--Andrew Harrison and Jerry Kermicle

Analysis of testcross data from the summer of 1987 has revealed that the R-g:15 allele is unusually sensitive to the paramutagenic action of R-st. A noncrossover derivative of R-g:15 governs anthocyanin production in the aleurone but not the various plant tissues pigmented by R-r:std (coleoptile, roots and anthers). The allele was included in a study designed to compare the behavior of several groups of paramutable alleles following different paramutation treatments.

Association with R-st in a heterozygote heritably alters the expression of many R alleles at high frequency. Brink called this allelic interaction paramutation and described those alleles susceptible to change as paramutable and those inciting change as paramutagenic (Annu. Rev. Genet. 7:129-152, 1973). When extracted from the heterozygote by testcrossing onto a W23 r-g tester (r-g is a null allele of R that is neither paramutable nor paramutagenic), paramutable R alleles that normally give a phenotype of darkly mottled aleurone in single dose exhibit reduced capacity for such pigmentation.

125-kernel samples having the genotype of interest were removed from testcross ears and prepared for analysis in the Agtron M-31-A, a device that records the spectral reflectance of a sample on a scale of 0-100 relative to calibration standards. Higher numbers denote greater reflectivity and thus lighter sample color. The average reading from four measurements was designated the Agtron score for a given sample; each Agtron score was converted to a quantity designated % full color (% fc.) to facilitate subsequent comparisons. Family averages for the r-g:Stadler and R-sc:n656 alleles constituted the colorless and full colored controls, respectively.

R-r:std is very sensitive to the paramutagenic action of R-st and thus serves as a convenient reference allele. Although R-r:std and R-g:15 exhibit similar levels of single dose nonparamutant pigmentation (mean % fc. scores of 86.9 and 83.8, respectively), the pigmenting potential of R-g:15 is driven essentially to zero by a single generation of heterozygosity with R-st. R-r:std, on the other hand, retains about 17% of nonparamutant pigmenting capacity after the same treatment (see Table 1). This is in sharp contrast to the results of Brown (Genetics 54:899-910, 1966), who observed no significant differences among 9 independently isolated R-g derivatives and R-r:std with respect to paramutational sensitivity.

Table 1. Testcross data.

The negative values listed for paramutant R-g:15 testcross families deserve comment. The Agtron scores for R-g:15 kernel samples were higher on average than those from the r-g:Stadler control family, and therefore the corresponding % fc. scores were negative. Why this should be the case is not known, perhaps variation in the level of carotenoid pigmentation contributes to the difference. Visual inspection of kernel samples from R-g:15 testcross ears revealed that most kernels retained a few small flecks of anthocyanin pigmentation. The Agtron is relatively insensitive to small differences in weakly pigmented samples, and evidently this very light mottling passed undetected. In any event the mean % fc. score for R-g:15 did not differ significantly from that of the colorless control.

Pairing a sensitive allele (paramutant or nonparamutant form) with an r-r or r-g allele tends to enhance the pigmenting action of that allele relative to controls maintained as homozygotes (Styles and Brink, Genetics 61:801-811, 1969). Paramutant forms of the R-r:std and R-g:15 alleles were compared with respect to this property of reversion. Both alleles were found to regain about 50% of the pigmenting potential lost in a prior generation of heterozygosity with R-st by experiencing a generation of pairing with r-g:Stadler (see Table 1). By this criterion, the mutational event that generated R-g:15 apparently increased its paramutational sensitivity without affecting its capacity for reversion. In contrast, after two generations of heterozygosity with R-st, the R-r:std allele (exhibiting a mean % fc. score not significantly different from that of R-g:15 following one generation of exposure to R-st action) was only restored by approximately 12% with similar treatment. R-g:15 has yet to be tested in this manner. The contrasting reversion behaviors of R-g:15 and R-r:std alleles reduced to similar levels of pigmenting action by interaction with R-st might reflect structural differences between the two alleles, or perhaps the effect of increasing generations of exposure to R-st is generally to reduce the capacity of a paramutant allele to revert towards a higher level of expression.

Return to the MNL 63 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage