There have been concerns about the reproducibility of banding profiles generated by RAPDs. These concerns have brought into question the reliability of the method to provide useful profile data either for genetic or varietal identification studies. With 200 primers (10mers from Operon Tech.), we have screened nine inbred lines of maize that collectively encompass a broad degree of genetic diversity for the Corn Belt. Using an extremely high stringency of selection, namely the presence of a few clear, well defined, and easily scorable bands per DNA/primer complex, and the ability to discriminate among at least 50% of the inbreds in this initial screen, we were able to identify 71 primers that are to be preferred for varietal identification in maize. This list of primers will be made available upon request.

An integral component of this screening process was to check the reproducibility of the banding profiles. DNA was extracted from each inbred line in duplicate and each DNA sample was then amplified twice giving four amplification products for each DNA/primer complex. Duplicate amplifications were either performed on the same thermocycling unit but on different days or on different thermocyclers during the same day. Visual comparisons of the profiles showed a very high degree of repeatability with only some of the faintly fluorescing bands showing presence and absence variation. The moderately to highly fluorescing bands were constant across replicates for each of the preferred primers against all maize genotypes that were included in the screening process.

One possible explanation for the sometimes observed lack of repeatability for bands that show a low level of fluorescence could be that they represent instances of inefficient priming, possibly due to a poor sequence homology between the primer and target site on the genomic DNA. These sites would then not be able to compete as effectively in comparison to other sites that have 100%, or near to 100%, sequence homology. A lack of repeatability for the faint bands could possibly be overcome, or at least reduced to a minimum, by increasing the stringency of the annealing conditions. Such a goal was a major component in the derivation of our current RAPD conditions for maize (see separate report by Chin and Smith, this Newsletter).

Once an ability to repeatably generate RAPD profiles has been established, it is then necessary to evaluate that capability for practical and routine purposes of varietal identification. In order to achieve that goal, it will be necessary to arrive at some objective criteria for the scoring and databasing of those profiles and to more thoroughly evaluate the repeatability of profile generation. Therefore, we are now directing the focus of our research into more objectively measuring the repeatability of molecular weights and fluorescent intensities generated by RAPDs. DNA extractions and amplifications have each been performed in duplicate by two laboratory personnel. RAPD profiles have been scanned by digital camera and recorded using Bioimage software and hardware. 

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage