The possible discovery of a non-Mendelian form of inheritance in the tiny mustard plant Arabidopsis thaliana – the lab rat of the plant world – has hit the news in a big way this week. See e.g. Carl Zimmer’s blog post “Move Over, Mendel (But Don’t Move Too Far)”, the New York Times story “Startling Scientists, Plant Fixes Its Flawed Gene,” a blurb from the NSF, “Cress overturns textbook genetics” at Nature News, and the actual March 24 Nature article, Lolle et al. (2005), “Genome-wide non-mendelian inheritance of extra-genomic information in Arabidopsis,” Nature, 434, 505-509.
Basically, there is a gene, HOTHEAD, abbreviated HTH, and the recessive null mutant of this gene, hth, produces plants with fused flowers. When HTH/hth heterozygotes are self-fertilized, the progeny phenotypes are 75% normal and 25% mutant, as Mendelian genetics predicts. However, when hth/hth plants are self-fertilized, instead of producing 100% hth phenotypes, up to 10% of the progeny are HTH/hth and have the HTH phenotype.
Lolle et al. eliminated several possible sources of error, and hypothesized that some form of extra-DNA, RNA-based heredity could be at work. If true, this would indeed be Big News. Most researchers are expressing skepticism for the moment.
Over on his personal blog, De Rerum Natura, PT regular and genetics grad student Reed Cartwright has suggested one of several possible alternative explanations, this one involving natural selection at the level of pollen grains. Basically the alternative explanation is this: imagine that during the generations of cell replication leading to pollen grains, or during the pollination process, HTH pollen has a large advantage over hth pollen. In this situation, the rare revertant mutations from hth to HTH might spread to substantial frequency in the “pollen population” of a flower fairly quickly, and/or it might successfully fertilize ova at a higher rate than hth pollen. If these advantages were large enough (in other words, if hth pollen is highly impaired), this could explain how HTH reappears at a frequency of 10% in the offspring of the hth homozygote.
You heard it first on De Rerum Natura. I have not read enough on this to reach much of a conclusion as to how likely Reed’s explanation is, but it does highlight the important fact that selection occurs at multiple levels: not only between organisms, but between gametes, and between the cell lineages leading to gametes. This “selfish” intra-organism competition between cells in a single organism is not necessarily beneficial to the organism as a whole, and thus some multicellular animals have evolved methods to restrain this competition, such as setting aside gamete cells early in development. This topic is discussed thoroughly in a 1987 book by Leo Buss, The Evolution of Individuality.