Expanding the Genetic Alphabet: Non-Epimerizing Nucleoside with the pyDDA Hydrogen-Bonding Pattern

Article abstract of DOI:10.1021/jo034900k

6-Amino-3-(2′-deoxy-β-D-ribofuranosyl)-5-nitro-1H-pyridin-2-one (4), a C-glycoside exhibiting the nonstandard pyDDA hydrogen-bonding pattern, was synthesized via Heck coupling. The nitro group greatly enhances the stability of the nucleoside toward acid-c

Full text of DOI:10.1021/jo034900k

Exp a n d in g th e Gen etic Alp h a bet:
Non -Ep im er izin g Nu cleosid e w ith th e
pyDDA Hyd r ogen -Bon d in g P a tter n
Daniel Hutter† and Steven A. Benner*,†,‡
Department of Chemistry and Department of Anatomy
and Cell Biology, University of Florida,
Gainesville, Florida 32611-7200
benner@chem.ufl.edu
Received J une 25, 2003
F IGURE 1. Nonstandard base pair with the pyDDA‚puAAD9
hydrogen-bonding pattern. Pyrimidine analogue 2 was believed
to suffer epimerization less likely than known analogue 1.3f
Abstr a ct: 6-Amino-3-(2′-deoxy-â-D-ribofuranosyl)-5-nitro-
1H-pyridin-2-one (4), a C-glycoside exhibiting the nonstand-
ard pyDDA hydrogen-bonding pattern, was synthesized via
Heck coupling. The nitro group greatly enhances the stabil-
ity of the nucleoside toward acid-catalyzed epimerization
without leading to significant deprotonation of the hetero-
cycle at physiological pH. These results make nucleoside 4
a promising candidate for an expanded genetic alphabet.
These conditions are fulfilled for the pyDDA‚ puAAD9
base pair (Figure 1). While the purine base and the
corresponding nucleoside 3 have been synthesized to
study their antiviral activity as well as their capability
to form purine‚purine base pairs,10 an unambiguous
pyDDA hydrogen-bonding pattern has so far only been
reported for the RNA-pyrazine analogue 1,3f while pseudo-
isocytidine is reported to exhibit this hydrogen-bonding
pattern under certain conditons only.5h In addition to
being a component of an expanded genetic information
system, these analogues also exhibit at neutral pH the
hydrogen-bonding pattern of protonated cytosine.
The pyDDA hydrogen-bonding pattern can be achieved
only with a C-glycoside, where the heterocycle is coupled
to the sugar moiety by a C-C bond rather than a C-N
bond, which is found in the standard nucleosides. A
special feature of C-nucleosides is their susceptibility to
epimerization if an electron-donating substituent is
present in a suitable position on the heterocycle (Scheme
1). This phenomenon was described for pseudouridine
The Watson-Crick nucleobase pair in DNA follows two
rules of complementarity: size complementarity (large
purines pair with small pyrimidines) and hydrogen-
bonding complementarity (hydrogen bond donors from
one base complement hydrogen bond acceptors from the
other base).1 A decade ago, we noted that 12 nucleobases
forming six base pairs joined by mutually exclusive
hydrogen-bonding patterns are possible within the Wat-
son-Crick restraints and that these might be function-
alized to enable a single biopolymer capable of both
genetics and catalysis.2 Expanded genetic alphabets have
now been further explored in a variety of laboratories,
and the possibility of a fully artificial system has been
advanced.3-7
Various analyses of the interaction between poly-
merases and their substrates suggest that the poly-
merase seeks two unshared pairs of electrons in the
minor groove, at position 3 of the purine (or analogue)
and at position 2 of the pyrimidine (or analogue).8 In
addition, the base pairs that form three hydrogen bonds
are expected to contribute more to duplex stability than
pairs joined by just two hydrogen bonds.
(5) Modified H-bonding patterns: (a) Seela, F.; Wei, C. F. Chem.
Commun. 1997, 1869. (b) Seela, F.; He, Y. J . Org. Chem. 2003, 68,
367. (c) Hirao, I.; Ohtsuki, T.; Mitsui, T. Yokoyama, S. J . Am. Chem.
Soc. 2000, 122, 6118. (d) Hirao, I.; Ohtsuki, T.; Fujiwara, T.; Mitsui,
T.; Yokogawa, T.; Okuni, T.; Nakayama, H.; Takio, K.; Yabuki, T.;
Kigawa, T.; Kodama, K.; Yokogawa, T.; Nishikawa, K.; Yokoyama, S.
Nature Biotechnol. 2002, 20, 177. (e) Searls, T.; McLaughlin, L. W.
Tetrahedron 1999, 55, 11985. (f) Chen, D. L.; McLaughlin, L. W. J .
Org. Chem. 2000, 65, 7468. (g) Roberts, C.; Banduru, R.; Switzer, C.
J . Am. Chem. Soc. 1997, 119, 4640. (h) Oyelere, A. K.; Strobel, S. A. J .
Am. Chem. Soc. 2000, 122, 10259.
(6) Hydrophobic base pairs: (a) Kool, E. T. Curr. Opin. Chem. Biol.
2000, 4, 602. (b) Kool, E. T.; Morales, J . C.; Guckian, K. M. Angew.
Chem., Int. Ed. 2000, 39, 990. (c) Kool, E. T. Acc. Chem. Res. 2002, 35,
936. (d) Ogawa, A. K.; Wu, Y.; McMinn, D. L.; Liu, J .; Schultz, P. G.;
Romesberg, F. E. J . Am. Chem. Soc. 2000, 122, 3274. (e) Berger, M.;
Luzzi, S. D.; Henry, A. A.; Romesberg, F. E. J . Am. Chem. Soc. 2002,
124, 1222.
(7) Metal chelation: (a) Weizman, H.; Tor, Y. J . Am. Chem. Soc.
2001, 123, 3375. (b) Zimmermann, N.; Meggers, E.; Schultz, P. G. J .
Am. Chem. Soc. 2002, 124, 13684. (c) Tanaka, K.; Tengeiji, A.; Kato,
T.; Toyama, N.; Shiro, M.; Shionoya, M. J . Am. Chem. Soc. 2002, 124,
12494.
(8) Steitz, T. In Biological Organization: Macromolecular Interac-
tions at High Resolution; Burnett, R. M., Vogel, H. J ., Eds.; Academic
Press: New York, 1987; pp 45-55.
* To whom correspondence should be addressed. Tel: +352-392-
7773. Fax: +352-392-7918.
† Department of Chemistry.
‡ Department of Anatomy and Cell Biology.
(1) (a) Watson, J . D.; Crick, F. H. C. Nature (London) 1953, 171,
737. (b) Watson, J . D.; Crick, F. H. C. Nature (London) 1953, 171, 964.
(2) (a) Switzer, C. Y.; Moroney, S. E.; Benner, S. A. J . Am. Chem.
Soc. 1989, 111, 8322. (b) Piccirilli, J . A.; Krauch, T.; Moroney, S. E.;
Benner, S. A. Nature (London) 1990, 343, 33.
(3) (a) Piccirilli, J . A.; Krauch, T.; MacPherson, L. J .; Benner, S. A.
Helv. Chim. Acta 1991, 74, 397. (b) Voegel, J . J .; Altorfer, M. M.;
Benner, S. A. Helv. Chim. Acta 1993, 76, 2061. (c) Voegel, J . J .; von
Krosigk, U.; Benner, S. A. J . Org. Chem. 1993, 58, 7542. (d) Heeb, N.
V.; Benner, S. A. J . Am. Chem. Soc. 1994, 116, 6929. (e) Voegel, J . J .;
Benner, S. A. J . Am. Chem. Soc. 1994, 116, 6929. (f) von Krosigk, U.;
Benner, S. A. J . Am. Chem. Soc. 1995, 117, 5361. (g) Voegel, J . J .;
Benner, S. A. Helv. Chim. Acta 1996, 79, 1863. (h) Kodra, J . T.; Benner,
S. A. Synlett 1997, 939. (i) J urczyk, S. C.; Battersby, T. R.; Kodra. J .
T.; Park, J . H.; Benner, S. A. Helv. Chim. Acta 1999, 82, 1005. (j)
J urczyk, S. C.; Horlacher, J .; Devine, K. G.; Benner, S. A.; Battersby,
T. R. Helv. Chim. Acta 2000, 83, 1517. (k) Rao, P.; Benner, S. A. J .
Org. Chem. 2001, 66, 5012.
(9) pyDDA: Pyrimidine analogue carrying a H-bond donor-donor-
acceptor motif proceeding from the major to the minor groove.
puAAD: purine analogue carrying an H-bond acceptor-acceptor-
donor motif.
(10) (a) Kim, S. H.; Bartholomew, D. G.; Allen, L. B.; Robins, R. K.;
Revankar, G. R.; Dea. P. J . Med. Chem. 1978, 21, 883. (b) Rosemeyer,
H.; Seela, F. J . Org. Chem. 1987, 52, 5136. (c) Seela, F.; Amberg, S.;
Melenewski, A.; Rosemeyer, H. Helv. Chim. Acta 2001, 84, 1996.
(4) (a) Service, R. F. Science 2000, 289, 232. (b) Wang, L.; Schultz,
P. G. Chem. Commun. 2002, 1.
10.1021/jo034900k CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/13/2003
J . Org. Chem. 2003, 68, 9839-9842
9839

SCHEME 1
SCHEME 2 a
several decades ago11 and for other C-nucleosides since.12
In our laboratory, we have shown that C-nucleosides with
5′- and 3′-hydroxyls epimerize, usually in either acidic
or basic medium, to a mixture of four isomers with the
pyranose forms being the most abundant.3e-g With the
hydroxyls protected or substituted, the epimerization
limits itself to the R/â-forms of the furanoses. This
epimerization is problematic, especially during the acidic
deprotection step of the 5′-position during solid-phase
oligonucleotide synthesis. But the nucleoside/oligonucle-
otide will also epimerize upon standing at pH 7 for a
prolonged period of time. Considering the mechanism in
Scheme 1, to prevent epimerization, a heterocycle was
sought that prevents the electron “push” from the bond
to this acidic proton. This could be achieved by putting
an electron withdrawing group at the 5-position13 (2)
(Figure 1).
a
Key: (a) m-CPBA, acetone, rt; (b) (1) KOAc, Ac2O, ∆, (2)
MeOH-NH3, rt; (c) NaOH, EtOH/H2O, ∆; (d) N-iodosuccinimide,
DMF, rt; (e) cat. Pd(OAc)2‚2Ph3As, Et3N, DMF, 60 °C; (f) TBAF,
THF, 0 °C; (g) NaBH(OAc)3, MeCN/MeCOOH, 0 °C.
Heck couplings of iodinated heterocycles to the glycal
13 and subsequent deprotection and reduction have been
reported previously by several groups.18-21 The bulky
TBDPS group at the 3′-position is assumed to direct
addition of the heterocycle to the â-face, and the free 5′-
hydroxyl leads to stereospecific reduction of the ketone
by complexation with the borate. According to these
literature procedures, palladium acetate with triphenyl-
arsine was used as the catalyst system for the Heck
coupling, with anhydrous DMF as the solvent. Triethyl-
amine was substituted for tributylamine as the base,
since it was easier to remove during purification, and 1.2
equiv of glycal proved to be sufficient. After several days
at 60 °C, â-nucleoside 14 was obtained in good yield.
Quick removal of the 5′-protective group with TBAF at
0 °C gave ketone 15 without any significant epimeriza-
tion. The crude ketone was reduced with NaBH(OAc)3
to the final nucleoside 4, part of which was purified by
reversed-phase HPLC for the physicochemical studies.
The pKa of 4 was measured by titrating an aqueous
solution of the nucleoside with dilute aqueous solutions
of HCl and NaOH and recording a UV spectrum at
The overall strategy for our synthesis of the nucleoside
(Scheme 2) focused on the Heck coupling of the iodinated
heterocycle with the correct hydrogen bonding pattern
(12) to the known14 glycal 13.
The initial attempt to synthesize the pyridinone 11
from 2-amino-3-nitropyridine (8) via Katada rearrange-
ment of N-oxide 9 was unsatisfactory. The N-oxide was
obtained easily and in good yield by oxidation with
m-CPBA.15 The subsequent rearrangement gave mostly
tar, however, and only about 10-15% of the pyridinone,
despite several variations in reaction conditions and
workup.16 We therefore switched to the less elegant route
starting with commercial 2,6-dichloro-3-nitropyridine.
Selective aminolysis led to the known 6-chloro derivative
10,17 which was hydrolyzed to 11 with aqueous sodium
hydroxide. Iodination at the 5-position with N-iodosuc-
cinimide in DMF gave 12.
(16) (a) Daeniker, H. U.; Druey, J . Helv. Chim. Acta 1958, 41, 2148.
(b) Taylor, E. C.; Driscoll, J . S. J . Org. Chem. 1960, 25, 1716. (c)
Markgraf, J . H.; Brown, J r., H. B.; Mohr, S. C.; Peterson, R. G. J . Am.
Chem. Soc. 1962, 85, 958. (d) McKillop, A.; Bhagrath, M. K. Hetero-
cycles 1985, 23, 1697. (e) Sato, N.; Miwa, N.; Suzuki, H.; Sakakibara,
T. J . Heterocycl. Chem. 1994, 31, 1229.
(17) Radl, S.; Hradil, P. Collect. Czech. Chem. Commun. 1991, 56,
2420.
(18) (a) Farr, R. N.; Outten, R. A.; Cheng, J . C.-Y.; Daves, G. D., J r.
Organometallics 1990, 9, 3151. (b) Zhang, H.-C.; Daves, G. D., J r. J .
Org. Chem. 1992, 57, 4690. (c) Zhang, H.-C.; Daves, G. D., J r.
Organometallics 1993, 12, 1499.
(19) (a) Hsieh, H.-P.; McLaughlin, L. W. J . Org. Chem. 1995, 60,
5356. (b) Chen, D. L.; McLaughlin, L. W. J . Org. Chem. 2000, 65, 7468.
(c) Searls, T.; Chen, D. L.; Lan, T.; McLaughlin, L. W. Biochemistry
2000, 39, 4375. (d) Lan, T.; McLaughlin, L. W. Bioorg. Chem. 2001,
29, 198.
(20) Coleman, R. S.; Madaras, M. L. J . Org. Chem. 1998, 63, 5700.
(21) With 5′-OH protected: (a) Chen, J . J .; Walker, J . A., II; Liu,
W.; Wise, D. S.; Townsend, L. B. Tetrahedron Lett. 1995, 36, 8363. (b)
Walker, J . A., II; Chen, J . J .; Hinkley, J . M.; Wise, D. S.; Townsend,
L. B. Nucleosides Nucleotides 1997, 16, 1999.
(11) (a) Cohn, W. E. J . Biol. Chem. 1960, 235, 1488. (b) Chambers,
R. W.; Kurkov, V.; Shapiro, R. Biochemistry 1963, 2, 1192.
(12) (a) Ohrui, H.; J ones, G. H.; Moffatt, J . G.; Maddox, M. L.;
Christensen, A. T.; Byram, S. K. J . Am. Chem. Soc. 1975, 97, 4602.
(b) De Bernardo, S.; Weigele, M. J . Org. Chem. 1976, 41, 287. (c)
Pankiewicz, K.; Matsuda, A.; Watanabe, K. A. J . Org. Chem. 1982,
47, 485.
(13) The different numbering systems of different heterocycles create
confusion when attempting to compare substituents placed upon them.
For this discussion, we use the pyrimidine numbering system by
analogy to the pyridine rings that we describe.
(14) (a) Ireland, R. E.; Thaisrivongs, S.; Vanier, N.; and Wilcox, C.
S. J . Org. Chem. 1980, 45, 48. (b) Larsen, E.; J orgensen, P. T.; Sofan,
M. A.; Pedersen, E. B. Synthesis, 1994, 1037. (c) Walker, J . A., II; Chen,
J . J .; Wise, D. S.; Townsend, L. B. J . Org. Chem. 1996, 61, 2219. (d)
Cameron, M. A.; Cush, S. B.; Hammer, R. P. J . Org. Chem. 1997, 62,
9065.
(15) Deady, L. W. Synth. Commun. 1977, 509.
9840 J . Org. Chem., Vol. 68, No. 25, 2003

F IGURE 3. Acid-catalyzed epimerization of nucleoside 4.
Natural log of the fraction of 4 remaining after time t versus
t at pH 2.0, 3.0, 4.1, and 5.0 (37 °C, aqueous).
F IGURE 2. Tritration of nucleoside analogue 4 (UV absor-
bance versus wavelength) showing a pKa of 7.8 ( 0.1.
different pH values. The resulting UV spectra, displaying
several isosbestic points, are collected in Figure 2.
Analysis of the curves gave a pKa value of 7.8 ( 0.1,
which is ca. 2 pKa units lower than those for the natural
nucleosides. Nevertheless, nucleoside 4 should be largely
(>80%) protonated at physiological pH and still more
protonated when embedded into a polyanionic DNA or
RNA chain. Such embedding generally elevates the pKa
of the nucleotide by about 0.5 units.22
To measure the rate of epimerization of the nucleoside
at neutral pH, a sample was dissolved in D2O and the
1H NMR was measured after several time intervals. The
spectrum did not change even after 1 week. In particular,
the NMR signal arising from H-6 of the heterocycle
remained unchanged. This suggested that the compound
was stable toward epimerization at neutral pH and
allowed for analysis by HPLC to measure the epimer-
ization rate at other pH values.
For those measurements, the pH of an aqueous solu-
tion of the nucleoside (ca. 5 mM) was adjusted with dilute
aqueous NaOH or HCl to the desired values. These
solutions were incubated at constant temperature. After
given time intervals, aliquots were removed, neutralized
and subjected to analysis by reversed-phase HPLC.
At high pH (pH 11.0), no epimerization was detected
at 23 °C even after 14 days. The HPLC trace showed still
only one peak, corresponding to the starting material.
This is consistent with the specific acid catalysis mech-
anism proposed by Voegel and Benner3g for their C-
glycoside and suggests that C-glycosides employ a specific
acid mechanism for epimerization somewhat more gener-
ally.
At low pH (2.0-5.0) and 37 °C (used to speed the
reaction), HPLC analysis showed the formation of four
different compounds, as expected from the mechanism
shown in Scheme 1. The rate of epimerization was
strongly dependent upon pH (Figure 3).
Plots of the log of the progress of the reaction versus
time were curved, however. We interpreted this as
evidence for a two-step process that forms I from 4 prior
to the formation of 5-7. While several alternatives are
possible, we prefer a model that involves the achievement
F IGURE 4. HPLC traces at different stages of the epimer-
ization process at pH 3: (a) t ) 0 h (pure 4); (b) t ) 6 h
(preequilibrium including 4 and 5); (c) t ) 72 h (appearance
of 6 and 7); (d) t ) 336 h (global equilibrium).
of a preequilibrium including 4, I, and 5, prior to the
achievement of a global equilibrium that includes 6 and
7.
This is because no evidence was obtained for the
presence of substantial amounts of I at any point in the
reaction cycle (see also Figure 2). This requires that the
rate for conversion of I to 4 must be smaller than the
rate for conversion of 4 to I under these conditions. The
curvature in the log progress plots can thus not be
explained by a preequilibrium between I and 4 alone but
rather suggests the achievement of a preequilibrium
including 4, I and 5.
The notion that the formation of a five membered
furanose ring is faster than the formation of a pyranose
ring, even though the pyranose product is more stable,
is not inconsistent with general concepts of reactivity in
similar systems.23
Further evidence for this two step model can be found
by comparing the HPLC traces at different stages of the
epimerization process, as shown for pH 3 in Figure 4.
(22) Saenger, W. Principles of Nucleic Acid Structure; Springer: New
York, 1984; p 107.
(23) Baldwin, J . E.; Thomas, R. C.; Kruse, L. I.; Silberman, L. J .
Org. Chem. 1977, 42, 3846.
J . Org. Chem, Vol. 68, No. 25, 2003 9841

pH is linear within experimental error with a slope of
-0.971 (Figure 5), suggesting that the rate process is first
order in hydrogen ion concentration consistent with
specific acid catalysis.
For the design of nonepimerizable nucleosides belong-
ing to the C-glycoside class, a tradeoff evidently must be
made between having a heterocycle that remains proto-
nated at physiological pH (that is, it has a high pKa) and
the stability of the nucleoside analogue with respect to
epimerization. A stronger electron withdrawing substitu-
ent at C-5 evidently lowers the rate of epimerization. But
it evidently lowers the pKa as well. The nucleoside analog
4 described here seems to strike that balance quite well.
Ack n ow led gm en t. Support of the National Insti-
tutes of Health (GM-54048) is gratefully acknowledged.
F IGURE 5. Replot of log kapparent versus pH, showing a first-
order dependence of the epimerization reaction with hydrogen
activity.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures, characterization data, and 1H and 13C NMR spectra
for all new compounds. Experimental details for the measure-
ment of pKa and epimerization rate of 4. This material is
available free of charge via the Internet at http://pubs.acs.org.
The HPLC traces at other pH values showed the same
pattern.
The initial rate of epimerization displays pseudo first-
order rate constants kpH2 ≈ 10-4 s-1 at pH 2.0 and kpH5
≈
2 × 10-7 s-1 at pH 5.0 (Figure 3). A replot of log k versus
J O034900K
9842 J . Org. Chem., Vol. 68, No. 25, 2003