Science Editor

One of the most formidable critics of the prion hypothesis still is not buying.

Similar to "the weapons of mass destruction" in Iraq, said Laura Manuelidis, professor and head of neuropathology at Yale University, the evidence for prions causing infection "is just not there."

"Misfolded prion proteins may just be empty shells," she told BioWorld Today.

The prion hypothesis is the idea that an endogenous protein can become an infectious agent that causes spongiform encephalopathies - such degenerative brain disorders as mad cow disease in cows, scrapie in sheep, and Creutzfeldt-Jacob disease (CJD) in humans. The putative infectious agent, prion protein, exists in two states: a normally folded one, whose function is unclear, and an abnormally folded one. An abnormally folded prion protein, when it comes into contact with other prion proteins, can induce a chain reaction of abnormal folding that leads to large protein aggregates, which in turn tear holes in the brain - hence the name "spongiform."

That's what a lot of people believe these days, anyway. But Manuelidis and her colleagues are not among them: In her opinion, the misfolded prion proteins are consequence rather than cause of infection. As for the nature of the infectious agent, Manuelidis said that "all the evidence points to a virus," though she also emphasized that "no one knows" what that infectious agent is.

In the Oct. 21, 2005, issue of Science, Manuelidis and her colleagues from Gifu University and Nagasaki University in Japan presented evidence that, while it does not identify such a putative virus, cannot be explained by the prion hypothesis either.

The scientists were able to develop a new assay system, consisting of co-cultured mouse hypothalamus cells that had been infected with a variety of strains of human CJD and sheep scrapie agents, to test whether pre-infection could protect cells against further infection by a different strain. The researchers found that while some strains do protect against subsequent infections with other strains, the relationship did not depend on the presence or absence of abnormally folded prion protein. They observed both protection without the presence of such protein, and a secondary "superinfection" in cells that were already infected with a disease strain that produced large amounts of abnormally folded prion protein.

The authors concluded in their paper that "these results are not compatible with the common assumption that TSE strains are encoded by some unresolved type of [prion protein] folding." Manuelidis elaborated that these data suggest infectivity "has to do with the strain of infectious agent, not with the presence or absence of prion protein."

Others disagree. Prion researcher Robert Peterson, professor at Case Western University in Cleveland, said that the paper does not challenge the prion hypothesis in the sense that what is being measured is "a change in conformational state, not infectivity" of the prion protein, and an alternative interpretation of Manuelidis' results is that "some forms [of prion protein] are more resistant to conformational change than others." He also noted some additional findings that are hard to reconcile with the idea of a virus as causal agent of prion misfolding diseases, including the fact that a change in the human gene encoding prion protein can lead to spongiform encephalitis.

Indeed, Manuelidis' opinion is running increasingly counter to the prevailing consensus of the prion research field. In the early years of the prion hypothesis, many scientists considered the idea that a protein could be an infectious agent preposterous, and many also felt the Nobel Prize Stanley Prusiner received in 1997 for work in that field undeserved, especially since it had not been shown at the time that a pure protein could actually infect a healthy animal.

In This Game, All Sides Refuse To Fold

The fact of the matter is that it still hasn't been shown that a pure protein can infect a healthy animal. While it apparently is easy enough to make prion proteins recombinantly, getting them to fold into the state characteristic of spongiform encephalitis is another matter altogether, and has to date been elusive.

But since 1997, scientists have come ever closer. The most recent advance came just a few months ago, when researchers from the University of Texas Medical Branch in Galveston, Universidad Aut noma de Madrid in Spain and the University of Chile in Santiago, reported in the April 22, 2005, issue of Cell that they were able to amplify abnormally folded prion proteins in a test tube until, homeopathy-like, the concentration of original material was too low to cause infection. Very unlike homeopathy, though, the method generated plenty of artificial prion protein, and the scientists were able to cause scrapie in a previously healthy hamster by injecting the animal with those artificial proteins.

An accompanying editorial, while admitting that "the possibility remains that RNA or other molecules present in the original brain preparation might be amplified along with [prion protein] and might play a critical role in conferring infectivity and specificity to the new [prion protein]," predicted that the PMCA method should allow the amplification of a recombinant protein within a few years. When such a recombinant protein manages to cause disease in a healthy animal, "the prion hypothesis will then be finally and irrefutably proven," the editorial said.

Manuelidis, for her part, noted that the Cell paper had its own unexplained findings, such as the fact that in vitro generated prion protein is far less infectious than its in vivo counterpart (a fact that the Cell paper's authors acknowledge is currently unexplained; they posit several possible explanations, including the fact that their method for amplification may lead to protein clusters that are a different size than in vivo and hence, may be less infective). Manuelidis, while she thinks the amplification technique could have great diagnostic value, said that the infectivity reported in the Cell paper could be due to contamination.

Manuelidis also said that the assay she and her colleagues described in their Science paper is simpler, faster and cheaper than current methodologies to study different strains of the agents, which should enable a larger number of researchers to enter the field, especially younger investigators still establishing themselves in the funding world. "Regardless of what one believes, the paper has a very new approach, and it opens things up to study by more people," she said.