Another round in the prion debate.

Virus or protein? The answer to that simple query torments researchers looking for the infectious agent that causes illnesses such as Creutzfeldt-Jakob disease in the human brain and similar disorders in sheep, cows, and other animals.

At first, researchers assumed that these neurodegenerative afflictions resulted from viruses--microscopic bundles of DNA or RNA wrapped in a coat of proteins. But all attempts to isolate and identify viruses from infected tissue proved fruitless.

Then in 1982, Stanley B. Prusiner of the University of California, San Francisco, School of Medicine launched a bombshell: He suggested that the infectious agent was a type of protein, which he called a prion. In the face of ridicule, Prusiner went on to identify a protein that could act as the hypothetical prion. His theory has gradually won a strong following (SN: 9/24/94, p.202).

Other scientists persisted in searching for viruses, arguing that prions cannot produce infections. Now, an analysis of brain tissue ravaged by Creutzfeldt-Jakob disease adds weight to that argument, report Laura Manuelidis and her colleagues at Yale University School of Medicine in the May 23 Proceedings of the National Academy of Sciences.

The Yale group ground up diseased brain tissue and ran it through sugar-laden gels, a method that separates components of tissue by either size or density. The separated fractions that contained most of the suspected prion proteins were not significantly infectious, whereas fractions with proteins bound to nucleic acids, either DNA or RNA, remained highly infectious. That suggested the presence of a virus.

In an attempt to rid the infectious fractions of any prions that might remain, the Yale team treated the samples with a chemical that breaks down proteins not bound to nucleic acids. The fractions stayed just as infectious. "The simplest explanation for all the data is that there is a virus that hasn't been found," asserts Manuelidis.