杏吧原创

MENDELIAN inheritance, the central tenet of genetics, is under attack from a few scrawny weeds that haven鈥檛 read the textbooks. The weeds are somehow inheriting DNA sequences from their grandparents that neither of their parents possessed 鈥� which is supposed to be impossible.

The orthodox view is that genes are passed down in the form of DNA, and all organisms have to make do with this parental DNA inheritance, mutations and all. Chemical or structural modifications to DNA can switch off genes, and these changes can pass from generation to generation, a phenomenon called epigenesis. But epigenetic changes do not alter the actual sequence of DNA.

Yet that is what seems to occur in the weedy cress Arabidopsis thaliana, the workhorse of plant biologists. Cress with two mutant copies of one gene seem to be able to correct the DNA they pass on, ensuring that at least a few of their offspring revert to normal. Robert Pruitt, whose team at Purdue University in West Lafayette, Indiana, made this extraordinary discovery, thinks that the mutant genes are being repaired using RNA templates inherited from earlier generations.

Other biologists are astonished by the findings. 鈥淚t鈥檚 amazing,鈥� says David Baulcombe, an expert on plant RNA at the John Innes Centre in Norwich, UK. 鈥淭he notion that RNA carries the information almost seems like the only way it could happen.鈥�

It is possible that the phenomenon is limited to this one plant. But in Nature (vol 434, p 505), Pruitt鈥檚 team speculates that it might be a more widespread mechanism that allows plants to 鈥渆xperiment鈥� with new mutations while keeping RNA spares as a back-up. If the mutations prove harmful, some plants in the next generation revert to their grandparents鈥� DNA sequence with the help of the RNA. 鈥淚t does make sense,鈥� Pruitt says.

Such a mechanism would be especially useful to plants that self-pollinate and so are not as genetically variable as other plants. But it might happen in all plants and even animals.

Pruitt鈥檚 team made the discovery after finding that some Arabidopsis refused to 鈥渂reed true鈥�. To Pruitt鈥檚 irritation as many as 1 in 10 of the offspring grew normally despite their parents having a mutation in both copies of the hothead gene, which causes petals and leaves to stick to one another. He assumed that normal seeds or pollen were contaminating his trials.

鈥淚t was our view that it was heresy when we started working on it, but we鈥檝e had time to get used to the idea鈥�

But a series of experiments ruled out contamination. They also ruled out other possibilities, including the gene spontaneously mutating back to the normal form, the existence of more than two copies of the hothead gene, or closely related DNA sequences providing a template for repairs. Eventually, Pruitt was left with one, unbelievable explanation: the normal offspring were somehow acquiring genetic information from ancestors other than their parents.

鈥淚t was our view that it was heresy when we started working on it, but we鈥檝e had time to get used to the idea now,鈥� he says. 鈥淚鈥檇 say I鈥檝e been the biggest sceptic all the way along, but every experiment has been done to find a conventional explanation and it鈥檚 as foolproof as we can make it. I have every confidence in the data, but I鈥檒l feel better about it when other people have seen similar things.鈥�

The team has also found that in hothead mutants, other faulty genes mysteriously revert to the sequence of earlier generations too. It may be that the phenomenon is caused by the hothead mutation and restricted to plants that carry it, says Ottoline Leyser, who studies plant developmental genes at the University of York in the UK. 鈥淧eople have been working on mutants for years, and they all behave in a Mendelian way,鈥� she says.

鈥淚t鈥檚 possible it is just related to this one gene,鈥� agrees Pruitt. 鈥淲e can鈥檛 rule it out, but I think it鈥檚 unlikely.鈥� Other researchers may simply have dismissed mutants that revert to an ancestral form as the product of contamination, Leyser says. 鈥淢aybe it has been under the radar.鈥�

Pruitt鈥檚 team is now trying to find the stash of RNAs from earlier generations that might provide the templates for repair, and work out how it is passed down. 鈥淢y guess is that it is in the nucleus somehow, or hitchhikes on chromosomes, but that鈥檚 just speculation,鈥� he says.

While the search goes on, Pruitt hopes other biologists will hunt for evidence of the phenomenon in plants, animals and even humans. 鈥淚f we can understand how these templates are used, we might be able to make our own to order,鈥� he says. That might help improve existing methods for repairing genes, which are not yet efficient enough to be used to treat genetic diseases.