We have been investigating the relation between quantitative trait loci (QTL) detected by RFLP mapping and loci defined by alleles with qualitative effects (mutants). As proposed by Robertson, DS (J. Theor. Biol. 117:1-10, 1985) qualitative and quantitative genetic variation may trace to the same loci. At a locus, it may be possible to recover alleles conferring a nearly continuous range of effects on the phenotype of interest. Alleles with extreme effects (possibly defective products or nulls) are easily recognized by their mutant phenotypes (dwarf plants, defective kernels, pigmentation, etc.). In contrast, alleles with more subtle effects (quantitative) would be more difficult to assess and characterize. The proposed relation between quantitative and qualitative variation may seem intuitively obvious to some; however, direct tests of the hypothesis have been difficult to design and conduct. We have attempted to conduct such tests by using clones of functionally defined genes as RFLP probes in mapping experiments of quantitative genetic variation. As the various maize maps (RFLP, mutant, and cytological) are merged and more clones of defined function become available, rigorous review of the presumed relationship should be possible. If the relationship is generally true, information from quantitative and qualitative approaches to mapping and genetic analysis should be complementary on the basis of their relative strengths for identifying interesting regions (via QTL mapping) and defining candidate loci (via mutants). In this note, we briefly report two cases supporting Robertson's hypothesis.

In the first case we used a genomic probe at or tightly linked to an1 as a marker in the analysis of genetic variation for plant height. The probe was provided by S. Briggs, S, Pioneer Hi-Bred Intl. The an1 locus is defined by a recessive, GA-responsive dwarf phenotype. The probe was mapped in a F2 population of 150 plants created by crossing inbreds Mo17 (ca. 167cm tall) and H99 (ca. 103cm). The same plants and their selfed progeny were evaluated in replicated experiments for several quantitative traits including plant height. The an1 probe was placed to the long arm of chromosome one and was closely linked to a QTL for plant height. The probe defined an interval on the genetic map which accounted for 40% of the phenotypic variation for plant height with additive effects estimated at 15cm (i.e. on average, plant height is predicted to increase by 15cm with each substitution of the "allele" from Mo17 for the "allele" from H99). We have observed similar effects on plant height by this region in other populations. Of course, there are other loci in the vicinity of the an1 QTL (e.g. D8) or other plant height QTL in other populations (Beavis et al., Theor. Appl. Genet. 83:141-145, 1991) that could contribute to variation in plant height. Determining which locus, if any, is the primary source of genetic variation in the population represents a major obstacle for direct tests of the hypothesis for most traits.

In the second case, we used the same materials and evaluated them for resistance to leaf blade feeding by the European corn borer. H99 and Mo17 are highly resistant and susceptible, respectively. The QTL with the largest contribution to resistance accounted for 17% of the variation in this population and was located (genetically) to an interval at the end of the short arm of chromosome four. This is also the vicinity of the bx1 locus. This locus is defined by a recessive phenotype lacking cyclic hydroxamates which inhibit corn borer larvae (Mo17 and H99 are +/+ at bx1). Currently, it is the only known maize locus with such effects. Therefore, alleles with quantitative effects at the bx1 locus seem to be plausible sources of genetic variation in this population.

A second putative QTL for resistance to leaf blade feeding was located to the long arm of chromosome one. The interval defining the QTL accounted for nearly 16% of the variation. To our knowledge, loci affecting resistance to European corn borer have not been described in this region. Certainly, mutant alleles with qualitative effects have not defined such loci (or molecules) in this region. In a crude way, QTL mapping studies may identify the functional significance of genetic regions in a very efficient manner. Once more efficient techniques have been developed for map-based cloning and for more thorough, directed evaluation of genomic regions, information from QTL mapping studies should provide important clues about starting positions and targets of the search. 

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