Peptide ribonucleic acids (PRNA). 2. A novel strategy for active control of DNA recognition through borate ester formation

Article abstract of DOI:10.1021/ja9935456

The effect of adding borax and boric acids on the nucleobase orientation and recognition behavior of novel mono- and oligomeric peptide ribonucleic acids (PRNAs) has been investigated. The base orientation of 5'-amino-5'-deoxyuridine and 5'-amino-5'-deoxy

Full text of DOI:10.1021/ja9935456

6900
J. Am. Chem. Soc. 2000, 122, 6900-6910
Peptide Ribonucleic Acids (PRNA). 2. A Novel Strategy for Active
Control of DNA Recognition through Borate Ester Formation
Takehiko Wada,*,† Narutoshi Minamimoto,† Yoshiaki Inaki,‡ and Yoshihisa Inoue*,†,§
Contribution from the Department of Molecular Chemistry and Department of Material and Life Science,
Osaka UniVersity, 2-1 Yamada-oka, Suita 565-0871, Japan and Inoue Photochirogenesis Project, ERATO,
JST, 4-6-3 Kamishinden, Toyonaka 565-0085, Japan
ReceiVed October 4, 1999
Abstract: The effect of adding borax and boric acids on the nucleobase orientation and recognition behavior
of novel mono- and oligomeric peptide ribonucleic acids (PRNAs) has been investigated. The base orientation
of 5′-amino-5′-deoxyuridine and 5′-amino-5′-deoxycytidine was shown by CD and NOE difference spectral
studies to switch from anti to syn in borate buffer or upon addition of borax. The origin of this phenomenon
is elucidated to be the cooperative effect of the cyclic borate esterification of the sugar’s cis-2′,3′-diol and the
hydrogen-bonding interaction between the sugar’s 5′-amino proton and the base’s 2-carbonyl oxygen. Because
this new strategy for switching the base orientation through the addition of borate is potentially applicable to
the recognition control of nucleic acids if the sugar’s 5′-proton and cis-2′,3′-diol remain unmodified, we
synthesized a series of PRNAs, in which the 5′-amino-5′-deoxypyrimidine ribonucleoside moiety was appended
to a mono- or oligo(γ-L-glutamic acid) backbone through the 5′-amino group. The orientation switching through
the addition of borate was also confirmed with the monomeric model, and the switching efficiency was enhanced
for oligomeric γ-PRNA. Finally, it was unambiguously demonstrated that the γ-PRNA 8-mer with an isopoly-
(L-glutamic acid) backbone can form a tight complex with DNA, and further, the recognition of DNA with
γ-PRNA 8-mer is controlled by the borate added as an external factor.
Introduction
istry is known to significantly influence the hybridization
efficiency.8 An inherent, crucial drawback of these model
systems is the lack of a direct means to actively control the
function of these nucleic acids.
To elucidate and mimic the versatile functions of DNA and
RNA, a wide variety of nucleic acid model compounds which
show sequence-specific recognition/binding to mRNA and
dsDNA have been designed.1 Several models of both normal
and antisense nucleic acids have been proposed, not only to
improve the stability of the oligomers and hybrids in the
presence of nuclease but also to enhance the hybridization
affinity,2 including phosphodiesters,3 phosphorothioate,4 alkyl-
phosphonate,5 and phosphoramidate.6 Even though it is difficult
in general to control the stereogenic center of the phosphorus
in the preparation of these model compounds,7 its stereochem-
It is well-documented that the recognition/binding abilities
of nucleic acids and their analogues are critically affected by
their conformational variations,9 especially through the syn-
anti orientational change of the nucleic acid base moiety.
Certainly, the anti orientation of the nucleobase is an essential
factor for efficient recognition, since syn-orientated nucleobases
are unfavorable for forming intermolecular hydrogen bonds with
the complementary base. Thus, if nucleobase orientation could
be switched by an additive, the external control of recognition
behavior can be readily applied. Unfortunately, no external agent
or factor, which affects the syn-anti orientation and therefore
the recognition behavior of nucleic acids and analogues has been
found or proposed to date, although the unusual syn orientation
is known to be induced by several internal factors, such as
increased steric hindrance and altered sugar puckering.10,11
† Department of Molecular Chemistry, Osaka University.
‡ Department of Material and Life Science, Osaka University.
§ Inoue Photochirogenesis Project.
(1) (a) For reviews, see: Uhlmann, E.; Peyman, A. Chem. ReV. 1990,
90, 543. (b) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science
1991, 254, 1497. (c) Egholm, M.; Buchardt, O.; Nielsen, P. E.; Berg, R. H.
J. Am. Chem. Soc. 1992, 114, 9677.
(2) For a review, see: Milligan, J. F.; Matteucci, M. D.; Martin, J. C. J.
Med. Chem. 1993, 36, 1923.
(3) Sonveaux, E. Bioorg. Chem. 1986, 14, 274.
(8) (a) Suska, A.; Grajkowski, A.; Wilk, A.; Uznanski, B.; Blaszczyk,
J.; Wieczorek, M.; Stec, W. J. Pure Appl. Chem. 1993, 65, 707. (b) Stec,
W. J.; Wilk, A. Angew. Chem., Int. Ed. Engl. 1994, 33, 709.
(9) (a) Almarsson, O.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 9542. (b) Gryaznov, S. M.; Lloyd, D. H.; Chen, J.-K.; Schultz, R. G.;
DeDionisio, L. A.; Ratmeyer, L.; Wilson, W. D. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 5798.
(10) (a) Winkley, M. W.; Robins, R. K. J. Org. Chem. 1968, 33, 2822.
(b) Schwizer, M. P.; Witkowski, J. T.; Robins, R. K. J. Am. Chem. Soc.
1971, 93, 277. (c) Suck, D.; Saenger, W. J. Am. Chem. Soc. 1972, 94, 6520.
(d) Schwizer, M. P. Banta, E. B.; Witkowski, J. T.; Robins, R. K. J. Am.
Chem. Soc. 1973, 95, 3770.
(4) (a) Eckstein, F. Annu. ReV. Biochem. 1985, 54, 367. (b) Matsukura,
M.; Shinozuka, K.; Zon, G.; Mitsuya, H.; Reitz, M.; Cohen, J. S.; Broder,
S. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 7706.
(5) (a) Miller, P. S.; Yano, J.; Yano, E.; Carroll, C.; Jayaraman, K.; Ts’O,
P. O. P. Biochemistry 1979, 18, 5134. (b) Miller, P. S.; McParland, K. B.;
Jayaraman, K.; Ts’O, P. O. P. Biochemistry 1981, 20, 1874.
(6) (a) Letsinger, R. L.; Singman, C. N.; Histand, G.; Salunkhe, M. J.
Am. Chem. Soc. 1988, 110, 4470. (b) Gryaznov, S. M.; Letsinger, R. L.
Nucleic Acids Res. 1992, 20, 3403. (c) Chen, J.-K.; Schultz, R. G.; Lloyd,
D. H.; Gryaznov, S. M. Nucleic Acids Res. 1995, 23, 2661.
(7) (a) Uznanski, B.; Niewiarowski, W.; Stec, W. J. Tetrahedron Lett.
1982, 23, 4289. (b) Lesnikowski, Z. J.; Jaworska, M.; Stec, W. J. Nucleic
Acids Res. 1988, 16, 11675. (c) Fujii, M.; Ozaki, K.; Kume, A.; Sekine,
M.; Hata, T. Tetrahedron Lett. 1986, 26, 935.
(11) (a) Schirmer, R. E.; Davis, J. P.; Noggle, J. H.; Hart, P. A. J. Am.
Chem. Soc. 1972, 94, 2561. (b) de Kok, A. J.; Romers, C.; de Leeuw, H.
P. M.; Altona, C.; van Boom, J. H. J. Chem. Soc., Perkin Trans. 2 1997,
487.
10.1021/ja9935456 CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/06/2000

PRNA and Control of DNA Recognition
J. Am. Chem. Soc., Vol. 122, No. 29, 2000 6901
Chart 1
Indeed, bulky substituents introduced at the 6-position of the
pyrimidine base induce the syn orientation in the pyrimidine
nucleosides,10 while ketalization of the cis-2′,3′-diol of uridine
(1) enhances the syn/anti ratio of the resulting 2′,3′-cUMP or
2′,3′-O-isopropylidene uridine through the imposed 2′,3′-planar-
O4′-exo-furanose structure.12 However, the energetically disfa-
vored syn orientation has not been induced by external agents.
Figure 1. CD spectra of 1 in borate buffer (0.1 M KH2PO4 - 0.05 M
Na2B4O7, pH 7.2; solid line) and 2 in phosphate buffer (0.033 M KH2-
PO4 - 0.033 M Na2HPO4, pH 7.2; dashed line) and borate buffer (chain
line).
It is also well-known that boric acids form cyclic esters with
a variety of cis-1,2-diols, including sugars and ribonucleosides,
and this esterification process is reversible in aqueous solutions
at moderate pH.13 Recently, we have reported that the nucleo-
base orientation of 5′-amino-5′-deoxyuridine (2) is switched
from anti to syn simply by using borate buffer in place of
conventional phosphate buffer, while the parent uridine 1 does
not undergo such an orientational change in borate buffer, as
determined by circular dichroism (CD) spectroscopy.14 This
unprecedented orientation-switching phenomenon is expected
to provide us with a unique methodology for externally
controlling nucleic acid recognition, if a nucleic acid analogue,
in which these functionalities are carrying a ribose unit with
free 2′,3′-diol and 5′-amino/amide proton, can be prepared, since
both are indispensable to the orientation-switching process.14
These structural and functional requirements are not satisfied
through conventional modifications of RNA/DNA or peptide
nucleic acids (PNA). For this purpose, we wish to propose a
new category of nucleic acid analogues, that is, peptide
ribonucleic acids (PRNA), in which the ribonucleoside units
are not directly incorporated into the main chain, as is the case
with RNA, but are instead attached to the peptide backbone as
pendant groups. In a previous paper,15 we have demonstrated
that this strategy works well with an R-PRNA which carries
5′-amino-5′-deoxyuridine units appended to poly(L-glutamic
acid). However, the efficiency of external control was not very
high as a result of the mismatched repetition distance of the
nucleobases. In the present study, 5′-amino-5′-deoxyuridine (2)
and 5′-amino-5′-deoxycytidine (3) units are tethered to the
isopoly(L-glutamic acid) backbone at the remaining R-carboxyl
group of glutamic acid through the 5′-amino group of 2 and 3
(γ-PRNA, Chart 1), thus reserving the ribose cis-2′,3′-diol for
cyclic borate formation as well as the 5′-amide proton for
hydrogen bonding with the pyrimidine nucleobase 2-carbonyl.
In the γ-PRNA molecule, the nucleobases are located at the
correct position for RNA/DNA recognition, and the intervening
ribose unit is expected to greatly improve the poor solubility
of the PNA and its RNA/DNA complex in aqueous solution.
Results and Discussion
5′-Amino-5′-deoxyuridine (2) and 5′-Amino-5′-deoxycyti-
dine (3) as Essential Units. In the recognition process involving
DNA/RNA, the base orientation of nucleoside attached to the
glycosyl group plays a major role, where an anti orientation is
essential for base recognition. Hence, if the syn-anti orientation
of nucleoside analogues can be switched by an external factor,
this will be a convenient and powerful tool for control of nucleic
acid recognition.
Pyrimidine nucleosides are known to favor an anti nucleobase
orientation in solution phase.16 One plausible strategy to induce
the unfavorable syn orientation is to utilize the cooperative
effects of a hydrogen-bonding interaction between 2-carbonyl
and 5′-hydroxyl groups and a bridging substitution of the cis-
2′,3′-diol. The use of the 5′-hydroxyl proton, which is inevitably
lost in conventional nucleotides and nucleic acid model com-
pounds, is unrealistic as a candidate for the possible hydrogen-
bonding interaction with 2-carbonyl. We therefore substituted
the 5′-hydroxyl group of uridine with an amino group in this
study, and elucidated the orientational change and its effect in
the resultant 5′-amino-5′-deoxyuridine in the presence/absence
of borate, which was added as an external switching agent.
A new route to the selective transformation of uridine to 5′-
amino-5′-deoxyuridine (2) was also developed, which is an
extension of the conventional synthetic procedure employed in
the synthesis of 2′-amino-2′-deoxyuridine starting from 2,2′-
cyclouridine.17
5′-Amino-5′-deoxycytidine (3) was synthesized from cytidine
(4) through the reduction and deprotection of 5′-azido-5′-deoxy-
N4-benzoylcytidine.18
CD Spectral Study on Monomers. CD spectroscopy has
been used to evaluate the syn-anti orientation of nucleobases.14
Hence, the CD spectra of uridine (1) and 5′-amino-5′-deoxyuri-
dine (2) were measured in phosphate and borate buffer solutions
at pH 7.2 in order to examine the effects of borate buffer and
the amino substitution. As can be seen from the CD spectra
shown in Figure 1, the [θ]max value of 1 decreased from 9700
deg cm2 dmol-1 in phosphate buffer to 7400 deg cm2 dmol-1
(76% of the original value) in a borate buffer. This value, which
is similar to that observed for 2′,3′-O-isopropylidene uridine in
phosphate buffer (6000 deg cm2 dmol-1),14 suggests that the
(12) (a) Follmann, H.; Pfeil, R.; Witzel, H. Eur. J. Biochem. 1977, 77,
451. (b) Lavallee, D. K.; Myers, R. B. J. Am. Chem. Soc. 1978, 100, 3907.
(13) (a) For reviews, see: James. T. D.; Sandanayake, K. R. A. S.;
Shinkai, S. Supramol. Chem. 1995, 6, 141. (b) Westmark, P. R.; Valencia,
L. S.; Smith, B. D. J. Chromatogr., A 1994, 664, 123. (c) Westmark, P. R.;
Smith, B. D. J. Pharm. Sci. 1996, 85, 266.
(14) Wada, T.; Minamimoto, N.; Inaki, Y.; Inoue, Y. Chem. Lett. 1998,
1025.
(15) Minamimoto, N.; Wada, T.; Inaki, Y.; Inoue, Y. Nucleic Acids Res.
Submitted.
(16) (a) Rhodes, L. M.; Scimmel, P. R. Biochemistry 1971, 10, 4426.
(b) Saenger, W. Angew. Chem., Int. Ed. Engl. 1973, 12, 591.
(17) Kimura, J.; Fujisawa, Y.; Sawada, T.; Mitsunobu, O. Chem. Lett.
1974, 691.
(18) Hata, T.; Yamamoto, I.; Sekine, M. Chem. Lett. 1975, 977.

6902 J. Am. Chem. Soc., Vol. 122, No. 29, 2000
Wada et al.
Figure 2. CD spectra of 2 in phosphate buffer (0.033 M KH2PO4 -
0.033 M Na2HPO4, pH 7; solid line) and borate buffer (0.1 M KH2PO4
- 0.05 M Na2B4O7, pH 7.2; dashed line) and 3 in phosphate buffer
(dotted line) and borate buffer (chain line).
borate added to the solution appreciably induces the syn
orientation through ester formation, although this effect is
moderate. Conversely, 5′-amino-5′-deoxyuridine (2) gave a
smaller [θ]max value (5100 deg cm2 dmol-1, 53% of that for 1
in phosphate buffer) than that observed for 1 even in a phosphate
buffer (9700 deg cm2 dmol-1), probably as a result of an
intramolecular hydrogen-bonding interaction. Interestingly, the
use of a borate buffer led to a further decrease of the [θ]max
value to 1600 deg cm2 dmol-1 (16% of that for 1 in phosphate
buffer) which is almost comparable to the value observed for
5′-amino-5′-deoxy-2′,3′-O-isopropylidene uridine (1200 deg cm2
dmol-1) in a phosphate buffer.14
Figure 3. 1H NMR and NOE differential spectra (600 MHz) of 2
obtained with presaturation at H6 (δH 7.6) in (a) phosphate buffer (0.033
M KH2PO4 - 0.033 M Na2HPO4, pH 7.2) and (b) borate buffer (0.1
M KH2PO4 - 0.05 M Na2B4O7, pH 7.2).
A similar tendency was observed for cytidine 4 and 5′-amino-
5′-deoxycytidine (3) (Figure 2). When the solvent was changed
from phosphate to borate buffer, the [θ]max value of 4 was
reduced from 14200 to 12200 deg cm2 dmol-1 (86% of the
original value). This suggests that the syn/anti ratio was
increased by borate ester formation, although the effect is
moderate. In the case of 5′-amino-5′-deoxycytidine (3), the
[θ]max value was 8400 deg cm2 dmol-1 in phosphate buffer,
which was smaller than that for 4 in phosphate buffer, probably
as a result of the intramolecular hydrogen bonding interaction.
In borate buffer, the [θ]max value of 3 was further decreased to
4900 deg cm2 dmol-1 (39% of that for 4 in phosphate buffer)
as a consequence of the combined effect of the borate ester
formation and the intramolecular hydrogen bonding interaction.
This demonstrates that the simultaneous use of the 5′-amino
substitution and the borate buffer is quite effective in inducing
a syn orientation in both uridine and cytidine derivatives.
Nuclear Overhauser Effect. The unique borate-driven
orientation switching of nucleobase from anti to syn was further
confirmed by NMR NOE measurements made on 2. Figure 3
Figure 3a. This NOE profile is entirely consistent with the anti
orientation of 2, shown in Figure 3 (top right).
In borate buffer, 2 afforded a completely different NOE
spectrum. As shown in Figure 3b, the presaturation of uracil’s
H6 gave NOE signals only with uracil’s H5 (14.5%) and
furanose’s H1′ (12.5%). This unique NOE profile, in addition
to the decreased CD intensity discussed above, provides
evidence in support of the predominant contribution of the syn
orientation for 2 in borate buffer, which is driven by the borate
ester formation, as illustrated in Figure 3 (bottom right).
In the case of 5′-amino-5′-deoxycytidine (3), the syn orienta-
tion induced by borate ester formation was unambiguously
confirmed by the NOE difference spectra obtained in similar
NMR measurements in phosphate and borate buffers at pH 7.2.
In addition to the chemical shift changes and the varied coupling
constants which are characteristic of the 2′,3′-planar puckering
of the sugar, the NOE difference spectra provide good evidence
in support of the syn orientation of the pyrimidine base of 3 in
borate buffer. Measurements made in phosphate and borate
buffers afforded distinctly different NOE difference spectra upon
presaturation of the cytosine’s H6. In phosphate buffer, several
significant NOE peaks were observed for the cytosine’s H5 at
δ 6.0 (15.2%) and/or furanose’s H1′ at δ 5.7 (12.1%), H2′ at δ
4.3 (7.2%), H3′ at δ 4.1 (2.8%), and H5′ at δ 3.0 (1.9%). These
NOE peaks are compatible with the anti-orientated pyrimidine
base in 3. In contrast, in borate buffer, only the cytosine’s H5
(7.9%) and furanose’s H1′ (12.8%) gave moderate NOE peaks
upon irradiation of cytosine’s H6. This characteristic NOE
profile, as well as the decreased [θ]max value of CD spectrum
observed in borate buffer, demonstrate that the syn orientation
of 3 is efficiently induced by the cooperation of the 2′,3′-cyclic
1
shows the H NMR and NOE difference spectra of 2 in (a)
1
phosphate and (b) borate buffers. It is noted that the H NMR
spectrum in borate buffer is similar to that of 2′,3′-O-isopro-
pylideneuridine,12 and the spin coupling constants for 2 of J2′,3′
(6.8 Hz) and J1′,2′ and J3′,4′ (8.2 Hz) in borate buffer are
consistent with those of 2′,3′-cUMP.19 This indicates that in
borate buffer the sugar puckering of 2 is quite similar to that of
2′,3′-cUMP. In the NOE spectra in phosphate buffer, where
uracil’s H6 at δ 7.6 is presaturated, uracil’s H5 and/or furanose’s
H1′ at δ 5.8 (26.7%), H2′ at δ 4.4 (7.9%), H3′ at δ 4.1 (2.7%),
and H5′ at δ 3.0 (2.0%) gave notable NOE peaks, as shown in
(19) Davies, D. B.; Sadikot, H. J. Chem. Soc., Perkin Trans. 2 1983,
1251.

PRNA and Control of DNA Recognition
J. Am. Chem. Soc., Vol. 122, No. 29, 2000 6903
control of the recognition and complex formation of natural
nucleic acids, since they do not possess 5′-amino or free 3′-
hydroxyl groups on the furanose unit, both of which are the
prerequisites for this borate-mediated switch. Nevertheless, the
switching phenomenon observed for 2 encouraged us to explore
the scope and limitations of this new methodology in the control
of the recognition behavior of nucleic acids and their analogues.
To satisfy the structural requirements described above, we
decided to synthesize a series of peptide ribonucleic acids
(PRNAs). In the present study, the 5′-amino-5′-deoxyuridine
(2) unit is tethered to an isopoly(L-glutamic acid) backbone at
the remaining R-carboxylic group of glutamic acid through the
5′-amino group of 2, thus reserving the cis-2′,3′-diol of ribose
for cyclic borate formation, as well as leaving the 5′-amide
proton free to hydrogen bond with the pyrimidine 2-carbonyl.
In addition to the recognition switching ability, PRNAs may
possess several advantages over a conventional peptide nucleic
acid (PNA),1a,b such as a flexible side chain, which has built-in
chirality, a structural similarity to native RNA, and an improved
water solubility.
Before considering a polymeric system, we examined whether
the same switching strategy would work well with a monomer
model of PRNA. For this purpose, we synthesized the protected
monomer 5 by reacting 2 with N- and γ-C-protected L-glutamic
acid in 88% yield, using benzotriazol-1-yloxytris(dimethyl-
amino)phosphonium hexafluorophosphate (BOP reagent), as
shown in Scheme 1.21 The benzyl ester of 5 was then deprotected
to give the PRNA monomer (6) (Scheme 3), in which the 5′-
amino proton was replaced by a more acidic amide proton
located at a position suitable for hydrogen bonding to the uracil’s
2-carbonyl. To confirm the possible orientation switching
induced by borate, CD and NMR spectral studies were
performed with 6.
The CD spectra of 6 were recorded in borate and phosphate
buffers under comparable conditions, but gave distinctly dif-
ferent molar ellipticities at the CD maxima at ∼270 nm, as
shown in Figure 5. Thus, the [θ]max value obtained in borate
buffer was 5000 deg cm2 dmol-1, which was significantly
smaller than the value of 6500 deg cm2 dmol-1 obtained in
phosphate buffer. This CD spectral behavior is similar to, but
less intense than, that observed for 2 (Figure 1). These
observations suggest that the borate-induced anti-to-syn orien-
tational change occurs in the PRNA monomer 6, but to a lesser
extent.
The orientation switching of 6 by borate was further verified
by NOE difference spectra recorded in (a) phosphate and (b)
borate buffers (Figure 6). In phosphate buffer, presaturation of
uracil’s H6 at δ 7.62 led to significant NOE peaks for uracil’s
H5 at δ 5.86 (19.7%) and furanose’s H1′ at δ 5.78 (5.7%), H2′
at δ 4.30 (8%) and H3′ at δ 4.08 (1%), verifying the anti
orientation of the base. In contrast, NOE enhancement was
observed only for H5 at δ 5.86 (10.5%) and H1′ at δ 5.76 (4.5%)
in borate buffer upon presaturation of H6 at δ 7.61, which is
consistent with a syn orientation. It is interesting to note that
small NOE peaks (<1% enhancement) at δ 4.1 and 4.3,
attributable to H3′ and H2′, are also observed for 6 in borate
buffer, although no such NOE peaks are observed for 2 under
the same conditions. These small NOE peaks, in conjunction
with the above-mentioned smaller CD spectral change observed
in borate buffer for 6 than formation for 2, indicate a less-
effective cooperation of the borate ester formation and the
hydrogen bonding interaction in the uridine-based 6. This lower
Figure 4. CD spectra changes of 2 with increasing concentrations of
(a) Na2B4O7 and (b) B(OH)3 in phosphate buffer (0.033 M KH2PO4 -
0.033 M Na2HPO4, pH 7.2); (a) [1] ) 1.0 × 10-4 M, (b) [1] ) 7.1 ×
10-5 M.
esterification and the hydrogen bonding interaction between the
5′-amino and 2-carbonyl groups.
Effects of Borate Ester Formation upon Orientation of
5′-Amino-5′-deoxyuridine (2). In agreement with the observa-
tion that 2 displays distinctly different CD spectra in phosphate
and borate buffers, the gradual addition of borax to the phosphate
buffer solution of 2 (pH 7.2) induced large CD spectral changes,
as shown in Figure 4a. Isosbestic points that are observed at
∼220 and 240 nm indicate that no multistep processes or
intermediates are involved but that a single step is responsible
for these changes. The molar ellipticity at the peak around 270
nm ([θ]max) was substantially reduced from 5200 to 1600 deg
cm2 dmol-1 as the borax concentration was increased from 0
to 50 mM. The final [θ]max value of 1600 deg cm2 dmol-1 is
close to that (1400 deg cm2 dmol-1) observed for 5′-amino-5′-
deoxy-2′,3′-O-isopropylidene uridine under the same condi-
tions.14 This confirms that the addition of borax enhances the
syn/anti ratio of nucleobase in 2 through the reversible formation
of a borate ester with the cis-2′,3′-diol of ribose. A quantitative
analysis of the CD spectral changes, using the nonlinear least-
squares fit to the curve for 1:1 stoichiometric complexation,
gave the equilibrium constant of 75 M-1 for the reversible
esterification of 2 with borax. This value is well within the
typical range (10-104 M-1) of the equilibrium constant for
borate ester formation with cis-1,2-diols in aqueous solutions
around pH 7.13b,20
Analogous CD spectral changes were observed upon addition
of boric acid of up to 84 mM concentration to a buffer solution
at pH 7.2 containing 2, again with accompanying isosbestic
points at the same wavelengths (Figure 4b). An identical final
[θ]max value of 1400 deg cm2 dmol-1 was obtained for 84 mM
boric acid, indicating a high syn content, induced by borate ester
formation. An equilibrium constant of 49 M-1 was calculated
for boric acid, and methylboric acid gave a similar equilibrium
constant of 52 M-1 in a buffer solution at pH 8.3. The addition
of phenylboric acid of up to 6 mM induced comparable CD
spectral changes at 270 nm, indicating a more efficient ester
formation, although the absorption of this reagent in the same
wavelength region could not permit us to make a quantitative
analysis of these results.
Peptide Ribonucleic Acid (PRNA) Monomer (5). The
above methodology, employing the reversible borate ester
formation to switch the orientation of the nucleobase, is
interesting and potentially useful. However, it is obvious that
this strategy cannot immediately be applied to the external
(20) (a) Soundararajan, S.; Badawi, M.; Kohlrust, C. M.; Hageman, J.
H. Anal. Biochem. 1989, 178, 125. (b) Singhal, R. P.; Ramamurthy, B.;
Govindraj, N.; Sarwar, Y. J. Chromatogr. 1991, 543, 17.
(21) (a) Rivaille, P.; Gautron, J. P.; Castro, B.; Milhaud, G. Tetrahedron
1980, 36, 3413. (b) Wenger, R. M. HelV. Chim. Acta 1984, 67, 502.

6904 J. Am. Chem. Soc., Vol. 122, No. 29, 2000
Wada et al.
Scheme 1a
a
Reagents and conditions: i, Boc-Glu(OBzl), BOP reagent, HOBt, (i-Pr)2NEt, DMF, rt, 1 h; ii, H2, Pd/C(10%), MeOH, rt, 2 h
affording 7 with a free amino terminus. The subsequent
condensation of 6 and 7 using the BOP reagent yielded the
dimeric PRNA (8). This synthetic cycle was repeated three times
to give the protected 8-mer (14). Each coupling process
proceeded in 70-98% yield. The benzyl group of 14 was
subsequently removed to give the N-protected 8-mer (15), to
the C-terminus of which was introduced Nꢀ-(2-chlorobenzyl-
oxycarbonyl)-O-benzyl-L-lysine (Lys(ClZ)-OBzl), as shown in
Scheme 3. 16 was completely deprotected using trimethylsilyl
triflate (TMSOTf),22 affording the γ-PRNA oligomer (17).
Using the above synthetic strategy, it is possible to introduce
different bases into the sequence of γ-PRNA, because the
oligomeric γ-PRNA was obtained by elongation of monomeric
γ-PRNA. In the present study, the γ-PRNA monomer (18),
which carries a cytidine unit, was also prepared in a manner
similar to that employed in the preparation of 5, and the Boc
group of 18 was removed by treatment with TFA to afford 19,
which has a free amino group. The BOP mediated condensation
of 19 with 6 yielded a dimeric γ-PRNA (20), which has a uridine
unit and a cytidine unit at the amino and carboxyl terminals of
glutamic acid, respectively. Repeated elongation, using 20,
followed by the subsequent introduction of lysine derivative and
final deprotection, afforded the 8-mer (21), which is composed
of alternate uridine and cytidine units with a lysine at the
C-terminus (Scheme 4). After gel filtration using Sephadex
G-25, the oligomeric γ-PRNAs 17 and 21 were purified by
reverse phase HPLC.
Figure 5. CD spectra of 6 in phosphate buffer (0.033 M KH2PO4 -
0.033 M Na2HPO4, pH 7.2; solid line) and borate buffer (0.1 M KH2-
PO4 - 0.05 M Na2B4O7, pH 7.2; dashed line).
CD Spectral Study on γ-PRNA Oligomers. The CD spectra
of γ-PRNA 8-mer (17) were measured in both borate and
phosphate buffers, under comparable conditions (Figure 7). In
phosphate buffer, the [θ]max value obtained (5200 deg cm2
dmol-1) was nearly the same as that observed for 5′-amino-5′-
dexoyuridine (2), and this is compatible with the preferred anti
orientation in phosphate buffer. In contrast, the [θ]max value of
this oligomer was greatly reduced to 650 deg cm2 dmol-1 in
borate buffer, which is significantly smaller than the value
observed for 2 in borate buffer (1600 deg cm2 dmol-1). The
reduced [θ]max value observed for PRNA 8-mer 17 in borate
buffer is important, since the use of the borate buffer, in place
of phosphate buffer, caused only a small decrease from 6500
to 5000 deg cm2 dmol-1 with the PRNA monomer (6). These
results strongly indicate the predominant syn orientation of
nucleobase in borate buffer, probably through cooperative effects
of adjacent nucleobases in this oligomer.
Figure 6. NOE differential spectra (600 MHz) of 6 obtained with
presaturation at H6 (δH 7.6) in (a) phosphate buffer (0.033 M KH2PO4
- 0.033 M Na2HPO4, pH 7.2) and (b) borate buffer (0.1 M KH2PO4 -
0.05 M Na2B4O7, pH 7.2).
efficiency may be the result of the steric effect of the
tert-butoxycarbonyl protecting group and the increased hydro-
philicity and extensive hydration around the peptide backbone.
Synthesis of Peptide Ribonucleic Acids with Oligo(γ-L-
glutamic acid) Backbone (γ-PRNA). A series of oligomeric
PRNAs were prepared by progressive elongation reactions
composed of a repeated selective deprotection-condensation
cycle, as shown in Scheme 2. In the first step, the benzyl group
of the protected PRNA monomer (5) was removed by catalytic
hydrogenation to give 6 with a free carboxyl terminus, while
the Boc group of 5 was removed by treatment with TFA,
A similar tendency was also observed with observed oligomer
21. As shown in Figure 8, the [θ]max value (2100 deg cm2
dmol-1) in borate buffer was much smaller than that observed
in phosphate buffer (7800 deg cm2 dmol-1). This indicates that
the change in base orientation from anti to syn in 21 is induced
by borate ester formation at each nucleoside unit.
(22) Fujii, N.; Otaka, A.; Ikemura, O.; Akaji, K.; Funakoshi, S.; Hayashi,
Y.; Kuroda, Y.; Yajima, H. J. Chem. Soc., Chem. Commun. 1987, 274.

PRNA and Control of DNA Recognition
J. Am. Chem. Soc., Vol. 122, No. 29, 2000 6905
Scheme 2a
a
Reagents and conditions: i, H2, Pd/C(10%), MeOH-DMF (10:0-1: 3 v/v), rt, 2-6 h; ii, TFA, 0 °C, 30 min; iii, BOP reagent, HOBt, (i-
Pr)2NEt, DMF, rt, 2-4 h
Scheme 3a
a
Reagents and conditions: i, Lys(CIZ)-OBzl, BOP reagent, HOBt, (i-Pr)2NEt, DMF, rt, 4 h; ii, m-cresol, PhSMe, TMSOTf, TFA, 0 °C, 1 h;
iii, 5% NH3 aq 0 °C, 30 min, then 4 N HCl/dioxane, 0 °C, 30 min
The CD spectra was also measured at various concentrations
of borax added to a phosphate buffer solution of 17. As shown
in Figure 9, the CD intensity continuously decreased as the
concentration of borax increased, leveling off between 3.7 and
10 mM. The CD spectra observed at borax concentrations of
borax >3.7 mM is almost superimposable upon that obtained
in borate buffer which contains 20 mM of borax. The gradual
CD spectral changes, accompanied by the isosbestic points
observed at ∼225 and 250 nm, indicate clearly that the reversible
anti-to-syn orientation switching process is caused by a coopera-
tive cyclic borate ester formation and hydrogen bonding
interaction, as observed for the 5′-amino-5′-deoxyuridine and
the γ-PRNA monomer 6. A quantitative treatment of these CD
spectral changes, using the nonlinear least-squares fitting to the
curve for 1:1 stoichiometric complexation, gave the equilibrium
constant of 750 M-1 for the formation of borate ester for each
ribonucleoside, which is exactly 10 times greater than that
obtained for 2 under the comparable conditions. The higher
binding constant and much reduced CD intensity at high
concentrations of borax observed for 17 indicate that the borate
ester formation and the hydrogen-bonding interaction syner-
getically enhance the anti-to-syn orientation switching of the
uracil moiety in oligomeric γ-PRNA. This may be attributed
to the increased hydrophobicity, and thus a reduced hydration
around the peptide backbone, which is thought to form a right-
handed, loose helical conformation based on the weak, positive
exciton coupling of the major CD band at 270 nm, and also
from the hypochromic change of 5% observed in the UV
spectrum.
Control of Hybridization of γ-PRNA with d(A)8 by Borate
Ester Formation. The hybridization ability of γ-PRNA 17 with
complementary d(A)8 was evaluated from the melting temper-
ature, Tm, which was determined from hypochromic changes
observed in the UV spectrum upon mixing 17 with the target
oligonucleotide 8-mer. The Tm value, taken as a quantitative
measure of the binding affinity, is known to be affected by the
conformation of a nucleic acid or its analogue,9 as well as the
ionic strength of the aqueous solution used. To elucidate the
effects of borate on the formation and stability of the PRNA-
DNA hybrid, the melting profiles for the γ-PRNA 17 pairs and
the reference compound (T)8 with the complementary d(A)8
were measured independently in phosphate buffer with and

6906 J. Am. Chem. Soc., Vol. 122, No. 29, 2000
Wada et al.
Scheme 4a
a
Reagents and conditions: i, Boc-Glu(OBzl), BOP reagent, HOBt, (i-Pr)2NEt, DMF, rt, 1 h; ii, TFA, 0 °C, 30 min; iii, 6, BOP reagent, HOBt,
(i-Pr)2NEt, DMF, rt, 2 h
Figure 7. CD Spectra of 17 in phosphate buffer (0.033 M KH2PO4 -
Figure 8. CD Spectra of 21 in phosphate buffer (0.033 M KH2PO4 -
0.033 M Na2HPO4, pH 7.2; solid line) and borate buffer (0.1 M KH2-
0.033 M Na2HPO4, pH 7.2; solid line) and borate buffer (0.1 M KH2-
PO4 - 0.05 M Na2B4O7, pH 7.2; dashed line).
PO4 - 0.05 M Na2B4O7, pH 7.2; dashed line).
without borax (20 mM). Table 1 summarizes the Tm values and
maximum hypochromicity observed in these two buffer solu-
tions. The stoichiometry of the complex with d(A)8 was
determined to be 1:1 in each case (uracil or thymine:adenine
unit ratio) using the Job plot of the hypochromic change upon
mixing (see Experimental), and triplex formation is therefore
not to occur under the conditions employed. Control runs using
noncomplementary (T)8 in place of d(A)8 were also carried out
under comparable conditions and found to show no hypochro-
micity or appreciable melting behavior. This confirms the base-
specific interaction of γ-PRNA 17 with the complementary
d(A)8. In borax-free buffer, the hybrid complex between
γ-PRNA 17 and d(A)8 gave a considerably higher Tm of 52.8
°C than the complementary (T)8-d(A)8 duplex (Tm ) 45.0 °C),
indicating a stronger interaction in the hybrid than in the natural
pair. To the best of our knowledge, this is the first example of
tight binding of the oligomers of nucleobases attached to a
γ-glutamyl peptide backbone.
hypochromic changes, while the complementary (T)8-d(A)8
duplex gave an even higher Tm of 56.9 °C and hypochromicity
of 40%, presumably due to the slight increase in the ionic
strength. This contrasting behavior between the natural and
hybrid pairs in the presence/absence of borax is most likely
attributable to the anti-to-syn orientation switching of the uracil
base in γ-PRNA 17, for which a cooperative borate ester
formation at the cis-2′,3′-diol and a hydrogen bonding interaction
between the 5′-amide proton and the 2-carbonyl oxygen are
responsible. However, the electrostatic repulsion between the
adjacent anionic borate esters makes some contribution, as
illustrated in Scheme 5.
The complexation behavior of γ-PRNA 21, which possesses
alternating uridine and cytidine units, was studied by using two
types of complementary oligonucleotides with a “parallel” or
5′-d(AG)4-3′ sequence and an “antiparallel” 5′-d(GA)4-3′ se-
quence. The antiparallel oligonucleotide is expected to form a
complex with γ-PRNA 21 from the N-terminus to the 3′-end,
while the parallel oligonucleotide should bind 21 from the
N-terminus to 5′-end. If γ-PRNA 21 is able to recognize the
In contrast, in the borax-containing buffer solution, the hybrid
complex did not exhibit any melting behavior above 0 °C, or

PRNA and Control of DNA Recognition
J. Am. Chem. Soc., Vol. 122, No. 29, 2000 6907
Conclusions
Current studies on antisense molecules are directed mostly
toward the simple inhibition of genetic information transfer.
Little effort has been devoted to the active control of the DNA/
RNA recognition processes, possibly as a result of a lack of
suitable fundamental strategy and the necessary practical tools.
In this and preceding studies,14,15 we have proposed a new
methodology and effective tools for controlling DNA/RNA
recognition through the use of an external agent. This strategy
employs novel nucleic acid analogues, that is, R- and γ-peptide
ribonucleic acids (R- and γ-PRNAs), as the recognition moiety
which has a built-in switch triggered by an external factor. The
formation of a borate ester of the ribose’s cis-2′,3′-diol and the
simultaneous hydrogen-bonding interaction between the 5′-
amide proton and the 2-carbonyl oxygen act as the external and
internal switching devices. The results obtained in both studies
are encouraging, demonstrating that R- and γ-PRNA form stable
hybrid complexes with the complementary oligonucleotides,
which are readily dissociated upon addition of borax or boric
acid. Certainly, a definition of the scope and limitations of this
strategy remain to be addressed, including the study of sequence
selectivity and the search for other switching devices/agents.
Moreover, the present concept of the external control of DNA/
RNA recognition has the potential to be widely employed in
the next generation of antisense molecules.
Figure 9. CD spectra changes of 17 with increasing concentrations
of Na2B4O7 in phosphate buffer (0.033 M KH2PO4 - 0.033 M Na2-
HPO4, pH 7.2); [17] ) 1.0 × 10-4 M.
Table 1. Melting Temperatures (Tm) of γ-PRNA-Oligonucleotide
Complexesa
Tm/°C
additive
γ-PRNA or
oligodeoxyribonucleotide
20 mM
none borax
complement
Experimental Section
17
17
(T)8
21
21
d(A)8
(T)8
d(A)8
52.8 <0
<0
<0
Standard abbreviations for amino acids and protecting groups are
as recommended by the IUPAC-IUB Commission on Biochemical
Nomenclature. Other abbreviations include: HOBt, 1-hydroxybenzo-
triazole; BOP, benzotriazol-1-yloxytris(dimethylamino)phosphonium
hexafluorophosphate; TFA, trifluoroacetic acid; isoGln(5′U), N5-(5′-
deoxy-5′-uridyl)-L-isoglutamine residue; ClZ, 2-chlorobenzyloxycar-
bonyl group. All starting materials, reagents, and solvents were
commercially available and used without further purification. Uridine
was purchased from Seikagaku Co. (Tokyo, Japan). Boc-L-amino acids
and 1-hydroxybenzotriazole (HOBt) were purchased from Peptide
Institute, Inc. (Osaka, Japan). Oligonucleotides were purchased from
Takara Co. (Kyoto, Japan). Other chemicals of guaranteed grade were
purchased Tokyo Kasei Kogyo Co. (Tokyo, Japan) or Wako Pure
Chemical Industries, Ltd. (Osaka, Japan). Column chromatography was
performed over 70-230 mesh silica gel from Merck. Gel filtration was
performed over Sephadex G-25 from Pharmacia. Nucleic acid model
was purified on a pre-packed Hiber column RT by elution with 10%
acetonitrile in water at a flow rate of 3 mL/min, using a recycling
preparative HPLC instrument LC-908 from Japan Analytical Industries.
45
56.9
d(ApGpApGpApGpApG) 41.6 <0
d(GpApGpApGpApGpA) 49.2 <0
a Solvent: phosphate buffer (0.033 M KH2PO4 - 0.033 M Na2HPO4,
pH 7.2).
Scheme 5. Proposed Recognition Control Model of 17 with
Complementary d(A)
1
IR spectra were recorded on a JASCO FT/IR-230 spectrometer. H
antiparallel orientation correctly, the parallel complex should
not be observed.23 As shown in Table 1, the antiparallel target
gives a Tm of 49.2 °C, which is appreciably higher than that
obtained for the parallel complex (Tm ) 41.6 °C). This result
may indicate that γ-PRNA 21 tightly binds to antiparallel target
sequence rather than parallel one, although the possibility of
one-base shifted binding to the parallel target in the antiparallel
fashion cannot rigorously be ruled out.
Interestingly, the addition of borax to these systems resulted
in a complete loss of Tm, indicating the dissociation of the
complex through the anti-to-syn orientation switching of the
base. It should be emphasized that oligomeric γ-PRNA and the
complementary antiparallel DNA form a complex that is more
stable than the equivalent natural DNA-DNA duplex. Further-
more, the on-off switching of nucleic acid recognition has been
realized for the first time using the borate-induced orientation
switching of γ-PRNA.
NMR spectra were obtained on a JEOL GSX-270 at 270 MHz or a
Varian INOVA-600 at 600 MHz. Chemical shifts are reported as δ
value in parts per million (ppm) relative to tetramethylsilane (δH
)
0.0 ppm) as internal standard. Mass spectral measurements were
performed on a JEOL AX-500 instrument by fast-atom bombardment
(FAB) ionization with nitrobenzyl alcohol (NBA) as a matrix and a
Voyager RP from PerSeptive Biosystems with R-cyano-4-hydroxycin-
namic acid (R-CHCA) or picolinic acid as a matrix (MALDI-TOF).
CD spectra were recorded on a JASCO J-720W spectrophotometer.
Thermal dissociation experiments were performed on a JASCO V-550
UV/vis spectrophotometer equipped with a temperature controller.
Absorbance data at 260 nm was collected at 10 s intervals upon heating
a solution of nucleic acid model (6 × 10-5 M) and oligonucleotide (6
× 10-5 M) at a rate of 0.5 °C/min to give the melting curve.
Equilibrium constants (KS) were determined for the borate ester
formation of γ-PRNAs from the quantitative analysis of the CD spectral
changes induced by adding boric acid or borax (3.7-84 mM) to a
borate-free buffer solution of 5′-aminonucleoside or γ-PRNA (0.7-1
× 10-4 M). Nonlinear least-squares analysis of the differential molar
ellipticity data (∆[θ]max) led to an excellent fit to the 1:1 stoichiometric
(23) Egholm, M.; Nielsen, P. E.; Buchardt, O.; Berg, R. H. J. Am. Chem.
Soc. 1992, 114, 9677.

6908 J. Am. Chem. Soc., Vol. 122, No. 29, 2000
Wada et al.
curve, affording the equilibrium constants for the esterification of 2,
3, and 17 with boric acid or borax.
After 30 s of stirring at 0 °C, 7 (1.68 g, 2.92 mmol) was added, and
the resultant mixture was stirred for 2 h at room temperature. The
solvent was removed under reduced pressure, and the residue was
purified by column chromatography on silica gel with chloroform-
Complex stoichiometry was determined from the Job plot, where
the interaction of PRNA and oligonucleotide was monitored by the
hypochromic changes at 260 nm in UV spectra, as reported previously.24
In these experiments, the molar fractions of γ-PRNA and oligonucle-
otide were continuously varied, while the combined concentration was
fixed at 10-4 M. In all cases examined, the induced hypochromicity
was maximized at a molar fraction of 0.5, indicating the formation of
1:1 complex between PRNA and oligonucleotide.
methanol (10:1 v/v) to give compound 8 as a powder (1.70 g, 70%),
1
ν
max(KBr)/cm-1 3420, 1680, 1540, 1380, 1270, and 1200; H NMR
(270 MHz, DMSO-d6) δ 1.36 (9 H, s, t-Bu-H), 1.80 (4 H, m, â-CH2),
2.15 (2 H, t, Jγ,â ) 7.1, γ-CH2), 2.35 (2 H, t, Jγ,â ) 7.8, γ-CH2), 3.83
(5 H, m, Boc-NHCH, 3′-H and 4′-H), 4.03 (2 H, q, J2′,1′ ) J2′,3′ ) 4.4,
2′-H), 4.26 (1 H, q, JCH,NH ) JCH,â ) 7.3, R-CH), 5.06 (2 H, s, PhCH2),
5.16 (2 H, s, 3′-OH), 5.39 (2 H, d, J OH,2′ ) 4.4, 2′-OH), 5.62 (1 H, d,
Nr-tert-Butoxycarbonyl-N5-(5′-deoxy-5′-uridyl)-L-isoglutaminBe en-
zyl Ester (Boc-isoGln(5′U)-OBzl) (5). To a solution of N-tert-
butoxycarbonyl-L-glutamic acid γ-benzyl ester (2.52 g, 7.47 mmol),
HOBt (1.01 g, 7.47 mmol), and BOP reagent (3.30 g, 7.47 mmol) in
DMF (100 mL) was added diisopropylethylamine (1.30 mL, 7.47
mmol). After 30 s of stirring at 0 °C, 5′-amino-5′-deoxyuridine (2)18
(2.00 g, 8.22 mmol) was added, and the mixture was stirred for 1 h at
room temperature. The solvent was removed under reduced pressure
and the residue was suspended in ethyl acetate (200 mL). The
suspension was successively washed with equal-volume portions of
4% sodium hydrogen carbonate, 5% potassium hydrogen sulfate, and
brine. The organic layer was dried over magnesium sulfate, filtered,
and concentrated under reduced pressure. The residue was purified by
column chromatography on silica gel with chloroform-methanol (30:1
v/v) to give compound 5 as a powder (3.70 g, 88%), νmax(KBr)/cm-1
J5,6 ) 8.3, 5-H), 5.64 (1 H, d, J5,6 ) 7.8, 5-H), 5.73 (2 H, d, J1′,2′
)
5.9, 1′-H), 6.85 (1 H, d, JNH,CH ) 8.3, Boc-NH), 7.35 (5 H, m, Ar-H),
7.63 (1 H, d, J6,5 ) 7.8, 6-H), 7.64 (1 H, d, J6,5 ) 7.8, 6-H), 7.95 (2 H,
m, 5′-NH), 8.14 (1 H, t, JNH,CH ) 10.7, R-NH), 11.33 (2 H, s, 3-NH);
MALDI-TOF HRMS (R-CHCA), m/z found 939.34 (M + Na),
calculated 939.335.
Boc-(isoGln(5′U))2-OH (9). Palladium on activated carbon (10%;
∼0.1 g) was added to a solution of 8 (0.850 g, 0.927 mmol) in methanol
(50 mL). After 4 h of continued stirring under hydrogen atmosphere
(1 bar) at room temperature, the reaction mixture was filtered, and the
filtrate was evaporated under reduced pressure to give compound 9 as
a powder (0.736 g, 96%), νmax(KBr)/cm-1 3390, 1680, 1540, 1470,
1390, and 1280; 1H NMR (270 MHz, DMSO-d6) δ 1.37 (9 H, s, t-Bu-
H), 1.67-1.81 (4 H, m, â-CH2), 2.14 (4 H, m, γ-CH2), 3.58 (4 H, m,
5′-H), 3.92 (4 H, m, 3′-H and 4′-H), 4.03-4.24 (3 H, m,2′-H and Boc-
NHCH), 4.42 (1 H, q, JCH,NH ) JCH,â ) 7.1, R-CH), 5.17-5.39 (4 H,
m, 2′-OH and 3′-OH), 5.67 (2 H, d, J5,6 ) 8.3, 5-H), 5.86 (2 H, d, J1′,2′
1
3420, 1680, 1520, 1460, 1390, 1250, and 1170; H NMR (270 MHz,
DMSO-d6) δ 1.37 (9 H, s, t-Bu-H), 1.77-1.88 (2 H, m, â-CH2), 2.36
(2 H, t, Jγ,â ) 7.8, γ-CH2), 3.29 (2 H, m, 5′-H), 3.83 (2 H, m, 3′-H and
4′-H), 3.94 (1 H, q, JCH,NH ) JCH,â ) 7.3, Boc-NHCH), 4.01 (1 H, q,
) 7.8, 1′-H), 6.99 (1 H, d, JNH,CH ) 8.8, Boc-NH), 7.67 (2 H, d, J6,5
)
8.8, 6-H), 7.88 (2 H, m, 5′-NH), 8.16 (1 H, m, R-NH), 11.35 (2 H, s,
3-NH); MALDI-TOF HRMS (R-CHCA), m/z found 849.29 (M + Na),
calculated 849.288.
J2′,1′ ) J2′,3′ ) 5.4, 2′-H), 5.07 (2 H, s, PhCH2), 5.18 (1 H, d, JOH,3′
)
4.9, 3′-OH), 5.40 (1 H, d, J ) 5.4, 2′-OH), 5.63 (1 H, d, J5,6 8.3,
OH,2′
5-H), 5.74 (1 H, d, J1′,2′ ) 5.9, 1′-H), 6.94 (1 H, d, JNH,CH ) 7.8, Boc-
NH), 7.35 (5 H, m, Ar-H), 7.64 (1 H, d, J6,5 ) 7.8, 6-H), 8.00 (1 H, t,
JNH,5′ ) 5.6, 5′-NH), 11.35 (1 H, s, 3-NH); Found: C, 55.06; H, 6.02;
N, 9.76. Calcd for C26H34N4O10‚1/4H2O: C, 55.07; H, 6.13; N, 9.88%;
MALDI-TOF HRMS (R-CHCA), m/z found 585.22 (M + Na),
calculated 585.217.
Nr-tert-Butoxycarbonyl-N5-(5′-deoxy-5′-uridyl)-L-isoglutamine (Boc-
isoGln(5′U)-OH) (6). Palladium on activated carbon (10%; ∼0.2 g)
was added to a solution of 5 (1.80 g, 3.20 mmol) in methanol (50 mL).
After 2 h of continued stirring under hydrogen atmosphere (1atm), the
reaction mixture was filtered, and the filtrate was evaporated under
TFA‚(isoGln(5′U))2-OBzl (10). 8 (0.850 g, 0.927 mmol) was
dissolved in TFA (5 mL), and the solution was kept at 0 °C for 30
min. TFA was removed under reduced pressure, and ether (100 mL)
was added to give compound 10 as a powder (0.846 g, 98%), νmax
-
(KBr)/cm-1 3420, 1680, 1540, 1470, 1280, 1200, and 1140; H NMR
(270 MHz, DMSO-d6) δ 1.89 (4 H, m, â-CH2), 2.25-2.34 (4 H, m,
γ-CH2), 3.59 (4 H, m, 5′-H), 3.86 (6 H, m, NH3+CH, R-CH, 3′-H and
4′-H), 4.11 (2 H, m, 2′-H), 5.08 (2 H, s, PhCH2), 5.20 (4 H, m, 3′-,
2′-OH), 5.70 (2 H, d, J5,6 ) 7.8, 5-H), 5.77 (2 H, d, J1′,2′ ) 5.9, 1′-H),
7.36 (5 H, m, Ar-H), 7.89 (1 H, m, 5′-NH), 8.10 (4 H, m, R-NH and
NH3+), 8.56 (1 H, m, 5′-NH), 11.32 (2 H, s, 3-NH); MALDI-TOF
HRMS (R-CHCA), m/z found 817.30 (M - CF3COO), calculated
817.301.
1
reduced pressure to give compound 6 as a powder (1.44 g, 95%), νmax
-
(KBr)/cm-1 3320, 1680, 1530, 1460, 1390, 1250, and 1170; H NMR
(270 MHz, DMSO-d6) δ 1.37 (9 H, s, t-Bu-H), 1.69-1.82 (2 H, m,
â-CH2), 2.19 (2 H, t, Jγ,â )7.3, γ-CH2), 3.29-3.39 (2 H, m, 5′-H),
3.83 (2 H, m, 3′-H and 4′-H), 3.92 (1 H, q, JCH,NH ) JCH,â ) 5.9, Boc-
NHCH), 4.02 (1 H, t, J2′,1′ ) J2′,3′ ) 4.9, 2′-H), 5.63 (1 H, d, J5,6 ) 8.3,
5-H), 5.73 (1 H, d, J1′,2′ ) 5.9, 1′-H), 6.92 (1 H, d, JNH,CH ) 8.3, Boc-
NH), 7.65 (1 H, d, J6,5 ) 7.8, 6-H), 7.98 (1 H, t, JNH,5′ ) 5.4, 5′-NH),
11.35 (1 H, s, 3-NH); MALDI-TOF HRMS (R-CHCA), m/z found
495.17 (M + Na), calculated 495.170.
1
Boc-(isoGln(5′U))4-OBzl (11). To a solution of 9 (0.662 g, 0.801
mmol), HOBt (0.108 g, 0.801 mmol), and BOP reagent (0.354 g, 0.801
mmol) in DMF (30 mL) was added diisopropylethylamine (0.29 mL,
1.68 mmol). After 30 s of stirring at 0 °C, 10 (0.820 g, 0.881 mmol)
was added, and the mixture was stirred for 2 h at room temperature.
The solvent was removed under reduced pressure, and ethanol was
added to give compound 11 as a powder (1.22 g, 94%), νmax(KBr)/
1
cm-1 3300, 1680, 1540, 1460, 1390, 1270, and 1080; H NMR (270
N5-(5′-Deoxy-5′-uridyl)-L-isoglutamine Benzyl Ester Trifluoro-
acetic Acid Salt (TFA‚isoGln(5′U)-OBzl) (7). 5 (1.80 g, 3.20 mmol)
was dissolved in TFA (10 mL), and the solution was kept at 0 °C for
30 min. TFA was removed under reduced pressure, and ether (200 mL)
was added to give compound 12 as a powder (1.81 g, 98%), νmax(KBr)/
MHz, DMSO-d6) δ 1.37 (9 H, s, t-Bu-H), 1.71-1.82 (8 H, m, â-CH2),
2.14 (6 H, m, γ-CH2), 2.35 (2 H, t, Jγ,â ) 7.6, γ-CH2), 3.26-3.44 (8
H, m, 5′-H), 3.83 (9 H, m, Boc-NHCH, 3′-H and 4′-H), 4.04 (4 H, m,
2′-H), 4.22 (3 H, m, R-CH), 5.06 (2 H, s, PhCH2), 5.16 (4 H, d, JOH,3′
1
cm-1 3420, 1680, 1560, 1460, 1270, 1200, and 1130; H NMR (270
) 4.4, 3′-OH), 5.39 (4 H, d, JOH,2′ ) 5.4, 2′-OH), 5.65 (4 H, d, J5,6
)
7.3, 5-H), 5.73 (4 H, d, J1′,2′ ) 3.9, 1′-H), 6.87 (1 H, d, JNH,CH ) 7.8,
Boc-NH), 7.35 (5 H, m, Ar-H), 7.65 (4 H, d, J6,5 ) 8.3, 6-H), 7.96 (4
H, m, 5′-NH), 8.12 (3 H, m, R-NH), 11.34 (4 H, s, 3-NH); MALDI-
TOF HRMS (R-CHCA), m/z found 1647.57 (M + Na), calculated
164.570.
Boc-(isoGln(5′U))4-OH (12). Palladium on activated carbon (10%;
∼0.1 g) was added to a solution of 11 (0.600 g, 0.369 mmol) in
methanol-DMF (1:1 v/v) (30 mL). After 4 h of continued stirring under
hydrogen atmosphere (1 atm) at room temperature, the reaction mixture
was filtered, and the filtrate was evaporated under reduced pressure to
give compound 12 as a powder (0.550 g, 97%), νmax(KBr)/cm-1 3420,
1680, 1540, 1490, 1380, 1210, and 1130; 1H NMR (270 MHz, DMSO-
d6) δ 1.37 (9 H, s, t-Bu-H), 1.69-1.82 (8 H, m, â-CH2), 2.15 (8 H, m,
MHz, DMSO-d6) δ 1.97 (2 H, m, â-CH2), 2.44 (2 H, t, Jγ,â ) 8.1,
γ-CH2), 3.82 (3 H, m, NH3+CH, 3′-H and 4′-H), 4.11 (1 H, q, J2′,1′
)
J2′,3′ ) 5.2, 2′-H), 5.08 (2 H, s, PhCH2), 5.24 (1 H, d, JOH,3′ ) 4.9,
3′-OH), 5.47 (1 H, d, J ) 5.4, 2′-OH), 5.63 (1 H, d, J5,6 ) 8.3,
OH,2′
5-H), 5.75 (1 H, d, J1′,2′ ) 5.4, 1′-H), 7.36 (5 H, m, Ar-H), 7.66 (1 H,
d, J6,5 ) 8.3, 6-H), 8.10 (3 H, br, NH3+), 8.62 (1 H, t, JNH,5′ 5.4, 5′-
NH), 11.37 (1 H, s, 3-NH); MALDI-TOF HRMS (R-CHCA), m/z found
463.17 (M - CF3COO), calculated 463.183.
Boc-(isoGln(5′U))2-OBzl (8). To a solution of 6 (1.25 g, 2.65 mmol),
HOBt (0.358 g, 2.65 mmol), and BOP reagent (1.17 g, 2.65 mmol) in
DMF (50 mL) was added diisopropylethylamine (0.97 mL, 5.57 mmol).
(24) Thomas, G. J.; Kyogoku, Y. J. Am. Chem. Soc. 1967, 89, 4170.

PRNA and Control of DNA Recognition
J. Am. Chem. Soc., Vol. 122, No. 29, 2000 6909
γ-CH2), 3.85 (9 H, m, Boc-NHCH, 3′-H and 4′-H), 4.04 (4 H, m, 2′-
H), 4.20 (3 H, m, R-CH), 5.21 (4 H, br, 3′-OH), 5.37 (4 H, br, 2′-OH),
5.64 (4 H, d, J5,6 ) 7.8, 5-H), 5.73 (4 H, d, J1′,2′ ) 5.9, 1′-H), 6.87 (1
H, d, JNH,CH ) 8.3, Boc-NH), 7.65 (4 H, d, J6,5 ) 8.3, 6-H), 7.95 (4 H,
m, 5′-NH), 8.13 (3 H, m, R-NH), 11.33 (4 H, s, 3-NH); MALDI-TOF
HRMS (R-CHCA), m/z found 1557.52 (M + Na), calculated 1557.523.
TFA‚(isoGln(5′U))4-OBzl (13). 11 (0.600 g, 0.369 mmol) was
dissolved in TFA (5 mL), and the solution was kept at 0 °C for 30
min. TFA was removed under reduced pressure, and ether (100 mL)
(8 H, m, R-NH), 11.33 (8 H, s, 3-NH); MALDI-TOF HRMS
(R-CHCA), m/z found 3360.13 (M + Na), calculated 3360.134.
HCl‚(isoGln(5′U))8
-Lys-OH (17). 16 (0.300 g, 0.090 mmol) was
dissolved in TFA (10 mL), and the solution was kept at -10 °C under
nitrogen atmosphere. Thioanisole (1.84 mL, 15.7 mmol), m-cresol (0.97
mL, 9.27 mmol), and trimethylsilyl triflate (3.00 mL, 16.6 mmol) were
successively added, and the mixture was stirred for 1 h at 0 °C. Ether
(200 mL) was added, and the precipitate was filtered. The precipitate
was dissolved in 5% aqueous ammonia. After 30 min, the solvent was
removed under reduced pressure, and the residue was treated with 4
M HCl in dioxane for 30 min. The solvent was removed under reduced
pressure, and the residue was purified by gel filtration and reverse phase
HPLC to give compound 17 as a powder (0.223 g, 82%), νmax(KBr)/
was added to give compound 13 as a powder (0.593 g, 98%), νmax
-
(KBr)/cm-1 3420, 1680, 1540, 1460, 1390, 1270, and 1130; H NMR
(270 MHz, DMSO-d6) δ 1.76-1.89 (8 H, m, â-CH2), 2.13-2.25 (6 H,
m, γ-CH2), 2.34 (2 H, t, Jγ,â ) 8.3, γ-CH2), 3.83 (9 H, m, NH3+CH,
3′-H and 4′-H), 4.05-4.22 (7 H, m, R-CH and 2′-H), 5.06 (2 H, s,
PhCH2), 5.16 (3 H, m, 3′-OH), 5.24 (1 H, d, JOH,3′ ) 4.9, 3′-OH), 5.42
(3 H, m, 2′-OH), 5.48 (1 H, d, JOH,2′ ) 5.4, 2′-OH), 5.64 (4 H, m,
5-H), 5.73 (4 H, m, 1′-H), 7.35 (5 H, m, Ar-H), 7.66 (4 H, m, 6-H),
7.95 (3 H, m, 5′-NH), 8.13 (6 H, m, R-NH and NH3+), 8.61 (1 H, m,
5′-NH), 11.35 (4 H, m, 3-NH); MALDI-TOF HRMS (R-CHCA), m/z
found 1548.52 (M - CF3COO + Na), calculated 1548.526.
1
1
cm-1 3420, 1680, 1540, 1460, 1390, 1270, and 1110; H NMR (270
MHz, DMSO-d6) δ 1.35 (4 H, m, γ-CH2(Lys) and δ-CH2(Lys)), 1.70-
1.83 (16 H, m, â-CH2(Glu)), 2.14 (16 H, m, γ-CH2(Glu)), 2.98 (2 H,
m, ꢀ-CH2(Lys)), 3.84 (17 H, m, NH3+CH, 3′-H and 4′-H), 4.06 (8 H,
m, 2′-H), 4.20 (8 H, m, R-CH), 5.17 (8 H, m, 3′-OH), 5.41 (8 H, m,
2′-OH), 5.65 (8 H, d, J5,6 ) 7.8, 5-H), 5.73 (8 H, d, J1′,2′ ) 5.9, 1′-H),
7.24 (2 H, br, ꢀ-NH2), 7.65 (8 H, d, J6,5 ) 7.8, 6-H), 7.95 (8 H, m,
5′-NH), 8.14 (11 H, m, R-NH and NH3+), 8.59 (1 H, m, 5′-NH), 11.34
(8 H, m, 3-NH); MALDI-TOF HRMS (R-CHCA), m/z found 2980.06
(M - HCl), calculated 2980.055.
Boc-(isoGln(5′U))8-OBzl (14). To a solution of 12 (0.468 g, 0.305
mmol), HOBt (0.041 g, 0.305 mmol), and BOP reagent (0.135 g, 0.305
mmol) in DMF (30 mL) was added diisopropylethylamine (0.11 mL,
0.641 mmol). After 30 s of stirring at 0 °C, 13 (0.550 g, 0.335 mmol)
was added, and the mixture was stirred for 4 h at room temperature.
The solvent was removed under reduced pressure, and methanol was
added to give compound 14 as a powder (0.854 g, 92%), νmax(KBr)/
Nr-t-Butoxycarbonyl-N5-(N4-benzoyl-5′-deoxy-5′-cytidyl)-L-iso-
glutamine benzyl ester (Boc-isoGln(5′BzC)-OBzl) (18). To a solution
of N-t-butoxycarbonyl-L-glutamic acid γ-benzyl ester (2.52 g, 7.47
mmol), HOBt (1.01 g, 7.47 mmol), and BOP reagent (3.30 g, 7.47
mmol) in DMF (100 mL) was added diisopropylethylamine (1.30 mL,
7.47 mmol). After 30 s of stirring at 0 °C, 5′-amino-5′-deoxycytidine
(3)18 (2.85 g, 8.22 mmol) was added, and the mixture was stirred for
1 h at room temperature. The solvent was removed under reduced
pressure, and methanol was added to precipitate a white solid. After
filtration, the product was recrystallized from methanol to give
compound 18 (4.48 g, 90%), νmax(KBr)/cm-1 3420, 1690, 1560, 1480,
1320, 1260, and 1140; 1H NMR (270 MHz, DMSO-d6) δ 1.36 (9 H, s,
t-Bu-H), 1.77-1.90 (2 H, m, â-CH2), 2.37 (2 H, t, Jγ,â ) 7.8, γ-CH2),
3.43 (2 H, m, 5′-H), 3.83 (1 H, q, J3′,2′ ) J3′,4′ ) 5.5, 3′-H), 3.92 (1 H,
q, J4′,3′ ) J4′,5′ ) 5.5, 4′-H), 3.98 (1 H, q, JCH,NH ) JCH,â ) 4.9, Boc-
NHCH), 4.07 (1 H, q, J2′,1′ ) J2′,3′ ) 4.7, 2′-H), 5.06 (2 H, s, PhCH2),
5.17 (1 H, d, JOH,3′ ) 5.9, 3′-OH), 5.49 (1 H, d, JOH,2′ ) 5.4, 2′-OH),
5.80 (1 H, d, J1′,2′ ) 3.9, 1′-H), 6.95 (1 H, d, JNH,CH ) 7.8, Boc-NH),
7.34 (5 H, m, Ar-H), 7.40 (1 H, d, J5,6 ) 7.8, 5-H), 7.52 (2 H, t, Jm,o
) Jm,p ) 7.6, Ar-m-H), 7.63 (1 H, t, Jp,m ) 7.6, Ar-p-H), 8.01 (2 H, d,
Jo,m ) 7.3, Ar-o-H), 8.07 (1 H, t, JNH,5′ ) 5.6, 5′-NH), 8.19 (1 H, d,
J6,5 ) 7.3, 6-H); MALDI-TOF HRMS (R-CHCA), m/z found 688.26
(M + Na), calculated 688.260.
1
cm-1 3420, 1680, 1530, 1460, 1400, 1270, and 1130; H NMR (270
MHz, DMSO-d6) δ 1.37 (9 H, s, t-Bu-H), 1.71-1.82 (16 H, m, â-CH2),
2.14 (14 H, m, γ-CH2), 2.34 (2 H, t, Jγ,â ) 7.8, γ-CH2), 3.25-3.42
(16 H, m, 5′-H), 3.83 (17 H, m, Boc-NHCH, 3′-H and 4′-H), 4.05 (8
H, m, 2′-H), 4.20 (7 H, m, R-CH), 5.06 (2 H, s, PhCH2), 5.15 (8 H, d,
JOH,3′ ) 3.4, 3′-OH), 5.39 (8 H, d, JOH,2′ ) 5.4, 2′-OH), 5.64 (8 H, d,
J5,6 ) 7.8, 5-H), 5.73 (8 H, d, J1′,2′ ) 5.4, 1′-H), 6.87 (1 H, d, JNH,CH
) 7.3, Boc-NH), 7.35 (5 H, m, Ar-H), 7.64 (8 H, d, J6,5 ) 7.8, 6-H),
7.95 (8 H, m, 5′-NH), 8.13 (7 H, m, R-NH), 11.33 (8 H, s, 3-NH);
MALDI-TOF HRMS (R-CHCA), m/z found 3064.04 (M + Na),
calculated 3064.041.
Boc-(isoGln(5′U))8-OH (15). Palladium on activated carbon (10%;
∼0.1 g) was added to a solution of 14 (0.800 g, 0.263 mmol) in
methanol-DMF (1:3 v/v) (50 mL). After 6 h of continued stirring under
hydrogen atmosphere (1 atm) at room temperature, the reaction mixture
was filtered, and the filtrate was evaporated under reduced pressure to
give compound 15 as a powder (0.745 g, 96%), νmax(KBr)/cm-1 3410,
1680, 1540, 1470, 1390, 1260, and 1110; 1H NMR (270 MHz, DMSO-
d6) δ 1.37 (9 H, s, t-Bu-H), 1.70-1.82 (16 H, m, â-CH2), 2.15 (16 H,
m, γ-CH2), 3.83 (17 H, m, Boc-NHCH, 3′-H and 4′-H), 4.05 (8 H, m,
2′-H), 4.19 (7 H, m, R-CH), 5.18 (8 H, m, 3′-OH), 5.29 (8 H, m, 2′-
OH), 5.65 (8 H, d, J5,6 ) 7.8, 5-H), 5.73 (8 H, d, J1′,2′ ) 5.9, 1′-H),
6.88 (1 H, d, JNH,CH ) 8.8, Boc-NH), 7.65 (8 H, d, J6,5 ) 8.3, 6-H),
7.95 (8 H, m, 5′-NH), 8.14 (8 H, m, R-NH), 11.33 (8 H, s, 3-NH);
MALDI-TOF HRMS (R-CHCA), m/z found 2973.99 (M + Na),
calculated 2973.994.
N5-(N4-benzoyl-5′-deoxy-5′-cytidyl)-L-isoglutamine benzyl ester
trifluoroacetic acid salt (TFA‚isoGln(5′BzC)-OBzl) (19). 18 (4.20
g, 6.31 mmol) was dissolved in TFA (50 mL), and the solution was
kept at 0 °C for 30 min. TFA was removed under reduced pressure,
and ether (200 mL) was added to give compound 19 as a powder (4.20
g, 98%), νmax(KBr)/cm-1 3400, 1680, 1560, 1480, 1320, 1250, 1200,
and 1140; 1H NMR (270 MHz, DMSO-d6) δ 1.79-1.92 (2 H, m,
â-CH2), 2.42 (2 H, t, Jγ,â ) 7.9, γ-CH2), 3.82 (3 H, m, NH3+CH, 3′-H
and 4′-H), 4.10 (1 H, q, J2′,1′ ) J2′,3′ ) 4.9, 2′-H), 5.07 (2 H, s, PhCH2),
5.20 (1 H, d, JOH,3′ ) 5.6, 3′-OH), 5.52 (1 H, d, JOH,2′ ) 5.4, 2′-OH),
Boc-(isoGln(5′U))8-Lys(ClZ)-OBzl (16). To a solution of 15 (0.700
g, 0.237 mmol), HOBt (0.032 g, 0.237 mmol), and BOP reagent (0.105
g, 0.237 mmol) in DMF (30 mL) was added diisopropylethylamine
(0.04 mL, 0.237 mmol). After 30 s of stirring at 0 °C, Nꢀ-2-
chlorobenzyloxycarbonyl-O-benzyl-L-lysine (0.106 g, 0.261 mmol) was
added, and the mixture was stirred for 4 h at room temperature. The
solvent was removed under reduced pressure, and methanol was added
to give compound 16 as a powder (0.712 g, 90%), νmax(KBr)/cm-1
5.78 (1 H, d, J1′,2′ 4.2, 1′-H), 7.33 (5 H, m, Ar-H), 7.41 (1 H, d, J5,6
7.7, 5-H), 7.51 (2 H, t, Jm,o ) Jm,p ) 7.6, Ar-m-H), 7.62 (1 H, t, Jp,m
)
)
)
7.6, Ar-p-H), 8.03 (2 H, d, Jo,m ) 7.4, Ar-o-H), 8.05 (1 H, t, JNH,5′
+
5.8, 5′-NH), 8.20 (4 H, m, NH3 and 6-H), 11.45 (1 H, s, 4-NH);
MALDI-TOF HRMS (R-CHCA), m/z found 566.22 (M - CF3COO),
calculated 566.225.
1
3330, 1680, 1540, 1470, 1390, 1270, and 1130; H NMR (270 MHz,
DMSO-d6) δ 1.36 (13 H, m, t-Bu-H, γ-CH2(Lys) and δ-CH2(Lys)),
1.57-1.84 (18 H, m, â-CH2(Glu) and â-CH2(Lys)), 2.14 (16 H, m,
â-CH2(Glu)), 2.98 (2 H, m, ꢀ-CH2(Lys)), 3.83 (17 H, m, Boc-NHCH,
3′-H and 4′-H), 4.05 (8 H, m, 2′-H), 4.20 (8 H, m, R-CH), 5.07 (2 H,
s, PhCH2), 5.08 (2 H, s, Cl-PhCH2), 5.16 (8 H, d, JOH,3′ ) 3.9, 3′-
OH), 5.39 (8 H, d, JOH,2′ ) 5.4, 2′-OH), 5.65 (8 H, d, J5,6 ) 7.8, 5-H),
5.73 (8 H, d, J1′,2′ ) 5.9, 1′-H), 6.87 (1 H, d, JNH,CH ) 7.3, Boc-NH),
6.99 (1 H, m, ꢀ-NH(Lys)), 7.35 (5 H, m, Ar-H(Bzl)), 7.45 (4 H, m,
Ar-H(Cl-Z)), 7.65 (8 H, d, J6,5 ) 7.8, 6-H), 7.96 (8 H, m, 5′-NH), 8.14
Boc-isoGln(5′U)-isoGln(5′BzC)-OBzl (20). To a solution of 9 (1.25
g, 2.65 mmol), HOBt (0.358 g, 2.65 mmol), and BOP reagent (1.17 g,
2.65 mmol) in DMF (50 mL) was added diisopropylethylamine (0.97
mL, 5.57 mmol). After 30 s of stirring at 0 °C, 19 (1.98 g, 2.92 mmol)
was added, and the mixture was stirred for 2 h at room temperature.
The solvent was removed under reduced pressure, and methanol was
added to give compound 20 as a powder (2.65 g, 98%), νmax(KBr)/
1
cm-1 3420, 1650, 1560, 1490, 1390, 1260, 1170, and 1080; H NMR
(270 MHz, DMSO-d6) δ 1.35 (9 H, s, t-Bu-H), 1.78-1.92 (4 H, m,

6910 J. Am. Chem. Soc., Vol. 122, No. 29, 2000
Wada et al.
â-CH2), 2.16 (2 H, m, γ-CH2), 2.36 (2 H, t, Jγ,â ) 7.8, γ-CH2), 3.42 (4
H, m, 5′-H(Urd) and 5′-H(Cyd)), 3.83 (3 H, m, 3′-H(Urd), 3′-H(Cyd)
and 4′-H(Urd)), 3.91 (2 H, m, Boc-NHCH and 4′-H(Cyd)), 4.03 (1 H,
q, J2′,1′ ) J2′,3′ ) 5.4, 2′-H(Urd)), 4.09 (1 H, q, J2′,1′ ) J2′,3′ ) 4.7,
2′-H(Cyd)), 4.29 (1 H, q, JCH,NH ) JCH,â ) 7.3, R-CH), 5.05 (2 H, s,
PhCH2), 5.17 (2 H, m, 3′-OH(Urd), and 3′-OH(Cyd)), 5.39 (1 H, d,
JOH,2′ ) 5.9, 2′-OH(Urd)), 5.51 (1 H, d, JOH,2′ ) 5.4, 2′-OH(Cyd)),
5.64 (1 H, d, J5,6 ) 7.8, 5-H(Urd)), 5.73 (1 H, d, J1′,2′ ) 5.9, 1′-H(Urd)),
5.78 (1 H, d, J1′,2′ ) 3.9, 1′-H(Cyd)), 6.87 (1 H, d, JNH,CH ) 8.3, Boc-
(2 H, br, ꢀ-NH2(Lys)), 7.32 (4 H, m, 5-H(Cyd)), 7.64 (4 H, m,
6-H(Urd)), 7.94 (4 H, m, 5′-NH(Urd)), 8.03 (11 H, m, R-NH and NH3+),
8.22 (8 H, m, 6-H(Cyd) and 5′-NH(Cyd)), 11.33 (8 H, m, 3-NH(Urd)
and 4-NH(Cyd)); MALDI-TOF HRMS (R-CHCA), m/z found 2976.12
(M - HCl), calculated 2976.119.
Acknowledgment. This paper is dedicated to the memory
of Dr. Narutoshi Minamimoto, who has made a great deal of
contribution to this work and unexpectedly deceased of cancer
on November 14, 1999. We are grateful to the generous support
of this work by a Grant-in-Aid for Scientific Research from
the Ministry of Education, Science, Sports, and Culture, and
also by a grant from the Asahi Glass Foundation. We would
also like to thank Dr. Simon R. L. Everitt of the Inoue
Photochirogenesis Project and Dr. George Lem of Columbia
University for assistance in the preparation of this manuscript.
NH), 7.34 (6 H, m, Ar-H and 5-H(Cyd)), 7.52 (2 H, t, Jm,o ) Jm,p
)
7.8, Ar-m-H), 7.63 (2 H, m, Ar-p-H and 6-H(Urd)), 7.95 (1 H, t, JNH,5′
) 5.9, 5′-NH(Urd)), 8.02 (3 H, m, Ar-o-H and R-NH), 8.20 (2 H, m,
6-H(Cyd) and 5′-NH(Cyd)), 11.31 (2 H, m, 3-NH(Urd) and 3-NH-
(Cyd)); MALDI-TOF HRMS (R-CHCA), m/z found 1042.38 (M + Na),
calculated 1042.377.
HCl‚(isoGln(5′U)-isoGln(5′C))4-Lys-OH (21). This compound was
synthesized in a similar manner as the synthesis of 17 using hetero
dimer 20 instead of homo dimer 8. Each deprotection and coupling
process was done with >95% and 80-95% yields, respectively; νmax
-
Supporting Information Available: Schemes of the syn-
thetic routes and the experimental procedures for 2 and 3; Tables
of the [θ]max values for 1, 2, 3, and 4 in phosphate and borate
buffers, the results of NOE experiments with 2 and 3 in
phosphate and borate buffers, and the [θ]max values for 2, 6,
17, and 21 in phosphate and borate buffers (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
(KBr)/cm-1 3400, 1660, 1550, 1480, 1400, 1230, 1150, and 1050; H
NMR (270 MHz, DMSO-d6) δ 1.34 (4 H, m, γ-CH2(Lys) and δ-CH2-
(Lys)), 1.76-1.90 (16 H, m, â-CH2(Glu)), 2.14 (16 H, m, γ-CH2(Glu)),
2.99 (2 H, m, ꢀ-CH2(Lys)), 3.43 (16 H, m, 5′-H(Urd), and 5′-H(Cyd)),
3.84 (17 H, m, NH3+CH, 3′-H(Urd), 3′-H(Cyd), 4′-H(Urd) and
4′-H(Cyd)), 4.04 (4 H, m, 2′-H(Urd)), 4.10 (4 H, m, 2′-H(Cyd)), 4.27
(8 H, m, R-CH), 5.17 (8 H, m, 3′-OH(Urd) and 3′-OH(Cyd)), 5.40 (4
H, m, 2′-OH(Urd)), 5.52 (4 H, m, 2′-OH(Cyd)), 5.64 (4 H, m,
5-H(Urd)), 5.72 (4 H, m, 1′-H(Urd)), 5.79 (4 H, m, 1′-H(Cyd)), 7.23
1
JA9935456