Polyploidy-associated paramutation in Arabidopsis is determined by small RNAs, temperature, and allele structure

Paramutation is a form of non-Mendelian inheritance in which the expression of a paramutable allele changes when it encounters a paramutagenic allele. This change in expression of the paramutable alleles is stably inherited even after segregation of both alleles. While the discovery of paramutation and studies of its underlying mechanism were made with alleles that change plant pigmentation, paramutation-like phenomena are known to modulate the expression of other traits and in other eukaryotes, and many cases have probably gone undetected. It is likely that epigenetic mechanisms are responsible for the phenomenon, as paramutation forms epialleles, genes with identical sequences but different expression states. This could account for the intergenerational inheritance of the paramutated allele, providing profound evidence that triggered epigenetic changes can be maintained over generations. Here, we use a case of paramutation that affects a transgenic selection reporter gene in tetraploid Arabidopsis thaliana. Our data suggest that different types of small RNA are derived from paramutable and paramutagenic epialleles. In addition, deletion of a repeat within the epiallele changes its paramutability. Further, the temperature during the growth of the epiallelic hybrids determines the degree and timing of the allelic interaction. The data further make it plausible why paramutation in this system becomes evident only in the segregating F2 population of tetraploid plants containing both epialleles. In summary, the results support a model for polyploidy-associated paramutation, with similarities as well as distinctions from other cases of paramutation. AUTHOR SUMMARYIn 1866, Gregor Mendel formulated the general principles of inheritance based on crossing experiments with pea plants. Curiously, in 1915, the progeny from crossing pea plants with a regular and a "rogue" leaf phenotype was lacking the expected segregation and recovery of the regular phenotype. This discovery was one of the first observations of non-Mendelian genetics and later demonstrated for more traits in other plants and termed paramutation. Paramutation is due to the epigenetic switch of an active gene to a silenced version which is then maintained in the inactive state in later generations. This demonstrates that acquired epigenetic changes can become permanent. Despite its early observation and numerous studies, mainly in maize and tomato, it is barely understood how paramutation is established and which parameters influence the process. We investigated a case of paramutation in Arabidopsis thaliana, crossing plants with genetically identical but epigenetically different alleles that result in resistance or sensitivity to an antibiotic in the growth medium. Paramutation did not become manifest immediately but only in the progeny of the hybrids, and only in plants with a doubled chromosome set. These features make this paramutation distinct from other cases. Our studies revealed several parameters that influence paramutation: an important role for sRNAs to initiate silencing, the sequence of the allele itself, the environmental conditions during growth of the hybrids, the developmental stage, and the copy number ratio between the alleles.