Toward amide-linked RNA mimics: Total synthesis of 3′-C branched uridine azido acid via an ene-iodolactonization approach

Article abstract of DOI:10.1021/ol027229z

(Matrix presented) A novel synthesis of 3′-C branched uridine azido acid has been accomplished by a sequence of stereoselective ene and iodolactonization reactions. Compared to traditional routes that start from carbohydrates, the present methodology is m

Full text of DOI:10.1021/ol027229z

ORGANIC
LETTERS
2003
Vol. 5, No. 2
181-184
Toward Amide-Linked RNA Mimics:
Total Synthesis of 3′-C Branched
Uridine Azido Acid via an
Ene−Iodolactonization Approach
Eriks Rozners* and Yang Liu
Department of Chemistry and Chemical Biology, Northeastern UniVersity,
Boston, Massachusetts 02115
e.rozners@neu.edu
Received November 4, 2002
ABSTRACT
A novel synthesis of 3′-C branched uridine azido acid has been accomplished by a sequence of stereoselective ene and iodolactonization
reactions. Compared to traditional routes that start from carbohydrates, the present methodology is more flexible and can be further optimized
by incorporation of novel future developments of synthetic chemistry. Because the key chemistry does not involve the 3′-C substituent, our
route is a general approach to 3′-C alkyl nucleoside analogues.
Gene therapy with ribozymes, small nucleic acids that can
catalytically cleave disease related messenger RNA mol-
ecules, is a promising alternative to conventional chemo-
therapy.1 However, the relatively low catalytic efficiency of
ribozymes currently prohibits clinical realization of this
potential. Compared to proteins, RNA is inherently less suited
for catalysis because of its limited folding, backbone
flexibility, and low diversity of functional groups.2 RNA has
negatively charged phosphates that are located on the surface
of the molecule in double helical form, and electrostatic
repulsion prevents helixes from forming the closely packed
tertiary structures needed for RNA to function as a catalyst.
The amide backbone has been used successfully in
synthetic biopolymer mimics. Peptide nucleic acids (PNA)
formed stable sequence-specific helixes with complementary
DNA and PNA.3 Short hybrid DNA-RNA duplexes, where
the DNA strand has a few selected phosphodiesters replaced
by neutral amide linkages, had thermal stability similar to
the nonmodified controls.4,5 One of us previously found that
amide linkages are equally well tolerated at some selected
positions in analogous RNA-RNA duplexes.6 Short uni-
formly modified amide-linked DNA also formed stable
duplexes with the complementary unmodified RNA and
DNA.7 These results suggest that the overall conformation
We propose that replacement of the negatively charged
phosphates by a neutral amide backbone will eliminate the
unfavorable electrostatic repulsion, and will considerably
improve the folding of RNA. Because of the improved
folding, we envision that the ribozymes built of amide-linked
RNA will be more efficient catalysts than their natural
counterparts.
(3) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991,
254, 1497. Egholm, M.; Buchardt, O.; Christensen, L.; Behrens, C.; Freier,
S. M.; Driver, D. A.; Berg, R. H.; Kim, S. K.; Norden, B.; Nielsen, P. E.
Nature 1993, 365, 566. Wittung, P.; Nielsen, P. E.; Buchardt, O.; Egholm,
M.; Norden, B. Nature 1994, 368, 561.
(4) De Mesmaeker, A.; Waldner, A.; Lebreton, J.; Hoffmann, P.; Fritsch,
V.; Wolf, R. M.; Freier, S. M. Angew. Chem., Int. Ed. Engl. 1994, 33, 226.
De Mesmaeker, A.; Lesueur, C.; Bevierre, M.-O.; Waldner, A.; Fritsch,
V.; Wolf, R. M. Angew. Chem., Int. Ed. Engl. 1996, 35, 2790. Idziak, I.;
Just, G.; Damha, M. J.; Giannaris, P. A. Tetrahedron Lett. 1993, 34, 5417.
(5) Lebreton, J.; Waldner, A.; Fritsch, V.; Wolf, R. M.; De Mesmaeker,
A. Tetrahedron Lett. 1994, 35, 5225.
(1) James, H. A.; Gibson, I. Blood 1998, 91, 371. Lewin, A. S.;
Hauswirth, W. W. Trends Mol. Med. 2001, 7, 221.
(2) Narlikar, G. J.; Herschlag, D. Annu. ReV. Biochem. 1997, 66, 19.
(6) Rozners, E.; Stro¨mberg, R. Nucleosides Nucleotides 1997, 16, 967.
10.1021/ol027229z CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/17/2002

of the amide linkage fits well in a nucleic acid double helix
causing no major disturbance.
a total synthesis of 3′-carbomethoxymethyl thymidine using
radical cyclization as the key step.
Amide-linked RNA can be prepared adopting the well-
developed peptide synthesis protocols (Figure 1). The
Our retrosynthetic analysis of 1 (Figure 1) first disconnects
the heterocyclic base and after adjustment of functional
groups reveals the iodolactone 2 as the first key intermediate.
Opening of the five-membered tetrahydofuran ring (iodolac-
tonization) identifies the unsaturated carboxylic acid 3 (or
derivative thereof) as the next key intermediate. These
disconnections reduce the complexity from two heterocycles
and four stereogenic centers in 1 to only two stereogenic
centers in the acyclic 3. The key intermediate 3 can be further
disconnected to simple organic compounds 4 and 5 to be
joined in a stereoselective ene reaction.
The synthetic realization of this plan started with protection
of the commercially available (Z)-3-penten-1-ol 6 as the tert-
butyldiphenylsilyl (TBDPS) ether 5 (Scheme 1). The tin
Scheme 1
Figure 1. Retrosynthetic analysis of amide-linked RNA.
synthetic challenge is the preparation of modified nucleosides
1, the amino acid equivalents required for the peptide type
couplings. Traditional syntheses of modified nucleosides
analogous to 1 focus on readily available chiral pool starting
materials: nucleosides and carbohydrates.4-8 The advantage
is that the stereochemical relationships are already set or can
be easily adjusted. However, such routes are frequently
lengthy and inefficient because the multifunctional and
sensitive starting materials limit the synthetic methodology
that can be used and require extensive protecting group
manipulations.
Several groups have reported syntheses of modified
nucleosides analogous to 1. Von Matt and co-workers
synthesized 2′-deoxy7 and 2′-OMe8b analogues of 1 starting
from nucleosides. Robins, Peterson, and co-workers8a pre-
pared a series of ribonucleosides 1 (base ) adenine, thymine)
starting from protected xylose. Both groups used the Wittig
reaction followed by stereoselective hydrogenation to form
the new carbon-carbon bond at C3′. Robins et al.9 also
reported a synthesis of an amide-linked uridine pentamer.
In this letter we report a novel total synthesis approach to
3′-C branched ribonucleosides 1. Our approach focuses on
reliable and stereoselective reactions that can be used to de
noVo synthesize the modified nucleosides 1 from simple
starting materials. Although conceptually similar approaches
have been used to prepare carbocyclic nucleosides,10 ap-
plications of total synthesis principles to modify natural
nucleosides are relatively rare.11,12 Lavallee et al.11 reported
tetrachloride promoted ene reaction of 5 with ethyl glyoxalate
4 gave the racemic ester 3a. The work of Mikami et al.13
has established good precedent for the regioselectivity and
the anti diastereoselectivity in the carbonyl ene reactions of
this type. The 1H NMR spectrum of 3a indicated the presence
of 5-10% of a minor compound, possibly the syn diaste-
reomer (not separable by flash chromatography at this stage).
Because the iodolactonization14 of 3a was impracticably
slow we transformed the ester into dimethyl amide 3b.
Iodolactonization of 3b in aqueous tetrahydrofuran in the
presence of sodium bicarbonate gave a mixture of trans and
cis iodolactones 2 in a ratio of 4:1. Attempts to optimize
temperature, solvent, base, and amide substituents did not
improve the trans/cis ratio. From literature precedents, it is
conceivable that the 4:1 preference for the trans isomer
(7) Von Matt, P.; De Mesmaeker, A.; Pieles, U.; Zu¨rcher, W.; Altmann,
K.-H. Tetrahedron Lett. 1999, 40, 2899.
(10) Shin, K. J.; Moon, H. R.; George, C.; Marquez, V. E. J. Org. Chem.
2000, 65, 2172. Trost, B. M.; Madsen, R.; Guile, S. D.; Brown, B. J. Am.
Chem. Soc. 2000, 122, 3151. Crimmins, M. T.; King, B. W.; Zuercher, W.
J.; Choy, A. L. J. Org. Chem. 2000, 65, 8499.
(11) Lavallee, J. F.; Just, G. Tetrahedron Lett. 1991, 32, 3469.
(12) Trost, B. M.; Shi, Z. J. Am. Chem. Soc. 1996, 118, 3037. Hager,
M. W.; Liotta, D. C. J. Am. Chem. Soc. 1991, 113, 5117. Svansson, L.;
Kvarnstrom, I.; Classon, B.; Samuelsson, B. J. Org. Chem. 1991, 56, 2993.
(13) Mikami, K.; Shimizu, M.; Nakai, T. J. Org. Chem. 1991, 56, 2952.
(14) Bartlett, P. A.; Myerson, J. J. Am. Chem. Soc. 1978, 100, 3950.
(8) (a) Robins, M. J.; Doboszewski, B.; Timoshchuk, V. A.; Peterson,
M. A. J. Org. Chem. 2000, 65, 2939. (b) Von Matt, P.; Lochmann, T.;
Kesselring, R.; Altmann, K.-H. Tetrahedron Lett. 1999, 40, 1873. (c) Robins,
M. J.; Sarker, S.; Xie, M.; Zhang, W.; Peterson, M. A. Tetrahedron Lett.
1996, 37, 3921. (d) Peterson, M. A.; Nilsson, B. L.; Sarker, S.; Doboszewski,
B.; Zhang, W.; Robins, M. J. J. Org. Chem. 1999, 64, 8183.
(9) Robins, M. J.; Doboszewski, B.; Nilsson, B. L.; Peterson, M. A.
Nucleosides, Nucleotides Nucleic Acids 2000, 19, 69.
182
Org. Lett., Vol. 5, No. 2, 2003

reflects the thermodynamic control14 in iodolactonization of
amide 3b.15 Similar cyclization of the carboxylic acid (X )
OH), expected to be under kinetic control, gave 2a and 2b
in a ratio close to 1:1. Because we could not efficiently
separate the trans and cis isomers of 2 (or later intermediates)
by flash column chromatography the following steps were
carried out on mixtures (for clarity only the major diaste-
reomer is shown in Scheme 2). Separation of isomers was
without purification of the intermediate lactols 8 (Scheme
2). Treatment of 9 with trimethylsilyl uracil in the presence
of tin tetrachloride gave the modified uridine nucleoside 10.
Although the 2′-O-acetyl group in principle would be
compatible with our planned synthesis of amide-linked RNA,
our initial experiments showed that it was too labile for the
final steps toward 1. Migration and cleavage of the acetyl
group complicated our attempts of selective removal of
TBDPS. Therefore, a protecting group change was necessary
to reveal the primary alcohol for the final oxidation and to
install the 2′-OTBS. Acid-catalyzed cleavage of both the
TBDPS and the acetyl group followed by disilylation with
tert-butyldimethylsilyl triflate (TBSOTf) gave 12 in 62%
yield for two steps without purification of the intermediate
diol 11. Treatment with HF-pyridine selectively removed the
primary TBS group and revealed the alcohol 13, setting the
stage for the final oxidation. At this point, flash silica gel
chromatography efficiently separated the products of the
deprotection reaction giving 13 (60%), 13-cis (9%), and
unreacted starting material 12 (21%). The isomeric purity
Scheme 2
1
of 13 was confirmed by H NMR.
A two-step oxidation of alcohol 13 gave the target
carboxylic acid 1a in a 97% yield after column chromatog-
raphy. An analytical sample of 1a was obtained by repeated
crystallization from hexanes or toluene containing small
amounts of methylene chloride.17
The identity of our final product was confirmed by
comparison of 1a and 1a-cis18 with a sample of chiral
carboxylic acid 1a that we prepared following the procedures
published by Robins and Peterson.8a The 1H NMR spectrum
of 1a prepared by our route matched exactly the spectrum
of independently prepared 1a, whereas the spectrum of 1a-
cis was distinctly different (see Supporting Information).
Our racemic synthesis of 1 (Schemes 1 and 2) can be
rendered asymmetric by using either chiral Lewis acid
catalyst19,20 or chiral auxiliary21 or a combination of both in
a double asymmetric mode. In preliminary experiments we
tested the chiral copper(II) bis(oxazolinyl) catalyst 14
developed by Evans and co-workers (Scheme 3).19 In contrast
achieved after synthesis of nucleoside and removal of the
TBDPS protection.
Scheme 3
Treatment of iodolactones 2 with sodium azide led to
azidolactones 7. Although inseparable by flash column
chromatography, 7 could be separated by HPLC on silica
gel column. The configuration of the diastereomeric products
trans 7a and cis 7b, initially assigned based on literature
precedents,15 was confirmed by NOESY experiments on the
separated diastereomers. We observed diagnostic NOEs
between the C4-methylene group and either the H5 of 7a or
the C5-azidomethylene protons of 7b (Scheme 1).
Using modifications of the published procedures,16 we
selectively reduced the lactone group of 7 (2.5 equiv of
DIBAL-H) and immediately acetylated the hydroxyl groups
to give the glycosyl acetates 9 in 86% yield for two steps
to original reports, the use of anhydrous 14 was essential to
obtain chiral nonracemic 3 in acceptable yields (70% after
4 days) and good enantiomeric purity (90% ee). The reaction
(15) Masaki, Y.; Arasaki, H.; Itoh, A. Tetrahedron Lett. 1999, 40, 4829.
Ha, H. J.; Lee, S. Y.; Park, Y. S. Synth. Commun. 2000, 30, 3645.
(16) Herdeis, C.; Schiffer, T. Tetrahedron 1996, 52, 14745. Mukaiyama,
T.; Suzuki, K.; Yamada, T.; Tabusa, F. Tetrahedron Lett. 1990, 46, 265.
(17) The synthetic intermediates in Schemes 1 and 2 were characterized
by 1H and 13C NMR and FT-IR spectroscopy. Elemental analysis of 1a
was in excellent agreement with theoretical calculation. See Supporting
Information for more detail.
Org. Lett., Vol. 5, No. 2, 2003
183

catalyzed by the aqua complex of 14 was unacceptably slow
(30% after 7 days). This result demonstrates the feasibility
of asymmetric synthesis of 1 using our route. The enantio-
meric purity can be further improved either by crystallization
of the final product 1 or by double asymmetric induction
using auxiliaries developed by Whitesell and co-workers.21
The total synthesis approach has the advantage that both
enantiomers of 1 can be prepared by choosing the appropriate
chiral catalyst or auxiliary thereby providing access to both
enantiomers of amide-linked RNA. This is important for
development of enantioselective ribozyme catalysis. For
example, Seelig et al.22 showed that artificially selected
ribozymes built of D- and L-nucleosides catalyzed Diels-
Alder reaction yielding enantiomeric products.
steps and 11% overall yield. Our route has possibilities for
further improvement and optimization using future develop-
ments of synthetic organic chemistry, e.g., improved variants
of iodolactonization and alternative syntheses of 3. Current
efforts in our laboratory are focused on enantioselective
synthesis of 1a.
Another feature of our synthesis is the generality. Because
the key substituent in the modified nucleoside, the 3′-C alkyl
branch, does not directly participate in the chemical trans-
formations, our route can be used to prepare other analogues
having different 3′-C substituents. For example, an ongoing
project in our laboratory requires modified nucleosides
having the 3′-C-CH2OH substituent. Currently we are work-
ing on adopting the present synthesis (Schemes 1 and 2) to
prepare these analogues starting from crotyl alcohol.
In summary, our synthetic route (Schemes 1 and 2)
provides racemic 3′-C branched uridine azido acid 1a in 13
Acknowledgment. We thank Northeastern University
(startup funds and RSDF grant to E.R.) and the donors of
the Petroleum Research Fund, administered by the American
Chemical Society (37599-AC1) for support of this research.
(18) A sample of 1a-cis was prepared from 13-cis in 76% yield by the
same two-step oxidation procedure as 1a (see Scheme 2).
(19) Evans, D. A.; Burgey, C. S.; Paras, N. A.; Vojkovsky, T.; Tregay,
S. W. J. Am. Chem. Soc. 1998, 120, 5824. Evans, D. A.; Tregay, S. W.;
Burgey, C. S.; Paras, N. A.; Vojkovsky, T. J. Am. Chem. Soc. 2000, 122,
7936
(20) Hao, J.; Hatano, M.; Mikami, K. Org. Lett. 2000, 2, 4059. Koh, J.
H.; Larsen, A. O.; Gagne, M. R. Org. Lett, 2001, 3, 1233.
(21) Whitesell, J. K.; Bhattacharya, A.; Aguilar, D. A.; Henke, K. J.
Chem. Soc., Chem. Commun. 1982, 989. Whitesell, J. K.; Lawrence, R.
M.; Chen, H.-H. J. Org. Chem. 1986, 51, 4779.
Supporting Information Available: Experimental pro-
1
cedures, spectral data, and copies of H and 13C NMR data
for compounds 1-3, 5, 7, 10, 12, and 13. This material is
available free of charge via the Internet at http://pubs.acs.org.
(22) Seelig, B.; Keiper, S.; Stuhlmann, F.; Ja¨schke, A. Angew. Chem.,
Int. Ed. 2000, 39, 4576.
OL027229Z
184
Org. Lett., Vol. 5, No. 2, 2003