Radiation Russian Roulette

The discovery that radiation damages DNA in a new and unexpected way has raised fears that it may cause a far wider range of diseases than was previously thought.

Rob Edwards reports
The New Scientist, 1997, October 11th,

EAT enough arsenic and you'll die. Death is swift and sure. Radiation on the other hand, is far less predictable. Expose yourself to even a low dose of radiation and it might or might not kill you some time in the future. This hit-and-miss effect on the body, along with the fact that it's invisible, and so mysterious, is why most people have a profound mistrust of radiation.

Epidemiological studies of survivors from Hiroshima and Nagasaki show that people started dying of leukemia five years after the bombs were dropped. It took another 15 years for cancers to develop in the lung, breast and urinary tract. Scientists have used these and other studies to reduce emissions from nuclear plants to levels that they predict will keep the likely number of deaths from radiation-induced disease to a vanishingly small figure, At present, it's internationally accepted that a member of the public should not receive more than 1 millisieverts a year. Yet, despite these safeguards, mistrust of radiation and the nuclear industry persists.

Now, radiation biologists are concluding that the public may well have been right all along, They have found a previously unknown pathway by which radiation can subvert living cells, Radiation, they say, may cause a much wider range at diseases than epidemiological studies predict. Even levels of exposure below 1 millisievert a year could be harmful, and thousands of people could face early death as a result, Worst of all, the small doses of radiation that millions habitually receive could be poisoning the human gene pool, wreaking damage on future generations. "It is a horrifying concept, says Eric Wright from the Medical Research Council at Harwell in Oxfordshire. "But we now have early indications that it may be happening."

Conventional wisdom says that when ionizing radiation hits a living cell, there are three possible outcomes. Either the cell is unharmed, or it is killed, or it survives if with its DNA damaged (see Diagram on page 26). If the cell is not mended by the cell's repair enzymes and the cell divides, the damage will be passed on to its daughter cells. Depending on the type of cell and which genes, if any are damaged, the result could be uncontrolled growth and eventually cancer.

But Wright, who is head of experimental hematology at the MRC's Radiation and Genome Stability Unit, has found a fourth possibility. "Radiation can also," he says, "inflict damage on cells that at the moment can only be detected after they have divided several times." He calls this radiation induced genomic instability.

The eventual effects of the instability include broken or misshapen chromosomes and mutated genes, and early cell death. Research from around the world has shown that it can be produced by neutrons, X-rays, gamma rays and alpha radiation. In the laboratory, a dozen cell divisions over a couple of weeks are enough to generate chromosomal defects in up to 30 per cent of an irradiated cell's progeny. "I regard the phenomenon as established," says Wright, "There is no doubt that genomic instability is a real consequence of radiation exposure."

Vulnerable cells

Inside the body, this process could have big implications. In an average lifetime, a human being will experience 1016 cell divisions, mostly in the first few years of life and during puberty. But stem cells in the bone marrow, which keep the blood replenished with red and white cells, as well as cells in the gut and skin, continue to divide throughout adult life. Likewise, sperm are constantly produced by cell division in adult males, In these cases, the potential for radiation induced instability to do its worst appears to be highest.

Wright, Munira Kadhim, and colleagues announced the discovery of genomic instability in 1992. They exposed stem cells from the bone marrow of mice to plutonium-238 giving them a dose of about 0.5 grays of alpha radiation, This is the equivalent of a single alpha particle passing through a cell, the lowest dose the cell could receive.

The cells were kept in Petri dishes for 11 days until they had divided between 10 and 13 times, each producing between 10,000 and 100,000 daughter cells. Wright found the progeny of the irradiated cells contained three and a half times as many chromosome aberrations as the descendants of cells that were not irradiated. In a letter to Nature, he concluded that the "relative biological effectiveness" - a measure of how damaging low-level radiation can be in the body - for isotopes that emit alpha particles is "effectively infinite".

In 1994, Wright repeated the experiment with stem cells taken from four people. After between 10 and 15 divisions, up to 25 percent of the progeny of cells from two of the individuals were riddled with broken and distorted chromosomes. The fact that cells from the other two subjects showed no signs of induced instability may mean that some people carry genes that protect them from this type of damage, Wright argues.

Mounting evidence

At least six other laboratories around the world have now found similar results. Bo Lambert from the Karolinska Institute in Stockholm, for example, showed that X-rays damage the chromosomes of the descendants of irradiated human lymphocytes. Robert Ullrich from the University of Texas in Galveston discovered chromosome aberrations in the offspring of human breast cells caused by neutron and gamma irradiation. Last year, researchers from NASA and the University of Naples in Italy reported that the offspring of skin cells developed chromosome aberrations after exposure to X-rays and alpha particles. They concluded that genomic instability could determine late genetic effects and should therefore be carefully considered in the evaluation of risk for space missions".

These studies all use cells grown in laboratories and so are open to the criticism that something different might happen in living animals. At least two experiments however suggest that radiation also causes genomic instability in vivo. In a previously unnoticed study in 1989, Christian Streffer from the University of Essen in Germany exposed fertilized eggs of mice to X-rays. When skin cells were taken from the growing fetuses, they' contained more chromosome aberrations than cells taken from unirradiated fetuses.


n addition, last year, Wright and his colleagues at the MRC irradiated stem cells from the bone marrow of male mice and transplanted them into female mice. (The transplants and their progeny contained a Y chromosome so could be easily distinguished from the females' cells.) The researchers detected "persisting chromosomal instability" in the male cell line up to a year later.

More recently, in order to test his suspicion that some people carry genes that predispose them to genomic instability, Wright has shown that some strains of mice are more vulnerable to genomic instability than others. In one experiment, he exposed bone marrow cells from three strains to radiation, Daughter cells from two of these strains went on to develop many chromosome aberrations than tens from the other. Wright and other radiobiologists are now searching for the mechanisms behind genomic instability. In one experiment, Wright found abnormal levels of highly reactive free radicals in cells derived from irradiated cells. There is good evidence that raised levels of free radicals can induce chromosomal damage, and Wright believes that a buildup of these chemicals over several generations could be the root cause of genomic instability. Keith Baverstock, a senior radiation scientist with the WHO, has a different theory. He believes radiation could damage a gene for one or two repair enzymes. DNA is not a static molecule but changes all the time, and repair enzymes constantly cut out damaged sections and patch them up. If radiation stops one of these enzymes doing its job, a subsequent error may not be properly repaired. When the cell divides, its progeny will inherit this imperfection along with the disabled enzyme, which will carry out further imperfect repairs, and so on, piling up flaws down the generations. Finally, the whole thing gets so bad that the whole thing just breaks up and you get instability," argues Baverstock. At this point the question becomes the same as that asked for all forms of DNA damage caused by radiation - how does the damage cause disease? Most work on this question has focused on cancer and scientists believe that certain genes may hold the key. If a gene that promotes cell division is damaged, for example, that cell can divide over and over again. Other possible contenders are genes such as p53, that normally suppress development of cancer. If a person's two copies of p53 are damaged, a tumor is likely to grow. All these suggestions could be different parts of the same complex puzzle. Baverstock compares the difficulties of identifying the biological mechanisms to a long car journey. "You may know that a car started in Glasgow and finished in Cambridge" he says, "but the number of different routes it could have taken in between is immense." Despite the holes in our understanding of induced genomic instability, Wright feels that we already know enough to start worrying. He believes that in addition to cancers such as leukemia, it may cause small increases in a wide range of other diseases. These could include developmental defects in fetuses, such as deformed limbs and cleft palates and brain disorders such as Alzheimer's, Parkinson's and motor neuron diseases. But he stresses that these suspicions are not yet backed up by experimental evidence. The amount of radioactivity needed to induce instability could be tiny. Wright's director at the MRC Unit, Dudley Good argues that a single alpha particle is enough to injure a cell and increase the risk of disease. Those who swallow just one atom of plutonium are hence marginally more likely to die early. "It's like Russian roulette." says Goodhead. Wright and Goodhead are not the only ones to be concerned. Two years ago this month, more than 30 radiobiologists and health specialists from around the world gathered in Helsinki for a workshop on the public health aspects of radiation induced genomic instability. They cite 26 studies which, they say suggest that the accepted rules about how to calculate the biological impact of radiation should be rewritten, "Genomic instability changes our way of thinking about how radiation damages cells and produces mutations," says Jack Little, professor of radiobiology at the Harvard School of Public Health in Boston, who attended the workshop.

Last year. participants in the workshop produced a report for the WHO and, although it was not published, New Scientist has obtained a copy. It suggests that instability is an early, key event in the process that leads to cancer. It points out that people with the inherited disorder Fanconi Anemia develop the same sort of chromosome aberrations seen in radiation induced instability and about 15 percent of these contract leukemia.

Instability is also a "plausible mechanism" for explaining illnesses other than cancer, the report says. "It would seem likely that if genomic instability led to health effects these would not be specific but may include developmental deficiencies in the fetus, cancer, hereditary disease, accelerated aging and such non-specific effectsa as loss of immune competence," Epidemiology would be "powerless" to detect any relationship between the incidence of such diseases and exposure to radiation, the report says because the number of people who would suffer from any single disease would be too low.

Baverstock, who was the main organizer of the Helsinki workshop, and Wright believe that the world should be more wary of low-level radiation. If genomic instability is causing unpredicted disease, and if some people are genetically predisposed to it, the regulatory system starts to look inadequate. Existing measures meant to protect people - argue Wright and Baverstock - are less than reassuring.

To check that people do not receive more than 1 millisievert a year, the British Ministry of Agriculture, Fisheries and Food monitors "critical groups" of people who, because of their lifestyles, are likely to receive the highest doses of radioactivity from nuclear plants. The Sellafield complex in Cumbria has been the largest emitter of radiation in Britain, discharging radioactive gases into the air and liquid into the Irish Sea. The critical groups here have included fishermen working in the Irish Sea, people who eat seaweed and occupants of houseboats moored on contaminated Cumbrian estuaries.


The underlying assumption is that everybody is equally vulnerable to radiation, and that possible health effects depend purely on levels of exposure, but if the critical groups do not contain people who are genetically predisposed to genomic instability, then this system will overestimate the level of radiation deemed safe.

These people could then be exposed to levels of radiation that could harm them. So the number of people to have died or suffer from radiation released from Sellafield, nuclear weapons tests, the Chernobyl accident, from medical X-rays and radon in buildings - could be much greater than anyone has dared to admit.

This is regarded as unscientific scaremongering by Britain's National Radiological Protection Board at Harwell. Roger Cox, head of the radiation effects department at the NRPB, does not dispute that his colleagues across the road at the MRC have found unstable changes in cells descended from irradiated cells, but he disagrees that they are likely to have any impact on health.

"The basic science is not the problem here, it is their interpretation of it," Cox argues. "There is no proof that genomic instability leads to cancer or other diseases, no studies that have shown an association between illness and instability and there is no hard evidence of any causal mechanisms. Even if instability causes an increased rate of illness, it would already be taken into account by existing safety limits. We're quite some way from having serious doubts about the risk estimates we make," he says.

In particular, Cox dismisses the suggestion that genomic instability can cause small increases in a wide range of diseases as "totally speculative". Although he admits that such an effect cannot be ruled out, he argues that if it exists, it must be very minor, contained within the statistical noise of epidemiological studies. "There is rigorous medical surveillance of Hiroshima and Nagasaki victims," he says. "It would be a surprise if there was any major effect on any aspect of health that had not been picked up."

Wright concedes that there is no proof that instability causes cancer, but he argues that it is "highly unlikely" to be irrelevant to the process. "Cox fails to appreciate," he says, "that the scattergun effect of instability - small increases in a wide range of diseases would by its very nature escape the notice of epidemiologists."

Wright also questions the relevance of studies of atom bomb survivors to the understanding of genomic instability. Extrapolating from a group of people exposed to a large acute dose of radiation to a group receiving small, chronic doses may not be valid. Two different mechanisms may be involved, and it's important to learn if this is the case. "Instead," says Wright, "the NPRB gets defensive and criticizes everything that does not fit their corner of the world."

His biggest concern is that instability could blight future generations. He has collaborated with Brian Lord from the Patterson Institute for Cancer Research at the Christie Hospital in Manchester in a study that is due to be published soon. It gives the first clear experimental evidence that instability can he passed from a male to his offspring in sperm.

Lord found that the pups of male mice exposed to alpha radiation suffered chromosome aberrations in their bone marrow likely to be associated with genomic instability. The finding lends support to the controversial theory advanced in 1990 by the late Martin Gardner from Southampton University that the children of fathers exposed to radiation at Sellafield run a higher than normal risk of contracting leukemia.

But Wright and Baverstock fear that the consequences could extend far beyond the leukemia cases. Millions of people worldwide are exposed to low level radiation. The damage inflicted on their DNA could be passed to their children, and to their childrens' children. The human gene pool could be permanently polluted.

Furthermore, argue Wright and Baverstock, there is no logical reason why such damage should be confined to ionizing radiation. Carmel Mothersill from the Dublin Institute of Technology told meeting in Toulouse and Oxford last month that the offspring of cells exposed to low levels of cadmium and nickel also suffer high rates of cell death-a tell-tale sign of genomic instability. Chemicals in tobacco smoke, air pollution or pesticides might also destabilize the genome.

These ideas are already irritating scientists working in radiation protection, who believe that existing safeguards are adequate. Wright and Baverstock themselves accept that institutional change will be slow and that there is much still to be learnt about the biology of genomic instability. In the meantime, they are minimizing their own exposure to radiation. Baverstock refused dental X-rays that were not medically necessary. Wright too avoids medical X-rays unless his dentist or doctor can convince him they are essential. And he does not eat fish from the Irish Sea for fear of contamination by plutonium from Sellafield. Further Reading: "Genomic Instability induced by ionizing radiation" by William Morgan and others, Radiation Research, vol 146, p 247. A full list of references is on Planet Science.

Source: New Scientist: 11 October 1997 Compliments of Proposition One Committee