Base moiety selectivity in cleavage of short oligoribonucleotides by di- and tri-nuclear Zn(II) complexes of azacrown-derived ligands

Article abstract of DOI:10.1039/b904828f

Cleavage of 6-mer oligoribonucleotides by the dinuclear Zn²⁺ complex of 1,3-bis[(1,5,9-triazacyclododecan-3-yl)oxymethyl]benzene (L ¹) and the trinuclear Zn²⁺ complex of 1,3,5-tris[(1,5,9- triazacyclododecan-3-yl)oxymethyl

Full text of DOI:10.1039/b904828f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Base moiety selectivity in cleavage of short oligoribonucleotides by di- and
tri-nuclear Zn(II) complexes of azacrown-derived ligands†
Maarit Laine, Kaisa Ketoma¨ki, Pa¨ivi Poija¨rvi-Virta* and Harri Lo¨nnberg
Received 9th March 2009, Accepted 9th April 2009
First published as an Advance Article on the web 19th May 2009
DOI: 10.1039/b904828f
Cleavage of 6-mer oligoribonucleotides by the dinuclear Zn2+ complex of 1,3-bis[(1,5,9-
triazacyclododecan-3-yl)oxymethyl]benzene (L1) and the trinuclear Zn2+ complex of
1,3,5-tris[(1,5,9-triazacyclododecan-3-yl)oxymethyl]benzene (L3) has been studied. The dinuclear
complex cleaves at sufficiently low concentrations ([(Zn2+)2L1] ≤ 0.1 mmol L-1) the 5¢NpU3¢ and 5¢UpN3¢
bonds (N = G, C, A) much more readily than the other phosphodiester bonds, but leaves the 5¢UpU3¢
site intact. The trinuclear (Zn2+)3L3 complex, in turn, cleaves the 5¢UpU3¢ bond more readily than any
other linkages, even faster than the 5¢NpU3¢ and 5¢UpN3¢ sites. Somewhat unexpectedly, the 5¢UpNpU3¢
site is cleaved only slowly by both the di- and tri-nuclear complex. The base-moiety selectivity remains
qualitatively similar, though slightly less pronounced, when the hexanucleotides are closed to hairpin
loops by three additional CG-pairs of 2¢-O-methylribonucleotides. Phosphodiester bonds within a
double helical stem are not cleaved, not even the 5¢UpU3¢ sites. Guanine base also becomes recognized
by (Zn2+)2L1 and (Zn2+)3L3, but the affinity to G is clearly lower than to U. The trinuclear cleaving
agent, however, cleaves the 5¢GpG3¢ bond only 35% less readily than the 5¢UpU3¢ bond.
moiety serves as a cleaving agent. With 5¢UpU3¢, both of the
azacrown moieties are engaged in the base moiety binding. The
Introduction
Dinuclear Zn2+ complexes have received considerable interest as
chemical models of enzymes catalyzing phosphoryl transfer and
as constituents of artificial RNases.1–9 Most of the studies have
been inspired by the concept of double Lewis acid activation, i.e.
the marked acceleration of phosphodiester cleavage by dinuclear
complexes capable of simultaneous binding to both of the non-
bridging oxygen atoms.10 Much less attention has been paid to the
fact that dinuclear complexes may simultaneously interact with
the base and phosphate moieties of RNA and, hence, induce base
moiety selective cleavage. In this respect, azacrown complexes of
Zn2+ are of particular interest, since they, besides their marked
cleaving activity,11 recognize uracil and thymine bases.12
catalytic activity of (Zn2+)2L1 is, hence, lost, but it can be restored
by addition of a third azacrown group, as with (Zn2+)3L3. For
comparison, di- and tri-nuclear Cu2+ complexes of calix[4]arenes
bearing azacrown ligands at the upper rim have been observed to
preferably cleave uracil containing dinucleoside-3,5-phosphates.13a
Besides uracil, guanine base offers a site of anchoring for
the Zn2+ complexes of ligands L2 and L3.8a Binding to N1-
deprotonated guanine is, however, considerably weaker than
binding to N3-deprotonated uracil. Hence, dinucleoside-3¢,5¢-
phosphates that consist of one guanosine and one adenosine
or cytidine are cleaved by the dinuclear Zn2+ complex of L2
approximately one order of magnitude less readily than their uracil
counterparts, but still three times as fast as the dimers that contain
only adenine and/or cytosine bases.
We have shown previously that Zn2+ complexes of di- and
tri-azacrown ligands exhibit as cleaving agents considerable
base moiety selectivity when dinucleoside-3¢,5¢-phosphates are
used as substrates.8a,8c Among the four possible dinucleoside-
3¢,5¢-phosphates derived from A and U, the heterodimers,
5¢ApU3¢ and 5¢UpA3¢, are cleaved by the dinuclear Zn2+ com-
plexes of 1,3- (L1) and 1,4- (L2) bis[(1,5,9-triazacyclododecan-3-
yl)oxymethyl]benzene up to two orders of magnitude more readily
than 5¢UpU3¢ or 5¢ApA3¢. The trinuclear complex of 1,3,5-tris[(1,5,9-
triazacyclododecan-3-yl)oxymethyl]benzene (L3), in turn, cleaves
5¢UpU3¢ as readily as 5¢ApU3¢ and 5¢UpA3¢, while the cleavage of
The present study has been undertaken to find out whether this
kind of base moiety selectivity also operates at oligonucleotide
level, i.e. in the presence of a plethora of binding sites. For this
purpose, a set of 6-mer oligoribonucleotides (1–10 in Scheme 1)
have been synthesized and their susceptibility to hydrolysis by the
Zn2+ complexes of ligands L1 and L3 has been studied. To find out
whether the situation within hairpin loops is similar to that with
relaxed single strands, some representatives of the hexanucleotides
have been closed to hairpin loops with three additional GC
pairs of 2¢-O-methylribonucleotides (11–13). In addition, the
cleavage of an 18-mer oligoribonucleotide hairpin (14) containing
1
5¢ApA3¢ remains slow.8c Kinetic, UV-spectrophotometric and H
NMR spectroscopic studies suggest that with 5¢ApU3¢ and 5¢UpA3¢,
one of the Zn2+-azacrown moieties anchors the cleaving agent to
deprotonated N3-site of the uracil base, while the other azacrown
a
5¢UpU3¢/3¢ApA5¢ site within the double helical stem has been
studied to find out whether the trinuclear (Zn2+)3L3 complex
could by binding to the UpU3¢ site displace the complementary
5¢
Department of Chemistry, University of Turku, FIN-20014, Turku, Finland.
E-mail: paipoi@utu.fi; Fax: +358-2-3336700; Tel: +358-2-3336771
† Electronic supplementary information (ESI) available: HPLC-
chromatograms and MS data. See DOI: 10.1039/b904828f
strand and then cleave the intervening phosphodiester bond. In
other words, could this cleaving agent be utilized to hydrolyze a
phosphodiester bond within a double helical stem, which is known
2780 | Org. Biomol. Chem., 2009, 7, 2780–2787
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Scheme 1 Di- and-tri-nucleating azacrown ligands and oligonucleotides used in the present study.
to withstand prolonged treatment with monomeric azacrown
chelates of Zn2+.11a,11b To the best of our knowledge, previous
studies on base moiety selective cleavage of oligonucleotides
by metal ion complexes are limited to the recent observation
that di- and tri-nuclear Cu2+ complexes of azacrown-decorated
calix[4]arenes prefer hydrolysis of CpA phosphodiester bonds.13b
(0.1 mol L-1, 65 ◦C, 30 min). The crude oligonucleotides were puri-
fied by RP-HPLC and their identity was verified by ESI-MS.
Kinetic measurements and determination of the cleavage site
As shown previously,14 the Zn2+-promoted cleavage of oligori-
bonucleotides proceeds by an attack of the 2¢-hydroxy group on
the phosphorus atom with concomitant departure of the 5¢-linked
nucleoside (Scheme 2). The role of the metal ion evidently is
to facilitate the proton transfer from the attacking nucleophile
(2¢-O) to the departing nucleophile (5¢-O), the latter step being rate-
limiting.15 The 2¢,3¢-cyclic phosphate obtained is finally hydrolyzed
to a mixture of 2¢- and 3¢-phosphates.
Results and discussion
Syntheses
The synthesis of the di- and tri-nucleating azacrown ligands, 1,3-
bis[(1,5,9-triazacyclododecan-3-yloxy)methyl]benzene (L1) and
1,3,5-tris[(1,5,9-triazacyclododecan-3-yl)oxymethyl]benzene (L3),
has been reported previously.8c Oligoribonucleotides 1–14 were
prepared by standard phosphoramidite protocol, using 2¢-O-
triisopropylsilyloxymethyl (TOM) building blocks and CPG-
succinyl-support. The TOM-protecting groups were removed
by triethylamine trihydrofluoride treatment in DMSO (65 ◦C,
30 min), followed by treatment with aq sodium acetate
The progress of the Zn2+-complex promoted cleavage of oligonu-
cleotides 1–14 was followed at pH 7.5 and 35 ◦C by withdrawing
aliquots from the reaction solution at suitable intervals and ana-
lyzing their composition by capillary electrophoresis. Potassium
4-nitrotoluenesulfonate was used as an internal standard. All
kinetic measurements were carried out in excess of the cleaving
agent, either (Zn2+)2L1 or (Zn2+)3L3. The concentration of the
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2780–2787 | 2781
©

View Article Online
Scheme 2 Mechanism of Zn2+-promoted cleavage of RNA phosphodiester bonds (a = apical ligand, e = equatorial ligand).
target oligonucleotide was 50 mmol L-1. The disappearance of
the starting material followed first-order kinetics, as exemplified
by Fig. 1. The kinetic data obtained is indicated in Tables 2–3 and
Figs. 2 and 3, and it is discussed in more detail below.
The preferred cleavage sites were determined for oligonu-
cleotides that were rapidly hydolyzed and, hence, selective cleavage
of some phosphodiester bonds by the base moiety anchored
cleaving agent was expected. The reactions were allowed to proceed
for 2–3 half-lives and the reaction mixtures were then subjected to
RP-HPLC separation. Major peaks were collected and identified
by ESI-MS. The HPLC-chromatograms and MS data are given as
the Supporting information. The predominant cleavage sites are
listed in Table 1.
Cleavage of oligonucleotides containing only A, C and U
Figs. 2 and 3 show the logarithmic first-order rate constants for
the cleavage of oligonucleotides 5¢CAACAC3¢ (1), 5¢CAAUAC3¢
(2), and 5¢CAUUCA3¢ (3) as a function of the logarithmic
concentration of (Zn2+)2L1 and (Zn2+)3L3, respectively. As seen,
all the oligonucleotides are cleaved approximately as readily at
the high concentration of the cleaving agent (2.0 mmol L-1), the
cleaving activity of (Zn2+)3L3 being 2- to 3-fold compared to that
of (Zn2+)2L1. Accordingly, the presence of one or two uracil bases
as the preferred site of anchoring of the cleaving agent does not
result in any observable rate acceleration, but random cleavage
of the phosphodiester bonds predominates. The higher cleaving
activity of the trinuclear complex may, at least partly, be attributed
to the 1.5 times higher overall concentration of the Zn2+ azacrown
chelate (Zn2+[12]aneN3) moieties. On going to low concentrations
of the cleaving agent (0.10 mmol L-1) the situation, however,
dramatically changes. When the dinuclear (Zn2+)2L1 is used as
a catalyst, oligonucleotide 2, containing one U, is cleaved 30 times
as fast as 1, containing only C and A, and almost 20 times as fast as
3, containing two adjacent U residues. According to MS-analysis
of the HPLC separated products, the cleavage of 2 predominantly
Fig. 1 Cleavage of hexanucleotide 5¢-CGAGCA-3¢ by the (Zn2+)3L3
complexes in HEPES buffer (0.10 mol L-1, pH 7.5, 35 ◦C, I
=
0.10 mol L-1).
Table 1 Predominant cleavage sites (↓) for the sequence-selective break-
down of 6-mer oligoribonucleotides by (Zn2+)2L1 or (Zn2+)3L3 in
0.10 mol L-1 HEPES buffer (0.10 mol L-1, pH 7.5, 35 ◦C, I = 0.10 mol L-1)
Oligonucleotide
Complex
Sites of cleavage
2
2
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
(Zn2+)3L3
5¢-CAA↓U↓AC-3¢
5¢-CAAU↓AC-3¢
3
5¢-CAU↓UAC-3¢
5
5¢-CUAC↓U↓A-3¢
5¢-CAAG↓AC-5¢
6
Fig. 2 Logarithmic rate constants for the breakdown of 5¢CAACAC3¢
(1;ꢀ), 5¢CAAUAC3¢ (2;᭹), and 5¢CAUUCA3¢ (3;ꢀ) as a function of
the logarithmic concentration of (Zn2+)2L1 at pH 7.5 and 35 ◦C (I =
0.1 mol L-1).
7
5¢-CAG↓GCA-3¢
5¢-CA↓G↓U↓CA-3¢
5¢-C↓G↓A↓G↓CA-3¢
5¢-C↓G↓AC↓G↓A-3¢
8
9
10
2782 | Org. Biomol. Chem., 2009, 7, 2780–2787
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
mental values: 3.3 ¥ 10-5 s-1, 3.9 ¥ 10-5 s-1 and 3.6 ¥ 10-5 s-1 obtained
with 1, 2 and 3, respectively. On going to the lowest concentration
of (Zn2+)2L1 (0.1 mmol L-1), the rate of the random cleavage should
be decelerated by a factor of 20 or 40, in the absence or presence of
the uracil-anchoring, respectively. In fact, the random background
cleavage appears to be slower, since the observed rate constant for
the disappearance of 1, viz. 2.4 ¥ 10-7 s-1, is considerably smaller
than the predicted value 1.6 ¥ 10-6 s-1. Anyway, the random cleav-
age becomes so slow that the effect of uracil-anchoring becomes
clearly visible. Qualitatively similar considerations apply to the
cleavage promoted by (Zn2+)3L3. At 2 mmol L-1 concentration
of the cleaving agent, the rate constants for the random cleavage
may be estimated to be 4.5 ¥ 10-5 s-1, in reasonable agreement
with the experimental value of 7.1 ¥ 10-5 s-1 for 1. At 0.1 mmol L-1
concentration, in turn, the predicted value for 1 is 2.3 ¥ 10-6 s-1 and
the observed one is again considerably smaller, viz. 4.4 ¥ 10-7 s-1.
It is worth noting that the lowest concentration of (Zn2+)2L1
and (Zn2+)2L2 (0.10 mmol L-1) in the cleaving experiments is only
double compared to the concentration of the oligonucleotide. Our
previous studies,8c however, show that the formation of ternary
complex between 5¢UpU3¢ and (Zn2+)2L1 is quantitative under such
conditions. In other words, half of (Zn2+)2L1 is engaged in the
ternary complex with the oligonucleotide. The rest uncomplexed
(Zn2+)2L1 undoubtedly is partially dissociated, since the binary
complex is considerably less stable than the ternary complex.8c The
concentration of uncomplexed Zn2+ must, however, remain below
0.1 mmol L-1 and, as discussed above, such a low concentration
does not result in marked cleavage.
Fig. 3 Logarithmic rate constants for the breakdown of 5¢CAACAC3¢
(1;ꢀ), 5¢CAAUAC3¢ (2;᭹), and 5¢CAUUCA3¢ (3;ꢀ) as a function of
the logarithmic concentration of (Zn2+)2L1 at pH 7.5 and 35 ◦C (I =
0.1 mol L-1).
takes place by rupture of the phosphodiester bonds neighboring
the U residue, i. e. the 5¢ApU3¢ and 5¢UpA3¢ bonds (Table 1).
Evidently anchoring of one of the Zn2+[12]aneN3 moieties of
(Zn2+)2L1 to the uracil base keeps the other Zn2+[12]aneN3 moiety
in the vicinity of the neighboring phosphodiester bonds, resulting
in site-selective cleavage. In the presence of two adjacent U
residues, both Zn2+[12]aneN3 moieties are engaged in nucleobase
binding and the cleavage of oligonucleotide 3, hence, remains
slow. Addition of a third Zn2+[12]aneN3 moiety to the cleaving
agent, however, changes the situation. As seen from Fig. 3,
oligonucleotide 3 is hydrolyzed in the presence of (Zn2+)3L3 as fast
as 2. Only the 5¢UpU3¢ linkage is now selectively cleaved (Table 1).
With 2, cleavage at the 3¢-side of U predominates. In summary, at
a low concentration of the cleaving agent, a selectivity comparable
to that previously8c reported for dinucleoside-3¢,5¢-phosphates is
achieved, while at high catalyst concentrations random cleavage
of phosphodiester bonds predominates.
Table 2 records the rate constants observed for the hydrolysis
of a larger selection of A, C and U containing oligonucleotides
at a low concentration (0.10 mmol L-1) of (Zn2+)2L1 or (Zn2+)3L3.
Among the oligomers studied, 4 and 5 contain two uracil bases,
like oligomer 3 discussed above, but separated by one (A) or two
(5¢ApU3¢) intervening nucleosides, respectively. As indicated before,
the presence of two adjacent uracil bases prevents the catalytic
action of the dinuclear (Zn2+)2L1 complex, but allows efficient
cleavage of the intervening phosphodiester bond by the trinuclear
(Zn2+)3L3 complex. The dinuclear (Zn2+)2L1 evidently bridges two
uracil bases even when they are separated by one nucleoside, since
The fact that the cleavage reaction exhibits base moiety selec-
tivity only at low concentrations of the cleaving agent is expected.
We have shown previously that the second-order rate constant
for the cleavage of individual phosphodiester bonds within an
oligonucleotide structure by the Zn2+[12]aneN3 complex fall in the
range 4.0 ¥ 10-4 L mol-1 s-1-2.6 ¥ 10-3 s-1 under the conditions
used in the present study.11a The cleaving activity of Zn2+ aqua ion
is very similar, the second order rate constants ranging from 3.9 ¥
10-4 L mol-1 s-1 to 2.5 ¥ 10-3 L mol-1 s-1. In addition, studies with
dinucleoside-3¢,5¢-phosphates indicate that neither of these cata-
lysts exhibit any marked base selectivity. 8a,c,14b Accordingly, an av-
erage second-order rate constant for the cleavage of a single phos-
phodiester bond by a Zn2+ azacrown complex or Zn2+ aqua ion is
1.5 ¥ 10-3 L mol-1 s-1 and, hence, the rate constant for the disappear-
ance of a 6-mer oligonucleotide is 7.5 ¥ 10-3 L mol-1 s-1. At the high-
est concentration of the cleaving agent, [(Zn2+)2L1] = 2.0 mmol L-1,
i.e. in 40-fold excess of the oligonucleotide target (50 mmol L-1),
the rate constant for the disappearance of 1–3 by random cleavage
(without any base moiety anchoring) may be expected to be 3 ¥
10-5 s-1. This estimation is in excellent agreement with the experi-
Table 2 First-order rate constants for the cleavage of oligoribonu-
cleotides containing A,
C and U
by (Zn2+)2L1 and (Zn2+)3L3
(0.10 mmol L-1) in HEPES buffer (0.10 mol L-1) at pH 7.5 and 35 ◦C
(I = 0.10 mol L-1)
Oligonucleotide
Complex
k/10-6 s-1
1
2
3
4
5
5¢CAACAC3¢
5¢CAAUAC3¢
5¢CAUUCA3¢
5¢CUAUCA3¢
5¢CUACUA3¢
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
(Zn2+)2L1
(Zn2+)3L3
0.24 0.03
0.44 0.03
7.70 0.05
15.6 0.2
0.44 0.05
19.8 0.4
0.27 0.02
1.84 0.03
1.12 0.16
16.2 0.5
0.48 0.20
0.68 0.05
3.15 0.07
2.57 0.09
0.39 0.05
5.66 0.50
11 5¢GCG-CAACAC-CGC3¢ (hairpin)
12 5¢GCG-CAAUCA-CGC3¢ (hairpin)
13 5¢GCG-CAUUCA-CGC3¢ (hairpin)
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2780–2787 | 2783
©

View Article Online
oligomer 4 containing a sequence 5¢UpApU3¢ remains stable in the
presence of this complex, similar to 3 containing a 5¢UpU3¢ site.
In striking contrast to the rapid cleavage of 3 by (Zn2+)3L3, 4 is
cleaved only slowly by this complex. The intervening nucleoside
(A) somehow appears to prevent the interaction of the third
Table 3 First-order rate constants for the cleavage of oligoribonu-
cleotides containing all four nucleotides by (Zn2+)2L1 and (Zn2+)3L3
(0.10 mmol L-1) in HEPES buffer (0.10 mol L-1) at pH 7.5 and 35 ◦C
(I = 0.10 mol L-1)
Oligonucleotide
Complex k/10-6 s-1
5¢
Zn2+[12]aneN3 moiety of the UpApU3¢-anchored cleaving agent
6
7
8
9
5¢CAAGAC3¢
5¢CAGGCA3¢
5¢CAGUCA3¢
5¢CGAGCA3¢
(Zn2+)2L1 1.02 0.06
(Zn2+)3L3 4.66 0.29
(Zn2+)2L1 1.92 0.12
(Zn2+)3L3 12.8 0.91
(Zn2+)2L1 2.17 0.05
(Zn2+)3L3 6.43 0.11
(Zn2+)2L1 1.39 0.06
(Zn2+)3L3 4.30 0.06
(Zn2+)2L1 1.93 0.04
(Zn2+)3L3 4.37 0.08
with the 5¢UpA3¢ and 5¢ApU3¢ bonds.
Oligomer 5, having the two uracil bases separated by two
nucleosides (5¢ApC3¢), behaves like oligomer 2, containing a single
U, in the sense that it is readily cleaved by (Zn2+)3L3. However,
even in this case simultaneous binding to both uracil bases seems
to play a role, since cleavage by the dinuclear (Zn2+)2L1 remains
relatively slow. The reason why among the two uridine residues the
one close to the 3¢-terminus constitutes the predominant cleavage
site of (Zn2+)3L3 (Table 1) remains obscure.
10 5¢CGACGA3¢
14 5¢GGC-GUUCAACGGAAC-GCC3¢ (hairpin) (Zn2+)2L1 0.48 0.20
(Zn2+)3L3 0.68 0.05
To find out whether the base-selectivity of the cleavage reaction
within hairpin loops is similar to that within relaxed single
strands, hexanucleotide loops (11–13) were generated using three
GC pairs of 2¢-O-methylribonucleotides to close the loop. It
has been reported16 that oligoribonucleotides consisting of three
closing GC-base pairs and an intervening non-self-complementary
sequence of 4–9 nucleotides adopt at 37 ◦C and pH 7.0 (I =
0.1 mol L-1 or 1.0 mol L-1) a hairpin loop structure. No
dimerization to structures containing an internal loop has been
observed at oligonucleotide concentrations less than 1 mmol L-1.
The melting point of these hairpins ranges from 55 to 70 ◦C. For
example, a hairpin 5¢GCGUGACAGCCGC3¢, closely resembling
oligonucleotides 11–13 has a melting point of 56.8 ◦C. For
comparison, oligomer 11 melts at 53.7 ◦C. Hairpin 14 is, in turn,
closed by four GC-pairs. We have shown previously11b that hexa-
and hepta-nucleotide loops closed with 4 CG base-pairs of 2¢-O-
oligoribonucleotides exhibit at pH 7.0 (I = 0.1 mol L-1◦ with NaCl)
concentration independent melting points of 87 1 C and 89
1 ◦C, respectively. Unfortunately, the melting temperatures cannot
be determined in the presence of the cleaving agents, since the
concomitant reasonably fast hydrolysis inevitably interferes with
the measurement at elevated temperatures. Binding of the cleaving
agent to a uracil base within the loop structure possibly induces
some conformational changes, but it hardly results in such a strain
that the hairpin structure would be opened. In all likelihood the
hairpin structure still predominates at 35 ◦C, i. e. 20 ◦C below its
inherent melting temperature.
Comparison of the cleavage rates of hairpins 11–13 with those
of the corresponding linear hexanucleotides 1–3 (Table 2) reveals
that the selectivity within loops is qualitatively similar to that
within linear single strands, although somewhat less pronounced.
For example, 5¢CAUUCA3¢ (3) is cleaved with (Zn2+)3L3 45 times
as rapidly as 5¢CAACAC3¢ (1). When the same sequences are
incorporated into a loop structure, the ratio of their cleavage
rates is 8.3. Replacement of the non-terminal C in 1 with U to
obtain 2, in turn, accelerates the cleavage by factors 32 and 35 on
using cleaving agents (Zn2+)2L1 and (Zn2+)3L3, respectively. Within
a hairpin loop, the corresponding accelerations (12 compared to
11) are only 6.6 and 3.8.
guanine base, although more weakly than with uracil base. To
find out whether this weak interaction is sufficient to result in
base-selective cleavage at oligonucleotide level, U in oligonu-
cleotides 2–5 was replaced with G. The kinetic data for the
cleavage of the oligonucleotides obtained (6–10) is given in
Table 3.
Several lines of evidence indicate that guanine base is recognized
by (Zn2+)2L1 and (Zn2+)3L3, but binding to G is clearly weaker than
to U. Firstly, oligonucleotide 2 incorporating a single U is cleaved
7.5 and 3.3 times as fast as its G-substituted counterpart 6 by
(Zn2+)2L1 and (Zn2+)3L3, respectively. The predominant cleavage
site of 6 is at the 3¢-side of the G residue (Table 1). Secondly,
5¢GpG3¢-containing 7 is cleaved by dinuclear (Zn2+)2L1 4.4 times
as fast as the 5¢UpU3¢-containing 3. In other words, (Zn2+)2L1 is
not arrested to a 5¢GpG3¢-site as quantitatively as to a 5¢UpU3¢-
site. In striking contrast to the slow hydrolysis of the 5¢UpU3¢-
containing 3 compared to its 5¢ApU3¢-containing counterpart 2,
the 5¢GpG3¢-containing 7 is cleaved even faster than the 5¢ApG3¢-
containing 6. With trinuclear (Zn2+)3L3, 7 is cleaved 35% more
slowly than 3, the cleavage expectedly taking place between the
two guanosines. Thirdly, replacement of the 5¢ApU3¢ site in 2 with
a 5¢GpU3¢ site (8) results in a 3.5-fold deceleration on the cleavage
by (Zn2+)2L1. The guanine base evidently competes with the
scissile phosphodiester linkages for the remaining Zn2+[12]aneN3
moiety of the uracil-anchored cleaving agent. (Zn2+)3L3-promoted
cleavage is decelerated by a factor of 2.4, most likely due to lower
affinity to the 5¢GpU3¢ site. Hydrolysis of the 5¢GpU3¢ bond clearly
predominates, but the neighboring phosphodiester bonds are also
cleaved to a detectable extent. Finally, oligonucleotides 9 and 10
containing two G residues separated by one and two nucleosides,
respectively, are cleaved by both the di- and tri-nuclear complex
approximately as rapidly as oligonucleotide 6 incorporating only
a single G. Phosphodiester bonds on both sides of the G
residues are cleaved. Accordingly, no indication of simultaneous
binding of the guanine bases to either of the cleaving agents is
obtained.
It has been known since the early studies of Usher and McHale17
that the phosphodiester bonds are hydrolyzed within a double
helix much less readily than within a single strand, because stack-
ing between the neighboring base-pairs prevents the 5¢-linked
nucleoside to adopt within the phosphorane intermediate an
Cleavage of oligonucleotides containing all four nucleotides
Previous studies8a with dinucleoside-3¢,5¢-phosphates have shown
that cleaving agents (Zn2+)2L1 and (Zn2+)3L3 also interact with
2784 | Org. Biomol. Chem., 2009, 7, 2780–2787
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
apical position required for the PO5¢ bond rupture. More re-
cent studies have shown that this is also the case on using
Zn2+[12]aneN3 as a cleaving agent.11a,11b Since Zn2+[12]aneN3 binds
to N3 and O4 of uracil, i.e. to atoms engaged in Watson-Crick
hydrogen bonding within a double helix, one might speculate
that (Zn2+)3L3 could possibly perform a chain invasion at a
5¢UpU3¢ site and cleave the intervening phosphodiester bond.
To find out whether this really is the case, an 18-mer hairpin
structure (14) consisting of a 5¢ApApCpG3¢ tetraloop of 2¢-O-
methylribonucleosides and a potential 5¢UpU3¢/3¢ApA5¢ cleavage
site in the stem was prepared. The stem additionally contained
one CG-pair on both sides of the 5¢UpU3¢/3¢ApA5¢ site and three
terminal GC-pairs of 2¢-O-methylribonucleotides. Both agents,
(Zn2+)2L1 and (Zn2+)3L3, cleaved this hairpin only slowly and at a
comparable rate. Evidently, neither the di- nor tri-nuclear complex
was able to cleave phosphodiester bonds within the double helical
stem.
Experimental section
Synthesis of oligonucleotides and cleaving agents
Ligands L1 and L3 were prepared as described previously.8c
Oligoribonucleotides 1–14 were synthesized by conventional
phosphoramidite strategy from 2¢-O-triisopropylsilyloxymethyl
(2¢-O-TOM) protected phosphoramidite building blocks (Glenn
Research) on a 1 mmol L-1 scale, following the standard RNA-
coupling protocol of ABI-392 DNA/RNA synthesizer. The 5¢-
O-DMTr group was removed on-support, except in case of
oligonucleotide 14. Detritylation was then carried out after
chromatographic purification by treatment with 80% aq acetic acid
(300 mL) at room temperature for 1 h. The solvent was removed
by evaporation and the residue was dissolved in water (500 mL).
The aqueous layer was washed with EtOAc (2 ¥ 500 mL). The
2¢-O-TOM protecting groups were removed with triethylamine
trihydrofluoride in DMSO (65 ◦C, 30 min) followed by treatment
◦
with aq sodium acetate (0.1 mol L-1, 65 C, 30 min). The crude
Conclusion
oligonucleotides were purified and desalted by RP-HPLC, except
oligonucleotide 14 which was desalted by NAPTM-5 column.
Chromatographic conditions: a Hypersil ODS column (250 ¥
4.6 mm, particle size 5 mm), flow rate 1 mL min-1, buffer A =
aq NH4OAc (50 mmol L-1), buffer B = 1:1 mixture of buffer A
and MeCN, a linear gradient from 0 to 25% buffer B in buffer A
in 25 min, followed by a linear gradient from 25 to 40% buffer
B in buffer A in 5 min. All aqueous solutions were prepared
in sterilized water and sterilized equipment was used for their
handling.
Cleavage of short oligoribonucleotides by a di- and tri-nuclear Zn2+
complex of [12]aneN3-derived ligands containing additionally
an ether oxygen as a potential H-bond acceptor site has been
studied. The results show that the marked base-moiety selectivity
previously8a,c reported for dinucleoside-3¢,5¢-phosphates is suffi-
cient to allow selective cleavage of short oligonucleotides in spite of
the presence of a plethora of potential binding sites. Accordingly,
the dinuclear complex cleaves at sufficiently low concentrations
([(Zn2+)2L1] ≤ 0.1 mmol L-1) the 5¢NpU3¢ and 5¢UpN3¢ bonds
(N = G, C, A) more readily than the other phosphodiester
bonds, but leaves the 5¢UpU3¢ sites intact. The trinuclear (Zn2+)3L3
complex, in turn, cleaves the 5¢UpU3 bonds more readily than
any other linkages, even faster than the 5¢NpU3¢ and 5¢UpN3¢ sites.
Interestingly, an intervening nucleoside between two U residues
(5¢UpNpU3¢), however, prevents the catalytic action. Within a
double helix, a 5¢UpU3¢ bond is not cleaved. In addition to uracil,
guanine base is also recognized by (Zn2+)2L1 and (Zn2+)3L3, but
the affinity to G is clearly lower than to U. The trinuclear cleaving
agent still cleaves the 5¢GpG3¢ bonds only 35% less readily than the
5¢UpU3¢ bonds. The base-moiety selectivity remains qualitatively
similar, though somewhat less pronounced, when the sequences
are closed to hexanucleotide hairpin loops by three additional
CG-pairs of 2¢-O-methylribonucleotides.
While the observations discussed above encourage one to
continue development of small molecule cleaving agents specific
for sequences of only a few bases, it is quite clear that the complexes
studied in the present paper do not yet allow manipulation of
really long RNA sequences. Introduction of a single uracil base
into a 6-mer oligonucleotide results in a 30-fold rate-acceleration.
Since the rate of the random background cleavage is proportional
to the number of phosphodiester bonds in the oligomer, the
site-selectivity of the cleavage is gradually lost on approaching
30 nucleotides long sequences. Similarly, though binding to two
contiguous uracil bases is preferred over binding to a single uracil,
the difference in stability is not sufficient for selective cleavage of
UpU sites within long RNA sequences. Nevertheless, the cleaving
agents studied may well find useful applications as chemical probes
for the structural elucidation of medium size RNA molecules, such
as tRNA.
The identity of 1–14 was verified by ESI-MS (Bruker
micrOTOF-Q ESI-MS system). 1: m/z obsd. (calcd.) 919.2 (919.6)
[M - 2H]2-, 612.4 (612.7) [M - 3H]3-, 459.1 (459.3) [M - 4H]4-. 2:
m/z obsd. (calcd.) 919.7 (920.0) [M - 2H]2-, 612.8 (613.1) [M -
3H]3-, 459.3 (459.5) [M - 4H]4-. 3: m/z obsd. (calcd.) 908.2 (908.6)
[M - 2H]2-, 605.1 (605.4) [M - 3H]3-, 453.6 (453.8) [M - 4H]4-. 4:
m/z obsd. (calcd.) 908.1 (908.6) [M - 2H]2-, 605.1 (605.4) [M -
3H]3-. 5: m/z obsd. (calcd.) 1817.3 (1818.2) [M - H]-, 908.1 (908.6)
[M - 2H]2-. 6: m/z obsd. (calcd.) 939.2 (939.6) [M - 2H]2-, 625.8
(626.1) [M - 3H]3-, 469.1 (469.3) [M - 4H]4-. 7: m/z obsd. (calcd.)
947.2 (947.6) [M - 2H]2-, 631.1 (631.4) [M - 3H]3-, 473.1 (473.3)
[M - 4H]4-. 8: m/z obsd. (calcd.) 927.7 (928.1) [M - 2H]2-, 618.1
(618.4), [M - 3H]3-, 463.3 (463.5). 9: m/z obsd. (calcd.) 947.2
(947.6) [M - 2H]2-, 631.1 (631.4) [M - 3H]3-, 473.1 (473.3) [M -
4H]4-. 10: m/z obsd. (calcd.) 947.2 (947.6) [M - 2H]2-, 631.1 (631.4)
[M - 3H]3-, 473.1 (473.3) [M - 4H]4-. 11: m/z obsd. (calcd.) 1290.9
(1290.7) [M - 3H]3-, 967.9 (967.7), [M - 4H]4-, 774.1 (774.0) [M -
5H]5-, 644.9 (644.8) [M - 6H]6-, 552.7 (552.7) [M - 7H]7-. 12: m/z
obsd. (calcd.) 1291.2 (1291.1) [M - 3H]3-, 968.2 (968.2) [M - 4H]4-,
774.3 (774.4) [M - 5H]5-, 645.1 (645.2) [M - 6H]6-, 552.8 (552.9)
[M - 7H]7-. 13: m/z obsd. (calcd.) 1283.5 (1283.7) [M - 3H]3-,
962.4 (962.5) [M - 4H]4-, 769.7 (769.8) [M - 5H]5-, 641.3 (641.3)
[M - 6H]6-, 549.5 (549.6) [M - 7H]7-. 14: m/z obsd. (calcd.) 1959.3
(1959.3) [M - 3H]3-, 1469.2 (1469.2) [M - 4H]4-, 1175.2 (1175.2)
[M - 5H]5-, 979.1 (979.1) [M - 6H]6-, 839.1 (839.1) [M - 7H]7-,
734.1 (734.1) [M - 8H]8-, 652.4 (652.4) [M - 9H]9-.
Kinetic experiments
The reactions were carried out in Eppendorf tubes which were
immersed in a thermostated water bath at 35
0.1 ◦C. The
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2780–2787 | 2785
©

View Article Online
solutions of the Zn2+ complexes were prepared by mixing stoi-
chiometric amounts of ligand L1 or L3 and Zn(NO3)2 in sterilized
water overnight at room temperature. HEPES buffer (0.10 mol L-1,
pH 7.5), the target oligoribonucleotide (50 mmol L-1) and potas-
sium 4-nitrobenzenesulfonate (internal standard, 0.10 mmol L-1)
were then added to the solution. The ionic strength was adjusted
to 0.10 mol L-1 with NaNO3. The total volume of the reaction
mixture was 200 mL. Aliquots of 20 mL were withdrawn at suitable
time intervals and cooled immediately on an ice-water bath. The
reaction was quenched by adding aq hydrogen chloride (1 mL of
1.0 mol L-1 solution). The aliquots were analyzed on a Beckman
Coulter P/ACE-MDQ Capillary Electrophoresis System using
a fused silica capillary (75 mm inner diameter, 50 cm effective
length) and citrate buffer (0.2 mol L-1, pH 3.1) as a background
electrolyte. The voltage applied was -20 kV and the temperature
of the capillary was kept at 25 ◦C. Between each analytical run, the
capillary was flushed for 3 min with water, aq HCl (10 mmol L-1)
and the background electrolyte. The samples were injected using
hydrodynamic injection with 2 psi for 8 s. The oligonucleotides
and the internal standard were detected by a UV detector at a
wavelength of 254 nm.
First-order rate constants were calculated for the disappearance
of the target oligoribonucleotide by applying the integrated first-
order rate law to the diminution of the signal of the starting
material. For normalizing, the integrated peak areas of the
starting material and internal standards were first divided by the
migration times of the corresponding peaks. Then the starting
material was divided by the internal standard and the first-
order rate constant k was calculated from the first-order rate
law.
Notes and references
1 (a) K. Selmeczi, C. Michel, A. Milet, I. Gautier-Luneau, C. Philouze,
J.-L. Pierre, D. Schnieders, A. Rompel and C. Belle, Chem.–Eur. J.,
2007, 13, 9093; (b) N. Mitic, S. J. Smith, A. Neves, L. W. Guddat, L. R.
Gahan and G. Schenk, Chem. Rev., 2006, 106, 3338; (c) J. Weston,
Chem. Rev., 2005, 105, 2151.
2 (a) M. Komiyama and J. Sumaoka, Curr. Opin. Chem. Biol., 1998,
2, 751; (b) S matsuda, A. Ishikubo, A. Kuzuya, M. Yashiro and M.
Komiyama, Angew. Chem., Int. Ed., 1998, 37, 3284; (c) M. Yashiro, A.
Ishikubo and M. Komiyama, J. Chem. Soc., Chem. Commun., 1997, 83;
(d) M. Yashiro, A. Ishikubo and M. Komiyama, J. Chem. Soc., Chem.
Commun., 1995, 1793.
3 (a) G. Feng, J. C. Mareque-Rivas and N. H. Williams, Chem.
Commun., 2006, 1845; (b) G. Feng, D. Natale, R. Prabaharan, J. C.
Mareque-Rivas and N. H. Williams, Angew. Chem., Int. Ed., 2006, 45,
7056.
4 (a) C. S. Rossiter, R. A. Mathews, I. M. A. del Mundo and J. R. Morrow,
J. Inorg. Biochem., 2009, 103, 64; (b) J. R. Morrow, T. L. Amyes and J. P.
Richard, Acc. Chem. Res., 2008, 41, 539; (c) M-Y. Yang, J. R. Morrow
and J. P. Richard, Bioorg. Chem., 2007, 35, 366; (d) A. O’Donoghue,
S. Y. Pyun, M.-Y. Yang, J. R. Morrow and J. P. Richard, J. Am. Chem.
Soc., 2006, 128, 1615; (e) J. R. Morrow and O. Iranzo, Curr. Opin. Chem.
Biol., 2004, 8, 192; (f) O. Iranzo, J. P. Richard and J. R. Morrow, Inorg.
Chem., 2004, 43, 1743; (g) O. Iranzo, T. Elmer, J. P. Richard and J. R.
Morrow, Inorg. Chem., 2003, 42, 7737; (h) O. Iranzo, A. Y. Kovalevsky,
J. R. Morrow and J. P. Richard, J. Am. Chem. Soc., 2003, 125, 1988;
(i) M.-Y. Yang, J. P. Richard and J. R. Morrow, Chem. Commun., 2003,
2832.
5 (a) F. Mancin and P. Tecilla, New J. Chem., 2007, 31, 800; (b) A. Scarso,
G. Zaupa, F. B. Houillon, L. J. Prins and P. Scrimin, J. Org. Chem.,
2007, 72, 376.
6 (a) R. Cacciapaglia, A. Casnati, L. Mandolini, D. N. Reinhoudt, R.
Salvio, A. Sartori and R. Ungaro, Inorg. Chim. Acta, 2007, 360, 981;
(b) R. Cacciapaglia, A. Casnati, L. Mandolini, D. N. Reinhoudt, R.
Salvio, A. Sartori and R. Ungaro, J. Org. Chem., 2005, 70, 624; (c) P.
Molenveld, J. F. J. Engbersen and D. N. Reinhoudt, Eur. J. Org. Chem.,
1999, 3269; (d) P. Molenveld, S. Kapsabelis, J. F. J. Engbersen and D. N.
Reinhoudt, J. Am. Chem. Soc., 1997, 119, 2948.
7 (a) A. A. Neverov, C. T. Liu, S. E. Bunn, D. Edwards, C. J. White,
S. A. Melnychuk and R. S. Brown, J. Am. Chem. Soc., 2008, 130, 6639;
(b) S. E. Bunn, C. T. Liu, Z.-L. Lu, A. A. Nevrov and R. S. Brown,
J. Am. Chem. Soc., 2007, 129, 16238; (c) A. A. Neverov, Z.-L. Lu, C. I.
Maxwell, M. F. Mohamed, C. J. White, J. S. W. Tsang and R. S. Brown,
J. Am. Chem. Soc., 2006, 128, 16398.
8 (a) Q Wang, E. Leino, A. Jancso´, I. Szila´gyi, T. Gajda, E. Hietama¨ki
and H. Lo¨nnberg, ChemBioChem, 2008, 9, 1739; (b) T. Niittyma¨ki, P.
Virta, K. Ketoma¨ki and H. Lo¨nnberg, Bioconjugate Chem., 2007, 18,
1583; (c) Q. Wang and H. Lo¨nnberg, J. Am. Chem. Soc., 2006, 128,
10716; (d) A. Jancso, S. Mikkola, H. Lo¨nnberg, K. Hegetschweiler and
T. Gajda, J. Inorg. Biochem., 2005, 99, 1283; (e) Q. Wang, S. Mikkola
and H. Lo¨nnberg, Chem. Biodiversity, 2004, 1, 1316; (f) A. Jancso´,
S. Mikkola, H. Lo¨nnberg, K. Hegetschweiler and T. Gajda, Chem.–
Eur. J., 2003, 9, 5404; (g) S. Albedyhl, D. Schnieders, A. Jancso, T.
Gajda and B. Krebs, Eur. J. Inorg. Chem., 2002, 1400; (h) T. Gajda, R.
Kra¨mer and A. Jancso, Eur. J. Inorg. Chem., 2000, 1635; (i) S. Mikkola,
Q. Wang, Z. Jori, M. Helkearo and H. Lo¨nnberg, Acta Chem. Scand.,
1999, 53, 453.
9 (a) M. Subat, K. Woinaroschy, C. Gerstl, W. Kaim and B. Ko¨nig, Inorg.
Chem., 2008, 47, 4661; (b) X. Sheng, X. Guo, X.-M. Lu, Y. Shao, F.
Liu and Q. Xu, Bioconjugate Chem., 2008, 19, 490; (c) C. Bazzicalupi,
A. Bencini, C. Bonaccini, C. Giorgi, P. Gratteri, S. Moro, M. Palumbo,
A. Simionato, J. Sgrignani, C. Sissi and B. Valtancoli, Inorg. Chem.,
2008, 47, 5473; (d) B. Bauer-Siebenlist, F. Meyer, E. Farkas, D. Vidovic
and S. Dechert, Chem.–Eur. J., 2005, 11, 4349; (e) C. Vichard and T.
Kaden, Inorg. Chim. Acta, 2004, 357, 2285; (f) K. Worm, F. Chu, K.
Matsumoto, M. D. Best, V. Lynch and E. V. Anslyn, Chem.–Eur. J.,
2003, 9, 741; (g) A. Bencini, E. Berni, A. Bianchi, C. Giorgi and B.
Valtancoli, Supramol. Chem., 2001, 13, 489.
Determination of the cleavage sites
The (Zn2+)3L3-promoted hydrolysis of hexanucleotides 2, 3, and
5–10, and the (Zn2+)2L1-promoted hydrolysis of 2, was allowed to
proceed from 2 to 3 half-lives in Eppendorf tubes immersed in a
◦
thermostated water bath at 35 0.1 C. The reaction conditions
were as described above, except that the total volume of the
reaction mixture was larger, either 400 mL or 600 mL. The reactions
were quenched by adding aq. hydrogen chloride solution (20 mL or
30 mL of 1.0 mol L-1 solution). The reaction mixtures were cooled
immediately on an ice-water bath and stored in a freezer before
analysis. The composition of the reaction mixtures were analyzed
by RP-HPLC. The cleavage products were separated and collected,
and finally verified by MS analysis. Chromatographic conditions:
Hypersil ODS column (250 ¥ 4.6 mm, particle size 5 mm), flow rate
1 mL min-1, buffer aq. NH4OAc (A = 0.05 mol L-1), buffer B = 1:1
mixture of buffer A and MeCN, a linear gradient from 0 to 15%
buffer B in buffer A in 25 min, followed by a linear gradient from
15 to 30% buffer B in buffer A in 10 min. The results obtained are
given as Supporting Information.
Melting temperature measurements. The melting curves (ab-
sorbance versus temperature) were measured at 260 nm on a
Perkin-Elmer Lambda 2 UV-vis spectrometer equipped with a
Peltier Temperature controller. The temperature was changed at
a rate of 0.5 ◦C/min (from 20 to 90 ◦C). The measurements were
performed in 10 mmol L-1 HEPES buffer (pH 7.5) containing
0.1 mol L-1 NaCl. The concentration of the oligonucleotide was
6 mmol L-1.
10 (a) N. H. Williams, B. Takasaki, M. Wall and J. Chin, Acc. Chem. Res.,
1999, 32, 485; (b) N. H. Williams, W. Cheung and J. Chin, J. Am. Chem.
Soc., 1998, 120, 8079; (c) N. H. Williams and J. Chin, Chem. Commun.,
1996, 131.
11 (a) U. Kaukinen, L. Bielecki, S. Mikkola, R. W. Adamiak and H.
Lo¨nnberg, J. Chem. Soc., Perkin Trans. 2, 2001, 1024; (b) I. Zagorowska,
2786 | Org. Biomol. Chem., 2009, 7, 2780–2787
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
S. Kuusela and H. Lo¨nnberg, Nucleic Acids Res., 1998, 26, 3392; (c) S.
Kuusela and H. Lo¨nnberg, J. Chem. Soc., Perkin Trans. 2, 1994, 2301;
(d) S. Kuusela and H. Lo¨nnberg, J. Phys. Org. Chem., 1993, 6, 347.
12 (a) S. Aoki and E. Kimura, Chem. Rev., 2004, 104, 769; (b) E. Kimura,
M. Kikuchi, H. Kitamura and T. Koike, Chem.–Eur. J., 1999, 5, 3113;
(c) E. Kimura, T. Shiota, M. Koike, M. Shiro and M. Kodama, J. Am.
Chem. Soc., 1990, 112, 5805.
13 (a) R. Cacciapaglia, A. Casnati, L. Mandolini, D. N. Reinhoudt, R.
Salvio, A. Santori and R. Ungaro, J. Am. Chem. Soc., 2006, 128, 12322;
(b) R. Cacciapaglia, A. Casnati, L. Mandolini, A. Peracchi, D. N.
Reinhoudt, R. Salvio, A. Santori and R. Ungaro, J. Am. Chem. Soc.,
2007, 129, 12512.
Soc., Perkin Trans. 2, 1996, 1895; (d) S. Kuusela and H. Lo¨nnberg,
J. Chem. Soc., Perkin Trans. 2, 1994, 2301.
15 (a) T. Lo¨nnberg and H. Lo¨nnberg, Curr. Opin. Chem. Biol., 2005, 9,
665; (b) S. Mikkola, U. Kaukinen and H. Lo¨nnberg, Cell Biochem.
Biophys., 2001, 34, 95; (c) S. Mikkola, E. Stenman, K. Nurmi, E.
Yousefi-Salakdeh, R. Stro¨mberg and H. Lo¨nnberg, J. Chem. Soc.,
Perkin Trans. 2, 1999, 1619.
16 (a) M. J. Serra, T. W. Barnes, K. Betschart, M. J. Gutierrez, K. J.
Sprouse, C. K. Riley, L. Stewart and R. E. Temel, Biochemistry,
1997, 36, 4844; (b) M. J. Serra, M. H. Lyttle, T. J. Axenson, C. A.
Schadt and D. H. Turner, Nucleic Acids Res., 1993, 21, 3845; (c) J.
SantaLucia, Jr. R. Kierzek and D. H. Turner, Science, 1992, 256,
217.
14 (a) S. Kuusela and H. Lo¨nnberg, Curr. Top. Solution Chem., 1997, 2, 29;
(b) S. Kuusela and H. Lo¨nnberg, Nucleosides Nucleotides Nucleic Acids,
1996, 15, 1669; (c) S. Kuusela, A. Guzaev and H. Lo¨nnberg, J. Chem.
17 D. A. Usher and A. H. McHale, Proc. Natl. Acad. Sci. U. S. A., 1976,
73, 1149.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2780–2787 | 2787
©