Baysal Bora and Joan Willett-Brozick found a mutation that causes paraganglioma tumors. Thats causing a stir because it looks as if the same defect may encourage the growth of common cancers.
12
PITTMED
COVER
STORY
WITH HELP FROM ANDEAN MOUNTAIN DWELLERS, AN OXYGEN-SENSING GENE IS LINKED TO TUMOR GROWTH BY ROBERT MENDELSON
UP FOR AIR
ccording to the clock on the laboratory wall, its 11 p.m. Besides that clock, there is only one other hint that the workday has long ago ended for most people. Its the lone window across the room. From this vantage point on the 14th oor of the University of Pittsburghs Western Psychiatric Institute and Clinic, Fifth Avenue looms in the distance. The thoroughfares perpetual trafc has dwindled to an occasional PAT bus and nocturnal automobile. But nobody bothers to peer out the lab window tonight. Nobody glances at the clock, either. Bora Baysal and Joan E. Willett-Brozick focus only on their data. They dont like what they see. They are searching for a genetic mutation within one single gene from a pool of what has been estimated to be 100,000 genes in the human genome. The search had been progressing well. Until now.
A
F
or Baysal, who today is an assistant professor of psychiatry and human genetics at Pitt, the search began in 1994, shortly after he left Ankara, Turkey, for Pittsburgh. His departure was the culmination of a career-altering decision. In 1988, he graduated from Gulhane Medical School in Ankara, and practiced medicine for two years as a family physician; but he found himself drawn to the unknowns of genetics, fascinated by the idea that answers to what makes one susceptible to certain conditions and diseases could be hidden away in the human genome. He began predoctoral training and received an international fellowship, which stipulated that he pursue his graduate education abroad. He decided on the University of Pittsburgh because, in large part, he had great respect for Robert E.
PORTRAIT PHOTOGRAPHY | JIM JUDKIS
Ferrell, who at the time chaired the Department of Human Genetics at the Graduate School of Public Health. After settling in at the University, Baysal needed to choose a dissertation topic, although in his field, he points out, choose isnt a totally appropriate word for selecting a project: Its not something you can entirely determine yourself.It depends on patient material, so availability of family material is one of the most important factors directing your research. You nd a very interesting genetic family and start your career. That is how hereditary paraganglioma entered his life. His PhD advisor, Charles W. Richard III, had contact with several families located in Pennsylvania with this rare hereditary disorder.
W W W. P E R U - E X P E D I T I O N S . C O M
Researchers found that the frequency of the tumors increased for inhabitants of higher elevations.
14
PITTMED
For those with hereditary paraganglioma, tumors generally start appearing after the age of 30, sometimes on the head, but usually along the carotid bodies (the tiny neurovascular workers located on the carotid artery that help the body respire). By the time the patient is 60 years old, some sort of tumor presence is highly probable. Its not a life-threatening disorder, although it may compress critical nerves and arteries in the neck, which can lead to health problems. Surgery in an advanced state is risky because of potential nerve damage and signicant disfigurement of the head and neck. Tumors can grow as large as half a skull. Richards access to several local families with the mutation was a boon to Baysal. These genetic families, he says, are precious resources for gene mapping. He had his project. Baysal wasnt alone in his search. This genetic condition is most prevalent in the Netherlands, so Dutch researchers had been searching for the genetic mutation, too. Through Richard, Baysal established a collaboration with a Dutch group. The Dutch had crudely identied the gene location, Baysal recalls. On the basis of their research, he focused on approximately 500 genes. We elaborated on this mapping process, and for years we and the Dutch tried to narrow down the critical region further and further and further. The mapping process wont be the pilot for a new television drama anytime soon. This process is extremely boring, he says. All you do is obtain new families and analyze chromosomes of affected individuals to see if you can narrow down the location. He and Willett-Brozick, a laboratory researcher, systematically did this day after day, week after week, year after year. Usually 10 to 12 hours daily, Monday through Friday; more hours on Saturday and Sunday. No vacations. Few distractions for the bachelor, other than a two-mile, straight-line jogging course from his Shadyside home to his Oakland laboratory. Bora lives this stuff; its his life, says Willett-Brozick, who conducted the gene-mapping experiments. The hard work paid off. By the beginning of 1997, the search had been narrowed by Baysal to approximately 175 genes. More good news. The Dutch had identied a specic region where they believed the mutated gene would be found; that region was located among the 175 genes still under Baysals scrutiny. It seemed they were on the right track. Seemed.
2 0 0 0 PAT O H A R A / STO N E
OCTOBER
15
Bora looked at the mutation not just mechanistically, not just technically; he thought, What would make this tumor happen at this ver y point with this type of tissue?
Richard, the principal investigator, is not convinced. The search goes on. More data. No progress. More frustration. More meetings. At last, Richard needs no more convincing. Baysals comprehensive collection of data speaks volumes. There is no sign of the mutation in the region identied by the Dutch. They redirect the search. However, more roadblocks surface. Its now the summer of 1997, and just as Baysal is about to receive his PhD, a job offer from a pharmaceutical company lures Richard away from Pitt. Meanwhile, the ultimate results from the project are far from a sure thing. Amid all the uncertainty, Ferrellthe man most responsible for Baysals leaving Turkey for Pittsburghreassures Baysal and for good reason: It was clear Bora was exceptional and Pitt Baysal found a mutation that causes mitochondria to act as though theyre in an oxygen-deprived would be foolish to let him founder environment. As a result, the tissue works overin any way. time producing cells. Pitt wasnt foolish. David Kupfer, chair of psychiatry, aysal and Willett-Brozick continue to supports Baysals lab operations, and Bernie stare at the data. At each other. At the Devlin comes on board as the new principal investigator. (Devlin, a statistical geneticist, data again. It doesnt make sense. Clearly, something is wrong. There can be consults as necessary, oversees the Dutch colonly two explanations, reasons Willett- laboration, and makes sure the project has Brozick. It is a technical problem on Baysals needed resources.) Baysal and Willett-Brozick end, or else the Dutch havent interpreted end up discounting the region identied by the Dutch and continue systematically to their data correctly. Baysal nds no discrepancies in his data. weed out possible genetic contenders. In the next two years, Baysal narrowed the At last they go home, the search narrowed no further. Tomorrow is no different. Or the search for the gene mutation to approximately 10 genes. Systematic research of those remainnext day. Or the next. Its time for a lab meeting. Willett- ing genesby Willett-Brozicks estimate Brozick, Baysal, and Richard talk. And, in should have taken another ve years, but the case of Baysal and Richard, they talk Baysal came up with a theory. He had come across some obscure autopsy loudly. Very loudly. Willett-Brozick isnt surprised. In fact, she is used to it. Charlie and reports from South America written nearly 30 Bora have really big hearts for science, she years ago. In those reports he noted an intersays. They just love it; they get very excited, esting parallel between those case and studies tense about it, not at each other, but excited his families with hereditary paraganglioma. The reports detailed how indigenous about their ideas. At this particular meeting Baysal is adamant. Its clear to him they are Peruvians and Bolivianswho lived among conducting their search in the wrong region. the peaks of the Andesfrequently acquired paraganglioma. Yet, these were not instances of hereditary paraganglioma; they were isolated cases. Moreover, researchers found that the frequency of these tumors increased for inhabitants of higher elevations. Their interpretations, naturally, focused on decreased oxygen levels at higher altitudes. They noted that the most common tumor site was located along the carotid bodies. This is the most oxygen-sensitive tissue in the whole body, says Baysal. When it detects a decrease in oxygen, it stimulates the heart and lungs to work harder so the body can adapt to lower oxygen levels. If you are in a continuously low oxygen environment, like the Indians living in high altitudes, this tissue keeps working and working and working and the number of cells increases and leads, eventually, to the Indians tumors. While there are hundreds of papers about carotid bodies and paragangliomas, says Baysal, I couldnt see anyone emphasizing the similarities: people with this genetic defect who lived in low altitudes experiencing the same kind of tumor that normally occurs in people living in high altitudes. From this coincidence springs Baysals theory: The problem for hereditary paraganglioma might be a defect in oxygen sensing. Immediately, Baysal and Willett-Brozick scan those genes in the remaining .4 megabase target region (i.e., approximately 10 genes) to nd out whether they might be involved in oxygen sensing. They nd one. Next, they need to run whats known as a single stranded conformational polymorphism test on one of their hereditary paraganglioma families to nd out if Baysals hypothesis is correct. The test, which nds mutations in small DNA fragments, normally takes a week to run, according to Willett-Brozick. She runs the test on Labor Day, 1999. We seem to work a lot of holidays, she says. The plans are to develop the lm a week later on the following Monday. Baysal cant wait. He goes to the lab on Saturday while Willett-Brozick attends to a previously booked weekend party at her home. Amid the Saturday afternoon revelry, her telephone rings. Its Baysal. Joan! We found it!
B
16
PITTMED
11cen.
stmy D11s1647 D11s1986
Baysal et al. 1999, 1.5 Mb
Heutink et al. 1994 ~11 Mb Van Schothorst et al. 1996 ~2 Mb
D11 s897 D11 s938 D11 s1885 D11 s1992
D11s1347
Baysal et al. 2000, SDHD Baysal et al. 1997, ~6 Mb
D11 s622
CD3D
11qter
The boxes portray polymorphic markers used to guide the search for the mutated gene causing hereditary paraganglioma. Black lines show how published works narrowed the search. Note that the 1996 paper was misleading, as evidenced by Baysals 1999 findings that pointed to other polymorphic markers.
What Baysal found was a mutation in the gene, named SDHD, that makes a protein eventually transported to the mitochondria the powerhouse and brains of the carotid bodies, according to Baysal. The mutation causes the mitochondria to react as if they were in an oxygen-deprived environment, even though that is not the case. Consequentlyas experienced by Andean Mountain dwellersthis tissue never stops working. The number of cells increases, and the increased mass eventually results in tumors. Baysal also found that genomic imprinting (an unexplained male/female inheritance protocol in the genome) passed on the mutation. In the case of hereditary paraganglioma, the disorder can be passed only from father to child. It was a moment of eureka, recalls Willett-Brozick. She marvels at how Baysal came up with his thesis: Bora looked at the mutation not just mechanistically, not just technically; he thought, What would make this tumor happen at this very point with this type of tissue? He looked at it as a disease state, through the eyes of a medical doctor. I dont think he could have done it without his medical training. He went from a physiological approach to a biochemical approach, and then we used our genetic techniques to determine that this gene is mutant. In February of this year, Science published the ndings. (Authors included Baysal, Ferrell, Elizabeth C. Lawrence, Willett-Brozick, Devlin, Richard, ve Dutch collaborators, and three MDs who provided the family materials:
David Myssiorek, Wendy S. Rubinstein, and Eugene N. Myers.) The paper drew international attention, and for good reason. Though the mutation is rare, Baysals identication of the SDHD genes inuence may well have a wide impact. Common cancers, not genetic in nature, have abnormalities where this gene is located, he explains. We are postulating that these alterations may be helping those common cancer developments, through the loss of our gene, because oxygen sensing is a critical function for cells. If you remove oxygen sensing capabilities, it may help cancer cells become more aggressive. It will take time to put Boras research into perspective, but I think what he discovered has the potential to be one of the more
important ndings to ever come out of the human genetics department, says Ferrell. A more immediate outcome of Baysals ndings is the ability to test for the SDHD mutation. Persons with the mutation have a 90 percent chance of developing the disorder. For those who test positive, Baysal strongly suggests certain lifestyle precautions to lessen the conditions impact. If patients can avoid putting themselves into situations that make their bodies work harder to obtain enough oxygen, such as smoking or living at high altitudes, he is convinced a paragangliomas growth could be slowed or even prevented. In addition, he recommends frequent screenings for the tumors. If they are discovered while in their infancy, they can be removed without great risk. Ferrell points out that Baysals findings impact not only hereditary paraganglioma and cancer research, but hypoxic conditions as well: Patients who are more or less similar concerning a pulmonary insufciency or some other medical condition that leads to hypoxia will have either a relatively benign clinical course, or, if their bodies dont respond, they will have a very poor outcome. This gene may give doctors a clue as to why some patients do well and other patients dont. For Baysal, many more late nights lie ahead. The real issue is just now starting for aspects of oxygen sensing and whether common cancers have anything to do with SDHD or other genes involved in oxygen sensing, says Baysal. Were trying to attack some of these questions and obtain some funding for the cancer aspect and for the genomic imprinting as well. Baysals six-year search may be over, but evidently his project has just begun. I
C O U RT E SY B AYS A L
MISBEHAVING GENES
Imprinting is a genetic battle of the sexes. In nature, a fetuss genes are derived from paternal and maternal copies of genes. Typically, the inherited parents genes are equal partners in the creation of the childs genes. However, for unknown reasons, a handful of genes receive instructions exclusively from either the maternal or paternal inheritance. This imprinting phenomenon occurs even though the chemical composition of the imprinted gene has the genetic inheritance of both parents. In the case of hereditary paraganglioma, the fathers gene dominates. As a result, when the genetic mutation is passed from a mother to her son, the mutation will be overridden by the fathers gene and the son will not be susceptible to the disorder. However, once the son has children, their genetic mutation will prevail over their mothers gene. Consequently, his daughters and sons will have a 90 percent chance of developing the tumors during their lifetime. The lucky 10 percent, according to Bora Baysal, of Pitts Departments of Psychiatry and Human Genetics, reflect environmental factors that may not be conducive to tumor growth. RM
OCTOBER
17