11
_______
THE FIRST SIGN OF A HUMAN CANCER GENE
On the heels of Witte’s breakthrough, research on the Abelson virus continued. The discovery that Gag/Abl was a tyrosine kinase upped the ante dramatically. The Abelson virus had arisen spontaneously in a laboratory and caused a fast and fatal leukemia in mice. Now, the protein responsible for that cancer, Gag/Abl, was revealed to be a tyrosine kinase.
Two distinct strands of research—src and the Abelson virus—had now come together. Erikson’s pursuit of the Src protein product had uncovered the role of kinases in cancer. Bishop and Varmus’s pursuit of the src gene’s origin had brought forth the notion of oncogene capture, and the understanding that healthy proto-oncogenes could convert to oncogenes through that capture. Src itself had nothing to do with the Abelson virus. But the concepts that had emerged from its study allowed Witte and others in the Baltimore lab to make sense of the giant protein inside cancerous mouse cells infected with the Abelson virus. The discovery led to an obvious question: Were tyrosine kinases some kind of universal mechanism for inducing cancer?
And that wasn’t the only question at hand. The researchers also wanted to learn more about the gene behind the tyrosine kinase. Because they had been able to create a noncancerous strain of Abelson virus, Witte and Rosenberg suspected that a single gene was causing the cancer. But what was the gene like? Why did it lead to cancer? How had it gotten into the Abelson virus?
Just as Rosenberg and Witte had arrived at the right time with the right talent, so did Steve Goff, who came to the Baltimore lab in 1978 with expertise in cloning genes. Goff had read about recombinant DNA while he was an undergraduate at Amherst College and knew right away that using this technology was what he wanted to do. “Cloning allows you to do things with . . . viruses that you could once only dream of doing,” said Goff, now at Columbia University. Today, the research techniques made possible by cloning—altering DNA, engineering mutations, combining DNA from two different sources—are as fundamental to science as letters are to reading. But in Goff’s eager postdoc days, these techniques were entirely new, and their advent was perfectly timed for unraveling the inner workings of the Abelson virus.
Goff knew the history of the Abelson virus when he began working on it in the Baltimore lab. He knew that it had been derived from the Moloney virus, which caused a type of cancer called T-cell leukemia. He knew that the Moloney virus would cause this leukemia in nearly every infected mouse, but that the cancer might take months to develop. He knew that when Herb Abelson had ablated the thymus in a bunch of mice and then infected them with the Moloney virus, some of them developed B-cell leukemia. He knew that the virus extracted from the B-cell tumor would cause more B-cell tumors, even in mice with an intact thymus. He knew that the B-cell tumors arose rapidly—not in months, but in days or weeks. “In a month, the mice were all dead,” said Goff. “So this was a really acute transforming virus.” He knew that this new virus—Abelson virus—was made up of the Moloney virus plus some unknown element. But he didn’t know what that unknown element could be; no one did.
Goff could use his recombinant DNA skills to find the added element in the Abelson virus. The place to start, though, was with the Moloney virus. Because its DNA was more abundant, and the lab already had probes for investigating its genome, Goff could more easily clone this virus. Afterward, he could compare the Moloney clone with the Abelson clone and zero in on where the two genomes differed. Whatever element was present in the Abelson clone but not in the Moloney clone had to be the one he was searching for.
Goff was far from alone in these efforts; the field of scientific research was crowded with competitors. With other labs outside of MIT also trying to clone the Moloney virus genome, the race was on to be the first. The laboratory of Inder Verma, a former member of the Baltimore lab, was the first to clone pieces of the genome. Then Goff, together with MIT researchers Chuck Shoemaker and Eli Gilboa, extracted Moloney DNA directly from infected cells and created the infectious clone.
That preliminary task accomplished, he moved on to cloning the Abelson virus, as labs elsewhere focused on other genomes from viruses known to cause cancer transformation. As those other genomes were investigated, a pattern began to emerge: The viruses had always acquired some cellular gene. Study after study confirmed the “cellular origin” of proto-oncogenes, as Bishop and Varmus had declared. As the evidence accrued, the notion of cancer as a genetic disease began to sink deeper into the minds of researchers. When Goff and the others dived into cloning Abelson, they were pretty certain about what they would find: a viral genome with a new gene.
They were right. As soon as the genome was cloned, they could see the difference between Abelson and its precursor. “The new one is shorter than Moloney,” Goff observed. “Pieces of Moloney are missing.” That was interesting: The genome of the Abelson virus had some but not all of the genome of the Moloney virus.
There was one more observation. “There’s unknown stuff also present,” Goff noted. The Abelson virus contained genes that were not present in the Moloney virus. As Goff, Baltimore, Rosenberg, and Witte stared at the cloning results, wondering what this mystery stuff could be, Bishop and Varmus’s proof that the RSV cancer transformation gene had been stolen from the host cell was echoing in their minds. “So we said, this has got to come from the mouse,” said Goff.
Goff and others in the Baltimore lab removed the portion of the Abelson virus genome that was different from the Moloney virus genome. The separated fragment could then be used as a probe to find its match in the normal mouse genome. If they were right, if the extra gene in the Abelson virus had come from the mouse, then their probe would find it in the mouse genome.
What they found was remarkable: The mice did indeed have this gene, now called abl, short for Abelson. But the mouse’s normal version—the proto-oncogene—was much bigger. The virus had a condensed version of the gene. Parts of the gene that did not pertain directly to the code for its product had been left out. Somehow, the virus had managed to acquire only the most essential information when it lifted the gene from the host cell. This made sense to Goff, who knew viruses had only very limited space for such codes. “Most genes are way too big to fit on a retrovirus,” said Goff. “[So] the way it works is, the retroviruses just carry the message.”
And they knew what the message was. Owen Witte had already uncovered that piece of information: The abl oncogene encoded a kinase.
How the virus had acquired that oncogene was now clear. When Herb Abelson injected the T-cell leukemia-causing Moloney virus into mice from which he’d removed the thymus, the virus just happened to scoop up a gene from the mice. That scooped-up abl gene just happened to be a proto-oncogene that, once it became part of the viral genome, turned into a full-on lethal oncogene. When this new virus, Abelson, infected mice, it caused a rapidly destructive B-cell leukemia. This was oncogene capture, just as Bishop and Varmus had described. Now, the mechanism by which the oncogene caused cancer was clear. The abl gene encoded a kinase, the activator, capable of setting in motion all manner of life-giving or -taking activities within. And this wasn’t just any kinase; it was a tyrosine kinase.
WHILE GOFF WAS isolating the culprit oncogene and finding its normal counterpart in mice, Witte traded his New England winters for California sun, accepting a position at UCLA, where he continued to study the Gag/Abl kinase. He fiddled with the Abelson virus genome to make mutants that lacked the abl gene. These viral creations, Witte showed, did not induce cancer in mice, further confirming the central role of the Abl portion of the mutant kinase in cancer transformation.
Then, late one night in 1983, Witte, who was now running his own lab, got a call from one of his students, who had noticed something strange. The student had been doing antibody experiments just like the one Witte had used to fish out Gag, the protein to which Abl had been attached. By now, Witte had made an antibody that specifically recognized Gag/Abl as its antigen. The stu
dent had been adding the new antibody to various groups of cells to see how they reacted with it. If the antibody gravitated toward a protein in the cells, then those cells, he knew, contained either Gag, Abl, or Gag/Abl.
That night, he noticed that one group of cells had a very large protein that was reacting strongly with the antibody. Just as Witte had seen the Gag antibody do in cells from Abelson-infected mice, the Abl antibody headed straight for this unknown protein, taking it for a foreigner and attacking immediately.
The instant recognition surprised the student. He knew the cells had not come from a mouse infected with the Abelson virus. What kind of cells were these? Where had they come from? The antibody to Gag/Abl had been created to study a virus known to cause cancer in mice. Why was it so active in cells that hadn’t come from an infected mouse?
He and Witte looked up the source of the cells and discovered that they had come from a patient with CML. That was a shock. Tyrosine kinases weren’t linked to any human cancers. In fact, no causative link between genetic mutations and cancer had ever been made.
An idea began taking shape in Witte’s mind. Why would an antibody to the Gag/Abl kinase react so strongly with human CML cells? Witte turned the matter over. Gag/Abl was a mutant protein encoded by a mutant gene that triggered B-cell leukemia in mice infected with the Abelson virus, itself an oddity that had arisen spontaneously in an NIH laboratory. What possible link could there be between Gag/Abl and CML? Oncogene research had not yet planted any stakes in the field of human cancer. Witte knew about the Philadelphia chromosome, but the idea that this abnormality could be connected to Abl was still out of reach. The thread of connection was so fragile and tentative, Witte was barely aware of the notion now taking form in his mind.
He knew that it was the addition of Abl that mattered. Goff’s clones had proven that this gene was the difference between Moloney and Abelson. In becoming Abelson, the virus had lifted the abl gene from the mouse. Once in the virus, abl became oncogenic, triggering B-cell leukemia in the mice.
Now Witte began to wonder: Was there an Abl kinase present in human CML cells? Is that what the Gag/Abl antibody was reacting with in the cells his student had called him about?
Was Abl involved in human CML?
12
_______
SPELLING OUT THE TRANSLOCATION
Witte and his growing crew of postdocs wasted no time in tearing into the mysterious CML protein. Which protein was reacting with the Gag/Abl antibody? What kind of protein was it? Was it involved in driving CML in people, as Gag/Abl was behind the B-cell leukemia of the Abelson virus?
In a completely separate pursuit, the Philadelphia chromosome was also coming under scrutiny. Rowley’s 1973 discovery of the translocation had not fallen on deaf ears; it just happened that those ears were very far away from her Chicago lab. For years, two Dutch scientists—Nora Heisterkamp and John Groffen—had wanted to clone chromosomes. They weren’t thinking of CML in particular, but they were aware of the increasingly recognized role of oncogenes in cancer, and wanted to dive into the work.
The development of cloning had engendered equal parts fear and hope around the world. Many federal governments restricted use of recombinant DNA—aka cloning—out of concern for what laboratory experiments could unleash. In this view, cloning was a biohazard disaster waiting to happen. It would bring new viruses that were more dangerous than the sum of their parts, animals with bizarre new traits, and whatever other unnatural catastrophes worry could conjure. But the promise of using cloning to understand human diseases was incredibly tantalizing.
At the time, an increasing number of diseases were being linked to genetic abnormalities. Hereditary disorders like hemophilia, in which blood clotting is impaired, and Klinefelter’s syndrome, which affects male development, had long been traced to variations in the X chromosome, one of the two chromosomes specific to sex. The 1959 discovery of the extra copy of chromosome 21 (known as trisomy 21) in people with Down syndrome fueled the already growing interest in the link between genes and disease, and brought the emerging field of cytogenetics into greater prominence. In the 1970s, Tay-Sachs disease, a well-known inherited disease most common in the Jewish population, was connected to mutations on chromosome 15.
Cloning portions of DNA using recombinant technologies would allow geneticists to explore those links more closely and to identify others. Although treatments targeting such genetic abnormalities—the prominent thrust of personalized medicine in the twenty-first century—were just hovering on the edges of anyone’s imagination, exploring the genetic underpinnings of human health and disease was considered essential scientific research. In 1976, the United States provided its first guidelines on the use of recombinant DNA, opening the floodgates ever so slightly. Other countries gradually followed suit.
In 1978, Groffen went to England to learn cloning techniques with another Dutch geneticist who’d left his home country, where recombinant DNA was still forbidden, to pursue his research elsewhere. Eventually, Groffen and Heisterkamp, who’d been postdocs in the same Netherlands lab, landed at the National Cancer Institute. Groffen’s assignment was to create a library of genes, noting their location on the human genome. The Baltimore lab’s work had been with mice, but by now geneticists knew that a gene in a mouse would almost certainly have a human genome counterpart. Having read papers about the abl oncogene from Baltimore’s lab, Groffen was interested to see if he could figure out the exact location of the normal abl gene on the human genome.
Just as Bishop and Varmus had so laboriously done with src before the invention of recombinant DNA, Groffen created an abl probe, a single strand of DNA containing only this gene. With that isolated sequence in hand, he could search for the exact location of the matching sequence within the human genome. Research at the Baltimore lab had found the fusion gene, gag/abl, responsible for the cancer-inducing capacity of the Abelson virus, and the fusion kinase Gag/Abl, encoded by the gene. At Owen Witte’s lab in California, signs of a similar protein in human CML cells had been spotted. The gag gene belonged exclusively to viruses, but abl was a mouse gene, and most likely the same gene existed in the human genome. But where among the entire array of human chromosomes was abl located? The work took months, but finally, in June 1982, he had the answer. The human abl gene was located on chromosome 9.
Over an evening glass of wine with another Dutch geneticist visiting the United States that summer, Heisterkamp and Groffen began wondering about whether abl might have some connection to CML. After all, chromosome 9, they knew, was involved in the Philadelphia chromosome translocation. Nowell and Hungerford had first spotted the abnormally small version of chromosome 22, and Janet Rowley had recognized this change to be part of a swap with genetic material on chromosome 9. More than 95 percent of people with CML had that genetic mutation. Could abl be involved in that translocation? It was just a musing. Each chromosome contains hundreds or even thousands of genes, so chances were slim that the translocated bit included abl. But by the time drinks were over, the notion had grown roots.
As it happened, that visiting geneticist had a brother, also a geneticist, who had just been given the task of cloning the Philadelphia chromosome breakpoint, that is, the exact spot in the chromosome where the breaks occurred in the swap. So, Heisterkamp and Groffen sent their abl probe—the mechanism for detecting abl in a genome—to the brother, Gerard Grosveld, back in Holland. Because the probe contained the genetic sequence of abl, Grosveld could use it to check for a match in the mutant Philadelphia chromosome.
He applied the probe to cells from CML patients. Soon enough, he found the matching gene. But it wasn’t on both copies of chromosome 9, as Heisterkamp and Groffen had found in normal cells. In one copy of chromosome 9, abl was there. But the other copy of chromosome 9—all chromosomes come in pairs, one inherited from each parent—didn’t have abl. A portion was there, but a large portion of it was missing. And that missing portion was on chromosome 22. The significance was immediately obvio
us: Grosveld knew that 22 was the other chromosome involved in the Philadelphia translocation. Groffen and Heisterkamp’s hunch had been correct: abl was located on the piece of chromosome 9 that switched places with a piece of chromosome 22.
Heisterkamp and Groffen set to work mapping chromosome 22 to see what exactly was going on. When abl moved to chromosome 22, what gene did it end up next to? What proteins did those genes encode? They knew that a mutation that combined two genes could result in two proteins being lumped together; Witte’s discovery of the Gag/Abl kinase, a fusion of two previously separate proteins, pointed to an underlying fusion of two previously separate genes. So checking out abl’s new neighbors made sense.
In 1984, they found the region on chromosome 22 where the break occurred when its tip translocated to chromosome 9. If the chromosome could be conceived to be a mile long, they had found the few inches in which the break always occurred. Of the nineteen CML patients whose DNA they studied, seventeen had a break at this same region of chromosome 22. In patient after patient, the break occurred at almost exactly the same point. They called it “breakpoint cluster region,” or bcr. In unraveling the sequence they’d homed in on, Heisterkamp and Groffen could see that bcr was a distinct gene. It had the telltale patterns in the A, C, T, G base pairings that mark the beginning and end of an individual gene sequence within the long, coiled strand of DNA. They had no clue what bcr did.
Just then, a third researcher entered the story. In Israel, Eli Canaani began investigating the protein product of abl—in genetics parlance, how abl was expressed in cells—in CML patients. As with so many scientific pursuits that catch fire, research on the abl gene and Abl kinase was lighting up a global map as lab groups around the world followed the onslaught of reports in scientific journals and began brandishing their own scientific swords as they joined the crusade. Each lab strived for the next significant breakthrough, fomenting competition but also speeding up the story.
The Philadelphia Chromosome Page 9
