A novel strategy of chemical modification for rate enhancement of 10-23 DNAzyme: A combination of A9 position and 8-aza-7-deaza-2′-deoxyadenosine analogs

Article abstract of DOI:10.1039/c1ob05065f

With the help of a divalent-metal ion, 10-23 DNAzyme cleaves RNA. Chemical modification of its catalytic loop to make a more efficient enzyme has been a challenge. Our strategy started from its five 2′-deoxyadenosine residues (A5, A9, A11, A12, and A15) i

Full text of DOI:10.1039/c1ob05065f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 5728
www.rsc.org/obc
PAPER
A novel strategy of chemical modification for rate enhancement of 10–23
DNAzyme: a combination of A9 position and
8-aza-7-deaza-2¢-deoxyadenosine analogs†
Junlin He,a Di Zhang,b Qi Wang,a Xia Wei,a Maosheng Cheng*b and Keliang Liu*a
Received 12th January 2011, Accepted 16th May 2011
DOI: 10.1039/c1ob05065f
With the help of a divalent-metal ion, 10–23 DNAzyme cleaves RNA. Chemical modification of its
catalytic loop to make a more efficient enzyme has been a challenge. Our strategy started from its five
2¢-deoxyadenosine residues (A5, A9, A11, A12, and A15) in the loop based on the capability of the N7
atom to form hydrogen bonds in tertiary structures. 8-Aza-7-deaza-2¢-deoxyadenosine and its analogs
with 7-substituents (3-aminopropyl, 3-hydroxylpropyl, or phenethyl) were each used to replace five dA
residues, respectively, and their effect on cleavage rate were evaluated under single-turnover conditions.
The results indicated that the N7 atom of five dA residues were necessary for catalytic activity, and the
N8 atom and 7-substituents were detrimental to the catalytic behavior of 10–23 DNAzyme, except that
all these modifications at A9 were favourable for the activity. Especially, DZ-3–9 with
7-(3-aminopropyl)-8-aza-7-deaza-2¢-deoxyadenosine (3) at A9 position gave a 12- fold increase of kobs
,
compared to the corresponding parent 10–23 DNAzyme. DZ-3–9 was supposed to catalyze the cleavage
reaction with the same mechanism as 10–23 DNAzyme based on their very similar pH-dependent and
divalent metal ions-dependent cleavage patterns. Introduction of functional groups at A9 position was
demonstrated to be a successful and feasible approach for more efficient 10–23 DNAzyme analogs.
with other natural or modified nucleosides, as well as backbone
Introduction
modifications. The nucleotide residues at the two edges of the
catalytic loop are crucial for its catalytic activity, and generally,
any substitution is detrimental to the catalytic efficiency. All
of the 2¢-deoxyguanosine residues are very important. Two 2¢-
deoxythymidine residues, T4 and T8, behave very differently. The
five 2¢-deoxyadenosine residues (A5, A9, A11, A12, and A15)
contribute to the catalytic reaction positively, albeit differently.
Among them, A5 and A15 are the most important residues.
However, it is not clear what roles these residues or their func-
tional groups play; a comprehensive understanding regarding the
mechanistic details of 10–23 DNAzyme is still absent. Therefore,
the basis for a rational modification in the catalytic motif of 10–
23 DNAzyme is currently far beyond reach. Nevertheless, in the
absence of detailed information of the catalytic active site in 10–
23 DNAzyme, chemical modification is still a fundamental tool
to have an insight into the catalytic power of this DNA molecule,
and for a more efficient version.
In our strategy for more efficient 10–23 DNAzyme analogs with
chemical modifications on its catalytic loop, we focused on the
five-membered ring part of 2¢-deoxyadenosine residues, because
their N7 atoms are also active in forming hydrogen bonds, such as
Hoogsteen and reversed Hoogsteen hydrogen-bonding involved in
various tertiary structures of nucleic acids,27 and nitrogen atom can
act as a ligand of metal ions, as testified in ribozyme structures.28
Therefore, five 2¢-deoxyadenosine analogs 1–5 were designed, and
The small deoxyribozyme 10–23 DNAzyme is an artificial catalytic
DNA molecule selected in vitro by Santoro and Joyce. It is com-
prised of a catalytic loop of 15 2¢-deoxyribonucleotides and two
flanking substrate-recognition domains.1,2 Because of its small size,
and chemical and enzymatic stability compared with ribozymes, it
has been broadly studied as a general-purpose genetic therapeutic
candidate against various disease-related mRNAs.3–13 However,
its fast cleavage on mRNA substrate is seriously dependent on the
concentration of divalent metal ion (Mg2+), thus, the low Mg2+
concentration in cells is an unfavorable factor for its practical
applications in vivo. Chemical modifications aimed at improving
its catalytic properties as well as stability have been carried
out since its appearance.14–26 The role of the catalytic motif has
been systematically studied by residue deletion or replacement
aBeijing Institute of Pharmacology and Toxicology, 27 Taiping Road, Beijing,
100850, China. E-mail: keliangliu55@126.com; Fax: +86-10-68211656;
Tel: +86-10-68169363
bSchool of Pharmaceutical Engineering, Shenyang Pharmaceutical Univer-
sity, 103 Wenhua Road, Shenyang, 110016, China
† Electronic supplementary information (ESI) available: MALDI-TOF
for DNAzyme oligodeoxynucleotides, melting data and example melting
curves for DNAzyme-substrate complexes, the cleavage pattern of 10-23
DNAzyme and DZ-3-9 in the presence of different Mg2+ concentrations
and different divalent metal ions, NMR spectra of compounds 3, 6, 10,
and 14. See DOI: 10.1039/c1ob05065f
5728 | Org. Biomol. Chem., 2011, 9, 5728–5736
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
each of them were incorporated into the five dA positions in
the catalytic loop, respectively. The modified 10–23 DNAzyme
analogs were evaluated against a DNA-RNA-DNA chimeric
substrate, 5¢-d(AGG TGC AGG)-rA-rU-d(GGA GAG CA)-3¢
(Fig. 1),29 in which the flanking DNA residues were used to avoid
ribonuclease cleavage.30 This study would help us to understand
the importance of the N7 atoms for the catalytic reaction, and
explore the possibility of finding more efficient deoxyribozymes by
introducing extra protein-like functional groups into the catalytic
loop of 10–23 DNAzyme.
coupling reaction of 7-iodo-8-aza-7-deaza-2¢-deoxyadenosine
with 3-trifluoroacetamidopropyne offered compound 6. It was
hydrogenated with 10% Pd/C to obtain compound 7, which
was deprotected in aq. ammonia to offer the nucleoside ana-
log 3. Compound 5 was obtained from the cross-coupling
reaction of 7-iodo-8-aza-7-deaza-2¢-deoxyadenosine with pheny-
35
lacetylene and subsequent hydrogenation (Scheme 2). The di-
n-butylaminomethylidene group was introduced to protect the 6-
amino group of compounds 7 and 5 to obtain compounds 8 and
12, respectively. These two compounds were then tritylated with
DMT-Cl (to give compounds 9 and 13) and further converted into
phosphoramidites 10 and 14, respectively. All new compounds
were fully characterized by elemental analysis or HRMS, 1H
NMR, 13C NMR, and 31P NMR. The coupling yields of all of
these phosphoramidites (0.15–0.2 M in acetonitrile) in the solid-
phase synthesis of modified deoxyribozymes were above 95% with
coupling time of 2 min. All of the DNA sequences were synthesized
in DMT-off mode. After deprotection with conc. aq. ammonia
(60 ◦C, 18 h), the sequences were purified with denaturing PAGE,
desalted with a SEP-PAK column. MALDI-TOF MS was used
for their characterization (ESI, Table S1†).
Fig. 1 Secondary structure of 10–23 DNAzyme and the DNA-RNA-
DNA substrate complex. Bold letters represent the RNA residues. The
arrow denotes the cleavage site on the substrate.
Cleavage reaction of modified 10-23 DNAzyme analogs
Results and discussion
In the catalytic loop of 10–23 DNAzyme, there are five 2¢-
deoxyadenosine residues, assigned as A5, A9, A11, A12, and A15,
with some residues contributing more than the others.24,25 Single
substitutions with 1–5 were conducted at these five positions. All
of the modified 10–23 DNAzyme analogs were shown in Table 1
and Table 2.
Chemistry
Nucleosides 1, 2, and 4 and their corresponding phospho-
ramidites were prepared according to the literature.31–34 Com-
pound 3 was prepared as described in Scheme 1. The cross-
Scheme 1 Synthesis of nucleoside analog 3 and phosphoramidite 10.
The Royal Society of Chemistry 2011 Org. Biomol. Chem., 2011, 9, 5728–5736 | 5729
This journal is
©

View Article Online
Table 1 Modified deoxyribozymes based on 10–23 DNAzyme with a single substitution in the catalytic loop with 2¢-deoxyadenosine analogs 1 and 2
and their kobs measured under single-turnover conditions
Name
Modified 10–23 DNAzyme analogs
kobs (min-1)
10–23 DNAzyme
DZ-1–5
DZ-1–9
DZ-1–11
DZ-1–12
DZ-1–15
DZ-2–5
5¢-d(tgc tct cca GGC TAG CTA CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC T1G CTA CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CT1 CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA C1A CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CA1 CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CAA CG1 cct gca cct)-3¢
5¢-d(tgc tct cca GGC T2G CTA CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CT2 CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA C2A CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CA2 CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CAA CG2 cct gca cct)-3¢
0.0037 0.0007
nda
0.0018 0.0001
nda
0.0022 0.0003
0.0027 0.0003
0.0022 0.0002
0.0089 0.0008
0.0020 0.0001
0.0032 0.0003
0.0008 0.00006
DZ-2–9
DZ-2–11
DZ-2–12
DZ-2–15
a For reactions with no cleavage or an observed rate constant of < 0.0001 min-1, the kobs is denoted as nd.
Table 2 Modified deoxyribozymes based on 10–23 DNAzyme with a single substitution in the catalytic loop with 2¢-deoxyadenosine analogs 3, 4, or 5
and their kobs measured under single-turnover conditions
Name
Modified 10–23 DNAzyme analogs
kobs (min-1)
DZ-3–5
DZ-3–9
DZ-3–11
DZ-3–12
DZ-3–15
DZ-4–5
DZ-4–9
DZ-4–11
DZ-4–12
DZ-4–15
DZ-5–9
5¢-d(tgc tct cca GGC T3G CTA CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CT3 CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA C3A CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CA3 CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CAA CG3 cct gca cct)-3¢
5¢-d(tgc tct cca GGC T4G CTA CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CT4 CAA CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA C4A CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CA4 CGA cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CTA CAA CG4 cct gca cct)-3¢
5¢-d(tgc tct cca GGC TAG CT5 CAA CGA cct gca cct)-3¢
0.0003 0.00001
0.045 0.004
0.0015 0.0001
0.0027 0.0006
0.0022 0.0002
0.0003 0.00006
0.025 0.006
0.0005 0.00002
0.0028 0.0002
0.0003 0.00006
0.0128 0.0018
Scheme 2 Synthesis of nucleoside analog 5 and phosphoramidite 14.
The Tm values of all of the deoxyribozyme-substrate complexes
rate constant kobs, and kobs reflected the rate of the chemical
with differently positioned modifications were around 51 ◦C in
a buffer for cleavage reaction (50 mM Tris-HCl, 2 mM Mg2+,
pH 7.4). Full DNA substrate sequence D19, 5¢-d (AGGTGC
AGG ATG GAG AGC A)-3¢ was used to avoid cleavage reaction
during the measurement. It was assumed that all of the substrate
was associated with the enzyme immediately under the reaction
conditions (37 ◦C), and the modifications in the catalytic loop
did not change DNAzyme-substrate complex stability. Single-
turnover conditions were used for the measurements of observed
cleavage step.36,37
The effect of N7 atoms of five 2¢-deoxyadenosines in the catalytic
loop of 10–23 DNAzyme screened with 2¢-deoxyadenosine analog 1
Substitution of dA with 7-deaza-2¢-deoxyadenosine (1) in the
catalytic loop of 10–23 DNAzyme produced five modified de-
oxyribozymes, each containing 1 instead of dA in a different
position in the catalytic loop (Table 1). As shown in Fig. 2A,
5730 | Org. Biomol. Chem., 2011, 9, 5728–5736
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
atom in 2¢-deoxyadenosine was more favorable than N8 atom of
compound 1. But for A15 located next to the cleavage site, it seems
that the new interactions related to N8 produced detrimental effect
on the catalytic reaction, a significant loss of activity of DZ-2–15
(kobs = 0.0008 min-1) (Table 1) was observed. Secondly, the N8
atom in A12 position (DZ-2–12) almost equally contributed to
the catalytic reaction as the N7 atom. The kobs of DZ-2–12 was
0.0032 min-1 under present single-turnover conditions, very close
to that of 10-23 DNAzyme. Finally, 2.4-fold faster cleavage was
observed with DZ-2–9 (kobs = 0.0089 min-1), indicating that the
N8 atom at A9 position was more appropriately located for its
positive role.
These primary screening studies with nucleoside analogs 1 and
2 demonstrated that the N7 atom of each 2¢-deoxyadenosine in
the loop is necessary for activity. At A9, its contribution could be
improved by shifting to the 8-position with 2. More importantly,
we learned that the catalytic potential of 10–23 DNAzyme could be
further optimized by chemical modification. We hypothesized that
A9 could be used as a potential position for further modification
and nucleoside analog 2 as the lead compound in the search for
more efficient 10–23 DNAzyme analogs.
Further modification of A9 with compounds 3–5 to introduce pro-
tein-like functional groups in the catalytic loop of 10–23 DNAzyme
Based on the small improvement on the cleavage rate of DZ-2–
9, it was recognized as the lead structure for further chemical
modifications with protein-like functional groups. Its residue
(compound 2) at A9 was substituted with compound 3 or 4, to
produce two novel modified DNAzymes, DZ-3–9 and DZ-4–9,
respectively. Thus, an amino or a hydroxyl group was introduced
to A9 of the catalytic loop. Amino and hydroxyl groups were
selected because they are well-known for hydrogen bonding and
proton transferring in catalytic reactions of protein enzymes. On
the other hand, the amino group is also known to offer electrostatic
stabilization of the transition state in the catalytic reaction of
ribozymes 28,38,39 and Ribonuclease A.40 As shown in Table 2 and
Fig. 3A, DZ-3–9 exhibited faster cleavage than DZ-2–9. Under
single-turnover conditions, kobs for DZ-3–9 was 0.045 min-1, which
was about 5 times faster than DZ-2–9, and 12 times faster than
10–23 DNAzyme. Interestingly, when such substitution was used
at the other four dA positions (A5, A11, A12, and A15), slower
cleavage than 10–23 DNAzyme was always observed. In other
words, the effect of compound 3 on the cleavage rate of the
modified deoxyribozymes was position-dependent.
The same tendency was observed with compound 4 (see Table
2 and Fig. 3B), the modified DNAzyme DZ-4–9 exhibited about
6.7 times faster cleavage than the parent 10–23 DNAzyme and
2.8 times faster than DZ-2–9. The replacement of compound 4
at the other four positions in the loop always resulted in a slower
cleavage than 10–23 DNAzyme.
The positive effect of compounds 3 and 4 at A9 position in
DZ-3–9 and DZ-4–9 suggested that A9 and 7-substituted 8-aza-7-
deaza-2¢-deoxyadenosine could be combined as a novel approach
for more efficient 10–23 DNAzyme analogs, DZ-2–9 could be
recognized as the new leading DNAzyme structure.
Fig. 2 The effect of the N7 atom of dA at five positions in the catalytic
loop of 10–23 DNAzyme was demonstrated with compounds 1 (A) and
2 (B) under single-turnover conditions. Solid triangles (A5), solid squares
(A9), open diamonds (A11), open squares (A12), open triangles (A15), and
solid diamonds (10–23 DNAzyme) were used to indicate the substitutions
at different positions.
under single-turnover conditions, almost complete loss of cleavage
activity was observed for DZ-1–5, indicating the most detrimental
effect of compound 1 at A5, compared with 10-23 DNAzyme. The
significant negative effect at A11 (DZ-1–11) was also observed.
This meant that the N7 atom was very important at these two
positions. In DZ-1–9, DZ-1–12, and DZ-1–15, the deletion of
N7 atom resulted in only moderate loss of activity, as shown by
kobs in Table 1. These results suggested that the N7 atom of each
2¢-deoxyadenosine residue contributes to the catalytic activity of
10–23 DNAzyme differently.
The influence of the shift of N7 to 8-position in 2¢-deoxyadenosine
on the catalytic reaction of DNAzymes
Based on above N7 atom deletion research, the interactions related
to the five N7 atoms of five 2¢-deoxyadenosine residues seem to be
favorable for the cleavage reaction of 10–23 DNAzyme. In order
to find if more favorable interactions could be produced by this
nitrogen atom, a micro-perturbation strategy at the N7 atom was
realized with compound 2 in which the N7 atom was shifted to
the 8-position without significantly altering the whole adenine
structure. As shown in Table 1 and Fig. 2B, the N8 atom had a
distinct position-dependent effect on the catalytic activity. Firstly,
the N8 atom at A5 (DZ-2–5) and A11 (DZ-2–11) compensated the
loss by the absence of N7 atom, to some extent, as evidenced by
their kobs. It implied that either N7 or N8 atom was necessary
for the catalytic reaction at these two positions, however, N7
This approach for more efficient 10–23 DNAzyme analog was
further testified with DNAzyme DZ-5–9, in which compound
5 was introduced to replace A9 in the catalytic loop (Table 2),
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5728–5736 | 5731
©

View Article Online
on the cleavage rates of the modified DNAzyme DZ-3–9 was
evaluated, and a comparison with the parent 10–23 DNAzyme
was made.
Under single-turnover conditions, it was observed that the
cleavage rates of DZ-3–9 and 10–23 DNAzyme were dependent
on the concentration of Mg2+ in the same way, at least from
0.1 mM to 50 mM used in our experiments, and no cleavage
was observed when EDTA was used as a scavenger of divalent
metal ions (ESI, Fig. S2†). In addition, the effect of two other
divalent metal ions, Mn2+ and Ca2+, was also examined for both
DNAzymes. As shown in Fig. 4, the effect of these three metal ions
are in the same order, Mn2+ > Ca2+ ª Mg2+ for both DZ-3–9 and
10–23 DNAzyme, and DZ-3–9 always reacted faster than 10–23
DNAzyme with each divalent metal ion. The presence of Mn2+ led
to the fastest cleavage.2 These results indicated that the divalent
metal ion plays a similar role in the reactions conducted by both
modified DNAzyme DZ-3–9 and the parent DNAzyme.
Fig. 3 The effect of 7-substituents of compound 2 at five positions in the
catalytic loop of 10–23 DNAzyme was demonstrated with compounds 3
(A) and 4 (B) under single-turnover conditions. Solid triangles (A5), solid
squares (A9), open diamonds (A11), open squares (A12), open triangles
(A15), and solid diamonds (10–23 DNAzyme) were used to indicate the
substitutions at different positions.
The kobs of DZ-5–9 was 0.0128 min-1, which was slower than
DZ-3–9 and DZ-4–9, but still about 3.5 times faster than 10–
23 DNAzyme and 1.4 times faster than DZ-2–9. It is notable
that a phenyl group at A9 was also favorable for the cleavage
reaction, although it could not participate in hydrogen-bonding as
an amino or hydroxyl group. The new hydrophobic interaction and
spacial occupation related to the phenyl group probably resulted
in a favorable conformational change. The different contributions
of compounds 1–5 at A9 demonstrated the importance of 7-
substituents, more efficient version of 10–23 DNAzyme could be
expected by optimizing the design of 7-substituents of 8-aza-7-
deaza-2¢-deoxyadenosine, hydrogen-bonding ability might be one
of the important properties to be considered in the design of the
7-substituents.
Fig. 4 The effect of divalent metal ions on the cleavage reactions of
10–23-DNAzyme (A) and DZ-3–9 (B). Mn2+ (solid triangle), Ca2+ (open
square), and Mg2+ (solid diamond) were used to indicate different metal
ions.
Preliminary mechanistic studies on the modified DNAzyme
DZ-3–9
The effect of pH on cleavage of DZ-3–9 was measured under
multiple-turnover conditions.1,2 With pH ranging from 7.5 to 8.5
for 10–23 DNAzyme (Fig. 5, A) and pH from 6.0 to 8.5 for DZ-
3–9 (Fig. 5, B), a roughly linear logkobs fashion against increasing
pH was observed for both DZ-3–9 and 10–23 DNAzyme,1,2 faster
cleavage was always observed under higher pH values. This pH-
dependence indicated that the deprotonation of the 2¢-hydroxyl
group was also part of the rate-limiting step of DZ-3–9, as reported
for 10–23 DNAzyme.1,2
10-23 DNAzyme has been characterized as a divalent metal ion-
dependent enzyme.1,2 The metal ion might act as a Lewis base
to strengthen the nucleophilicity of the 2¢-hydroxyl group or as
a Lewis acid to coordinate the 5¢-oxygen atom to facilitate the
cleavage of the 5¢-end fragment, as well as play a structural role
to stabilize and/or pre-organize the active catalytic conformation.
The log-linear relationship of catalytic rate versus pH indicated
that the deprotonation of 2¢-OH is embedded in the rate-limiting
step.1,2,38,41 Therefore, the influence of pH and divalent metal ions
The same effect of the two critical factors (divalent metal ion
and pH) indicated that 10–23 DNAzyme and DZ-3–9 probably
5732 | Org. Biomol. Chem., 2011, 9, 5728–5736
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Conclusion
Based on the chemical modifications with 2¢-deoxyadenosine
analogs 1–5 on the catalytic loop of 10–23 DNAzyme, the
importance and potential of the five-membered ring part of 2¢-
deoxyadenosine residues for catalytic activity were explored for
the first time. A9 is the only position for a positive effect of
these funtional groups: N8 atom, amino, hydroxyl, and phenyl
groups which were located in the five-membered ring part of 2¢-
deoxyadenosine. A9 position and 7-substituted 8-aza-7-deaza-2¢-
deoxyadenosine analogs were combined to offer a prosperous
approach for more efficient 10–23 DNAzyme analogs. More
efficient deoxyribozymes will undoubtedly offer a better choice
in deoxyribozyme applications, such as for genetic therapeutics.
Experimental section
General
Most of the commercially available chemicals were used without
further purification. Pyridine was dried by refluxing with CaH2.
Thin layer chromatography (TLC) was run on HS GF254 (Yantai
Institute of Chemical Industry, China). Silica gel (200–300 mesh,
Qingdaohaiyang Chemicals Co., China) was used for flash column
chromatography. 1H, 13C and 31P NMR were recorded on a JNM-
ECA-400 spectrometer (JEOL, Japan), internal trimethylsilane
(TMS) (1H, 13C) or external 85% H3PO4 (31P) was used as the
standard. The J values are given in Hz. Elemental analysis on a
Fisons-1108 (Fisons, Italy), and HR-MS on Q-FT-MS (Apex Qe,
Brucker) were performed by the National Center of Biomedical
Analysis (Beijing, China).
Fig. 5 pH-dependence of 10–23-DNAzyme (A) and DZ-3–9 (B) under
multiple-turnover conditions with different pH values ranging from 6 to
8.5.
conducted the cleavage reaction in the same way. We speculated
that the functional groups at A9 might be involved in the opti-
mization of catalytic conformation, and their different influence
(NH2 > OH > phenyl) might reflect their ability on driving
conformational changes, probably through forming new hydrogen
bonds or hydrophobic interactions in the catalytic loop.
Oligodeoxynucleotides
Normally protected phosphoramidites of canonical residues were
purchased from Proligo (Sigma–Aldrich). Oligodeoxynucleotide
synthesis was run on an ABI 392 DNA/RNA synthesizer (Applied
Biosystems, USA) on a 1 mmol scale with the DMT-off mode
according to the User Protocol. The DMT-off oligodeoxynu-
cleotides were deprotected in conc. aq. ammonia for 18 h at
60 ◦C. All were then purified by a 20% polyacrylamide/7 M
urea denaturing gel and extracted with 0.3 M sodium acetate.20
Desalting was conducted using SEP-PAK cartridges (Oasis MAX,
C18, Waters, USA) with repeated washing with sterilized doubly-
distilled water. The product was lyophilized and stored at -18 ◦C.
The chimera substrate was purchased from Takara (Dalian,
China). A Cary-100 Bio UV-Visible spectrophotometer equipped
with a Cary temperature controller (Varian, USA) was used for
Tm measurements.
The homogeneous oligodeoxynucleotides were characterized by
MALDI-TOF with 2¢,4¢,6¢-trihydroxyacetophenone (THAP) as
the matrix. MALDI-TOF MS were acquired on an AXIMA-CFP
plus mass spectrometer (KRATOS Analytical, Shimadzu Group
Company, Japan) equipped with a nitrogen laser (337.1 nm) in
positive ion and linear mode, with an acceleration voltage of
20 kV and averaged over 100 laser shots. Each spectrum was
externally calibrated with Cytochrome C (m/z 12361) providing
mass measurement accuracies of approximately 500 ppm across
the 4 kDa ~20 kDa mass range.
In the cleavage reaction of 10–23 DNAzyme, three steps are
thought to be mainly responsible for the whole cleavage profile:
association of its binding arms with the target sequence, the
chemical cleavage step, and the dissociation between the binding
arms and the cleaved products.27 The first and the third steps are
related to the length of the binding arms and their hybridization
affinity to the substrate. Therefore, the length of the binding arms
has to be balanced between the rate and specific recognition. It
has been reported that the incorporation of LNA (locked nucleic
acid) residues in the binding arms significantly increase the rate
by improving hybridization affinity to the substrate.24 The present
chemical modification on the catalytic loop of 10–23 DNAzyme
conferred a faster chemical cleavage step; therefore, the modified
DNAzymes are intrinsically faster than 10–23 DNAzyme. LNA
modification also might be applied to these modified DNAzymes
for a further increase in reaction rates.
Chemical modification was demonstrated again to be a powerful
tool for more efficient deoxyribozymes. Currently, much efforts
have been being devoted to explore the catalytic potential of
nucleic acid structures by chemical modifications with protein-
like functional groups,42–45 although it is not known whether the
maximum catalytic power of deoxyribozyme or ribozyme could
approach to that of nucleases.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5728–5736 | 5733
©

View Article Online
4-Amino-1-(2-deoxy-b-D-erythro-pentofuranosyl)-3-(3-trifluoro-
acetamido-propyn-1-yl)-1H-pyrazolo[3,4-d]pyrimidine (6). 7-
flash chromatography. The product was obtained as a colorless
solid (2.1 g, 82%). Rf (CH2Cl2/CH3OH 20 : 1) 0.35. H NMR
1
Iodo-8-aza-7-deaza-deoxyadenosine (0.8 g, 2.1 mmol) was
dissolved in dried DMF (30 ml). To the solution was added
Pd(Ph3)2Cl2 (0.14 g, 2.0 mmol), CuI (0.1 g, 0.6 mmol), the mixture
was stirred at r.t. under nitrogen atmosphere. Triethylamine
(2 ml, 10.5 mmol) was added to the reaction mixtrure, followed
by N-propynyltrifluoroacetamide (0.4 g, 2.5 mmol). The
reaction mixture was then stirred at r.t. overnight. DMF was
vacuum-evaporated off and the residue was subjected to flash
chromatography. The product was obtained as colorless solid
(0.74 g, 87.5%). Rf (CH2Cl2/CH3OH 9 : 1) 0.4. 1H NMR (DMSO-
d6): d 2.25 (m, 1 H, C2¢–Ha), 2.76 (m, 1 H, C2¢–Hb), 3.36, 3.51 (2 m,
2 H, C5¢–H), 3.82 (m, 1 H, C4¢–H), 4.42 (m, 3 H, C3¢–H, CH2), 4.75
(t, J = 5.6, 1 H, C5¢–OH), 5.28 (d, J = 4.5, 1 H, C3¢–OH), 6.55 (t, J =
6.3, 1 H, C1¢–H), 8.26 (s, 1 H, C2–H), 6.74, 8.17 (2 br, 2 H, NH2),
10.16 (s, 1 H, NH). 13C NMR (DMSO-d6): d 30.0, 38.9, 62.3, 70.9,
74.3, 84.0, 87.7, 90.2, 101.0, 126.1, 153.7, 156.7, 157.6. HRMS
Calcd for C15H16F3N6O4 (MH+): 401.1180, Found: 401.1181.
(DMSO-d6): d 0.92 [2 t, J = 7.4, 6 H, CHN(CH2CH2CH2CH3)2],
1.32 (m,
4 H, CHN(CH2CH2CH2CH3)2], 1.61 (m, 4 H,
CHN(CH2CH2CH2CH3)2], 2.00 (m, 2 H, 5-CH2CH2CH2), 2.24
(m, 1 H, C2¢–Ha), 2.82 (m, 1 H, C2¢–Hb), 3.03 (m, 2 H, 5-
CH2CH2CH2), 3.23–3.60 [m, 8 H, CHN(CH2CH2CH2CH3)2, C5¢–
H, 5-CH2CH2CH2], 3.82 (m, 1 H, C4¢–H), 4.45 (m, 1 H, C3¢–
H), 4.77 (t, J = 5.7, 1 H, C5¢–OH), 5.24 (d, J = 4.5, 1 H, C3¢–
OH), 6.55 (t, J = 6.4, 1 H, C1¢–H), 8.42 (s, 1 H, C2–H), 8.98
[s, 1 H, CHN(CH2CH2CH2CH3)2], 9.41 (m, 1 H, NHCO). 13C
NMR (DMSO-d6): d 13.5, 13.6, 19.1, 19.8, 25.7, 27.3, 28.7, 30.4,
37.9, 45.4, 51.4, 62.5, 71.2, 83.8, 87.5, 106.0, 111.6, 114.5, 117.4,
120.2, 146.5, 155.4, 157.2, 162.5. Anal. Calcd for C24H36F3N7O4
(M 543.58): C, 53.03; H, 6.68; N, 18.04. Found: C, 52.65; H, 6.49;
N, 17.87.
1-[2-Deoxy-5-O-(4,4¢-dimethoxytriphenylmethyl)-b-D-erythro-
pentofuranosyl]-4-{[(di-n-butylamino)methylidene]amino}-3-(3-
trifluoroacetamido-prop-1-yl)-1H-pyrazolo[3,4-d]pyrimidine (9).
Compound 8 (1.0 g, 1.85 mmol) was co-evaporated with
dried pyridine three times before it was dissolved in dried
pyridine (3 ml). 4,4¢-Dimethoxytrityl chloride (DMT-Cl)
(0.75 g, 2.2 mmol) was added to the solution in portions.
After stirring at r.t. for 1 h, methanol (3 ml) was added to
the reaction mixture. The solution was concentrated for flash
chromatography. The product was obtained as a colorless
solid (0.95 g, 62.5%). Rf (CH2Cl2/CH3OH 20 : 1) 0.7. 1H
NMR (DMSO-d6): d 0.93 [t, 6 H, CHN(CH2CH2CH2CH3)2],
4-Amino-1-(2-deoxy-b-D-erythro-pentofuranosyl)-3-(3-trifluoro-
acetamido-prop-1-yl)-1H-pyrazolo[3,4-d]pyrimidine (7). The sus-
pension of compound 6 (2 g, 5 mmol) and Pd/C (10%, 1.0 g) in
methanol (200 ml) was sealed in an oven under hydrogen (5 atm).
The mixture was stirred at 30 ◦C for 5 h. The catalyst was filtered
off, and the filtrate was concentrated to obtain a colorless solid
(1.9 g, 95%). Rf (CH2Cl2/CH3OH 10 : 1) 0.45. 1H NMR (DMSO-
d6): d 1.87 (m, 2 H, 5-CH2CH2CH2), 2.20 (m, 1 H, C2¢–Ha), 2.79
(m, 1 H, C2¢–Hb), 2.97 (m, 2 H, 5-CH2CH2CH2), 3.30 (m, 2 H,
5-CH2CH2CH2), 3.37, 3.52 (2 m, C5¢–H), 3.79 (m, 1 H, C4¢–H),
4.42 (m, 1 H, C3¢–H), 4.78 (m, 1 H, C5¢–OH), 5.22 (d, J = 4.8, 1
H, C3¢–OH), 6.50 (t, J = 6.6, 1 H, C1¢–H), 7.35 (br, 2 H, NH2),
8.16 (s, 1 H, C2–H), 9.41 (m, 1 H, NHCO). 13C NMR (DMSO-d6):
d 25.2, 27.3, 37.9, 62.5, 71.2, 83.7, 87.5, 98.8, 111.6, 114.5, 117.4,
120.2, 145.0, 154.8, 155.9, 158.1. Anal. Calcd for C15H19F3N6O4
0.25 H2O (M 408.85): C, 44.07; H, 4.81; N, 20.56. Found: C, 44.13;
H, 4.82; N, 20.06.
1.33 (m,
4
H, CHN(CH2CH2CH2CH3)2), 1.62 (m,
4 H,
CHN(CH2CH2CH2CH3)2), 1.77 (m,
2
H, 5-CH2CH2CH2),
2.30 (m, 1 H, C2¢–Ha), 2.78 (m, 1 H, C2¢–Hb), 2.89 (m, 2 H,
5-CH2CH2CH2), 3.03–3.13 [m, 4 H, CHN(CH2CH2CH2CH3)2],
3.50–3.60 (m, C5¢–H, 5-CH2CH2CH2], 3.67, 3.69 (2 s, 2 OCH3),
3.95 (m, 1 H, C4¢–H), 4.52 (m, 1 H, C3¢–H), 5.27 (m, 1 H,
C3¢–OH), 6.60 (m, 1 H, C1¢–H), 6.73–7.30 (m, arom. H), 8.44 (s,
1 H, C2–H), 9.00 [s, 1 H, CHN(CH2CH2CH2CH3)2], 9.37 (m, 1
H, NHCO). 13C NMR (DMSO-d6): d 13.5, 13.6, 19.1, 19.8, 25.7,
27.2, 28.6, 30.4, 38.2, 45.4, 51.4, 54.8, 54.8, 64.9, 71.3, 83.4, 85.1,
85.4, 106.0, 112.8, 112.9, 114.5, 117.4, 126.3, 127.5, 127.6, 129.5,
129.6, 135.6, 145.0, 146.5, 155.4, 157.1, 157.8, 157.9, 162.4. Anal.
Calcd for C45H54F3N7O6 0.25 H2O (M 850.45): C, 63.55; H, 6.46;
N, 11.53. Found: C, 63.32; H, 6.20; N, 11.40.
4-Amino-3-(3-amino-prop-1-yl)-1-(2-deoxy-b-D-erythro-pento-
furanosyl)-1H-pyrazolo[3,4-d]pyrimidine (3). The suspension of
compound 7 (100 mg, 0.25 mmol) in aq. ammonia (10 ml)
was sealed in a bottle. After stirring at r.t. for 4 h, the
solution was concentrated and the product was crystallized
from the residue as a colorless solid (70 mg, 93.3%). Rf
(CH2Cl2/CH3OH/NH3·H2O, 1 : 1 : 0.02) 0.26. 1H NMR (DMSO-
d6): d 1.76 (m, 2 H, 5-CH2CH2CH2), 2.21 (m, 1 H, C2¢–Ha), 2.67
(m, 2 H, 5-CH2CH2CH2), 2.78 (m, 1 H, C2¢–Hb), 2.99 (m, 2 H,
5-CH2CH2CH2), 3.38, 3.54 (2 m, 1 H, C5¢–H), 3.80 (m, 1 H, C4¢–
H), 4.43 (m, 1 H, C3¢–H), 4.78 (m, 1 H, C5¢–OH), 5.22 (d, J =
4.8, 1 H, C3¢–OH), 6.50 (t, J = 6.6, 1 H, C1¢–H), 7.35 (br, 2 H,
NH2), 8.15 (s, 1 H, C2–H). 13C NMR (DMSO-d6): d 24.8, 30.6,
37.9, 62.5, 71.2, 83.6, 87.5, 98.8, 145.5, 154.7, 155.8, 158.2. HRMS
Calcd for C13H21N6O3 (MH+): 309.1670. Found: 309.1671.
1-[2-Deoxy-5-O-(4,4¢-dimethoxytriphenylmethyl)-b-D-erythro-
pentofuranosyl]-4-{[(di-n-butylamino)methylidene]amino}-3-(3-tri-
fluoroacetamido-prop-1-yl)-1H-pyrazolo[3,4-d]pyrimidine 3¢-[(2-
Cyanoethyl) N,N-diisopropylphosphoramidite] (10). To the
solution of compound 9 (0.45 g, 0.55 mmol) in dichloromethane
(5 ml), diisopropylamine tetrazolide (0.17 g, 1.1 mmol) and
2-cyanoethyl
N,N,N¢,N¢-tetraisopropylphosphorodiamidite
(0.55 g, 1.1 mmol) were added. The reaction mixture was
stirred at rt for 30 min. It was washed with 5% aq. NaHCO3
and subsequently brine, and dried with anhydrous Na2SO4.
The solution was concentrated for flash chromatography
to afford the product as colorless foam (0.45 g, 80.5%). Rf
(CH2Cl2/CH3OH 40 : 1) 0.48, 0.50. 1H NMR (DMSO-d6): d
0.90–1.81 [m, 28 H, CHN(CH2CH2CH2CH3)2, 5-CH2CH2CH2,
2 CH(CH3)2], 2.47–3.18 [m, 10 H, C2¢–H, 5-CH2CH2CH2),
CHN(CH2CH2CH2CH3)2, CH2CH2CN], 3.35–3.77 [m, 14 H,
1-(2-Deoxy-b-D-erythro-pentofuranosyl)-4-{[(di-n-butylamino)-
methylidene]amino}-3-(3-trifluoroacetamido-prop-1-yl)-1H-pyr-
azolo[3,4-d]pyrimidine (8). To the solution of compound 7
(1.9 g, 4.7 mmol) in methanol (40 ml) was added N,N-di-n-
butylformamide dimethyl acetal (1.75 ml, 7 mmol). The solution
was stirred at 40 ◦C for 4 h. Then, it was concentrated for
5734 | Org. Biomol. Chem., 2011, 9, 5728–5736
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
C5¢–H, 5-CH2CH2CH2, 2 OCH3, 2 CH(CH3)2, CH2CH2CN],
4.10 (m, 1 H, C4¢–H), 4.78 (m, 1 H, C3¢–H), 6.61–6.75, 7.14,
7.28 (3 m, 14 H, C1¢–H, arom. H), 8.47 (2 s, 1 H, C2–H), 9.01
[s, 1 H, CHN(CH2CH2CH2CH3)2], 9.40 (m, 1 H, NHCO). 31P
NMR (CDCl3): 147.16, 147.83. 13C NMR (CDCl3): d 13.6, 13.8,
19.0, 19.7, 20.0, 20.3, 22.8, 22.9, 24.4, 24.5, 25.6, 26.9, 29.1, 31.0,
37.5, 39.2, 43.0, 43.1, 45.2, 46.0, 46.5, 52.4, 55.1, 58.0, 58.2, 64.4,
84.1, 85.0, 85.2, 85.3, 85.8, 107.0, 112.8, 116.9, 117.5, 126.4,
127.5, 128.1, 128.2, 130.0, 136.0, 144.9, 146.9, 155.8, 156.0, 156.7,
157.0, 158.2, 162.9. HRMS Calcd for C54H72F3N9O7P (MH+):
1046.5239. Found: 1046.5253.
4 H, CHN(CH2CH2CH2CH3)2], 3.63 (2 s, 6 H, 2 OCH3), 3.98 (m,
1 H, C4¢–H), 4.54 (m, 1 H, C3¢–H), 4.81 (t, J = 5.5, 1 H, C5¢–OH),
5.31 (d, J = 4.8, 1 H, C3¢–OH), 6.63 ~7.18 (m, 19 H, arom. H, C1¢–
H), 8.47 (s, 1 H, C2–H), 9.01 [s, 1 H, CHN(CH2CH2CH2CH3)2].
13C NMR (DMSO-d6): d 13.6, 19.2, 19.6, 28.6, 29.8, 30.4, 34.0,
38.4, 45.3, 51.4, 54.8, 54.9, 65.1, 71.4, 83.4, 85.2, 85.6, 106.0,
112.9, 125.8, 126.4, 127.6, 127.7, 127.9, 128.2, 129.5, 129.7, 135.7,
141.5, 145.1, 146.9, 155.4, 157.2, 157.8, 157.9. Anal. Calcd for
C48H56N6O5 (M 797.00): C 72.34, H 7.08, N 10.54. Found C 72.04,
H 7.08, N 10.27.
1-[2-Deoxy-5-O-(4,4¢-dimethoxytriphenylmethyl)-b-D-erythro-
pentofuranosyl)-4-{[(di-n-butylamino)methylidene]amino}-3-(2-
phenylethyl)-1H -pyrazolo[3,4-d]pyrimidine3¢-[(2-cyanoethyl) N,
N-diisopropylphosphoramidite] (14). Using the same procedure
as for compound 10, compound 13 (0.25 g, 0.31 mmol) was treated
with 2-cyanoethyl N,N-di-isopropylchlorophosphoramidite
(0.1 g, 0.42 mmol) in dried dichloromethane (2 ml) in the presence
of N,N-diisopropylethylamine (DIEA) (1 ml, 5.75 mmol). The
product was obtained after flash chromatography as a colorless
solid (120 mg, 38.3%). Rf (CH2Cl2/CH3OH 30 : 1): 0.61, 0.67. 1H
NMR (CDCl3): d 0.80, 0.95 [2 t, 6 H, CHN(CH2CH2CH2CH3)2],
1.07–1.72 [m, 20 H, 2 CH(CH3)2, CHN(CH2CH2CH2CH3)2],
2.40 ~3.10 (m, 6 H, C2¢–H, CH2CH2CN, 5-CH2CH2), 3.24–
3.82 [m, 16 H, 2 OCH3, C5¢–H, CH2CH2CN, 5-CH2CH2,
CHN(CH2CH2CH2CH3)2], 4.23 (m, 1 H, C4¢–H), 4.83 (m, 1
H, C3¢–H), 6.66–7.28 (m, 19 H, aromatic H, C1¢–H), 8.49 (s, 1
H, C2–H), 8.88 [s, 1 H, CHN(CH2CH2CH2CH3)2]. 31P NMR
(CDCl3): 148.75, 148.57. 13C NMR (CDCl3): d 13.7, 13.6, 19.8,
20.1, 20.3, 24.4, 24.5, 24.6, 37.6, 43.0, 43.1, 43.2, 74.0, 74.2, 74.5,
74.7, 84.1, 85.0, 85.2, 85.9, 125.6, 126.4, 126.5, 127.5, 128.1,
128.2, 130.0, 136.2, 142.0, 144.9, 147.7, 155.5, 156.0, 156.6, 158.2,
162.9, HRMS Calcd for C57H74N8O6P (MH+): 997.5464. Found:
997.5478.
1-(2-Deoxy-b-D-erythro-pentofuranosyl)-3-(2-phenylethyl)-1H-
pyrazolo[3,4-d]pyrimidin-4-amine (5). Compound 11 (1.8 g,
5.12 mmol) and 10% Pd/C (0.66 g) in methanol (50 ml) was
sealed under hydrogen (5 kg) in an oven. The mixture was stirred
at 30 ◦C for 6 h. The catalyst was filtered off, and the filtrate
was concentrated to obtain a colorless solid (1.7 g, 93.4%). Rf
1
(CH2Cl2/CH3OH 9 : 1) 0.38. H NMR (DMSO-d6): d 2.19 (m, 1
H, C2¢–Ha), 2.74 (m, 1 H, C2¢–Hb), 3.01 (m, 2 H, 5-CH2CH2), 3.31
(2 m, 4 H, C5¢–H, 5-CH2CH2), 3.80 (m, 1 H, C4¢–H), 4.43 (d, J =
4.2, 1 H, C3¢–H), 4.79 (t, J = 5.7, 1 H, C5¢–OH), 5.22 (d, J = 4.2,
1 H, C3¢–OH), 6.50 (t, J = 6.5, 1 H, C1¢–H), 7.27 (m, 5 H, arom.
H), 8.2 (s, 1 H, C2–H). 13C NMR (DMSO-d6): d 29.4, 33.5, 38.0,
62.6, 71.3, 83.6, 87.6, 98.9, 125.9, 128.2, 128.6, 141.1, 145.1, 154.8,
155.8, 158.2. Anal. Calcd for C48H52N6O5 (M 355.39): C 60.83, H
5.96, N 19.71. Found C 60.56, H 5.99, N 19.22.
1-(2-Deoxy-b-D-erythro-pentofuranosyl)-4-{[(di-n-butylamino)-
methylidene]amino}-3-(2-phenylethyl)-1H-pyrazolo[3,4-d]pyrimi-
dine (12). Using the same procedure as for 8, compound 5 (1.0 g,
2.85 mmol) was treated with N-di-n-butylformamide dimethyl
acetal (0.58 g, 2.85 mmol) in methanol (4 ml) at 40 ◦C. After
purification by flash chromatography, an oily product was ob-
tained (1.13 g, 94.0%). Rf (CH2Cl2/CH3OH 15 : 1) 0.57. 1H NMR
(DMSO-d6): d 0.67, 0.91 [2 t, 6 H, CHN(CH2CH2CH2CH3)2],
1.16, 1.29 [2 m, 4 H, CHN(CH2CH2CH2CH3)2], 1.56 [m, 4 H,
CHN(CH2CH2CH2CH3)2], 2.22 (m, 1 H, C2¢–Ha), 2.78 (m, 1 H,
C2¢–Hb), 3.09 (m, 2 H, 5-CH2CH2), 3.33, 3.48 [2 m, 8 H, C5¢–H,
CHN(CH2CH2CH2CH3)2, 5-CH2CH2], 3.80 (m, 1 H, C4¢–H), 4.43
(m, 1 H, C3¢–H), 4.81 (t, J = 5.5, 1 H, C5¢–OH), 5.25 (d, J = 4.5, 1
H, C3¢–OH), 6.54 (t, J = 6.4, 1 H, C1¢–H), 7.24 (m, 5 H, arom. H),
8.42 (s, 1 H, C2–H), 8.98 [s, 1 H, CHN(CH2CH2CH2CH3)2]. 13C
NMR (DMSO-d6): d 30.1, 34.7, 34.8, 37.9, 40.8, 62.5, 71.2, 83.7,
87.5, 106.6, 125.9, 128.2, 128.3, 141.4, 146.8, 155.4, 157.6, 162.5.
Anal. Calcd for C27H38N6O3·0.5H2O (M 499.13): C 64.91, H 7.71,
N 16.83. Found C 65.08, H 7.97, N 16.33.
Tm measurement
An equimolar mixture of deoxyribozyme (1.3 mM) and full DNA
substrate sequence D19, 5¢-d(AGG TGC AGG ATG GAG AGC
A)-3¢ (1.3 mM) in a buffer (1 ml) was used for Tm measurements.
The buffer consisted of 50 mM Tris-HCl (pH 7.5) and 2 mM
MgCl2. The solution was heated to 85 ◦C. After halting for 10 min,
it was cooled to 10 ◦C at a rate of 0.5 ◦C/min. The UV absorbance
was recorded at 260 nm during the cooling process. The Tm values
were obtained from the melting curves, and each melting curve
was fit to a non-self-complementary two-state model.
Radio-labeling of the chimeric substrate
1-[2-Deoxy-5-O-(4,4¢-dimethoxytriphenylmethyl)-b-D-erythro-
pentofuranosyl)-4-{[(di-n-butylamino)methylidene]amino}-3-(2-
phenylethyl)-1H-pyrazolo[3,4-d]pyrimidine (13). Using the same
procedure as for compound 9, compound 12 (0.87 g,
2.07 mmol) was tritylated with DMT-Cl (0.78 g, 2.28 mmol)
in dried pyridine (2 ml). The colorless product was ob-
tained after purification by flash chromatography (1.1 g,
66.7%). Rf (CH2Cl2/CH3OH 20 : 1) 0.45. 1H NMR (DMSO-
d6): d 0.74, 0.91 [2 t, 6 H, CHN(CH2CH2CH2CH3)2], 1.10,
1.30 [2 m, 4 H, CHN(CH2CH2CH2CH3)2], 1.56 (m, 4 H,
CHN(CH2CH2CH2CH3)2], 2.33 (m, 1 H, C2¢–Ha), 2.75 (m, 3 H,
C2¢–Hb, 5-CH2CH2), 3.13 (m, 4 H, C5¢–H, 5-CH2CH2), 3.48 [m,
The substrate 5¢-d(AGG TGC AGG)-rA-rU-d(GGA GAG CA)-
3¢ (11 nmol) was 5¢-32P-labeled with 50 mCi of [g-32P] ATP and
10 U of T4 polynucleotide kinase (Takara, Dalian, China) at
37 ◦C for 60 min. After deactivation at 70 ◦C for 5 min, the
radioactive substrate was extracted using a SEP-PAK column.
After washing with sterilized water several times, it was eluted with
methanol/water (70/30, v/v), lyophilized, and stored at -30 ◦C.
The cleavage reaction under single-turnover conditions
The reaction between deoxyribozyme (2 mM) and the substrate
(20 nM) was conducted in a buffer of 50 mM Tris-HCl (pH 7.4)
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 5728–5736 | 5735
©

View Article Online
containing 2 mM MgCl2.1 The mixture of deoxyribozyme and the
substrate was incubated in 50 mM Tris-HCl (pH 7.4) at 90 ◦C
for 3 min. After cooling to 37 ◦C, an equal volume of 50 mM
Tris-HCl (pH 7.4) containing Mg2+ (4 mM) was added to initiate
the reaction. The reaction was monitored by taking aliquots from
the reaction mixture at different time points that were quenched
by an equal volume of stopping solution (100 mM EDTA, 8 M
Urea). The cleavage process was analyzed by electrophoresis on a
20% polyacrylamide gel containing 7 M urea, and quantified by
the densitometry of the gel images with a Molecular Dynamics
Storm 840 Phosphoimager.14 The equation P% = P•% - C·exp
[-kobst] was used to calculate the observed rate constant, where
P% is the cleavage percentage of product at time t, P•% is the
final percentage of the product at t = •, C is the difference in
P% between t = • and t = 0, and k is the observed rate constant.
The data was given as the averaged results of three independent
experiments. Less than 20% variation was observed for identical
experiments performed on different days.
8 E. A. Nunamaker, H.-Y. Zhang, Y. Shirasawa, J. N. Benoit and D.
A. Dean, Am. J. Physiol. Heart Circ. Physiol., 2003, 285, H2240–
H2247.
9 A. Hurtado, H. Podinin, M. Oudega and B. Grimpe, Brain, 2008, 131,
2596–2605.
10 A. Mitchell, C. R. Dass, L. Q. Sun and L. M. Khachigian, Nucleic
Acids Res., 2004, 32, 3065–3069.
11 J. B. Trepanier, J. E. Tanner and C. Alfieri, Virology, 2008, 377, 339–344.
12 S. K. Silverman, Nucleic Acids Res., 2005, 33, 6151–6163.
13 B. Vester, L. B. Lundberg, M. D. Sørensen, B. R. Babu, S. Douthwaite
and J. Wengel, Biochem. Soc. Trans., 2004, 32, 37–40.
14 B. Nawrot, K. Widera, M. Sobczak, M. Wojcik and W. J. Stec, Curr.
Org. Chem., 2008, 12, 1004–1009.
15 V. M. Jadhav, V. Scaria and S. Maiti, Angew. Chem., Int. Ed., 2009, 48,
2557–2560.
16 H. Asanuma, H. Hayashi, J. Zhao, X. Liang, A. Yamazawa, T.
Kuramochi, D. Matsunaga, Y. Aiba, H. Kashida and M. Komiyama,
Chem. Commun., 2006, 5062–5064.
17 D. D. Young, M. O. Lively and A. Deiters, J. Am. Chem. Soc., 2010,
132, 6183–6193.
18 B. Wang, L. Q. Cao, W. Chiuman, Y. F. Li and Z. Xi, Biochemistry,
2010, 49, 7553–7562.
19 T. Kubo, K. Takamori, K. Kanno, B. Rumiana, H. Ohba, M.
Matsukisono, Y. Akebiyama and M. Fujii, Bioorg. Med. Chem. Lett.,
2005, 15, 167–170.
20 D. Smuga, K. Majchrzak, E. Sochacka and B. Nawrot, New J. Chem.,
2010, 34, 934–948.
Analysis of pH dependence under multiple-turnover conditions
The reactions were conducted in different buffers according to
the required pH ranging from 6.0 to 8.5. Multiple-turnover
conditions were employed.2 A mixture of the substrate (200 nM)
and deoxyribozyme (20 nM) was incubated in 50 mM Tris-HCl
buffer containing 2 mM Mg2+. Aliquots were withdrawn after
various time intervals for analysis as described above.
21 H. Lusic, D. D. Young, M. O. Lively and A. Deiters, Org. Lett., 2007,
9, 1903–1906.
22 S. Schubert, D. C. Gu¨l, H. P. Grunert, H. Zeichhardt, V. A. Erdmann
and J. Kurreck, Nucleic Acids Res., 2003, 31, 5982–5992.
23 B. Vester, L. B. Lundberg, M. D. Sørensen, B. R. Babu, S. Douthwaite
and J. Wengel, J. Am. Chem. Soc., 2002, 124, 13682–13683.
˙
24 Z. Zaborowska, J. P. Fu¨rste, V. A. Erdmann and J. Kurreck, J. Biol.
Chem., 2002, 277, 40617–40622.
˙
25 Z. Zaborowska, S. Schubert, J. Kurreck and V. A. Erdmann, FEBS
Analysis of divalent metal ion dependence under single-turnover
conditions
Lett., 2005, 579, 554–558.
26 B. Nawrot, K. Widera, M. Wojcik, B. Rebowska, G. Nowak and W. J.
Stec, FEBS J., 2007, 274, 1062–1072.
The cleavage rate of deoxyribozymes in the presence of either
Mg2+, Mn2+, or Ca2+ (2 mM) were measured under single-turnover
conditions in a buffer of 50 mM Tris-HCl (pH 7.5).2 A mixture
of the substrate (20 nM) and deoxyribozyme (2 mM) in a buffer
27 G. M. Blackburn, in Nucleic Acids in Chemistry and Biology, 2nd Ed.,
ed. G. M. Blackburn and M. J. Gait, Oxford University Press, Oxford,
1996, pp. 56–58.
28 C. Macelrevey, J. D. Salter, J. Krucinska and J. Wedekind, RNA, 2008,
14, 1600–1616.
◦
containing different divalent metal ions was incubated at 37 C.
29 L. Zhang, W. J. Gasper, S. A. Stass, O. B. Ioffe, M. A. Davis and A. J.
Mixson, Cancer Res., 2002, 62, 5463–5469.
Aliquots were withdrawn after various time intervals for analysis
as described above.
30 N. Ota, M. Warashina, K. Hirano, K. Hatanaka and K. Taira, Nucleic
Acids Res., 1998, 26, 3385–3391.
31 F. Seela and A. Kehne, Tetrahedron, 1985, 41, 5387–5392.
32 F. Seela and H. Steker, Helv. Chim. Acta, 1985, 68, 563–570.
33 F. Seela and K. Kaiser, Helv. Chim. Acta, 1988, 71, 1813–1823.
34 D. Zhang, Y. M. Li, L. Xu, M. S. Cheng, Y. Zhou, J. L. He and
K. L. Liu, Nucleosides, Nucleotides Nucleic Acids, 2010, 29, 734–
747.
Acknowledgements
We thank the reviewers of this paper for many valuable comments
and suggestions. This work was supported by the National Key
Technologies R&D Program for New Drugs 2009ZX09301-002
and National Natural Science Foundation of China 21072229.
The authors thank Prof. Ningsheng Shao and Guifang Dou for
their help in the 32P-labelling experiments.
35 F. Seela and M. Zulauf, J. Chem. Soc., Perkin Trans. 1, 1998, 3233–
3239.
36 M. J. Cairns, A. King and L. Q. Sun, Nucleic Acids Res., 2003, 31,
2883–2889.
37 R. R. Breaker, G. M. Emilsson, D. Lazarev, S. Nakamura, I. J. Puskarz,
A. Roth and N. Sudarsan, RNA, 2003, 9, 949–957.
38 R. C. Spitale, R. Volpini, M. G. Heller, J. Krucinska, G. Cristalli and J.
E. Wedekind, J. Am. Chem. Soc., 2009, 131, 6093–6095.
39 J. A. Grasby, K. Mersmann, M. Singh and M. J. Gait, Biochemistry,
1995, 34, 4068–4076.
Notes and references
1 S. W. Santoro and G. F. Joyce, Proc. Natl. Acad. Sci. U. S. A., 1997, 94,
4262–4266.
40 R. T. Raines, Chem. Rev., 1998, 98, 1045–1065.
41 M. Emilsson, S. Nakamura, A. Roth and R. R. Breaker, RNA, 2003,
9, 907–918.
42 R. Ting, L. Lermer, J. Thomas, Y. Roupioz and D. M. Perrin, Pure
Appl. Chem., 2004, 76, 1571–1577.
43 S. W. Santoro, G. F. Joyce, K. Sakthivel, S. Gramatikova and C. F.
Barbas III, J. Am. Chem. Soc., 2000, 122, 2433–2439.
44 M. Hollenstein, C. J. Hipolito, C. H. Lam and D. M. Perrin, Nucleic
Acids Res., 2009, 37, 1638–1649.
45 J. M. Thomas, J.-K. Yoon and D. M. Perrin, J. Am. Chem. Soc., 2009,
131, 5648–5658.
2 S. W. Santoro and G. F. Joyce, Biochemistry, 1998, 37, 13330–13342.
3 C. R. Dass, Trends Pharmacol. Sci., 2004, 25, 395–397.
4 L. Q. Sun, M. J. Cairns, E. G. Saravolac, A. Baker and W. L. Gerlach,
Pharmacol. Rev., 2000, 52, 325–347.
5 E. Hjiantoniou, S. Iseki, J. B. Uney and L. A. Phylactou, Biochem.
Biophys. Res. Commun., 2003, 300, 178–181.
6 Y. Y. Xie, X. D. Zhao, L. P. Jiang, H. L. Liu, L. J. Wang, P. Fang, K.
L. Shen, Z. D. Xie, Y. P. Wu and X. Q. Yang, Antiviral Res., 2006, 71,
31–41.
7 T. Kuwabara, M. Warashina, T. Tanabe, K. Tani, S. Asano and K.
Taira, Nucleic Acids Res., 1997, 25, 3074–3081.
5736 | Org. Biomol. Chem., 2011, 9, 5728–5736
This journal is
The Royal Society of Chemistry 2011
©