Phosphine-Free Stille-Migita Chemistry for the Mild and Orthogonal Modification of DNA and RNA

Article abstract of DOI:10.1002/chem.201404843

An optimized catalyst system of [Pd₂(dba)₃] and AsPh₃ efficiently catalyzes the Stille reaction between a diverse set of functionalized stannanes and halogenated mono-, di- and oligonucleotides. The methodology allows for the facile conjugation of short and long nucleic acid molecules with moieties that are not compatible with conventional chemical or enzymatic synthesis, among them acid-, base-, or fluoride-labile protecting groups, fluorogenic and synthetically challenging moieties with good to near-quantitative yields. Notably, even azides can be directly introduced into oligonucleotides and (deoxy)nucleoside triphosphates, thereby giving direct access to "clickable" nucleic acids.

Full text of DOI:10.1002/chem.201404843

DOI: 10.1002/chem.201404843
Full Paper
&
Chemical Biology
Phosphine-Free Stille–Migita Chemistry for the Mild and
Orthogonal Modification of DNA and RNA
[
a]
Andrꢀ Krause, Alexander Hertl, Fabian Muttach, and Andres Jꢁschke*
Abstract: An optimized catalyst system of [Pd (dba) ] and
tional chemical or enzymatic synthesis, among them acid-,
base-, or fluoride-labile protecting groups, fluorogenic and
synthetically challenging moieties with good to near-quanti-
tative yields. Notably, even azides can be directly introduced
into oligonucleotides and (deoxy)nucleoside triphosphates,
thereby giving direct access to “clickable” nucleic acids.
2
3
AsPh efficiently catalyzes the Stille reaction between a di-
3
verse set of functionalized stannanes and halogenated
mono-, di- and oligonucleotides. The methodology allows
for the facile conjugation of short and long nucleic acid mol-
ecules with moieties that are not compatible with conven-
Introduction
Results and Discussion
Strategies for the functionalization of nucleic acids are of enor-
mous interest for applications in modern molecular biotech-
To start with a simple system, we investigated the Stille modifi-
cation of ribo- or mixed ribo-/deoxyribodinucleotides
(Scheme S1 in the Supporting Information), which we had re-
[
1]
[2]
nology and medical research. Although numerous methods
exist to introduce modifications co- or postsynthetically into
[11]
cently identified as efficient “initiator dinucleotides” that are
selectively incorporated by RNA polymerases at the 5’ terminus
of a transcript. In these molecules with the general structure
5’-(d)UpG-3’, the guanosine nucleoside is required for recogni-
tion by the polymerase, while the “carrier nucleoside” (d)U can
be modified at the non-Watson–Crick face; preferably at posi-
tion C-5. Ideally, the C-5 substituted target compound should
be synthesized from a universal, unprotected dinucleotide pre-
cursor within a single selective and orthogonal transformation,
giving access to a broad chemical space of modifications at
minimal synthetic effort.
[
3]
chemically synthesized oligonucleotides, several limitations
[
4]
apply to chemosynthetic methods. Co-synthetic approaches
require the stability of the modified monomers under coupling
[
5]
and deprotection conditions. Moreover, standard solid-phase
synthetic procedures for oligonucleotides, and in particular for
RNA, demand for a set of mutually orthogonal protecting
groups, thereby restricting the use of acid-, base- and fluoride-
labile moieties for conjugation purposes. Enzymatic strategies,
in contrast, capitalize on the high accuracy and processivity of
[
6]
DNA and RNA polymerases, but offer only limited opportuni-
ties for site-specific modification.
Due to the hydrolytic lability of RNA, a particularly mild
catalytic system had to be identified (Table 1).
0
Pd -catalyzed cross-coupling reactions have become stan-
[
7]
[8]
dard tools for the modification of nucleosides, nucleotides
Stille–Migita reactions have previously been reported in
polar solvent systems such as NMP and DMF with [Pd (dba) ]
[
9]
and oligonucleotides. Yet, there are no reports on cross-cou-
pling reactions on unprotected RNA, which is not surprising,
given the need for strong bases in most of these reactions, in-
evitably leading to hydrolysis of the internucleotide bonds.
Among the common reactions, the Stille–Migita coupling
2
3
0
[12]
as a Pd source and P(Fu) or AsPh as a ligand. Compared
3
3
to phosphine-based ligands, AsPh -based catalyst systems ex-
3
[13]
I
hibit higher catalytic rates. Cu as an additive was discussed
to bind excessive ligand and thereby increase the activity of
[
10]
0
[14]
stands out for its mild reaction conditions, encouraging us
to attempt the coupling of stannylated moieties to 5-iodinated
pyrimidine nucleosides and oligonucleotides without addition
of bases.
the catalytic Pd species. In an initial attempt, 5-I-dUpG in
the presence of 0.5 equiv [Pd (dba) ], 2 equiv AsPh , 5 equiv
2
3
3
CuI and 12 equiv of stannane 1, reacted at 808C for 2 h, fur-
nished deoxydinucleotide 1a in 66% yield (Table 1, and Fig-
ure S1A in the Supporting Information). HPLC analysis indicat-
ed 34% unmodified starting material. Application of these con-
ditions to ribodinucleotide 5-I-UpG resulted, however, in com-
plete decomposition of the starting material (Figure S1B in the
[
a] Dipl.-Biochem. A. Krause, A. Hertl, F. Muttach, Prof. Dr. A. Jꢀschke
Institute of Pharmacy and Molecular Biotechnology
Heidelberg University, Im Neuenheimer Feld 364, 69120 Heidelberg
jaeschke@uni-hd
Supporting Information). A different catalytic system with
[9d,12a]
P(Fu) as a ligand
was only slightly more efficient, furnish-
3
ing 10% of the desired product 1b along with numerous
decomposition products like guanosine and deglycosylated
uracil derivative 1c.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404843.
Chem. Eur. J. 2014, 20, 16613 – 16619
16613
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tected ethynyl stannane 8. 5-I-UpG was converted
Table 1. Identification of a suitable catalyst system for the Stille–Migita reaction on
the unprotected 5-I-(d)UpG dinucleotide.
quantitatively within 30 min at 0.5 equiv [Pd (dba) ]
2
3
in the presence of stannanes 5 and 6 without any
sign of dinucleotide decay (Figure S5 in the Support-
ing Information). Stannane 7 reacted slower, and
a higher catalyst loading (1 equiv [Pd (dba) ]) was re-
2
3
quired to reach near-quantitative consumption of the
starting material. Besides the expected coupling
product 7a (44%) we observed additional species
eluting as a broad peak (37%) during HPLC separa-
tion. High resolution ESI mass spectrometry revealed
the masses and isotope patterns of three different
Pd-containing compounds (Figure S6 in the Support-
ing Information), supporting the assumption of an
Dinucleotide
Ligand/
co-catalyst
Yield
[%]
Starting
material [%]
Total side-product/
(G /B ) [%]
[
d]
[d]
[e] [f]
[d]
II
oxidative addition intermediate where Pd is attached
[
a]
5
5
5
5
5
5
5
-I-dUpG
AsPh
AsPh
3
/5 equiv CuI
/5 equiv CuI
66
–
10
68
99
34
–
<1
1
1
6
–
[
[
[
a]
to C-5 of the 5’-uridine moiety, as well as a hydrated
and a ligand-free analogue. For TMS-acetylene-de-
rived stannane 8, a moderate ratio of reduction to
UpG (11%), associated with some guanosine forma-
tion (25%) was observed, yielding 51% of 8a.
-I-UpG
-I-UpG
-I-UpG
3
complete decay
89/(75/14)
31/(23/8)
–
2/(2/0)
–
a]
a]
PFu
3
/–
AsPh
AsPh
AsPh
AsPh
3
/–
3
/–
3
/–
3
/–
[
a]
-I-dUpG
[
b]
c]
-I-UpG
-I-UpG
92
>95
[
4
III
As exhibits a higher stability towards oxidation
[a] Reaction performed at 808C, 2 h; [b] reaction performed at 808C, 15 min; [c] reac-
III
than P in phosphine-derived ligands. We therefore
tion performed at 608C, 30 min; [d] based on integration of the UV signal at 254 nm;
[e] guanosine; [f] deglycosylated uracil coupling product 1c.
assumed that Staudinger-type reactions between
[5,18]
phosphines and azides,
which prevent the direct
Considering the incompatibility of nucleic acids
[
15]
with various copper species,
we now tried
Table 2. Scope of the Stille–Migita reaction on unprotected 5-I-UpG.
a copper-free [Pd (dba) ]/AsPh catalytic system. Both
2
3
3
5-I-dUpG and 5-I-UpG were consumed quantitatively,
resulting in near-complete product formation for 5-I-
dUpG and 68% for 5-I-UpG. From this encouraging
starting point, we attempted to further improve the
yield of the desired products by varying tempera-
tures and reaction times. Indeed, a reduced reaction
time of 15 min at 808C resulted in 92% yield of 1b
at essentially suppressed dinucleotide decomposi-
tion, while near-quantitative conversion could be
achieved at 608C within 30 min (Figure 1 and Figur-
es S2 and S3 in the Supporting Information).
To investigate the generality of this approach,
commercially available furan-, pyridine- and N-meth-
ylindole-derived stannanes were subjected to Stille–
Migita coupling on 5-I-UpG (Table 2 and Figure S4 in
the Supporting Information). The catalytically chal-
Stannane
[Pd
AsPh
2
(dba)
3
]/
Yield
[%]
Starting
material [%]
UpG
[%]
Side-products/
G [%]
[
16]
lenging pyridine-derived stannane 2 yielded 23%
of the desired coupling product, while near-quantita-
tive product formation was obtained for stannanes 3
and 4.
[a]
[a]
[a,b]
[c]
[a]
3
[equiv]
2
2
3
4
5
6
7
7
8
9
9
0.5/2
1/4
20
23
>99
98
99
99
8
44
51
47
14
3
–
–
–
–
81
5
<1
51
–
59
63
–
<2
<1
<1
<1
11
7/2
11/3
–
–
–
0.5/2
0.5/2
0.5/2
0.5/2
0.5/2
1/4
0.5/2
0.5/2
1/4
Since standard procedures in RNA oligonucleotide
synthesis are associated with alkaline and acidic de-
protection steps, as well as with exposure toward
fluoride ions, the chemical space for the introduction
of labile functionalities is limited. Yet, it is of consider-
able interest to generate RNAs bearing acid-, base- or
fluoride-labile moieties for further site-specific trans-
–
10/<1
40/3
37/25
<1
12/8
11
<1
6
82
[
3c,17]
[a] Based on integration of the UV signal at 254 nm; [b] product of reduction;
formations.
We therefore synthesized stannanes
[
c] guanosine.
of various aryl dioxolanes (5–7), as well as TMS-pro-
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org
16614
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Enzymatic 5’ incorporation of modified 5-substituted UpG dinu-
cleotides into RNA by T7 RNAP. A 19-nt test RNA was synthesized in vitro in
the presence of C-5-substituted UpG dinucleotides or GMP and UpG as ref-
erence initiators. Open and filled arrows indicate the positions of unmodified
and modified transcripts, respectively. Incorporation efficiencies and relative
RNA yields were normalized to GMP and UpG are indicated below the corre-
sponding lanes.
ly demanding 5-substituents gave lower yields of the desired
transcription product. The highest incorporation efficiency was
observed for 9a, which gave RNA yields similar to unmodified
UpG. Thus, the Stille–Migita coupling products gave access to
a diverse set of site-specifically modified RNA molecules.
The successful coupling on the dinucleotide system prompt-
ed us to investigate the expansion of this methodology to the
functionalization of oligonucleotides. 5-I-dU-, 5-I-dC- or 5-I-U-
bearing DNA and RNA sequences (Figure S9 in the Supporting
Information) were synthesized by phosphoramidite chemistry
and, prior to standard deprotection by ammonia, subjected to
coupling reactions with the corresponding stannanes while
still attached to the solid support (Table 3). 5-I-dU-bearing ON1
was found to couple with stannanes 1, 8 and 9 in yields of 38,
54 and 57%, respectively (Figure S10 in the Supporting Infor-
mation). On alkaline deprotection of the TMS-ethynyl-modified
DNA precursor, the corresponding ethynyl-bearing oligonu-
cleotide was generated, thus providing an efficient alternative
to the direct incorporation of expensive 5-ethynyldeoxyuridine
Figure 1. Optimization of reaction temperature and time for the Stille–
Migita reaction between 5-I-UpG and 1 in the presence of 0.5 equiv
2 3 3
[Pd (dba) ] and 2 equiv AsPh . HPLC analysis revealed near-quantitative cou-
pling product formation at full dinucleotide stability at 608C for 30 min.
incorporation of azide groups in phosphoramidite solid phase
synthesis, could be circumvented in an AsPh -based catalyst
3
system. This would allow for the direct, postsynthetic introduc-
tion of azides into the dinucleotide using an azide-bearing
stannane. Indeed, under the conditions optimized above,
azido-stannane
9 was coupled to 5-I-UpG at 0.5 equiv
[21]
[
Pd (dba) ] loading with 47% efficiency, showing no sign of
phosphoramidites into oligonucleotides. Once more, azido-
stannane 9 was found to be transferred to both terminal and
internal positions of the oligonucleotide (Figures S11 and S12
in the Supporting Information). Coupling yields were found to
be dependent on the accessibility of the iodinated residue,
ranging from 23% for internal to 57% for terminal modifica-
tions. Double modification of ON4 yielded 24% of the target
compound. 5’-modified 5-I-dU or 5-I-U-bearing RNA oligonucle-
otides exhibited similarly efficient coupling properties, yielding
61 and 59% of the respective coupling product (Figure S13 in
the Supporting Information). Internally modified RNA oligonu-
cleotides of identical base sequence to ON2, however, turned
out to be essentially unreactive, most likely due to steric
2
3
azide decomposition (Table 2). An increased catalyst loading of
equiv [Pd (dba) ] furnished the target compound in 82%
1
2
3
yield (Figure S7 in the Supporting Information).
In order to move from small to larger RNAs, HPLC-purified
coupling products of stannanes 1, 3, 5, 6, 8 and 9 with
5
-I-UpG were then subjected to transcriptional priming experi-
[
19]
ments with T7 phage RNA polymerase (RNAP), applying gua-
nosine monophosphate (GMP) and the unmodified dinucleo-
[
11,20]
tide UpG
as reference initiator nucleotides. All dinucleoti-
des were found to be accepted as transcriptional starters by
T7 RNAP (Figure 2, Figure S8 and Table S1 in the Supporting In-
formation) at similar incorporation ratios of 68 to 77%. Sterical-
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org
16615
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
N -CTP was directly synthesized from 5-I-CTP at an
3
Table 3. Synthesis of C-5-substituted DNA and RNA oligonucleotides conflicting with
isolated yield of 58% (Figure S14A in the Supporting
Information). It was found to be a well-accepted sub-
strate for the T7 RNAP, allowing the in vitro synthesis
of internally modified RNA (Figure S14B–D in the
Supporting Information).
standard phosphoramidite synthesis and deprotection chemistry on solid support.
Conclusion
In conclusion, our study explores, for the first time,
the scope and utility of the Stille–Migita reaction on
mono-, di-, and oligonucleotides and the utilization
of the reaction products by DNA and RNA poly-
merases. The unique conditions of the Stille reaction
allow for the direct introduction of sensitive function-
alities that would be incompatible with the tradition-
al procedures of solid phase oligonucleotide synthe-
sis and phosphoramidite chemistry. A careful choice
of the ligand conferred stability onto azide moieties
under coupling conditions as well, providing a novel
and general approach for the synthesis of “clickable”
DNA and RNA by chemical or chemoenzymatic
means. Triorganotin, as well as triorganoarsenic com-
pounds have been subject to debates for their toxici-
ty (see the Supporting Information). Work with these
compound classes should always be carried out with
care under appropriate laboratory conditions. Mild-
[
c]
ON
Sequence
Stannane
Yield [%]
M
calcd
M
found
[
[
[
[
[
[
[
[
[
[
[
a]
a]
a]
a]
a]
a]
a]
a]
a]
b]
b]
1
1
1
2
3
4
5
6
7
8
9
dU*ATAGGAGCT
dU*ATAGGAGCT
dU*ATAGGAGCT
TATAdU*GAGCT
TATAdU*GAGCTCAGCT
dU*AdU*AGGAGCT
C*ATAGGAGCT
1
8
9
9
9
9
9
9
9
9
9
38
54
57
23
25
24
46
23
23
61
59
3183.5474
3075.5433
3146.5918
3121.5691
4645.8348
3227.6250
3145.5949
3120.6018
4644.8202
3262.5019
3278.5196
3183.5488
3075.5440
3146.5775
3121.5853
4645.8106
3227.6222
3145.6083
3120.6030
4644.8507
3262.5153
3278.5102
[d]
TATAC*GAGCT
TATAC*GAGCTCAGCT
dU*AUAGGAGCU
U*AUAGGAGCU
[
a] DNA oligonucleotide; [b] RNA oligonucleotide; [c] based on integration of UV
signal at 254 nm; [d] alkaline deprotection of coupling product results in TMS-depro-
tection and yields the 5-ethynyl-modified conjugate.
shielding by the bulky 2’-O-TBDMS groups of protected RNAs
on solid support. Thus, Stille–Migita chemistry provides an ele-
gant route towards azido-modified oligonucleotides for click
chemistry, a class of substances not easily accessible to solid-
phase synthesis.
ness, reliability and tolerance towards most functional groups
make us expect Stille–Migita chemistry to become a standard
tool for the synthesis of complex oligonucleotides and nucleo-
side triphosphates.
Finally, we investigated whether this chemistry could be
exploited to directly functionalize (deoxy)nucleoside triphos-
Experimental Section
[
22]
phates, a delicate class of compounds that rapidly decom-
poses outside a narrow range of pH, salt and solvent condi-
Stille–Migita coupling on 5-I-dUpG and 5-I-UpG
[
23]
In a 1 mL vial, 5-I-dUpG or 5-I-UpG (1 mmol) and stannane
tions. Recently, Rao et al.
described the synthesis of an
(12 mmol) were dissolved in DMF (250 mL). In a second vial a ten-
azido-modified UTP analogue over six synthetic steps. Apply-
ing a nonprotic solvent system of DMF/DMSO (1:1), Stille cross-
fold-concentrated [Pd (dba) ]/AsPh catalyst solution in DMF
2
3
3
(500 mL) was prepared under alternating vacuum/argon cycles. Pre-
coupling on 5-I-dCTP with 0.5 equiv [Pd (dba) ] and 2 equiv
2
3
mixed catalyst solution (50 mL) was transferred to the degassed re-
action mixture under argon. The reaction was allowed to proceed
at temperatures and times indicated for the individual experiment.
Reactions were partitioned between water (500 mL) and DCM
AsPh at 608C for 45 min gave 5-((E)-4-azidobut-1-enyl)-dCTP
3
(
5-N -dCTP) in only one step in 54% isolated yield (Figure 3A).
3
HPLC analysis indicated clean and quantitative conversion of
the starting material without any side-product formation or
decay (Figure 3B). Spectroscopic properties were found to be
(
500 mL, for acid-sensitive compounds 500 mL buffer A and 500 mL
A-buffered DCM) and the organic layer was extracted with water
2ꢃ500 mL). Reaction products were analyzed by HPLC (gradient
buffer A/buffer B; 0 min: 2% buffer B; 10 min: 2% buffer B; 20 min:
0% buffer B; 45 min: 100% buffer B; 52 min: 2% buffer B). Yields
(
identical to those of 5-N -dCTP synthesized according to the
3
conventional route from 5-N -dC with subsequent triphosphor-
3
1
ylation (Figure 3C). 5-N -dCTP was subjected to PCR experi-
3
were derived by integration of UV signals at 254 nm. For prepara-
tive purpose, coupling products were purified by semipreparative
HPLC (gradient buffer A/buffer B; 0 min: 10% buffer B; 10 min:
25% buffer B; 25 min 60% buffer B; 30 min: 100% buffer B;
ments with the KOD-XL DNA polymerase and 5-N -dCTP as the
3
sole dCTP source (Figure 3D). The amount of amplified DNA in
PCR reactions using 5-N -dCTP was found to be comparable to
3
3
5 min: 10% buffer B; 40 min: 10% buffer B).
that of dCTP-containing reactions. The amplicon could be
biotinylated quantitatively by SPAAC-biotinylation with sulfo-
DBCO-biotin (S11), yielding a DNA-conjugate of higher molecu-
lar weight. On addition of streptavidin, a quantitative gel shift
was observed, proving the reactivity of the azide handle.
Under identical cross-coupling conditions the ribo-analogue 5-
Stille–Migita coupling on solid support
In a 1 mL vial, CPG-bound DNA or RNA, corresponding to about
200 nmol, and stannane (12 mmol) were suspended in 250 mL DMF.
In a second vial [Pd (dba) ] (5.8 mg, 6 mmol) and AsPh (7.8 mg,
2
3
3
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org
16616
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. A) Alternative routes for the synthesis of 5-N
3
-dCTP by Stille–Migita coupling: a) 5 equiv 9, 0.5 equiv [Pd
], 20% AsPh , DMF, 808C, 3 h, 85%; c) 1.2 equiv POCl , PO(OMe) , 08C, 1 h; d) i. 5 equiv (NHBu
, DMF, 08C, 2 h; ii. TEAB pH 7.5, 40%. B) HPLC chromatogram of 5-I-dCTP (upper graph) and reaction mixture after 45 min (lower graph).
2
(dba)
3
], 2 equiv AsPh
3
, DMF/DMSO (1:1),
6
4
08C, 45 min, 54%; b) 1.2 equiv 9, 5% [Pd
.2 equiv NBu
2
(dba)
3
3
3
3
3
)H
2 2 7
P O ,
3
3
1
C) Comparison of P NMR spectra of 5-N
D) PCR-amplification of a 70-mer dsDNA template by KOD-XL DNA polymerase using 5-N
synthetically obtained (**) 5-N -dCTP were applied. 5-N -dC-modified DNA was sequentially SPAAC-biotinylated using sulfo-DBCO-biotin (S11) and incubated
with streptavidin. Stepwise bandshifts indicate successful incorporation of 5-N
3
-dCTP synthesized by Stille coupling on 5-I-dCTP (upper graph) and triphosphorylation of 5-N
3
-dC (lower graph).
3
-dCTP as sole dCTP source. Batches of conventionally (*) and post-
3
3
3
-dCTP and efficient SPAAC-modification of the azide moiety.
2
4 mmol) were dissolved in DMF (500 mL). The catalyst solution was
added and the mixture was incubated for 10 min, followed by ad-
degassed, back-filled with argon and 50 mL (0.6 mmol [Pd (dba) ]
dition of Et
by centrifugation and washed with Et
2
O (1 mL) at RT. A precipitate formed that was collected
O (300 mL). The air-dried
2
3
and 2.4 mmol AsPh ) were transferred to the reaction mixture that
2
3
was subsequently degassed and back-filled with argon again. The
mixture was stirred at 608C for 3 h and diluted to 500 mL with
DMF. The CPG material was recovered by filtration through a centri-
fuge filter and washed with 100 mm EDTA (2ꢃ500 mL, pH 8), DMF
pellet was redissolved in water (200 mL). Analytics was performed
as described above for 5-I-(d)UpG (gradient of buffer A/buffer B;
0 min: 10% buffer B; 10 min: 10% buffer B; 35 min: 35% buffer B;
4
0 min: 100% buffer B; 45 min: 10% buffer B).
(
2ꢃ500 mL) and diethyl ether (2ꢃ200 mL). Oligonucleotides were
deprotected in 28–30% aqueous NH (200 mL) at room tempera-
3
ture for 2 h for 5-I-(d)U-bearing sequences or at 558C for 3 h for 5-
I-dC-modified sequences. The aqueous solutions were recovered
and the CPG material was washed with water (3ꢃ200 mL). The
2’-Deoxy-5-((E)-4-azidobut-1-enyl)cytidine (5-N
-dC)
3
5
-I-dC (95 mg, 0.27 mmol), stannane 9 (125 mg, 0.33 mmol), AsPh3
(
16.5 mg, 55 mmol) and [Pd (dba) ] (12.5 mg, 13.5 mmol) were dis-
2 3
combined aqueous layers were washed with CHCl (200 mL). Exces-
3
solved in anhydrous DMF (5.5 mL). The solution was degassed
under vacuum and back-filled with argon. After being stirred at
sive ammonia was removed by evaporation in a rotary concentra-
tor. RNA oligonucleotides were desilylated in DMF/Et N·3HF (1:1,
3
8
08C for 3 h, the solvent was removed and the residue was sub-
1
20 mL) at 658C for 2 h. Isopropoxytrimethylsilane (240 mL) was
jected to flash column chromatography on a 12 g silica column
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org
16617
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(
gradient DCM/MeOH (1:0) to (7:1)), yielding 73 mg (85%) of the
mixed in anhydrous DMSO (365 mL) in a sealed 1 mL vial under
argon atmosphere. [Pd (dba) ] (13.3 mg, 14.5 mmol) and AsPh
3
1
title compound. R =0.26 (DCM/MeOH 95:5); H NMR (500 MHz,
f
2
3
[
D ]DMSO): d=8.07 (s, 1H, H6), 6.34 (d, J=15.7 Hz, 1H, H7), 6.15 (t,
(17.8 mg, 58 mmol) were dissolved in anhydrous DMF (730 mL) in
a separate vial. The vial was degassed and back-filled with argon.
The catalyst solution (365 mL) was added to the reaction mixture,
followed by purging with argon and stirring at 608C for 45 min.
Analytical HPLC revealed near-quantitative consumption of the
starting material. The reaction mixture was diluted to 4.5 mL in
buffer A and extracted with A-buffered CHCl3 (3ꢃ1.5 mL). The
aqueous phase was freeze-dried and the residue was purified by
reversed-phase chromatography on a 23 g C18 column, gradient
6
J=6.5 Hz, 1H, H1’), 5.86 (dt, J=15.7 Hz, 6.9 Hz, 1H, H8), 5.19 (d,
J=4.3 Hz, 1H, OH-3’), 5.07 (t, J=4.8 Hz, 1H, OH-5’), 4.26–4.21 (m,
1
3
2
1
(
(
3
H, H3’), 3.78 (q, J=3.5 Hz, 1H, H4’), 3.74–3.60 (m, 1H, H5a’), 3.60–
.50 (m, 1H, H5b’), 3.42 (t, J=6.9 Hz, 2H, H10), 2.38 (q, J=7.0 Hz,
H, H9), 2.14 (ddd, J=13.0, 6.0, 3.8 Hz, 1H, H2a’), 2.02 ppm (td, J=
13
3.0, 6.4 Hz, 1H, H2b’); C NMR (126 MHz, [D ]DMSO): d=163.1
6
C4), 154.2 (C2), 137.1 (C6), 125.6 (C8), 123.1 (C7), 103.9 (C5), 87.1
C4’), 85.0 (C1’), 69.9 (C3’), 60.8 (C5’), 49.9 (C10), 40.6 (C2’),
+
2.1 ppm (C9); HR-ESI-TOF-MS: m/z calcd for [M+Na ]
buffer A/buffer B (0–5 min: 0% buffer B; 45 min: 45% buffer B, 1%
1
C H N O Na 345.1282, found 345.1277.
per min). Yield: 7.2 mg, 58%. H NMR (500 MHz, D O): d=7.90 (s,
1
3
18
6
4
1
2
1
1
4
3
H, H6), 6.27 (d, J=15.82 Hz, 1H, H7), 6.18 (dt, J=15.7, 6.74 Hz,
H, H8), 6.03 (d, J=5.1 Hz, 1H, H1’), 4.44 (t, J=5.0 Hz, 1H, H3’),
.40 (t, J=5.2 Hz, 1H, H2’), 4.35–4.18 (m, 3H, H4’ and H5’), 3.56–
.44 (m, 2H, H10), 3.20 (q, J=7.3 Hz, 18H, N-(CH -CH ) ), 2.55–2.51
2
’-Deoxy-5-((E)-4-azidobut-1-enyl)cytidine 5’-O-triphosphate
(5-N -dCTP)
3
2
3 3
1
3
Method A: 5-N -dC (40 mg, 0.125 mmol) was dried at 808C under
high vacuum and suspended in freshly distilled PO(MeO) (315 mL).
The mixture was cooled to 08C and POCl (13.5 mL, 0.15 mmol) was
added, followed by stirring for 2 h at 08C. An ice-cold solution of
(m, 2H, H9), 1.28 ppm (t, J=7.3 Hz, 27H, N-(CH
(126 MHz, D O): d=164.0 (C4), 156.9 (C2), 137.55 (C6), 131.9 (C8),
121.3 (C7), 107.8 (C5), 88.9 (C1’), 83.1 (d, J=9.2 Hz,1C, C4’), 74.0
(C2’), 69.5 (C3’), 65.0 (d, J=5.2 Hz, C5’), 50.1 (C10), 46.6 (N-(CH
-CH ) ); C NMR
2 3 3
3
3
2
3
-
2
31
(
NBu H) H P O (340 mg, 0.62 mmol) and NBu (125 mL, 0.52 mmol)
CH ) ), 32.2 (C9), 8.2 ppm (N-(CH -CH ) ); P NMR (202 MHz, D O,
3 3 2 3 3 2
3
2
2
2
7
3
in anhydrous DMF (1.25 mL) was added and the reaction mixture
was stirred for 1 h at 08C. The reaction was stopped by adding 2m
aqueous TEAB (850 mL, pH 7.5). The solvents were partially re-
moved by threefold co-evaporation with water. The residue was
subjected to ion-exchange chromatography on a Sephadex A25
column (40 mL) in a linear gradient of TEAB in water (pH 7.5,
10 mm phosphate buffer pD=7.0 (d=2.35 ppm)): d=À8.34 (d, J=
19.6 Hz, Pg), À9.56 (d, J=20.2 Hz, Pa), À21.00 ppm (t, J=19.6 Hz,
À
Pb); HR-ESI-TOF-MS: m/z calcd for [MÀH ] C13
H N O P 577.0256,
20 6 14 3
found 577.0266.
5
0 mm to 1.2m over 4 h). Product-containing fractions were freeze- Acknowledgements
dried and purified by reversed phase chromatography on a 23 g
C18 column, gradient buffer A/buffer B (0–5 min: 0% buffer B;
4
Work in the Jꢁschke laboratory is supported by the Helmholtz
Initiative on Synthetic Biology, and by the Biotechnology 2020
program of the BMBF.
5 min: 45% buffer B, 1% per min). Yield: 38 mg, 40%. Method B:
-I-dCTP (14.4 mg, 14.5 mmol) and stannane 9 (28.1 mg, 72.5 mmol)
5
were mixed in anhydrous DMSO (375 mL) in a sealed 1 mL vial
under argon atmosphere. In a separate vial, [Pd (dba) ] (13.3 mg,
2
3
1
4.5 mmol) and AsPh (17.8 mg, 58 mmol) were dissolved in anhy-
3
Keywords: click chemistry · cross-coupling · DNA · palladium ·
drous DMF (750 mL), degassed and back-filled with argon. The cata-
lyst solution (375 mL) was added to the reaction mixture, followed
by purging with argon and stirring at 608C for 45 min. Analytical
HPLC revealed near-quantitative consumption of the starting mate-
rial. The reaction mixture was diluted to 4.5 mL in buffer A and ex-
RNA
[
1] a) R. W. Sinkeldam, N. J. Greco, Y. Tor, Chem. Rev. 2010, 110, 2579–2619;
b) F. Wachowius, C. Hçbartner, ChemBioChem 2010, 11, 469–480; c) S.
Sasaki, K. Onizuka, Y. Taniguchi, Chem. Soc. Rev. 2011, 40, 5698–5706;
d) S. G. Srivatsan, A. A. Sawant, Pure Appl. Chem. 2011, 83, 213–232.
2] a) R. L. Juliano, X. Ming, O. Nakagawa, Acc. Chem. Res. 2012, 45, 1067–
tracted with A-buffered CHCl (3ꢃ1.5 mL). The aqueous phase was
3
freeze-dried and 5-N -dCTP was purified by semipreparative HPLC,
3
[
yielding 7.5 mg (54%) of the title compound as triethylammonium
1
076; b) F. T. M. D. Vicentini, L. N. Borgheti-Cardoso, L. V. Depieri, D. D.
1
salt. H NMR (500 MHz, D O): d=7.88 (s, 1H, H6), 6.32 (t, J=6.9 Hz,
Mano, T. F. Abelha, R. Petrilli, M. V. L. B. Bentley, Pharm. Res. 2013, 30,
2
1
H, H1’), 6.27 (d, J=15.7 Hz, 1H, H7), 6.14 (dt, J=15.6, 6.8 Hz, 1H,
915–931.
H8), 4.68–4.60 (m, 1H, H3’), 4.43–3.97 (m, 3H, H4’, H5’), 3.52–3.48
[3] a) J. Goodchild, Bioconjugate Chem. 1990, 1, 165–187; b) K. W. Hill, J.
Taunton-Rigby, J. D. Carter, E. Kropp, K. Vagle, W. Pieken, D. P. C. McGee,
G. M. Husar, M. Leuck, D. J. Anziano, D. P. Sebesta, J. Org. Chem. 2001,
66, 5352–5358; c) M. A. Podyminogin, E. A. Lukhtanov, M. W. Reed, Nu-
cleic Acids Res. 2001, 29, 5090–5098; d) N. Minakawa, Y. Ono, A. Matsu-
da, J. Am. Chem. Soc. 2003, 125, 11545–11552; e) T. S. Seo, Z. M. Li, H.
Ruparel, J. Y. Ju, J. Org. Chem. 2003, 68, 609–612; f) C. C. Y. Wang, T. S.
Seo, Z. M. Li, H. Ruparel, J. Y. Ju, Bioconjugate Chem. 2003, 14, 697–701;
g) K. Gutsmiedl, C. T. Wirges, V. Ehmke, T. Carell, Org. Lett. 2009, 11,
2405–2408; h) J. Schoch, M. Wiessler, A. Jꢁschke, J. Am. Chem. Soc.
2010, 132, 8846–8847; i) P. van Delft, N. J. Meeuwenoord, S. Hoogen-
doorn, J. Dinkelaar, H. S. Overkleeft, G. A. van der Marel, D. V. Filippov,
Org. Lett. 2010, 12, 5486–5489; j) J. B. Ravnsbaek, M. F. Jacobsen, C. B.
Rosen, N. V. Voigt, K. V. Gothelf, Angew. Chem. Int. Ed. 2011, 50, 10851–
(
m, 2H, H10), 3.19 (q, J=6.9 Hz, 12H, N-(CH -CH ) ), 2.52 (q, J=
2 3 3
6
.2 Hz, 2H, H9), 2.41 (ddd, J=14.1, 6.4, 3.5 Hz, 1H, H2a’), 2.36–2.31
13
(
(
m, 1H, H2b’), 1.27 ppm (t, J=7.3 Hz, 18H, N-(CH -CH ) ); C NMR
126 MHz, D O): d=164.8 (C4), 157.5 (C2), 138.4 (C6), 132.6 (C8),
2 3 3
2
1
6
3
22.2 (C7), 108.4 (C5), 86.6 (C1’), 86.2 (d, J=9.3 Hz, C4’), 69.3 (C3’),
5.9 (d, J=5.3 Hz, C5’), 50.7 (C10), 47.2 (N-(CH -CH ) ), 39.9 (C2’),
2
3 3
31
2.8 (C9), 8.8 ppm (N-(CH -CH ) ); P NMR (202 MHz, D O, 10 mm
2
3 3
2
phosphate buffer pD=7.0 (d=2.35 ppm)): d=À7.37 (d, J=
1
7.6 Hz, Pg), À10.41 (d, J=19.7 Hz, Pa), À21.35 ppm (t, J=18.5 Hz,
À
Pb); HR-ESI-TOF-MS: m/z calcd for [MÀH ] C H N O P 561.0307,
13
20
6
13 3
found 561.0326.
10854; Angew. Chem. 2011, 123, 11043–11046.
5
CTP)
-((E)-4-Azidobut-1-enyl)cytidine 5’-O-triphosphate (5-N3-
[4] a) C. B. Reese, Org. Biomol. Chem. 2005, 3, 3851–3868; b) R. I. Hogrefe,
B. Midthune, A. Lebedev, Isr. J. Chem. 2013, 53, 326–349.
[
5] M. Aigner, M. Hartl, K. Fauster, J. Steger, K. Bister, R. Micura, ChemBio-
Chem 2011, 12, 47–51.
The reaction was carried out as for 5-N -dCTP, method B. 5-I-CTP
3
(
14.1 mg, 13.9 mmol) and stannane 9 (28.1 mg, 72.5 mmol) were
[6] M. Hocek, M. Fojta, Org. Biomol. Chem. 2008, 6, 2233–2241.
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org 16618
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[7] a) D. E. Bergstrom, J. L. Ruth, J. Am. Chem. Soc. 1976, 98, 1587–1589;
b) L. A. Agrofoglio, I. Gillaizeau, Y. Saito, Chem. Rev. 2003, 103, 1875–
[14] V. Farina, S. Kapadia, B. Krishnan, C. J. Wang, L. S. Liebeskind, J. Org.
Chem. 1994, 59, 5905–5911.
1
2
916; c) G. Hervꢀ, G. Sartori, G. Enderlin, G. Mackenzie, C. Len, RSC Adv.
014, 4, 18558–18594.
8] a) A. L. Casalnuovo, J. C. Calabrese, J. Am. Chem. Soc. 1990, 112, 4324–
330; b) L. H. Thoresen, G. S. Jiao, W. C. Haaland, M. L. Metzker, K. Bur-
[15] a) D. S. Sigman, D. R. Graham, V. D’Aurora, A. M. Stern, J. Biol. Chem.
1979, 254, 12269–12272; b) M. K. Stern, J. K. Bashkin, E. D. Sall, J. Am.
Chem. Soc. 1990, 112, 5357–5359.
[16] K. L. Billingsley, S. L. Buchwald, Angew. Chem. Int. Ed. 2008, 47, 4695–
4698; Angew. Chem. 2008, 120, 4773–4776.
[17] a) P. M. E. Gramlich, S. Warncke, J. Gierlich, T. Carell, Angew. Chem. Int.
Ed. 2008, 47, 3442–3444; Angew. Chem. 2008, 120, 3491–3493; b) P.
Kielkowski, H. Macickova-Cahova, R. Pohl, M. Hocek, Angew. Chem. Int.
Ed. 2011, 50, 8727–8730; Angew. Chem. 2011, 123, 8886–8889.
[
4
gess, Chem. Eur. J. 2003, 9, 4603–4610; c) E. C. Western, J. R. Daft, E. M.
Johnson, P. M. Gannett, K. H. Shaughnessy, J. Org. Chem. 2003, 68,
6
2
767–6774; d) P. Capek, R. Pohl, M. Hocek, Org. Biomol. Chem. 2006, 4,
278–2284.
[
9] a) T. Strube, O. Schiemann, F. MacMillan, T. Prisner, J. W. Engels, Nucleo-
sides Nucleotides Nucleic Acids 2001, 20, 1271–1274; b) T. Kottysch, C.
Ahlborn, F. Brotzel, C. Richert, Chem. Eur. J. 2004, 10, 4017–4028; c) A.
Omumi, D. G. Beach, M. Baker, W. Gabryelski, R. A. Manderville, J. Am.
Chem. Soc. 2011, 133, 42–50; d) L. Wicke, J. W. Engels, Bioconjugate
Chem. 2012, 23, 627–642; e) H. Cahovꢄ, A. Jꢁschke, Angew. Chem. Int.
Ed. 2013, 52, 3186–3190; Angew. Chem. 2013, 125, 3268–3272; f) L.
Lercher, J. F. McGouran, B. M. Kessler, C. J. Schofield, B. G. Davis, Angew.
Chem. Int. Ed. 2013, 52, 10553–10558; Angew. Chem. 2013, 125, 10747–
[
18] a) J. Steger, D. Graber, H. Moroder, A. S. Geiermann, M. Aigner, R.
Micura, Angew. Chem. Int. Ed. 2010, 49, 7470–7472; Angew. Chem.
2
010, 122, 7632–7634; b) T. Santner, M. Hartl, K. Bister, R. Micura, Bio-
conjugate Chem. 2014, 25, 188–195.
[
19] a) B. Seelig, A. Jꢁschke, Bioconjugate Chem. 1999, 10, 371–378; b) R.
Fiammengo, K. Musilek, A. Jꢁschke, J. Am. Chem. Soc. 2005, 127, 9271–
9
276.
20] C. Pitulle, R. G. Kleineidam, B. Sproat, G. Krupp, Gene 1992, 112, 101–
05.
21] a) D. Graham, J. A. Parkinson, T. Brown, J. Chem. Soc. Perkin Trans.
1998, 1131–1138; b) J. Gierlich, G. A. Burley, P. M. E. Gramlich, D. M.
[
[
1
10752.
[
10] a) M. Kosugi, Y. Shimizu, T. Migita, Chem. Lett. 1977, 6, 1423–1424; b) D.
Milstein, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 3636–3638; c) J. K.
Stille, Angew. Chem. Int. Ed. Engl. 1986, 25, 508–524; Angew. Chem.
1
Hammond, T. Carell, Org. Lett. 2006, 8, 3639–3642.
22] J. Dadovꢄ, P. Vidlꢄkovꢄ, R. Pohl, L. Havran, M. Fojta, M. Hocek, J. Org.
Chem. 2013, 78, 9627–9637.
[
[
1
2
986, 98, 504–519; d) P. Espinet, A. M. Echavarren, Angew. Chem. Int. Ed.
004, 43, 4704–4734; Angew. Chem. 2004, 116, 4808–4839.
23] H. Rao, A. A. Sawant, A. A. Tanpure, S. G. Srivatsan, Chem. Commun.
[
[
[
11] A. Samanta, A. Krause, A. Jꢁschke, Chem. Commun. 2014, 50, 1313–
316.
2
012, 48, 498–500.
1
12] a) V. Farina, S. I. Hauck, Synlett 1991, 157–159; b) H. F. Song, X. H. Li,
Y. T. Long, G. Schatte, H. B. Kraatz, Dalton Trans. 2006, 4696–4701.
13] V. Farina, B. Krishnan, J. Am. Chem. Soc. 1991, 113, 9585–9595.
Received: August 13, 2014
Published online on October 16, 2014
Chem. Eur. J. 2014, 20, 16613 – 16619
www.chemeurj.org
16619
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim