A set of core RFLP markers for maize

--Jack Gardiner, E.H. Coe, Susan Melia-Hancock, D.A. Hoisington, and Shiaoman Chao

Maize researchers are fortunate in having not one but several well developed RFLP linkage maps. Many of these maps have numerous markers in common, which allows an approximate correlation of the various public and private sector maps. Still there is not a complete correlation between maps and choosing the proper RFLP probes to flank a trait of interest is often a trial and error process dependent upon finding a suitable probe-enzyme combination. Clearly, a standardized set of chosen RFLP markers would be a useful starting point for any maize researcher wanting to map a particular trait. In this article, we would like to propose a set of core markers that define a series of bins and a core map.

To establish materials for a core map, an immortalized Tx303/CO159 F2 population consisting of 56 individuals was developed. This was accomplished by planting 40 F3 seeds from each of 56 F2 ears in two 20-plant rows, each planted with half delayed 7 days. Pollen from 5 F3 plants in one row was bulked in one pollination cycle and used to pollinate 5 F3 plants in the other row. Using this approach for two or more pollination cycles, as many as possible of the 40 F2 ears were pollinated. This procedure allows the heterozygosity of the 56 individual F2 ears to be maintained, minimizing drift and selection, while at the same time allowing the production of large amounts of seed. A minimum of 20 ears were bulked in order to recapture the heterozygous constitution of the F2. Using this procedure only 40 of the 56 F2 families are complete, due to localized poor stands in 1990. The remaining 16 families will be completed in the summer of 1991. Limited distribution of seed will be available to anyone interested in doing their own RFLP mapping. All that one need do to recapture the heterozygosity of the F2 mapping population is to plant out 10 or more seeds from each family - by bulking samples at the leaf collection stage it is only necessary to isolate 56 DNA samples.

A complete UMC RFLP map consisting of 187 markers was generated on a SUN 4.0 workstation using Mapmaker V. 2.0, the genetic mapping program developed by Eric Lander that uses maximum likelihood equations to generate a multipoint map. A maximum allowable recombination fraction of 0.4 and a LOD score of 3.0 were used.

In the present immortalized Tx/Co F2 population, of the 187 RFLP and isozyme markers that have been mapped, 94 were selected as core markers (please see the accompanying map). The selection criteria were as follows. Ideally, a core marker should be a single copy probe which gives an easily interpretable pattern and is polymorphic across a variety of inbred lines. For the most part, core markers fulfill this requirement but a few represent the only probe available in a sparsely marked region of the genome. Secondly, a strong preference has been given to those probes that are widely used and are available to researchers in both the public and private sectors. Finally, core markers were chosen to be spaced no more than 30cM (22% recombination) apart, with an optimal spacing of 15-20cM. In 13 cases, lack of a suitable core marker that fit the above criteria necessitated regions of greater than 30cM to be defined between a pair of core markers. It is anticipated that over the next year these regions will be subdivided by core markers, which are in the process of being determined.

Even with our best refinement of the data, a few orders must be considered uncertain because of low LOD scores (i.e., data are not definitive). Specifically the end of chromosome 1L is uncertain and we have chosen the order on the basis of previous work and consensus with other maps.

The designation of 94 core markers in addition to the 13 to be designated allows the maize genome to be divided into a series of 116 "bins". We appreciate the suggestions and encouragement of Tim Helentjaris toward this approach. Each bin is defined by a pair of RFLP markers. For example, bin 1.01 is defined by BNL5.62 and UMC157. Bins are numbered according to the chromosome on which they reside, with further subdivisions following the decimal point being created by numbering sequentially from the most distal short arm core marker to the most distal long arm core marker. As stated above, 13 regions spanned a distance greater than 30cM. These bins will be subdivided upon finding a suitable core marker that bisects this region. This bin system has the advantage that any genetic trait can be localized to a small region of the genome by using a limited number of agreed-upon markers. Once localized to a particular bin, the bin number serves as a computer sortable tag which will allow all markers in the same bin to be identified, should more precise map order within a bin be desired. The other advantage of a bin system for collection, organization and retrieval of genetic information is that it is expandable and allows bins to be infinitely subdividable. For example, bin 6.10, defined by UMC62 and UMC134, can be subdivided into 9 computer sortable sub-bins by appending 0.1 - 0.9 to 6.10 to define bins 6.10.1 - 6.10.9. In the future this will become a necessity as more RFLP probes are mapped.

We look forward to receiving comments or suggestions on the Core Map and the utility of ýbinsţ.


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

Return to the MNL 65 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page