Epigenetic and Gene Regulatory Consequences of Repeated Tetraploidies in Plants

Dr. Michael Freeling
University of California at Berkeley
Tuesday, March 25, 2014 - 4:00pm
Marley 230

Ancient allopolyploids carry subgenomes that are distinguished because duplicate genes are differentially lost (biased fractionation) and that subgenome with the fewest genes expresses it’s genes to a comparatively lower mRNA level (genome dominance). Said in another way: “the dominant subgenome has the most genes in it”. This rule holds for the ancient plant polyploids sequenced to date. I will present data from ancient allopolyploids maize and Brassica rapa supporting the conclusions: 1) Genome dominance is heritable over tens of millions of years, and passes thru sequential polyploidies. 2) Only allopolyploids express biased fractionation/genome dominance. 3) 24nt smallRNAs target near genes on the recessive subgenome preferentially: smRNA coverage marks the recessive subgenome even around genes that have “switched” expression from the expected (genes on the dominant subgenome expressed more) to the unexpected (gene on the dominant subgenome expressed less). So, smRNA coverage (and a “silenceability”) makes an excellent subgenome marker. 4) Transposon coverage marks lowered nearby gene expression, and is a poor subgenome marker. Since 24nt RNAs originate from transposons, making the aforementioned distinctions was not trivial. We hypothesize that the parent with the lower transposon load will become the dominant subgenome, and read avidly papers from the B. Gaut laboratory suggesting “tradeoffs” between transposon silencing and lowered gene expression. Genome dominance provides a window into a poorly understood aspect of gene regulation: the fine-tuning and inheritance of gene quantitative output. We have shown in both maize and brassicas that genes simply must have components that can quickly raise or lower their baseline transcriptional power. Since gene product balance has been shown to be particularly important for fitness, this component -- called a “rheostat”—is expected to operate as a part of all genes that encode dose-sensitive functions, and keep networked gene products in optimal balance. Transposons comprise this component. Once our use of the word “balance” is understood, beautiful, testable solutions to both the heterosis problem and the C-value paradox emerge from this work.