Developmental changes in the methylation of P-pr
--O. Prem Das and Joachim Messing

A new allele of the P locus, P-pr, has been isolated that is associated with a change in methylation of the P gene (preceding note, this issue). Molecular analysis of leaf DNA from 126 progeny segregating for P-pr and P-ww had identified 90 individuals carrying P-pr. These 90 samples displayed a wide range in the relative intensities of the higher MW bands corresponding to full, partial and basal methylation defined by SalI digests. Basal methylation (represented by bands of 1.2, 3 and 3.4kb) is equivalent to P-rr, partial methylation (bands of 4.2 and 4.6kb) arises from methylation of two SalI sites in cognate positions in the two pairs of repeats detected by probe 15, and full methylation gives two bands of >10kb (see Fig. 2 of previous note for details). Variation ranged from predominantly full methylation (low levels of partial, and undetectable basal bands) to predominantly basal (low levels of partial, and undetectable full methylation). However, all leaf DNA samples were chimeric, i.e. contained more than one set of bands.

Chimeric methylation in a tissue may result either from an increase or a decrease in methylation during development. Immature embryos and endosperms showed predominantly full methylation, and the bands corresponding to basal and partial methylation showed a modest increase with kernel development. Mature embryos and endosperms showed a further increase. This suggested a decrease rather than an increase in methylation with development. All tissues of mature plants, including brace roots, cob, culm, husks, pericarp, silks, tassel stems and tassel glumes, were chimeric for methylation. Data from sections of leaf and silk tissue were also consistent with a decrease in methylation during development. Silks and brace roots showed more methylation than tissues of earlier origin, suggesting that methylation did not decrease uniformly with time or with ongoing cell division. These changes may resemble the developmental changes in the activity and methylation of Spm (e.g. Banks et al., Genes Dev. 2:1364, 1988; Fedoroff, N and Banks, Genetics 120:559, 1988) and Mu (e.g. Chandler, V and Walbot, V, PNAS 83:1767, 1986; Martienssen, R et al., Genes Dev. 4:331, 1990). However, we detect a decrease in methylation, rather than the increase seen with transposons.

Animal and plant genes whose activity correlates with methylation are usually demethylated in expressing tissue. In contrast, demethylation in P-pr occurs in all tissues, despite a phenotype confined to floral tissues and pericarp. Demethylation in leaf DNA (ratio of partial and basal methylation to full methylation) correlated qualitatively with pigmentation in pericarp. Furthermore, DNA from heavily pigmented pericarp of mature kernels was less methylated than from lightly pigmented kernels. This was true in comparisons among ears, and even between sectors differing in pigmentation on a single ear. These correlations suggest a role for demethylation in the patterned phenotype of P-pr. Preliminary data show reduced P transcripts in pericarps of plants carrying P-pr compared to P-rr.

The origin and properties of P-pr can be accounted for by the following model. P-pr may have originated in a somatic event which changed the methylation state of the P-rr gene. This state was transmitted through meiosis at least twice, in the two inceptions of this allele. The properties of this allele are such that after fertilization, it is gradually converted to a demethylated state through development. This occurs in both endosperm and plant tissues; in the latter, demethylation appears to be controlled independently in individual organs. Expression in clones of pericarp cells carrying a demethylated gene and not in those carrying a methylated gene may account for the variegated phenotype. Demethylation may occur after the separation of gamete precursors from other cells, enabling normal transmission of the methylated state. 


Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page