In a startling discovery, geneticists at Purdue University say they have found plants that possess a corrected version of a defective gene inherited from both their parents, as if some handy backup copy with the right version had been made in the grandparents' generation or earlier.
The finding implies that some organisms may contain a cryptic backup copy of their genome that bypasses the usual mechanisms of heredity. If confirmed, it would represent an unprecedented exception to the laws of inheritance discovered by Gregor Mendel in the 19th century. Equally surprising, the cryptic genome appears not to be made of DNA, the standard hereditary material.
The discovery also raises interesting biological questions - including whether it gets in the way of evolution, which depends on mutations changing an organism, not being put right by a backup system.
"It looks like a marvelous discovery," said Elliott Meyerowitz, a plant geneticist at the California Institute of Technology. David Haig, an evolutionary biologist at Harvard, described the finding as "a really strange and unexpected result," which would be important if the observation holds up and applies widely in nature.
The new result, reported online yesterday in the journal Nature by Robert E. Pruitt, Susan J. Lolle and colleagues at Purdue, has been found in a single species, the mustard-like plant called arabidopsis that is the standard laboratory organism of plant geneticists.
But there are hints that the same mechanism may occur in people, according to a commentary by Detlef Weigel of the Max Planck Institute for Developmental Biology in Tuebingen, Germany. Weigel describes the Purdue work as "a spectacular discovery."
The finding grew out of a research project started three years ago in which Pruitt and Lolle were trying to understand the genes that control the plant's outer skin, or cuticle. As part of the project, they were studying plants with a mutated gene that made the plant's petals and other floral organs clump together. Because each of the plant's two copies of the gene was in mutated form, they had virtually no chance of having normal offspring.
But up to 10 percent of the plants' offspring kept reverting to normal. Various rare events can make this happen, but none involves altering the actual sequence of DNA units in the gene. Yet when the researchers analyzed the mutated gene, known as hothead, they found it had changed, with the mutated DNA units being changed back to normal form.
"That was the moment when it was a complete shock," Pruitt said.
A mutated gene can be put right by various mechanisms that are already known, but all require a correct copy of the gene to be available to serve as the template. The Purdue team scanned the DNA of the entire arabidopsis genome for a second, cryptic copy of the hothead gene but could find none.
Pruitt and his colleagues argue that a correct template must exist, but because it is not in the form of DNA, it probably exists as RNA, DNA's close chemical cousin. RNA performs many important roles in the cell, and is the hereditary material of some viruses. But it is less stable than DNA, and so has been regarded as unsuitable for preserving the genetic information of higher organisms.
Pruitt said he favored the idea that there is an RNA backup copy for the entire genome, not just the hothead gene, and that it might be set in motion when the plant was under stress, as is the case with those having mutated hothead genes.
He and other experts said it was possible that an entire RNA backup copy of the genome could exist without being detected, especially since there has been no reason until now to look for it.
Scientific journals often take months or years to get comfortable with articles presenting novel ideas. But Nature accepted the paper within six weeks of receiving it. Christopher Surridge, a biology editor at Nature, said the finding had been discussed at scientific conferences for quite a while, with people saying it was impossible and proposing alternative explanations. But the authors had checked all these and disposed of them, Surridge said.
As for their proposal of a backup RNA genome, "that is very much a hypothesis, and basically the least mad hypothesis for how this might be working," Surridge said.
Haig, the evolutionary biologist, said that the finding was fascinating but that it was too early to try to interpret it. He noted that if there was a cryptic template, it ought to be more resistant to mutation than the DNA it helps correct. Yet it is hard to make this case for RNA, which accumulates many more errors than DNA when it is copied by the cell.
He said that the mechanism, if confirmed, would be an unprecedented exception to Mendel's laws of inheritance, since the DNA sequence itself is changed. Imprinting, an odd feature of inheritance of which Haig is a leading student, involves inherited changes to the way certain genes are activated, not to the genes themselves.
The finding poses a puzzle for evolutionary theory because it corrects mutations, which evolution depends on as generators of novelty. Meyerowitz said he did not see this posing any problem for evolution because it seems to happen only rarely. "What keeps Darwinian evolution intact is that this only happens when there is something wrong," Surridge said.
The finding could undercut a leading theory of why sex is necessary. Some biologists say sex is needed to discard the mutations, almost all of them bad, that steadily accumulate on the genome. People inherit half of their genes from each parent, which allows the half left on the cutting-room floor to carry away many bad mutations.
Pruitt said the backup genome could be particularly useful for self-fertilizing plants, as arabidopsis is, since it could help avoid the adverse effects of inbreeding. It might also operate in the curious organisms known as bdelloid rotifers that are renowned for not having had sex for millions of years, an abstinence that would be expected to seriously threaten their Darwinian fitness.