Nature brings us a groundbreaking article(SF Gate article) on a new process of inheritance in mammals, paramutation, whereby a phenotypic change is heritable from one generation to the next without the transfer of an allele. In other words, you inherit a trait without inheriting the DNA. A neat trick, but also a violation of mendelian genetics, and contrary to the dogma of molecular biology.

The researchers in this case have mice with a defective copy of a gene called Kit. The gene in a heterozygote gives the mouse white feet and a white tip of their tail. Here's a cutout from the first figure of their paper of the heterzygous mice, and their genetically normal, but white-tailed offspring.



Further, subsequent generations of these totally genetically normal mice carried the trait, with eventual fading after many generations. The researchers then showed if you transfered RNA from a heterzygous carrier mouse (either somatic or from the sperm) into an embryo you could reproduce the effect indicating RNA is responsible for the hereditary nature of the phenomenon.

Pretty freaking neat! It's still not clear to me exactly what the biochemical mechanism of the trait is. They indicate through transfer of a micro-RNA that degrades the Kit gene that they also reproduce the effect, but that raises the question, does the defective gene lead to the expression of the microRNA (miR21 or miR22)? Is the altered expression of this existing microRNA responsible for the trait?

I think it must be, because they generated the phenotype artificially through a knockout trangenic mouse, and there is no reason the knockout insert should have generated this effect. Probably a combination of a susceptibility of the Kit gene to this effect, due to targeting microRNAs that exist naturally in the cell, combined with whatever bizarre effect the knockout allele causes leads to this phenotype. Maybe the cell is sensitive to transcriptional overactivity at the Kit locus. Also very troubling about the paper, they were able to get the white-tail phenotype (albeit at much lower frequency) just by transferring normal RNA to embryos. This makes even less sense.

Lots to think about, however, no matter what the mechanism, heredity without DNA in mammals is new and interesting. It also indicates a possibly new mechanism for hereditary disease, or an explanation for why some genetic disorders don't always end up causing disease even if people end up inheriting the gene, also known as variable penetrance of genetic disorders.