Distinct combination of purification methods dramatically improves cohesive-end subcloning of PCR products.

For most purposes, cloning protocols based on the polymerase chain reaction (PCR) using either the template-independent terminal transferase activity of Taq DNA polymerase in TA cloning schemes (8,9) or blunt-endbased procedures involving Pwo or Pfu DNA polymerase or polished Taq-generated PCR products are quite satisfactory. Nevertheless, both techniques have been proven less straightforward than anticipated and yield low numbers of recombinant clones. To remedy this obvious disadvantage, many ligationindependent cloning schemes have been introduced (1,10). In contrast, subcloning of PCR products from genomic or plasmid DNA with restriction sites of the scientist’s choice is often desired. To use cohesive-end cloning schemes, oligonucleotide primers, which exhibit convenient restriction endonuclease recognition sequences, are commonly used. Despite many improvements made in the past, many laboratories describe this approach as difficult, usually failing or inefficient for unknown reasons (1,5,10), even with the recognition sites placed at a sufficient distance from the end of the PCR primer [because many restriction endonucleases fail to cleave when their recognition sequences are located within a few bases of the end of a DNA fragment (6)]. As a consequence, most researchers complain that no reliable protocol is available, especially when using high-fidelity proofreading enzymes or enzyme mixtures for amplification. Recently, evidence was provided that remnants of Taq DNA polymerase surviving successive purification routines in a fully active state [e.g., phenol/chloroform extraction, ethanol precipitation, gel electrophoresis, and column purification (2,3)] are a major cause of low cloning efficiency. Also, because significant numbers of dNTPs survive such treatments, residual DNA polymerase efficiently counteracts subsequent restriction enzyme digestions; i.e., it continuously fills up 5′ staggered ends (2). In our hands, the clonability of PCR products was not significantly improved by modified procedures suggested to eliminate all carryover traces of DNA polymerase, primers and dNTPs before restriction endonuclease digestion, such as removal of polymerase by digestion with proteinase K (3), precipitation of PCR products with polyethylene glycol (PEG) and high salt (2,7) or clean-up of PCRs with widely available column purification kits from different commercial providers (Figure 1). We present a novel protocol for sticky-end cloning of PCR products using a combination of purification steps that reliably eliminate all carryover traces of DNA polymerase, interfering dNTPs, primers and salts, thus markedly improving the cloning efficiency of all PCR products tested so far. The following protocol exemplifies ready-to-go purification for a 3253-bp fragment containing the nls-lacZ reporter gene [nuclear localization sequence linked E. coli lacZ gene (4)] with restriction endonuclease sites of the researcher’s choice. Amplification reactions were carried out on a GeneAmp System 9600 Thermal Cycler (Perkin-Elmer, Weiterstadt, Germany) in a total volume of 100 μL consisting of 50 mM KCl, 10 mM TrisHCl (pH 8.6), 2 mM MgCl2 in the presence of 200 μM each dNTP, 100 nM of forward and reverse primer, 0.1–10 ng of template DNA (ca. 106 copies of target DNA) and 3.5 U of Expand High Fidelity DNA Polymerase (Boehringer Mannheim GmbH, Mannheim, Germany). Each 30-nucleotide (nt) primer contained 10 nt of noncomplementary sequence at the 5′ end consisting of a 6-nt endonuclease restriction site and a 4-nt clamp added to improve endonuclease activity in subsequent PCR product digestion. Samples were initially heated to 94°C for 1 min to denature the template DNA. PCR was carried out for 30 cycles, each at 94°C for 15 s, 61°C for 15 s and 72°C for 2.5 min, followed by a final extension step for 5 min at 72°C.