Human Genome Sciences Inc. will unveil today a protein library it hopes will lead to new drugs and which provides information it wants to sell to other companies.
Human Genome will tell the Biotechnology Industry Organization's annual meeting in New York that it has produced what it considers to be the world's largest database of genes that encode about 10,000 "expressed" or "secreted" proteins.
Those proteins ultimately determine virtually all cell processes -- including the development of many diseases.
Being able to replicate or manipulate those proteins could prove an enormous benefit to medicine -- and to those who possess the information. That's because many diseases are believed to be triggered and controlled by abnormalities in the messages proteins deliver.
"This is one of the most exciting programs we have at Human Genome," said Craig A. Rosen, senior vice president for research and development at the Rockville company.
The publicly held company says it took about two years to amass the library. The company decided to unveil the database because it had amassed so much data and the information has already led to testing several proteins on animals.
"We are fairly confident now it will be a resource that will lead to new drug candidates in the future," said Rosen.
William A. Haseltine, Human Genome's chief executive officer, said the company hopes to develop new drugs from the database and move those into clinical trials. And it hopes to sell access to this database to drug companies trying to speed development of their own medicines.
Rosen said the expressed protein database is so large he doubts the company will ever be able to fully mine it on its own.
One possible customer for the protein library: The consortium of five drug companies, including SmithKline Beecham PLC, attempting to develop drugs from Human Genome's gene database.
Just 10 percent of all human genes, or an estimated 10,000 to 12,000 genes, encode secreted proteins.
Haseltine said the company program has resulted in a database of 80 percent to 90 percent of all secreted proteins. The company, he said, has deciphered the complete "sequence" or genetic code for 75 percent of those, or about 9,000 secreted proteins.
What proteins do
He said he believes the company's protein identification program has put it at the forefront of the race to move beyond compiling a complete database of all human genes, to beginning to understand what specific genes and proteins do and how to put that information to use in developing a new generation of medicines and diagnostic tools.
Human Genome has three experimental drugs in human clinical trials, which arose out of another of the company's gene research efforts. They include drugs to trigger new blood vessel growth around heart muscles, to speed wound healing and to help cancer patients bounce back from chemotherapy.
The company now has a team screening the protein library for potential candidates to use in medicine, said Rosen. But it plans to be very selective at first in choosing which ones to move forward as drug candidates.
"This is a very select set of information that we have. It's a very new approach to drug development," he said.
Haseltine, whose company already deals with almost a dozen companies for access to its database of genetic information, said: "It's another tool for us to use in our effort to address unmet medical needs."
Whether this new tool for mining molecular information results in big things remains to be seen.
Promising new drugs
But one thing is clear -- a growing number of experimental drugs, such as a new class of anti-cancer compounds that turn off the blood supply to tumors, are being developed from recently discovered proteins. While biotechnology companies lead this effort, several large pharmaceutical houses are embracing the approach.
Developing drugs from proteins is not a new concept. Insulin, for example, is made from a copy of the protein that the body produces to regulate glucose. Industry experts estimate that drugs made by copying the body's chemistry generate $10 billion in sales annually.
For now, only two other major biotechnology companies, Genentech of South San Francisco, Calif., and American Home Products Corp.'s Genetics Institute, have put a heavy focus on secreted proteins as a source for developing new drugs.
Genentech has developed five experimental drug candidates thus far from its expressed protein program called "Speedy," including an experimental diabetes treatment. The company anticipates up to 20 new drug candidates emerging from the program over the next two years.
Other biotechnology companies are taking different approaches to discovering genetic information for use in developing medical treatments.
Myriad Genetics Inc., a Salt Lake City genomics company, uses a database of known proteins and a proprietary technology to search for other proteins that might have a role in the disease process.
The company sells access to its ProNet protein database to drug companies looking to develop new drugs.
So far, the company has signed on three ProNet customers, including Schering AG and Bayer AG, in deals worth more than $115 million, paid in stages as drugs are being developed.
Pub Date: 2/25/99