New images of an L-shaped molecule on the surface of a mouse leukemia virus could help scientists realize the promise of human gene therapy -- the effort to cure disease by inserting genes directly into human cells.
The images, published in the September 12 issue of Science, show the crystal structure of a piece of the virus's envelope protein -- the piece required to recognize and bind to receptors on the surface of a mammalian cell.
"This is the first high-resolution structure of any retrovirus receptor-binding domain," said Dr. Peter S. Kim, professor of biology and a member of the Whitehead Institute and associate investigator of the Howard Hughes Medical Institute (HHMI). "Knowing the structure is critical for understanding how retroviruses bind to and enter cells and should help improve the effectiveness of human gene therapy."
"Retroviruses are simultaneously a profound human medical problem and a potential medical solution," said Dr. Deborah Fass, first author of the Science paper and a postdoctoral fellow in Dr. James Berger's laboratory at Whitehead. "Natural retroviruses cause AIDS, leukemia and other diseases in both humans and animals. But remodeled retroviruses, stripped of their ability to cause disease, can be ideal vehicles for gene therapy."
Retroviruses are designed by nature to transport their genetic information into the nucleus, or command center, of the cells they infect. The retrovirus DNA inserts itself into the cell's DNA where it is duplicated every time the cell divides. In a natural infection, the retrovirus DNA eventually subverts the cell's command machinery, forcing it to make thousands of new virus particles.
Almost 20 years ago, researchers discovered that they could disable some retroviruses and trick them into ferrying human genes instead of their own genetic material into cells. These tiny delivery vehicles, produced from a mouse leukemia virus, lack the genes required to make new virus particles. They have the potential to combat disease by inserting normal genes into tissues of patients with severe genetic diseases or cancer -- a strategy called somatic cell gene therapy.
Retroviral vectors have been used in clinical trials to treat severe combined immune deficiency disease (SCID, the disease made famous by the "bubble boy"), and a few other diseases, including several forms of cancer. Efforts to apply the technique more broadly have been limited, in part because of the inability to target retroviral vectors to specific types of cells. Ideally, if one were treating a genetic disease involving the nervous system, for example, one would want a gene delivery vehicle that carried its cargo directly and exclusively to nerve cells in the brain or spinal cord.
Researchers have not had much success targeting retroviral vectors, in part because no one knew which part of the retrovirus's envelope protein was responsible for recognizing receptors on the surface of target cells. There had to be a "key" that allowed the retrovirus to lock on to one type of cell and bypass others.
The current studies, conducted jointly by the Kim and Berger laboratories at Whitehead and by Dr. James Cunningham's HHMI laboratory at Brigham and Women's Hospital, have identified that key. On the outer surface of the retrovirus envelope protein (on the short leg of the "L" that makes up the binding domain), two helices and a series of loops fit together to form a precise pattern of ridges and valleys. This pattern is the key that determines which cells are accessible to the virus and which are not.
"By adding cell-specific hormones or other factors to the critical loops, gene therapy researchers can begin to reprogram retroviruses to bind to specific target cells," Dr. Cunningham said. "Most importantly, we can also answer fundamental questions about the role of the receptor-binding apparatus in the retrovirus life cycle."
Researchers have been trying to find the elusive receptor-binding site for more than a decade. "The breakthrough came when Robert Davey, a postdoctoral fellow in Jim Cunningham's laboratory, purified the critical domain of the binding subunit and devised a way to obtain crystals using large quantities of the domain," Dr. Fass said.
The research collaboration was supported by the HHMI. The X-ray crystallography studies were conducted in the W.M. Keck Foundation X-ray Crystallography Suite at the Whitehead Institute.
A version of this article appeared in MIT Tech Talk on September 24, 1997.
