High-efficiency human B-cell cloning using hygromycin B-resistant feeder cells.

efficiency was not affected by DNA source or PCR condition. Blue colonies contained a blunt-end-ligated plasmid without insert. Twenty randomly chosen white colonies were sequenced. Nineteen of them contained inserts ended with a primer and an A/T pair formed by complementary overhangs, and one contained an abnormally formed blank plasmid that was missing a part of the original sequence. The number of recombinant clones generated by our vector and by the Original TA Cloning Kit (Invitrogen) was about the same. Our observations are quite different from those reported by Mead et al. (4). These differences could be a result of different quality of the enzymes used, different stability of the 3′ overhangs or different efficiency of the complementary 3′ adenylate extension of the PCRgenerated fragment. Detailed analysis of the results presented by Mead et al. (4) indicates that the quality of the XcmI enzyme was likely the main reason for the low cloning efficiency observed for the vector. The authors mentioned the poor quality of the enzyme they used, and also noted that 15 of 18 junctions they had analyzed contained unexpected DNA fusion. In all cases, the boundary thymidine nucleotide was missing. The vector showed 20.2% efficiency of self-ligation and 68.1% efficiency of a blunt-end cloning, whereas the PCR-generated fragments were cloned with efficiency not higher than 15%. For the HphI-based vector tested at the same time, these numbers were 3.9%, 1.0% and up to 32.7%, respectively. Our results rehabilitate the use of XcmI-based vectors for cloning PCR products. We suggest that the low performance shown by Mead et al. (4) for the XcmI-based vector was a result of poor quality of the enzyme used, and we conclude that a highly efficient vector for cloning Taq DNA polymerasegenerated PCR products can be prepared from the plasmid described here by its single-step digestion with XcmI.