Happy Accidents: Serendipity in Major Medical Breakthroughs in the Twentieth Century (23 page)

It was amid this scientific ferment that Bishop and Varmus came on the scene. In 1970 Harold Varmus, who was interested in the genetic basis of cancer, began working as a postdoctoral fellow in the laboratory of virologist and biologist J. Michael Bishop at the University of California at San Francisco (UCSF). Following up on Rous's newly appreciated work, Bishop and Varmus set out to study the cancer-causing gene within the Rous virus. “Harold's arrival changed my life and career,” Bishop said. Their relationship soon became one of equals, and they would make their major discoveries as a team.

Viruses are subcellular forms of life—basically mere packets of genes, generally fewer than a dozen. In 1975 Bishop and Varmus made a startling observation. The virus they were studying had somehow appropriated a gene that sparked malignant growth in the host cells of chickens. The Rous sarcoma virus contains only four genes. Three of these are used to reproduce the virus; the fourth is the gene that induces the cancerous growth. They had come upon an elemental secret of cancer, a gene that can switch a cell from normal to cancerous growth.

What was the source of the gene? Does a cell itself harbor such genes? A piece of the mystery unfolded (like “lifting a corner of the veil,” in Einstein's famous phrase) when they found that the gene was present in healthy chicken cells as well as infected ones. The gene was not a native component of the virus, but rather at some point during the virus's cellular passages—either as it moved from cell to cell in one host animal or as it passed from one chicken to another—it had picked up an RNA copy of a chicken gene.

The California scientists undertook a manic search for this gene,
screening every species they could get their hands on. “I for one,” Bishop acknowledged, “failed to foresee the eventual outcome.”
⁵
They explored the DNA of ducks, turkeys, geese, even one of the world's largest and most primitive birds, the flightless Australian emu, and encountered startling results. All had the gene.
⁶
They looked at the cells of mice, cows, rabbits, and fish. The gene that was once thought of as a “chicken gene” was, in fact, present in every one of these species. Finally, they screened human DNA and were excited to find it there as well. (The researchers were startled by the vigorous skepticism of some whose incredulous response was “Are you trying to tell us that a chicken gene is also in humans?” Bishop was flabbergasted by this biological naiveté on the part of accomplished scientists. Had Darwin, he wondered, labored in vain?)

By now, reality was staring them in the face. Clearly, the gene is present throughout the evolutionary scale, having persisted for half a billion years, and belongs to the normal genome. The gene is not inherently cancer-causing but functions as a regular part of the cellular machinery, probably in connection with regulation of its development or growth. Retroviruses picked up these normal cellular genes and instigated changes that caused them to become cancerous. Such viruses thus carry a mutated cancer-promoting gene donated by a previously infected cell and may trigger the process of cancer when they infect other cells. The Rous virus, through an accident of nature during the course of viral propagation, served as the vector for the cancer-causing genetic hitchhiker, originating in a cell.

“From these findings,” wrote Varmus, “we drew conclusions that seem even bolder in retrospect, knowing they are correct, than they did at the time.”
⁷
Their new model points to the cellular origin of cancer-causing genes. When the ordinary gene malfunctions, it is transformed into an “oncogene” that can cause tumor formation. (“Onco-” stems from the Greek
onkos
, meaning “mass” or “bulk.”) A whole new paradigm was identied. Oncogenes are normal cellular genes that control cellular growth and when mutated become cancer-causing.

Thus, in each cell there is a potential time bomb, latent cancer genes that may be activated immediately or many years later. Mutations
may set off the time bomb, directing one renegade cell to overmultiply and become malignant. Cancer represents a series of events at the level of the genome. Susceptibility is sometimes genetic and heritable, but cancer can also arise through chance or, more often, when promoted by chemical and mutagenic carcinogens or retroviruses.

Bishop and Varmus shared the Nobel Prize in 1989, only thirteen years after their major discovery.
⁸
Their new model of the oncogene is a milestone that probes the ultimate origins of cancer, directing attention to a site deep within the cell. Their totally unexpected finding triggered a revolution in cancer research that continues today. It set in motion an avalanche of research on fundamental factors that govern the normal growth of cells and new insights into the complex group of diseases that we call cancer. It led to the discovery of scores of different oncogenes and their functioning through protein messengers that offer targets for drug treatment and, someday, it is anticipated, ways of preventing cancer altogether.

Like Riding a Bicycle

In 1993 Varmus described how strongly research gripped him: “It's an addiction. It's a drug. It's a craving. I have to have it.”
⁹
A long-distance bicyclist, Varmus drew a compelling metaphor for scientific research from this activity: “Long flat intervals. Steep, sweaty, even competitive climbs. An occasional cresting of a mountain pass, with the triumphal downhill coast. Always work. Sometimes pain. Rare exhilaration. Delicious fatigue and well-earned rests.”
¹⁰
After winning the Nobel Prize, Varmus became director of the National Institutes of Health for a number of years and then went on to head the Memorial Sloan-Kettering Cancer Center in New York.
¹¹

18

A Contaminated Vaccine Leads to Cancer-Braking Genes

On April 12, 1955, following nationwide clinical trials on 1.8 million American schoolchildren, Dr. Thomas Francis Jr. of the University of Michigan announced that the Salk vaccine against polio—at the time the leading crippler of children—was “safe, effective, and potent.” Euphoria swept the country. As Richard Carter noted in his biography of Jonas Salk, “People observed moments of silence, rang bells, honked horns, blew factory whistles, fired salutes, kept their traffic lights red in brief periods of tribute, took the rest of the day off, closed their schools or convoked fervid assemblies, therein drank toasts, hugged children, attended church, smiled at strangers, forgave enemies.”
¹

Up until 1961, the culture for preparing the lifesaving polio vaccine used kidney cells from rhesus monkeys from India. When a measles infection (from human contact) spread throughout the monkey colony, Merck switched to using African green monkeys.

Such an event might seem unfortunate, but what happened next is another example of how serendipity works its magic. Thorough testing uncovered the surprising finding that the first monkey species harbored a previously undetectable virus that was harmless to them but capable of killing and replicating in the cells of the second species. The conclusion was both startling and distressing: the Salk polio vaccine, in use since 1953, was badly contaminated with a virus. Had it
not been for the chance measles infection and the subsequent switch to African green monkeys, the virus would not have been detected.

Alarm spread when both Merck and NIH researchers showed that the virus caused cancer when injected into newborn hamsters. The shocking reality could not be dismissed: The polio vaccine, which acted to reduce or eliminate poliovirus epidemics all over the world, was contaminated with a virus that initiated cancer in hamsters. Mass immunizations with injections of three doses of vaccine were so popular that about 450 million doses had been administered in the years 1955–59. The United States Bureau of Biologics acted swiftly to eliminate the virus from poliovirus seed stocks and grow poliovirus solely in the African green monkey cells.

Meanwhile, amid the intellectual ferment in molecular biology of this era in cancer research, Arnold Levine, a molecular biologist at Princeton University, was intrigued by this newly discovered virus shown to transform normal cells into cancer cells. Levine started by thinking that the virus's protein shell might induce an antibody in the blood. The antibody could be detected as a biomarker to indicate the presence and severity of the infection and even the results of treatment.

Levine's search led him in 1979 to unexpectedly discover a new protein in the blood. He named it simply p53 because the protein has the molecular weight of 53,000 hydrogen atoms.
²
For ten years after its discovery, Levine followed the p53 gene and its protein down a false trail, believing it to be an oncogene, a cancer-accelerating gene. He found that putting the gene into normal cells (using rat embryos) made them cancerous. In 1983 he cloned the gene for the first time and demonstrated what appeared to be its tendency to induce tumors. But Levine was wrong about the actual nature of p53.

Finally, in 1989, the real breakthrough occurred. While pursuing the genetic causes of colorectal cancer in humans, Bert Vogelstein at Johns Hopkins University found a mutation in the p53 gene.
³
As Levine reviewed the data, he experienced what can only be described as a “Eureka!” moment. In a conceptual leap, he realized that p53 was not a gene that stimulates cancer but a gene that
suppresses
tumors.
⁴
He had been led astray because the clones he had used in the early experiments
were unwittingly composed of mutants. It was as though he was struck by lightning when he came to understand that the p53 gene had to be mutated in order to foster cancer transformation within a cell—that is, its normal function is to
inhibit
division and growth of tumor cells.

The normal p53 protein acts as an emergency brake to arrest any runaway growth of cells that may have acquired cancerous tendencies and as a damage-control specialist to induce cell death under specific circumstances, such as in the presence of DNA damage. It has been dubbed the “guardian of the genome” because of its vital biological functions. After ten years of dedicated research, “suddenly,” Levine ruefully noted, “p53 became a hot ticket in cancer research.” A single mutation in the 135th position of its 393 amino acids can eliminate the surveillance capability of the protein and allow a cancer to grow. In its mutated form, it is found in more than 50 percent of human cancers, including most major ones. Among the common tumors, about 70 percent of colorectal cancers, 50 percent of lung cancers, and 40 percent of breast cancers carry p53 mutations. The p53 gene is also linked to cancers of the blood and lymph nodes. Put another way, of the ten million people diagnosed with cancer each year worldwide, about half have p53 mutations in their tumors. The gene's inactivation through mutation enables tumor cells, in a famous phrase, “to reach for immortalization.”

In 1993
Science
magazine named p53 the Molecule of the Year. It has generated much enthusiasm for the development of strategies for the diagnosis, prevention, and cure of cancers.

Arnold Levine is a round-faced, high-energy, fast-thinking, fast-talking individual who commands deep affection from his laboratory team. He went on to become the president of Rockefeller University in New York. Would he consider the discovery of p53 serendipitous? “Absolutely,” he frankly responds, noting the frequency of chance discoveries and citing the observations of Richard Feynman, the Nobel laureate in physics, on how scientists commonly write their articles in such a way as to “cover their tracks” when it comes to how they stumbled upon the truth.
⁵

P
ROBING THE
G
ENOME FOR THE
R
ENEGADE
P
ATHWAY

A major focus resulting from such advances in recognizing the influences of genetic mutations upon cancer is the field of targeted drug therapy. In the human cell's intricate inner circuitry are dozens of molecular chains of communication, or “signaling pathways,” among various proteins. It is now understood that there are roughly ten pathways that cells use to become cancerous and that these involve a variety of crucial genetic alterations.

Research is actively directed toward developing drugs able to interfere with the molecular mechanisms that drive the growth in a tumor and that are attributable to these mutated genes. A handful have been approved for use. The best known is Gleevec, which has a dramatic effect on an uncommon kind of leukemia called chronic myelogenous leukemia (CML) and an even more rare stomach cancer, gastrointestinal stromal tumor (GIST). CML, however, is exceptional in that it is due to a single gene mutation, and the drug is able to block its specific tumor-signaling mechanism. Unfortunately, over time, further mutations circumvent the molecular signal that Gleevec blocks, building drug resistance. Most cancers are more complex, with multiple mutations that require a multipronged attack.

Picking the right targets is critical, and the most promising new diagnostic technology is the DNA microarrary, or “gene chip.” Gene chips can identify mutations in particular genes, as well as monitor the activity of many genes at the same time. It is hoped that this precise molecular biology will lead to the design of a range of drugs that target highly specific signaling pathways of cancer.

Cancer was long thought of as one disease expressed in different parts of the body—breast, lung, brain—but researchers are now teasing apart the myriad genes and proteins that differentiate cancer cells, not just from healthy cells but also from each other. The aim is to classify cancers mainly by their genetic characteristics, by which pathway is deranged, not by where in the body they arise or how they look under a microscope. The ideal is to fine-tune the diagnosis and treatment of cancer for more “personalized medicine.”