When two plants are grafted together, they share much more than water and minerals: They also swap genetic material, according to a study published in tomorrow's (May 1) issue of Science. These findings muddy the distinction between naturally-occurring gene transfer in plants and the human-mediated mechanisms we generally refer to as genetic engineering.

Image: Science/AAAS

Ever since Soviet and Western scientists in the 1960s denounced Trofim Lysenko -- the infamous Stalinist agronomist who rejected Mendelian genetics and maintained that heredity can be changed by grafting and other non-genetic techniques -- the prevailing dogma has been that genetic materials do not mix at grafting junctions. But Ralph Bock and Sandra Stegemann of the Max Planck Institute of Molecular Plant Physiology in Potsdam, Germany, weren't convinced. In 2003, they showed that genes are transferred at a high frequency from chloroplasts to the nucleus within tobacco plants, suggesting that genetic material might be "leaky" and flow between cellular components. So, they wondered, might DNA flow between the cells of grafted plants as well?

The researchers reciprocally grafted two transgenic tobacco plant varieties that carried distinct genetic markers in different cellular compartments -- one in the nucleus, the other in the chloroplast. They attempted 74 grafts, and showed that 45 contained cells harboring markers from both original plants -- evidence that intracellular transfer had indeed occurred.

Further experiments showed that the hereditary handover always flowed from the chloroplast to the nucleus, which Bock and Stegemann argued is probably mediated by the transfer of large chunks of DNA or even entire chloroplast genomes. The researchers also tested distant leaves but never found any evidence of gene transfer far from the graft sites. "From all we can say, it is a process that is probably restricted to the cells that are in contact with each other," Bock told The Scientist. Even so, shoots growing out of the graft site could stably transmit transferred DNA to progeny tobacco plants.

Farmers wishing to propagate their crops are not the only purveyors of grafting techniques. Stocks and scions can fuse naturally when the stems or roots of different trees grow into each other, too. And if they then swap genes, "this is a way for nature to produce transgenic plants," Bock said. "It's sort of natural genetic engineering." The findings "further blur the boundary" between natural gene transfer and human-mediated genetic engineering techniques, he added.

"It's remarkable that they were able to get as many transfers almost every time that they tried," Aaron Richardson, an evolutionary biologist at the University of Georgia who was not involved in the study, told The Scientist. But Richardson noted that the study only demonstrated gene transfer between "two things that are very closely related too each other." Thus, it remains to be seen how widespread grafting-mediated genetic exchange is between unrelated species. "In terms of wide transfer, this may not be as important," he said.

Bock is currently running experiments to test whether he can use grafting to move genes from tomato into potato and from tobacco into tomato. "In theory it should be possible to transfer genes between different species by grafting," he said. If Bock is correct, then grafting could provide a new tool to manipulate plants that are impervious to traditional genetic engineering techniques. Plus, it could explain evolutionary evidence showing that genes have been shuffled around between unrelated plant species.


Related stories:

Plant-to-plant horizontal gene transfer
[10th July 2003]

No horizontal transfer
[21st June 2001]