When prions are transferred from one species to another—like from sheep and cows to mice in the laboratory or to humans in the case of the fatally neurodegenerative variant Creutzfeldt-Jakob disease—new forms of the infectious proteins can emerge over time that make them deadly to the new host. A new study examines the emergence and persistence of prion mutations, which allow prions to grow in infected cells in the presence of anti-prion compounds.

In the classic sense, prions, which are misfolded versions of the brain protein PrP, cannot mutate because they do not contain DNA or RNA. They can, however, give rise to variants with different properties, possibly due to differences in the folding, or shape, of the proteins. In the study, published December 31 in Science Express, researchers estimated the rate at which prion mutants can appear in cultured human nerve cells. In addition, the study suggests that once variants appear, they persist at low levels, giving rise to a heterogeneous prion population.

"On the face of it, you have exactly the same process of mutation and adaptive change in prions as you see in viruses," said Dr. Charles Weissmann in a prepared statement. Weissmann, who is the head of the Scripps Florida Department of Infectology in Jupiter, Fla., led the study.

To track prion mutation, Weissmann's team mixed one prion-infected human nerve cell with 1,000 uninfected human nerve cells in each petri dish. The infected cell contained a single prion that was susceptible to a drug called swainsonine, or swa. Then the team let the cells grow in the presence of swa. Because the team knew that each dish started with a single, swa-sensitive prion, the researchers knew that any additional prion-positive cells would mean that swa-resistant strains had developed that got released from the infected cells and spread to other cells in the dish. Based on the number of times the cells divided and the number of prion-positive cells in the dish, the group could roughly estimate how quickly prions became swa-resistant.

The researchers began to see new prion-infected cells after leaving the swa-sensitive prion in the drug for 22 cell divisions, which took about 22 days. In other petri dishes, drug resistant strains did not emerge until the cells had doubled over 50 times, or for 50 days. From these results, Weissmann's team approximated that one swa-resistant prion will emerge for every one million new prions that are formed.

The researchers pointed out that traits other than swa-resistance could be selected for in a population, and that the mutation rate they estimated only reflects new prion strains that acquired swa-resistance. "The overall mutation rate could be even greater and the prion population more diverse, comprising a multiplicity of 'substrains,'" the authors wrote.

Weissmann and his collaborators found that, similar to how viruses respond after antiviral drug treatment is stopped, prions reacquire drug susceptibility in the absence of swa. When the researchers removed swa from the medium in which prion-infected cells were grown, they found that the cells began to secrete swa-sensitive, instead of swa-resistant strains, after the cells had divided about 10 times. It is unclear if these susceptible variants came from newly formed prions or if they had been present in low levels during cell growth in swa medium. Nonetheless, even after about a month of growing cells in the absence of swa, the group found that 0.5 percent of prions remained drug resistant, indicating that this resistant variant could persist as a minority species in the population.

The fact that new prion "substrains" can appear and spread among cells in just a couple dozen cell divisions suggests that drug-resistance could easily develop in the lifetime of the host, from mouse to man. As a result, therapeutic attempts to inhibit the prion form of PrP are "likely to be thwarted," the authors wrote. A more promising approach, they suggest, would be to stabilize or reduce the expression of the normal form of PrP. Earlier studies by Weissmann and his collaborators showed that mice genetically engineered to lack the PrP gene were healthy, suggesting that the protein is not essential, at least for these animals. Different groups are pursuing the possibility of silencing the PrP gene or using antibodies that bind normal PrP protein as a treatment for prion disease.

Image of brain tissue showing neuron loss from variant Creutzfeldt-Jakob disease courtesy of  the CDC