Be careful with the primers when screening your clones by polymerase chain reaction.

Polymerase chain reaction (PCR) is a powerful tool with many different applications in Molecular Biology and related disciplines. Amplification of the inserts cloned into a vector directly from the bacterial colony is one of them. This technique allows a fast screening of many different clones in order to detect which one of them contains the desired construction. The primers used to direct this amplification process usually are designed to anneal to sequences flanking the polylinker of the vector. This approach allows the screening of different constructions regardless of the sequence of the DNA inserts, with the only exception of constructions containing very long DNA fragments. However, occasionally, mainly when using vectors that are not currently employed in the laboratory and due to the absence of specific primers flanking the polylinker of these vectors, it is tempting to use primers complementary to sequences located in the insert such as, for example, those used for PCR amplification of the DNA to be cloned. In this paper we show that the use of insertspecific primers instead of primers hybridizing to the regions flanking the polylinker in the vector may generate many false positive results due to direct amplification of noncloned DNA inserted in the competent bacteria. In consequence, a combination of vector-specific and insert-specific primers is strongly recommended to determine, in a single experiment, both the presence of the insert into the vector and its orientation. Fig. 1 shows the results obtained after direct PCR amplification of the DNA fragments presumably cloned in pX63 vector from transformed Escherichia coli DH5a bacteria. DNA was amplified under standard conditions (total volume of 50ll containing 0.25mM dNTPs, 2.5mM MgCl2, 16mM ðNH4Þ2SO4, 67mM Tris–Cl, pH 8.8, 0.01% Tween 20, 2.5U EcoTaq–Ecogen-, and 100 pmol oligonucleotide primers). DNA from the colony was added to the PCR tubes directly by gentle agitation of a plastic tip that had previously been used to take the colony from the LB-agar plate. Our desired constructions should contain the coding regions of three apoptosis related genes—Egl 1 [1], Bcl XL [2], and Hrk [3]—inserted into pX63 Neo, a Leishmania expression vector [4]. Four ampicillin-resistant E. coli DH5-a colonies were selected for each one of the constructions and the presence or absence of the insert was determined by PCR using SP6 and T7 primers. All the DNA fragments to be cloned contained sequences complementary to these primers at their ends. According to the results obtained, all the selected clones included the desired DNA sequences into the vector (Fig. 1A). However, the analysis of the plasmid DNA extracted from the bacterial cell cultures revealed that none of them contained the inserts, as shown by the presence of an unique DNA fragment, identical in all cases to the intact vector, after HindIII digestion (Fig. 1B). The appropriate controls were included to check that the cultures were fully ampicilline selected prior to DNA preparation (data not shown). Apparently, the DNA fragments to be cloned were lost during bacterial growth. In order to check this hypothesis, single colonies that were shown to be positive clones according to the first analysis were extended over a second ampicillin-containing LB plate and a single colony from the overnight culture was analyzed again under the same conditions as the original colonies. No specific amplified band was observed in any of the selected clones, indicating that the DNA fragments were present in the original colony but were lost during bacterial growth (data not shown). Among the several different explanations that we could find for this behavior, the most feasible one was that the DNA fragments that were amplified in the original colony were not really cloned into the vector Analytical Biochemistry 308 (2002) 189–191