The pl gene undergoes paramutation
--Garth I. Patterson, Etienne Kaszás, Karen Cone, E. H. Coe, Jr. and Vicki Chandler

The expression of the biosynthetic genes of the anthocyanin pathway in maize is regulated by the genes pl, c1, b, and r. Two of these genes, r and b, are known to undergo paramutation. Paramutation can be simply defined as follows: one allele of a gene heritably alters the other allele when the two alleles are present in a heterozygote. For example, in b paramutation, when B' and B-I are present in a heterozygote, the B' allele changes B-I into B', so that only B' alleles are transmitted. In this report, we describe the isolation and characterization of an allele of pl that causes paramutation. This allele, which we call Pl'-mahogany (Pl'-mah), changes the Pl-rhoades (Pl-rh) allele into Pl'-mah, so that only Pl'-mah alleles are transmitted from a Pl'-mah/Pl-rh plant. (A note on nomenclature: Pl-rhoades is the standard Pl+ allele in our stocks. The phenotype produced by this allele is not unusual--according to the old nomenclature, this allele would simply be known as Pl).

An unusual variant was isolated in a family of genotype B-I/B-I, R-r/R-r, Pl-rh/Pl-rh (W23 background). The expected phenotype for this genotype is intense color in sheath, culm, and husk due to the action of B-I and Pl-rh, and intense purple color in the anthers due to the action of R-r and Pl-rh. The unusual variant had color in all of these tissues, but was different in two ways. First, the variant had less total pigment in each tissue. Second, the formation of pigment was more "sun-red", that is, the formation of pigment was substantially less in tissues not exposed to the sun. For example, inner husks of the variant were mostly green, whereas inner husks of the normal plant had substantial purple color. We call this variant phenotype "mahogany". This isolation is not a unique event; plants with a similar phenotype are seen occasionally in our B-I/B-I, R-r/R-r, Pl-rh/Pl-rh (W23 background) stock. We have not carefully measured the frequency, but we estimate that a few percent of these plants have this phenotype.

Initial crosses indicated that one of the genes in this stock was undergoing paramutation. The variant plant was thought to be mutant for either B-I, R-r, or Pl-rh alleles, but, since it was a self progeny of B-I/B-I, R-r/R-r, Pl-rh/Pl-rh, the plant should also have had at least one normal copy of each of these genes. When the variant was crossed to various genotypes, however, 100% (60/60) of the progeny had the mahogany phenotype, i.e. were light-colored and sun-red. None of the progeny had the normal phenotype that should have been seen if the original variant plant was due to a simple dominant mutation. The failure of a normal allele to segregate when it is heterozygous with certain alleles is a hallmark of paramutation.

We suggest the following hypothesis for the mahogany phenotype. The original variant had a new pl allele--Pl'-mah. The genotype of the variant plant was B-I/B-I, R-r/R-r, Pl-rh/Pl'-mah. The Pl'-mah allele produces a lighter, more sun-dependent phenotype, and is dominant to Pl-rh. Paramutation in this plant would change Pl-rh into Pl'-mah, therefore, all progeny of the variant would receive the Pl'-mah allele. This hypothesis explains the fact that all progeny of the variant plant had the mahogany phenotype.

Two lines of experimentation support this hypothesis. First, the mahogany phenotype segregates with the pl locus, as shown in the figure. The variant plant, of putative genotype Pl'-mah/Pl-rh, was crossed to a pl/pl tester, to produce Pl'-mah/pl heterozygotes. The r-r allele indicated in the figure gives the same anther phenotype as the R-r allele. Pl'-mah/pl plants were crossed to Pl-rh/Pl-rh plants, and the progeny were grown and classified by anther phenotype. 24/43 progeny had uniform purple anthers, indicative of a Pl-rh/pl genotype, 16/43 progeny had light anthers (mahogany phenotype), indicative of a Pl'-mah/Pl-rh genotype, and 3/42 plants had anthers with intermediate color, and were of uncertain genotype. This segregation of approximately 1:1 is that predicted by the hypothesis that the mahogany phenotype is caused by a Pl'-mah allele that can suppress a normal Pl-rh allele. To verify that the segregation of phenotypes was correlated with pl segregation, we used pl probes on Southern blots to determine the genotypes of the plants. The Pl-rh and pl alleles can be distinguished on Southerns, but the Pl'-mah and Pl-rh alleles are indistinguishable with the enzymes tested to date. Pl'-mah/Pl-rh can be distinguished from pl/Pl-rh, since the pl probe hybridizes to two bands in the latter, and only one in the former. The Southern analysis demonstrated that the 24 plants with uniform purple anthers were, in fact, of Pl-rh/pl genotype, and the 16 plants with light anthers were, in fact, of Pl'-mah/Pl-rh genotype. The three plants with an intermediate phenotype were Pl'-mah/Pl-rh genotype. Thus, 40/40 plants with a clear phenotype had the genotype predicted by the hypothesis that the mahogany phenotype is due to pl paramutation. (Figure)

The other evidence that the mahogany phenotype is due to pl paramutation is that the change in phenotype is correlated with a change in pl expression. Siblings that were either Pl'-mah/Pl-rh or pl/Pl-rh were grown, tissue collected and total RNA prepared. Husk and anther tissue from plants with a mahogany phenotype had substantially less pl message than similar tissues from plants with a normal phenotype. In these same samples, the amount of b and r message was similar in all plants, regardless of whether the phenotype was mahogany or normal.

Thus, we have seen a novel anthocyanin phenotype that is linked with the pl locus and associated with changes in pl expression. We propose that this novel phenotype is due to a pl mutation, which we call Pl'-mah. This mutant has an unusual property, in that, when Pl'-mah is heterozygous with Pl-rh, only Pl'-mah is transmitted. This exclusive transmission is a hallmark of paramutation, indicating that pl should be added to the short list of maize genes that undergo paramutation. We hope that further work on this system, and comparison of results from this system with b and r paramutation will shed light on this interesting phenomenon. 

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage