Low radiation hurts bystander cells. (Science News of the week).

Particles that radiate from decaying radon atoms can ravage the living cells they strike and increase the likelihood that those cells will later become cancerous. Researchers have now directly demonstrated that neighboring cells not suffering direct hits can be harmed, too. They've also taken a step toward showing how this type of radiation, called alpha particles, indirectly hurts those bystanders.

Radon derives from the decay of uranium and seeps naturally into the air from the ground. It's the primary environmental source of alpha particles, which contribute to cancer risk by causing aberrations in DNA. Alpha particles from inhaled radon are second only to smoking as a cause of lung cancer (SN: 3/7/98, p. 159).

Because a person's exposure to alpha particles typically is low, researchers have had to estimate public health threats from radon by extrapolating from the effects of higher doses of alpha radiation. Such data comes primarily from studies of survivors of the atomic bombs that destroyed Hiroshima and Nagasaki in Japan. The customary extrapolation, called the linear no-threshold model, assumes that cancer risk is proportional to the dose of radiation even at low doses.

According to a team of scientists led by Tom K. Hei of Columbia University, that model underestimates the risks from low-dose radiation. In the Dec. 4 PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, the researchers demonstrate more clearly than before that alpha particles striking and damaging the nuclei of a small fraction of the cells in a population can do enough indirect damage to nearby cells to increase cancer risk almost as much as if all the cells had been hit.

The researchers used a precision microbeam device to fire alpha particles into nuclei of human-hamster hybrid cells in petri dishes.

When the researchers irradiated all the nuclei with exactly one alpha particle each, 98 mutations of a certain gene occurred per 100,000 surviving cells. Zapping only 5 percent of the nuclei produced 57 such mutations per 100,000 cells, rather than the 5 mutations that a linear model predicts. Irradiating 20 percent of the nuclei produced more than 80 mutations, almost as many as resulted from 100 percent irradiation. Those data "suggest the need to reconsider the validity of the linear extrapolation," the researchers say.

Cell-to-cell communication channels called gap junctions appear to play a role in causing mutations in bystander cells. When the researchers bathed cells in a chemical that inhibits gapjunction communication and then irradiated the nuclei, they found fewer mutations among bystanders. Almost no bystanders were damaged in another experiment in which the cells lacked gap junctions.

"It's unequivocal that there's a bystander effect," says Eric J. Hall, the director of radiation research at Columbia, who wasn't an author on the paper. "The beauty of the microbeam technique is that you know which cells have been hit" and can observe mutations in nontargeted cells, he says. Previous studies using other techniques have found a bystander effect, but this is the first to directly demonstrate mutations, which are a cancer risk.

Philippe Duport, a radiation researcher at the University of Ottawa in Ontario, calls the study "very well designed and conducted," but sees "reasons to doubt" that risks are disproportionately high at low doses.

Radiation's effects in cell cultures don't necessarily reflect what happens in "a whole organism, with its full range of defense-repair mechanisms," says Duport. Processes such as DNA repair and cell death triggered by radiation damage could cancel the effect on bystander cells observed in the lab, he suggests.

Furthermore, while a bystander effect can contribute to cancer, other cell-to-cell interactions in living tissues "may mitigate against increased risk," says Barry D. Michael, a radiation biophysicist at the Gray Cancer Institute in Northwood, England. One of these interactions halts cell division and hence cancer. "The jury is still out on whether [cell-to-cell] effects lead to a greater or lower risk," Michael says.