General Deoxyribozyme-Catalyzed Synthesis of Native 3′-5′ RNA Linkages
Whitney E. Purtha, Rebecca L. Coppins, Mary K. Smalley, and Scott K. Silverman*
Department of Chemistry, UniVersity of Illinois at UrbanasChampaign, 600 South Mathews AVenue,
Urbana, Illinois 61801
Received May 23, 2005; E-mail: firstname.lastname@example.org
Deoxyribozymes (DNA enzymes) are DNA catalysts for a variety
of chemical reactions that typically involve nucleic acid substrates.1
Our laboratory has focused on the in vitro selection of DNA
enzymes for RNA ligation.2A highly challenging goal has been
deoxyribozymes that synthesize native 3′-5′ RNA linkages rapidly
and in high yield for a wide variety of RNA sequences, rather than
for only a limited set of substrates.2fWe recently described a
selection strategy that favors native RNA ligation by incorporating
a stringently 3′-5′-selective step into the selection rounds.2hWe
have now applied this strategy to identify two RNA ligase
deoxyribozymes that rapidly form high yields of 3′-5′ linkages
with modest sequence requirements for the two RNA substrates,
thereby fulfilling all requirements for useful RNA ligase reagents.
Because RNA ligation by protein enzymes3does not always provide
acceptable yields,4,5the identification of general DNA enzymes for
3′-5′ RNA ligation enables alternative synthetic routes that will
be useful for practical biochemistry.
Our recently described selection methodology2awas used to
identify deoxyribozymes that join a 2′,3′-diol to a 5′-triphosphate
(Figure 1). Previously, such ligations led to 2′,5′-branched RNA
by reaction of an internal 2′-hydroxyl group,2b,eor they led to linear
3′-5′ RNA but with restrictive and impractical sequence require-
ments.2fHere, 3′-5′ selectivity during ligation was enforced by
incorporating the RNA-cleaving 8-17 deoxyribozyme6into the
selection procedure,2hstarting at either round 2 (for selections using
40 mM Mg2+) or round 5 (for selections using 1 mM Zn2+). In
both cases, >95% of the ligation products from each uncloned
selection pool had 3′-5′ linkages (Figure S1). When the ligation
activities had stopped increasing, individual deoxyribozymes were
cloned. On the basis of a preliminary survey of activities, two
clones, named 9DB1 (from round 9 of the Mg2+selection) and
7DE5 (from round 7 of the Zn2+selection) were examined further.
By cleaving the ligation products from each of the two new
deoxyribozymes with 8-17, which is highly selective for 3′-5′
RNA linkages,2aboth 9DB1 and 7DE5 were verified to create
3′-5′ linkages (Figure S1).
The 9DB1 and 7DE5 deoxyribozymes ligate RNA under practical
in vitro incubation conditions. As shown in Figure 2, 9DB1 provides
60-70% yield of ligated RNA with kobs∼ 0.04 min-1(t1/2∼ 15
min) at 40 mM Mg2+, pH 9.0, and 37 °C. Similarly, 7DE5 has
40-50% yield of ligated RNA with kobs∼ 0.02 min-1(t1/2∼ 30
min) at 1 mM Zn2+, pH 7.5, and 23 °C. The 9DB1 deoxyribozyme
is also effective at pH 7.5, where kobsis ∼0.2 h-1(t1/2∼ 4 h; data
not shown). Incubation at the lower pH value of 7.5 instead of 9.0
should be useful for synthesis of larger RNAs that may experience
more nonspecific degradation during an overnight incubation period,
particularly at higher pH.
In the selection design that led to 9DB1 and 7DE5, only one
RNA nucleotide of each substrate was not base-paired with the
DNA binding arms (the AVG nucleotides that flank the ligation site;
Figure 1). Experience from other selections suggested that base-
paired RNA nucleotides are likely to tolerate simple RNA:DNA
Watson-Crick covariation without demanding particular RNA
bases at the paired positions.2a,c,d,gIndeed, both 9DB1 and 7DE5
permit almost any changes to their RNA substrates away from the
ligation site (Figure 2). Comprehensive assays revealed that 9DB1
requires only DVRA (D ) A, G, or U; R ) A or G), and 7DE5
needs only AVR (Figures S2 and S3).7For comparison, these
practical DVRA and AVR sequence requirements are each less
restrictive than that of the 8-17 deoxyribozyme (AVG), which along
with related DNA enzymes, such as 10-23 (RVY; Y ) U or C), is
widely used as a general RNA-cleaving biochemical tool.8
To demonstrate the utility of the new deoxyribozymes for
synthesis of biologically derived RNAs, we used 7DE5 to prepare
the Tetrahymena group I intron P4-P6 domain, a representative
and often-studied RNA.5,9,10As shown in Figure 3A, synthesis of
P4-P6 by 7DE5 was readily achieved in good yield, even though
P4-P6 is completely unrelated to the short RNA substrates that
5′-triphosphate substrates. Shown are the sequences of the 40-nucleotide
enzyme regions of the 9DB1 and 7DE5 deoxyribozymes.
DNA-catalyzed 3′-5′ RNA ligation using 2′,3′-diol and
Figure 2. Kinetic assays for the 9DB1 (Mg2+-dependent) and 7DE5 (Zn2+-
dependent) deoxyribozymes for 3′-5′ RNA ligation. Incubation conditions
were 40 mM Mg2+, pH 9.0, 37 °C and 1 mM Zn2+, pH 7.5, 23 °C. The
mutated RNA substrate sequences (dashed bars) differ from the original
sequences (solid bars) at every nucleotide except those near the ligation
site, as indicated in the legend. See Supporting Information for compre-
hensive generality assays. Values of kobsfor 9DB1: original substrates, 0.036
( 0.006 min-1(n ) 7, mean ( standard deviation); mutant substrates,
0.064 ( 0.016 min-1(n ) 3). Values of kobsfor 7DE5: original substrates,
0.019 ( 0.005 min-1(n ) 7); mutant substrates, 0.012 ( 0.002 min-1(n
Published on Web 09/03/2005
13124 9 J. AM. CHEM. SOC. 2005, 127, 13124-13125
10.1021/ja0533702 CCC: $30.25 © 2005 American Chemical Society
were used during the selection procedure that led to the identifica-
tion of 7DE5. The P4-P6 synthesized by 7DE5 was shown
conclusively to have a native 3′-5′ linkage at the ligation junction
created by the deoxyribozyme (Figures S6 and S7), and the synthetic
P4-P6 folds like wild-type P4-P6 as assayed by nondenaturing
PAGE (Figure 3B).5,10To further demonstrate ligation generality,
we used both 9DB1 and 7DE5 to prepare the 72-nucleotide core
of the xpt G-riboswitch11(Figure 3C). In addition to showing the
generality of the new deoxyribozymes, these results demonstrate a
simple approach to synthesize the riboswitch with modifications
for structure-function studies, as has been achieved for P4-P6.5,10
In summary, we have identified the 9DB1 and 7DE5 deoxy-
ribozymes that rapidly create native 3′-5′ RNA linkages in useful
yield and in a sequence-general fashion. These deoxyribozymes
should be of immediate practical utility alongside the familiar but
imperfect protein-mediated splint ligation methodology.3In addition,
9DB1 and 7DE5 are conceptually interesting in the context of their
3′-5′ selectivities, ligation mechanisms, and three-dimensional
structures, all of which will be investigated in other studies.
Recently, extensive selection experiments provided a large family
of DNA enzymes similar to 8-17 that collectively cleave almost
any RNA dinucleotide junction.12A similar effort for RNA ligation
using a comprehensive set of RNA substrate sequences and the
approach described here is anticipated to lead to an analogous family
of 3′-5′ RNA ligase deoxyribozymes. Such experiments are in
progress in our laboratory.
Acknowledgment. This work was supported by the Burroughs
Wellcome Fund, the March of Dimes, the National Institutes of
Health, the American Chemical Society Petroleum Research Fund,
the David and Lucile Packard Foundation, Sigma Xi, and the UIUC
Department of Chemistry. We thank Yangming Wang for advice
on applying 8-17 during the selections, and Chandra Miduturu
for materials used in the P4-P6 ligation experiment.
Supporting Information Available: Selection details; experiments
demonstrating the generality of 9DB1 and 7DE5 for RNA substrate
sequences; MALDI-MS data for oligonucleotide transcripts; procedures
for P4-P6 and riboswitch ligation reactions; nondenaturing PAGE
details; and P4-P6 linkage assays. This material is available free of
charge via the Internet at http://pubs.acs.org.
(1) (a) Emilsson, G. M.; Breaker, R. R. Cell. Mol. Life Sci. 2002, 59, 596-
607. (b) Lu, Y. Chem.sEur. J. 2002, 8, 4589-4596. (c) Silverman, S.
K. Org. Biomol. Chem. 2004, 2, 2701-2706. (d) Achenbach, J. C.;
Chiuman, W.; Cruz, R. P.; Li, Y. Curr. Pharm. Biotechnol. 2004, 5, 321-
(2) (a) Flynn-Charlebois, A.; Wang, Y.; Prior, T. K.; Rashid, I.; Hoadley, K.
A.; Coppins, R. L.; Wolf, A. C.; Silverman, S. K. J. Am. Chem. Soc.
2003, 125, 2444-2454. (b) Wang, Y.; Silverman, S. K. J. Am. Chem.
Soc. 2003, 125, 6880-6881. (c) Wang, Y.; Silverman, S. K. Biochemistry
2003, 42, 15252-15263. (d) Ricca, B. L.; Wolf, A. C.; Silverman, S. K.
J. Mol. Biol. 2003, 330, 1015-1025. (e) Coppins, R. L.; Silverman, S.
K. Nat. Struct. Mol. Biol. 2004, 11, 270-274. (f) Coppins, R. L.;
Silverman, S. K. J. Am. Chem. Soc. 2004, 126, 16426-16432. (g) Coppins,
R. L.; Silverman, S. K. J. Am. Chem. Soc. 2005, 127, 2900-2907. (h)
Wang, Y.; Silverman, S. K. Biochemistry 2005, 44, 3017-3023. (i)
Hoadley, K. A.; Purtha, W. E.; Wolf, A. C.; Flynn-Charlebois, A.;
Silverman, S. K. Biochemistry 2005, 44, 9217-9231. (j) Wang, Y.;
Silverman, S. K. Angew. Chem., Int. Ed. 2005, 44, in press.
(3) (a) Moore, M. J.; Sharp, P. A. Science 1992, 256, 992-997. (b) Moore,
M. J.; Query, C. C. Methods Enzymol. 2000, 317, 109-123.
(4) (a) Han, H.; Dervan, P. B. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4955-
4959. (b) Zahler, A. M.; Roth, M. B. Proc. Natl. Acad. Sci. U.S.A. 1995,
92, 2642-2646. (c) Tarn, W. Y. Biochimie 1996, 78, 1057-1065. (d)
Strobel, S. A.; Ortoleva-Donnelly, L. Chem. Biol. 1999, 6, 153-165. (e)
Nishikawa, F.; Shirai, M.; Nishikawa, S. Eur. J. Biochem. 2002, 269,
(5) Silverman, S. K.; Cech, T. R. Biochemistry 1999, 38, 8691-8702.
(6) Santoro, S. W.; Joyce, G. F. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 4262-
(7) For these assays, right-hand RNA substrates with 5′-triphosphate pyrim-
idines (VY) could not be tested readily because these nucleotides cannot
be incorporated at the 5′-terminus of RNA by transcription using T7 RNA
polymerase. Therefore, our conclusions of the VR sequence requirements
for 9DB1 and 7DE5 are conservative, practical estimates. In either case,
it is possible that VY could be accepted by the deoxyribozyme, if the
corresponding 5′-triphosphate substrate were prepared and tested.
(8) (a) Pyle, A. M.; Chu, V. T.; Jankowsky, E.; Boudvillain, M. Methods
Enzymol. 2000, 317, 140-146. (b) Joyce, G. F. Methods Enzymol. 2001,
(9) (a) Murphy, F. L.; Cech, T. R. Biochemistry 1993, 32, 5291-5300. (b)
Murphy, F. L.; Cech, T. R. J. Mol. Biol. 1994, 236, 49-63. (c) Cate, J.
H.; Gooding, A. R.; Podell, E.; Zhou, K.; Golden, B. L.; Kundrot, C. E.;
Cech, T. R.; Doudna, J. A. Science 1996, 273, 1678-1685. (d) Juneau,
K.; Cech, T. R. RNA 1999, 5, 1119-1129. (e) Juneau, K.; Podell, E.;
Harrington, D. J.; Cech, T. R. Structure 2001, 9, 221-231. (f) Schwans,
J. P.; Cortez, C. N.; Olvera, J. M.; Piccirilli, J. A. J. Am. Chem. Soc.
2003, 125, 10012-10018. (g) Schwans, J. P.; Li, N. S.; Piccirilli, J. A.
Angew. Chem., Int. Ed. 2004, 43, 3033-3037.
(10) (a) Silverman, S. K.; Cech, T. R. Biochemistry 1999, 38, 14224-14237.
(b) Silverman, S. K.; Zheng, M.; Wu, M.; Tinoco, I., Jr.; Cech, T. R.
RNA 1999, 5, 1665-1674. (c) Silverman, S. K.; Deras, M. L.; Woodson,
S. A.; Scaringe, S. A.; Cech, T. R. Biochemistry 2000, 39, 12465-12475.
(d) Silverman, S. K.; Cech, T. R. RNA 2001, 7, 161-166. (e) Young, B.
T.; Silverman, S. K. Biochemistry 2002, 41, 12271-12276. (f) Doherty,
E. A.; Batey, R. T.; Masquida, B.; Doudna, J. A. Nat. Struct. Biol. 2001,
8, 339-343. (g) Matsumura, S.; Ikawa, Y.; Inoue, T. Nucleic Acids Res.
2003, 31, 5544-5551.
(11) (a) Batey, R. T.; Gilbert, S. D.; Montange, R. K. Nature 2004, 432, 411-
415. (b) Serganov, A.; Yuan, Y. R.; Pikovskaya, O.; Polonskaia, A.;
Malinina, L.; Phan, A. T.; Hobartner, C.; Micura, R.; Breaker, R. R.; Patel,
D. J. Chem. Biol. 2004, 11, 1729-1741.
(12) Cruz, R. P. G.; Withers, J. B.; Li, Y. Chem. Biol. 2004, 11, 57-67.
Figure 3. Application of the 9DB1 and 7DE5 deoxyribozymes to prepare
large RNAs. (A) Ligation assay with 7DE5 for the 160-nucleotide P4-P6
RNA. The 32-nt left-hand RNA substrate was 5′-32P-radiolabeled. The 128-
nt right-hand RNA substrate was unradiolabeled. P denotes a product
standard of wild-type P4-P6 that was prepared independently by transcrip-
tion. Timepoints were taken at t ) 0.5, 10, 40, and 180 min (full 6% PAGE
image is shown in Supporting Information). The ligation yield at the final
timepoint was 65%. (B) Functional assay of the ligated P4-P6 RNA. Mg2+-
dependent nondenaturing (native) PAGE5,10demonstrates that P4-P6
synthesized by 7DE5 folds equivalently to wild-type P4-P6. Unf, unfolded
control mutant of P4-P6; wt, wild-type P4-P6; lig, P4-P6 by 7DE5
ligation. See Supporting Information for details. (C) Ligation assays for
the 72-nt xpt G-riboswitch with 9DB1 and 7DE5 (t ) 0.5, 30, 60, and 180
min; full 12% PAGE image is shown in Supporting Information). Ligation
yields at the final timepoints were 55, 68, and 42%.
C O M M U N I C A T I O N S
J. AM. CHEM. SOC. 9 VOL. 127, NO. 38, 2005 13125