In the course of screening the maize genomic library for the work described above, a troublesome phenomenon was encountered which may be of interest to others attempting to isolate maize genomic cognates using heterologous cDNA clones. Initial screening of the library was performed at low stringency with nick-translated fragments containing entire Lilium EMPR cDNA inserts, including the oligo-dG/oligo-dC tails created during the cloning procedure, or nick-translated holo-plasmid DNA. Both approaches resulted in the isolation of a large number of positively-hybridizing phage which after plaque purification showed approximately the same signal strengths under either low (50% Formamide, 5X SSPE, 0.2% SDS-30 C), intermediate or high (same salts, 37 C or 42 C) stringency probe hybridization conditions. Southern blot analysis of restriction enzyme digested phage DNAs showed that this homology could be mapped to specific DNA fragments from the cloned maize DNA inserts of these phage. Upon further examination, however, it was found that all homology to the lily cDNAs vanished when these clones were probed with internal segments from the cDNA inserts, even though these internal segments contained the most conserved portions of the cDNA clones. Screening with these internal segments identified instead a much smaller subset of maize clones showing plaque cross-reaction only at lower stringencies, though the signal seen under the 30 C condition was somewhat stronger than that seen for the clones described above. When a Southern blot comparison was performed with equivalent amounts of purified, restricted DNA for such a positive and some of the clones described above at the lower stringency, the homologous maize segment from the second type of positive gave a much stronger signal even with whole insert or holo-plasmid probe, while it was the only fragment which showed a signal when the blot was probed with a fragment from within a lily EMPR insert. These observations suggested that the first type of positive clone was due to annealing of the oligo-dG/oligo-dC tracks of whole cloned cDNA inserts or holo-plasmids to oligo-nucleotide tracks in the maize genomic clones. This has since been confirmed by probing Southern blots with end-labelled synthetic oligo-dG, which anneals specifically to the same specious positive bands as the plasmid probe.
These observations indicate that care should be exercised when screening maize genomic libraries with heterologous cDNA clones, due to the existence of genomic sequences with homology to the oligo-nucleotide tails created during the widely-used procedure for inserting cDNAs in the PstI site of pBR322 and other vectors. The length of oligo-dG/dC necessary to obtain such specious signals is not great: sequencing has shown that my usual lily probe has a tail of only 16 Gs, while the synthetic oligo-dG probe contained oligo-dG pieces from 12 to 18 nucleotides in length. Fidelity of the specious hybridization is quite good; in the present case signals were obtained at stringencies where true positives would not cross-react. It should be noted that the relative intensities of Benton plaque signals can be quite deceptive, particularly at intermediate stringencies where the cross-reaction of real but diverged homologous regions is marginal. For all these reasons, it is advisable to use fragments derived from the internal portion of cloned heterologous cDNA inserts if these are to be used to screen maize genomic clones.
Robert A. Bouchard and Michael McKeown
Return to the MNL 60 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page