Construction of a genetic linkage map in maize using restriction fragment polymorphisms

As reported at the meeting last year, we are in the process of constructing a genetic linkage map for maize using restriction fragment polymorphisms (RFPs) as our source of markers. The potential uses for such a tool have been well documented by others and ourselves and include applications to problems in basic genetics and plant improvement programs. For instance, we have already utilized them to map large duplicated areas of the maize genome to different chromosomal locations and to dissect quantitative traits into their individual genetic components. The ability of RFPs to detect variability in a wide range of lines, and hence be informative in many different crosses, may serve as a unifying mechanism to bring together genetic information of many different types such as morphological markers, isozymes, cytological data, quantitative traits, etc.

As our source of markers, we have utilized both cDNAs prepared from total leaf mRNA and unique sequence genomic clones less than 2kb in length. These were both initially screened for hybridization signal intensity, complexity of signal, and informativeness against two parents, H427 and 761, supplied by Tim Murphy of Northrup King. Those that satisfied our criteria, i.e. they consistently yielded a strong hybridization signal, were not too complex in their hybridization pattern, and differentiated the two lines used, were then analyzed for their inheritance in 50 F2 plants derived from these two lines. A number of clones for identified genes, such as shrunken-1 and waxy, were also obtained from other researchers whenever possible and similarly tested. Loci detected by these clones were examined for cosegregation by maximum likelihood analysis and placed into linkage groups. To date we have identified 117 loci detected by RFPs and have arranged them into just over 20 distinct linkage groups.

To assign these linkage groups to defined chromosomes and begin to correlate the map we are producing with that derived by other methods, we are using several approaches. Three linkage groups were assigned to chromosomes on the basis that they contained clones of genes with known genomic locations: Adh1 on chromosome 1, a1 on 3, Adh2 on 4, and sh1, bz1, and wx1 on 9. The remaining groups were assigned by analyzing individual clones by hybridization to a set of monosomics. Using the r-x1 deficiency system, we obtained monosomic plants corresponding to eight of the ten maize chromosomes, excluding only 1 and 5, and produced genomic DNA from them. By hybridizing our probes against Southern blots containing DNA from these monosomics and testing for loss of signal contributed by the female parent, we have been able to assign these loci and linkage groups to their chromosomal origins. Seventy-five of our loci were tested by this method to insure the accuracy of these assignments. An additional 38 loci, that were uninformative in our original cross, were also assigned to chromosomes by this method and are awaiting inheritance analysis in other populations. We have also tested for cosegregation of our RFP loci with known morphological markers in a Mangelsdorf backcross population. Linkage of some of our loci to bm2 on chromosome 1, su1 on 4, pr1 on 5, Y1 on 6, gl1 on 7, and g1 and R1 on 10 have yielded further information on the orientation of our map with respect to the conventional map.

The current version of our map, so derived, is shown in the accompanying figure. Chromosome designations are along the left side, loci designations along the top of the horizontal lines with map distances below the lines. Map distances in parentheses are tentative as they are probably beyond the resolution of our original analysis and need to be verified in a larger population. Loci assigned to chromosomes by monosomic analysis without linkage information are set along the right side of the figure. Symbols for known markers determined morphologically are shown in brackets, while those that were determined through the use of RFPs are shown without.

Future work will center on several areas to improve on these results as this version of the map must be viewed as work in progress. First, more markers will be accumulated and mapped to improve resolution of the maize genome and insure that most areas are covered by informative marker sets. Secondly we will utilize a larger F2 population to obtain better estimates of map distances and arrangements of closely linked loci. We are currently working with C. Stuber and M. Edwards at NCSU, to evaluate these marker sets in a different population of approximately 200 F2s which have also been characterized by isozyme analysis. This will yield better resolution as well as improve our correlation with known genomic locations by testing for cosegregation with isozyme loci. Finally we plan to improve our correlation with the conventional map by testing the inheritance of our loci with B-A translocation sets and other morphological markers to establish the presence of centromeres. Our goals then are to produce a detailed set of RFP loci that cover most of the maize genome, to further correlate these loci with the conventional map, and to produce a tester set of RFPs that can be made available to other researchers for various applications.


Tim Helentjaris, Scott Wright and Dave Weber

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

Return to the MNL 60 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page