by Robert Sanders
Fifty years after scientists first traced the genetic disorder phenylketonuria to a malfunctioning enzyme, scientists from Berkeley and Norway have reconstructed part of the enzyme that is knocked out in the disease.
The new picture of the enzyme not only explains why nearly half of the known 280 mutations in the gene inactivate the enzyme, but also provides a starting point for chemists to engineer a more stable form of the enzyme that could be given to stave off the serious effects of phenylketonuria, or PKU.
"The question is, based on our structure determination, can we make a stable form of the enzyme that those with the disorder can ingest?" said Raymond Stevens, assistant professor of chemistry here and a researcher in the Physical Sciences Division at Lawrence Berkeley National Laboratory. "It's a great approach that has worked in the past, but we don't know how easy it will be unless we try."
Other enzymes that have been stabilized so they can be given orally include lactase, taken by those who cannot digest milk proteins, and alpha-galactosidase, taken by those who have difficulty digesting fiber.
Stevens and his colleagues at Berkeley, along with a team from the University of Tromsø in Norway led by Torgeir Flatmark, report the three-dimensional crystal structure of a major portion of the enzyme, phenylalanine hydroxylase, in the December issue of Nature Structural Biology.
Phenylketonuria, though rare, is the most common inherited disease of amino acid metabolism, affecting one in every 10,000-15,000 infants. Before its cause was recognized, it inevitably led to moderate to severe mental retardation.
Only in 1934 did Norwegian biochemist Asbjørn Følling discover that it was caused by a build-up of the amino acid phenylalanine in the blood. Phenylalanine is one of the 20 amino acid building blocks used by the body to construct proteins, including the proteins that act as catalysts-the enzymes.
Scientists later found the cause to be an inactivated enzyme-phenylalanine hydroxylase that results from any of a number of single mutations in the gene for the enzyme. Phenylalanine hydroxylase normally converts phenylalanine to the amino acid tyrosine, and without it, the accumulated phenylalanine is eventually converted to toxic ketones that damage nerves and irreversibly affect the structure of the brain.
Perhaps one in a thousand people are carriers of a mutation in one of the two genes for phenylalanine hydroxylase, but symptoms normally do not occur unless a mutated gene is inherited from each parent.
Today infants are screened at birth for high levels of phenylalanine in the blood. Each year in the United States, several hundred babies diagnosed with phenylketonuria are put on a very strict diet that, since all protein contains phenylalanine, is essentially devoid of concentrated sources of protein. If a restricted diet is begun early and well-maintained throughout life, affected children can expect normal development and a normal life span.
Despite attempts since 1952 to crystallize the enzyme for biochemical study, success eluded scientists until last year, when a group at the University of Tromsø in Norway achieved the feat. They inserted the phenylalanine hydroxylase gene into bacteria, which produced enough of the enzyme to purify and crystallize. However, they were unable to take these crystals and analyze the three-dimensional structure of the protein.
"It's a very difficult protein to work with," Stevens said. "It tends to get destabilized quickly, so it's very touchy to work with."
Stevens and his Berkeley group decided to tackle the problem after they successfully determined the structure of a similar enzyme, tyrosine hydroxylase-a result they published earlier this year.
Most scientists predicted that the majority of the approximately 280 known mutations that deactivate the enzyme would be in the small active site, which latches onto phenylalanine and catalyzes its conversion to tyrosine. To Stevens' surprise, the mutations were found to be scattered randomly throughout a much larger area called the catalytic domain, which represents about a third of the whole enzyme.
"The majority of the mutations are in the catalytic domain and only a few are in the active site itself," Stevens said. "This protein seems to be extremely sensitive to single point mutations, which destroy its activity. It doesn't take much to knock this enzyme out."
In all, Stevens found that 107 of the known mutations were in the catalytic domain, including those most frequently encountered in African-Americans and southeastern Europeans, and the two mutations responsible for almost half of the cases of PKU in Caucasians.
Stevens plans to continue the collaboration with the Flatmark group in Tromsø, as they attempt to solve the complete protein structure of the enzyme and analyze its structure and function in an effort to understand its catalytic activity and the effect of mutations leading to PKU.
In addition, Stevens and his laboratory colleagues at Berkeley are working on ways to stabilize the enzyme structure without knocking out its active site.
"This has been a wonderful collaboration, combining our structural approach with the Norwegian group's genetic approach," Stevens said. Flatmark and his lab have been working on the disorder PKU for more than 30 years.
One of Stevens' main areas of research at Berkeley involves the structure and function of the many proteins and enzymes involved in the nervous system. He is looking too at the other enzymes involved in the metabolic pathway that includes tyrosine hydroxylase and dopamine hydroxylase.
Coauthors of the paper, aside from Stevens and Flatmark, are Fabrizia Fusetti of the chemistry department (now at the University of Groningen, The Netherlands); Heidi Erlandsen and Edward Hough of the University of Tromsø's Department of Chemistry; and Aurora Martinez of the University of Bergen, Norway.
The research was supported by the Physical Sciences Division at Lawrence Berkeley National Laboratory and various Norwegian funding agencies and foundations.