Reproducibility of the differential amplification between leaf and root DNAs in soybean revealed by RAPD markers

 Random amplified polymorphic DNA (RAPD) was used to determine whether such markers can be employed for detecting genomic modification during plant development or under certain stress environments. Pairwise comparisons in RAPD patterns of leaf and root DNA amplifications were studied for 11 soybean accessions representing different origins. Hydroponic culture was used for the ease of harvesting roots. From a total of 40 primers screened, it was found that 16 can detect leaf DNA polymorphism, 19 for root DNA polymorphism, while 10 show a greater consistency for detecting polymorphism between leaf and root (L/R) DNAs. Nevertheless, problems were encountered when the newly synthesized oligo-primers and different thermal cyclers were used to check the data. Several factors were then tested for their reproducibility. The results indicated that the amplified differences between root and leaf DNAs are mostly not affected by template DNA concentrations. The addition of DMSO (dimethyl sulphoxide) or TMAC (tetramethyl-ammonium chloride) also did not mask the L/R differences. However, DNA polymerase and oligo-primers synthesized from different manufacturers, as well as the thermal cyclers, reacted differently sometimes. Regardless of the general problems of reproducibility in RAPD patterns, some amplified differences remain between the L/R DNAs. The most distinct patterns involve differences in the relative intensity of amplified bands. Differential amplification might have occurred during plant leaf and root development. Southern hybridization of the eluted polymorphic bands against restriction digestion of total genomic DNA confirms their being homologous to soybean DNA fragments. Polymorphism of these specific L/R differences also exists among varieties. RAPD should be a useful tool in detecting genomic alterations during plant development or under certain stress environments, as long as the factors affecting the reproducibility of RAPD patterns can be properly controlled. An additional cycle of selection would be possible if such a type of polymorphism is proved to be correlated with certain developmental characters. Received: 7 October 1996 / Accepted: 20 May 1997