Secret agent: hidden helper lets fungus save plants from heat.

The story of a fungus that keeps plants from withering in hot soil turns out to have been missing a character--the virus that makes it all work.

The fungus Curvularia protuberata grows inside plant tissues without damaging them. In 2002, researchers working in Yellowstone National Park reported that grass colonized by the fungus thrived in soils that simmer at over 40[degrees]C (104[degrees]F) all summer.

A closer look now shows that the fungus alone doesn't protect plants from heat, says virologist Marilyn Roossinck of the Samuel Roberts Noble Foundation in Ardmore, Okla. The fungus itself has to be infected with a previously unknown agent, which she and her colleagues have named Curvularia thermal-tolerance virus, the group reports in the Jan. 26 Science.

Researchers haven't found many three-partner mutual-benefit societies, and this is the first plant-fungus collaboration known to have a virus as a third party, Roossinck says. She speculates that new ways to protect crops from heat might eventually result from understanding this threesome.

"I would hope that it changes people's thinking about viruses," Roossinck says. Scientists have primarily chosen to study viruses that cause disease, but she says that she suspects that most viruses don't have ill effects on their hosts. "There's a huge world out there that hasn't been looked at" she says.

Researchers discovered the original grass-fungus arrangement in a species of what's called panic grass, Dichanthelium lanuginosum.

While working on a different project in 2003, Roossinck looked through the Yellowstone Curvularia samples. She found signs of viral infection in the fungi in hot spot grasses but not in fungi from cooler places. Yellowstone soil can heat up to 50[degrees]C.

Roossinck and her colleagues isolated the virus and tested its powers in both the grass and tomato plants. One of the challenges that she faced was the failure of standard techniques to cure the fungus of its viral infection. However, when bringing fungal samples out of storage, Roossinck serendipitously discovered that freezing destroys the virus but not its host.

After Roossinck removed the virus by freezing the fungus, the latter no longer offered even limited protection to tomato plants. When she reinfected the fungus, its protective powers returned.

The new study "nicely demonstrates the complexity of plant-microbial interactions, says Stan Faeth of Arizona State University in Tempe, who studies grass and their live-in fungi.

The newly described three-way partnership reminds Nancy Moran of the aphids that she studies, which depend on resident bacteria for defense against parasites. Moran, who's at the University of Arizona in Tucson, showed that a virus in the bacteria provides the genes for toxins that could protect the host.

Live-in helpers "are a way for multicellular hosts, such as plants and animals, to acquire new ecological capabilities without actually incorporating foreign genes directly into their genomes, she says. "And viruses collectively have the greatest diversity of any genomes."