Berkeley -- A University of California, Berkeley, scientist is challenging one of the central tenets of cancer research, that cancer results from a chance series of genetic mutations that drive a cell into wild, uncontrolled growth.
Molecular biologist Peter Duesberg, better known for his claim that the human immunodeficiency virus is not the cause of AIDS, contends that mutation is not the cause of cancer. Rather, he says, cancer results from disruption of the normal number of chromosomes in a cell, primarily duplication of one or more chromosomes.
Called aneuploidy, this type of chromosomal abnormality is found in nearly every solid cancer studied to date, but has always been considered a side effect of cancer, not the cause itself.
In a peer-reviewed article in the March 28 issue of the journal Proceedings of the National Academy of Sciences, Duesberg and his colleagues at UC Berkeley argue instead that aneuploidy is the primary cause of cancer and explains many aspects of cancer that the genetic mutation theory cannot.
For some 25 years, Duesberg has pointed out problems with the genetic mutation theory of cancer. Now that the theory has become almost dogma, he is ratcheting up his criticism and receiving support from numerous other scientists
"Peter won't let the field stand still, which is extremely important," said Avery A. Sandberg, chief editor of the journal Cancer Genetics and Cytogenetics. "Once scientists think of one theory as being the de facto theory, we're in great trouble."
If Duesberg is right, it would overthrow a theory that has dominated thinking for the past 15 years, guiding research and dictating how doctors detect and prevent cancer. One field that would feel the impact is cancer screening.
"Rather than looking for mutations in biopsied cells, we should look for aneuploidy as a sign of early cancer," said Duesberg, a professor of molecular and cell biology.
A group of physicians at UC San Francisco is now screening for a type of skin cancer called melanoma by looking for chromosomal anomalies in skin cells.
Duesberg said that the principal problem with the mutation theory of cancer is that no one has successfully turned a normal human cell into a cancer cell by inserting mutated genes. Such a demonstration would definitively prove that mutations cause cancer.
"No one has found, even once, a combination of genes from any cancer that when inserted into normal cells turns them into cancer cells," he said.
He finds particular fault with a report last year in the journal Nature by Robert Weinberg and others of MIT's Whitehead Institute, in which the authors claimed to have accomplished just that. They took normal human cells and inserted two cancer-causing genes, called oncogenes, plus another gene that makes cells grow unchecked, and generated cancerous cells. This showed, they wrote, that these genetic mutations "suffice to create a human tumor cell."
Not so, says Duesberg. He requested samples of the cancer cells from Weinberg and found that all of them also had numerical chromosome alterations, or aneuploidy.
"The cause could have been either aneuploidy or genetic mutation," he said.
Another argument against the genetic mutation hypothesis, said Duesberg, is that nearly half of all cancer-causing chemicals appear not to cause mutations at all. Asbestos, arsenic, some hormones, urethane, nickel and polycyclic aromatic hydrocarbons all are known to induce cancer in humans, but none are mutagens.
"The mutation hypothesis predicts that all carcinogens are mutagenic," he said. "Yet half of all carcinogens are not mutagenic, so how do they cause cancer?"
Plus, Duesberg argues, if genetic mutations cause cancer, then cancer should arise immediately after a mutation. Instead, cancers appear decades after exposure to a carcinogen.
"A hallmark of carcinogens is that they have a very long latency period," he said. "Scientists argue that this is because cancer is a multistep, epigenetic phenomenon, but that exactly describes aneuploidy."
A major argument for aneuploidy over genetic mutation, Duesberg says, is that the cellular disruption caused by having too many copies of an entire chromosome is much greater than that expected from a handful of mutated genes, and is much more likely to affect the many cellular processes known to be fouled up in cancer cells.
"What's more likely to cause cancer," asked Duesberg's colleague and coauthor David Rasnick, a visiting scientist in his laboratory, "the tens or hundreds of genes screwed up by aneuploidy, or the several genes screwed up by a few genetic mutations?"
Experimental evidence analyzed by Rasnick and Duesberg and reported last year shows that cancer cells exhibit massive overproduction and underproduction of a large number of proteins. They found thousands of proteins whose expression was doubled in cancer cells.
"It's not a small number of genes that have a large increase in expression, but a large number with a small increase in expression that transforms a normal cell into a cancer cell," Rasnick said.
Support comes from a number of scientists, including Athel Cornish-Bowden, director of research at the CNRS Institut Fédératif de Recherche in Marseilles, France. In a July 1999 analysis piece in Nature Biotechnology he wrote, "Not only is the association between aneuploidy and cancer so close as to be virtually exact, but the predicted metabolic effect of over-expressing a large and arbitrary set of genes is just the collapse of normal regulation seen in cancer. Altering just one enzyme activity rarely produces much effect, ... but simultaneous alteration in many activities can overwhelm the normal controls."
In 1998, Nature Biotechnology science editor Harvey Bialy noted that the only solid tumors -- so-called to distinguish them from leukemia and similar immune system cancers -- whose cells contain a normal number of chromosomes are those very rare ones caused by retroviruses. Otherwise, some 5,000 known solid tumors have chromosome disruptions.
Bialy concluded that in a new, dynamic theory of the genome, "the oncochromosome may come to supplant the oncogene as the primary experimental focus."
According to Sandberg, "There's a lot to be said for Peter's theory. No one has disproved it, and since the genetic mutation theory was proposed 15 years ago, people have found lots of exceptions.
"I predict that both theories will turn out to be right, because cancer will be found to result from multiple pathways."
According to Duesberg's scenario, carcinogens enter cells and disrupt the spindle apparatus that drags chromosomes apart during cell division. Just as unbalanced cables twist a suspension bridge out of shape, so unbalanced spindles twist and improperly separate chromosomes during mitosis, causing duplication and loss of entire chromosomes. The damage such duplication can cause is evidenced by diseases such as Down's syndrome, in which cells have three rather than two copies of chromosome 21. That disease is characterized by severe developmental problems and retardation, plus many metabolic problems, a 100-fold increased risk of leukemia, sterility and an average life span of only 30 years.
The chromosome disruptions of aneuploid cells only worsen in subsequent generations of the cells. Luckily, Duesberg said, most such cells die, which means cancer is rare. Occasionally, however, the chromosome abnormalities will generate a cell that survives better than the normal cell, and it will grow into a cancer.
Such a scenario, not unlike the evolution of a new species, explains the slow development of most cancers, Duesberg said. Natural evolution of a new species also is based on chromosome number variation, he pointed out.
Duesberg and his colleagues have conducted several experiments that have produced support for his aneuploid theory of cancer. Among these are tests of nonmutagenic carcinogens that show that most of them do cause aneuploidy, even though they do not create genetic mutations.
One problem, he claims, is that funding agencies refuse to fund experiments in the area. Of 13 proposals Duesberg has sent to federal, state and local funding agencies, none has been approved. Nevertheless, he continues to build support for the aneuploid theory of cancer.
"This is an old concept that was abandoned too soon," he said, noting that scientists 50 years ago seriously considered aneuploidy a possible cause of cancer.
"Today, aneuploidy is not in the textbooks, it's not in the scientific repertoire," Duesberg said. "Two years ago, even some of my Berkeley colleagues didn't know about aneuploidy. Now they do."