Rolling Circle DNA Synthesis: Small Circular Oligonucleotides as Efficient Templates for DNA Polymerases

Abstract

We report that small, single-stranded circular DNA oligonucleotides 26 to 74 nucleotides (nt) in size can behave as catalytic templates for DNA synthesis by several DNA polymerase enzymes. The DNA products are repeating end-to-end multimeric copies of the synthetic circular DNAs, and range from 1 000 to > 12 000 nucleotides in length. Several aspects of this reaction are unusual: first, the synthesis proceeds efficiently despite the curvature and small size of the circles, some of which have diameters significantly smaller than that of the enzyme itself. Second, the synthesis can proceed hundreds of times around the circle, while rolling replication of larger circular plasmid DNAs requires other proteins for processive synthesis. Finally, the synthesis scheme produces multiple copies of the template without the requirement for either heating or cooling cycles and requires less than stoichiometric amounts of primer, unlike other DNA synthesis methods. We report on the scope of this reaction, and demonstrate that the multimeric products can be cleaved enzymatically to short, sequence-defined oligodeoxynucleotides. This new approach to DNA synthesis may be a practical way to produce useful repeating DNAs, and combined with DNA cleavage strategies, it may represent a useful enzymatic approach to short, sequence-defined oligodeoxynucleotides.

Figure 1

(A) Schematic of rolling circle DNA synthesis, in which small circular oligodeoxynucleotides act as templates for DNA polymerases. Multimeric product can also be cleaved into monomer-length oligonucleotides if desired, as shown. (B) Model of a circular 26-base oligodeoxynucleotide next to the crystal structure of the KF polymerase, showing relative sizes. The internal diameter of the circle is ~40 Å; the enzyme dimensions are ~65 × 65 × 85 Å.

Figure 2

Sequences of the circular oligodeoxynucleotides used as templates for polymerases in this study. Shown underneath each circle is the corresponding primer used to initiate synthesis. Arrows indicate 5′ to 3′ strand orientation; mC (used in circle 1) refers to 5-methylcytosine.

Figure 3

Rolling circle DNA synthesis with 34mer circular DNA template 2 and KF enzyme. Shown is an autoradiogram from 20% denaturing PAGE gel analysis of products, which are radiolabeled either by (lane 1) incorporation of α-³²P-dTTPs or (lane 2) by initiating with ³²P-labeled primer. Banding patterns produced during synthesis (lane 1) arise from limiting concentration of dTTP in reaction, causing pauses once per turn, since only one A is present in the circle.

Figure 4

Rolling circle DNA synthesis from 34mer circle 2 (lane 1) and control experiments (lanes 2–6) showing the requirement for circularity, template, primer, enzyme, and dNTP’s for successful reaction. Shown is an autoradiogram from 1% agarose gel analysis of products, which are radiolabeled by initiating with ³²P-labeled primer. Residual counts at the origin (lanes 5 and 6) are artifacts arising from ethanol precipitation.

Figure 5

Comparison of the abilities of several DNA polymerases to process 34mer circular DNA template 2. Shown is an autoradiogram from 1% agarose gel analysis of products, which are radiolabeled by initiating with ³²P-labeled primer. See Experimental Section for details.

Figure 6

Time course showing the increase in length of rolling circle products. Shown is an autoradiogram from 1% agarose gel analysis of products at several time points in one reaction with 42mer circle 3 as template. Products are radiolabeled by initiating with ³²P-labeled primer.

Figure 7

Quantitative analysis of nucleotide uptake into polymeric products, as a function of circle and primer concentrations and ratios, dNTP concentrations, and enzyme activity. (A) The effect of [circle•primer] concentration on nucleotide uptake, using 1:1 circle:primer ratio. Legend: [circle•primer] = 1 nM (◇), 10 nM (○), 100 nM (□). (B) The effect of varied circle:primer ratio on nucleotide uptake. The ratios are 5:1 (□), 1:1 (○), and 1:5 (◇). (C) The effect of lowering dNTP concentration on nucleotide uptake. [dNTP] = 1 mM (◇) and 0.5 mM (□). (D) The effect of initial enzyme amount on polymer yield (as percent nucleotide uptake) at 3 h of reaction. The nucleotide fraction incorporated into polymeric strands was quantitated by UV absorbance of reaction products after removal of free nucleotides by equilibrium dialysis (see Experimental Section).

Figure 8

The effect of size of circular oligodeoxynucleotides on rolling synthesis products with the KF enzyme. Shown is an autoradiogram from 1% agarose gel analysis of products, which are radiolabeled by initiating with ³²P-labeled primer.

Figure 9

Effect of addition of a second aliquot of KF polymerase to a reaction mixture with 42mer circle 3 after 24 h. Quantitation of nucleotides taken up during rolling circle synthesis was performed by equilibrium dialysis (see Experimental Section).

Figure 10

Cleavage of multimer products arising from three different circles (3–5) with restriction endonuclease Taq Iα. Shown is an autoradiogram from denaturing PAGE gel analysis of products, which are radiolabeled by incorporation of α-³²P-dTTP. Banding patterns produced during synthesis (lanes 1, 3, 5) arise from limiting concentration of dTTP in reaction. After cleavage, predicted products are half-length copies of the circles, which contain two copies of the same sequence.