Hydrogelation and spontaneous fiber formation of 8-aza-7-deazaadenine nucleoside 'click' conjugates

Article abstract of DOI:10.1016/j.tet.2011.07.015

Nucleoside hydrogels based on benzyl azide 'click' conjugates of 8-aza-7-deaza-2′-deoxyadenosine bearing 7-ethynyl, 7-octa-(1,7-diynyl), and 7-tri-prop-2-ynyl-amine side chains were synthesized (1, 3, 4). The cycloaddition adduct with the shortest linker

Full text of DOI:10.1016/j.tet.2011.07.015

Tetrahedron 67 (2011) 7418e7425
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Hydrogelation and spontaneous fiber formation of 8-aza-7-deazaadenine
nucleoside ‘click’ conjugates
,
b
,
a
,
b
c
Frank Seela a , Suresh S. Pujari , Andreas H. Schafer
*
€
a
€
Laboratory of Bioorganic Chemistry and Chemical Biology, Center for Nanotechnology, Heisenbergstraße 11, 48149 Munster, Germany
Laboratorium fur Organische und Bioorganische Chemie, Institut fur Chemie, Universitat Osnabruck, Barbarastraße 7, 49069 Osnabruck, Germany
nanoAnalytics GmbH, Center for Nanotechnology, Heisenbergstraße 11, 48149 Munster, Germany
b
c
€
€
€
€
€
€
a r t i c l e i n f o
a b s t r a c t
Nucleoside hydrogels based on benzyl azide ‘click’ conjugates of 8-aza-7-deaza-20-deoxyadenosine
bearing 7-ethynyl, 7-octa-(1,7-diynyl), and 7-tri-prop-2-ynyl-amine side chains were synthesized (1, 3,
4). The cycloaddition adduct with the shortest linker (1) yields the most powerful hydrogelator forming
stable gels at a concentration of 0.3 wt % of 1 in water. One molecule of 1 catches 7500 water molecules.
Cycloaddition of the 8-aza-7-deaza-7-azido-20-deoxyadenosine (9) and 3-phenyl-1-propyne (10) leads to
the isomeric conjugate 2, with a CeN connectivity between the nucleobase and triazole moiety. This gel
is less stable than that of the adduct 1. Both gels show a similar stability over a wide pH range (4.0e10.0).
Xerogels of 1 and 2 studied by scanning electron microscopy (SEM) reveal that both click adducts (1 and
2) form long fibers spontaneously.
Article history:
Received 9 May 2011
Received in revised form 21 June 2011
Accepted 6 July 2011
Available online 23 July 2011
Keywords:
Modified nucleosides
Hydrogelation
‘Click’ conjugates
Pyrazolo[3,4-d]pyrimidines
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
certain positions and/or linked to residues, which can cause a var-
iation in the hydrophilicehydrophobic forces, gels are formed.
Low molecular weight gelators (LMWGs)1e3 have received at-
tention because of their potential use in tissue engineering,4 tar-
geted drug delivery,5 light harvesting,6 or as nanomaterials.7e9 In
such ‘physical’ gels small subunits are linked non-covalently, which
is different from the more common gels that are based on poly-
meric gelators. The dynamic balance between hydrophilic and hy-
drophobic interactions of gelator molecules plays a defining role in
the aggregation of molecules in a given solvent. The self-assembly
of low molecular weight building blocks, which results from su-
pramolecular entanglements of molecules or chemically cross-
linked species into a network, traps liquid in the network leading
to a self-supporting gel.10e12 Such hydrogels are of great impor-
tance because of their capability to entrap a large number of water
molecules. Hydrogels can form nanotubes or nanofibers, which find
extensive applications in material science, diagnostics, medicine
and supramolecular chemistry.4,5,7e9
Guanine was the first and foremost nucleobase tried and tested for
molecular assemblies. In 1904, Bang31 showed that guanosine and
its analogs, mainly guanylic acid (GMP), form gels in aqueous so-
lution. In 1962, Gellert32 proposed that four guanines are related to
each other by the operation of fourfold rotation axis, leading to
a planar tetramer arrangement. Later in 1971, Guschlbauer33 and
others34e36 showed that, not only guanosine but also several of its
analogs form gels. Further, it was demonstrated that isoguanosine37
can also form gels; our laboratory showed that gels of 20-deoxy-
isoguanosine formed in 0.5 M NaCl are stable.38 This gel formation
is cation dependent. Most of the hydrogel forming nucleosides
were found by serendipity and not by rational design. Nucleosides
can easily bind water molecules as they contain a variety of donor
and acceptor centers.
Over the years, a number of LMW gelators have been syn-
thesized,1e3,22e26 and this concept is extended to nucleoside based
gelators because of their excellent biocompatibility. Click conjugates
of 20-deoxyuridine, prepared by the HuisgeneMeldaleSharpless
cycloaddition were reported to form stable hydrogels.24 Here we
selected the artificial nucleoside 8-aza-7-deaza-20-deoxyadenosine,
which was decorated with various side chains at the 7-position for
gelation studies.39e41 Triazole moieties with a lipophilic residuewere
introduced by click conjugation, and the gelation behavior was
studied in water. Each derivative is an amphiphile featuring multiple
Among the monomeric naturally occurring compounds forming
gels are amino acids,13ꢀ18 sugars,19e21 nucleosides22e26 and other
chemical entities.27e30 Nucleosides themselves seldom offer the
possibility of gelation but when these nucleosides are modified at
* Corresponding author. Tel.: þ49 0 251 53406 500; fax: þ49 0 251 53406 857;
e-mail address: Frank.Seela@uni-osnabrueck.de (F. Seela).
URL: http://www.seela.net/
functionalities derived from
a combination of lipophilic and
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.07.015

F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
7419
hydrophilic units (Fig. 1). While nucleoside conjugates 1 and 2 have
an almost identical structure, compound 3 represents a molecule
with a long linker arm and 4 with a branched side chain decorating
the nucleoside with two proximal benzyl and 1,2,3-triazole moieties.
With those compounds in hand, the relationship of nucleoside
structure with gel formation was studied, xerogels were formed and
analyzed by scanning electron microscopy. Equimolar mixtures of
isomeric nucleosides (1 and 2) behave differently than the pure
components. The influence of base modification will be presented
along with experiments using nucleoside conjugates with related or
complementary nucleobases.
the amino group protons thereby forming a seven-membered ring
system. This assumption is further supported by compound 4,
which does not show a separation of the proton signals (Fig. S14)
due to the distant position of the triazole moiety. These phenomena
might influence the gelation process.
The click conjugation and the nature of the connectivity of tri-
azole system have decisive impact on the pattern of UV/vis spectra.
As shown in Fig. 2, when the triazole ring is connected to the base
through a CeC bond (1), a strong absorption maximum at 254 nm
was observed along with a shoulder at 289 nm. But when the tri-
azole is connected to the base through a CeN bond as in case of 2,
Fig. 1. The benzyltriazole appended nucleosides with variable linker arms and branched side chain used in this study.
2. Results and discussion
the absorption maximum is found at 287 nm with a 265 nm
shoulder. So, the electronic properties of the nucleobase will be
changed by the connectivity of the triazole system. This might
cause changes in the stacking properties of the base moiety.
2.1. Synthesis and characterization of monomers
The nucleosides bearing short (7), long (11),41 and branched (12)
linker arms were prepared by Sonogashira cross coupling reaction
of common iodo precursor 539e41 and respective alkynes as shown
in Scheme 1. At first, the trimethylsilyl compound 6 was obtained
by Sonogashira cross coupling from the iodo precursor 5. Further
deprotection of the silyl group with anhydrous K2CO3 gave the free
nucleoside 7 (85%). Then the [3þ2] cycloaddition click reac-
tion42e45 was performed in THF/H2O/t-BuOH (3:1:1) with benzyl
azide 8 to give the click adduct 1 in 88% yield. Its regioisomer 2 was
synthesized by reacting iodo-precursor 5 with the alkyne 10 under
microwave assistance.46 Nucleoside azide 9 was formed as in-
termediate in situ, which was not isolated. This intermediate sub-
sequently undergoes ‘click’ reaction with 3-phenyl-1-propyne 10,
to yield 2 as the N-regioisomer of 1 in 83% yield. Nucleoside 12
bearing a branched side chain, was prepared in 67% yield, which
was further clicked with benzyl azide as described above to give the
click adduct 4 in 76% yield. The click adduct 3 and its precursor 11
were prepared as reported in literature41 and characterized.
The closely related regioisomers 1 and 2 can be easily identified
by characteristic signals in 1H and 13C NMR spectra. In click adduct
1 the amine protons are found at 8.13 and 9.10 ppm whereas its
regioisomer 2 shows the respective protons at 8.19 and 8.36 ppm. In
13C NMR the C5-carbon for click adduct 1 appears at 97.9 ppm,
2.2. Water gelation of nucleoside conjugates
In the above mentioned click adducts, like in conventional
amphiphiles, the nucleoside conjugates display a polar head group
(sugar moiety), which is responsible for the water solubility and
other more or less hydrophobic units (nucleobase, 1,2,3-triazole
moiety and benzyl residues) that offer aid to aggregation. In all
compounds investigated in this study, the molecules form a glyco-
sylic linkage to a 20-deoxyribofuranosyl moiety, which is directly
connected to the nucleobase; the triazole system is either directly
attached (1 and 2) or linked by branched or non-branched arms
(conjugates 3 and 4). Regarding base modification, we studied 8-
aza-7-deazaadenine derivatives along with conjugates displaying
a 7-deazaadenine base (13) (Scheme 2); the corresponding uridine
compound 1424 (Supplementary data) was used in base pairing
studies and for comparison.
2.3. Stability and rheology of gels
To study the gelation ability of adducts 1e4, they were sus-
pended in a given volume of water, heated to a clear solution and
cooled. Gel formation was established when the gel passes the
tube-inversion test.24 The click adduct 1 forms a robust gel in
a given volume of water with a minimum gelation concentration
(MGC) of 3 mg/mL, thereby demonstrating its excellent gelation
behavior toward water (Fig. 3a). Further increases in the concen-
tration of the click adduct had a further positive effect on gel sta-
bility. This has been tested up to a concentration of 11 mg/mL of
compound 1, which leads to an even more stable gel. Next, the
gelation behavior of the click adducts bearing long (3) and
branched (4) linker arms was studied. Conjugate 3 did not form
a gel at all, neither at low nor at high concentrations (Table 1).
Apparently, the increased hydrophobic character arising from the
whereas it is shifted to 91.7 ppm in its regioisomer
2
(Supplementary data). Furthermore, it was observed in the 1H NMR
spectrum that the protons of the amino groups of compounds 1 and
2 are sharp and clearly separated (Supplementary data, Figs. S8 and
S10) while those of the 7-deazapurine conjugate 13 form a broad
signal (Fig. S16). From that we conclude that one of the NH2 protons
of compound 1 and 2 is involved in hydrogen bonding with a tri-
azole nitrogen as acceptor site. This is a consequence of the higher
acidity of the 6-amino group protons of the 8-aza-7-deazaadenine
conjugates 1 and 2 compared to that of the 7-deazapurine analog
13. An additional factor is the proximity of the triazole ring to one of

7420
F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
Scheme 1. Reagents and conditions: (i) Trimethylsilylacetylene, CuI, [Pd0[P(Ph3)4], Et3N, DMF, rt, 12 h; (ii) anhydrous K2CO3, MeOH, rt, 2 h; (iii) benzyl azide, CuSO4$5H2O,
Na-ascorbate, THF/H2O/t-BuOH, rt, 12 h; (iv) NaN3, L-proline, CuSO4, Na-ascorbate, Na2CO3$10H2O, DMF/H2O (4:1), MW, 30 min, 130 ꢁC; (v) tripropargylamine, CuI, [Pd0[P(Ph3)4],
Et3N, DMF, rt, 12 h; (vi) CuSO4$5H2O, Na-ascorbate, EtOH/H2O (1:1), rt, 12 h; (vii) octa-(1,7)-diyne, CuI, [Pd0[P(Ph3)4], Et3N, DMF, rt, 12 h.
additional methylene groups in the linker units perturbed the
balance between the hydrophilicehydrophobic properties. Then,
the branched nucleoside adduct 4 was tested. With an MGC of
2 mg/mL, conjugate 4 forms a colloidal solution, which turns to
a clear solution in 24 h. Within 2e3 weeks it forms very fine fibers
as shown in Fig. 3d. In contrary, the nucleosides 5, 7, 11, and 12
bearing no clicked side chain did not form a gel.
Next, the gelation ability of compound 2 (Fig. 3b) was compared
with conjugate 1 (Fig. 3a). Even though molecules 1 and 2 are of the
same size and have an almost identical structure, conjugate 2 be-
haves differently with regard to gelation than nucleoside 1. The gel
is less stable at a low nucleoside concentration and an increase of
the nucleoside concentration did not increase the stability signifi-
cantly, while conjugate 1 became more stable at higher nucleoside
concentration. This indicates that the triazole ring connectivity
formed during click reaction plays a decisive role in the gelation
process. Then, the gelation ability of conjugate 1 in the presence of
conjugate 2 in an equimolar mixture was tested (Fig. 3c). As in-
dicated in Table 1, the gelation ability of adducts 1 and 2 was found
to be existing in the mixture of both conjugates. However,
while individual conjugates formed stable gels at a concentration of
3 mg/mL, very unstable gels were formed by the mixture, which did
not pass the ‘tube-inversion’ test. The hydrogel behavior in the
tube-inversion test is demonstrated in Fig. 3aec. A Tgel of 61 ꢁC was
determined for compound 1 (c¼3 mg/mL) by a dropping ball ex-
periment as described in literature.28
As it is known that hydrogen bonding plays a vital role in the
balance of counteracting forcesdhydrophilic hydrophobic force-
sdwe wanted to study whether base pairing influences the gel
formation.22,35 For this, the click adduct of benzyl azide and

F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
7421
Table 1
Minimum gelation concentrations (MGCs) of nucleoside click conjugates forming
hydrogelsa
Compd
MGC (mg/mL)
Gel
Compd
MGC (mg/mL)
Gel
1
0.1
0.2
0.3e0.5
0.1
0.2
u
u
s
u
u
s
1þ2
1þ4
2þ4
3
0.1e0.5
0.1e0.5
0.3
0.1e0.5
0.1e0.3
u
n
n
n
f
2
4
0.3e0.5
a
s: stable gel; u: unstable gel; n: no gel. f: fiber. MGCs provided in parentheses
(wt %).
conjugates 1 and 2 was studied. Under all those pH values, no
changes in gel stability were observed in case of ‘adenine’ click
conjugate 1. So, there is no pH response within a pH range of 4e10
in the case of gel 1 (pKa value of 1 is 3.89). But when the pH de-
pendent gelation property of the already known ‘uridine’ adduct
1424 was compared, destabilization of 14 occurred at higher pH
values (pH¼10) due to the deprotonation of the nucleobase (pKa
value of 14 is 8.74; see Supplementary data, Fig. S2).
Fig. 2. UV profiles of the click nucleoside conjugates 1 and 2. The measurements were
performed in methanol (0.98 M).
m
The transition phenomenon of viscous fluid (sol) to an elastic
solid (gel), which comprises a 3D-supramolecular network is re-
ferred as gelation. In order to determine the viscoelastic properties
of gels, rheological experiments were performed. The storage
modulus and loss modulus were measured and are expressed as G0
and G00, respectively. For a gel, the elastic modulus G0 must be rel-
atively independent of frequency of deformation, and G0 must be
greater than G00.16 As it is demonstrated in Fig. 4, these two char-
acteristics are observed where G0 and G00 were depicted as a func-
tion of frequency.
In Fig. 4aec, the storage modulus G0 is always higher than the
loss modulus G00 for both, compound 1 and its N-isomer compound
2, and even for the mixture of compounds 1 and 2, when measured
at 25 ꢁC. As the temperature was increased, the difference between
G0 and G00 decreases and eventually both merges together in case of
compound 1 (see Supplementary data) depicting that at higher
temperature the transition from gel to sol occurs leading to a de-
struction of the gel. But the situation is somewhat different in case
of the N-isomer 2. The crossover between G0 and G00 exists at 50 Hz
at 25 ꢁC, and an additional crossover point appears around 10 Hz at
75 ꢁC (see Supplementary data). This demonstrates that the gel is
less stable at such frequencies. On the contrary, the mixture of
compounds 1 and 2 does not show the crossover point at 25 ꢁC
(Fig. 4c), but the magnitude of both moduli is lower than for the
individual compounds. Taken together, the gel formed by
Scheme 2.
5-ethynyl-20-deoxyuridine (14) was synthesized as reported (see
Supplementary data), and its gelation ability was already con-
firmed.24 Then, equimolar amounts of compound 1 and the com-
plementary uridine adduct 14 (both dissolved in water) were tested
for gelation. The mixture of 1 and 14 resulted into a clear solution,
whereas these individual conjugates form stable gels. The results of
all gel formation studies are summarized in Table 1.
Changes of pH of the gelating systems have been found to alter
the properties of gels.47e49 This prompted us to investigate the
hydrogelation of the nucleoside conjugates at different pH values.
Slightly acidic (pH 4.0 and pH 5.0) as well as alkaline conditions (pH
9.0 and 10.0) were chosen, and gel formation of the nucleoside
Fig. 3. Tube-inversion test depicting gelation: Gelation test for (a) click adduct 1. (b) Click adduct 2. (c) Equimolar mixture of 1 and 2. (d) Image of branched adduct 4 depicting fiber
formation rather than gelation.

7422
F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
Fig. 4. Evolution of G0 and G00 as a function of the frequency sweep. Measurement of compound (a) 1 at 25 ꢁC; (b) 2 at 25 ꢁC and (c) 1þ2 at 25 ꢁC.
compound 1 is much stronger than the gel of compound 2 as well as
its mixture with 1.
stable hydrogel. Experiments with closely related molecules
reveal that the click conjugates respond sensitively to changes
on the modification of the triazole moiety and the nucleobase.
The short linker armed click adducts such as 1 with a C-linked
triazole moiety efficiently gelates in water whereas that with
the isomeric N-linked triazole (2) forms gels of lower stability.
Long or branched linkers bearing two triazole moieties within
one molecule do not lead to gels at all. So, it seems to be
important that the triazole moiety is directly linked to the
nucleobase.
Apart from structural changes in the triazole moiety the effect
of nucleobase structure on gel formation was studied. For this,
a shape mimic of compound 1 was used. Replacement of the
pyrazolo[3,4-d]pyrimidine skeleton by a pyrrolo[2,3-d]pyrimi-
dine moiety with one ring nitrogen less resulted in the adduct 13,
which was synthesized and tested for gelation. This 7-
deazaadenine click conjugate 13, does not form a gel at any
given concentration. In order to detect conformational changes
on both click conjugates, molecular dynamics simulations were
performed using Hyperchem 8.0 in the absence of water mole-
cules. The results for compounds 1, 2, 3, and 13 are shown in
Fig. 6. From this, it can be concluded that compounds 1e3 lock
rather similar. Nevertheless, compounds 1 and 2 form a gel while
compound 13 does not. As expected the overall structure of
conjugate 3 with the long linker looks rather different. It stays to
prove why so significant changes in the gelation properties are
observed for compounds of so similar molecules (1 and 13).
Earlier, protecting group experiments performed on amino
groups demonstrated that the amide character of the 6-amino
group is significantly higher for the 8-aza-7-deazapurines 1 and
2 than for the 7-deazapurine nucleoside 13. Consequently, the
amino groups of 1 and 2 are better proton donors and can par-
ticipate in hydrogen bonding. The more hydrophobic character of
2.4. Fiber formation of xerogel click conjugates 1 or 2
To get an insight into the microstructures, scanning electron
microscopy (SEM) was performed on xerogels (20 days, room
temperature dried gels). The figures shown are typical SEM pic-
tures of compounds 1 or 2 at different magnification. Fig. 5a
displays the picture with the magnification (9000ꢂ) of the con-
jugate 1. Careful inspection of Fig. 5a indicates that the fibers are
formed spontaneously during gel formation. The fibers of 1 are
hollow with small channels inside (nanotubes). These channels
can be seen very clearly from high resolution magnification pic-
tures (Fig. 5b). Lower magnification pictures confirm that these
fibers are inter-tangling with each other forming supramolecular
networks. Compound 2 (Fig. 5def) forms also fibers. However,
these fibers appear to be solid rather than hollow. For compari-
son, SEM measurements of 2’-deoxyuridine click conjugate 14
were performed (see Supplementary data, Fig. S1).24 The struc-
ture of the molecular assembly of compounds 1 and 2 within the
fibers is unknown. Nevertheless, hydrogen bonding as well as
stacking interactions are anticipated to be the main forces hold-
ing the molecules together.
2.5. Influence of nucleobase structure, triazole connectivity
and linker units on gel assembly and molecular modeling
Molecular forces hold the monomeric gelator molecules to-
gether thereby forming
a supramolecular assembly. These
forces are hydrophobic interactions, stacking interactions, and
also hydrogen bonding. This is clearly evidenced by the in-
fluence of structural changes of compound 1 forming the most

F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
7423
Fig. 5. SEM images of compound 1 with the scale bars 20
mm of (a) compound 1 (b) highlighting the channels of fibers encircled in the image of compound 1. (c) Compound 1 with
different magnification. (d)e(f) Compound 2 with the scale bar 20
mm with different magnifications.
the 7-deazaadenine base of 13 together with a lower amino
group acidity as well as weaker stacking interactions may ac-
count for this behavior.
3. Conclusion
We have demonstrated the ability of 8-aza-7-deazaadenosine
‘click’ conjugate 1 to form a stable gel at a minimum gel concen-
tration of 3 mg in 1 mL of water. Calculations on individual mole-
cules reveal that one molecule of 1 can catch 7500 water molecules.
The hydrogel of the closely related 2 is less stable; both form
nanofibers. Equimolar mixtures of 1 and 2 and those containing
complementary nucleobases (1) and thymine click conjugate 14
behave differently as the pure conjugates and do not form stable
gels in mixtures. The pyrrolo[2,3-d]pyrimidine shape mimic of 1,
with one nitrogen atom less in the ‘purine’ moiety, does not form
a gel at any given concentration. The rheology of the gel formed by
conjugate 1 is entirely different from the rheology of its closely
related conjugate 2, which displays crossover points signifying the
gel formed by conjugate 2 is much weaker than that of conjugate 1.
Moreover, the regioisomeric mixture of conjugate 1 and 2 forms
a very unstable gel, which cannot pass the tube-inversion test. Both,
storage and loss moduli of the mixture are much lesser than for the
individual compounds. The click adducts 1 and 2 exhibit good gel
Fig. 6. Molecular dynamics (MD) simulation models of conjugate 1, conjugate 2,
conjugate 3 and conjugate 13. The molecules were constructed using Hyperchem 8.0
and energy minimized using OPLS force field calculations.

7424
F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
stability over a wide range of pH, making them promising candi-
dates for various purposes, such as targeted drug delivery, wound
healing50 or material science.51
C50eOH), 5.27e5.28 (1H, d, J¼4.8 Hz, C30eOH), 6.52e6.56 (1H, m,
C10eH), 8.24 (1H, s, C2eH). Anal. Calcd for C12H13N5O3 (275.26): C,
52.36; H, 4.76; N, 25.44. Found: C, 52.33; H, 4.69; N, 25.31.
4. Experimental section
4.2.3. 1-[2-Deoxy-
triazol-40-yl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (1). To a solu-
tion of 7 (0.1 g, 0.36 mmol) and benzyl azide (0.058 g, 54 L,
0.43 mmol) in THF/H2O/t-BuOH (3:1:1, 5 mL) was added sodium
ascorbate (362 L, 0.36 mmol) of a freshly prepared 1 M solution
in water, followed by the addition of copper(II) sulfate penta-
hydrate 7.5% in water (312 L, 0.08 mmol). The emulsion was
b-D
-erythro-pentofuranosyl]-3-[(1-benzyl-10,20,30-
4.1. General materials and methods
m
All chemicals were purchased from Acros, Fluka or SigmaeAldrich
(SigmaeAldrich Chemie GmbH, Deisenhofen, Germany). Solvents
were of laboratory grade. Thin-layer chromatography (TLC) was
performed on TLC aluminum sheets covered with silica gel 60 F254
(0.2 mm). Flash column chromatography (FC): silica gel 60 (VWR,
Germany) at 0.4 bar; UV spectra: U-3200 spectrometer (Hitachi,
Tokyo, Japan); NMR spectra: Avance-DPX-300 spectrometer (Bruker,
Rheinstetten, Germany), at 300 MHz for 1H and 75.48 MHz for 13C;
m
m
stirred for 12 h at room temperature and the solution was
evaporated and the residue was applied to FC (silica gel, column
10ꢂ3 cm, CH2Cl2/MeOH 90/10). From the main zone, com-
pound 1 (0.075 g, 88%) was isolated as a colorless solid. TLC
(silica gel, CH2Cl2/MeOH 9:1). Rf 0.5. UV lmax (MeOH)/nm 254
d
in parts per million relative to Me4Si as internal standard. All the
(
/dm3 molꢀ1 cmꢀ1 14,900), 289 (10,600). 1H NMR [DMSO-d6,
3
synthesized compounds were characterized by elemental analyses,
300 MHz]: d
2.24e2.32 (1H, m, C20-Ha), 2.79e2.88 (1H, m, C20-
1H and 13C NMR spectra. Elemental analyses were performed by
Hb), 3.38e3.46 (1H, m, C50eH), 3.53e3.64 (1H, m, C50eH),
3.82e3.87 (1H, m, C40eH), 4.48e4.49 (1H, d, J¼4.2 Hz, C30eH),
4.79e4.83 (1H, t, J¼5.7 Hz, C50eOH), 5.28e5.29 (1H, d, J¼4.5 Hz,
C30eOH), 5.73 (2H, s, CH2), 6.58e6.62 (1H, t, J¼6.3 Hz, C10eH),
7.33e7.41 (5H, m, AreH), 8.13 (1H, br s, NHa), 8.24 (1H, s, tri-
azoleeH), 8.83 (1H, s, C2eH), 9.10 (1H, br s, NHb). Anal. Calcd for
C19H20N8O3 (404.81): C, 55.88; H, 4.94; N, 27.44. Found: C, 55.98;
H, 4.85; N, 27.35.
€
Mikroanalytisches Laboratorium Beller (Gottingen, Germany).
4.2. General procedure for gelation
The click adduct (3 mg) was suspended in 1 mL of water in
a glass vial. The mixture was heated until it became a clear solution.
After complete dissolution, the solution was gradually allowed to
cool to room temperature. Then after the sample was subjected to
a ‘tube-inversion test’.24 The compound is said to be a gelator, when
the sample passes tube-inversion test and was characterized as
a gel. When the formed gel flows neither freely like clear solution
nor self-supporting, it is called partial gel.
4.2.4. 4-Amino-7-(2-deoxy-b-D-erythro-pentofuranosyl)-5-[1-
benzyl-10,20,30-triazol-40-yl]-7H-pyrrolo-[2,3-d]pyrimidine
(13). Procedure as described above for compound 1. 4-Amino-7-(2-
deoxy-
b
-
D
-erythro-pentofuranosyl)-5-(ethynyl)-7H-pyrrolo[2,3-d]
pyrimidine52 12a, (0.82 g, 0.3 mmol), benzyl azide (78
mL,
4.2.1. 1-(2-Deoxy-
b
-
D
-erythro-pentofuranosyl)-3-[(trimethylsilanyl)
0.36 mmol), sodium ascorbate (0.238 g, 1.2 mmol), and copper(II)
sulfate pentahydrate (0.075 g, 0.3 mmol). EtOH/H2O (1:1, 5 mL).
Compound 13 was isolated as a colorless solid. Yield (0.067 g, 55%).
TLC (silica gel, CH2Cl2/MeOH 9:1). Rf 0.7. UV lmax (MeOH)/nm 245
ethynyl]-1H-pyrazolo[3,4-d] pyrimidin-4-amine (6). To a solution of
compound 539,40 (0.95 g, 2.51 mmol) in dry DMF (25 mL), CuI
(0.096 g, 0.50 mmol), Pd(PPh3)4 (0.29 g, 0.25 mmol), dry Et3N
(0.51 g, 0.7 mL, 5.05 mmol), and 6 equiv of trimethylsilylacetylene
(1.48 g, 2.14 mL, 15.1 mmol) were added. The reaction mixture,
which slowly turned to black was stirred under inert atmosphere
for 12 h (TLC monitoring). The reaction mixture was evaporated and
the oily residue was adsorbed on silica gel and subjected to FC
(silica gel, column 15ꢂ3 cm, eluted with CH2Cl2/MeOH
95:5/90:10) affording one main zone. Evaporation of the solvent
gave 6 (0.7 g, 80%) as a colorless solid. TLC (silica gel, CH2Cl2/MeOH
(
/dm3 molꢀ1 cmꢀ1 14,400), 275 (10,800). 1H NMR [DMSO-d6,
3
300 MHz]:
d
2.15e2.24 (1H, m, C20-Ha), 2.41e2.45 (1H, m, C20-Hb),
3.45e3.63 (2H, m, C50eH), 3.79e3.83 (1H, m, C40eH), 4.33e4.35
(1H, t, J¼6.0 Hz, C30eH), 5.02 (1H, br s, C50eOH), 5.28 (1H, br s,
C30eOH), 5.68 (2H, s, CH2), 6.51e6.55 (1H, dd, J¼6.2 Hz, C10eH),
7.33e7.44 (5H, m, AreH), 7.86 (1H, s, C8eH), 8.08 (1H, s, C2eH),
8.511 (1H, s, triazoleeH).
9:1). Rf 0.6. UV lmax (MeOH)/nm 246 (
3
4.2.5. 1-[2-Deoxy-
triazol-10-yl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (2). Compound
5 (0.050 g, 0.13 mmol), 3-phenyl-1-propyne 10 (0.026 g, 27 L,
0.22 mmol), NaN3 (0.013 g, 0.20 mmol), -proline (0.004 g,
b-D
-erythro-pentofuranosyl]-3-[(4-benzyl-10,20,30-
285 (12,800). 1H NMR [DMSO-d6, 300 MHz]:
d
2.20e2.28 (1H, m, C20-Ha), 2.72e2.80 (1H, m, C20-Hb), 3.48e3.54
(1H, m, C50eH), 3.78e3.83 (1H, m, C40eH), 4.38e4.43 (1H, m,
C30eH), 4.76e4.80 (1H, t, J¼5.7 Hz, C50eOH), 5.29e5.31 (1H, d,
J¼4.5 Hz, C30eOH), 6.52e6.56 (1H, t, J¼6.3 Hz, C10eH), 8.25 (1H, s,
C2eH). Anal. Calcd for C15H21N5O3Si (347.44): C, 51.85; H, 6.09; N,
20.16. Found: C, 51.86; H, 5.96; N, 20.01.
m
L
0.03 mmol, 20 mol %), Na2CO3$10H2O (0.012 g, 0.04 mmol,
30 mol %), and sodium ascorbate (0.014 g, 0.07 mmol) were sus-
pended in a DMF/H2O (4:1, 2.5 mL) mixture in a 2e5 mL microwave
vessel. To the stirred mixture, CuSO4 (20 mol %, as a 1 M solution in
H2O) was added and the vessel was sealed and subjected to mi-
crowave irradiation at 130 ꢁC for 30 min (ramp time: 45 s, pre-
stirring: 20 s, high). After completion of the reaction (monitored
by TLC), the reaction mixture was cooled to room temperature. The
solvent was evaporated to dryness and subjected to FC (silica gel,
column 10ꢂ3 cm, CH2Cl2/MeOH 90/10). Isolation of main zone
afforded 2 (0.045 g, 83%) as a colorless solid. TLC (silica gel, CH2Cl2/
4.2.2. 1-(2-Deoxy-b-D-erythro-pentofuranosyl)-3-ethynyl-1H-pyr-
azolo[3,4-d]pyrimidin-4-amine (7). To the solution of compound 6
(0.06 g, 0.17 mmol) in methanol (6 mL), anhydrous K2CO3 (0.006 g,
0.04 mmol) was added and stirred at room temperature for 2 h.
After completion of the reaction (monitored by TLC), solvent was
evaporated and the residue was subjected to FC (silica gel, column
15ꢂ3 cm, eluted with CH2Cl2/MeOH 94:6/90:10). Evaporation of
the solvent afforded compound 7 (0.04 g, 85%) as a colorless solid.
TLC (silica gel, CH2Cl2/MeOH 9:1). Rf 0.4. UV lmax (MeOH)/nm 246
MeOH 9:1). Rf 0.5. UV lmax (MeOH)/nm 245 (
3
7800), 286 (13,800). 1H NMR [DMSO-d6, 300 MHz]:
d
(1H, m, C20-Ha), 2.77e2.85 (1H, m, C20-Hb), 3.40e3.43 (1H, m,
C50eH), 3.50e3.56 (1H, m, C50eH), 3.80e3.85 (1H, m, C40eH), 4.16
(2H, s, CH2), 4.47 (1H, br s, C30eH), 4.73e4.77 (1H, t, J¼5.7 Hz,
C50eOH), 5.29e5.31 (1H, d, J¼4.2 Hz, C30eOH), 6.61e6.65 (1H, t,
J¼6.3 Hz, C10eH), 7.20e7.25 (1H, m, AreH), 7.29e7.36 (4H, m,
(
/dm3 molꢀ1 cmꢀ1 10,400), 285 (12,900). 1H NMR [DMSO-d6,
3
300 MHz]:
d
2.21e2.29 (1H, m, C20-Ha), 2.72e2.80 (1H, m, C20-Hb),
3.37e3.54 (2H, m, C50eH), 3.80e3.81 (1H, m, C40eH), 4.40e4.45
(1H, m, C^CH), 4.68 (1H, m, C30eH), 4.74e4.78 (1H, t, J¼5.7 Hz,

F. Seela et al. / Tetrahedron 67 (2011) 7418e7425
7425
AreH), 8.20 (1H, br s, NHa), 8.30 (1H, s, triazoleeH), 8.36 (1H, br s,
References and notes
NHb), 8.72 (1H, s, C2eH). Anal. Calcd for C19H20N8O3 (404.81): C,
55.88; H, 4.94; N, 27.44. Found: C, 56.05; H, 4.86; N, 27.20.
1. Terech, P.; Weiss, R. G. Chem. Rev. 1997, 97, 3133e3159.
2. Estroff, L. A.; Hamilton, A. D. Chem. Rev. 2004, 104, 1201e1218 and references
cited therein.
4.2.6. 1-(2-Deoxy-
b
-D-erythro-pentofuranosyl)-3-[di(prop-2-ynl)
3. George, M.; Weiss, R. G. Acc. Chem. Res. 2006, 39, 489e497 and references cited
therein.
4. Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869e1879 and references cited
therein.
5. Langer, R. Acc. Chem. Res. 2000, 33, 94e101.
6. Ajayaghosh, A.; Praveen, V. K.; Vijayakumar, C. Chem. Soc. Rev. 2008, 37,
109e122 and references cited therein.
7. Gao, P.; Zhan, C.; Liu, L.; Zhou, Y.; Liu, M. Chem. Commun. 2004, 1174e1175.
8. Xu, B. Langmuir 2009, 25, 8375e8377.
9. Sahoo, P.; Kumar, D. K.; Trivedi, D. R.; Dastidar, P. Tetrahedron Lett. 2008, 49,
3052e3055.
10. Flory, P. J. Faraday Discuss. 1974, 57, 7e18.
amino]prop-1-ynyl]-1H-pyrazolo[3,4-d] pyrimidin-4-amine (12). A
solution of compound 5 (0.1 g, 0.26 mmol) in dry DMF (5 mL), was
treated with CuI (0.01 g, 0.05 mmol), Pd(PPh3)4 (0.031 g,
0.03 mmol), dry Et3N (0.053 g, 73
tri(prop-2-ynyl)amine (0.35 g, 480
mixture, which slowly turns to black was stirred under inert at-
mosphere for 12 h. After the completion of the reaction (TLC
monitoring), the reaction mixture was evaporated and the oily
residue was adsorbed on silica gel and subjected to FC (silica gel,
column 15ꢂ3 cm, eluted with 98:2/94:6) affording one main
zone. Evaporation of the solvent gave 12 (0.065 g, 67%) as a color-
less foam. TLC (silica gel, CH2Cl2/MeOH 9:1). Rf 0.5. UV lmax
mL, 0.52 mmol) and 6 equiv of
mL, 2.6 mmol). The reaction
11. Wang, R.; Geiger, C.; Chen, L.; Swanson, B.; Whitten, D. G. J. Am. Chem. Soc. 2000,
122, 2399e2400.
12. Sakurai, K.; Jeong, Y.; Koumoto, K.; Friggeri, A.; Gronwald, O.; Sakurai, S.;
Okamoto, S.; Inoue, K.; Shinkai, S. Langmuir 2003, 19, 8211e8217.
13. Bhuniya, S.; Park, S. M.; Kim, B. H. Org. Lett. 2005, 7, 1741e1744.
14. Suzuki, M.; Owa, S.; Yumoto, M.; Kimura, M.; Shirai, H.; Hanabusa, K. Tetrahe-
dron Lett. 2004, 45, 5399e5402.
15. Ryan, D. M.; Anderson, S. B.; Nilsson, B. L. Soft Matter 2010, 6, 3220e3231.
16. Dutta, S.; Shome, A.; Debnath, S.; Das, P. K. Soft Matter 2009, 5, 1607e1620.
17. Karbarz, M.; Romanski, J.; Michniewicz, K.; Jurczak, J.; Stojek, Z. Soft Matter
2010, 6, 1336e1342.
(MeOH)/nm 248 (
3
[DMSO-d6, 300 MHz]:
d
m, C20-Hb), 3.27 (2H, s, CH2), 3.48e3.53 (4H, m, CH2), 3.75 (2H, s,
C^CH), 3.78e3.83 (1H, m, C40eH), 4.40e4.43 (1H, m, C30eH),
4.78e4.82 (1H, t, J¼5.7 Hz, C50eOH), 5.29e5.30 (1H, d, J¼4.5 Hz,
C30eOH), 6.52e6.56 (1H, t, J¼6.3 Hz, C10eH), 8.24 (1H, s, C2eH).
Anal. Calcd for C19H20N6O3 (380.4): C, 59.99; H, 5.30; N, 22.09.
Found: C, 59.94; H, 5.24; N, 21.96.
18. de Loos, M.; van Esch, J. H.; Kellogg, R. M.; Feringa, B. L. Tetrahedron 2007, 63,
7285e7301.
19. Srivastava, A.; Ghorai, S.; Bhattacharjya, A.; Bhattacharya, S. J. Org. Chem. 2005,
70, 6574e6582.
20. Jung, J. H.; John, G.; Masuda, M.; Yoshida, K.; Shinkai, S.; Shimizu, T. Langmuir
2001, 17, 7229e7232.
4.2.7. 1-[2-Deoxy-b-D-erythro-pentofuranosyl]3-{3-[bis(1-benzyl-
21. Amanokura, N.; Kanekiyo, Y.; Shinkai, S.; Reinhoudt, D. N. J. Chem. Soc., Perkin
10,20,30-triazol-4-ylmethyl)amino]prop-1-ynyl}-1H-pyrazolo[3,4-d]
pyrimidin-4-amine (4). Procedure as described above for com-
pound 1. Compound 12 (0.1 g, 0.27 mmol), benzyl azide (0.084 g,
Trans. 2 1999, 1995e2000.
22. Araki, K.; Yoshikawa, I. Top. Curr. Chem. 2005, 256, 133e165 and references cited
therein.
23. Park, S. M.; Lee, Y. S.; Kim, B. H. Chem. Commun. 2003, 2912e2913.
24. Park, S. M.; Shen, Y.; Kim, B. H. Org. Biomol. Chem. 2007, 5, 610e612.
78
mL, 0.63 mmol), sodium ascorbate (262
m
L, 0.26 mmol), and
L, 0.06 mmol).
ꢀ ꢀ
25. Godeau, G.; Barthelemy, P. Langmuir 2009, 25, 8447e8450.
copper(II) sulfate pentahydrate 7.5% in water (226
EtOH/H2O (1:1, 5 mL). Compound 4 was isolated as a colorless solid.
Yield (0.133 g, 76%). TLC (silica gel, CH2Cl2/MeOH 9:1). Rf 0.6. UV
lmax (MeOH)/nm 255 (
NMR [DMSO-d6, 300 MHz]:
m
26. Yoshikawa, I.; Yanagi, S.; Yamaji, Y.; Araki, K. Tetrahedron 2007, 63,
7474e7481.
27. de Loos, M.; Friggeri, A.; van Esch, J.; Kellogg, R. M.; Feringa, B. L. Org. Biomol.
Chem. 2005, 3, 1631e1639.
28. Brinksma, J.; Feringa, B. L.; Kellogg, R. M.; Vreeker, R.; van Esch, J. Langmuir
2000, 16, 9249e9255.
/dm3 molꢀ1 cmꢀ1 32,400), 291 (24,900). 1H
d
2.21e2.29 (1H, m, Ha), 2.76e2.85 (1H,
m, C20-Hb), 3.49e3.57 (2H, m, C50eH), 3.81 (5H, s, C40-H, 2x CH2),
4.42e4.45 (1H, m, C30eH), 4.80e4.84 (1H, t, J¼5.7 Hz, C50eOH),
5.29e5.31 (1H, d, J¼4.5 Hz, C30eOH), 5.59 (4H, s, CH2), 6.54e6.58
(1H, t, J¼6.3 Hz, C10eH), 7.28e7.39 (10H, m, AreH), 8.13 (2H, s,
triazoleeH), 8.25 (1H, s, C2eH). Anal. Calcd for C33H34N12O3
(646.7): C, 61.29; H, 5.30; N, 25.99. Found: C, 61.57; H, 5.18; N, 25.85.
29. Ribot, J. C.; Guerrero-Sanchez, C.; Hoogenboom, R.; Schubert, U. S. Chem.
Commun. 2010, 6971e6973.
30. Meister, A.; Drescher, S.; Karlsson, G.; Hause, G.; Baumeister, U.; Hempel, G.;
Garamus, V. M.; Dobner, B.; Blume, A. Soft Matter 2010, 6, 1317e1324.
31. Bang, I. Beitr. Chem. Physiol. Path. 1904, 5, 317e320.
32. Gellert, M.; Lipsett, M. N.; Davies, D. R. PNAS 1962, 48, 2013e2018.
33. Chantot, J. F.; Sarocchi, M.-T.; Guschlbauer, W. Biochemie 1971, 53, 347e354.
34. Davis, J. T. Angew. Chem., Int. Ed. 2004, 43, 668e698 and references cited
therein.
35. Lena, S.; Masiero, S.; Pieraccini, S.; Spada, G. P. Chem.dEur. J. 2009, 15,
7792e7806.
Acknowledgements
36. Sreenivasachary, N.; Lehn, J.-M. PNAS 2005, 102, 5938e5943.
37. Ravindranathan, R. V.; Miles, H. T. Biochim. Biophys. Acta 1965, 94, 603e606.
38. Seela, F.; Jiang, D. 2010, personal communication.
39. Seela, F.; Zulauf, M. J. Chem. Soc., Perkin Trans. 1 1999, 479e488.
40. Seela, F.; Zulauf, M. J. Chem. Soc., Perkin Trans. 1 1998, 3233e3239.
41. Seela, F.; Pujari, S. S. Bioconjugate Chem. 2010, 21, 1629e1641.
42. Huisgen, R. Pure Appl. Chem. 1989, 61, 613e628.
We are especially grateful to P. Ding for the synthesis, charac-
terization and supply of the 7-deazapurine click conjugate 13. We
would like to thank Dr. S. Budow and Dr. P. Leonard for helpful
discussions and support. We also thank S.A. Ingale for reading of
the manuscript. Financial support by ChemBiotech, Muenster,
Germany, is highly appreciated.
€
43. Huisgen, R.; Szeimies, G.; Mobius, L. Chem. Ber. 1967, 100, 2494e2507.
44. Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057e3064.
45. Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int. Ed.
2002, 41, 2596e2599.
Supplementary data
ꢀ
46. Klein, M.; Diner, P.; Dorin-Semblat, D.; Doerig, C.; Grøtli, M. Org. Biomol. Chem.
2009, 7, 3421e3429.
13C NMR chemical shifts and coupling constants of the 8-aza-7-
deaza-20-deoxyadenosine derivatives. SEM images. pKa profiles of
click conjugates. Rheological measurements. 1H NMR and 13C NMR
spectra of new compounds (Figs. S4eS17). Supplementary data
associated with this article can be found in online version at
doi:10.1016/j.tet.2011.07.015.
47. Jeong, J. H.; Schmidt, J. J.; Cha, C.; Kong, H. Soft Matter 2010, 6, 3930e3938.
48. Roy, B.; Saha, A.; Esterrani, A.; Nandi, A. K. Soft Matter 2010, 6, 3337e3345.
49. Pal, A.; Shrivastava, S.; Dey, J. Chem. Commun. 2009, 6997e6999.
50. Grinstaff, M. W. Biomaterials 2007, 28, 5205e5214.
51. Mortisen, D.; Peroglio, M.; Alini, M.; Eglin, D. Biomacromolecules 2010, 11,
1261e1272.
52. Seela, F.; Zulauf, M. Synthesis 1996, 726e730.