Making colony PCR easier by adding gel-loading buffer to the amplification reaction.

Because of its simplicity, PCR performed directly from bacterial colonies on agar plates has been extensively used as a first step in the identification of recombinant clones (1–3,6,7). Instead of making DNA minipreps and restriction digestion from several putative recombinant colonies, colony PCR allows the identification of the positive clones, and so reduces these procedures to one clone per ligation experiment. When several ligations are being performed simultaneously or when the recombinant frequency is expected to be low and no blue/white screening is available, it is necessary to perform several colony PCR amplifications. We realized that colony PCR could be made easier if the gel-loading buffer was included in the amplification mix instead of being added individually to each tube just before the electrophoresis. We evaluated the effects of three widely used gel-loading buffers on the amplification of targets with different sizes, from different sources and using several primer combinations. Plasmids containing inserts with different sizes and coming from different sources were introduced into Escherichia coli strain DH5αTM (Life Technologies, Gaithersburg, MD, USA) by a calcium chloride method (5). Bacterial lysate was obtained from ampicillin-resistant colonies on LB plates by a slight modification of Güssow and Clackson (2): colonies (0.5–1.5 mm diameter) were picked and transferred to 50 μL water, boiled for 5 min and stored at 4°C until used. Gel-loading buffers (5) were prepared as 6× concentrated stocks: for buffer I, 0.25% (wt/vol) bromophenol blue, 0.025% (wt/vol) xylene cyanol FF and 40% (wt/vol) sucrose in water; for buffer II, 0.25% (wt/vol) bromophenol blue, 0.025% (wt/vol) xylene cyanol FF and 15% (wt/vol) Ficoll (Type 400) in water and for buffer III, 0.25% (wt/vol) bromophenol blue, 0.025% (wt/vol) xylene cyanol FF and 30% (vol/vol) glycerol in water. PCR amplifications were carried out by mixing 2 μL lysate with 8 μL of a premixture so that the final concentration of each reaction was 50 mM KCl, 1.5 mM MgCl2, 10 mM Tris-HCl, pH 9.0, 200 μM each dNTP, 0.4 μM of each primer, 0.5 U of Taq DNA polymerase (Amersham Pharmacia Biotech, São Paulo, Brazil) and gelloading buffer to final concentrations of 0, 0.25 and 0.5×, respectively. (These low concentrations still cause DNA samples to sink into the gel wells.) The inserts were amplified in 0.2 mL tubes for 30 cycles at 94°C for 30 s, 55°C for 30 s and 72°C for 2 min, followed by a 7 min final extension at 72°C using a GeneAmp PCR System 9700 thermal cycler (PE Biosystems, Foster City, CA, USA). The reactions were separated by electrophoresis on a 1% TAE agarose gel. To evaluate the effects of the three gel-loading buffers on target amplification efficiency, six reactions were performed in parallel, using 2 μL of nondiluted lysate and lysate dilutions of 1:100, 1:1000, 1:5000, 1:10 000 and 1:20 000 with water. The DNA construct used, pA14, contains a 359 bp insert corresponding to the 3′ untranslated region from an aluminum-induced maize gene (our unpublished results) cloned in pGEM-T Easy (Promega, Madison, WI, USA). Figure 1 shows that colony PCR performed without gel-loading buffers allowed abundant target amplification even when the bacterial lysate was diluted to 1:1000. The overall amplification in the presence of 0.25× of buffers I, II or III was similar to the control reaction (no loading buffer in the PCR mixture). At a 5000-fold dilution, a faint band was still observed in both the control and in the samples containing any of the gel-loading buffers. With a 0.5× final concentration of any of the three buffers, the target was still amplified in those samples starting from non-diluted bacterial lysate. However, in the samples amplified with buffer I, the decrease in the amplification efficiency was higher. The effect of the 0.5× buffer concentration was clearly seen in the samples amplified from the lysate dilutions because at a 100-fold dilution, a faint band could be observed only with buffer III. At higher dilutions, no amplification could be detected with any of the three gel-loading buffers. To verify the performance of the loading buffers on the amplification of larger inserts, a PCR was performed using the lysate obtained from colonies transformed with p715 (4). This construct contains a 3046 bp insert that corresponds to the maize HRGP promoter fused to the GUS coding region Benchmarks