sealPurdue News
____

October 1999

Research puts sting into drama of virus-cell fusion

WEST LAFAYETTE, Ind. -- Researchers at Purdue University have identified the role of a protein segment that allows some cancer-causing viruses to latch onto and infect cells.

Analysis of the protein segment -- which bears a striking resemblance to bee venom -- may increase scientists' understanding of how retroviruses and some other viruses enter cells. The work may lead to new therapies aimed at thwarting such viruses.

Retroviruses can cause tumors and leukemia in animals and humans. The AIDS virus also is a retrovirus.

"The new data have important implications not only for the membrane fusion processes that lead to the entry of retroviruses and other viruses such as influenza viruses and the Ebola virus, but also to normal cellular processes such as fertilization of an egg by a sperm," says David Sanders, assistant professor in Purdue's Department of Biological Sciences.

Working with doctoral student Gwen Taylor of Monaca, Pa., Sanders discovered the importance of an amino acid sequence that is part of a protein that was previously thought to be unassuming.

Their study shows that this segment of the protein is essential for membrane fusion, and that changing just one amino acid in the sequence could eliminate fusion between the virus and its host.

The findings, published in the September issue of Molecular Biology of the Cell, may lead to novel treatments to block the entry of retroviruses and other membrane-possessing viruses, Sanders says.

Sanders studies the structure of proteins that stud the membrane surrounding a retrovirus. These proteins, called envelope proteins, allow the virus to bind to a host cell and promote the fusion of the virus and cell membranes. It is through this fusion process that the virus enters the cell.

Though the process of cell fusion has long been recognized, the structure of these membrane-spanning proteins include regions that were poorly understood, Sanders says.

"The specific amino-acid sequence of these regions, for example, was not considered critical for viral fusion because, in looking across all the different types of viruses, there is a tremendous amount of variety in this region, indicating that the amino-acid sequence may be random," he says.

Looking at a family of retroviruses that includes feline and human T-cell leukemia viruses, however, Sanders found that the membrane-spanning domain was not random, and in fact shared similarities with some other types of viruses.

Using the Moloney murine leukemia virus as a model, Sanders and Taylor showed that by replacing part of the sequence they could eliminate membrane fusion, while having no effect on any other known function of the protein.

"In fact, we showed that by changing just one amino acid, we could prevent fusion," Sanders says. "This is the first time that anyone has made a mutation in this region and seen this kind of effect."

Their analysis of the protein also showed that the amino acid sequence in this region shares some important characteristics with the active agent of bee venom, which works by forming holes in membranes.

"This is exactly what the bee venom and some other toxins look like," Sanders says. "And the way they work is to form a pore, or hole, in the membrane."

The new findings suggest that this region of the retrovirus envelope protein may be forming a similar hole in the membrane of the host cell during the fusion process, Sanders says.

These insights may point scientists toward new ways of thwarting viruses and help redefine scientific models of how membrane fusion occurs, Sanders says.

The Moloney murine leukemia virus used in the study is related to the feline and human T-cell leukemia viruses and is the major retroviral agent used for gene therapy procedures.

Funding for the study was provided by the National Institutes of Health, Purdue University Research Foundation and the Leukemia Research Foundation.

Source: David A. Sanders, (765) 494-6453; retrovir@bragg.bio.purdue.edu

Writer: Susan Gaidos, (765) 494-2081; susan_gaidos@purdue.edu

Purdue News Service: (765) 494-2096; purduenews@purdue.edu

NOTE TO JOURNALISTS:  Copies of the scientific study mentioned in this story are available from Purdue News Service, (765) 494-2096.

ABSTRACT

The role of the membrane-spanning-domain sequence in viral-glycoprotein-mediated membrane fusion
Gwen M. Taylor and David A. Sanders
The role of glycoprotein membrane-spanning domains in the process of membrane fusion is poorly understood. It has been demonstrated that replacing all or part of the membrane-spanning domain of a viral fusion protein with sequences that encode signals for glycosylphosphatidylinositol-linkage attachment abrogates membrane-fusion activity. It has been suggested, however, that the actual amino-acid sequence of the membrane-spanning domain is not critical for the activity of viral fusion proteins. We have examined the function of Moloney murine leukemia virus envelope proteins with substitutions in the membrane-spanning domain. Envelope proteins bearing substitutions for proline-617 are processed and incorporated into virus particles normally and bind to the viral receptor. However, they possess greatly reduced or undetectable capacities for the promotion of membrane fusion and infectious virus-particle formation. Our results imply a direct role for the residues in the membrane-spanning domain of the murine leukemia virus envelope protein in membrane fusion and its regulation. They also support the thesis that membrane-spanning domains possess a sequence-dependent function in other protein-mediated membrane-fusion events.


* To the Purdue News and Photos Page