Last year I suggested that some quantitative genes might be alleles of qualitative mutants and that it might be possible to isolate genic DNA for individual quantitative genes (MGCNL 58:10-11, 1984). A more detailed description of this proposed relationship has been submitted for publication.

I would like to suggest here a possible way of testing this proposal. Seed size is probably the trait most amenable to study at the present time since it is an important quantitative trait and there are numerous qualitative mutants that affect seed size. These mutants range from small seeds to those with completely empty pericarps. Neuffer and Sheridan (Genetics 95:929-960, 1980) have reported that 27% of EMS-induced mutants were recessive kernel mutants, and a large proportion of these fell into the defective category. We have found in our Mutator stocks that kernel mutants are one of the most prevalent seed mutants. By using a transposable DNA system like Mu or Ac, etc. to induce qualitative seed-size mutants, it should be possible to locate the genic DNA for many of them. The DNA from these in turn can be used to locate the wildtype alleles (quantitative genes?). If the wildtype alleles are indeed responsible for quantitative variation, then corn varieties that differ markedly in seed size might be expected to have different wildtype alleles at some of these loci, which possibly could be distinguished at the molecular level. In early tests it might be wise to study strains that show extreme differences in seed size, such as Argentine, Strawberry or other small seeded popcorns and a present day dent inbred or maybe even a large-seeded variety, such as Cuzco.

Once the genic DNAs are isolated, a variety of molecular investigations can be initiated. Differences in the DNAs in these diverse lines can be studied by restriction endonuclease mapping and DNA sequencing. Differences in gene expression and regulation also can be studied through the analysis of DNA transcripts, and eventually even differences in the final gene products could be investigated. If indeed a qualitative defective mutant gene is the allele of a wildtype quantitative gene, isolation of the quantitative (wildtype) alleles at this locus from lines differing in seed size should yield genes or gene products, in some instances at least, that differ in measurable ways.

There must be many gene loci other than those more or less directly responsible for seed size (i.e., the de loci) that also are involved in the expression of this trait in less direct ways--for example, genes involved in such traits as time of maturity, number of rows of seeds per ear, number of ears per plant, etc. Thus, a study of the defective mutants will not provide the complete picture but has the potential for testing the original hypothesis and, if it proves valid, providing information and material that might be useful in a breeding program.

Donald S. Robertson

Return to the MNL 59 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage