Photo: Larry Canner
One day in the early 1990s, Bert Vogelstein was showing fellow cancer researcher Sandy Markowitz around his lab at the Johns Hopkins Medical Institutions in Baltimore. "Suddenly, from halfway across the lab, this big hoot and holler goes up," says Scott Kern, who'd joined Vogelstein's lab as a postdoc after completing his medical training. People were huddled around a gel, excited by some eagerly anticipated result. Recalls Kern: "Sandy looked over and said something like, 'What's going on?"' To which Vogelstein replied, "Oh, they do that every day."
The eruption in question, Vogelstein says, was most likely another step along the road to identifying
FOUND IN TRANSLATION
Now the pair is busy transforming the lab into a vehicle for translational research. They are identifying genes and proteins that can serve as drug targets, devising more efficient and less invasive ways to diagnose cancer early, and exploring compounds and therapeutic approaches that can be tested in clinical trials. "We often look across the street to the clinical cancer center and think about how what we discover can impact the folks over there," says Victor Velculescu, a former MD-PhD student who's now a junior faculty member in the Kinzler-Vogelstein group.
Last year, Samuels, Velculescu, and others in the lab discovered that the gene for PI3K, an enzyme central to a signaling pathway that controls cell growth, is mutated in 30% of colorectal cancers and 25% of breast cancers. They turned up these alterations using high-throughput sequencing combined with a computer program designed to search whole families of genes for mutations present in tumors but not in normal tissues – a method that evolved in the Kinzler-Vogelstein lab.
Now some members of the team are attempting to determine how the mutations affect gene function. Others are working on understanding what the mutations mean for patients: whether cancers with altered PI3K are more aggressive or more likely to metastasize, for example. Still others are trying to develop drugs targeted to the mutant protein, "doing what a pharmaceutical company would do, but in an academic setting," says Velculescu.
GIVING CANCER GANGRENE
While part of the lab works on treatments that are still a ways in the future, others are focused on a therapy that sounds like it could be straight out of the past. Former postdoc Shibin Zhou, now a junior faculty member, and graduate student Ian Cheong and others on the "bug team" are using a soil-dwelling, anaerobic bacterium called
The idea for the project came when Vogelstein and Saurabh Saha tried to purify DNA from tumor samples fresh from the pathology lab, but then found themselves thwarted by vast areas of necrosis. "After seeing 25 or 30 cases with all these dead cells, rather than looking at it as just a pain in the butt making our lives difficult, we thought: Is this something that can be exploited therapeutically?" says Vogelstein.
The team continues to work on improving the bacteria's tumor-dissolving abilities and making sure the therapy is safe. In the meantime, the project has already generated a spinoff – monitoring infections in real time – that might have more immediate clinical application. "We wanted to see if we could track the bacteria as they were eating the tumor," says Luis Diaz, the newest faculty member in the group. So they developed a method using SPECT (single-photon emission computed tomography) scanning to monitor a radiolabeled marker that
Such unconventional projects can be risky undertakings, particularly for trainees, because the results and subsequent publications are not guaranteed. "It's one thing for me and Ken, who already have established careers, to embark on some of these crazy things," says Vogelstein. "But it's quite another for a grad student or postdoc to have the guts to try. They deserve a lot of respect for doing that."
But the craziness, it seems, is a big part of the draw. "I read the first bug paper in
FORMING A PARTNERSHIP
Zuhair Kareem/Path Photo
And then there are the patients. Like many members of their lab, Kinzler and Vogelstein were attracted to cancer research by a concern for people with cancer. Kinzler did his undergraduate training in toxicology, where he focused on assessing cancer risk. And one of Vogelstein's first patients, during his training as an MD, was a 4-year-old girl with a brain tumor. "She left an indelible impression on me," he says. The worst part was trying to offer explanations to the family. "We had no idea what was going on. Cancer could have come from outer space," says Vogelstein. "They'd ask, 'Is it something we did?' Of course you say 'no.' But if you don't know anything about the disease, how do you know it's not something they did?"
Vogelstein had already set up his lab when Kinzler enrolled at Hopkins as a graduate student and ended up joining the lab on a rotation. "Within one day I realized that Ken was probably the smartest guy I'd ever met," says Vogelstein. "He had a knack for understanding what was important." Kinzler finished his degree, stayed for a postdoc, and eventually became the youngest full professor in the department. "He was promoted in less time than I was," notes Vogelstein.
A partnership seemed only natural. "Two heads are better than one," says Vogelstein, "especially if they're reasonably good heads." Not that those heads necessarily see eye to eye. "We're always fighting," says Vogelstein. "If I have a bad idea, Ken will be the first one to tell me it's the worst idea he's ever heard." On that point, Kinzler agrees. "We like sharing ideas, but we really enjoy a good argument."
"It's fascinating to see," says postdoc Yardena Samuels. "I think they purposely disagree on lots of things."
"You'd hear them raising their voices at each other when the door was closed," notes Kern, who now runs his own lab at Hopkins. The ruckus might seem intimidating, he says, "but if you went in, you'd see two people with big smiles on their faces having a great discussion."
The intellectual sparring, the partners posit, allows them to hone their hypotheses and figure out what to do next. "One person has an idea that's not so good; the other one takes the idea, recognizes a germ of something interesting, and modifies it so it becomes practical and realizable," says Vogelstein. "It's not only more productive, but also a more fun way to do science."
READING LISTS AND ROBOTS
Aside from their mutual love of argument, Kinzler and Vogelstein bring different strengths to the enterprise; for example, their reading habits. "I think I approach problems more Talmudically," says Vogelstein, explaining, "I read everything I can possibly get my hands on."
"He has a huge suitcase he brings home on weekends," says Cheong. "Not a briefcase, an actual suitcase. Loaded with journals." Kinzler, on the other hand, once confessed that the only periodical he reads cover to cover is
Then there's the difference in experimental approach. "Bert loves shiny things," says postdoc Devin Dressman. "Fluorescent molecules, fluorescent beads, anything fluorescent. He does mostly wet work. Ken loves building stuff." In the late '80s, Kinzler assembled a homemade PCR "robot" out of "three water baths, a broom handle, a solenoid from a washing machine, and a little timer," says Vogelstein. Rubber bands attached a test tube rack to the broom handle and Kinzler programmed the contraption to dip the tubes in the baths, which were set at different temperatures.
"It really worked," says Baker, who used the device to verify
In February, Kinzler installed a metal lathe in his office. "It's not practical to wait a few days each time we need a new piece," he explains. "For the cost of a new Pipetteman, we got a milling machine so we could make the pieces ourselves." He's currently using the lathe to build bits needed for "beaming," a technique that involves performing PCR on DNA samples fixed to one-micron beads inside an emulsion. Dressman and others worked out the approach, which allows the team to analyze millions of individual molecules "with high specificity and sensitivity and in a cost-effective way," says Kinzler. Those qualities make it perfect for diagnostics such as testing stool samples for mutations in cancer-causing genes (e.g.,
"Who knows whether what we're doing now will ever result in real advances for patients," says Vogelstein. "Maybe we'll succeed, maybe we'll fail. But at some point I think you have to try. At some point you have to make that jump."