coplanarity.
Acknowledgements
The Natural Sciences and Engineering Research Council
(NSERC) of Canada is gratefully acknowledged for financial
support this work and SHARCNET (Western) for computational
time. We also thank Dr Yogesh Sanghvi of Rasayan Inc. for
material support. Dr Mark S. Workentin is thanked for helpful
discussion.
Conclusions
We have prepared ribonucleoside analogues possessing substi-
tuted triazole aglycons by CuAAC. This represents a convergent
6524 | Org. Biomol. Chem., 2012, 10, 6521–6525
This journal is © The Royal Society of Chemistry 2012

View Article Online
18 (a) Z. Janeba, J. Balzarini, G. Andrei, R. Snoeck, E. De Clercq and
M. J. Robins, Can. J. Chem., 2006, 84, 580–586; (b) R. H. E. Hudson
and J. M. Moszynski, Synlett, 2006, 18, 2997–3000.
19 (a) R. H. E. Hudson and A. K. Dambenieks, Heterocycles, 2006, 68,
1325–1328; (b) P. Kielkowski, R. Pohl and M. Hocek, J. Org. Chem.,
2011, 76, 3457–3462.
Notes and references
1 K. L. Seley, L. Zhang and A. Hagos, Org. Lett., 2001, 3, 3209–
3210.
2 K. L. Seley, L. Zhang, A. Hagos and S. Quirk, J. Org. Chem., 2002, 67,
3365–3373.
20 S. A. Al-Taweel, Phosphorus, Sulfur Silicon Relat. Elem., 2002, 177,
1041–1045.
3 M. Polak, K. L. Seley and J. Plavec, J. Am. Chem. Soc., 2004, 126,
8159–8166.
21 F. Wojciechowski and R. H. E. Hudson, Nucleosides, Nucleotides Nucleic
Acids, 2007, 26, 1199–1202.
4 A. B. Bardon and S. D. Wetmore, J. Phys. Chem. A, 2004, 109, 262–
272.
22 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, Angew.
Chem., 2002, 114, 2708–2011.
5 K. L. Seley, S. Quirk, S. Salim, L. Zhang and A. Hagos, Bioorg. Med.
Chem. Lett., 2003, 13, 1985–1988.
23 A. Vintiloiu and J. C. Leroux, J. Controlled Release, 2008, 125, 179–
192.
24 S. M. Park, Y. S. Lee and B. H. Kim, Chem. Commun., 2003, 2912–
2913.
6 S. Quirk and K. L. Seley, Biochemistry, 2005, 44, 13172–13178.
7 K. L. Seley, S. Salim and L. Zhang, Org. Lett., 2005, 7, 63–66.
8 K. L. Seley, S. Salim, L. Zhang and P. I. O’Daniel, J. Org. Chem., 2005,
70, 1612–1619.
25 D. W. Dodd, N. D. Jones and R. H. E. Hudson, Artif. DNA: PNA XNA,
2010, 1, 90–95.
26 Unpublished observations.
9 D. W. Dodd, K. N. Swanick, J. T. Price, A. L. Brazeau, M. J. Ferguson,
N. D. Jones and R. H. E. Hudson, Org. Biomol. Chem., 2010, 8, 663–
666.
27 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov,
J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma,
V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich,
A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, GAUSSIAN 09 (Revision C.01), Gaussian, Inc., Wallingford
CT, 2010.
10 F. Amblard, J. H. Cho and R. F. Schinazi, Chem. Rev., 2009, 109, 4207–
4220.
11 U. Pradere, V. Roy, T. R. McBrayer, R. F. Schinazi and L. A. Agrofoglio,
Tetrahedron, 2008, 64, 9044–9051.
12 K. El Akri, K. Bougrin, J. Balzarini, A. Faraj and R. Benhida, Bioorg.
Med. Chem. Lett., 2007, 17, 6656–6659.
13 S. Hou, W. Liu, D. Ji and Z. Zhao, Bioorg. Med. Chem. Lett., 2011, 21,
1667–1669.
14 E. J. Amigues, E. Armstrong, M. Dvorakova, M. E. Migaud and
M. Huang, Nucleosides, Nucleotides Nucleic Acids, 2009, 28, 238–
259.
15 Reviewed in: Y. N. Teo and E. T. Kool, Chem. Rev., 2012, DOI: 10.1021/
cr100351g, web published.
16 A. Štimac and J. Kobe, Carbohydr. Res., 2000, 324, 149.
17 C. Richardson, C. M. Fitchett, F. R. Keene and P. J. Steel, Dalton Trans.,
2008, 2534–2537.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6521–6525 | 6525