Use of PCR and direct sequencing to compare DNA sequences from homologous regions for several cultivars --Tim Helentjaris, Donna Shattuck-Eidens and Russell Bell We have had the opportunity to produce genetic linkage maps based upon restriction fragment length polymorphisms (RFLPs) in several species. We are intrigued by the relatively high rate of polymorphism exhibited in maize and Brassica when compared to other species such as tomato and melon. Virtually any unique sequence clone in maize can reveal multiple polymorphisms amongst a selection of domesticated maize inbreds using only a few restriction enzymes. In tomato the occurrence of RFLPs is much lower; less than 5% of the clones tested reveal any polymorphism and many more enzymes were tested. Since RFLP analysis simply reflects the DNA sequence variation between individuals, we reasoned that an examination at the nucleotide sequence level of species exhibiting RFLPs at differing rates, would yield a better explanation of this difference. Previously available technologies made it impractical to economically approach this problem.

With the development of the polymerase chain reaction (PCR, Saiki et al., Science 230:1350-1354, 1985), it has become possible to amplify homologous regions from many different individuals and sequence them directly to compare any DNA sequence variation in the amplified regions. Inserts from clones used previously to detect RFLPs amongst maize cultivars (Helentjaris, Trends Genet. 3:217-221, 1987) were sequenced completely. Pairs of oligonucleotide primers were synthesized from each end of the inserts, approximately 500 to 1000bp apart, so that their 5' to 3' orientations were directed towards each other. Using standard conditions from the original papers, these primers were used to amplify homologous regions from 7 to 9 maize cultivars. The PCR products were then purified from agarose gels and sequenced using a Sequenase kit.

Some sample data for maize are shown in the accompanying table. It can be seen for these two regions that there are multiple DNA sequence variations between these cultivars. We estimate from our current data that there are 30 to 50 alterations per 1000bp. Both base pair changes and sequence rearrangements of various sizes are seen. Interestingly, a tandem duplication of 17bp in the 451 region of group 4 (underlined) is followed by a very similar 8bp tandem duplication less than 200bp away. In the group 5 lines neither of these sequences is duplicated. A large insertion of several hundred base pairs was seen in the 288 region in B14A (not shown) but was not found to be related to any of the transposable element sequences in the Genbank database. We are currently expanding both the amount of sequence information on specific regions and the number of individuals evaluated.

Using this type of analysis, we feel that it will now be much more practical to compare homologous regions of numerous individuals at the DNA sequence level. It is also interesting that in melon we have found by similar analysis that the rate of sequence variation is about tenfold less than that seen for maize. Since short tandem duplications and rearrangements can be generated by transposable element activity, it is tempting to speculate that the high level of polymorphism observed in maize is the result of transposable element activity. To date no natural transposable elements have been identified in tomato and melon, and both exhibit little polymorphism. It should also be noted that the species which exhibit higher levels of RFLP variation reproduce primarily by outcrossing whereas the less polymorphic species are primarily self-pollinated. The different modes of reproduction may affect the level of transposable element activity itself or it may be that a high mutation frequency resulting from transposable element activity or any other mechanism can not be tolerated in a self-pollinated species and is therefore selected against. At this point we remain at the speculation stage but the PCR method of analyzing sequences will allow us to examine the question "Why do different species exhibit different levels of RFLP variation?".

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 63 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page