I have found that many plant breeders feel that modern genetic engineering techniques will be useful for locating genic DNA for qualitative traits, such as disease resistance. However, they do not foresee much hope of isolating genic DNA for the genes controlling quantitative traits. Because the expression of quantitative traits involves many genes, each contributing only a small effect to the total phenotype, they feel it will be impossible to screen for the effect of a single quantitative gene. If quantitative genes are in a class by themselves and are not related to those with which the geneticist deals (i.e., qualitative genes), this might be true. To me, however, it seems unreasonable (to say nothing of wasteful) to assume a plant has two sets of loci, one for quantitative traits and one for qualitative traits, when one set could account for both patterns.
Let us, for instance, consider a trait like plant height. If one tests the inheritance of plant height in crosses between a tall and a short genotype, it would be found that plant height is indeed a quantitatively inherited trait. The geneticist, however, knows of many loci with qualitative mutants for plant height (e.g., dwarfs, brachytics, etc.). The numbers of such loci approach those suggested for the number of quantitative loci assumed. I would like to suggest that the qualitative mutants of the geneticist may be just null or near null alleles of the quantitative loci of the plant breeders. Molecular geneticists have shown that at the DNA level there is probably no one wild type allele (Johns, Strommer and Freeling, Genetics 105:733-743, 1983; Burr, Evola, Burr and Beckman, Genetic Engineering: Principles and Methods, Vol. 5, 1983). For those loci in which the DNA has been isolated, it has been determined that there is extensive polymorphism at the base pair level in the wild type alleles. All of these alleles produce active gene products resulting in what is recognized as wild types. Undoubtedly, there is variation in the relative efficiency of the gene products of these wild type alleles. If two inbred lines have different sets of wild type alleles producing functional products with different efficiencies, it seems reasonable to expect a quantitative pattern of inheritance. Thus, variation at the base pair level that does not eliminate the production of a functional gene product produces quantitative alleles. However, a mutation at one of these loci resulting in a gene product that has lost all or most of its function will result in a qualitative mutant.
If this suggested relationship is real between qualitative and quantitative genes, it would be possible to isolate the DNA for quantitative genes. If one uses Mutator or a similar system to induce a dwarf mutant, one would potentially be able to isolate the DNA for the wild type allele of the dwarf locus and, hence, the DNA for a quantitative gene for plant height.
Similar experiments could be done for genes involved in yield. Seed size is one important component of yield. Possibly the many defective seed mutants are qualitative alleles of loci responsible for the quantitative inheritance of seed size. Perhaps the photosynthetic mutants (hcf) are qualitative mutants for quantitative genes controlling photosynthetic efficiency.
This is only a brief description of a hypothesized concept of the relationship between quantitative and qualitative genes and some of the implications of this model. There are many other aspects of this relationship, which have not been considered here, but will be incorporated in a more in-depth paper that is being prepared.
Donald S. Robertson
Return to the MNL 58 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page