Improved deoxyribozymes for synthesis of covalently branched DNA and RNA

Abstract

A covalently branched nucleic acid can be synthesized by joining the 2'-hydroxyl of the branch-site ribonucleotide of a DNA or RNA strand to the activated 5'-phosphorus of a separate DNA or RNA strand. We have previously used deoxyribozymes to synthesize several types of branched nucleic acids for experiments in biotechnology and biochemistry. Here, we report in vitro selection experiments to identify improved deoxyribozymes for synthesis of branched DNA and RNA. Each of the new deoxyribozymes requires Mn²(+) as a cofactor, rather than Mg²(+) as used by our previous branch-forming deoxyribozymes, and each has an initially random region of 40 rather than 22 or fewer combined nucleotides. The deoxyribozymes all function by forming a three-helix-junction (3HJ) complex with their two oligonucleotide substrates. For synthesis of branched DNA, the best new deoxyribozyme, 8LV13, has k(obs) on the order of 0.1 min⁻¹, which is about two orders of magnitude faster than our previously identified 15HA9 deoxyribozyme. 8LV13 also functions at closer-to-neutral pH than does 15HA9 (pH 7.5 versus 9.0) and has useful tolerance for many DNA substrate sequences. For synthesis of branched RNA, two new deoxyribozymes, 8LX1 and 8LX6, were identified with broad sequence tolerances and substantial activity at pH 7.5, versus pH 9.0 for many of our previous deoxyribozymes that form branched RNA. These experiments provide new, and in key aspects improved, practical catalysts for preparation of synthetic branched DNA and RNA.

Figure 1.

Covalently branched nucleic acids. (A) General structure and synthesis strategy for covalently branched nucleic acids. Two substrates are ligated: the left-hand and right-hand substrates, or L and R. The branch-site ribonucleotide, rX, of the L substrate is colored red. The R substrate (with 5′-adenylated or 5′-triphosphorylated nucleotide Z) is colored blue. B = nucleobase. (B) Chemical structure of covalently branched DNA (R = H) and RNA (R = OH). The goal of this study is to identify improved deoxyribozymes for synthesis of these branched nucleic acids.

Figure 2.

Three-helix-junction (3HJ) complexes between deoxyribozymes and their left-hand and right-hand (L and R) substrates for synthesis of covalently branched nucleic acids. Loops A and B are initially random at the outset of in vitro selection but have well-defined sequences for each particular deoxyribozyme. Paired regions P1 through P4 have Watson–Crick base pairs between deoxyribozyme and substrate. Note that formation of P4 draws together the nucleophile (branch-site nucleotide) and electrophile (5′-adenylate or 5′-triphosphate) such that they are juxtaposed at the intersection of a 3HJ composed of P1, P2, and P4. (A) The 7S11 deoxyribozyme that forms branched RNA (23,33). (B) The 15HA9 deoxyribozyme that forms branched DNA (6). (C) The new in vitro selection experiment for branched DNA synthesis, with expanded N₃₃ loop A and P1 region (9 bp) relative to 15HA9. The new selection for branched RNA synthesis was similar, except the L substrates were DNA rather than RNA, and the R substrate was 5′-triphosphorylated. The covalent linkage between deoxyribozyme and R substrate as required during selection is shown explicitly (the 3′-terminal rGrGrA allows ligation by T4 RNA ligase), whereas for all of the deoxyribozyme assays performed after selection, this linkage was absent. Not depicted fully is a 3′-extension ( … AGCTGATCCTGATGG-3′) on the deoxyribozyme, to enable PCR primer binding after each selection step.

Figure 3.

The 8LV13 deoxyribozyme for branched DNA synthesis. (A) PAGE images for assays performed with systematic sequence changes in either the P2 region of the L substrate (left) or the P3 region of the R substrate (right). The specified number of nucleotides adjacent to the branch-site ribonucleotide were unchanged, whereas all other nucleotides of the indicated region were changed by Tv1. Assays were performed in 50 mM HEPES, pH 7.5, 20 mM MnCl₂ 150 mM NaCl and 2 mM KCl at 37°C. t = 0, 2, 7, 30 and 60 min. (B) Kinetic plot for the assays in (A). Data for 8LV1 and 8LV9 were similar (data not shown).

Figure 4.

Establishing generality of 8LV13 for branched DNA synthesis by comprehensive sequence changes in both the L and R DNA substrates. Assays were performed as in Figure 3. (A) PAGE image, with timepoints at t = 0, 2, 7, 30 and 60 min. (B) Kinetic plots. k_obs values (min⁻¹): Par 0.14, Tv1 0.07, Tsn 0.19, Tv2 0.10.

Figure 5.

Branch-site nucleotide tolerance of 8LV13. Substrate sequences were as shown in Figure 2C except for variation in the branch-site nucleotide of the L substrate.

Figure 6.

Assays of 8LX deoxyribozymes for branched RNA synthesis. (A) Kinetic plots for assays of 8LX1, with systematic sequence changes in the P1 and P2 regions of the L substrate. (B) Kinetic plots for assays of 8LX6. Data for 8LX8 (not shown) was similar to that shown for 8LX6.

Figure 7.

Assays of 8LX1 and 8LX6 with nucleotide changes in the L and R substrates, to determine the maximum allowed changes. (A) Assays of 8LX1 and 8LX6 to determine the maximum allowed changes to the R substrate. Each deoxyribozyme was assayed with Tv1 changes to L (all nucleotides of P1 and P2) as well as R (keeping the specified number of nucleotides unchanged beginning at the 5′-terminus). (B) Assays of 8LX6 with Tsn changes in the L substrate. Both P1 and P2 of the L substrate were changed by Tsn, except for the indicated number of nucleotides in the P1 and P2 regions. In all cases, full Watson–Crick base pairing was retained in both P1 and P2 between L and the deoxyribozyme.

Figure 8.

Establishing generality of 8LX6 for branched RNA synthesis by comprehensive sequence changes in both the L and R RNA substrates. For Tv1 and Tv2 changes, all nucleotides in L (P1 + P2) were changed, and all nucleotides except the 5′-G in R (P3 + P4) were changed. For Tsn changes, all L and R nucleotides were similarly changed, except the first two nucleotides adjacent to the branch site of P2 were unmodified. (A) PAGE image, with timepoints at t = 0, 4, 30 min; 4, 11 and 24 h. Sequence-dependent loss of signal due to 5′-phosphate hydrolysis is evident and likely unavoidable. Comparable yields were observed when the reactions were performed on the 0.5 nmol scale using nonradiolabeled substrates. (B) Kinetic plots. k_obs values: Par 17.1 h⁻¹ (0.29 min⁻¹), Tv1 0.47 h⁻¹, Tsn 0.07 h⁻¹, Tv2 0.16 h⁻¹.

Figure 9.

Branch-site nucleotide tolerance of (A) 8LX1, and (B) 8LX6. Substrate sequences were unchanged except for variation at the branch-site nucleotide of the L substrate.

Figure 10.

Substrate sequence requirements for the new deoxyribozymes. (A) 8LV13, for synthesis of branched DNA. (B) 8LX1 and 8LX6, for synthesis of branched RNA. For all three deoxyribozymes, see Table 1 for the 33 nt loop A sequences. The same color scheme as in Figure 2 is used, except that substrate nucleotides depicted here in black have restricted identities. Those nucleotides in color (P1 and P2, gray; P3 and P4, blue; branch site, red) can be changed freely as long as Watson–Crick covariation is maintained. For both 8LX1 and 8LX6, the 4 nt immediately following loop A (5′-GGGC-3′ in both cases) are different from those at the outset of in vitro selection (Figure 2C) and are fixed even when the L substrate sequence is changed. For 8LX6, there may be some restriction on the L sequence for specific nucleotide changes involving transitions relative to the parent sequence (see Figure 7). For 8LX1 and 8LX6, branch-site rG is tolerated with modest efficiency (see Figure 9).