Boronic acid-based autoligation of nucleic acids: Influence of the nature of the 3′-end ribonucleotidic strand

Article abstract of DOI:10.1007/s00706-012-0881-7

The development of synthetic systems displaying dynamic and adaptive characteristics is a formidable challenge with wide applications from biotechnology to therapeutics. Recently, we described a dynamic and programmable nucleic acid-based system relying on the formation of reversible boronate internucleosidic linkages. The DNA- or RNA-templated system comprises a 5′-ended boronic acid probe connecting a 3′-ended ribonucleosidic oligonucleotide partner. To explore the dominant factors that control the reversible linkage, we synthesized a series of 3′-end modified ribonucleotidic strands. Evidence suggests that geometric and steric factors are key features for controlling the equilibria.

Full text of DOI:10.1007/s00706-012-0881-7

Monatsh Chem (2013) 144:495–500
DOI 10.1007/s00706-012-0881-7
ORIGINAL PAPER
Boronic acid-based autoligation of nucleic acids: influence
0
of the nature of the 3 -end ribonucleotidic strand
Renaud Barbeyron
Jean-Jacques Vasseur
•
Jesper Wengel
Michael Smietana
•
•
Received: 28 September 2012 / Accepted: 13 November 2012 / Published online: 14 December 2012
Ó Springer-Verlag Wien 2012
Abstract The development of synthetic systems displaying
dynamic and adaptive characteristics is a formidable
challenge with wide applications from biotechnology to
therapeutics. Recently, we described a dynamic and pro-
grammable nucleic acid-based system relying on the
formation of reversible boronate internucleosidic linkages.
widely accepted that these processes are the outcome of
simple chemical systems that evolved into functional self-
organized architectures [1, 2]. One of the many issues
scientists face concerning the structure of contemporary
nucleic acids is the selection of phosphodiester internu-
cleosidic linkages [3]. Indeed, non-enzymatic synthesis of
oligonucleotides from simple building blocks requires
phosphate activation, which often leads to mixtures of
compounds [4–7]. To overcome these problems, non-
enzymatic DNA- and RNA-templated ligation systems
displaying high ligation fidelity have been developed.
Relying on the hybridization of two half-length oligomers
along a template, many ligation chemistries have been used
to generate natural or modified backbones [8–18]. While
remarkable, these systems lack the adaptive and corrective
parameters necessary for the genetic polymers to evolve. In
turn, it has been suggested that these parameters could be
achieved using reversible linkages able to respond to
external stimuli [19–24]. In this context, we recently
developed a dynamic DNA- and RNA-templated ligation
system based on the reversible formation of cyclic five-
membered boronate internucleosidic linkages. The ligation
that does not require any added reagent occurs through the
reaction of two half-stranded probes, one carrying a boro-
0
The DNA- or RNA-templated system comprises a 5 -ended
0
boronic acid probe connecting a 3 -ended ribonucleosidic
oligonucleotide partner. To explore the dominant factors
that control the reversible linkage, we synthesized a series
0
of 3 -end modified ribonucleotidic strands. Evidence sug-
gests that geometric and steric factors are key features for
controlling the equilibria.
Keywords Boron ꢀ Complexation ꢀ DNA ꢀ
Dynamic processes ꢀ Template synthesis
Introduction
The transfer of genetic information undoubtedly relies on
selection, amplification, and correction processes. It is now
0
nic acid at its 5 -end, while its partner would feature a
0
natural 3 -end ribonucleotide (Fig. 1). We also demon-
strated that the system displayed adaptative behavior
responding to external stimuli such as temperature, pH,
diol concentration, and cyanide ions [25–28]. In order to
explore the factors that influence the sensitivity and sta-
bility of the boronate internucleosidic linkage, we surveyed
R. Barbeyron ꢀ J.-J. Vasseur ꢀ M. Smietana (&)
Institut des Biomol e´ cules Max Mousseron, UMR 5247
CNRS-Universit e´ Montpellier 1, Universit e´ Montpellier, 2,
place Eug e` ne Bataillon, 34095 Montpellier, France
e-mail: msmietana@um2.fr
0
its formation by varying the nature of the 3 -end ribonu-
J. Wengel
cleotidic strand. We report here our investigation aimed at
delineating the optimum structural features that drive the
dynamic system.
Nucleic Acid Center, Department of Physics,
Chemistry and Pharmacy, University of Southern Denmark,
Campusvej 55, 5230 Odense M, Denmark
123

4
96
R. Barbeyron et al.
HO
B
However, many issues could affect the boronic acid/diol
complexation. For the diol moiety, it has been demon-
strated that low pKA values, small O–C–C–O dihedral
angles, steric hindrance, and restricted rotations around the
C–C bond of the diol are key factors for optimal binding
O
OH
HO
O
HO
O
O
C
G
T
A
[
29, 32, 39].
Associated with structural restrictions imposed by the
double helix, it is crucial to understand the intrinsic pref-
erence that governs the boronic acid-based DNA-templated
autoligation process.
T (°
T
C)
T (°C)
T
0
The series of 3 -end sequences investigated is shown in
Fig. 2. Along with natural uridine (1), we probed confor-
mational, electronic, chemical, and optimal spacing
features of the hybridizing probes. Derivatives 2 and 3
were selected to evaluate two contrasted flexible and rigid
modifications. Aminonucleosides 4 and 5 were chosen for
their ability to trigger the formation of oxazaborolidines
along the template. Finally, to investigate the effects of
relative positions of hybridization, we synthesized
sequences in which bases were deleted (compound 6 and
O
O
O
B
O
O
O
C
G
T
A
7a) or had an added cis-diol motif (compound 7b).
Our experimental design involves a 14mer DNA- or
pH
pH
0
RNA-template (ODND and ORNR respectively), a 5 -
boronic acid heptathymidinylate sequence (ODNB), and
HO
O
0
O
the 3 -end modified sequences ODNX. The two half-
O
B
stranded probes exhibit purposefully major differences in
affinity for the complementary strand (Table 1, entries 1
and 2; T = 14.8 °C for ODND/ODNB and T = 38.7 °C
O
O
O
C
G
m
m
T
for ODND/ODN1). Control experiments carried out with
A
(dT) (ODNC), an analog of ODNB without the boronic
7
acid modification, are given for comparison. All curves
featured a double sigmoidal transition corresponding to the
transition of the half-duplexes with the template. Our goal
was to explore whether the templated ligation between
ODNB and ODNX could occur and to what extent. Since
the boronic acid functionality is carried by the less stable
half-duplex, the formation of a novel boronate-linked full
duplex mainly influences the lower transition. Considering
that the boronate/diol complex could be stabilized at high
pH, we first probed the pH effect on the different systems.
Results are presented in Table 1. As expected, ODN1
triggered the autoligation at pH 7.5, leading to a boronate-
ligated duplex having a melting temperature 6.1 °C higher
DNA or RNA template
Fig. 1 Mechanism of the boronic acid-based autoligation process
resulting in boronate ester
Results and Discussion
It is well known that boronic acids react reversibly with
1
,2- and 1,3-substituted Lewis base donors and particularly
cis-diol groups to form boronic esters [29, 30]. The overall
process is an equilibrium that can be shifted by azeotropic
distillation of the water or with the use of a dehydrating
agent. In water, the formation of a diol boronate complex
can be favored at high pH as a consequence of a rehybri-
dation of the boronic ester (neutral trigonal boron, sp2
hybridized) to a hydroxy boronate ester (anionic tetrahedral
boron, sp3 hybridized) [31, 32]. At neutral pH simple
Lewis bases such as fluoride [33–35] and cyanide ions are
also known to trigger the formation of sp3 species [36].
These properties have notably been used for the develop-
ment of boronic acid-based sensors [29, 34, 37, 38].
than that of the nicked dsDNA (T = 20.9 vs. 14.8 °C,
m
Table 1, entry 3). This stabilization goes up to 16.0 °C at
pH = 9.5 (T = 30.8 vs. 14.8 °C, Table 1, entry 3) as a
m
consequence of the rehybridation of boron from sp2 to sp3.
In contrast, ODN2, which features an acyclic RNA mimic
0
0
lacking a bond between the C2 and C3 atoms of the ribose
ring, was unable to induce the autoligation irrespective of
the pH of the experiments (Table 1, entry 4). Despite the
destabilization of duplexes generally induced by acyclic
nucleosides, it has been demonstrated that Watson-Crick
1
23

Boronic acid-based autoligation of nucleic acids
497
Fig. 2 Structures of sequences
considered
O
O
O
U
T
U
O
O
O
HO
HO
HO
HO
= U
HO
HO
1
2 = UNA-T
3 = 2'-MeU
O
O
O
T
O
U
O
O
O
OH
HO
HO
HO
HO
N
H N
2
HO
H
7
7
a =Glyceryl 6mer
b =Glyceryl 7mer
4
= 2'-NH U 5 = 2-amino-LNA-T 6 = Abasic
2
Table 1 UV thermal denaturation data of DNA-templated autoligation experiments
a
b
c
Entry
Sequences
Control
T
m
/°C
pH
7
pH
8.5
pH
9.5
.5
0
1
2
ODNC
3 -TTTTTTT
14.8
38.7
14.8
16.8
12.1
13.5
14.7
16.0
–
–
–
–
0
0
0
0
0
0
0
ODN1
1GCTGCC-5
1GCTGCC-5
2GCTGCC-5
3GCTGCC-5
4GCTGCC-5
5GCTGCC-5
7GCTGCC-5
–
–
d
0
bn
bn
bn
bn
bn
bn
3
ODNB/ODN1
ODNB/ODN2
ODNB/ODN3
ODNB/ODN4
ODNB/ODN5
ODNB/ODN7b
0
3 -TTTTTTT
0
20.9
13.7
11.0
12.1
13.4
13.0
27.9
13.6
12.1
12.4
13.9
12.9
30.8
13.4
16.1
12.3
13.7
10.9
d
4
3 -TTTTTTT
0
d
5
3 -TTTTTTT
0
d
6
3 -TTTTTTT
0
d
7
3 -TTTTTTT
0
d
8
3 -TTTTTTT
0
Template ODND is 5 -d(AAAAAAAACGACGG)-3
a
b
c
bn
T
refers to boronothymidine
values indicated refer to the control ODNC/ODNX at pH 7.5. No major variation could be observed at higher pHs
T
m
Melting temperatures refer to the melting of the corresponding sequence(s) with ODND and were obtained from the maxima of the first
derivative of the melting curve (A260 vs. temperature) recorded in a buffer containing 0.5 M NaCl and 10 mM of sodium cacodylate, DNA
concentration 3 lM of each strand. Curve fit data were averaged from fits of three denaturation curves
d
m
T values indicated refer only to the lowest temperature-dependent transition
base-pairing rules are still obeyed with UNA monomers
0
the optimal structural requirements. As we previously
noted, an RNA-like conformation on the ribonucleoside
partner is preferred to promote an efficient autoligation
[
40]. Thus, although UNA contains a 2 -hydroxyl group, it
appears that its highly flexible nature prevents the forma-
tion of a stable boronate linkage. This observation could
also be explained in terms of the higher energy required for
the system to generate an eight-membered ring. ODN3,
0
process [26]. The estimated percentage of the north (C3
0
endo) conformation in 2 -C-methyl uridine is 90 % (mea-
1
sured by H NMR) [41]. Thus, despite the preorganization
0
which carries an added methyl group at the 2 position,
of the cis-diol motif in ODN3, neither the trigonal nor the
tetrahedral complex is particularly stable. Molecular
dynamics simulations have shown that tetrahedral inter-
nucleosidic linkages almost exclusively prefer a single
conformer (structure A, Fig. 3) [27]. It is thus possible that
induced a slight destabilization at pH = 7.5 (T = 11.0 vs.
m
1
2.1 °C, Table 1, entry 5) and proved inefficient for liga-
tion of the probes at pH = 8.5. However, the stabilization
observed at pH = 9.5, although modest, demonstrates that
the ligation could proceed and provide information about
0
the addition of a methyl group at the 2 -position creates an
123

4
98
R. Barbeyron et al.
unavoidable axial interaction that results in significant
complex destabilization with the adjacent ribose as repre-
sented in structure B (Fig. 3), thus suggesting that
geometric and steric factors appear to exert significant
control of the complex stability.
T = 32.9 vs. 21.5 °C, DT = 9.9 °C for ORNR).
m
m
Although less marked, an identical trend was observed with
ODN3. Its RNA-like structure allows a slight stabilization
of the resulting duplex at pH = 7.5 and a minor increased
pH dependency (T = 16.1 vs. 11.0 °C, DT = 5.1 °C for
m
m
We next shifted our attention to examine the potential
ability of the templated system to generate oxazaboroli-
ODND and T = 17.3 vs. 14.0 °C, DT = 4.0 °C for
m
m
ORNR), but inevitable steric congestion does not permit a
higher degree of complex stability. As was the case in the
DNA-templated experiments, all other modifications were
unable to promote ligated duplexes (Table 2).
dine linkages with ODN4 and ODN5, which exist in 25 and
0
1
00 % C3 -endo conformations, respectively [42, 43].
While it is true that most oxazaborolidines are known to be
sensitive to water, studies describing the formation of sp3
hybridized boroxazolidines in the presence of water have
been reported [44, 45]. Considering that upon hybridization
the template significantly increases the effective molarity
of the reactants, we anticipated the possibility of generating
stable oxazaborolidine linkages. Unfortunately, neither
ODN4 nor ODN5 could mediate such linkages (Table 1,
entries 6 and 7).
Finally, ODN1 and ODN3 were evaluated in the pres-
ence of cyanide ions as an external stimulus to uphold
stable cyanoboronate junctions at pH = 7.5. Experiments
performed in the presence of NaCN (3 mM) at pH = 7.5
on ODN1 and ODN3 confirmed that a high level of
stabilization could be achieved at neutral pH both in
DNA- and RNA-templated systems (Table 3).
In summary, the data presented in this article reveal that
geometric and steric structural features are the prevailing
factors that control the stability of boronate internucleosi-
dic linkages. If more stable complexes inside a double
helix are to be created, these factors will require close
attention, but at this stage natural ribonucleotides appear to
be the most effective partners. A second feature is the
destabilizing role of added diol linkers, emphasizing the
importance of adjusted base pairs. These findings com-
plement our earlier studies devoted to the assembly of
stimuli-responsive architectures and highlight the selective
association of natural ribonucleotidic partners over modi-
fied ones.
In the case of ODN6 and ODN7a, no transition could be
observed whatever the pH because of the destabilization
induced by the removal of one base pair (data not shown).
The proper number of bases if necessary is nevertheless not
sufficient to induce a boronate linkage with the glyceryl
partner, as can be seen Table 1, entry 8. These results are
very useful for evaluating the reaction scope of the DNA-
templated autoligation system and highlight the importance
of structural mimicry of the DNA backbone.
RNA-templated autoligation experiments performed in
the presence ORNR confirmed the results obtained with its
DNA counterpart. Indeed, ODN1 and ODN3 were the only
sequences able to induce a boronate junction. Compared to
DNA, the RNA template displayed higher stabilizations
against the nicked system at pH = 9.5 with ODN1 (T =
m
Experimental
3
0.8 vs. 14.8 °C, DT = 17.9 °C for ODND and
m
T = 32.9 vs. 15.0 °C, DT = 16.0 °C for ORNR).
Reactions were conducted in oven-dried glassware. All
reagents were purchased from Aldrich or local suppliers
and used without purification. All reactions were monitored
by TLC, and visualization was effected with UV and/or by
developing in vanillin. Chromatography refers to open
column chromatography on silica gel (Merck, 40–63 lm).
NMR spectra were recorded on a Bruker DRX-300 MHz
m
m
However, the pH dependency decreased slightly (T =
3
m
0.8 vs. 20.9 °C, DT = 11.4 °C for ODND and
m
O
O
1
13
spectrometer at 300 MHz ( H) and 75 MHz ( C) at
O
H
O
H
298 K. Chemical shifts are reported in ppm relative to the
1
solvent residual peak CDCl as internal standard ( H and
OH
OH
T
T
B
B
3
O
O
1
3
1
C). Data for H are reported as follows: chemical
O
H
O
shift, multiplicity (s = singlet, d = doublet, t = triplet,
m = multiplet), coupling constants (J) in Hz, and inte-
gration. Oligonucleotides were synthesized on a 1-lmol
scale using standard phosphoramidite coupling chemistry
and analyzed by RP-HPLC (Dionex Ultimate 3000) with a
Nucleodur 100-3 C18 column (75 9 4.6 mm; Macherey–
Nagel) and by MALDI-TOF MS (Voyager Perseptive
Biosystems) using trihydroxyacetophenone (THAP) or
H
H
O
H
O
U
U
O
O
a
b
Fig. 3 Conformational representations of the sp3 linkages with
ODN1 (a) and ODN3 (b)
1
23

Boronic acid-based autoligation of nucleic acids
499
Table 2 UV thermal denaturation data of RNA-templated autoligation experiments
a
b
c
Entry
Sequences
Control
T
m
/°C
pH
7
pH
8.5
pH
9.5
.5
0
1
2
ODNC
3 -TTTTTTT
–
–
–
–
0
0
0
0
0
0
0
ODN1
1GCTGCC-5
1GCTGCC-5
2GCTGCC-5
3GCTGCC-5
4GCTGCC-5
5GCTGCC-5
7GCTGCC-5
–
–
d
0
bn
bn
bn
bn
bn
bn
3
ODNB/ODN1
ODNB/ODN2
ODNB/ODN3
ODNB/ODN4
ODNB/ODN5
ODNB/ODN7b
0
3 -TTTTTTT
0
15.0
15.0
12.3
13.0
14.9
17.0
21.5
15.0
14.0
12.0
14.9
12.1
31.9
15.0
15.1
11.9
10.9
12.1
32.9
14.0
17.3
7.7
d
4
3 -TTTTTTT
0
d
5
3 -TTTTTTT
0
d
6
3 -TTTTTTT
0
d
7
3 -TTTTTTT
0
10.9
11.1
d
8
3 -TTTTTTT
0
Template ORNR is 5 -r(AAAAAAAACGACGG)-3
a
b
c
bn
T
refers to boronothymidine
values indicated refer to the control ODNC/ODNX at pH 7.5. No major variation could be observed at higher pHs
T
m
Melting temperatures refer to the melting of the corresponding sequence(s) with ORNR and are obtained from the maxima of the first
derivative of the melting curve (A260 vs. temperature) recorded in a buffer containing 0.5 M NaCl and 10 mM of sodium cacodylate, DNA
concentration 3 lM of each strand. Curve fit data were averaged from fits of three denaturation curves
d
m
T values indicated refer only to the lowest temperature-dependent transition
a
b
Table 3 Effect of cyanide ions on the UV thermal denaturation data of DNA-template and RNA-template autoligation experiments
c
d
Entry
Sequences
m
T /°C
a,e
0
bn
bn
bn
bn
0
0
0
0
1
ODNB/ODN1
ODNB/ODN1
ODNB/ODN3
ODNB/ODN3
3 -TTTTTTT
0
1GCTGCC-5
1GCTGCC-5
3GCTGCC-5
3GCTGCC-5
24.9
27.9
18.9
19.7
b,e
2
3 -TTTTTTT
0
a,e
3
3 -TTTTTTT
0
b,e
4
3 -TTTTTTT
a
0
0
Template ODND is 5 -d(AAAAAAAACGACGG)-3
0
b
c
d
0
Template ORNR is 5 -r(AAAAAAAACGACGG)-3
bn
T
refers to boronothymidine
Melting temperatures refer to the melting of the corresponding sequence(s) with ODND or ODNR and are obtained from the maxima of the
first derivative of the melting curve (A260 vs. temperature) recorded in a buffer containing 3 mM NaCN, 0.5 M NaCl, 10 mM of sodium
cacodylate, and DNA/RNA concentration 3 lM of each strand. Curve fit data were averaged from fits of three denaturation curves
e
m
T values indicated refer only to the lowest temperature-dependent transition
3
-hydroxypiconilic acid (HPA) as matrix and ammonium
with 50–80 % ethyl acetate in cyclohexane to yield
1
a yellow foam (206 mg, 90 %). H NMR (400 MHz,
citrate as co-matrix. The boronothymidine phosphorami-
dite was synthesized as previously described [27].
CDCl
.42–7.25 (m, 10H), 6.87–6.85 (m, 4H), 6.07 (s, 1H), 5.25
d, J = 8.3 Hz, 1H), 4.02–4.15 (m, 2H), 3.79 (s, 6H), 3.58
): d = 10.58 (s, 1H), 8.27 (d, J = 8.1 Hz, 1H),
3
7
0
0
0
2
-Methyl-5 -O-(4,4 -dimethoxytrityl)uridine
C H N O )
(
(
(
31
32
2
8
13
s, 2H), 3.15 (d, J = 9.4 Hz, 1H), 1.34 (s, 3H) ppm;
C
After 4 coevaporations with anhydrous pyridine, 105 mg
3
2
NMR (100 MHz, CDCl ): d = 163.8, 158.7, 151.2, 144.5,
0
3
-methyluridine (0.41 mmol) was dissolved in 2.5 cm
140.7, 135.3, 135.1, 130.3, 130.1, 129.2, 128.2, 128.0,
anhydrous pyridine and mixed with 167 mg dimethoxytri-
127.2, 113.3, 102.6, 92.0, 87.1, 82.2, 78.7, 73.2, 55.2 ppm.
tyl chloride (0.49 mmol) under argon atmosphere for
3
4
0
5 min. The reaction was quenched by addition of 3 cm
CPG-succinyl-2 -C-methyluridine
Succinyl-CPG resin (250 mg), 28 mg 2 Me rU (0.05 mmol),
0
water, and the mixture was extracted with ethyl acetate
3
2 9 5 cm ). The organic layer was washed with a
3
(
3 mg 4-DMAP (0.025 mmol), 20 mm Et N, and 96 mg
3
3
saturated solution of sodium hydrogen carbonate (3 9 5
3
cm ), dried over Na SO , and concentrated. The crude
EDCꢀHCl (0.5 mmol) were mixed in 3 cm of anhydrous
pyridine in a sealed tube and shaken overnight. The
reaction mixture was filtrated, and the resin was washed with
2
4
product was purified by column chromatography on silica
123

5
00
R. Barbeyron et al.
dichloromethaneanddried underreduced pressure. Inasealed
3
tube, the resin was shaken for 1 h with 1 cm acetic anhydride
3
and 1 cm methylimidazole. After filtration, the resin was
21. Li ZY, Zhang ZYJ, Knipe R, Lynn DG (2002) J Am Chem Soc
124:746
2
2. Engelhart AE, Cafferty BJ, Okafor CD, Chen MC, Williams LD,
Lynn DG, Hud NV (2012) Chem Bio Chem 13:1121
washed with dichloromethane and dried under reduced
23. Bean HD, Anet FAL, Gould IR, Hud NV (2006) Origins Life
Evol B 36:39
-1
pressure. The obtained loading was 39.7 lmol g
.
2
2
2
2
2
4. Ura Y, Beierle JM, Leman LJ, Orgel LE, Ghadiri MR (2009)
Science 325:73
5. Luvino D, Baraguey C, Smietana M, Vasseur JJ (2008) Chem
Commun 44:2352
6. Martin AR, Mohanan K, Luvino D, Floquet N, Baraguey C,
Smietana M, Vasseur JJ (2009) Org Biomol Chem 7:4369
7. Martin AR, Barvik I, Luvino D, Smietana M, Vasseur JJ (2011)
Angew Chem Int Ed 50:4193
8. Martin AR, Vasseur JJ, Smietana M (2012) Pure Appl Chem
8
Acknowledgments We thank the Agence Nationale de la Recher-
che (ANR PRODYGY-11-JS07-005-01) and the R e´ gion Languedoc-
Roussillon (Programme Chercheur d’Avenir) for providing financial
support for our research.
References
4:1659
29. Jin S, Cheng Y, Reid S, Li M, Wang B (2010) Med Res Rev
30:171
1
2
3
4
5
6
7
. Eigen M (1971) Naturwissenschaften 58:465
. Eigen M, Schuster P (1977) Naturwissenschaften 64:541
. Westheimer FH (1987) Science 235:1173
. Orgel LE (1995) Acc Chem Res 28:109
30. Nishiyabu R, Kubo Y, James TD, Fossey JS (2011) Chem
Commun 47:1106
. Orgel LE (1998) Trends Biochem Sci 23:491
31. Hall DG (2005) Boronic acids. Preparation and applications in
organic synthesis and medicine, Wiley-VCH, Weinheim
32. Springsteen G, Wang BH (2002) Tetrahedron 58:5291
33. DiCesare N, Lakowicz JR (2002) Anal Biochem 301:111
34. Galbraith E, James TD (2010) Chem Soc Rev 39:3831
35. Xu Z, Kim SK, Han SJ, Lee C, Kociok-Kohn G, James TD, Yoon
J (2009) Eur J Org Chem, p 3058
36. Badugu R, Lakowicz JR, Geddes CD (2005) Dyes Pigm 64:49
37. Luvino D, Smietana M, Vasseur JJ (2006) Tetrahedron Lett
47:9253
. Orgel LE (2000) Proc Natl Acad Sci USA 97:12503
. Miyakawa S, Joshi PC, Gaffey MJ, Gonzalez-Toril E, Hyland C,
Ross T, Rybij K, Ferris JP (2006) Origins Life Evol B 36:343
. Dolinnaya NG, Sokolova NI, Ashirbekova DT, Shabarova ZA
8
9
(
1991) Nucleic Acids Res 19:3067
. Herrlein MK, Nelson JS, Letsinger RL (1995) J Am Chem Soc
17:10151
0. Rohatgi R, Bartel DP, Szostak JW (1996) J Am Chem Soc
18:3332
1
1
1
1
1
1
1
1. Xu Y, Kool ET (2000) J Am Chem Soc 122:9040
2. Xu YZ, Kool ET (1999) Nucleic Acids Res 27:875
3. Vonkiedrowski G (1986) Angew Chem Int Ed 25:932
4. Kleiner RE, Brudno Y, Birnbaum ME, Liu DR (2008) J Am
Chem Soc 130:4646
5. Mattes A, Seitz O (2001) Chem Commun 37:2050
6. Summerer D, Marx A (2002) Angew Chem Int Ed 41:89
7. Peng XH, Li H, Seidman M (2010) Eur J Org Chem, p 4194
8. El-Sagheer AH, Brown T (2011) Chem Commun 47:12057
9. Lehn JM (2007) Chem Soc Rev 36:151
38. Luvino D, Gasparutto D, Reynaud S, Smietana M, Vasseur JJ
(2008) Tetrahedron Lett 49:6075
39. Yan J, Springsteen G, Deeter S, Wang B (2004) Tetrahedron
60:11205
40. Pasternak A, Wengel J (2011) Org Biomol Chem 9:3591
41. Beigelman LN, Ermolinsky BS, Gurskaya GV, Tsapkina EN,
Karpeisky MY, Mikhailov SN (1987) Carbohydr Res 166:219
42. Guschlbauer W, Jankowski K (1980) Nucleic Acids Res 8:1421
43. Campbell MA, Wengel J (2011) Chem Soc Rev 40:5680
44. Futatsugi K, Yamamoto H (2005) Angew Chem Int Ed 44:1484
45. Rico AR, Tlahuextl M, Flores-Parra A, Contreras R (1999) J
Organomet Chem 581:122
1
1
1
1
1
2
0. Engelhart AE, Hud NV (2010) Cold Spring Harbor Perspect Biol
2
:a002196
1
23