Radiation can cause a previously unrecognized genetic change in cells that produce blood cells, perhaps setting the stage for leukemia later on, a test-tube study suggests.
The result fits with the idea that very low levels of particular kinds of radiation might cause leukemia, even though standard theory would not predict that outcome, study co-author Eric Wright said.
He said the standard theory of how radiation affects blood cells does not explain reported clusters of leukemia cases near some nuclear installations, for example.
Another scientist called the laboratory result interesting, but cautioned against any link to leukemia clusters.
Wright is head of the division of radiation oncogenesis at the Medical Research Council Radiobiology Unit near Oxford, England. He and colleagues present the work in today's issue of the journal Nature.
They studied the effect of alpha radiation, which is a stream of tiny particles, on stem cells from the bone marrow of mice. Stem cells give rise to a variety of blood cells.
In the experiment, most stem cells struck by a single alpha particle died.
The experiment focused on the survivors. Some descendants of the survivors showed abnormalities in their chromosomes, the string-like structures that carry genes, Wright said in a telephone interview.
The aberrations appeared in only some cells descended from a given stem cell, and often, different abnormalities appeared in various cells that descended from a single stem cell, Wright said.
He said the radiation apparently produced some kind of change in the stem cell that was inherited by the descendants, in which it could produce chromosomal abnormalities many generations later.
No cell had leukemia, but the experimental procedure would not have allowed that result to occur, Wright said.
Researchers still must find out whether the abnormalities could lead to cancer, he said.
Kenneth Moss of Arizona State University, a radiobiologist and president-elect of the Health Physics Society, said the result was worth further study.
But he said it did not demonstrate any link to leukemia clusters or to cancer in general.