Synthesis of oligonucleotides with a 2'-cap at the 3'-terminus via reversed phosphoramidites.

Article abstract of DOI:10.1021/ol020212w

[reaction: see text] A method is presented for the synthesis of single compounds or small combinatorial libraries of oligonucleotides with 2'-acylamido-2'-deoxyuridine residues at the 3'-terminus. Selection experiments identified the residue of anthraquin

Full text of DOI:10.1021/ol020212w

ORGANIC
LETTERS
2003
Vol. 5, No. 3
247-250
Synthesis of Oligonucleotides with a
2′-Cap at the 3′-Terminus via Reversed
Phosphoramidites
William H. Connors,†,‡ Sukunath Narayanan,†,§ Olga P. Kryatova,‡ and
,†,‡,§
Clemens Richert*
Departments of Chemistry, UniVersity of Constance, 78457 Konstanz, Germany,
Tufts UniVersity, Medford, Massachusetts 02155, and Institute of Organic Chemistry,
UniVersita¨t Karlsruhe (TH), D-76131 Karlsruhe, Germany
cr@rrg.uka.de
Received October 17, 2002
ABSTRACT
A method is presented for the synthesis of single compounds or small combinatorial libraries of oligonucleotides with 2′-acylamido-2′-
deoxyuridine residues at the 3′-terminus. Selection experiments identified the residue of anthraquinone-2-carboxylic acid as a “molecular cap“
that increases the UV melting point of the duplex (5′-ACGCGU-3′)2 by up to 28 °C compared to the unmodified control duplex.
Oligonucleotides with modifications at the termini play
important roles in molecular biology, biotechnology, and the
emerging field of molecular medicine. Among the most
common non-nucleic acid moieties appended to termini are
chromophores such as the “big dyes” used to label strands
produced in the chain terminator method of DNA sequenc-
ing,1 cyanine dyes for labeling strands employed in DNA
chip experiments,2 and the fluorophore/quencher combina-
tions employed in molecular beacons.3 Further, linkers for
the immobilization of oligonucleotides on flat surfaces are
central to the preparation of oligonucleotide microarrays.4
Non-nucleic acid moieties appended to the termini can also
be used to increase the affinity for target strands5 or modulate
the biological fate of oligonucleotides.6 Even mRNAs have
been terminally modified to allow conjugation to proteins.7
Recently, it was demonstrated that carboxylic acid residues
directly appended to an oligonucleotide with a terminal
aminodeoxynucleotide can have substantial effects on the
stability of DNA duplexes.8 Work in these laboratories has
focused on 5′-modified oligonucleotides, but oligonucleotides
with a 3′-terminal 2′-deoxy-2′-nalidixoylamidouridine residue
(1NA, Figure 1) have also been described.9 The duplex (5′-
Figure 1.
* To whom correspondence should be addressed at his Karlsruhe address.
Phone: 49-(0)721-608 2091. Fax: 49 (0) 721 608 4825.
† University of Constance.
ACGCGU-NA-2′)2, where NA is a nalidixic acid residue,
has a UV melting point (Tm) up to 22 °C higher than that of
unmodified control duplexes (ACGCGU)2 and (ACGCGT)2.
This melting point increase is lower than that of the best
‡ Tufts University.
§ Universita¨t Karlsruhe (TH).
10.1021/ol020212w CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/15/2003

Scheme 1a
a Reaction conditions: (a) Alloc-Phe-OH, HBTU, DIEA, DMF; (b) DMT-Cl, DMAP, TEA, pyridine; (c) succinic anhydride, DMAP,
pyridine; (d) LCAA-cpg, HBTU, HOBT, DIEA, DMF; (e) Pd(PPh3)4, [Et2NH2]+HCO3-, PPh3, CH2Cl2; (f) R-CO2H, HBTU, HOBT, DIEA,
DMF; (g) DNA synthesis via 3′-phosphoramidite protocol; (h) NH4OH.
5′-cap identified to date, where a ∆Tm of up to 27.8 °C has
been realized.10 Therefore, it was decided to search for 3′-
terminal uridine residues of general structure 2 (Figure 1),
where the residue R provides greater duplex stabilization than
the residue of nalidixic acid. This search was to involve
combinatorial synthesis and spectrometrically monitored
selection assays (SMOSE).21 Here we report on a new route
to oligonucleotides with 2′-acylamido-2′-deoxyuridine resi-
dues at their 3′-termini that allows for combinatorial syn-
theses and on the identification of a residue with a greater
duplex-stabilizing effect than nalidixic acid.
2′-Acylamidonucleotides have previously been incorpo-
rated in oligoribonucleotides and oligodeoxyribonucleotides.9-14
Two routes to small combinatorial libraries of oligonucleo-
tides with 3′-terminal residues of general structure 2 were
tested (Schemes 1-3). Both start from 2′-amino-2′-deoxy-
uridine (3).15-17 The first route (Scheme 1) produced oligo-
nucleotides where a phenylalanine residue links the acid
“cap” to the aminouridine and employed conventional 3′-
phosphoramidites for the unmodified portion of the DNA.
The phenylalanine residue was inserted to avoid O-to-N acyl
migration of the succinyl linker to the solid support. Such a
migration could occur via a five-membered intermediate
similar to that of ribonuclease-catalyzed hydrolysis of RNA
strands18 during the coupling of activated carboxylic acids.
The route, involving protected nucleoside 4, amino-
acylated solid support 5, and support-bound coupling product
6, produced modified strands of general structure 7R, where
R were the residues of nalidixic acid (NA) and an-
thraquinone-2-carboxylic acid (AQ) (Figure 1 and Scheme
3). Attempts to prepare small combinatorial libraries via
mixed couplings21 by this route were thwarted by massive
side reactions discovered in the MALDI-TOF mass spectra
of crudes. These were presumably due to the low reactivity
of the amino groups located close to the surface of the solid
support and side reactions during DNA synthesis following
the installation of the acid residues.
Therefore, an alternative route employing “reversed phos-
phoramidites”, i.e., 5′-phosphoramidites,19 was established.
This synthesis involves chain assembly from the 5′- to the
3′-terminus of the DNA strand (Schemes 2 and 3). It places
the amino group of the 2′-amino-2′-deoxyuridine residue at
the distal end of the DNA chain attached to the solid support,
ensuring high reactivity in solid-phase-based couplings to
the amine and avoids side reactions with phosphoramidites
during chain extension. The route required the synthesis of
5′-phosphoramidite with 2′,3′-protection of 2′-amino-2′-
deoxyuridine (3). Several protection schemes were tested,
including one where the 5′-hydroxyl was first protected with
a dimethoxytrityl group, followed by installation of Alloc
groups on the 3′-hydroxyl and 2′-amino groups. This route
proved to be low-yielding because dimethoxytritylation of
3 was not sufficiently selective and resulted in concomitant
protection of the 2′-amino group. Even the silylation of the
5′-position of 3 with a TBDPS group as a first step was
unsatisfactory due to a lack of regio- and chemoselectivity.
Best results were obtained with the route shown in Scheme
2. It involved 2′-protection of 3 with an Fmoc-group to give
8, and N,O-acetal formation to give 9. Phosphitylation of
this double-protected nucleoside gave 10 in 54% overall yield
from 3.
Starting from commercial support 11 and using 5′-
phosphoramidites, chain assembly via 12 allowed incorpora-
tion of 10 in oligonucleotides in high yield, producing 13,
the isopropylidene group of which was made acid labile by
removal of the Fmoc group with piperidine (Scheme 3).
Scheme 2a
a Reaction conditions: (a) Fmoc-OSu, DMF, 74%; (b) 2,2-dimethoxypropane, CSA, acetone, 96%; (c) NCC2H4O-P(NiPr2)2, DIPAT,
CH3CN, 76%.
248
Org. Lett., Vol. 5, No. 3, 2003

Scheme 3a
a Reaction conditions: (a) DNA synthesis with reverse phosphoramidites; (b) coupling cycle with 10; (c) piperidine, DMF; (d) AcOH,
H2O; (e) R-CO2H, HBTU, HOBT, DIEA, DMF; (f) NH4OH; (g) carboxylic acid mixture, HBTU, HOBT, DIEA, DMF.
Treatment with 20% aqueous acetic acid for 15 min gave
amino alcohol 14 without residual 2′/3′-protected species,
as detected in MALDI-TOF mass spectra of deprotected
crudes. Coupling of activated carboxylic acids to 14 pro-
ceeded in high yield, generating compounds of general
structure 15R, whose deprotected crudes again showed no
uncoupled DNA hexamers. This encouraged us to perform
combinatorial couplings with small reactivity-adjusted mix-
tures of carboxylic acids.20 The mixtures employed are shown
in Figure S9 (Supporting Information). When 14 was allowed
to react with these mixtures of activated carboxylic acids,
seven libraries of general structure chemset 1 were formed,
deprotection of which with NH4OH gave chemsets 2. The
MALDI-TOF mass spectra of the crude libraries are shown
in Figures S1-7 (Supporting Information).
The crude oligonucleotide libraries of general structure
chemset 2 were then employed in nuclease survival selection
experiments.21 These subject small chemical libraries to the
Table 1. UV Melting Points and Hyperchromicities of DNA Duplexes
Tm
(°C)b
∆Tm
(°C)c
hyperchromicity
(%)b
buffer
150 mM NH4OAc
duplexa
ref
(ACGCGT)2 (17)2
33.6 ( 1.0
48.0 ( 0.9
48.1 ( 0.8
50.9 ( 1.0
53.6 ( 0.5
33.9 ( 0.7
34.1 ( 0.8
50.5 ( 2.0
49.6 ( 0.2
52.0 ( 0.3
56.2 ( 2.7
20.6d
9.0 ( 0.3
16.5 ( 0.4
18.3 ( 1.0
11.5 ( 0.6
13.5 ( 0.9
8.5 ( 0.2
9.0 ( 0.2
13.7 ( 1.9
16.5 ( 0.7
9.5 ( 0.9
19.1 ( 1.4
6.3d
9
9
(ACGCGU-NA)2 (16NA)2
(ACGCGU-Phe-NA)2 (7NA)2
(ACGCGU-Phe-AQ)2 (7AQ)2
(ACGCGU-AQ)2 (16AQ)2
(ACGCGT)2 (17)2
+14.4
+14.5
+17.3
+20.0
this work
this work
this work
1000 mM NH4OAc
9
9
9
(ACGCGU)2 (10)2
+0.2
(ACGCGU-NA)2 (16NA)2
(ACGCGU-Phe-NA)2 (7NA)2
(ACGCGU-Phe-AQ)2 (7AQ)2
(ACGCGU-AQ)2 (16AQ)2
(ACGCGT)2 (17)2
(ACGCGU-NA)2 (16NA)2
(ACGCGU-AQ)2 (16AQ)2
(ACGCGT)2 (17)2
(ACGCGU-NA)2 (16NA)2
(ACGCGU-AQ)2 (16AQ)2
(ACGCGT)2 ((17)2
(ACGCGU-NA)2 (16NA)2
(ACGCGU-AQ)2 (16AQ)2
+16.6
+15.7
+18.1
+22.3
this work
this work
this work
10 mM phosphate buffer
9
9
35.2d
+14.6
+27.4
16.6d
48.0 ( 0.4
31.7 ( 0.7
52.6 ( 1.3
56.3 ( 0.2
31.1 ( 0.3
53.1 ( 0.3
59.2 ( 0.3
9.4 ( 0.4
7.9 ( 2.0
17.5 ( 1.7
8.6 ( 1.3
6.6 ( 1.4
16.7 ( 0.7
9.8 ( 0.2
this work
150 mM NaCl, 10 mM phosphate buffer
1 M NaCl, 10 mM phosphate buffer
9
9
+20.9
+24.6
this work
9
9
+22.0
+28.1
this work
a Sequences are given from the 5′- to the 3′-terminus. b Average of four melting points ( SD at a 3.5 ( 0.6 µM strand concentration. c Melting point
difference to control strand. d Average of two melting curves.
Org. Lett., Vol. 5, No. 3, 2003
249

attack of single-strand-specific nucleases. In the present case,
calf spleen phosphodiesterase (E.C. 3.1.16.1) was employed,
which digests single-stranded oligonucleotides from the 5′-
terminus, i.e., the terminus where the strands are unmodified.
Modified strands with 2′-appendages strongly stabilizing their
duplexes could be expected to be in the single-stranded state
to the smallest extent, thus resisting nuclease attack the
longest. To specifically select for strands with “caps” better
than the known nalidixic acid residue, 1 equiv of compound
16NA was added to every library. The MALDI-TOF mass
spectra of mixtures undergoing nuclease selection were
screened for peaks of full length oligonucleotides surviving
longer than 16NA. While the libraries produced with
mixtures 1-6 did not yield compounds exceeding 16NA in
lifetime, the selection of the library prepared with mixture 7
did show that 16AQ was a “lone survivor” (Figure S8,
Supporting Information).
Accordingly, 16AQ was synthesized individually, HPLC
purified, and subjected to UV melting experiments under
various buffer conditions (Table 1). Compared to control
duplex (ACGCGT)2 (17)2, that of lead compound
(ACGCGU-NA)2 (16NA)2, and those of the two phenyl-
alanine-linked hybrids (ACGCGU-Phe-NA)2 (7NA)2 and
(ACGCGU-Phe-AQ)2 (7AQ)2, the duplex of 16AQ melted
at higher temperatures under all conditions tested. The
melting point increase (∆Tm) was greatest in phosphate
buffered saline solution, reaching +28.1 or +14 °C per
modification at 1 M NaCl and 10 mM phosphate buffer. This
is the highest melting point increase observed to date for an
acylamido/deoxy-modified DNA hexamer, slightly above the
27.8 °C realized for a 5′-appended quinolone, which reached
this value at 10 mM salt, i.e., conditions less favorable for
hybridizations on DNA chips.22,23
It is interesting to note that the “winner cap” identified in
this exploratory combinatorial study has been tested as a 5′-
cap in our earlier work21b and did not come up as a strongly
stabilizing residue. Nor did cholic acid, the residue of which
is known to tightly bind to terminal base pairs when
appended to the 5′-terminus as an acylamido substituent,8b
produce a hit, suggesting that the duplex-stabilizing effect
of the anthraquinone carboxylic acid residue is not an
unspecific effect due to its hydrophobicity. Modeling and
force field minimizations performed with an anthracene
carboxylic acid moiety suggest that a residue of this shape
can stack on the terminal base pair formed by the aminode-
oxyuridine without disruption of base pairing.24
In conclusion, the results presented here show that
oligonucleotides with 2′-acylamido-2′-deoxyuridine residues
at the 3′-terminus can be prepared via a short and rugged
route allowing for the generation of small combinatorial
libraries. Anthraquinone carboxylic acid as the building block
for the acyl portion of a modified hexamer leads to
significantly enhanced duplex stability.
Acknowledgment. This work was supported by Deutsche
Forschungsgemeinschaft, Grants RI 1063/1-2 and FOR 434.
The authors thank Jan Rojas Stu¨tz for help with the acquisi-
tion of MALDI-TOF mass spectra, Siegfried Herzberger for
synthetic support, and A. Friemel for the acquisition of NMR
spectra.
(1) Rosenblum, B. B.; Lee, L. G.; Spurgeon, S. L.; Khan, S. H.; Menchen,
S. M.; Heiner, C. R.; Chen, S. M. Nucleic Acids Res. 1997, 25, 4500-
4504.
(2) Hedge, P.; Qi, R.; Abernathy, K.; Gay, C.; Dharap, S.; Gaspard, R.;
Hughes, J. E.; Snesrud, E.; Lee, N.; Quackenbush, J. BioTechniques 2000,
29, 548-563.
(3) (a) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303-308.
(b) Kuhn, H.; Demidov, V. V.; Coull, J. M.; Fiandaca, M. J.; Gildea, B.
D.; Frank-Kamenetskii, M. D. J. Am. Chem. Soc. 2002, 124, 1097-1103.
(4) (a) Maskos, U.; Southern, E. M. Nucleic Acids Res. 1992, 20, 1679-
1684. (b) Lamture, J. B.; Beattie, K. L.; Burke, B. E.; Eggers, M. D.; Ehrlich,
D. J.; Fowler, R.; Hollis, M. A.; Kosicki, B. B.; Reich, R. K.; Smith, S. R.
Nucleic Acids Res. 1994, 22, 2121-2125. (c) Pirrung, M. C.; Odenbaugh,
A. L.; Davis, J. D. Langmuir 2000, 16, 2185-2191.
(5) Montenay-Garestier, T.; Sun, J. S.; Chomilier, J.; Mergny, J. L.;
Takasugi, M.; Asseline, U.; Thuong, N.; Rougee, M.; Helene, C. In
Molecular Basis of Specificity in Nucleic Acid-Drug Interactions; Pullman,
B., Jortner, J., Eds.; Kluwer Academic Publishers: Norwell, MA, 1990;
pp 275-290.
(6) Manoharan, M.; Tivel, K. L.; Condon, T. P.; Andrade, L. K.; Barber-
Peoch, I.; Inamati, G.; Shah, S.; Mohan, V.; Graham, M. J.; Bennett, C. F.;
Crooke, S. T.; Cook, P. D. Nucleosides Nucleotides 1997, 16, 1129-1138.
(7) Roberts, R. W.; Szostak, J. W. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 12297-12302.
(8) (a) Sarracino, D. A.; Steinberg, J. A.; Vergo, M. T.; Woodworth, G.
F.; Tetzlaff, C. N.; Richert, C. Bioorg. Med. Chem. Lett. 1998, 8, 2511-
2516. (b) Bleczinski, C. F.; Richert, C. J. Am. Chem. Soc. 1999, 121,
10889-10894.
(9) Kryatova, O. P.; Connors, W. H.; Bleczinski, C. F.; Mokhir, A. A.;
Richert, C. Org. Lett. 2001, 3, 987-990.
(10) Tuma, J.; Connors, W. H.; Stitelman, D. H.; Richert, C. J. Am. Chem.
Soc. 2002, 124, 4236-4246.
(11) Cohen, S. B.; Cech, T. R. J. Am. Chem. Soc. 1997, 119, 6259-
6268.
(12) Milne, L.; Perrin, D. M.; Sigman, D. S. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 3136-3141.
(13) Yamana, K.; Mitsui, T.; Nakano, H. Tetrahedron 1999, 55, 9143-
9150.
(14) Hendrix, C.; Devreese, B.; Rozenski, J.; van Aerschot, A.; De Bruyn,
A.; van Beeumen, J.; Herdewijn, P. Nucleic Acids Res. 1995, 23, 51-57.
(15) Verheyden, J. P. H.; Wagner, D.; Moffat, J. G. J. Org. Chem. 1971,
36, 250-254.
(16) McGee, D. P. C.; Vargeese, C.; Zhai, Y.; Kirschenheuter, G. P.;
Settle, A.; Diedem, C. R.; Pieken, W. A. Nucleosides Nucleotides 1995,
14, 1329-1339.
(17) McGee, D. P. C.; Sebesta, D. P.; O’Rourke, S. S.; Martinez, R. L.;
Jung, M. E.; Pieken, W. A. Tetrahedron Lett. 1996, 37, 1995-1998.
(18) Fersht, A. Enzyme Structure and Mechanism, 2nd ed.; W. H.
Freeman and Company: New York, 1992.
(19) (a) Uhlmann, E.; Engels, J. Tetrahedron Lett. 1986, 27, 1023-1026.
(b) Sande, v. d. J. H.; Ramsing, N. B.; Germann, M. W.; Elhorst, W.;
Kalisch, B. W.; Kitzing, v. E.; Pon, R. T.; Clegg, R. C.; Jovin, T. M. Science
1988, 241, 551-557. (c) Koga, M.; Moore, M. F.; Beaucage, S. L. J. Org.
Chem. 1991, 56, 3757-3759. (d) Beier, M.; Stephan, A.; Hoheisel, J. D.
HelV. Chim. Acta 2001, 84, 2089-2095.
(20) Dombi, K. L.; Steiner, U. E.; Richert, C. J. Combin. Chem. 2003,
in press.
(21) (a) Altman, R. K.; Schwope, I.; Sarracino, D. A.; Tetzlaff, C. N.;
Bleczinski, C. F.; Richert, C. J. Comb. Chem. 1999, 1, 493-508. (b) Mokhir,
A. A.; Tetzlaff, C. N.; Herzberger, S.; Mosbacher, A.; Richert, C. J. Combin.
Chem. 2001, 3, 374-386.
Supporting Information Available: Experimental pro-
cedures and MALDI-TOF mass spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL020212W
(22) Selected recent reviews: (a) Ramsay, G. Nature Biotechnol. 1998,
16, 40-44. (b) Southern, E. M.; Mir, K.; Shchepinov, M. Nature Genetics
1999, 21, 5-9. (c) Lockhardt, D. J.; Winzeler, E. A. Nature 2000, 405,
827-836.
(23) (a) Evertsz, E. M.; Au-Young, J.; Ruvolo, M. V. Lim, A. C.;
Reynolds, M. A. BioTechniques 2001, 31, 1182-1192. (b) Dombi, K. L.;
Griesang, N.; Richert, C. Synthesis 2002, 816-824.
(24) Mokhir, A.; Richert, C. Unpublished results.
250
Org. Lett., Vol. 5, No. 3, 2003