Solution-phase synthesis of branched DNA hybrids via H -phosphonate dimers

Article abstract of DOI:10.1021/jo202508n

A method for the solution-phase synthesis of branched oligonucleotides with tetrahedral or pseudo-octahedral geometry is described that involves the coupling of 3′-H-phosphonates of protected dinucleoside phosphates and organic core molecules. The dimer building blocks are produced by a synthesis that requires no chromatographic purification and that produces the dimer H-phosphonates in up to 44% yield in less than three days of laboratory work. A total of seven different branched hybrids were prepared, including a new hybrid of the sequence (CG)₄TBA, where TBA stands for tetrakis(p- hydroxybiphenyl)adamantane that assembles into a material from micromolar aqueous solution upon addition of MgCl₂.

Full text of DOI:10.1021/jo202508n

Article
pubs.acs.org/joc
Solution-Phase Synthesis of Branched DNA Hybrids via
H-Phosphonate Dimers
Arunoday Singh,‡ Mariyan Tolev,‡ Christine I. Schilling,§ Stefan Brase,§ Helmut Griesser,‡
̈
and Clemens Richert*,‡
‡Institute for Organic Chemistry, University of Stuttgart, 70569 Stuttgart, Germany
§DFG-Center for Functionalized Nanostructures (CFN) and Institute of Organic Chemistry, Karlsruhe Institute of Technology,
76131 Karlsruhe, Germany
S
* Supporting Information
ABSTRACT: A method for the solution-phase synthesis of branched oligonucleotides with tetrahedral or pseudo-octahedral
geometry is described that involves the coupling of 3′-H-phosphonates of protected dinucleoside phosphates and organic core
molecules. The dimer building blocks are produced by a synthesis that requires no chromatographic purification and that
produces the dimer H-phosphonates in up to 44% yield in less than three days of laboratory work. A total of seven different
branched hybrids were prepared, including a new hybrid of the sequence (CG)4TBA, where TBA stands for tetrakis-
(p-hydroxybiphenyl)adamantane that assembles into a material from micromolar aqueous solution upon addition of MgCl2.
diesters toward basic aqueous conditions.12 However, the main
reason why H-phosphonates have fallen from favor is probably
INTRODUCTION
The first use of an H-phosphonate for the assembly of
■
internucleotide bonds dates back to the 1950s.1 Todd and
the availability of extremely efficient automated DNA synthe-
sizers for chain assembly using phosphoramidite chemistry that,
colleagues used a 5′-H-phosphonate of 2′,3′-isopropylideneur-
idine as a building block and diphenyl phosphorochloridate as
activating agent for the solution-phase synthesis of a dinucleo-
once established, have outcompeted other methods.
We became interested in the solution-phase synthesis of
branched DNA chains with organic branching elements, or
side phosphate. Efficient solid-phase syntheses of oligonucleo-
“cores”, as part of a program to generate nanoporous DNA-
tides using H-phosphonate building blocks were reported in
the 1980s,2,3 producing chains of more than 100 nucleotides in
based materials. For rigid cores, DNA chains as short as dimers
suffice to drive assembly processes that lead to materials.13
A
favorable cases. Difficult coupling reactions have been per-
formed repeatedly with H-phosphonate monomers,4 including
solid-phase5 and solution-phase syntheses of oligonucleotides
with phosphorothioate linkages.6 New protocols for solid-phase
chain assembly have been published since the breakthrough
methodology of Matteucci and colleagues in 1986,7,8 and the
usefulness of H-phosphonates for the preparation of nucleoside
triphosphates has been demonstrated.9 However, compared to
phosphoramidite-based solid-phase synthesis,10 the H-phosphonate
method of chain assembly has found few followers in the field of
nucleic acid chemistry.
convergent solution-phase synthesis of such hybrids requires
the preparation and handling of dimer building blocks. Since
H-phosphonates are more stable toward hydrolysis than the
corresponding phosphoramidites and thus easier to handle, we
decided to study H-phosphonates of dimers, using an adapta-
tion of the recently published procedure of Jones et al.14 for the
synthesis of cyclic diguanosine monophosphate (c-di-GMP).
The methodology uses inexpensive, commercially available
phosphoramidites of nucleosides. It treats H-phosphonates both
as unreactive, latent phosphates and as readily activable forms
of electrophiles that undergo coupling with alcohols. Here, we
One reason for this lack of attention may be the potential
for side reactions during coupling, such as the formation of
branched chains and symmetrical phosphite anhydrides.11
Another may be the hydrolytic lability of the H-phosphonate
Received: December 8, 2011
Published: February 27, 2012
© 2012 American Chemical Society
2718
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
Figure 1. Target molecules of this study.
demonstrate that this approach produces branched hybrids with
short sticky ends in convenient solution-phase syntheses that
involve coupling of H-phosphonates dimers to aliphatic or
aromatic alcohols. Our study also included the synthesis of a
new branched hybrid with trimer nucleoside arms. The method-
ology presented here is comparable in its yields to that using
dimer phosphoramidites, presented in the accompanying publi-
cation,15 but the starting materials are more readily available and
the procedure is less time-consuming. The molecules prepared
are distantly related to branched oligonucleotides16 with two
DNA chains attached to a branching element and more closely
related to oligonucleotide dendrimers.17 Both branched oligonu-
cleotides and oligonucleotide dendrimers have thus far been
prepared via solid-phase syntheses, based on phosphoramidites
of nucleosides as building blocks.
as core,18,19 and three different types of tetrahedral hybrids.
The latter were the tetrahedral hybrids (CG)4TPM (4), with
tetrakis(p-hydroxyphenyl)methane (TPM)13 as branching
element, (CGT)4TTPA (6) and (CGG)4TTPA (7) as more
rigid and less charged four-arm hybrids, based on tetrakis-
(triazoylphenyl)adamantane (TTPA) as core,20 and (CG)4TBA
(1), (TC)4TBA (2), and (GA)4TBA (3) that are oligonucleo-
tide-bearing versions of tetrakis(p-hydroxybiphenyl)adaman-
tane (TBA). The new core “TBA” was prepared because of its
long and rigid biphenyl linker, designed to keep the DNA arms
far apart from each other, reducing the steric crowding and
the electrostatic repulsion upon assembly into materials. The
sequence 5′-CG-3′ was chosen based on our previous work,
showing that this dimer acts as a “sticky end” or “zipper” that is
strong enough to induce material formation via multivalent
interactions in aqueous solution.13 The non self-complementary
sequences 5′-TC-3′ and 5′-GA-3′ were chosen to access an
alternative, binary assembly system. Similar non self-comple-
mentary sticky ends were employed by Seeman and co-workers
to generate designed DNA crystals.21 Finally, the trimer
sequence 5′-CGG-3′ was included to investigate the influence
RESULTS AND DISCUSSION
■
Figure 1 shows the target molecules of our study. Four different
types of branched oligonucleotides were synthesized. The
group of hybrids includes six-arm, pseudo-octahedral hybrid
(CG)6HPX, based on hexakis(p-hydroxyphenyl)xylene (HPX)
2719
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
Scheme 1. Synthetic Route to Hybrid 4 Using H-Phosphonate Chemistry
of a spacer residue (G) on the effect of the “CG-zippers” during the
assembly process and to expand the methodology to trimer chains.
Recently, Jones and co-workers reported the synthesis of
cyclic diguanosine monophosphate (c-di-GMP) as a one-pot
process in solution, using a combination of phosphoramidite
and H-phosphonate chemistry.14 Although other syntheses of
cyclic diguanosine monophosphate using H-phosphonate
chemistry were reported previously,22 the recent method was
particularly attractive, as it uses no chromatographic purifica-
tion steps up to the level of the linear dimer. The method
of Jones et al. employs ribonucleotides instead of deoxyribo-
nucleotides. A ribonucleoside phosphoramidite is hydrolyzed
by treatment with pyridinium trifluoroacetate in acetonitrile
containing 2 equiv of water, followed by treatment with tert-
butylamine to remove the cyanoethyl group. Dichloroacetic
acid (DCA) is then used for DMT deprotection, followed by
addition of 2 equiv of pyridine to generate an H-phosphonate
with a free 5′-hydroxy group. The subsequent coupling step is
induced by the addition of a phosphoramidite to the reaction
mixture, followed by oxidation with TBHP. Neither step ap-
peared unsuitable for 2′-deoxynucleosides, and the assembly of
the dimer H-phosphonate was expected to be easier than that
of the dimer 3′-phosphoramidites because the polarity of the
product should allow the removal of excess reagents via
precipitation. Finally, phosphodiesters are more easily liberated
from their cyanoethyl-protected precursors than their methyl-
protected counterparts, possibly avoiding complications during
the deprotection of phenolic ester linkages.
Jones et al. mentioned above. The coupling to the core was carried
out by addition of TPM (13) to the dimer H-phosphonate
11a in pyridine, using diphenyl chlorophosphate (DPCP) as
condensing agent, followed by oxidation with iodine. Excess
iodine was removed by an aqueous washing step with sodium
bisulfite solution. The next step (detritylation) was carried out
with DCA, followed by removal of the base protecting groups
in ammonium hydroxide. The success of the reactions was con-
firmed by MALDI-TOF mass spectrometry. The mass spec-
trum gave low intensity peaks but showed the desired product
along with a number of side products, including a three-arm
hybrid.
Because the process was a one-pot procedure and did not
include any purification steps, the presence of intermediates like
9g was expected to cause side reactions. Therefore, we syn-
thesized the dimer H-phosphonates separately, as shown in
Scheme 2, purified them, and then coupled them to the core
molecules, followed by oxidation and deprotection. The synthe-
sis started with commercially available phosphoramidites 8a−g,
which were hydrolyzed by treatment with pyridinium trifluoro-
acetate in acetonitrile containing 2 equiv of water, followed by
treatment with tert-butylamine to remove the cyanoethyl group
and detritylation with DCA in dichloromethane in the presence
of 10 equiv of water. Quenching with methanol prevented retri-
tylation and afforded crude H-phosphonates 9a−g. Residual
reagents were removed by precipitating again from a mixture of
diethyl ether/ethyl acetate (1:1, v/v), starting from a concen-
trated solution in dichloromethane. After coevaporation with
acetonitrile, 9a−g were obtained in sufficient purity, as judged
First, we attempted to synthesize the DNA hybrids in a one-
pot procedure, using H-phosphonate chemistry. To establish
the methodology, hybrid 4 was explored first (Scheme 1).
For this, the 5′-DMT-protected dimer H-phosphonate 11a
was generated, using an adaptation of the methodology of
1
by ESI MS and 31P NMR. The H NMR spectrum showed
residual DMT alcohol, which could not be removed completely
by reprecipitation, and that proved inconsequential for the sub-
sequent couplings with phosphoramidites 10c−t. The following
2720
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
Scheme 2. Synthesis of H-Phosphonates of Protected Dimers
Scheme 3. Synthesis of Branched DNA Hybrids (CG)4TBA (1), (TC)4TBA (2), and (GA)4TBA (3)
three dimer sequences were prepared to access DNA hybrids
with self-complementary or non-selfcomplementary arms:
5′-CG-3′, 5′-TC-3′, and 5′-GA-3′.
showed peaks for products with only two or three DNA arms.
Pivaloyl chloride and 2-chloro-5,5-dimethyl-1,3,2-dioxaphos-
phorinane 2-oxide (DMOCP) gave lower yields still and were
therefore not tested further. Using DPCP, the reaction tem-
perature was reduced to −40 °C as recommended by Reese and
Song for other H-phosphonate couplings.24 Because oxidation
with I2 in pyridine/H2O led to hydrolysis of the aryl nucleoside
H-phosphonate diester, the reaction mixture was first treated
with I2 in dry pyridine, followed by addition of water after 1 min.23
The fully protected crude hybrid (which itself could not be
detected in MALDI spectra, but was detectable after removal of
the DMT groups) was then treated with dichloroacetic acid in
dichloromethane to give 5′-deprotected hybrid 15, followed
by treatment with ammonium hydroxide to yield crude 1.
Purification by reversed-phase HPLC on a C-8 column at
50 °C gave 1 in 30% yield. A large-scale synthesis of (CG)4TBA
(12 μmol) was also performed. It gave 10 mg of pure 1 (30%
overall yield of isolated pure material).
Coupling with tetrazole as activator in dioxane as solvent
followed by oxidation with tert-butyl hydroperoxide (TBHP)
gave dimers 11a−c. The dimer H-phosphonates were purified
by two precipitations from ethyl acetate, starting from solutions
in dichloromethane/MeOH (9:1, v/v) to afford 11a−c in
1
42−44% overall yield. The H NMR spectra showed a purity
greater than 90%, and the dimers were then used for coupling
to cores, without further purification.
With the dimer H-phosphonates in hand, we turned to
coupling them to the branching elements or cores to generate
branched hybrids. The synthesis of TBA hybrids was performed
as shown in Scheme 3. The synthesis of 1 started with the
coupling of dimer H-phosphonate 11a to TBA core 12 in a
mixture of pyridine and CH3CN (1:1, v/v), using DPCP as
condensing agent.23 Initial attempts at room temperature did
not give satisfactory results, and the MALDI-TOF mass spec-
trum of the crude reaction mixture, measured after detritylation,
Using similar methodology, two other hybrids with non-
selfcomplementary arms, namely (TC)4TBA (2) and
2721
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
hybrid with a free hydroxy group.15 All hybrids were success-
fully purified via one-stage HPLC purification at 50 °C.
We then extended our work to hybrids with trimer DNA
arms (Scheme 5). We chose (CGT)4TTPA (6) and (CGG)4TTPA
(7) as targets. Both feature a nucleoside as spacer between
the tetrakis(triazolylphenyl)adamantane core and the CG
zipper sticky ends. Studies on DNA-mediated crystalliza-
tion of gold nanoparticles have shown that linkers can be critical
for obtaining highly ordered crystallites.25 The first nucleoside
was attached to the acetylenic core via Cu(I)-catalyzed Huisgen
cycloaddition.26 The protocols reported earlier for hybrids with
dimer arms20 were optimized for 22. Here, 3′-azido-3′-deoxyth-
ymidine (AZT, 21t) was coupled to tetrakis(4-ethynylphenyl)-
adamantane (20) in the presence of catalytic amounts of CuSO4
and sodium ascorbate in DMSO/water. Using an equimolar ratio
of copper and reducing agent led to a reaction time of 1 h, com-
pared to 10 h required for the earlier protocol.20,19 Further,
dropwise addition of a 5% aqueous EDTA solution at the end of
the reaction produced 22 as a colorless precipitate that could be
readily isolated from all other components of the reaction mix-
ture by centrifugation. Compound 23 was prepared as reported.19
Following the protocol developed for 1−3, hybrids 6 and 7
were prepared by coupling the H-phosphonate building block
of the CG dimer (11a) to 22 or 23 with DCPC as coupling
agent in a mixture of pyridine/CH3CN at −40 °C. Higher
acetonitrile content led to incomplete solubilization of the
starting materials. The subsequent oxidation with TBHP was
also performed at −40 °C, as higher temperatures led to partial
decomposition. Detritylation with 80% aqueous AcOH and
washing with hexanes gave the 5′-deprotected hybrids. The final
basic deprotection was induced by treating with concentrated
aqueous ammonia, followed by addition of the so-called AMA
mixture (ammonium hydroxide/methylamine, 1:1). Adding AMA
from the start led to side products with a mass 14 Da higher than
that of 6/7 that may have been formed by nucleophilic substi-
tution involving methylamine. Pure hybrids were again obtained
after a single chromatographic step, employing a gradient of
CH3CN in 10 mM TEAA buffer and a C8 column either at
23 or 70 °C.
Figure 2. MALDI-TOF mass spectra of (CG)4TBA (1): (a) crude
product 1 after precipitation from ethyl acetate; (b) after single-stage
HPLC purification.
(GA)4TBA (3), were synthesized in 28% and 20% yield,
respectively. MALDI-TOF spectra of crude and purified 1 are
shown in Figure 2. After the acidic step for the removal of
DMT groups, approximately 5−10% of the three-arm product
was observed, possibly because of incomplete conversion,
as only one coupling cycle was employed. The final basic
deprotection step to remove the nucleobase and phospho-
diester protecting groups caused approximately 10% side pro-
1
duct with three DNA arms and one phosphate group. A H
NMR spectrum of (GC)4TBA (1), acquired under denaturing
conditions (100 mM NaOH, 60 °C) showed the expected set
of sharp signals (Figure 3).
To test the scope of the method, other branched DNA
hybrids were synthesized. The first synthesis employed the
smaller tetrahedral TPM core (13), leading to (CG)4TPM (4)
in 25% overall yield (Scheme 4). The second hybrid prepared
was based on the pseudo-octahedral six-arm HPX core (14)
and gave (CG)6HPX (5) in 20% overall yield. The synthesis of
four-arm TPM hybrids proceeded similarly to that of TBA
hybrids, but in the case of the six-arm HPX hybrid, slightly
larger amounts of side products were observed. After complete
deprotection, approximately 10% phosphorylated five-arm prod-
uct was detected, along with approximately 15−20% five-arm
We then proceeded to studying the assembly process for the
new hybrids 1, 2, and 7. Figure 4 shows association and melting
profiles for 1 and control compound 2, measured at 260 nm.
Assays were started at 10 mM NaOH and 95 °C to ensure
full denaturation, followed by neutralization at 95 °C. Each
cooling and heating cycle was repeated at least twice to ensure
reproducibility. At 10 μM hybrid concentration and low salt
concentration, hybrid 1 showed a melting point (Tm) of 65 °C,
whereas control hybrid 2 with non-selfcomplementary arms did
not show any transition (Figure 4a). As expected, addition of
NaCl shifted the melting transition for 1 to higher temperatures.
1
Figure 3. H NMR spectrum of purified 1 (0.1 M NaOH/D2O 9:1, 500 MHz, 60 °C, with solvent suppression).
2722
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
Scheme 4. Synthesis of the Branched DNA Hybrids (a) (CG)4TPM (4) and (b) (CG)6HPX (5)
Scheme 5. Synthesis of the Branched DNA Hybrids (CGT)4TTPA (6) and (CGG)4TTPA (7)
Since the high temperature baseline could no longer be esta-
blished, the melting point could not be determined (Figure 4b).
After addition of 100 mM MgCl2, a sharp drop in absorption
was observed below 25 °C for the solution of 1 (Figure 4c),
accompanied by the formation of a visible precipitate. During
the subsequent heating, the precipitate melted at a higher tem-
perature (starting at 35 °C) than the temperature of onset of
precipitation, as expected for the formation of large assemblies
whose kinetics of formation and disassembly are much slower
than those of individual duplexes. At 10 mM MgCl2, the onset
2723
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
Figure 4. Thermal association and melting profiles of (CG)4TBA (1) and (TC)4TBA (2) with non-selfcomplementary DNA arms at 10 μM hybrid
concentration in 10 mM phosphate buffer, 1.5 mM NaOAc, and cooling or heating rates of 0.5 °C/min. Samples were dissolved by using 10 mM
NaOH (100 μL), heated to 95 °C, and neutralized with acetic acid solution (10 mM, 100 μL), followed by dilution to 650 μL through addition of
sodium phosphate buffer and water. The first data point after neutralization and dilution is indicated on the graphs. Data from cooling curves are
shown with filled symbols, subsequent heating curves are shown with open symbols: (a) curves without additional salt, (b) plus 150 mM NaCl, (c)
plus an additional 100 mM MgCl2. The drop of absorbance in (c) was accompanied by formation of a precipitate, as shown in (d) (photograph of
cuvettes) for blank buffer, non-self-complementary hybrid 2, and 1.
of precipitation for 1 was shifted to lower temperatures (10−
12 °C), than the 25−30 °C range found at 100 mM MgCl2 for
this compound, and the extent of precipitation was decreased.
Again, no melting was observed for 2 under any of the salt con-
ditions tested.
Compared to the other branched hybrids studied thus far,13,19
compound 1 is the first hybrid for which the earliest signs of duplex
formation (Figure 4a) and formation of a macroscopically visible
material occur at very different temperatures/salt concentrations.
We hypothesize that this is due to slower kinetics, caused by the
long and stiff biphenyl linkers that place the CG dimers much
further away from each other than in hybrids 4 and 5.
Compound (CGG)4TTPA (7) behaved similarly to (CG)4TTPA
that lacks the “linker nucleoside”.19 In the absence of magnesium
ions, melting transitions were observed at 25 °C (no NaCl) and
51 °C (150 mM NaCl, Figure 5). Upon addition of MgCl2, even at
85 °C, an off-white material precipitated that did not redissolve
when the sample was heated again after the end of the cooling
curve. A magnesium concentration of 10 mM sufficed to induce
precipitation, but at 100 mM MgCl2 insoluble material formed
that did not redissolve upon heating.
CONCLUSIONS
■
In conclusion, we present a convenient route to branched
oligonucleotide hybrids with organic cores that can produce
10 mg batches of pure hybrids within days. The methodology
may be scaled up to produce larger quantities for material
sciences applications.
The results are noteworthy, both on a synthetic level and on
the level of the assembly properties of the hybrids. Synthetically,
Figure 5. Thermal association curves for (CGG)4TTPA (7), at 5 μM
hybrid concentration in 10 mM phosphate buffer, pH 7, and NaOAc
⧫
(1.5 mM), (a) buffer only ( ), +150 mM NaCl (Δ), or (b) plus
●
additional 100 mM MgCl2 ( ), (c) photograph of cuvette showing
precipitates of hybrid 7 in presence of 100 mM MgCl2.
2724
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
(CH2Cl2/CH3OH = 9:1 v/v); Rf = 0.13; HRMS (ESI-TOF) m/z calcd
for C17H18N5O6P [M − H]− 418.0, m/z obsd 418.0; 31P NMR
(DMSO-d6, 121.5 MHz) δ = 1.05; 1H NMR (DMSO-d6, 300 MHz) δ
(ppm) = 11.28 (brs, 1H), 8.38 (d, J = 8.3 Hz, 1H), 8.01 (s, 1H), 7.75
(d, JP−H = 619 Hz, 1H), 7.68−7.10 (m, 5H), 6.23−6.09 (m, 1H),
4.83−4.71 (m, 1H), 4.14−4.05 (m, 1H), 3.69−3.59 (m, 2H), 2.33 −
2.11 (m, 2H).
the H-phosphonate route presented here is convenient, as it
produces the dimer building blocks quickly and from inexpensive
starting materials. The H-phosphonate moiety of the 3′-terminal
nucleoside remains unreactive throughout the coupling to the
5′-terminal nucleoside and the oxidation, thus avoiding the need
for a protecting group. The coupling of dimer H-phosphonates
with the cores gave satisfactory yields, without repetitive pro-
cedures otherwise used to convert residual unreacted phenolic
hydroxy groups. The fact that the unprotected phosphodiester
linkage between DNA chain and aromatic core does not
entirely eliminate the low-level fragmentation for purine-
containing sequences15 further confirms the assumption that
these reactions are not induced during deprotection of phos-
phates, but by another base-induced fragmentation process
favored by the dendridic structure. Precipitation produces
dimers of sufficient purity to couple them to the cores, thus
avoiding a difficult and expensive chromatographic step that
would most probably result in the partial loss of cyanoethyl
protecting groups.15
From the point of view of new materials, the two separate
transitions observed for the new, stiff and expanded hybrid
(CG)4TBA (1) in the assembly experiments are interesting.
One transition occurs at low salt concentration (Figure 4a),
confirming the strong multivalency underlying the hybridiza-
tion process (the predicted melting point for the linear duplex
of 5′-CG-3′ is −90.8 °C at 10 μM DNA concentration with
150 mM NaCl and 100 mM MgCl2).27 Only upon addition of
MgCl2 did the assemblies precipitate (Figure 4c).
It is tempting to speculate that the latter process is caused
by a closer packing of duplexes or “condensation”, increasing
the density and thus driving precipitation. The “two-transition-
phenomenon” is in contrast to the properties that the cor-
responding hybrids with TPM core (4) or HPX core (5) show
(broad transitions at low salt concentration and precipitation
accompanying hybridization in the presence of MgCl2).13,19 We
suspect that the stiffer core and the longer, more rigid linker
produces more porous assemblies that do not condense readily
upon addition of the divalent cation. Thus, the unusual
properties observed for hybrids such as 1 further broaden our
understanding of this fascinating class of compounds.
N4-Benzoyl-2′-deoxycytidine-3′-yl H-Phosphonate (9c). The
reaction was performed following the general protocol A, starting from
[N4-benzoyl-5′-O-(4,4-dimethoxytrityl)-2′-deoxycytidine-3′-yl](N,N′-
diisopropyl)cyanoethylphosphoramidite (8c, 500 mg, 600 μmol) to
yield (212 mg, 530 μmol, 89%) of 9c as an off-white solid: TLC
(CH2Cl2/CH3OH = 9:1 v/v); Rf = 0.11; HRMS (ESI-TOF) m/z calcd
for C16H18N3O7P [M − H]− 394.0, m/z obsd 394.0; 31P NMR
(DMSO-d6, 121.5 MHz) δ = 1.05; 1H NMR (DMSO-d6, 300 MHz) δ
(ppm) = 11.10 (brs, 1H), 8.38 (d, J = 7.6 Hz, 1H), 8.02 (s, 1H), 7.74
(d, JP−H = 616 Hz, 1H), 7.69−7.24 (m, 5H), 6.20−6.10 (m, 1H),
4.86−4.66 (m, 1H), 4.19−4.02 (m, 1H), 3.75−3.57 (m, 2H), 2.28−
2.12 (m, 2H).
N2-Isobutyryl-2′-deoxyguanosine-3′-yl H-Phosphonate (9g).
The reaction was performed following general protocol A, starting
from [5′-O-(4,4′-dimethoxytrityl)-N2-isobutyryl-2′-deoxyguanosine-3′-yl]-
(N,N′-diisopropyl)cyanoethylphosphoramidite (8g, 500 mg, 590 μmol)
to yield (220 mg, 550 μmol, 93%) of 9g as an off-white solid: TLC
(CH2Cl2/CH3OH = 9:1 v/v); Rf = 0.12; HRMS (ESI-TOF) m/z calcd
for C14H20N5O7P [M − H]− 400.1, m/z obsd 400.1; 31P NMR (DMSO-
1
d6, 121.5 MHz) δ = 0.77; H NMR (DMSO-d6, 300 MHz) δ (ppm) =
12.04 (brs, 1H), 11.82 (brs, 1H), 8.23 (s, 1H), 7.77 (d, JP−H = 623 Hz,
1H), 6.28−6.19 (m, 1H), 4.91−4.80 (m, 1H), 4.09−4.00 (m, 1H),
3.63−3.50 (m, 2H), 2.82−2.59 (m, 3H), 1.12 (d, J = 6.8 Hz, 6H).
General Protocol B (Synthesis of Dimer H-Phosphonates).
Monomer H-phosphonate 9a, 9c, or 9g, previously dried at 0.1 mbar,
was used without further purification. After addition of molecular
sieves 3 Å (5 bulbs), 9a/9c or 9g (500 μmol) was dissolved in dioxane
(2.5 mL) by heating to 50 °C. The mixture was allowed to cool to
room temperature. Then a solution of previously dried phosphor-
amidite 10c/10 g or 10t (600 μmol, 1.2 equiv) in dioxane (2.5 mL)
was added, followed by tetrazole solution (0.45 M in acetonitrile,
650 μmol, 1.3 equiv). After 30 min, TLC showed completion of the
reaction (CH2Cl2/MeOH, 90/10, v/v), and the mixture was cooled to
0 °C. tert-Butyl hydroperoxide (TBHP) (1.5 mmol, 3 equiv, 5.5 m in
decane, over molecular sieves) was added, and the reaction mixture
was allowed to reach room temperature within 15 min. The mixture
was concentrated in vacuo, and the resulting foam was dissolved in
CH2Cl2/CH3OH (9/1 v/v, 0.5 mL) followed by precipitation with a
mixture of diethyl ether and ethyl acetate (3.0 mL, 1/1, v/v). The
precipitate was isolated by centrifugation, and the supernatant was
aspirated. The solid was redissolved in CH2Cl2/CH3OH (9/1, v/v,
0.5 mL), and precipitated again by the addition of ethyl acetate
(3.0 mL). This process was repeated three more times to give the
dimer H-phosphonates 11a, 11b, and 11c as pale yellow solids.
5′-DMT-CBz-PO(OCH2CH2CN)-GiBu-3′H-phosphonate (11a).
The reaction was carried out according to general protocol B, starting
from 9g (200 mg, 500 μmol) and 10c (500 mg, 600 μmol, 1.2 equiv).
Precipitation with ethyl acetate afforded 240 mg (210 μmol, 42%) 11a
as a pale yellow solid. The isolated product was used for coupling to
the core without any further purification (purity >95%, detected by
NMR and mass spectrometry): TLC (CH2Cl2/CH3OH, 9:1 v/v) Rf =
0.12; MS (MALDI-TOF) calcd for C54H56N9O16P2 [M − H]− 1148.3,
m/z obsd 1147.1; 31P NMR (DMSO-d6, 121.5 MHz) δ = −0.10,
−2.55, −2.67; 1H NMR (DMSO-d6, 300 MHz) δ (ppm) = 12.02 (brs,
1H), 11.86 (brs, 1H), 11.27 (brs, 1H), 8.25−8.22 (m, 2H), 8.00 (d, J =
7.5 Hz, 2H), 7.62 (m, 1H), 7.51 (t, J = 7.5 Hz, 2H), 7.40−7.11 (m,
10H), 6.92−6.82 (m, 4H), 6.67 (d, JP−H = 573 Hz, 1H, 6.33−6.22,
6.20−6.10 (2 m, 2H), 5.09−4.95, 4.80−4.69 (2 m, 2H), 4.30−4.11 (m,
4H), 3.71, 3.72 (2 s, 6H), 3.30−3.27 (m, 5H), 2.88 (t, J = 5.7 Hz, 2H),
2.78−2.66 (m, 2H), 2.62−2.50 (m, 2H), 1.16−1.03 (m, 6H).
EXPERIMENTAL SECTION
■
General Protocol A (Synthesis of Monomer H-Phosphonates).
To a solution of commercially available phosphoramidite (500 μmol)
in CH3CN (2.5 mL) and H2O (18.0 μL, 1.0 mmol, 2.0 equiv) was
added pyridinium trifluoroacetate (115.8 mg, 600 μmol, 1.2 equiv).
After 5 min, tert-butylamine (2.5 mL) was added, and the reaction
mixture was stirred for 15 min at room temperature. The mixture was
then concentrated in vacuo to a foam. The residue was dissolved in
CH3CN (5.0 mL) and concentrated again to a foam, and this process
was repeated one more time. To the residue dissolved in CH2Cl2
(6.0 mL) was added H2O (90 μL, 5.0 mmol, 10 equiv), followed by
6% dichloroacetic acid in CH2Cl2 (6.0 mL, 4.4 mmol). After 10 min,
the reaction was quenched with CH3OH (2.0 mL) and concentrated
in vacuo to a small volume (500 μL). The crude product was then
precipitated by addition of diethyl ether and centrifuged, and the
supernatant solution was aspired. The solid was redissolved in
CH2Cl2/CH3OH (9/1, v/v, 500 μL), precipitated again by the addi-
tion of diethyl ether (3.0 mL), and washed twice with ethyl acetate/
diethyl ether (1/1, v/v, 3.0 mL) to give the desired product.
N6-Benzoyl-2′-deoxyadenosine-3′-yl H-Phosphonate (9a).
The reaction was performed following general protocol A, starting
from [N6-benzoyl-5′-O-(4,4-dimethoxytrityl)-2′-deoxyadenosine-3′-yl]-
(N,N′-diisopropyl)cyanoethylphosphoramidite (8a, 500 mg, 580 μmol)
to yield 225 mg (540 μmol, 93%) of 9a as an off-white solid: TLC
5′-DMT-T-PO(OCH2CH2CN)-CBz-3′H-phosphonate (11b). Gen-
eral protocol B was used, starting from 9c (200 mg, 510 μmol) and 10t
(454 mg, 610 μmol, 1.2 equiv). Yield 235 mg (220 μmol, 44%) of 11b
2725
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
as a yellowish solid. The product was used for coupling to the core
without further modification and had a purity >95%, as detected by
NMR and MS: TLC (CH2Cl2/CH3OH, 9:1, v/v) Rf = 0.10; MS
(MALDI-TOF) calcd for C50H51N6O16P2 [M − H]− 1053.3 m/z obsd
1052.2; 31P NMR (DMSO-d6, 121.5 MHz) δ = 0.33, −2.49, −2.70; 1H
NMR (DMSO-d6, 300 MHz) δ (ppm) = 11.37 (brs, 2H), 8.19−8.10
(m, 1H), 8.00 (d, J = 7.3 Hz, 2H), 7.62 (t, J = 8.1 Hz, 1H), 7.55−
7.46 (m, 3H), 7.40−7.10 (m, 10H), 6.92−6.82 (m, 4H), 6.28−6.18,
6.17−6.08 (2 m, 2H), 5.17−5.06, 4.81−4.56 (2 m, 2H), 4.40−4.14 (m,
4H), 3.72, 3.70 (2 s, 6H), 3.31−3.28 (m, 6H), 2.99−2.84 (m, 2H),
1.11 (s, 3H).
aqueous phosphate buffer (5.0 mL, 0.2 M, pH 7). The aqueous phase
was separated and back-extracted with CH2Cl2 (5 mL). The combined
organic layers were dried over Na2SO4, filtered, and concentrated in
vacuo. The residue was coevaporated twice from toluene and dissolved
in a minimal amount of CH2Cl2/CH3OH (95/5, v/v), followed
by precipitation with ethyl acetate. After centrifugation (3500 rpm,
5 min) the residue was washed three times with ethyl acetate in an
ultrasonic bath for 1 min and centrifuged. The solid obtained was
dried at 0.001 mbar and 40 °C to yield the crude fully protected
hybrid, which was deprotected without further purification. For this,
the protected hybrid was dissolved in CH2Cl2 (5 mL) and H2O
(10 equiv), followed by addition of 6% dichloroacetic acid (DCA) in
CH2Cl2 (5 mL, 3.5 mmol). After 10 min, the reaction was quenched
by addition of CH3OH (2 mL). The solution was then concentrated,
and the hybrid was precipitated by addition of diethyl ether/ethyl
acetate (3 mL, 1/1, v/v). The precipitate was separated by centri-
fugation, redissolved in a minimal volume of CH2Cl2/CH3OH
(0.5 mL, 9/1, v/v), and precipitated again by addition of ethyl acetate
(3 mL). This process was repeated three more times to give the DMT-
deprotected hybrid (15, 16, 17, 18, or 19). To remove the cyanoethyl
groups and the protecting groups of the nucleobases, the product was
treated with ammonium hydroxide (5 mL) for 5 h at 55 °C. Excess
ammonia was removed by passing a stream of N2 over the surface until
the sample was odorless. The remaining solution was evaporated to
dryness by lyophilization to yield crude hybrid 1, 2, 3, 4, or 5. The
crude was then purified by reversed-phase HPLC (C8 column) using a
gradient of 5−40% CH3CN in 10 mM TEAA buffer at 55 °C.
(CG)4TBA (1). General protocol C was used, starting from TBA
core 12 (10 mg, 12.4 μmol) and 11a (200 mg, 175 μmol). After
complete deprotection, the crude hybrid was dissolved in 10 mM
TEAA buffer containing 5% CH3CN and subjected to HPLC puri-
fication (C8 column, 250 × 20 mm), using a gradient of CH3CN in
10 mM TEAA buffer, 5−30% in 60 min at 55 °C. Hybrid 1 eluted at
tR = 35 min, yield 30%: MS (MALDI-TOF) calcd for C134H144N32-
O52P8, [M − H]− 3280, obsd 3280; 31P NMR (122 MHz, NaOH
5′-DMT-GiBu-PO(OCH2CH2CN)-ABz-3′H-phosphonate (11c).
General protocol B was used, starting from 9a (200 mg, 0.47 mmol)
and 10g (481 mg, 0.57 mmol, 1.2 equiv). Yield 218 mg (39%,
0.19 mmol) of 11c as pale yellow solid. The dimer building block
(purity >95%, as detected by NMR and MS) was used for coupling
to the core without further modification: TLC (CH2Cl2/CH3OH, 9:1,
v/v); Rf = 0.09; MS (MALDI-TOF) calcd for C55H56N11O15P2 [M −
H]− 1172.3, m/z obsd 1171.3; 31P NMR (DMSO-d6, 121.5 MHz): δ =
2.02, −2.66, −2.80; 1H NMR (DMSO-d6, 300 MHz) δ (ppm) = 12.16,
11.76, 11.20 (3 brs, 3H), 8.47−8.02 (m, 2H), 7.68− 7.61 (m, 2H),
7.52 (t, J = 7.3 Hz, 3H), 7.40−7.15 (m, 11H), 6.95 − 6.82 (m, 4H),
6.37−6.17 (2 m, 2H), 5.04, 4.79 (2 m, 2H), 5.26−5.02, 5.01−4.73
(2 m, 2H), 4.30−4.11 (m, 4H), 3.74, 3.73 (2 s, 6H), 3.42−3.13
(m, 6H), 2.96−2.86 (m, 1H), 2.81 − 2.68 (2 m, 2H), 2.62−2.50 (2 m,
2H), 1.15−1.07 (m, 6H).
Synthesis of Cores. The synthesis of the TPM and the HPX core
followed literature protocols.13,19 The TBA core was obtained from
1,3,5,7-tetrakis(4-methoxybiphen-4-yl)adamantane28 via demethylation
with boron tribromide, as described below.
1,3,5,7-Tetrakis(4′-hydroxy[1,1′-biphenyl]-4-yl)adamantane
(TBA) (12). 1,3,5,7-Tetrakis(4-methoxybiphen-4′-yl)adamantane
(600 mg 690 μmol, 1.00 equiv) was placed in a Schlenk flask and
dissolved in chloroform (75 mL). After being cooled to 0 °C, BBr3
(1.04 mL, 11.1 mmol, 15.0 equiv) was added dropwise (exothermic
reaction). The mixture was stirred for an additional 2 h at this tem-
perature, and allowed to react for another 16 h at room temperature
while a precipitate formed. The mixture was treated three times with
methanol (3 × 50 mL) and evaporated in vacuo to dryness.
Purification via column chromatography (silica gel, 50 g, CH2Cl2/
methanol 15:1 to 10:1, v/v) yielded 465 mg (84%) of 1,3,5,7-
tetrakis(4′-hydroxy[1,1′-biphenyl]-4-yl)adamantane as an off-white
1
(0.1 M)/D2O 9:1 v/v) δ −1.05, −5.61; H NMR (500 MHz, NaOH
(0.1 M) in D2O; 9:1 v/v, 60 °C) δ 8.39 (s, 4 × 1H, H8G), 8.20−8.02
(m, 24H, Ar-H), 7.89 (d, J = 7.8 Hz, 4 × 1H, H6C), 7.81 (d, J = 9.4 Hz,
6H, Ar-H), 6.77−6.69 (m, 4 × 1H, H1′G), 6.50−6.43 (m, 4 × 1H,
H1′C), 6.25 (d, J = 8.0 Hz, 4 × 1H, H5C), 5.53−5.46 (m, 4 × 1H,
H3′G), 4.51−4.33 (m, 12H, H4′C/H5′/5″G), 4.10 (dd, J5′−4′ = 4.4,
J5′−5″ = 11.6, 4 × 1H, H5′C), 4.01 (dd, J5″−4′ = 5.8, J5″−5′ = 12.6, 4 × 1H,
H5″C), 3.28−3.19 (m, 4 × 1H, H2′G), 3.17−3.09 (m, 4 × 1H, H″G),
2.67−2.56 (m, 4 × 1H, H2′C), 2.71 (s, 12H, core-H), 2.31−2.21 (m,
4 × 1H, H2″C).
(TC)4TBA (2). Following general protocol C, 12 (4 mg, 5 μmol)
was reacted with 11b (73.7 mg, 70 μmol). Crude 2 was taken up in
10 mM TEAA buffer containing 5% CH3CN, and HPLC purified (C8
column, 250 × 4 mm) using a gradient of CH3CN in 10 mM TEAA
buffer, 5−35% in 60 min at 55 °C. Hybrid 2 eluted at tR = 37 min,
yield 28%; MS (MALDI-TOF) calcd for C134H148N20O56P8 [M − H]−
3180, obsd 3180.
(GA)4TBA (3). Following general protocol C, 12 (2.0 mg, 2.5 μmol)
was reacted with 11c (41 mg, 35 μmol). The crude was taken up in
10 mM TEAA buffer containing 5% CH3CN, and HPLC purified (C8
column, 250 × 4 mm), using a gradient of CH3CN in 10 mM TEAA
buffer, 5−35% in 60 min at 55 °C, with elution at tR = 40 min, yield
20%; MS (MALDI-TOF) calcd for C138H143N39O48P8 [M − H]−
3361, obsd 3360.
(CG)4TPM (4). General protocol C was used, starting from TPM
core 13 (2 mg, 5.2 μmol) and 11a (83.5 mg, 72.8 μmol, 3.5 equiv per
OH group of core). The crude hybrid was dissolved in 10 mM TEAA
buffer containing 5% CH3CN and HPLC purified (C8 column, 250 ×
4 mm), using a gradient of CH3CN in 10 mM TEAA buffer, 5−18% in
45 min at 55 °C. Hybrid 4 eluted at tR = 16 min, yield 25%; MS
(MALDI-TOF) calcd for C101H117N32O52P8 [M − H]− 2857, obsd
2856.
1
solid: Rf = 0.31 (CH2Cl2/methanol 15:1, v/v); H NMR (400 MHz,
MeOH-d4) δ = 1.89 (s, 12 H), 6.89 (d, 3J = 8.5 Hz, 8 H), 7.22 (d, 3J =
8.3 Hz, 8 H), 7.40−7.42 (m, 16 H) ppm; 13C NMR (100 MHz,
MeOH-d4) δ = 40.1, 48.5, 116.7, 126.7, 127.4, 129.1, 133.7, 139.9,
̃
149.1, 158.0 ppm; IR (DRIFT) ν = 3437 [br, w, νAr(OH)], 3029 [w,
νAr(CH)], 2925 (w), 2851 [w, νAd(CH2)], 1891 (vw), 1610 (w), 1498
[m, νAr(CC)], 1398 (w), 1355 [w, δAr(OH)], 1260 [w, δAr(CH)],
1172 (w), 1104 (w), 1003 (w), 819 [m, νAr(p-subst.)], 786 (w), 704
(w), 583 (w), 532 (w), 456 (w), 418 (w) cm−1; MS (FAB, 3-NBA),
m/z 809 (55) [M + H]+, 808 (100) [M+]; HRMS (C58H48O4) calcd
808.3552, obsd 808.3555.
General Protocol C (Synthesis of DNA Hybrids). The core
[1,3,5,7-tetrakis(4-hydroxybiphen-4-yl)adamantane (TBA, 12),
tetrakis(4-hydroxyphenyl)methane (TPM, 13), or 1,4-phenylenebis-
[tri(4′-hydroxyphenyl)methane] (HPX, 14)] was dried for 1 h at
100 °C and 0.001 mbar. The core (5.0 μmol) was then mixed with
5′-O-DMT-dimer-H-phosphonate (11a, 11b, or 11c, 3.5 equiv per OH
group of core), and the mixture was coevaporated from dry pyridine
(2 × 2 mL). After addition of molecular sieve 3 Å (5 beads), a mixture
of dry pyridine and CH3CN (0.5 mL, 2/3, v/v) was added in an argon
atmosphere. The reaction mixture was then cooled to −40 °C.
Diphenyl chlorophosphate (DPCP) (1.50 equiv) was added, and the
mixture was allowed to stir for 30 min at −40 °C. Then a solution of
iodine in dry pyridine (2 M, 1.5 equiv to dimer) was added, followed
by H2O (5 equiv) after 1 min. The reaction mixture was slowly
allowed to reach room temperature and was stirred for further 30 min.
After addition of CH2Cl2 (10 mL), the mixture was washed with a
solution of aqueous sodium thiosulfate (5 mL, 10%, w/w) and
(CG)6HPX (5). Following general protocol C, HPX core 14 (2 mg,
3.0 μmol) was reacted with 11a (73.3 mg, 63.8 μmol, 3.5 equiv per
OH group of core). The crude hybrid was taken up in 10 mM TEAA
2726
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
buffer containing 5% CH3CN and HPLC purified on a C8 column
(250 × 4 mm) with a gradient of CH3CN in 10 mM TEAA buffer, 5−
18% in 40 min at 55 °C. Hybrid 5 eluted at tR = 22 min, yield 20%; MS
(MALDI-TOF) calcd for C158H178N48O78P12 [M − H]− 4368, obsd
4367.
temperature. After 30 min, the solution was diluted with CH2Cl2
(2 mL), and was poured into aqueous Na2S2O3 solution (600 μL,
0.1 M). The organic layer was washed with water (2 × 1 mL), dried
over Na2SO4, filtered, and concentrated in vacuo to dryness. The
residue was coevaporated from toluene (2 × 2 mL), taken up in a
mixture of THF/H2O (400 μL, 2:1 v/v), and precipitated with ethyl
acetate (2 mL). The precipitate was washed with ethyl acetate (4 ×
1 mL), and was dried for 30 min at 0.1 mbar. The solid was taken up
in acetic acid (80%, 150 μL), and the mixture was stirred at room
temperature. The solution was extracted with hexane (8 × 400 μL).
After 1 h, the reaction mixture was diluted with water (500 μL) and
was washed again with hexane (2 × 400 μL). Residual hexane was
removed with a gentle stream of air, directed onto the surface of the
solution. After 20 min, the aqueous solution was lyophilized to
dryness. The hybrid was then dissolved in saturated aqueous ammonia
(28%, 200 μL), and was shaken for 2 h. Then, aqueous methylamine
(40%, 200 μL) was added, and the reaction mixture was shaken for
3 h. Water (300 μL) was added, and excess ammonia and methylamine
were removed with a gentle stream of air directed onto the surface of
the solution. After 60 min, the aqueous solution was lyophilized to
dryness. Samples were purified by reversed-phase HPLC (C8 column),
using a gradient of CH3CN in 10 mM TEAA buffer (5−15% in
20 min, 15−40% in 60 min). Hybrid 7 eluted at a retention time (tR)
of 47 min: yield 28%; MS (MALDI-TOF) calcd for C158H175N64-
Tetrakis[4-(1,3′-deoxythymidin-3′-yl-1,2,3-triazol-4-yl)-
phenyl]adamantane (22). The following is a modification of a
literature protocol.20 3′-Azido-3′-deoxythymidine (21a, 52.0 mg,
0.195 mmol) and 1,3,5,7-tetrakis(4-ethinylphenyl)adamantane (20)
(21.4 mg, 0.039 mmol) were dissolved in DMSO (0.5 mL). A slurry of
CuSO4·5H2O (5.1 mg, 0.020 mmol) and sodium L-(+)-ascorbate (4.4
mg, 0.022 mmol) in water (0.1 mL) was added, and the resulting
mixture was stirred at room temperature. To monitor the reaction, a
small sample (1 μL) was removed, diluted 200 fold with DMSO, and
analyzed by MALDI-TOF-MS. After 1.5 h, the educt was fully con-
verted, and the green solution was poured dropwise into an aqueous
solution of EDTA (2 mL, 5%). The precipitate was washed with water
(5 × 0.5 mL) and dried over P4O10 at 0.01 mbar, yielding 58 mg (90%,
0.036 mmol) of title compound 22 as a solid. The analytical data agreed
with the literature.20
Tetrakis[4-(1-N2-isobutyryl-2′,3′-dideoxyguanosin-3′-yl-
1,2,3-triazol-4-yl)phenyl]adamantane (23). The synthesis fol-
lowed the protocol given in the literature,19 starting from 21b
(265 mg, 730 μmol) and 20 (61.0 mg, 110 μmol), and proceeded with
an overall yield of 93%.
−
−
O60P8 [M − H] 4177.0; obsd. 4178.0.
(CGT)4TTPA (6). The TTPA hybrids 6 and 7 were synthesized
using a modification of General Protocol C. To a vacuum-dried flask
with 22 (5.30 mg, 3.30 μmol), 11a (43.1 mg, 37.5 μmol) was added.
The mixture was coevaporated from dry pyridine (2 × 200 μL). Then,
molecular sieves (3 Å, 5 beads) were added, and the residue was
dried for 1 h at 0.001 mbar and 60 °C. After addition of dry pyridine
(300 μL) under argon atmosphere, the clear solution was diluted with
dry CH3CN (300 μL). The reaction mixture was cooled to −40 °C,
and diphenyl chlorophosphate (19.3 mg, 72.0 μmol) was added. The
mixture was stirred for 1 h at this temperature. Then, a mixture of
TBHP (5.5 min dodecane, 10 μL) and oxidizer solution for DNA
synthesis (0.02 M I2 in pyridine/H2O, 90 μL) was added dropwise
to the reaction solution, and the solution was stirred at −40 °C for
20 min. Aliquots (4 × 200 μL) of the reaction mixture were added to
ethyl acetate (1 mL each), leading to precipitation. The combined
slurries were centrifuged, and the residue was washed with ethyl
acetate (4 × 300 μL). The solid containing the fully protected hybrid
was dissolved in acetic acid (100 μL, 80% aqueous solution), and the
mixture was stirred at room temperature. The solution was washed
with hexane (8 × 300 μL). After 1 h, the reaction mixture was diluted
with water (600 μL) and was washed again with hexane (3 × 200 μL).
Residual hexane was removed with a gentle stream of N2, and the
aqueous solution was lyophilized to dryness. The product was then
dissolved in saturated aqueous ammonia (28%, 200 μL), and was
shaken for 2 h at room temperature. Then, aqueous methylamine
(40%, 200 μL) was added, and the reaction mixture was shaken again
for another 3 h. Water (400 μL) was added, and the excess of ammo-
nia and methylamine was removed with a gentle stream of N2, directed
onto the surface of the solution. The remaining solution was evapo-
rated to dryness by lyophilization to yield crude hybrid 6. Samples for
assembly studies were purified by reverse phase HPLC (C8 column)
using a gradient of CH3CN in 10 mM TEAA buffer (10% for 5 min,
10−15% in 10 min, 15−18% in 15 min) at 70 °C. Hybrid 6 eluted at
tR = 18 min, yield 42%: MS (MALDI-TOF) calcd for C158H180N52-
O64P8 [M − H]− 4077, obsd 4078.
ASSOCIATED CONTENT
* Supporting Information
Collection of NMR spectra, HPLC chromatograms, and
MALDI-TOF mass spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*Tel.: 49 (0) 711 685 64311. Fax: 49 (0) 711 685 64321.
E-mail: lehrstuhl-2@oc.uni-stuttgart.de.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Th. Sabirov for contributing to the synthesis of
intermediates and C. Gerlach and Dr. B. Claasen for help with
the acquisition and interpretation of NMR spectra. This work
was supported by DFG (Grant No. RI 1063/13-1 to C.R.) and
the Center for Functional Nanostructures at KIT.
REFERENCES
■
(1) Hall, R. H.; Todd, A.; Webb, R. F. J. Chem. Soc. 1957, 3291−
3296.
(2) (a) Froehler, B. C.; Matteucci, M. D. Tetrahedron Lett. 1986, 27,
469−472. (b) Froehler, B. C.; Ng, P. G.; Matteucci, M. D. Nucleic
Acids Res. 1986, 14, 5399−5407.
(3) (a) Garegg, P. J.; Lindh, I.; Regberg, T.; Stawinski, J.; Stromberg,
̈
R. Tetrahedron Lett. 1986, 27, 4051−4054. (b) Garegg, P. J.; Lindl, I.;
Regberg, T.; Stawinski, J.; Stromberg, R.; Henrichson, C. Tetrahedron
̈
Lett. 1986, 27, 4055−4058.
(4) See: (a) Li, H.; Fedorova, O. S.; Trumble, W. R.; Fletcher, T. R.;
Czuchajowski, L. Bioconjugate Chem. 1997, 8, 49−56. (b) Wada, T.;
Sato, Y.; Honda, F.; Kawahara, S.-I.; Sekine, M. J. Am. Chem. Soc. 1997,
119, 12710−12721. (c) An, H.; Wang, T.; Maier, M. A.; Manoharan,
M.; Ross, B. S.; Dan Cook, P. J. Org. Chem. 2001, 66, 2789−2801.
(d) Asseline, U.; Cheng, E. Tetrahedron Lett. 2001, 42, 9005−9010.
(e) Zain, R.; Stawinski, J. J. Org. Chem. 1996, 61, 6617−6622.
(5) (a) Reese, C. B.; Song, Q. L. J. Chem. Soc., Perkin Trans. 1 1999,
1477−1486. (b) Johansson, T.; Stawinski, J. Tetrahedron 2004, 60,
389−395.
(CGG)4TTPA (7). To 23 (10.1 mg, 5.08 μmol) was added 11a
(70.1 mg, 61.0 μmol). The mixture was coevaporated from pyridine
(1 mL) and CH3CN (1 mL). Molecular sieves (3 Å) were added,
followed by drying for 1 h at 0.001 mbar. After addition of pyridine
(300 μL) under argon atmosphere, the solution was diluted with
CH3CN (270 μL). The reaction mixture was cooled to −40 °C, and
diphenyl chlorophosphate (24.7 mg, 92 μmol) was added. After 1 h,
the hybrid was oxidized by addition of a solution of I2 in pyridine
(50 μL, 2 M), and was stirred for another 10 min. Then, water (50 μL)
was added, and the reaction mixture was allowed to reach room
2727
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728

The Journal of Organic Chemistry
Article
(6) Adamo, I.; Dueymes, C.; Schonberger, A.; Navarro, A.-E.; Meyer,
̈
A.; Lange, M.; Imbach, J.-L.; Link, F.; Morva, F.; Vasseur, J.-J. Eur. J.
Org. Chem. 2006, 436−448.
(7) Sakatsume, O.; Yamane, H.; Takaku, H.; Yamamoto, N. Nucleic
Acid Res. 1990, 18, 3327−3331.
(8) Ott, G.; Arnold, L.; Smrt, J.; Sobkowski, M.; Limmer, S.;
Hofmann, H.-P.; Sprinzl, M. Nucleosides Nucleotides 1994, 13, 1069−
1085.
(9) Sun, Q.; Edathil, J. P.; Wu, R.; Smidansky, E. D.; Cameron, C. E.;
Peterson, B. R. Org. Lett. 2008, 10, 1703−1706.
(10) (a) Caruthers, M. H. Science 1985, 230, 281−285. (b) Caruthers,
M. H. In Synthesis and Applications of DNA and RNA; Narang, S. A.,
Ed.; Academic: Orlando, 1987; pp 47−94. (c) Beaucage, S. L.; Iyer,
R. P. Tetrahedron 1992, 22, 2223−2311.
(11) Blackburn, G. M., Gait, M. J., Loakes, D., Williams, D. M., Eds.
Nucleic Acids in Chemistry and Biology, 3rd ed.; RSC Publishing:
Cambridge, 2006; pp 149−150.
(12) (a) Westheimer, F. H.; Huang, S.; Covitz, F. J. Am. Chem. Soc.
1988, 110, 181−185. (b) Reese, C. B. Tetrahedron 2002, 58, 8893−
8920.
(13) Meng, M.; Ahlborn, C.; Bauer, M.; Plietzsch, O.; Soomro, S. A.;
Singh, A.; Muller, T.; Wenzel, W.; Brase, S.; Richert, C. ChemBioChem
̈
2009, 10, 1335−1339.
(14) Gaffney, B. L.; Veliath, E.; Zhao, J.; Jones, R. A. Org. Lett. 2010,
12, 3269−3271.
(15) Griesser, H.; Tolev, M.; Singh, A.; Sabirov, T.; Gerlach, C.;
Richert, C. J. Org. Chem. 2012, 77, DOI: 10.1021/jo202505h.
(16) (a) Hudson, R. H. E.; Uddin, A. H.; Damha, M. J. J. Am. Chem.
Soc. 1995, 117, 12470−12477. (b) Vargas-Baca, I.; Mitra, D.; Zulyniak,
H. J.; Banerjee, J.; Sleiman, H. F. Angew. Chem., Int. Ed. 2001, 40,
4629−4632.
(17) (a) Hudson, R. H. E.; Damha, M. J. J. Am. Chem. Soc. 1993, 115,
2119−2124. (b) Hudson, R. H. E.; Robidoux, S.; Damha, M. J.
Tetrahedron Lett. 1998, 39, 1299−1302. (c) Shchepinov, M. S.; Mir,
K. U.; Elder, J. K.; Frank-Kamenetskii, M. D.; Southern, E. M. Nucleic
Acids Res. 1999, 27, 3035−3041. (d) Shchepinov, M. S.; Udalova, I. A.;
Bridgman, A. J.; Southern, E. M. Nucleic Acids Res. 1997, 25, 4447−
4454.
(18) Hatano, T.; Kato, T. Tetrahedron 2008, 64, 8368−8380.
(19) Singh, A.; Tolev, M.; Meng, M.; Klenin, K.; Plietzsch, O.;
Schilling, C. I.; Muller, T.; Nieger, M.; Brase, S.; Wenzel, W.; Richert, C.
̈
Angew. Chem., Int. Ed. 2011, 50, 3227−3231.
(20) Plietzsch, O.; Schilling, C. I.; Tolev, M.; Nieger, M.; Richert, C.;
Muller, T.; Brase, S. Org. Biomol. Chem. 2009, 7, 4734−4743.
̈
(21) Zheng, J.; Birktoft, J. J.; Chen, Y.; Wang, T.; Sha, R.;
Constantinou, P. E.; Ginnell, S. L.; Mao, C.; Seeman, N. C. Nature
2009, 461, 74−77.
(22) (a) Zang, Z.; Gaffney, B. L.; Jones, R. A. J. Am. Chem. Soc. 2004,
126, 16700−16701. (b) Zeng, F.; Jones, R. A. Nucleosides Nucleotides
1996, 15, 1679−1686. (c) Yan, H.; Aguilar, A. L. Nucleosides
Nucleotides Nucl. Acids 2007, 26, 189−204.
(23) Ciesl
́
ak, J.; Jankowska, J.; Sobkowski, M.; Wenska, M.;
́
Stawinski, J.; Kraszewski, A. J. Chem. Soc., Perkin Trans. 2002, 31−37.
(24) Reese, C. B.; Song, Q. Bioorg. Med. Chem. Lett. 1997, 7, 2787−
2792.
(25) (a) Nykypanchuk, D.; Maye, M. M.; van der Lelie, D.; Gang, O.
Nature 2008, 451, 549−552. (b) Park, S. Y.; Lytton-Jean, A. K. R.; Lee,
B.; Weigand, S.; Schatz, G. C.; Mirkin, C. A. Nature 2008, 451, 553−
556. (c) Macfarlane, R. J.; Lee, B.; Jones, M. R.; Harris, N.; Schatz,
G. C.; Mirkin, C. A. Science 2011, 334, 204−208.
(26) (a) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem.
2002, 66, 3057−3064. (b) Rostovtsev, V. V.; Green, L. G.; Fokin,
V. V.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 2596−2599.
(27) Calculated using Hyther by Peyret, N.; Santa Lucia, J., Jr.,
Wayne State University.
(28) Schilling, C. I.; Plietzsch, O.; Nieger, M.; Muller, T.; Brase, S.
̈
Eur. J. Org. Chem. 2011, 1743−1754.
2728
dx.doi.org/10.1021/jo202508n | J. Org. Chem. 2012, 77, 2718−2728