The finding is reported in the July 30 issue of Science.
In the study, the researchers created a large fragment of the normal prion protein -- a harmless protein found in all mammals examined. They then folded this fragment into the abnormal shape that they suspected would give it the infectious properties of the prion. Next, they injected the folded protein fragment into the brains of mice genetically engineered to over express the same fragment, but with the shape of the normal prion protein. After a year, the mice developed prion disease and brain tissue from the inoculated mice was injected into wild-type mice that subsequently developed prion disease in about half a year.
"Our study demonstrates that misfolding a particular segment of the normal prion protein is sufficient to transform the protein into infectious prions," says the lead author of the study, Giuseppe Legname, PhD, UCSF assistant adjunct professor of neurology in the laboratory of the senior author Stanley B. Prusiner, MD, UCSF professor of neurology and director of the UCSF Institute for Neurodegenerative Diseases.
"A great deal of evidence indicates that prions are composed only of protein, but this is the first time that this has been directly shown in mammals. The challenge in the last few years has been to figure out exactly how to demonstrate that prions are made entirely of protein."
SPONTANEOUS PRION DISEASES
The discovery that a small change in the condition of a cell can cause the development of a prion offers an explanation, says Prusiner, for the sporadic form of Creutzfeldt Jakob disease (CJD), which is responsible for 85 percent of cases of prion disease in humans (occurring in 1 or 2 people per million) and is believed to develop spontaneously. It also supports his belief, he says, that sporadic forms of prion disease are caused by prion strains that are different from the one causing bovine spongiform encephalopathy (BSE) in cattle in Britain. He says he thinks that sporadic BSE will be found in one to five cattle per million and predicts such numbers will be found with increased testing for BSE.
"The finding represents a renaissance in prion biology," says Prusiner. "For the first time, we can create prions in the test tube, which will change the way scientists do experiments in the field. We now have a tool for exploring the mechanism by which a protein can spontaneously fold into a shape that causes disease."
More broadly, he says, the advance may lead to similar changes in the way studies are conducted for other neurodegenerative diseases that involve protein misprocessing, including Alzheimer's disease, Parkinson's disease and amyotrophic lateral sclerosis. While it is not clear that a protein changes shape in all these diseases, each involves a particular protein that undergoes some form of misprocessing, in terms of a shape change, metabolism or degradation, or proteolysis. It is not clear which of these forms of misprocessing occurs in each disease, says Prusiner. However, as in prion diseases, the misprocessing involves a profound conformational change that most often occurs spontaneously.
"The insights that scientists have made into the spontaneous misprocessing of prion proteins have already aided progress in studies of other neurodegenerative diseases," says Prusiner. "But we hope that our new findings with synthetic prions will help scientists investigating other neurodegenerative diseases to move one step further in understanding how misprocessing is spontaneously initiated, and how it progresses."
The production of synthetic prions is the latest milestone in the 30-year effort by UCSF scientists to move in on the biochemical composition of the elusive agent, which causes a variety of similar rare, fatal, brain-destroying diseases, including sporadic CJD and variant CJD, in humans, BSE, or "mad cow" disease, in cattle, scrapie in sheep, and like illnesses in deer, elk and mink.
PRION'S FATAL DANCE
The researchers have long maintained that a prion does not contain nucleic acid, the genetic material of life (DNA or RNA). Viruses, which have a nucleic acid core, replicate by high jacking the machinery of a cell and using it to synthesize more nucleic acid. In contrast, prions are an aberrant form of a normal protein (thus composed of amino acids) that form when a particular segment of normal prion protein in the brain's nerve cells, or neurons, loses its corkscrew-shape structure (known as an alpha helix) and flattens into so-called beta sheets. They suspect that individual normal prion proteins (PrPC) occasionally misform in all people and relevant animals, but are routinely "cleared," or removed, from brain cells. However, in rare cases, they suspect, the abnormal protein, or prion (PrPSc), is not cleared.
Once conversion occurs, they hypothesize, the prion moves on to other normal prion proteins, pinning and flattening their spirals, and thus, initiating a process that occurs repeatedly, akin to a deadly Virginia reel in the brain. The accumulation and aggregation of the flattened beta sheets leads to structural damage of the nerve cells, causing cell degradation that generally leads to death in less than a year. Prions can arise spontaneously, result from an inherited mutation in prion protein gene or develop through infection from an exogenous source.
When the protein-only theory was postulated by Prusiner, in 1982, it was met with skepticism. In subsequent years, the UCSF scientists and numerous other groups have reported substantial evidence to support the hypothesis, reflected in the fact that Prusiner was awarded the Lasker Prize, in 1994, and the Nobel Prize in Physiology or Medicine, in 1997, for the discovery "of prions -- a new biological principal of infection," which he named prion (PREE-on), for proteinacious infectious protein. Still, despite the wealth of scientific studies producing evidence to support the theory, a direct, straightforward test the prion theory has eluded researchers until now.
The goal has been to create a bonafide prion in the lab, which would be proven to be such by its ability to infect animals and cause a fatal illness. The challenge, says Prusiner, has been the inability to determine the details of the prion's three-dimensional structure at the atomic level. "If we knew this," he says, "we could have designed a physical assay that would tell us we are now making PrPSc in the test tube."
The new study represents the latest tactic by Prusiner and his colleagues to get around this block: working from the belief that beta-sheet-rich structures harbor prion infectivity, but without knowing which segments are responsible, the team set out to create a synthetic agent made up of a subset of beta-sheet-rich structures that assemble into amyloid fibers, which they hypothesized might contain some prion infectivity.
Following the strategy used to establish that a virus or bacterium is the cause of a particular infectious disease, UCSF scientists reported 20 years ago that prions purified from brains of rodents that were clumped into amyloid fibrils triggered disease when injected into the brains of healthy animals, and produced mad cow-like brain pathology.
Because prions are unprecedented, UCSF scientists wanted to go one step further and produce synthetic prions. The scientists chose to produce a fragment of the normal PrP in E. coli since bacteria are known not to carry prions. The fragment was chosen because it corresponds in length to the truncated PrP that assembles infectious amyloid fibrils when purified from infected brains.
The scientists purified the PrP fragment from E. coli and then altered its conformation so that it might become an infectious prion. They did this by taking the segment of the protein that they know has the capacity to form amyloid, and placing it in a shaking device to promote amyloid formation. They tracked the process with thioflavin T, a dye that fluoresces in the presence of amyloid.
After 40 hours, amyloid was detected. To accelerate the reaction time, the team then took some of these amyloid fibrils, and used them as a "seed" for the production of nascent amyloid fibrils in a second shaking tube. This time amyloid fibrils were detected after 10 hours. These were called "seeded" amyloid fibrils.
To determine if the PrP fragment was infectious, Prusiner and his colleagues used transgenic mice making the same PrP fragment. Importantly, the mice express truncated PrP at 16 times the level that PrP is normally made in wild-type mice. The over expression of the PrP fragment in these mice shortens the incubation times, which already approach the lifespan of mice. Prusiner and his colleagues also thought that if the truncated PrP expressed in the transgenic mice corresponded precisely to the PrP fragment produced in bacteria, this would provide the most sensitive system for detecting newly formed prions.
Notably, amyloid is a structure that, depending on the protein it contains, has been implicated in a number of brain diseases including Alzheimer's and Parkinson's diseases. After about 300 days, with none of the transgenic mice sick, the researchers were ready to declare the study a failure. But then, at 380 days, one of the mice showed symptoms of a prion-like disease. Eventually, all of the inoculated mice showed neurologic disease, the last one 660 days after injection.
Prusiner and his colleagues then inoculated more transgenic as well as wild-type mice with brain extract prepared from one of the sick mice. The prions in the brain extract caused disease in about 150 days in the wild-type mice and in about 90 days in transgenic mice expressing full-length PrP.
In each case, on the primary passage, the scientists detected four hallmarks of prion disease -- (1) clinical signs of neurologic dysfunction (ataxia, or loss of motor coordination, and rigidity), (2) neuropathologic changes in the brain (vacuolation, deposits of PrPSc and astrocytic gliosis), (3) resistance of PrPSc to breakdown by protease and (4) most importantly serial transmission of prion infectivity to wild-type and other transgenic mice.
While the results strongly indicate that the subset of beta-sheet-rich structure represented by amyloid harbors prion infectivity, the scientists report that they have preliminary evidence that other beta-sheet-rich structures may also harbor prion infectivity. And they are interested in producing prion preparations that have much higher levels of prion infectivity than the two reported.
"Our findings gives us the opportunity to start exploring prions on a new level," says Legname.
Co-authors of the study were Ilia V. Baskakov, PhD, who, at the time the study was started was a postdoctoral fellow in the UCSF Institute for Neurodegenerative Diseases (IND), and is now at University of Maryland in Baltimore; Hoang-Oanh B. Nguyen, a staff research assistant in the UCSF/IND; Detlev Riesner, professor of biochemistry at the Institut fur Physikalische Biologie, Heinrich-Heine Universitat, Dussedldorf, Germany; Fred E. Cohen, PhD, UCSF adjunct professor of pharmacology, and Stephen J. DeArmond, PhD, UCSF professor of pathology and neuropathology.
The study was funded by the National Institutes of Health grants, as well as by gifts from the G. Harold and Leila Y. Mathers Charitable Foundation, the Dana Foundation and the Sherman Fairchild Foundation.