Search

The preparation of two complete series of RNA nucleoside phosphoramidities, each having an appropriately protected amine or thiol functional group is presented. The new phosphoramidites would be particularly valuable for enabling site-specific introduction of biophysical probes and constraints into RNA.

In the ligation reaction, a 2'-5' RNA phosphodiester linkage is created from a 2', 3E-cyclic phosphate and a 5'-hydroxyl group. In addition to providing interesting new examples of nucleic acid catalysts, deoxyribozymes that ligate RNA are anticipated to be a valuable practical complement to protein-mediated splint ligation from preparation of site-specifically modified RNAs.

Research has been conducted on the GAAA tetraloop-receptor motif of the intron RNA P4-P6 domain. Results indicate the existance of the differences between the thermodynamic and kinetic roles of this motif in Mg (super)2+ -dependent folding of P4-P6 tetraloop.

Research has been conducted on the folding kinetics of the 160-nucleotide P4-P6 domain of the Tetrahymena group I intron RNA. The study of this domain has been conducted via stopped-flow fluorescence.

RNA tertiary folding has been monitored by fluorescence of covalently and site-specifically attached pyrene, made possible by availability of 2'-amino-2'-deoxynucleoside phosphoramidites. This kind of internal labeling gives a much wider array of possible derivatization sites vs only one 5'-end and one 3'-end per molecule. Pyrene was chosen because the fluorescence intensity depends especially strongly it it on environment when attached to nucleic acids. A single site-specifically tethered chromophore can report formation of global structure of a large RNA molecule and make it possible to monitor the equilibrium progress and the real-time kinetics of RNA tertiary folding.

Ab initio computations on pyridine and pyridinium analogues of m-xylylene show that the neutral heterocycle is essentially similar to benzene as a ferromagnetic coupling unit, while the cationic pyridiniums act differently. A protonated pyridine functions either as a ferromagnetic coupling unit or an antiferromagnetic coupling unit, depending on the substitution pattern. Bis(trimethylenemethane) analogues of the pyridine and pyridinium biradicals are studied by electron paramagnetic resonance spectroscopy to test the theoretical predictions.

The article explains the optimization of 10MD5, a DNA-hydrolyzing deoyribozyme, achieved by the compromises taking place among pH tolerance, site specificity and sequence tolerance. The compound is shown to be highly sequence, as well as site-specific.

An ensemble and single-molecule fluorescence resonance energy transfer (FRET) study was performed on three orthogonally labeled variants of the xpt guanine riboswitch aptamer domain. The data obtained elucidated the biological role of guanine as stabilizing factor for the globally folded aptamer domain conformation at physiologically relevant [Mg.sup.2+] concentrations ([less than]1 mM), while much higher [Mg.sup.2+] concentrations are required to induce aptamer folding event in absence of guanine.

An energetically substantial stabilization of the RNA tertiary structure is obtained by the strategic attachment of a DNA constraint that has selectively stabilized the folded conformation of the RNA. The results have shown that significant stabilizations are seen when various attachment sites are used on the P4-P6 RNA and energetically nontrivial RNA stabilization is introduced without needing extensive contacts to the RNA itself.

In vitro selection is used for identifying new deoxyribozymes that has catalyzed the Diels-Alder reaction between anthracene and maleimide substrates. The comparison of both rate constants and rate enhancements has shown that DAB22 has catalyzed the Diels-Alder reaction as efficiently as 39M49, thus showing that deoxyribozymes are as efficient as the ribozymes for catalysis of C-C bond formation.

The synthesis of a thymidine phosphoramidite that has the 5'-tethered aldehyde masked as a 1,2-diol and a convenient protocol developed for the oxidative cleavage of the 1,2-diol to the aldehyde using NaIO4 is described. Both the procedures would find applicability in attaching DNA to biomolecules.

The role of the leaving group in the 7S11-catalyzed reaction by altering the 5'-phosphorylation state and the length of the second RNA substrate is comprehensively examined. The observation of 7S11-catalyzed branch formation with an oligonucleotide leaving group strengthens this resemblance to natural RNA splicing, with the oligonucleotide playing the role of the 5'-exon in the first step.

A brief report on Zn(super 2 +)-dependent deoxyribozymes that ligate RNA is presented. The DNA enzymes are identified by in vitro selection and ligate RNA at 1 mM Zn(super 2 +) and 23 Celsius, pH 7.9 that is faster.

The 7S11 DNA enzyme mediates the nucleophilic attack of an adenosine 2'-hydroxyl group at a 5'-triphosphate, forming 2',5'-branched RNA in a reaction that resembles the first step of in vivo RNA splicing. The 7S11 characterization experiments that have two important implications for nucleic acid chemistry and biochemistry are described.

The results related to natural selection, in which abrupt environmental variation can provide a rapid change in selection pressure. The new approach to obtain 3'-5' selective RNA ligase deoxyribozymes is particularly important for ongoing selection efforts. Linear 3'-5' RNA linkages are an important practical objective because the native backbone is desirable in site-specifically modified ribozymes assembled by ligation.

A study is conducted to obtain deoxyribozymes that instead create linear 3'-5'-linked RNA, a strategically modified the selection approach is proposed by embedding the nascent ligation junction within an RNA:DNA duplex region. These RNA ligase deoxyribozymes are the first that create native 3'-5' RNA linkages, which are highly elusive via other selection approaches.

Characterization of DNA enzymes with respect to their branch forming activity is discussed on the basis the type of branched RNA synthesized by it. The results support the utility of deoxyribozymes in creating synthetic 2', 5' branched RNAs for investigation of group II intron splicing and other biochemical reactions.

Branched and lariat RNA can be synthesized by deoxyribozymes that have significant potential for convenient and practical generalization is demonstrated. Efforts are being made to expand the scope of branched and lariat RNA formation by the new deoxyribozymes, including variation of loop size and composition for the lariats.

The evolution of deoxyribozymes with RNA ligase activity is reported. The deoxyribozymes that emerged from the process ligate RNA with highly selective formation of 2'-5' phosphodiester linkages.

Deoxyribozymes 9Db1 and 7DE5 are identified that rapidly create native 3' - 5' RNA linkages in useful yield and in a sequence-general fashion. 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 further studies.

A conceptually novel approach to DNA nanotechnology is described in which a DNA constraint actively controls the conformation of an attached RNA macromolecule. The balance between the competing RNA and DNA structures is controlled by the Mg(super 2+) concentration when the native Mg(super 2+)-dependent RNA conformation and the DNA constraint cannot exist simultaneously.

Conspectus: Catalysis is a fundamental chemical concept, and many kinds of catalysts have considerable practical value. Developing entirely new catalysts is an exciting challenge. Rational design and screening have provided many new small-molecule catalysts, and directed evolution has been used to optimize or redefine the function of many protein enzymes. However, these approaches have inherent limitations that prompt the pursuit of different kinds of catalysts using other experimental methods. Nature evolved RNA enzymes, or ribozymes, for key catalytic roles that in modern biology are limited to phosphodiester cleavage/ligation and amide bond formation. Artificial DNA enzymes, or deoxyribozymes, have great promise for a broad range of catalytic activities. They can be identified from unbiased (random) sequence populations as long as the appropriate in vitro selection strategies can be implemented for their identification. Notably, in vitro selection is different in key conceptual and practical ways from rational design, screening, and directed evolution. This Account describes the development by in vitro selection of DNA catalysts for many different kinds of covalent modification reactions of peptide and protein substrates, inspired in part by our earlier work with DNA-catalyzed RNA ligation reactions. In one set of studies, we have sought DNA-catalyzed peptide backbone cleavage, with the long-term goal of artificial DNA-based proteases. We originally anticipated that amide hydrolysis should be readily achieved, but in vitro selection instead surprisingly led to deoxyribozymes for DNA phosphodiester hydrolysis; this was unexpected because uncatalyzed amide bond hydrolysis is 105-fold faster. After developing a suitable selection approach that actively avoids DNA hydrolysis, we were able to identify deoxyribozymes for hydrolysis of esters and aromatic amides (anilides). Aliphatic amide cleavage remains an ongoing focus, including via inclusion of chemically modified DNA nucleotides in the catalyst, which we have recently found to enable this cleavage reaction. In numerous other efforts, we have investigated DNA-catalyzed peptide side chain modification reactions. Key successes include nucleopeptide formation (attachment of oligonucleotides to peptide side chains) and phosphatase and kinase activities (removal and attachment of phosphoryl groups to side chains). Through all of these efforts, we have learned the importance of careful selection design, including the frequent need to develop specific "capture" reactions that enable the selection process to provide only those DNA sequences that have the desired catalytic functions. We have established strategies for identifying deoxyribozymes that accept discrete peptide and protein substrates, and we have obtained data to inform the key choice of random region length at the outset of selection experiments. Finally, we have demonstrated the viability of modular deoxyribozymes that include a small-molecule-binding aptamer domain, although the value of such modularity is found to be minimal, with implications for many selection endeavors. Advances such as those summarized in this Account reveal that DNA has considerable catalytic abilities for biochemically relevant reactions, specifically including covalent protein modifications. Moreover, DNA has substantially different, and in many ways better, characteristics than do small molecules or proteins for a catalyst that is obtained "from scratch" without demanding any existing information on catalyst structure or mechanism. Therefore, prospects are very strong for continued development and eventual practical applications of deoxyribozymes for peptide and protein modification. [ABSTRACT FROM AUTHOR]

Research was conducted to quantify the tertiary structure contributions of individual hydrogen bonds in a 'ribose zipper' motif of the crystallized Tetrahymena group I intron P4-P6 domain. The nucleotides were substituted with their 2'-deoxy or 2'-methoxy analogues and the energetic effects on tertiary folding were analyzed with a quantitative Mg2+-dependent nondenaturing-gel assay. Results demonstrate that entropically favorable hydrophobic interactions balance enthalpically unfavorable hydrogen bond deletions and steric clashes for multiple 2'methoxy substitutions.

RNA molecules have numerous functions including catalysis and small molecule recognition, which typically arise from a tertiary structure. There is increasing interest in mechanisms for the thermostability of functional RNA molecules. Sosnick, Pan, and co-workers introduced the notion of "functional stability" as the free energy of the tertiary (functional) state relative to the next most stable (nonfunctional) state. We investigated the extent to which secondary structure stability influences the functional stability of nucleic acids. Intramolecularly folding DNA triplexes containing alternating T•AT and C+•GC base triples were used as a three-state model for the folding of nucleic acids with functional tertiary structures. A four-base-pair tunable region was included adjacent to the triplex-forming portion of the helix to allow secondary structure strength to be modulated. The degree of folding cooperativity was controlled by pH, with high cooperativity maintained by lower pH (5.5), and no cooperativity by higher pH (7.0). We find a linear relationship between functional free energy and the free energy of the secondary structure element adjacent to tertiary interactions, but only when folding is cooperative. We translate the definition of functional stability into equations and perform simulations of the thermodynamic data, which lend support to this model. The ability to increase the melting temperature of tertiary structure by strengthening base-pairing interactions separate from tertiary interactions provides a simple means for evolving thermostability in functional RNAs. [ABSTRACT FROM AUTHOR]

We recently used in vitro selection to identify 7511, a deoxyribozyme that synthesizes 2',5'- branched AMA. The 7511 DNA enzyme mediates the nucleophilic attack of an adenosine 2'-hydroxyl group at a 5'-triphosphate, forming 2',5'-branched AMA in a reaction that resembles the first step of in vivo AMA splicing. Here, we describe 7S1 i characterization experiments that have two important implications for nucleic acid chemistry and biochemistry. First, on the basis of a comprehensive analysis of its substrate sequence requirements, 7S1 1 is shown to be generally applicable for the synthesis of a wide range of 2',5'-branched RNAs. Strict substrate sequence requirements are found at the two AMA nucleotides that directly form the branched linkage, and these requirements correspond to those nucleotides found most commonly at these two positions in natural spliced RNA5. Outside of these two nucleotides, most substrate sequences are tolerated with useful ligation activity, although rates and yields vary. Because chemical synthesis approaches to branched AMA are extremely limited in scope, the deoxyribozyme-based route using 7511 will enable many experiments that require branched RNA. Second, comprehensive nucleotide covariation experiments demonstrate that 7S11 and its AMA substrates adopt a three-helix-junction structure in which the branch-site nucleotide is located at the intersection of the three helices. Because many natural ribozymes have multi-helix junctions, 7S11 is an interesting model system for catalytic nucleic acids. [ABSTRACT FROM AUTHOR]

Previous experiments have identified numerous RNA ligase deoxyribozymes, each of which can synthesize either 2',5'-branched RNA, linear 2'-5'-linked RNA, or linear 3'-5'-linked RNA. These products may be formed by reaction of a 2'-hydroxyl or 3'-hydroxyl of one RNA substrate with the 5'-triphosphate of a second RNA substrate. Here the inherent propensities for nucleophilic reactivity of specific hydroxyl groups were assessed using RNA substrates related to the natural sequences of spliceosome substrates and group II introns. With the spliceosome substrates, nearly half of the selected deoxyribozymes mediate a ligation reaction involving the natural branch-point adenosine as the nucleophile. In contrast, mostly linear RNA is obtained with the group II intron substrates. Because the two sets of substrates differ at only three nucleotides, we conclude that the location of the newly created ligation junction in DNA-catalyzed branch formation depends sensitively on the RNA substrate sequences. During the experiment that led primarily to branched RNA, we abruptly altered the selection strategy to demand that the deoxyribozymes create linear 3'-5' linkages by introducing an additional selection step involving the 3'-5'-selective 8-17 deoxyribozyme. Although no 3'-5' linkages (⩽ 1%) were detectable in the pool products at the point that the 3'-5' selection pressure was applied, deoxyribozymes that specifically create 3'-5' linkages quickly emerged within a few selection rounds. Our success in obtaining 3'-5' linkages via this approach shows that the outcome of deoxyribozyme selection experiments can be dramatically redirected by strategic changes in the selection procedure, even at a late stage. These results relate to natural selection, in which abrupt environmental variation can provide a rapid change in selection pressure. Linear 3'-5' RNA linkages are an important practical objective because the native backbone is desirable in site-specifically modified ribozymes assembled by ligation. Therefore, this new approach to obtain 3'-5'-selective RNA ligase deoxyribozymes is particularly important for ongoing selection efforts. [ABSTRACT FROM AUTHOR]

We recently reported deoxyribozymes (DNA enzymes) that synthesize 2',5'-branched RNA. The in vitro-selected 9F7 and 9F21 deoxyribozymes mediate reaction of a branch-site adenosine 2'-hydroxyl on one RNA substrate with the 5'-triphosphate of another RNA substrate. Here we characterize these DNA enzymes with respect to their branch-forming activity. Both 9F7 and 9F21 are much more active with Mn[sup2+] than with Mg[sup2+] The K[subd,app](Mg[sup2+]) > 400 mM but K[subd,app](Mn[sup2+]≃ 20-50 mM, and the ligation rates k[subobs] are orders of magnitude faster with Mn[sup2+] than with Mg[sup2+] (e.g., 9F7 ≃ 0.3 min[sup-1] with 20 mM Mn[sup2+] versus 0.4 h[sup-1] with 100 mM Mg[sup2+], both at pH 7.5 and 37 °C). Of the other tested transition metal ions Zn[sup2+], Ni[sup2+], Co[sup2+], and Cd[sup2+], only Co[sup2+] supports a trace amount of activity. 9F7 is more tolerant than 9F21 of varying the RNA substrate sequences. For the RNA substrate that donates the adenosine 2'-hydroxyl, 9F7 requires YUA, where Y = pyrimidine and A is the branch site. The 3'-tail emerging from the branch-site A may have indefinite length, but it must be at least one nucleotide long for high activity. The 5'-triphosphate RNA substrate requires several additional nucleotides with varying sequence requirements (5'-pppGRMWR). Outside of these regions that flank the ligation site, 9F7 and 9F21 tolerate any RNA substrate sequences via Watson-Crick covariation of the DNA binding arms that interact directly with the substrates. 9F7 provides a high yield of 2',5'-branched RNA on the preparative nanomole scale. The ligation reaction is effectively irreversible; the pyrophosphate leaving group in the ligation reaction does not induce 2',5'-cleavage, and pyrophosphate does not significantly inhibit ligation except in 1000- fold excess. Deleting a specific nucleotide in one of the DNA binding arms near the ligation junction enhances ligation activity, suggesting an interesting structure near this region of the deoxyribozyme- substrate complex. [ABSTRACT FROM AUTHOR]

We describe an unprecedented DNA-catalyzed DNA cleavage process in which a radical-based reaction pathway cleanly results in excision of most atoms of a specific guanosine nucleoside. Two new deoxyribozymes (DNA enzymes) were identified by in vitro selection from N 40 or N 100 random pools initially seeking amide bond hydrolysis, although they both cleave simple single-stranded DNA oligonucleotides. Each deoxyribozyme generates both superoxide (O 2 -• or HOO • ) and hydrogen peroxide (H 2 O 2 ) and leads to the same set of products (3'-phosphoglycolate, 5'-phosphate, and base propenal) as formed by the natural product bleomycin, with product assignments by mass spectrometry and colorimetric assay. We infer the same mechanistic pathway, involving formation of the C4' radical of the guanosine nucleoside that is subsequently excised. Consistent with a radical pathway, glutathione fully suppresses catalysis. Conversely, adding either superoxide or H 2 O 2 from the outset strongly enhances catalysis. The mechanism of generation and involvement of superoxide and H 2 O 2 by the deoxyribozymes is not yet defined. The deoxyribozymes do not require redox-active metal ions and function with a combination of Zn 2+ and Mg 2+ , although including Mn 2+ increases the activity, and Mn 2+ alone also supports catalysis. In contrast to all of these observations, unrelated DNA-catalyzed radical DNA cleavage reactions require redox-active metals and lead to mixtures of products. This study reports an intriguing example of a well-defined, DNA-catalyzed, radical reaction process that cleaves single-stranded DNA and requires only redox-inactive metal ions.

DNA catalysts (deoxyribozymes) for a variety of reactions have been identified by in vitro selection. However, for certain reactions this identification has not been achieved. One important example is DNA-catalyzed amide hydrolysis, for which a previous selection experiment instead led to DNA-catalyzed DNA phosphodiester hydrolysis. Subsequent efforts in which the selection strategy deliberately avoided phosphodiester hydrolysis led to DNA-catalyzed ester and aromatic amide hydrolysis, but aliphatic amide hydrolysis has been elusive. In the present study, we show that including modified nucleotides that bear protein-like functional groups (any one of primary amino, carboxyl, or primary hydroxyl) enables identification of amide-hydrolyzing deoxyribozymes. In one case, the same deoxyribozyme sequence without the modifications still retains substantial catalytic activity. Overall, these findings establish the utility of introducing protein-like functional groups into deoxyribozymes for identifying new catalytic function. The results also suggest the longer-term feasibility of deoxyribozymes as artificial proteases.

This study focuses on the development of DNA catalysts (deoxyribozymes) that modify side chains of peptide substrates, with the long-term goal of achieving DNA-catalyzed covalent protein modification. We recently described several deoxyribozymes that modify tyrosine (Tyr) or serine (Ser) side chains by catalyzing their reaction with 5'-triphosphorylated RNA, forming nucleopeptide linkages. In each previous case, the side chain was presented in a highly preorganized three-dimensional architecture such that the resulting deoxyribozymes inherently cannot function with free peptides or proteins, which do not maintain the preorganization. Here we describe in vitro selection of deoxyribozymes that catalyze Tyr side chain modification of tethered and free peptide substrates, where the approach can potentially be generalized for catalysis involving large proteins. Several new deoxyribozymes for Tyr modification (and several for Ser modification as well) were identified; progressively better catalytic activity was observed as the selection design was strategically changed. The best new deoxyribozyme, 15MZ36, catalyzes covalent Tyr modification of a free tripeptide substrate with a k(obs) of 0.50 h(-1) (t(1/2) of 83 min) and up to 65% yield. These findings represent an important advance by demonstrating, for the first time, DNA catalysis involving free peptide substrates. The new results suggest the feasibility of DNA-catalyzed covalent modification of side chains of large protein substrates and provide key insights into how to achieve this goal.

We recently reported the identification by in vitro selection of 10MD5, a deoxyribozyme that requires both Mn2+ and Zn2+ to hydrolyze a single-stranded DNA substrate with formation of 5′-phosphate and 3′-hydroxyl termini. DNA cleavage by 10MD5 proceeds with kobs=2.7 h(−1) and rate enhancement of 10(12) over the uncatalyzed P−O hydrolysis reaction. 10MD5 has a very sharp pH optimum near 7.5, with greatly reduced DNA cleavage rate and yield when the pH is changed by only 0.1 unit in either direction. Here we have optimized 10MD5 by reselection (in vitro evolution), leading to variants with broader pH tolerance, which is important for practical DNA cleavage applications. Because of the extensive Watson−Crick complementarity between deoxyribozyme and substrate, the parent 10MD5 is inherently sequence-specific; i.e., it is able to cleave one DNA substrate sequence in preference to other sequences. 10MD5 is also site-specific because only one phosphodiester bond within the DNA substrate is cleaved, although here we show that intentionally creating Watson−Crick mismatches near the cleavage site relaxes the site specificity. Newly evolved 10MD5 variants such as 9NL27 are also sequence-specific. However, the 9NL27 site specificity is relaxed for some substrate sequences even when full Watson−Crick complementarity is maintained, corresponding to a functional compromise between pH tolerance and site specificity. The site specificity of 9NL27 may be restored by expanding its “recognition site” from ATGT (as for 10MD5) to ATGTT or larger, i.e., by considering 9NL27 to have reduced substrate sequence tolerance relative to 10MD5. These findings provide fundamental insights into the interplay among key deoxyribozyme characteristics of tolerance and selectivity, with implications for ongoing development of practical DNA-catalyzed DNA hydrolysis.

We report a deoxyribozyme (DNA enzyme) that catalyzes the convergent and general synthesis of branched DNA. The 15HA9 deoxyribozyme mediates nucleophilic attack of the 2'-hydroxyl group of a ribonucleotide embedded within one DNA substrate into a 5'-adenylate of the second DNA substrate. This approach can be used to synthesize multiply branched DNA with a wide range of DNA sequences.

Self-splicing group II intron RNAs catalyze a two-step process in which the intron is excised as a lariat by two successive phosphodiester exchange reactions. The reversibility of the first step has been hypothesized to act as a proofreading mechanism for improper 5'-splice site selection. However, without synthetic access to mis-spliced RNAs, this hypothesis could not be tested. Here, we used a deoxyribozyme to synthesize several branched RNAs that are derived from the ai5gamma group II intron and mis-spliced at the 5'-splice site. Unlike the correctly spliced ai5gamma RNAs, the mis-spliced RNAs are observed not to undergo the reverse of the first step. This is well-controlled negative evidence against the hypothesis that first-step reversibility is a proofreading mechanism for 5'-splice site selection. In a reaction equivalent either to the hydrolytic first step of splicing or to the hydrolytic reverse of the second step of splicing, a mis-spliced 5'-exon can be "trimmed" to its proper length by the corresponding mis-spliced intron, and in one case, the trimmed 5'-exon was observed to proceed correctly through the second step of splicing. These findings are the first direct evidence that this second proofreading mechanism can occur with a group II intron RNA that is mis-spliced at the 5'-splice site. On the basis of the likely structural and evolutionary relationship between group II introns and the spliceosome, we suggest that this second proofreading mechanism may be operative in the spliceosome.

We report Zn(2+)-dependent deoxyribozymes that ligate RNA. The DNA enzymes were identified by in vitro selection and ligate RNA with k(obs) up to 0.5 min(-)(1) at 1 mM Zn(2+) and 23 degrees C, pH 7.9, which is substantially faster than our previously reported Mg(2+)-dependent deoxyribozymes. Each new Zn(2+)-dependent deoxyribozyme mediates the reaction of a specific nucleophile on one RNA substrate with a 2',3'-cyclic phosphate on a second RNA substrate. Some of the Zn(2+)-dependent deoxyribozymes create native 3'-5' RNA linkages (with k(obs) up to 0.02 min(-)(1)), whereas all of our previous Mg(2+)-dependent deoxyribozymes that use a 2',3'-cyclic phosphate create non-native 2'-5' RNA linkages. On this basis, Zn(2+)-dependent deoxyribozymes have promise for synthesis of native 3'-5'-linked RNA using 2',3'-cyclic phosphate RNA substrates, although these particular Zn(2+)-dependent deoxyribozymes are likely not useful for this practical application. Some of the new Zn(2+)-dependent deoxyribozymes instead create non-native 2'-5' linkages, just like their Mg(2+) counterparts. Unexpectedly, other Zn(2+)-dependent deoxyribozymes synthesize one of three unnatural linkages that are formed upon the reaction of an RNA nucleophile other than a 5'-hydroxyl group. Two of these unnatural linkages are the 3'-2' and 2'-2' linear junctions created when the 2'-hydroxyl of the 5'-terminal guanosine of one RNA substrate attacks the 2',3'-cyclic phosphate of the second RNA substrate. The third unnatural linkage is a branched RNA that results from attack of a specific internal 2'-hydroxyl of one RNA substrate at the 2',3'-cyclic phosphate. When compared with the consistent creation of 2'-5' linkages by Mg(2+)-dependent ligation, formation of this variety of RNA ligation products by Zn(2+)-dependent deoxyribozymes highlights the versatility of transition metals such as Zn(2+) for mediating nucleic acid catalysis.

Branched RNA molecules with a 2',5'-phosphodiester linkage are important biochemical intermediates. Lariat RNA is a particular type of branched RNA that is formed during intron splicing in vivo. Synthesis of branched and lariat RNA is challenging, and there are few general approaches that are applicable in vitro. Here we report the identification of divalent metal-dependent deoxyribozymes (DNA enzymes) that synthesize branched and lariat RNA. In vitro selection was used to obtain deoxyribozymes that selectively join an internal RNA 2'-hydroxyl with a 5'-terminal triphosphate in a convenient "binding arms" format. At least 85% yield of 2',5'-branched RNA is obtained at 37 degrees C and 20 mM Mn2+, pH 7.5 in