Two isolates of a new allele of the P gene, termed P-pr (for patterned pericarp, red cob), that originated by transmission of somatic epimutation of the P-rr allele were described previously (MNL 67; Das and Messing, Genetics, in press). P-pr was similar to P-rr in sequence, but was more methylated at both CG and CNG motifs. Its phenotype was characterized by variegated pigmentation in pericarp, but not cob, variability in pigmentation between siblings, and the presence of large clonal sectors that differed in pericarp pigmentation. In heterozygotes with this allele, the uniform red pigmentation conditioned by P-rr was reduced, and rendered qualitatively similar to that of P-pr in all the above respects. However, the reduction was variable, and many ears appeared fully red. To quantitatively demonstrate the reduction in pigmentation, we have used the assay described in the preceding note, and the following genetic schemes.
Two genetic schemes (Fig. 1, next note) were used to control for genetic background effects on P-pr-1. In one (Table 1, left columns), 8 plants of the genotype P-ww/P-pr, 4 sibling plants of genotype P-ww/P-ww and one P-pr/P-rr plant (all in the original background that P-pr-1 was isolated in) were crossed to P-rr in the W22 inbred background. The genotype of the resulting plants (control genotype of P-ww/P-rr or test genotype of P-pr/P-rr) was determined by segregation in the next generation, and five random kernels from each of the resulting ears were used for quantitation of pigmentation. In the second scheme (right two columns), 5 plants of the genotype P-pr/P-rr were crossed to P-rr in the W22 background, and again, the two resulting genotypes P-pr/P-rr and P-rr'/P-rr were distinguished by segregation in the next generation (P-rr' is used to designate the P-rr allele that has interacted with P-pr in the previous generation).
Relative to P-ww, P-pr reduces pigmentation of P-rr by two-fold; relative to P-rr', reduction is closer to 2.5-fold. The true value is likely intermediate, since P-pr is capable of generating, by itself, more pigmentation than P-ww and less than P-rr' (see next note). The differences between the P-pr/P-rr mean values in the two schemes may reflect the genetic background, since the second scheme uses two successive crosses to W22 before the testcross, whereas the first uses only one. However, in each scheme, factors unlinked to P should be equalized between test and control. Individual pigmentation values in each data set fit a normal distribution; when grouped by frequency in ascending intervals, they approximated the expected bell-shaped curve. Quantitative measures of the fit, obtained by comparing the number of entries within 0.5, 1, 2 and 3 standard deviations of the mean in each data set to expected values, did not differ from expectations at a confidence level of 90%.
Table 1. Reduction of pericarp pigmentation* of P-rr by P-pr.
*Expressed as OD/mg dry tissue from pericarps of 5 random kernels of
to the MNL 68 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page