By William J. Cromie

Gazette Staff

Eugene Shakhnovich invented a better way to discover chemical compounds that lead to promising new drugs. The professor of chemistry and chemical biology then faced the tough but common question of what to do next.

His invention takes the form of a computer program that designs or finds small molecules that bind to proteins. Virtually all diseases involve malfunctioning proteins, so finding ways to turn them on and off with such molecules is the goal of every drug and biotechnology company in the world.

"It takes only minutes to generate thousands of molecules on a personal computer," Shakhnovich says of his system. "Users can then take advantage of a powerful scoring system to quickly determine which molecules are most likely to bind tightly with the target protein."

Now, here's the tough question. Should Shakhnovich license this software to a drug company, form a partnership with one, or start his own company? If he wanted to form a fresh company, how would he do it? Where would he get money? To find answers, Shakhnovich consulted friends, colleagues, and the Harvard Office for Technology and Trademark Licensing (OTTL).

Elizabeth Hess of that Office had the marvelous idea of bringing in Business School students to help Shakhnovich and Robert DeWitte, a former post-doctoral affiliate and co-inventor of the software. Hess and colleague Frances Toneguzzo contacted Associate Professor of Business Administration Josh Lerner about getting the help of students from the School's Field Study Program. The program gives students the opportunity to obtain firsthand experience with actual problems, rather than working with textbook examples.

"This is the first time we've had such a collaboration, and it's been very productive so far," notes Toneguzzo. "It gives students valuable hands-on experience in the real world, and allows inventors to tap into the resources and expertise of the Business School."

"We see this project as promoting us out of coursework and into the reality of creating a business and helping it survive," says Christopher Michel, one of the students. He, Ann Dwane, and Thomas Goundrey are writing business plans and researching opportunities for the new company, Initio Pharmaceuticals.

"There is a business side to all of this which we must learn," Shakhnovich admits. "Sometimes we disagree on what should be done, but working with them has been helpful and positive."

Fitting Proteins

In 1995, Shakhnovich and DeWitte began looking at sites on proteins where small molecules can dock and, by so doing, change a protein's activity. Such a mating is often referred to as a key that fits a door, opening or locking it.

Proteins make all living parts work: muscles, organs, senses, metabolism -- the whole ball of life. Whenever a person has a disease, his or her illness can almost always be traced to a protein that isn't working as it should. Alzheimer's, AIDS, heart diseases, cancers, diabetes, obesity, you name it and proteins are somehow involved. Viruses, bacteria, and other infectious invaders, as well as the immune systems humans have evolved to protect against them, are all protein dependent. So is the switching on and off of genes. Therefore, drugs that place the right molecules into the right crevices, folds, or pockets in proteins hold the potential of treating, curing, even preventing most diseases.

When drug companies attempt to identify and work out the structure of a protein involved in an illness, they often spend millions of dollars and months, if not years, trying to find a key molecule that will increase or decrease the protein's activity. Company scientists go through a prodigious trial-and-error process of trying to fit every possible molecule they know into a keyhole that will change a protein's behavior. When they find more than one molecule that fits, their chemists make them all, then by more trial-and-error they narrow things down to the best match.

Shakhnovich and DeWitte computerized this process using software that produces virtual molecules, computer-designed compounds custom-made for the job. They call it SMoG, for Small Molecule Generator. Harvard filed a patent on it in September 1996, and Initio was incorporated a year later.

The idea of computer-generated proto-drugs appealed to Alain Gilbert, a French-born entrepreneur with more than 20 years of experience in the drug industry. He currently serves as chairman of Initio and is attempting to raise money in the United States and Europe.

"We are having discussions with investors, and we expect money to start coming in soon," says Shakhnovich.

"We also have had expressions of interest from large pharmaceutical companies that are potential customers," reports Ann Dwane, one of the Business School students.

Designing Drugs

Meanwhile, Initio is negotiating a licensing agreement with OTTL.

"Technology developed by Harvard researchers doesn't automatically go to companies formed by the researchers," Toneguzzo points out. "The University seeks bids from all companies that might be interested in licensing the technology. Our policy is to find the best company to develop the technology both from Harvard's and the public's point of view."

Initio placed the only bid, but Harvard still had to determine if it is the best company to develop SMoG. The University decided that it is. The soon-to-be-signed licensing agreement gives Harvard stock in Initio.

"Although we received no bids from large pharmaceutical companies, several of them said they would be interested when the technology becomes available in a user-friendly form," Toneguzzo says.

Shakhnovich and DeWitte are now trying to make that happen.

After some false starts, they worked with George Whitesides, Mallinckrodt Professor of Chemistry, on a protein involved in glaucoma, a common vision disorder caused by a buildup of pressure in the eye. A student in Whitesides' lab, Bartosz Grzybowski, used SMoG to generate a molecule designed to inactivate a protein known as HCAII, which is involved in this buildup. Robert Chapman, a post-doctoral affiliate in chemistry, made the actual molecule and tested it.

"Our results are still preliminary, but it seems to be successful," Chapman says.

Shakhnovich and DeWitte have produced two other molecules, which they refer to as "commercially viable," but these compounds have not yet been made in a lab. One may enhance proteins that suppress cancer. The other is involved in signal systems that switch genes on and off, thus it could be useful in treating a variety of genetic illnesses.

Faster, Cheaper, Better

These are the first examples of molecules that Initio plans to produce faster and cheaper than any similar existing software.

At first, the company plans to sell computer blueprints for building its molecules. Eventually, Shakhnovich intends to make the actual compounds in a chemistry lab, probably in Cambridge, Mass. "By next summer we expect to have a laboratory and about five employees," he says.

"Offering actual molecules, instead of computer blueprints for making the items, increases the value of the product dramatically," points out Goundrey of the Business School. "You save your customers much more time and money, and increase your revenues."

"There are many difficulties to be overcome, especially that of finding good scientists and managers," Shakhnovich notes. "But in the long run, we feel we will be successful.

In helping Initio survive, Goundrey, Dwane, and Michel believe they are doing more than getting valuable experience for themselves. "We want to make a difference while we're at the Business School," Michel says. "We want to leave here with a legacy that makes things better for the people who follow us."

What they are doing, they feel, could be a model for future collaborations that bring disparate parts of Harvard together in ways that assist students, faculty, and the University. Eventually, they see establishing "blockbuster teams of students from the Business, Law, and Medical schools to quickly transfer new technology from the laboratory to the marketplace."