In Kurt Vonnegut’s Cat’s Cradle, scientists create a highly stable form of crystalline water called “ice-nine” that stays frozen even at high temperatures. Ice-nine instantly freezes any liquid water it touches. Its accidental release into nature solidifies the oceans and all contiguous bodies of water, and global catastrophe threatens our existence. Luckily for us, ice-nine is fictitious. But its biological counterpart, unfortunately, is not. The misfolded proteins known as prions are very real.

Prions are proteinaceous infectious particles, formed when normal proteins misfold and clump together. Biochemists Byron Caughey of the National Institute of Allergy and Infectious Diseases and Peter Lansbury of Brigham and Women’s Hospital were among the first to explore the analogy between Vonnegut’s ice-nine and prions in their 1995 review of scrapie, an infectious and deadly neurological disease of sheep.1 Like ice-nine, the particles that spread scrapie consist of highly...

The chilling similarity between the modus operandi of ice-nine and prions is an apt illustration of the long-standing and well-deserved reputation of prions as catastrophic agents. Researchers are identifying more and more cases of prion-like protein misfolding that cause neurodegenerative diseases.

But a different side of prions is also coming to light. Many newly discovered prions and prion-like proteins do not appear to cause disease at all. On the contrary, some even protect against it. Still other prions are turning out to be key players in basic biological processes. (See illustration.) These discoveries are driving a new appreciation for prions as versatile components in the machinery of life, a paradigm that has fostered conceptual advances in fields as diverse as signal transduction, memory formation, and evolution.

Prions as killers

Proteins that can act in a prion-like manner are referred to as prion proteins or prion-forming proteins, whether or not they are in the prion conformation. A prion is the infectious particle itself, not the proteins that make it up. A prion particle is thought to be composed of one or more amyloid fibers or oligomers, which are themselves composed of prion proteins.

Like other infectious particles, such as bacteria and viruses, prions can spread from one organism to another. Oral uptake is the most common natural form of transmission. Humans have also become infected through blood transfusions, human hormone injections, and surgery with contaminated instruments. Prions exhibit different variants, or strains, each with distinct molecular features and clinical manifestations. But what most fascinates scientists and the public alike is that, in contrast to viruses and all living organisms, prions lack the canonical information-storage molecules—DNA and RNA—yet are still able to copy and transmit biological information.

The idea that proteins could act in a manner previously ascribed only to nucleic acids was greeted with skepticism and ridicule when it was first championed by Stanley Prusiner in 1982.2 Even today there are those who maintain that prion diseases are actually caused by viruses. But as often seems to be the case in scientific discourse, what was once heretical is now dogma. The “protein-only” hypothesis, which posits that a string of amino acids is sufficient for disease transmission, steadily gained acceptance and, in the last three years, achieved irrefutable status when Jiyan Ma of Ohio State University College of Medicine and colleagues generated bona fide infectious particles from recombinant prion proteins.3

How do proteins pull off this remarkable feat? Like the water molecules in ice-nine, proteins within prion particles are arranged into a dense, highly organized lattice. But unlike ice crystals, this lattice grows in only one dimension—from either end—resulting in a proteinaceous fiber. (See illustration.) Every protein subunit takes on the exact configuration of those it flanks. The subunits at the ends of the fiber are exposed. Each of those exposed surfaces acts as a sticky template that recruits the next subunit, locking it down and contorting it into the same configuration. That new subunit now acquires the property of the original, and the process repeats ad infinitum.

Crystalline fibers, usually referred to as amyloids, have a long association with disease in their own right, beginning in 1639 with the first description of an autopsied spleen harboring white stones that would later be recognized as amyloid.4 Currently, amyloids are associated with more than two dozen incurable human diseases, including Alzheimer’s and variant Creutzfeldt-Jakob disease (the human version of mad cow disease), as well as Huntington’s, Parkinson’s, and type 2 diabetes.

Since prions are composed of amyloids, are these amyloid diseases also, in effect, prion diseases? Increasingly, the answer appears to be “Yes.” An avalanche of recent discoveries has revealed that some of the most prevalent and devastating neurodegenerative diseases, such as Alzheimer’s, Parkinson’s, and Huntington’s, all spread throughout the brain in a prion-like manner, involving self-templated protein deposition.5 Alzheimer’s and Parkinson’s diseases can even be transmitted between laboratory animals (albeit deliberately), through the inoculation of diseased brain samples into the brains of healthy recipients. Similar results have emerged in animal models of Huntington’s and other diseases. Thus, the boundary between amyloid and prion diseases is rapidly dissolving. The prospect that a similar process could occur in the real world—with common dementias transmitting between individuals, effectively becoming infectious diseases—is now under debate in the scientific community.

Prions as agents of inheritance

PROMISCUOUS PRIONS: Prion proteins recruit other proteins of the same sequence as they grow into a neatly organized lattice. When a new monomer arrives, it links to the fibril and assumes the exact shape of its neighbor. Fibrils can ultimately cluster together to form large deposits, or plaques, though the relevance of these clusters is not clear.
See full infographic: JPG | PDF
Thankfully, there is also a bright side to prions. The chain reaction of protein misfolding at the heart of prion propagation has also been harnessed by some organisms for a variety of benefits, including the regulation of gene expression and of the immune response.

This expanded view of prions traces its origin to a bizarre trait in the budding yeast, Saccharomyces cerevisiae, that had puzzled geneticists for decades due to its non-Mendelian mode of inheritance. When cells expressing the trait, which manifests as an enhanced ability to utilize poor nutrient sources, are mated with cells that do not, the trait is inherited by all of the resulting progeny, rather than the one-half expected if the trait had resulted from a classical genetic mutation. This pattern of inheritance had only been observed previously for genetic elements housed outside the nucleus, such as in mitochondrial and viral genomes. Unlike those elements, however, this trait can also arise spontaneously in yeast. And the phenotype it confers closely mimics that produced by an absence of the nuclear gene called URE2. Paradoxically, URE2 must be expressed for the trait to occur in yeast cells.

In 1994, Reed Wickner of the National Institutes of Health made a key observation: the bizarre yeast trait appeared at a higher frequency when URE2 was overexpressed. He further determined that the Ure2 proteins specifically, rather than the DNA and RNA molecules encoding them, were responsible for this effect. Wickner then offered the only explanation compatible with the seemingly incongruous results: the information responsible for the trait’s inheritance was encoded by an alteration in the Ure2 protein itself.6 In other words, Ure2 forms a prion, which reproduces itself through the conversion of other Ure2 molecules in the cytoplasm. During cell division, the prion particles divide among daughter cells, thereby perpetuating the trait in a non-Mendelian manner. (See illustration.)

Wickner’s insight broadened the prion concept for the first time beyond mammals, in which prions were first identified as the agents of mad cow and related diseases, and beyond the idea of a singular prion protein associated with those diseases. Moreover, yeast cells that harbored these elements grew at comparable rates to those that did not, suggesting that prions may not be universally pathogenic and may even act as protein-based elements of inheritance in healthy organisms.

Since Wickner’s original discovery, scientists have exploited the superior tractability and genetic resources of yeast to look for additional prion-forming proteins. To date, researchers have determined that approximately two dozen proteins can form prions in this organism. And at least some of them appear to do so quite frequently. At least one-third of wild yeast isolates harbor traits that can be attributed to prions.7 These traits include changes in cell-cell adhesion, resistance to antibiotics, and alterations in the way yeast cells use nutrients. Why might yeast harbor so many of these elements? More importantly, do other organisms also commonly harbor nonpathogenic prions?

The proteins that form prions in yeast tend to be involved in processes that regulate the flow of genetic information in the cell, such as transcription, RNA processing, and translation. Consequently, alterations in these processes by prion formation affect the way that such information is expressed, causing changes in phenotype that can have dramatic consequences for the cell. Sometimes the changes are beneficial. Other times they are detrimental. In either case, prions provide an added level of variation that may give the population a greater chance of surviving an otherwise dooming environmental change. In this sense, populations of yeast cells appear to be exploiting prions as phenotypic “bet-hedging” devices, to ensure that they collectively harbor the phenotypic diversity necessary to survive a dynamic and often harsh environment. Of course, with the risk that prions can turn pathogenic, this raises the question of why cells might take such measures for this purpose, rather than relying on common and relatively innocuous sources of phenotypic diversity, such as genetic variation and transcriptional “noise.”

In my own lab at the University of Texas (UT) Southwestern Medical Center, we sought to answer this question by looking at the types of genetic information most heavily regulated by prions. Not surprisingly, we found that prion-forming transcription factors each target a disparate group of genes. However, one particular yeast gene, FLO11, had a pronounced tendency to be subject to prion regulation. FLO11 encodes a protein involved in cell surface adhesion, functioning like molecular Velcro to enable cells to stick to one another in various arrangements. In effect, the Flo11p protein allows yeast cells to temporarily abandon their normally solitary lifestyles in favor of a more communal existence.8

Prions may not be universally pathogenic and may even act as protein-based elements of inheritance in healthy organisms.

The benefits of Flo11p-dependent multicellularity are multifold.8 In some cases, Flo11p allows cells to remain attached after they divide, enabling them to collectively orient their growth toward more hospitable environments. Other times, such as during the fermentation of Sherry wine, yeast cells form Flo11p-dependent mats that catch their own carbon dioxide bubbles, allowing the yeast to rise to the surface of the grape must where oxygen is more plentiful. Importantly, all Flo11p-dependent multicellular behaviors require the participating cells to act in a cooperative fashion. If some cells do not produce the adhesion protein, they risk compromising the integrity of the entire structure. Therefore, the success of the strategy—and the fitness payoff for adopting it—hinges on the commitment of each cell to a stable, multigenerational developmental program.

The self-perpetuating nature of prions is an ideal way for cooperating lineages to enforce that commitment. We found that when the transcription factor Mot3p converts to a prion form, it activates FLO11 (among other genes) and stably perpetuates the multicellular state in subsequent generations.9 By virtue of their common descent and shared inheritance of the prions, all cells in the lineage act together to achieve a common goal. Mot3p, along with multiple other prion proteins that regulate FLO11, is tuned to respond to environmental signals, such as changes in pH or depletion of nutrients, and to exert unique regulatory responses. This likely affords yeast a measure of plasticity in the way that they deploy multicellularity, enabling lineages of cells to physically morph to fit the unique demands of new environments.

Prions as saviors

DIVERSE EFFECTS OF PRIONS: Prions are most well known for their role in disease. Spongiform encephalopathies, such as mad cow disease and scrapie in sheep, are the result of a toxic accumulation of prions in the brains of these animals. In humans, prions have been identified as the cause of the fatal Creutzfeldt-Jakob disease (the human version of mad cow disease), and scientists speculate that they underlie many other neurodegenerative disorders, including Alzheimer’s, Huntington’s, and Parkinson’s. But studies in yeast point to a wide range of prion activity in healthy cells, suggesting the reactive proteins may perform functions critical to normal cellular life.
See full infographic: JPG | PDF
For many aspects of cell biology, what has been true in yeast is often true in higher organisms and even humans. Do human cells also take advantage of prion switches?

The first hint that we may indeed harbor beneficial prions surfaced in 2003, when Eric Kandel and colleagues at Columbia University noticed that a neuronal protein called cytoplasmic polyadenylation element binding protein (CPEB) has a sequence that is remarkably reminiscent of yeast prion proteins. Working with an isoform of CPEB from the sea slug Aplysia californica, they discovered that the protein plays a key role in the formation of long-term memories. Following a hunch, the researchers then used yeast cells to test whether the protein could in fact form prions. It did.10 Since then, evidence has accumulated that CPEB and functionally related proteins may indeed behave like prions in the neurons of Aplysia, fruit flies, and even humans.11 These prions presumably not only perpetuate themselves, but also localize the synthesis of important proteins to synapses in response to neurotransmitter stimulation. In other words, neurons appear to have co-opted the self-sustaining prion switch as a key mechanism for the formation of memories.

Since this discovery, there have been a number of other suspected prion proteins identified in higher organisms. The most definitive evidence for a functional prion switch in humans was recently provided by James Chen and colleagues here at the UT Southwestern Medical Center. Chen’s team studies the body’s innate immune system, the ability to distinguish self from nonself and to mount the appropriate responses to defeat pathogens before they take hold. In the course of their research, the team found that one particular protein, called MAVS, which is key to our innate ability to fight certain viral infections, acquires a self-perpetuating fibrillar form in cells that have become infected with virus.12 The rapid prion-like templated conversion of MAVS amplifies the cellular alarm signal and ultimately induces the production of interferons that recruit macrophages and other immune factors to combat the infection.

GROWING FIBRILS: Prion proteins come together to form amyloid fibers (pictured here), which themselves can cluster into prion particles.
Here, a prion-based molecular switch may be exactly what the situation calls for. The self-sustaining and irreversible nature of prions endows MAVS with an executive role in the immune response. Once a virus is detected, the cell must resolve to stop that virus from hijacking its own machinery, even if it means signaling its own destruction in the process. Failing to do so risks compromising the health of the entire organism. Other proteins in the same superfamily as MAVS also seem to have prion-like properties.13 And two other proteins that are not related but have similar functions—the RIP1 and RIP3 kinases—polymerize into filamentous amyloids that are necessary to induce programmed necrosis, a form of cellular suicide important for normal body development. Thus, ironically, it is now clear that the same process that causes devastating neurodegenerative diseases also lies at the heart of our innate ability to fend off other diseases.

Across biological systems, prions appear to determine cell fate, for better or for worse. For some proteins in the human brain, the aberrant formation of prions leads to an irreversible and agonizing decline in cognitive capacity and ultimately to death. For others, prion formation may be critical to the function of the brain itself. Elsewhere in our body, prion-like switches commit cells to different forms of programmed suicide that are critical for normal development and that protect us from infectious disease. And in yeast, prion formation appears to be linked to one of the most significant evolutionary transitions in life’s history—the emergence of multicellularity. Perhaps the real-life ice-nine is not all bad after all. 

Randal Halfmann is a Sara and Frank McKnight independent postdoctoral fellow at the University of Texas Southwestern Medical Center in Dallas.


  1. P.T. Lansbury, Jr., B. Caughey, “The chemistry of scrapie infection: implications of the ‘ice 9’ metaphor,” Chem Biol, 2:1-5, 1995.
  2. S.B. Prusiner, “Novel proteinaceous infectious particles cause scrapie,” Science, 216:136-44, 1982.
  3. F. Wang et al., “Generating a prion with bacterially expressed recombinant prion protein,” Science, 327:1132-35, 2010.
  4. R.A. Kyle, “Amyloidosis: a convoluted story,” Br J Haematol, 114:529-38, 2001.
  5. C. Soto, “Transmissible proteins: expanding the prion heresy,” Cell, 149:968-77, 2012.
  6. R.B. Wickner, “[URE3] as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae,” Science, 264:566-69, 1994.
  7. R. Halfmann et al., “Prions are a common mechanism for phenotypic inheritance in wild yeasts,” Nature, 482:363-68, 2012.
  8. S. Bruckner, H.U. Mosch, “Choosing the right lifestyle: adhesion and development in Saccharomyces cerevisiae,” FEMS Microbiol Rev, 36:25-58, 2012.
  9. D.L. Holmes et al., “Heritable remodeling of yeast multicellularity by an environmentally responsive prion,” Cell, 153:153-65, 2013.
  10. K. Si et al., “A neuronal isoform of the Aplysia CPEB has prion-like properties,” Cell, 115:879-91, 2003.
  11. M.G. Thomas et al., “Synaptic control of local translation: the plot thickens with new characters,” Cell Mol Life Sci, doi:10.1007/s00018-013-1506-y, 2013.
  12. F. Hou et al., “MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response,” Cell, 146:448-61, 2011.
  13. H. Wu, “Higher-order assemblies in a new paradigm of signal transduction,” Cell, 153:287-92, 2013.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?