Synthesis of modified DNA by PCR with alkyne-bearing purines followed by a click reaction

Article abstract of DOI:10.1021/ol7026015

(Chemical Equation Presented) Alkyne-bearing deazapurine triphosphates were prepared and successfully incorporated into DNA using the polymerase chain reaction (PCR). The obtained alkyne-labeled DNA was successfully used in a click reaction with galactose

Full text of DOI:10.1021/ol7026015

ORGANIC
LETTERS
2008
Vol. 10, No. 2
249-251
Synthesis of Modified DNA by PCR with
Alkyne-Bearing Purines Followed by a
Click Reaction
Philipp M. E. Gramlich, Christian T. Wirges, Johannes Gierlich, and
Thomas Carell*
Department of Chemistry, Ludwig-Maximilians UniVersity Munich, Butenandtstr. 13,
D-81377 Munich, Germany
thomas.carell@cup.uni-muenchen.de
Received October 26, 2007
ABSTRACT
Alkyne-bearing deazapurine triphosphates were prepared and successfully incorporated into DNA using the polymerase chain reaction (PCR).
The obtained alkyne-labeled DNA was successfully used in a click reaction with galactose azide.
The utilization of DNA for the assembly of functional
materials has received considerable interest in recent years.1
To this end, modified DNA was created featuring a multitude
of novel functions which, for example, allows one to
accelerate the detection of an analyte,2 to sequence DNA,3
to create DNA- and RNA-based receptors (aptamers),4 or to
assemble them into novel nanomaterials.5 In order to
introduce the modifications in a highly controllable manner
in long DNA strands, the exchange of the natural triphos-
phates by chemically modified building blocks in enzymatic
reactions such as the PCR reaction is the most common
approach. This method is, however, not generally applicable
because it cannot be predicted if and how a specific
triphosphate is accepted by different DNA polymerases.6,7
We therefore decided to develop a two-step procedure in
which a nucleoside precursor is incorporated into DNA by
PCR, which can be subsequently converted into a variety of
derivatives. For a similar approach using the Staudinger
ligation, see ref 8. For the functionalization step, we exploited
the recently described Cu-catalyzed Huisgen-Meldal-
Sharpless reaction,9 termed click reaction. We already used
this chemistry to efficiently modify DNA-bearing alkyne
functionalities at thymidines or cytidines.2,10,11 We herein
* To whom correspondence should be addressed. Fax: +49 (0)89 2180
77756. Tel: +49 (0)89 2180 77750.
(1) (a) Condon, A. Nat. ReV. Genet. 2006, 7 (7), 565-575. (b) Feldkamp,
U.; Niemeyer, C. M. Angew. Chem., Int. Ed. 2006, 45 (12), 1856-1876.
(2) Hammond, D. M.; Manetto, A.; Gierlich, J.; Azov, V. A.; Gramlich,
P. M. E.; Burley, G. A.; Maul, M.; Carell, T. Angew. Chem., Int. Ed. 2007,
46 (22), 4184-4187.
(3) Seo, T. S.; Bai, X.; Kim, D. H.; Meng, Q.; Shi, S.; Ruparel, H.; Li,
Z.; Turro, N. J.; Ju, J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102 (17), 5926-
5931.
(4) Sidorov, A. V.; Grasby, J. A.; Williams, D. M. Nucleic Acids Res.
2004, 32 (4), 1591-1601.
(5) (a) Braun, E.; Eichen, Y.; Silvan, U.; Ben-Yoseph, G. Nature 1998,
391, 775-778. (b) Keren, K.; Krueger, M.; Gilad, R.; Ben-Yoseph, G.;
Sivan, U.; Braun, E. Science 2002, 297 (5578), 72-75. (c) Gu, Q.; Cheng,
C.; Gonela, R.; Suryanarayanan, S.; Anabathula, S.; Dai, K.; Haynie, D. T.
Nanotechnology 2006, 17, R14-R25.
(6) Ja¨ger, S.; Rasched, G.; Kornreich-Leshem, H.; Engeser, M.; Thum,
O.; Famulok, M. J. Am. Chem. Soc. 2005, 127 (43), 15071-15082.
(7) (a) Kuwahara, M.; Nagashima, J.-i.; Hasegawa, M.; Tamura, T.;
Kitagata, R.; Kazuo, H.; Hososhima, S.-i.; Kasamatsu, T.; Ozaki, H.; Sawai,
H. Nucleic Acids Res. 2006, 34 (19), 5383-5394. (b) Gourlain, T.; Sidorov,
A.; Mignet, N.; Thorpe, S. J.; Lee, S. E.; Grasby, J. A.; Williams, D. M.
Nucleic Acids Res. 2001, 29 (9), 1898-1905.
(8) (a) Wang, C. C. Y.; Seo, T. S.; Li, Z.; Ruparel, H.; Ju, J. Bioconjugate
Chem. 2003, 14 (3), 607-701. (b) Weisbrod, S. H.; Marx, A. Chem.
Commun. 2007, 18, 1828-1830.
10.1021/ol7026015 CCC: $40.75
© 2008 American Chemical Society
Published on Web 12/13/2007

wish to report the extension of the technology to the purine
bases dA and dG.
7-iodo-2′-deoxyguanosine 4 was synthesized in 10 steps
according to literature procedures.14 Sonogashira coupling
of 4 with 1-TMS-protected 1,7-octadiyne gave the key
intermediate 5 in good yield. After deprotection of the alkyne
and cleavage of the sugar protecting groups to 6, the
monomer was successfully converted into triphosphate 7,15
suitable for enzymatic incorporation.
In order to limit structural perturbation of the DNA duplex
to a minimum, we placed the alkyne group in the major
groove.6,12 To achieve this, we prepared the alkyne-bearing
derivatives of the 7-deazapurine bases 7-deaza-dA and
7-deaza-dG (Figure 1). The adenine building block was
To examine how the triphosphates 3 and 7 are accepted
by polymerases, we first performed primer extension studies.
To this end, a fluorescently labeled primer was hybridized
to a 30mer template. The elongation of the primer was
followed using PAGE. In the initial test runs, we used two
different templates containing each of the canonical bases
at least one time.16 As expected from literature precedents,6
the unnatural alkyne-bearing purines 3 and 7 were able to
fully replace the canonical bases dA and dG in typical primer
extension experiments. In further experiments, we could also
replace the nucleobases dT and dC by the alkyne-modified
bases 9 and 11, allowing us to prepare a DNA derivative in
which each nucleobase carries an alkyne-bearing functional
group. In all cases, we obtained a clean, full-length product,
proving that all triphosphates were efficiently incorporated
(Supporting Information).
Encouraged by this result, we started to incorporate our
building blocks 3 and 7 using PCR. To this end, we used a
300mer template DNA to create two modified DNA strands,
DNA300‚2 (dATP replaced by 3) and DNA300‚6 (dGTP
replaced by 7). Detection of the PCR products obtained with
the building block 7 turned out to be difficult because the
ethidium bromide fluorescence is typically quenched by
deazapurine derivatives.17 We found that the PCR product
DNA300‚2, however, was stainable with ethidium bromide
and also with SYBR green. Detection of DNA300‚6 required
the use of a fluorescein-labeled primer strand. With the help
of the KOD XL polymerase, we could indeed prepare
DNA300‚2 and DNA300‚6 efficiently without the use of any
additives (Figure 2). The obtained amplicons comprise 170
(DNA300‚2) or 103 (DNA300‚6) alkynes attached to the
purine bases. Simple screenings of the different temperature
steps (annealing, elongation, and denaturation) of the PCR
cycle were sufficient to achieve this goal. An attempted PCR
incorporation of all four alkyne-modified triphosphates into
a 300mer DNA unfortunately failed. In order to investigate
the structure of the modified DNA in more detail, we
performed UV and CD measurements (Supporting Informa-
tion). Both DNA strands exhibited an additional absorption
band at longer wavelengths due to conjugation of the
nucleobase to the internal alkyne. This observation is in
accord with previous results obtained by us.11 This additional
Figure 1. Syntheses of the triphosphates 3 and 7. The alkyne-
bearing pyrimidine triphosphates 9 and 11 as well as the galactose
azide 12 were reported previously.
synthesized by a Sonogashira coupling on the unprotected
7-deaza-7-iodo-2′-deoxyadenosine 1 to give the free nucleo-
side 2.13,14 The synthesis of the triphosphate 3 was achieved
by the method of Kovacs et al.15 The corresponding 7-deaza-
(9) (a) Huisgen, R. 1,3-Dipolar Cycloaddition Chemistry; Wiley: New
York, 1984; pp 1-176. (b) Tornoe, C. W.; Christensen, C.; Meldal, M. J.
Org. Chem. 2002, 67 (9), 3057-3064. (c) Kolb, H. C.; Finn, M. G.;
Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004-2021.
(10) (a) Gierlich, J.; Burley, G. A.; Gramlich, P. M. E.; Hammond, D.
M.; Carell, T. Org. Lett. 2006, 8 (17), 3639-3642. (b) Burley, G. A.;
Gierlich, J.; Mofid, M. R.; Nir, H.; Tal, S.; Eichen, Y.; Carell, T. J. Am.
Chem. Soc. 2006, 128 (5), 1398-1399.
(13) (a) Seela, F.; Zulauf, M. Synthesis 1996, 726-730. (b) Seela, F.;
Sirivolu, V. R., Chem. BiodiVersity 2006, 3, 509-514.
(14) (a) Davoll, J. J. Chem. Soc. 1960, 1960, 130-138. (b) Buhr, C. A.;
Wagner, R. W.; Grant, D.; Froehler, B. C. Nucleic Acids Res. 1996, 24
(15), 2974-2980.
(15) Kovacs, T.; O¨ tvo¨s, L. Tetrahedron Lett. 1988, 29 (36), 4525-4528.
(16) Held, H. A.; Benner, S. A. Nucleic Acids Res. 2002, 30 (17), 3857-
3869.
(11) Gierlich, J.; Gutsmiedl, K.; Gramlich, P. M. E.; Schmidt, A.; Burley,
G.; Carell, T. Chem.sEur. J. 2007, 13, 9486-9494.
(17) (a) Li, H.; Seela, F. Nucleosides Nucleotides 2005, 24 (5-7), 865-
868. (b) Weiss, J.; Zucht, H.-D.; Forssmann, W.-G. PCR Methods Appl.
1994, 4, 124-125. (c) Latimer, L. J. P.; Lee, J. S. J. Biol. Chem 1991, 266
(21), 13849-13851.
(12) Capek, P.; Cahova´, H.; Pohl, R.; Hocek, M.; Gloeckner, C.; Marx,
A. Chem.sEur. J. 2007, 13, 6196-6203.
250
Org. Lett., Vol. 10, No. 2, 2008

by HPLC, and the modified bases 2 and 6 were co-injected.
These spectra show that the DNA indeed contains the
modified purine bases 2 and 6. Only a small amount of the
unsubstituted purine nucleosides dA and dG were detected,
which derive from the commercial primers that were used
for the PCR reactions. This result proves that our triphos-
phates were efficiently incorporated into 300DNA‚2 and
300DNA‚6. Further proof for the presence of the modified
bases was obtained from ESI-MS/MS and UV spectra of the
eluting compounds.
Having achieved the successful incorporation of 3 and 7
into the PCR amplicons, we wanted to know if the modified
DNA features sequence changes. To this end, we used
300DNA‚2 and 300DNA‚6 as templates for a PCR reaction
with the natural dNTPs and either KOD XL or Deep Vent
exo- polymerase. The resulting amplicons were sequenced
(Supporting Information). We observed that the 300DNA‚
N strands retained the original sequence, proving that the
modified bases do not reduce the polymerase fidelity in our
PCR reaction but are faithfully copied.
Figure 2. Five percent PAGE of the PCR amplicons 300DNA‚2
(lane 2) and 300DNA‚6 (lane 3). Lane 1 contains a 100 bp DNA
ladder.
band is also visible in the CD spectra, which exhibit an extra
minimum at 310 nm. The overall shape of the curve is similar
to a typical B-form DNA, which is an indication that both
DNA strands DNA300‚2 and DNA300‚6 adopt a B-type
conformation as duplexes.
In order to prove the presence of the modifications in the
DNA strands, we performed enzymatic digests of 300DNA‚
N (unmodified), 300DNA‚2, and 300DNA‚6 (Figure 3). The
In order to prove the proficiency of the alkyne-containing
DNA in the click reaction, we used DNA‚2 and DNA‚6 as
a reaction partner in the Cu(I)-catalyzed click reaction with
galactose azide 12. CuBr and TBTA ligand18 were added to
perform the reaction. The reaction products were purified
from excess reactants by simple ethanol precipitation and
were digested enzymatically. HPLC chromatography of the
resulting nucleoside mixtures combined with UV and MS/
MS analysis clearly showed the quantitative yield of the
postsynthetic click functionalization, which proceeds without
formation of any detectable side product. No traces of
unreacted alkynes could be found (Figure 3).
In summary, we present novel alkyne-bearing purines,
which can be incorporated into DNA via a PCR reaction.
These building blocks act as selective reaction partners in
the click reaction with azides. Quantitative conversion of
more than 100 alkynes in DNA was again observed.
Acknowledgment. We thank BASF AG and the Deutsche
Forschungsgemeinschaft (SFB 486) for financial support.
P.M.E.G. thanks the Cusanuswerk for a Ph.D. scholarship.
C.T.W. and J.G. thank the Fonds of the Chemical Industry
for Ph.D. scholarships. We are indebted to Markus Spallek,
Ana Varja, Veronika Ehmke, Markus Horn. and Simon
Ku¨nzel from the LMU Munich for their synthetic contribu-
tions.
Figure 3. HPLC chromatograms of the enzymatic digest of
DNA300‚2 (top) and DNA300‚6 (bottom) before (left) and after
(right) the click reaction with galactose azide 12. The inset shows
the HPLC trace at 310 nm. Deoxyinosine (dI) is formed by
deamination of dA during the enzymatic digest. The click products
form double peaks due to anomerization of the sugar moiety.
Supporting Information Available: Detailed experi-
mental protocols and spectral data for all compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL7026015
DNA was treated with nuclease P1, calf spleen phosphodi-
esterase, snake venom phosphodiesterase, and alkaline phos-
phatase. The obtained mixture of nucleosides was separated
(18) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Org. Lett.
2004, 6 (17), 2853-2855.
Org. Lett., Vol. 10, No. 2, 2008
251