*M.C.B.'s current address is Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA.
**D.A.D.'s present address is Alnylam Pharmaceuticals, Cambridge, MA, USA.
Figure 1: Removal of nicked plasmid DNA from supercoiled plasmid preparations using acid phenol extraction.
A rigorous understanding of the biological function of superhelical tension in cellular DNA requires the development of new tools and model systems for study. To this end, an ethidium bromide–free method has been developed to prepare large quantities of either negatively or positively super-coiled plasmid DNA. The method is based upon the known effects of ionic strength on the direction of binding of DNA to an archaeal histone, rHMfB, with low and high salt concentrations leading to positive and negative DNA supercoiling, respectively. In addition to fully optimized conditions for large-scale (>500 µg) supercoiling reactions, the method is advantageous in that it avoids the use of mutagenic ethidium bromide, is applicable to chemically modified plasmid DNA substrates, and produces both positively and negatively supercoiled DNA using a single set of reagents.
DNA supercoiling is ubiquitous in prokaryotic and eukaryotic genomes (1,2,3,4) and has clear effects on many cellular functions (5,6,7,8,9,10). The study of superhelical tension in DNA, however, has been hampered by the paucity of methods to prepare sizeable quantities of supercoiled plasmid DNA as a model system. Several methods exist to prepare either positively (11) or negatively (12,13) supercoiled plasmid DNA, but these methods are limited by low superhelical density, small preparation size, availability of crucial enzymes, or the need to use highly mutagenic ethidium bromide which causes significant nicking of the plasmid DNA. To date, no single method provides both positively and negatively supercoiled plasmid in parallel and biochemically similar reactions for use in comparative studies of supercoiled DNA.
To address this problem, we modified a positive supercoiling method developed previously (14) to allow the preparation of both positively and negatively supercoiled plasmid DNA. In the original low–ionic strength method, topologically relaxed plasmid is bound to tetramers of the archaeal histone rHMfB, which constrains positively supercoiled toroids, while compensatory negatively supercoiled plectonemes form in the unbound portion of the DNA (Figure 1A). Treatment with topoisomerase then removes the unconstrained plectonemes while leaving the bound toroids intact (Figure 1B), resulting in plasmid with positive supercoiling upon removal of the archaeal protein (Figure 1C). We now expand this method to allow parallel production of negatively supercoiled plasmid. This was achieved by exploiting a published observation (15) that the direction of wrapping of DNA around rHMfB is dictated by the ionic strength of the buffer. With a switch from low to high ionic strength, the direction of wrapping of DNA around the rHMfB histones is reversed to constrain negative toroids and thus negative supercoiling. We have also added an acid phenol extraction step to remove non-closed-circular (i.e., nicked) plasmid from the preparation.
The method begins with preparation of plasmid substrate, a topoisomerase-containing extract and rHMfB protein. Plasmid pUC18 prepared by alkaline lysis (16) was dialyzed against 3 M NaCl in TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0) using high molecular weight (100,000 Da) tubing, followed by exhaustive dialysis against TE buffer alone. A topoisomerase-containing extract of chicken erythrocyte nuclei (referred to as chicken blood extract, or CBE) was prepared from fresh, sterile, citrate-treated chicken blood (Rockland Immunochemical, Gilbertsville, PA) as previously described (17), aliquoted, and stored at -80°C. One unit (U) of CBE was defined as the volume of extract required to relax 1 µg of pUC18 under standard relaxation conditions (see next paragraph). Inactivated CBE for use in preparing sham-treated DNA substrates was prepared by heating CBE to 95°C for 20 min. Isolation of rHMfB was performed as described previously (18), except that cells were lysed by two passages through an Emulsiflex-C5 homogenizer (Avestin, Ottawa, ON, Canada) at 20,000 psi, and a Hi-Prep 16/10 Heparin FF column (GE Healthcare, Piscataway, NJ, USA) was used for affinity purification. Fractions containing rHMfB were dialyzed against 4× rHMfB binding buffer (40 mM Tris, 4 mM EDTA, 8 mM K2HPO4, 200 mM NaCl, pH 8.0), and stored at 4°C.
The first step in the method involves relaxation of negatively supercoiled plasmid (50 µg/mL) by treatment with 1 U CBE/1 µg plasmid in 200 mM NaCl, 20 mM Tris (pH 8.0), 0.25 mM EDTA, 5% glycerol for 1.5 h at 37°C. The reaction was adjusted to 1% SDS and 150 µg/mL proteinase K, and incubated at 37°C for 1.5 h. DNA was further purified by phenol/chloroform/isoamyl alcohol extraction followed by chloroform extraction and ethanol precipitation. After resuspension in TE, the DNA was desalted by passage over a NAP-25 column (GE Healthcare), and quantified spectrophotometrically.
The next step is to determine the optimal proportions of rHMf B and DNA, which is crucial to maximizing the supercoiling; either too little or too much rHMfB will result in lower overall super-coiling (Figure 2A). The proportions of rHMfB and DNA were optimized by mixing 0.5 µg relaxed plasmid (25 µg/mL final concentration) with varying quantities of rHMfB (0.5–1.5 mass equivalents) in 1× rHMfB reaction buffer in the presence (for negative super-coiling) or absence (for positive super-coiling) of 350 mM potassium glutamate. Unrestrained compensatory supercoiling was removed by adding 10 U CBE per µg of plasmid following adjustment of the buffer by addition of 1/10 volume of 590 mM Tris, 8 mM K2HPO4, 22 mM EDTA, 100 mM NaCl and pH 8.0, and further addition of 1/10 volume of either 1.4 M potassium glutamate (for negative supercoiling) or water (for positive super-coiling). Reactions were incubated again for 1.5 h at 37°C. DNA was purified as described earlier and the degree of super-coiling assessed by one-dimensional agarose gel electrophoresis. Batches of ≥500 µg supercoiled plasmid DNA were then prepared by scaling the optimized reactions for the larger quantities and volumes.
Several salt species were tested for efficacy in this protocol, with potassium glutamate found to be the most consistent for production of large quantities of negatively supercoiled plasmid (Table 1). Of the salt species that did not result in supercoiled plasmid, all were found to interfere with activity of the topoisomerase added to remove compensatory supercoils (data not shown).
The proportion of nicked plasmid was minimized to <5% using a modification of the acidic phenol extraction method of Zasloff et al. (19). Plasmid was resus-pended in TE and desalted on a NAP-25 column. The DNA solution (<0.5 mg/mL) was adjusted to 75 mM NaCl and 50 mM sodium acetate, pH 4.0, followed by addition of 1 volume phenol, previously equilibrated with sodium acetate (50 mM, pH 4.0), and vortexing for 5 min. Following centrifugation (6,000× g, 5 min), the aqueous layer was immediately removed and neutralized by addition to 1/10 volume 2 M Tris, pH 8.0. Residual phenol was removed by chloroform extraction, and plasmid was precipitated with 1 volume isopropanol and resus-pended in TE. Abasic sites were quantified by gel electrophoresis following treatment of plasmid DNA with 0.1 M putrescine dihydrochloride (pH 7.0; Sigma-Aldrich, St. Louis, MO, USA) at 37°C for 30 min to convert abasic sites to strand breaks (20). By limiting exposure of the plasmid to acidic conditions, we are able to reduce the fraction of nicked plasmid in the preparation (Supplementary Figure 1A) without DNA depurination (Supplementary Figure 1B).
After purification, topoisomer content and direction of supercoiling in the substrates were determined by one- and two-dimensional gel electrophoresis as described in detail elsewhere (21), with the fraction of nicked molecules determined by gel electrophoresis in the presence of ethidium bromide (21). As shown in Figure 2B, controlled variation of the salt concentration leads to a similar degree of plasmid supercoiling of both signs, with >50% of the molecules possessing linking number changes (ΔLk) from ± 7 to 15 (data not shown). This yields superhelical density (σ) values of ±0.03–0.06 in plasmid pUC18 (2686 bp). The maximally resolved ΔLk for negatively supercoiled plasmid is -10, while that for positively supercoiled plasmid runs up to +15 (data not shown). This degree of supercoiling is similar to that reported using the published methods for preparing positively supercoiled (14) and negatively supercoiled plasmid DNA (13). Individual topoisomer populations can be isolated by extraction of plasmid DNA from excised agarose gel bands or by gradient centrifugation.
In addition to fully optimized conditions for large-scale (>500 µg) reactions, this improved method for preparing plasmid DNA with high levels of positive or negative supercoiling has the advantages of avoiding the use of mutagenic ethidium bromide, of applicability to chemically modified plasmid DNA substrates, and of producing both positively and negatively supercoiled DNA with high superhelical density and low levels of nicking using a single set of reagents. The resulting DNA substrates are ideal for sensitive biochemical and biophysical studies of DNA supercoiling.
The authors thank Kathleen Sandman and John Reeve (Ohio State University) for the rHMfB expression vector and members of the Dedon laboratory for experimental assistance and critical reading of the manuscript. This work was supported by funding from the National Cancer Institute (NCI; grant nos. CA072936, CA110261, and CA103146) and a Center Grant from the National Institute of Environmental Health Sciences (ES002109). M.C.B. was supported by a National Defense Science and Engineering Graduate Fellowship. This paper is subject to the NIH Public Access Policy.
The authors declare no competing interests.
Address correspondence to Peter C. Dedon, Department of Biological Engineering, NE47-277, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA, USA, 02139. email: [email protected]