A universal, photocleavable DNA base: Nitropiperonyl 2′-deoxyriboside

Article abstract of DOI:10.1021/jo001594r

A universal, photochemically cleavable DNA base analogue would add desirable versatility to a number of methods in molecular biology. A novel C-nucleoside, nitropiperonyl deoxyriboside (NPdR, P*), has been investigated for this purpose. NPdR can be converted to its 5′-DMTr-3′-CE-phosphoramidite and was incorporated into pentacosanucleotides by conventional synthesis techniques. The destabilizing effect on hybrid formation with a complementary strand when this P*base opposes A, T, and G was found to be 3-5 kcal/mol, but 9 kcal/mol when it opposes C. Brief irradiation (λ > 360 nm, 20 min) of DNA containing the P*base and piperidine treatment causes strand cleavage giving the 3′- and 5′-phosphates. Two significant recent interests, universal/non-hydrogen-bonding base analogues and photochemical backbone cleavage, have thus been combined in a single molecule that serves as a light-based DNA scissors.

Full text of DOI:10.1021/jo001594r

J . Org. Chem. 2001, 66, 2067-2071
2067
A Un iver sa l, P h otoclea va ble DNA Ba se: Nitr op ip er on yl
2′-Deoxyr ibosid e
Michael C. Pirrung,* Xiaodong Zhao, and Shannon V. Harris
Department of Chemistry, Levine Science Research Center, Duke University,
Durham, North Carolina 27708-0317
pirrung@chem.duke.edu
Received November 9, 2000
A universal, photochemically cleavable DNA base analogue would add desirable versatility to a
number of methods in molecular biology. A novel C-nucleoside, nitropiperonyl deoxyriboside (NPdR,
P*), has been investigated for this purpose. NPdR can be converted to its 5′-DMTr-3′-CE-
phosphoramidite and was incorporated into pentacosanucleotides by conventional synthesis
techniques. The destabilizing effect on hybrid formation with a complementary strand when this
P* base opposes A, T, and G was found to be 3-5 kcal/mol, but 9 kcal/mol when it opposes C. Brief
irradiation (λ > 360 nm, 20 min) of DNA containing the P* base and piperidine treatment causes
strand cleavage giving the 3′- and 5′-phosphates. Two significant recent interests, universal/non-
hydrogen-bonding base analogues and photochemical backbone cleavage, have thus been combined
in a single molecule that serves as a light-based DNA scissors.
In tr od u ction
terrupted double-stranded nucleic acid would thereby be
shortened, facilitating conversion to single strands. Al-
ternatively, the ability to selectively destroy a short
oligonucleotide in the spatially and temporally defined
manner uniquely permitted by light could enable power-
ful studies of chemical, biochemical, or biological pro-
cesses. For example, with no added reagents, catalysts,
or treatments, oligonucleotide fragmentation could be
selectively promoted inside cells or during PCR reactions.
Many workers have developed functional analogues of
the heterocyclic bases in nucleic acids. For example,
Bergstrom7 and Brown8 have advanced nitropyrrole and
nitroindole as heterocycles that will be accepted opposite
any of the natural bases in double stranded nucleic acid.
Such “universal” bases9 have uses including simplified
syntheses of short oligonucleotides that are used as
probes for hybridization and primers for DNA sequencing
and nucleic acid amplification. Hydrophobic, self-pairing
bases10 developed by Schultz and Romesberg serve as
universal terminators.11 On the other hand, Kool has
shown that high-fidelity complementation of an un-
natural base that cannot participate in hydrogen bonding
can occur during DNA polymerase reactions. Difluoro-
toluene (F) deoxyriboside12 specifically substitutes for
thymidine (T) opposite a natural adenine (A) either in
the template strand13 or as an incoming triphosphate.14
Most techniques for DNA manipulation in molecular
biology involve polymerase chain reaction (PCR) or
processing by restriction enzymes and ligases. Chemical
methods to manipulate DNA can offer advantages in
terms of substrate generality and enzyme-free reaction
products. For example, chemical ligation of oligonucle-
otides through an SN2 reaction of 5′-phosphorothioate and
phosphoroselenoate DNAs has been recently reported by
Kool.1 Photochemical processes to manipulate nucleic
acids offer a further advantage of being fully reagent-
free. For example, photochemical generation of single-
strand breaks2 and “sticky ends” has been reported by
Taylor.3
A DNA base that recognizes the four native bases in
DNA equally and is photochemically cleavable could have
a number of uses in nucleic acid fragmentation. A variety
of new techniques for nucleic acid analysis/sequencing,
such as mass spectrometry4 and DNA “chips”,5 best
utilize relatively short, single-stranded analytes. Existing
fragmentation methods6 have drawbacks in their se-
quence bias and processing steps needed to remove
reagents/byproducts. A nucleoside analogue incorporated
stochastically into a nucleic acid chain and activated for
backbone cleavage by long-wavelength UV light would
create single-strand nicks. The average length of unin-
(7) Bergstrom, D. E.; Zhang, P.; J ohnson, W. T. Nucleic Acids Res.
1997, 25, 1935-1942. Hoops., G. C.; Zhang, P.; J ohnson, W. T.; Paul,
N.; Bergstrom, D. E.; Davisson, V. J . Nucleic Acids Res. 1997, 25,
4866-4871. Bergstrom, D. E.; Zhang, P.; Toma, P. H.; Andrews, P. C.;
Nichols, R. J . Am. Chem. Soc. 1995, 117, 1201-9.
(1) Xu, Y.; Kool, E. T. J . Am. Chem. Soc. 2000, 122, 9040. Xu, Y.;
Kool, E. T. Nucleic Acids Res. 1999, 27, 875-881.
(2) Ordoukhanian, P.; Taylor, J .-S. J . Am. Chem. Soc. 1995, 117,
9570-1.
(3) Zhang, K.; Taylor, J .-S. J . Am. Chem. Soc. 1999, 121, 11579-
11580. Ordoukhanian, P.; Taylor, J .-S. Bioconjug. Chem. 2000, 11, 94-
103. Zhang, K.; Taylor, J .-S. Biochemistry 2001, 40, 153-159.
(4) Nordhoff, E. Trends Anal. Chem. 1996, 15, 240-250. Nordhoff,
E.; Kirpekar, F.; Roepstorff, P. Mass Spectrom. Rev. 1996, 15, 69-
138.
(5) Ramsay, G. Nat. Biotechnol. 1998, 16, 40-44. Marshall, A.;
Hodgson, J . Nat. Biotechnol. 1998, 16, 27-31.
(6) Vaughan, P.; McCarthy, T. V. Nucleic Acids Res. 1998, 26, 810-
5.
(8) Loakes, D.; Brown, D. M.; Linde, S.; Hill, F. Nucleic Acids Res.
1995, 23, 2361-6. Loakes, D.; Brown, D. M. Nucleic Acids Res. 1994,
22, 4039-43.
(9) Hill, F.; Loakes, D.; Brown, D. M. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 4258-4263.
(10) McMinn, D. L.; Ogawa, A. K.; Wu, Y.; Liu, J .; Schultz, P. G.;
Romesberg, F. E. J . Am. Chem. Soc. 1999, 121, 11585-11586.
(11) Berger, M.; Wu, Y.; Ogawa, A. K.; McMinn, D. L.; Schultz, P.
G.; Romesberg, F. E. Nucleic Acids Res. 2000, 28, 2911-4.
(12) Schweitzer, B. A.; Kool, E. T. J . Org. Chem. 1994, 59, 7238-
42.
10.1021/jo001594r CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/01/2001

2068 J . Org. Chem., Vol. 66, No. 6, 2001
Pirrung et al.
Ch a r t 1
This work was aimed at combining the concepts of
universal bases with nucleotide substitutes that can be
cleaved by irradiation into two portions (and thereby
cleave the backbone of the nucleic acid). Because the
known photocleavable compounds do not resemble natu-
ral nucleotides, they cannot form duplex structures and
could not be DNA polymerase substrates. We designed
the C-nucleoside called P* (nitropiperonyl) based prima-
rily on the substitution pattern known to provide optimal
nitrobenzyl photochemistry and without regard to base
pairing. Given its monocyclic structure with an oxygen
at C4, P* resembles T. The C4 carbonyl of thymine
participates in a hydrogen bond with the amino group of
adenosine in the natural T-A pair, but the C2 carbonyl
is not involved in any hydrogen bonding. The nitro group
of P* is at the position corresponding to C2, is expected
to play no role in pairing, and can occupy free space.
Because of the resemblance of the nitroaromatic to the
nitroheterocyclic universal bases, P* offers the potential
to be accepted by nucleic acid modifying enzymes. Pre-
liminary reports on some aspects of this work have
appeared.15
Ta ble 1. Obser ved NOE Cr oss P ea k s in NOESY of
2′-Deoxy-1′-(2-n itr op ip er on yl)-D-r ibofu r a n ose
protons
H1′a
H2′Ra
H2′âa
H3′a
H1′
++
+
++
-
-
++
â-anomer
H2′R (2.63 ppm)
H2′â (1.92 ppm)
H3′
++
+
-
++
+
-
++
++
++
H1′
-
-
++
R-anomer
H2′R (1.85 ppm)
H2′â (3.00 ppm)
H3′
-
++
-
++
-
++
a
++, strong; +, detectable; -, not detectable.
deprotected C-deoxyriboside and NOESY studies of both
R and â anomers of 1. They show very distinct patterns
of NOE cross-peaks (Table 1). In the R anomer, one of
the H2′ protons (R) has coupling with H1′ but not H3′.
The other H2′ proton (â) has coupling with both H3′ and
H1′. In the â anomer, the H2′R proton has strong coupling
with H1′ and weak coupling with H3′. The other H2′
proton (â) has strong coupling with H3′ and weak
coupling with H1′. Unlike the observation in the R
anomer, the longer distance couplings of H2′R-H3′ and
H2′â-H1′ appear as weak NOESY peaks rather than
being completely absent. This may be due to the flex-
ibility of the 2′-deoxyribofuranose conformation of the â
anomer, since AM1 calculations (Spartan) of the â
anomer suggest that there may be a steric interaction
between the 5′-CH2 and the aryl group. It is also worth
noting that H2′â (δ ) 1.92 ppm) in the â anomer and
H2′R (δ ) 1.85 ppm) in the R anomer resonate at higher
field due to diamagnetic shielding by the aromatic ring.
The nitropiperonyl-2′-deoxyriboside was crystallized from
methanol solution and subjected to X-ray crystallogra-
phy.20 The structure demonstrates most of the properties
of natural nucleosides, with the base in an anti confor-
mation (based on analogy to T; vide infra) and a 3′-exo
sugar pucker in the 2′-deoxyribose. The AM1 calculation
also suggests the most stable conformation of the base
is anti.
Resu lts
The target for synthesis was (nitropiperonyl)deoxyri-
boside. It is prepared by treatment of ditoluoylribosyl
chloride16 with the diarylcadmium17 reagent derived from
benzodioxol-4-yl Grignard. This method for preparation
of the C-nucleoside was chosen because of its superiority
(for simple aryl nucleosides) to some other known meth-
ods.18 The coupling reaction was examined at three
different temperatures, 4 °C, room temperature, and
refluxing THF. Higher reaction temperature may cause
side reactions, as colored products result. Reaction at
room temperature results in the highest yield (87%),
providing a mixture of epimers (R/â ) 10/1). Reaction at
4 °C does not show a notable improvement of yield, and
the ratio of R/â epimers is independent of reaction
temperature. Kool originally performed similar couplings
at room temperature and saw anomer ratios favoring R
by 5:1 to 10:1. Using Kool’s p-TsOH/toluene equilibration
procedure,19 our initial mixture is converted to a 1:1.55
R/â ratio that is rectified by chromatography. The R
anomer can be recycled to give good overall throughput
to the â anomer (76%). Nitration (Cu(NO3)2/Ac2O, 90%)
yields 1 (λmax 360 nm, H2O) (Chart 1). There is no
detectable R/â isomerization under these nitration condi-
tions, which were the first examined.
Using conventional methods,21 1 can be converted to
the 5′-dimethoxytrityl-3′-cyanoethyl(diisopropyl)phospho-
ramidite 2 (52%) and used in automated DNA synthesis
(Chart 1). It was used to prepare the pentacosanucleotide
4, which substitutes P* for T in parent sequence 3.
Hybridization properties of 4 were examined against the
four variants of sequence 5, giving all four pairings with
P*. Using known methods,22 the thermodynamics of
melting were determined with the base pairs in the boxed
region. The results, summarized in Table 2, support the
idea that P* resembles T as well as exhibiting universal
Before examining the incorporation of P* into DNA,
we address the issue of whether the synthesized artificial
base has the desired structural features. The structure
was confirmed through X-ray crystallography of the
(13) Moran, S.; Ren, R. X.-F.; Kool, E. T. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 10506-10511. Moran, S.; Ren, R. X.-F.; Rumney, S.; Kool,
E. T. J . Am. Chem. Soc. 1997, 119, 2056-2057.
(14) Liu, D.; Moran, S.; Kool, E. T. Chem. Biol. 1997, 4, 919-926.
(15) Pirrung, M. C.; Zhao, X. Phosph. Sulf. 1999, 146, 371-372.
Pirrung, M. C.; Zhao, X. Harris, S. V. Abstracts of Papers, 218th
National Meeting of the American Chemical Society, Aug 22-26, 1999;
New Orleans, LA; American Chemical Society: Washington, DC, 1999;
ORGN 609.
(16) Hoffer, M. Chem. Ber. 1960, 93, 2777-2781.
(17) Chaudhuri, N. C.; Kool, E. T. Tetrahedron Lett. 1995, 36, 1795-
1798.
(18) Hsieh, H. P.; Mclaughlin, L. W. J . Org. Chem. 1995, 60, 5356-
5359.
(19) Ren, R. X.-F.; Chaudhuri, N. C.; Paris, P. L.; Rumney, S.; Kool,
E. T. J . Am. Chem. Soc. 1996, 118, 7671-7678.
(20) The structure was solved at -100 °C by Dr. P. White, University
of North Carolina. The crystal belongs to the monoclinic P21 space
group. Data were deposited in the Cambridge Structural Database.
(21) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1992, 48, 2223-2311.
(22) Marky, L. A.; Breslauer, K. J . Biopolymers 1987, 26, 1601-
1610. Gelfand, C. A.; Plum, G. E.; Grollman, A. P.; J ohnson, F.;
Breslauer, K. J . Biochemistry 1998, 37, 7321-7327.

Universal, Photocleavable DNA Base
J . Org. Chem., Vol. 66, No. 6, 2001 2069
observed products. Very well precedented nitrobenzyl
photochemistry23 should lead to 8, which is also similar
to known pathways for creation of abasic sites through
formation of a C-1′ radical.24 Ring-chain tautomerism
produces ketone 9, which can undergo â-elimination of
the 3′-phosphate to 10. To generate the product derived
F igu r e 1. Photochemical backbone cleavage. Oligonucleotide
4 was 32P-end labeled with polynucleotide kinase. A solution
of 0.1 pmol of 4 in 10 µL of TE was irradiated in a Rayonet
reactor at 400 µW/cm2 for 20 min. Aqueous piperidine (90 µL
of 1 M solution) was added, and the reaction was heated to 90
°C for 30 min. The solvent was removed under vacuum, and
the products were suspended in formamide/loading dye and
then subjected to 12% denaturing polyacrylamide gel electro-
phoresis. Lane 1 is starting 4. Lane 2 is the irradiated product
before backbone cleavage. Lane 3 is the cleaved product, which
has the mobility of a 13-mer, indicating it is a 12-mer bearing
a 3′-phosphate.
Ta ble 2. Th er m od yn a m ics of Hybr id iza tion of
Oligon u cleotid es 3 a n d 4 (Con ta in in g th e P * Ba se) to
Oligon u cleotid e 5
from the segment of oligonucleotide to the 5′ of P* (its
3′-phosphate), a second elimination reaction likely occurs.
Extensive efforts to establish this mechanism through
isolation of intermediates and products in the irradiation
of 1 were unsuccessful. Support for the mechanism comes
from direct MALDI-TOF mass spectrometry of 4, wherein
laser excitation cleaves the backbone. The observed ions
correspond to the 5′-fragment (11; before phosphate
elimination) and the 3′-fragment (12).
base pair
(3 or 4-5)
∆H°
(kcal/mol)
∆S°
(kcal/mol‚K)
∆G°
(kcal/mol)
T
m (°C)
T-A
62.0
57.0
55.0
55.0
53.5
56.0
-195.9
-187.7
-163.9
-154.4
-143.1
-145.6
-0.555
-0.538
-0.469
-0.440
-0.408
-0.412
-30.5
-27.2
-24.1
-23.2
-21.5
-22.7
P*-A
P*-T
P*-G
P*-C
T-T
base properties. It pairs best with A, but also fairly well
with G and T. Qualitatively, a P*-A pair leads to a ∼5
°C depression in Tm compared to a T-A pair.
The photochemical behavior of P* in DNA strands was
next examined. The 5′-32P-labeled 4 was subjected to
irradiation at 350 nm, followed by base treatment (pip-
eridine, 90 °C), resulting in a 12-mer oligonucleotide
fragment bearing a 3′-phosphate (Figure 1) (and presum-
ably an unlabeled 12-mer). A second 25-mer 6 was
synthesized that places P* asymmetrically to ensure
reliable identification of the fate of the strands 5′ of P*
and 3′ of it. Irradiation of its 5′-end-labeled derivative
and base treatment leads to a 6-mer product as the 3′-
phosphate (5′-GTA GAAp-3′) as established by PAGE
(data not shown). A similar experiment with unlabeled
oligonucleotide gives two products as established by
HPLC (C18, 100 mM Et3NHOAc/MeCN 0-50% gradient),
5′-GTA GAAp-3′ (14.8 min), and 5′-TC TTT TCC TTC
TAG ATC G-3′ (28.0 min).
Discu ssion
The relatively good stability of DNA duplexes contain-
ing the P* base is likely a consequence of its nitroaryl
group, which is known from the nitropyrrole and nitroin-
(23) Pillai, V. N. R. Synthesis 1980, 1.
(24) Tronche, C.; Greenberg, M. M. Chem. Biol. 1998, 5, 263-271.
Meijler, M. M.; Zelenko, O.; Sigman, D. S. J . Am. Chem. Soc. 1997,
119, 1135-6. Hwang, J .-T.; Tallman, K. A.; Greenberg, M. M. Nucleic
Acids Res. 1999, 19, 3805-3810.
A mechanism can be proposed for photochemical cleav-
age of oligonucleotides containing P* (7) to give the

2070 J . Org. Chem., Vol. 66, No. 6, 2001
Pirrung et al.
by adding a few I2 crystals and kept at ∼40 °C for 1 h.
Cadmium chloride powder (450 mg, 2.5 mmol) was added.
After being refluxed for 1.5 h, the mixture was cooled to room
temperature, 1′R-chloro-3′,5′-di-O-toluoyl-2′-deoxyribose (1.50
g, 3.8 mmol), prepared as reported by Hoffer, was added, and
the resulting mixture was allowed to stir for 6 h. The reaction
was quenched with aqueous NH4Cl and extracted with ethyl
acetate. The extracts were washed with saturated aqueous
NaHCO3 and saturated NaCl solution and dried over anhy-
drous Na2SO4. The mixture was filtered, and the solvent was
removed. The products were purified by flash column chro-
matography on silica gel with 12% ethyl acetate/hexane as the
eluant. The products (1.60 g, 3.3 mmol) were obtained as clear
oil (epimer R/â, 10:1) in 87% yield.
Ch a r t 2
Equ ilibr a tion to 2′-Deoxy-1′â-p ip er on yl-3′,5′-d i-O-tolu -
oyl-D-r ibofu r a n ose. To 75 mL of toluene were added the R/â
epimer mixture (1.00 g), a catalytic amount of p-toluenesulfonic
acid (100 mg), one drop of concentrated sulfuric acid, and five
drops of water. After the mixture was refluxed for 2 h, the
toluene was removed and the residue was neutralized with
saturated aqueous NaHCO3 and extracted with ethyl acetate.
The extracts were washed with saturated NaCl solution and
dried over anhydrous Na2SO4. The R and â epimers were
isolated by flash chromatography on silica gel with 12.5% ethyl
acetate/hexane as the eluent. R Epimer: 0.30 g (30%); 1H NMR
(300 MHz, CDCl3) δ 7.95 (2H, d, J ) 8.2 Hz), 7.78 (2H, d, J )
8.2 Hz), 7.23 (2H, d, J ) 8.3 Hz), 7.20 (2H, d, J ) 8.3 Hz),
6.97-6.76 (3H, m), 5.94 (2H, s, CH2), 5.58 (1H, m, H1′), 5.26
(1H, m, H3′), 4.66 (1H, m, H4′), 4.56 (2H, m, H5′,5′′), 2.89 (1H,
m, H2′â), 2.40 (6H, s, 2 × CH3), 2.11 (1H, m, H2′R); IR (thin
film, cm-1) 2922, 2852, 1720, 1271, 1104. â Epimer, 0.46 g
(46%): 1H NMR (300 MHz, CDCl3) δ 7.97 (2H, d, J ) 8.3 Hz),
7.96 (2H, d, J ) 8.3 Hz), 7.27 (2H, d, J ) 8.0 Hz), 7.23 (2H, d,
J ) 8.0 Hz), 6.90-6.74 (3H, m), 5.94 (2H, s, CH2), 5.58 (1H,
dd, J ) 1.3, 4.8 Hz, H1′), 5.16 (1H, dd, J ) 5.0, 10.9 Hz, H3′),
4.63 (2H, m, H5′5”), 4.50 (1H, m, H4′), 2.49 (1H, m, H2′R), 2.44
(3H, s, CH3), 2.41 (3H, s, CH3), 2.22 (1H, m, H2′â); IR (thin
film, cm-1) 2921, 1719, 1611, 1271, 1105; HRMS (FAB) m/z
calcd for C28H26O7 474.1678, found 474.1682.
2′-Deoxy-1′â-(2-n itr op ip er on yl)-3′,5′-d i-O-tolu oyl-D-r i-
bofu r a n ose. To a solution of 2′-deoxy-1′â-piperonyl-3′,5′-di-
O-toluoyl-D-ribofuranose (1.40 g, 2.95 mmol) in 175 mL of
acetic anhydride immersed in an ice-water bath was added
Cu(NO3)2‚3H2O (2.57 g, 9.38 mmol).28 The mixture was stirred
for 25 min and then poured into 200 mL of saturated aqueous
NaHCO3 and extracted with ethyl acetate. The extracts were
washed with saturated NaCl solution and dried over anhy-
drous Na2SO4. The product was purified by flash chromatog-
raphy on silica gel with 18% ethyl acetate/hexane as the
eluant. The product (1.38 g, 2.65 mmol) was obtained as a
yellow oil in 90% yield: 1H NMR (300 MHz, CDCl3) δ 7.98
(2H, d, J ) 8.2 Hz, Tol-H), 7.93 (2H, d, J ) 8.2 Hz, Tol-H),
7.54 (1H, s, Ar-H), 7.37 (1H, s, Ar-H), 7.27 (2H, d, J ) 8.0 Hz,
Tol-H), 7.22 (2H, d, J ) 8.0 Hz, Tol-H), 6.09 (1H, d, J ) 6.6
Hz, CH2), 6.08 (1H, d, J ) 6.6 Hz, CH2), 5.77 (1H, dd, J ) 5.1,
10.2 Hz, H1′), 5.58 (1H, dd, J ) 1.8, 4.8 Hz, H3′), 4.72 (2H, m,
H5′,5”), 4.51 (1H, m, H4′), 2.94 (1H, ddd, J ) 1.2, 5.1, 6.4 Hz,
H2′R), 2.43 (3H, s, CH3), 2.40 (3H, s, CH3), 2.05 (1H, m, H2′â);
IR (thin film, cm-1) 2922, 2853, 1712, 1611, 1503, 1482, 1268,
1177, 1106; HRMS (FAB) m/z calcd for C28H25NO9 519.1529,
found 519.1539.
dole precedents to facilitate stacking interactions even
when hydrogen bonding across the helix is not possible.
The nitro group thus serves a dual purpose, as it also
imparts the photochemical reactivity required to cleave
the backbone. The hybridization of P*-containing oligo-
nucleotides in several sequence contexts shows some
selectivity for P*-A pairs. Progression from the T-A
perfect match to the P*-A pairing to the T-P* pairing
(Chart 2) results in a ∼3 kcal/mol difference in duplex
stability at each step, perhaps reflecting the loss of a
hydrogen bond. Generally, P* is accepted in duplexes
through base stacking, fulfilling the intent if not the
definition of a universal base. However, given that earlier
workers have not found a correlation between thermo-
dynamic duplex stability and the acceptance of unnatural
bases by enzymes such as DNA polymerases, further
experimentation will be required to define the true nature
of P* as a universal base.
The biophysical evidence suggests that our synthetic
compound has many desirable features for an artificial
nucleoside. The photochemistry of P* leads to DNA
strand cleavage exclusively at the modified position,
demonstrating that the behavior of P* is not due to a
general DNA cleaving property of nitro compounds under
irradiation.25 The photochemistry of this P* base parallels
chemistry recently reported for the photochemical intro-
duction of deoxyribonolactone lesions into oligodeoxyri-
bonucleotides through irradiation of nitroindole deoxyri-
bosides, though this work was not extended to backbone
cleavage.26
2′-Deoxy-1′â-(2-n itr op ip er on yl)-D-r ibofu r a n ose. To a
solution of 2′-deoxy-1′â-(2-nitropiperonyl)-3′,5′-di-O-toluoyl-D-
ribofuranose (0.22 g, 0.42 mmol) in 30 mL of methanol was
added 0.5 M sodium methoxide in methanol (7.5 mL). The
mixture was stirred for 45 min and then quenched by the
addition of NH4Cl powder (1.5 g). The mixture was filtered,
concentrated, and flash chromatographed on silica gel with
80% ethyl acetate/hexane. The product was obtained as yellow
solid (0.12 g, 0.42 mmol) in a quantitative yield: UV (H2O)
Exp er im en ta l Section
2′-De oxy-1′-p ip e r on yl-3′,5′-d i-O-t olu oyl-D-r ib ofu r a -
n ose. To 5 mL of dry tetrahydrofuran (THF) and magnesium
turnings (120 mg, 5 mmol) was added 4-bromo-1,2-methyl-
enedioxybenzene (1.03 g, 5 mmol).27 The reaction was initiated
(25) Nielsen, P. E.; J eppesen, C.; Egholm, M.; Buchardt, O. Bio-
chemistry 1988, 27, 6338-43. Nielsen, P. E.; Egholm, M.; Koch, T.;
Christensen, J . B.; Buchardt, O. Bioconjug. Chem. 1991, 2, 57-66.
(26) Kotera, M.; Bourdat, A.-G.; Defranq, E.; Lhomme, J . J . Am.
Chem. Soc. 1998, 120, 11810-11811.
λmax 360.5 nm, ꢀ
1.3 × 104, UV (CH3OH) λmax 346.0 nm, ꢀ
max
(27) Hurd, C. D.; Bonner, W. A. J . Am. Chem. Soc. 1945, 67, 1972-
1977.
(28) Klein, R. S.; Kotick, M. P.; Watanabe, K. A.; Fox, J . J . J . Org.
Chem. 1971, 36, 4113-4116.

Universal, Photocleavable DNA Base
J . Org. Chem., Vol. 66, No. 6, 2001 2071
1
1.4 ×104; mp 133-134 °C; H NMR (300 MHz, CDCl3) δ
calculated 7669.04, only the two fragment peaks at 3996.21
(calcd 3993.67) and 3597.82 amu (calcd 3596.40).
max
7.54 (1H, s, Ar-H), 7.31 (1H, s, Ar-H), 6.12 (1H, d, J ) 3.0
Hz, CH2), 6.11 (1H, d, J ) 3.0 Hz, CH2), 5.68 (1H, dd, J ) 6.0,
9.0 Hz, H1′), 4.40 (1H, m, H3′), 4.02 (1H, m, H4′), 3.93 (2H,
m, H5′,5”), 2.63 (1H, ddd, J ) 3.0, 6.0, 8.9 Hz, H2′R), 1.92 (1H,
m, H2′â); IR (thin film, cm-1) 3396 (br), 2931, 2857, 1520, 1504,
Melt in g Tem p er a t u r e. Double-stranded oligonucleotide
25-mers (1-20 µM) were dissolved in a buffer containing 100
mM NaCl, 10 mM sodium phosphate, pH 7.0. Absorbance was
measured in quartz cuvettes with optical path lengths of 1 cm
or 1 mm using an Aviv 62DS spectrophotometer. The absor-
bance of the oligonucleotides at 260 nm was in the range of
0.4-1.2. The samples were heated in the spectrophotometer
from 40 to 75 °C at increments of 0.5. Each incremental data
point was equilibrated for 30 s with a maximum fluctuation
of 0.1 °C. The melting temperature (Tm) of each duplex
oligonucleotide was determined at three to four different
concentrations by taking the first derivative of the melting
curve (absorbance vs temperature). Curve fitting was per-
formed with SigmaPlot using a simplified nonlinear least-
squares model, and the data were plotted in MS Excel.
Thermodynamic data were calculated from a two-state model
using the method of Marky & Breslauer. ∆G was determined
for 25 °C (298 K), and Tm was calculated at 1 µM.
MALDI-TOF MS. Oligonucleotides were precipitated with
ammonium acetate/ethanol. After resuspension in 1:1 aceto-
nitrile/water at ∼20 pmol/µL, 0.8 µL was placed on a MALDI
plate and allowed to air-dry. An 0.4 µL aliquot of AG-50W X
ion-exchange resin that had been previously treated with 5M
ammonium acetate was placed on top of the DNA spot and
allowed to dry. Matrix (0.8 µL of 6-aza-2-thiothymine solution,
0.1 g/mL with ammonium citrate) was added and allowed to
dry. MALDI mass spectra were acquired in the linear, nega-
tive-ion mode using a nitrogen laser (337 nm). All spectra were
collected using an acceleration voltage of 25 kV, a grid voltage
of 23 kV, a guide wire voltage of 75 V, and a delay time of 225
ns. Each spectrum was the sum of 156 to 256 laser shots. The
raw intensity versus time data in each mass spectrum were
smoothed using a Savitsky-Golay routine prior to mass
calibration using internal standards.
Ir r a d ia tion Exp er im en ts. Oligonucleotides 4 or 6 were
treated with T4 polynucleotide kinase (NEB) and ATP at 37
°C for 2 h in kinase buffer. The products were size-separated
on 19% PAGE, and the desired product was excised from the
gel, eluted into 1 × TE buffer (10 mM Tris‚HCl, 1 mM EDTA,
pH 7.5), and precipitated with ethanol. The 5′-labeled oligo-
nucleotide 25-mer (0.1 pmol) was suspended in 10 µL of TE
and irradiated at 350 nm in a Rayonet photochemical reactor
maintained close to room temperature by a fan. A Pyrex filter
was used to block light <300 nm. Photolysis was complete at
a fluence of 400 µW/cm2 for 20 min, after which 90 µL of 1 M
piperidine in water was added and the reaction was heated to
90 °C for 30 min. The solvent was removed under vacuum,
and the products were resuspended in 10 µL of formamide/
dye and resolved on a 12% denaturing polyacrylamide gel.
1482, 1329, 1255, 1033; HRMS (FAB) m/z calcd for C12H13
NO7 283.0692, found 283.0686.
-
2′-deoxy-1′R-(2-nitropiperonyl)-D-ribofuranose was prepared
similarly from 2′-deoxy-1′R-piperonyl-3′,5′-di-O-toluoyl-D-ribo-
furanose: 1H NMR (300 MHz, CDCl3) δ 7.56 (1H, s, Ar-H),
7.35 (1H, s, Ar-H), 6.12 (1H, d, J ) 4.6 Hz, CH2), 6.11 (1H, d,
J ) 4.6 Hz, CH2), 5.68 (1H, t, J ) 7.1 Hz, H1′), 4.44 (1H, m,
H3′), 4.22 (1H, m, H4′), 3.76 (2H, m, H5′,5′′), 3.00 (1H, m,
H2′â), 1.85 (1H, m, H2′R).
2′-Deoxy-1′â-(2-n it r op ip er on yl)-5′-O-d im et h oxyt r it yl-
D-r ibofu r a n ose. To a solution of 2′-deoxy-1′â-(2-nitropipero-
nyl)-D-ribofuranose (0.11 g, 0.39 mmol), triethylamine (0.15
mL), and N,N-dimethylaminopyridine (5 mg, 0.041 mmol) in
10 mL of pyridine was added dimethoxytrityl chloride (0.194
mg, 0.50 mmol). After being stirred for 6 h, the reaction was
poured into 50 mL of water and extracted with ethyl acetate.
The extracts were washed with saturated aqueous NaHCO3
and saturated NaCl solution and dried over anhydrous Na2-
SO4. The product was obtained as a yellow foam (0.16 g, 0.27
mmol) in 69% yield after flash chromatography on silica gel
with 27.5% ethyl acetate/hexane as the eluant: 1H NMR (300
MHz, CDCl3) δ 7.55-7.19 (11H, m, Ar-H), 6.84 (4H, d, J ) 8.8
Hz), 6.09 (2H, 2 d, J ) 8.4 Hz, CH2), 5.69 (1H, dd, J ) 6.2, 8.8
Hz, H1′), 4.38 (1H, m, H3′), 4.06 (1H, m, H4′), 3.79 (6H, s,
2×OCH3), 3.38 (2H, m, H5′,5′′), 2.62 (1H, ddd, J ) 3.1, 6.1,
9.3 Hz, H2′R), 1.96 (1H, m, H2′â); IR (thin film, cm-1) 2927,
1511, 1480, 1257, 1034; HRMS (FAB) m/z calcd for C33H31
NO9 585.1998, found 585.1983.
-
2′-Deoxy-1′â-(2-n itr op ip er on yl)-3′-O-p h osp h or a m id ite-
5′-O-d im et h oxyt r it yl-D-r ib ofu r a n ose. 2′-Deoxy-1′â-(2-ni-
tropiperonyl)-5′-O-dimethoxytrityl-D-ribofuranose (60 mg, 0.103
mmol) was dissolved in 4 mL of CH2Cl2 and 1 mL of triethy-
lamine, and 2-cyanoethyl N,N-diisopropylchlorophosphora-
midite (0.10 mL, 0.44 mmol) was added. After the mixture was
stirred for 3 h, the solvent was removed under vacuum and
the products were purified by column chromatography on silica
gel eluted with 30% ethyl acetate/hexane. A mixture of two
isomeric products was obtained as a pale yellow oil (60 mg,
0.076 mmol) in 76% yield: 1H NMR (300 MHz, CDCl3) δ 7.55-
7.20 (11H, m, Ar-H), 6.84 (4H, m), 6.10 (2H, m, CH2), 5.67
(1H, m, H1′), 4.46 (1H, m), 4.21 (1H, m), 3.86 (1H, m), 3.80
(3H, s, OCH3), 3.79 (3H, s, OCH3), 3.62 (2H, m), 3.32 (1H, m),
2.78 (1H, m), 2.65 (2H, m, CH2), 2.43 (2H, m, CH2), 1.92 (1H,
m), 1.17 (12H, d, J ) 6.8 Hz, 4 × CH3); 31P NMR (162 MHz,
CDCl3, referenced to TMP at 0.00) δ 149.24, 148.33; IR (thin
film, cm-1) 2958, 2923, 2853, 1605, 1511, 1480, 1462, 1256,
1033; HRMS (FAB) m/z calcd for C42H48N3O10P 785.3077, found
785.3020.
Oligon u cleotid e Syn th esis. Oligonucleotides were syn-
thesized on an Applied Biosystems 392 DNA/RNA synthesizer
with standard phosphoramidite chemistry. An additional 10
min was added to the coupling time for the P* nucleoside to
ensure complete incorporation, though it was not established
that this long a coupling was necessary. The oligomers were
purified by either HPLC or 12% polyacrylamide gel electro-
phoresis. MALDI analysis of 3 gave an observed mass of
7617.85 amu, comparable ((0.1%) to the calculated value of
7610.0 amu. MALDI analysis of 4 gave no parent ion at the
Ack n ow led gm en t. Financial support provided by
NIH GM 46720. The assistance of Dr. G. Dubay and
access to the MS instruments of Prof. M. Fitzgerald is
gratefully acknowledged. The assistance of L. LaBean
in administrative support of this work is appreciated.
Su p p or tin g In for m a tion Ava ila ble: Melting curves for
six duplexes and ORTEP plot of P* nucleoside. This material
is available free of charge via the Internet at http://pubs.acs.org.
J O001594R