Improved Electrophoretic Resolution of DNA Fragments in Samples Containing High Concentrations of Salts

It is common knowledge among DNA researchers that large amounts of salt in DNA samples compromise the resolution of agarose gel electrophoresis. The adverse salt effects are illustrated in Figure 1, using the electrophoretic separation of Lambda DNA-BstE II digest (New England Biolabs, Beverly, MA, USA) as an example. Lanes 7 and 8 of Figure 1, respectively, show the effects of using either 1 M NaCl or 3× REACT (100 mM Tris-HCl at pH 7.6, 10 mM MgCl2, 150 mM NaCl; Life Technologies, Gaithersburg, MD, USA) (1) as sample buffers. In these cases, the resolution of DNA fragments is poor because of their altered mobilities. Poor resolution of DNA fragments was even more pronounced if samples contained higher salt concentrations or if the electrophoresis was carried out at higher voltages (e.g., 100, 120 or 140 V; data not shown). We also noticed under our experimental condition of moderate input of DNA with high salts, that the DNA fragments separated better in 1.2% agarose than in 0.8% agarose (data not shown), which is consistent with the results of a previous investigation (3). Many common experimental manipulations result in DNA samples with high concentrations of salts. For a fine fractionation of DNA fragments, these salts are usually removed before electrophoresis by one of the following procedures: dialysis against buffers of low salts, ethanol precipitation, chromatography or centrifugal filtration. However, besides requiring additional time and resources, these methods inevitably cause at least the partial loss of DNA samples. This is particularly undesirable and inconvenient in cases when the samples are limited in quantity or if quantitative recoveries are imperative. We have minimized the salt effect by adding a simple “timeout” step in setting up the agarose gel electrophoresis. The timeout step consisted of resting the samples in the wells for 30 min before the electrophoresis is begun. This allowed the salts in the samples to diffuse into the surrounding TBE buffer (1× TBE buffer has a salt concentration of 0.091 M), thus reducing the salt concentration in the DNA samples. Consequently, the electrophoretic separation of lambda DNA-BstE II fragments in highsalt sample buffer is better after a 5-min timeout step (Figure 1, lanes 4 and 5) and is improved further after a 30-min timeout step (Figure 1, lanes 2 and 3). We did not detect any diffusion of the DNA fragments during the timeout. In addition, we find that a high-salt electrophoresis buffer reverses significantly the poor resolution obtained from use of high-salt sample buffers, since the separation of the DNA fragments is better in 2× TBE (Figure 2A, lanes 1–3), yet worse in 0.5× TBE (Figure 2B, lanes 1–3), compared with the resolution obtained using 1× TBE buffer (Figure 1, lanes 6–8). The observations described can be used as a reference for streamlining a number of routine molecular biology protocols; e.g., in genomic Southern analysis after the complete restriction of genomic DNA in a large volume, the samples can be concentrated using a vacuum concentrator before proceeding directly to gel electrophoresis. The commonly used protocol for concentrating the samples by ethanol-precipitation before electrophoresis is no longer necessary, and thus sample preparation can be simplified (2).