What Are Prions? Infectious Proteins Without DNA

A prion is an infectious agent that contains no DNA and no RNA. It is a protein that has folded into the wrong shape and can then push normal copies of the same protein into that shape too. Each converted molecule becomes a new seed, so a few misfolded proteins can grow into enough to wreck brain tissue. Prions cause scrapie in sheep, kuru in highland Papua New Guinea, bovine spongiform encephalopathy (mad cow disease) in cattle, and Creutzfeldt-Jakob disease (CJD) in people. This article covers how the idea was proven, how the mechanism works, and what is still debated.

This is a mechanism explainer, not medical guidance. Anyone worried about a specific symptom or exposure should talk to a physician.

An infection with no genome

Every other infectious agent known by the 1980s carried nucleic acid. Bacteria have DNA, viruses have DNA or RNA, and even the odd little plant pathogens called viroids are short strands of RNA. Heredity ran through nucleic acid, a conclusion built on landmark work such as the Hershey-Chase experiment and the Avery-MacLeod-McCarty experiment. The way DNA stores and passes on instructions was the foundation of molecular biology.

Against that background, the diseases called transmissible spongiform encephalopathies looked strange. Scrapie, a degenerative disease of the central nervous system in sheep and goats, could be passed from one animal to another after a long incubation period. The brain tissue of affected animals developed a sponge-like look under the microscope. Researchers assumed a slow virus was responsible, and they could not find one.

Prusiner's 1982 paper

In April 1982, Stanley Prusiner published "Novel proteinaceous infectious particles cause scrapie" in Science. The abstract reports that the scrapie agent is inactivated by alkali but that five procedures aimed more specifically at modifying nucleic acids failed to inactivate it. Six lines of evidence, including sensitivity to protein-digesting enzymes, indicated that the agent contains a protein required for infectivity. Because its behavior differed from viruses, plasmids, and viroids, Prusiner proposed a new term, "prion," for a small proteinaceous infectious particle resistant to most procedures that modify nucleic acids.

The claim that an infection could be spread by a protein alone was a big departure from established biology, and it took years of work to persuade the field. Prusiner received the 1997 Nobel Prize in Physiology or Medicine for his work on prions, a point recorded in a history of kuru in Hektoen International.

Kuru: the disease that made the case

Before prions had a name, an epidemic in the highlands of Papua New Guinea showed what they could do. Kuru affected the Fore people, and the word means trembling. According to the Hektoen history, oral testimony places the first appearance of the disease in the 1910s, and patrol officers and anthropologists reported many ill women and children in the 1950s. The Fore attributed it to sorcery. The press called it the laughing sickness because some patients displayed a fixed grin.

In 1957, district medical officer Vincent Zigas introduced Carleton Gajdusek to the problem, and Gajdusek published the first medical description. Michael Alpers joined the work in 1961. Researchers noticed that kuru resembled scrapie: similar symptoms, a sponge-like brain, transmissibility, and very long incubation periods.

Why women and children

Kuru fell most heavily on women and children, and the explanation lay in funeral custom. The University of Utah's genetics education site notes that the disease spread through rituals in which women and children ate their dead relatives, which is why men were rarely infected. Hektoen's history adds that the Fore believed this carried the essence of the dead into living relatives, and that Australian authorities banned the practice in the late 1950s.

The proof of transmissibility came from primates. Brain tissue from Kigea, an 11-year-old patient, was injected into a chimpanzee named Daisey. Daisey and a second chimpanzee, Georgette, later developed kuru. That experiment established that the disease was infectious and that the agent resided in brain tissue.

Gajdusek received the 1976 Nobel Prize for his work. He had framed the agent as an "unconventional virus," because nothing then known could explain it otherwise.

What the long incubation shows

Incubation periods in kuru were very long, in some cases longer than a human lifespan, according to the Hektoen article. No one born after 1960 developed kuru, and the last cases were reported in 2003, 2005, and 2009. In 1970, Alpers proposed that the epidemic might have begun from a spontaneous case of CJD, then spread through the feasts. That makes kuru a chain of infection that ran on the practice of eating brain tissue, and ended when the practice did.

A genetic twist

A 2009 paper in the New England Journal of Medicine identified a protective variant of the prion protein gene, called G127V, among Fore people in the kuru region, and reported that families carrying it had a significantly lower incidence of kuru. An earlier 2003 paper in Science found that the epidemic had imposed strong selection on the Fore, essentially eliminating people with two identical copies (homozygotes) of the gene at position 129. Later researchers disputed the paper's separate worldwide claim of ancient balancing selection. The disease left a mark on the population's DNA, which is ironic for an agent that has none of its own.

The normal protein and the wrong shape

The gene that matters is called PRNP, and it codes for the prion protein, PrP. Every healthy person makes it. The Utah genetics page calls the normal version PrP-sen and the misfolded version PrP-res. Researchers more often write PrPC for the cellular, normal form and PrPSc for the scrapie form.

The difference between the two is shape, not sequence. They have the same amino acids in the same order. Prusiner's 1998 review in PNAS describes the normal protein being converted into the disease form through a post-translational process in which it acquires a high beta-sheet content, and describes the disease form acting as a template on which the normal protein is refolded. A beta-sheet is a flat, ribbon-like arrangement of a protein chain, and beta-sheets tend to stack on each other. Think of a pile of flat plates that sticks together versus loose tangled string.

Templated conversion, step by step

In plain terms, the misfolded protein clings to normal prion protein and changes its conformation, then builds up in brain tissue. Prions contain no genetic information and need no genes to reproduce themselves. The steps run like this:

1. A misfolded PrP molecule meets a normal one on or near the surface of a neuron.
2. The misfolded molecule binds it and nudges it into the misfolded arrangement.
3. The new misfolded molecule can now do the same to another normal one.
4. Clumps accumulate, and neurons die, leaving holes in the brain tissue that give "spongiform" disease its name.

A useful comparison is the domino rule. Knock one over and each one that falls knocks over the next. Unlike dominoes, though, each fallen piece is also a new striker, so the number of active strikers grows. That is how a tiny seed can expand into a large pile of misfolded protein.

Feature	Normal PrP (PrPC)	Misfolded PrP (PrPSc)
Amino acid sequence	Same	Same
Dominant fold	Mostly alpha-helical	High beta-sheet content
Present in healthy people	Yes	No
Behavior	Stays as single molecules	Binds and converts normal copies, and aggregates
Contains nucleic acid	No	No

What is still unclear

The Utah page points out that, in 2005, researchers at the University of Texas made the misfolded form in a test tube and injected it into healthy brains, causing spongiform disease, which supports the protein-only idea. The same page says that it is still unclear what triggers the normal protein to fold into the disease form in the first place. In most cases of CJD the cause of that first misfolding event is unknown, a fact that drives the sporadic cases discussed below.

What the disease looks like from the outside

Seen from outside the brain, prion diseases share a recognizable arc: a long silent stretch, then a rapid decline. The CDC lists the signs of CJD as dementia, trouble walking, sudden jerky movements, and visual disturbances. Kuru had its own presentation, which the Utah genetics site summarizes this way: victims first lost the ability to walk, then the ability to swallow or chew, and weight loss led to death.

The names of the diseases come from what pathologists see under the microscope. Misfolded prion proteins damage the brain by killing neurons, and the loss of cells leaves small holes in the tissue, which gives the sponge-like appearance that "spongiform" describes. Notice the order of events: the protein aggregates first, the neurons die next, and the tissue changes show up last. By the time the holes are visible, the damage is far advanced.

Why the delay is so long

A long incubation fits the logic of templated conversion. One plausible reading is that, early on, few misfolded molecules exist and the total builds slowly, which would be consistent with what the BSE and kuru data show: very long incubation periods, in some kuru cases longer than a human lifespan, followed by a clinical course that the CDC says usually ends in death within a year for CJD.

This arithmetic has a consequence for epidemiology. When an exposure ends, as it did with the feed ban in 1988 and the end of the funeral feasts, cases continue for years or decades because people and animals already carry the seed. Declines lag behind the intervention that causes them. Pattison's review notes that clinical BSE cases fell only about five years after the feed ban, while infections themselves dropped immediately.

What is not yet settled

Researchers do not agree on exactly how misfolded protein kills neurons. The aggregates might be toxic themselves, or an intermediate form might be, or the loss of the normal protein's function might contribute. The Utah page says that it remains unclear what triggers the first misfolding event, and the evidence summarized in the tau review shows that even in related diseases the link between a spreading protein and neuron death is not firmly established. Stating that plainly is part of describing the field accurately: the templating mechanism is well supported, and the details of toxicity are not fully resolved.

How strong is the protein-only evidence?

Because the claim was so unexpected, it is worth listing what supports it, in the order that it accumulated:

• Nucleic acid treatments fail. In 1982 Prusiner reported that five procedures aimed at nucleic acids did not inactivate the scrapie agent, while procedures that damage proteins did reduce infectivity.
• A specific protein is involved. The same paper reported six lines of evidence that a protein is required for infectivity.
• Conformation differs. The normal protein is largely alpha-helical, while the disease form has a high beta-sheet content, as reviewed in Prusiner's 1998 article.
• Making it in a tube works. In 2005, University of Texas researchers produced the misfolded form in a test tube, injected it into healthy brains, and caused spongiform disease, according to the Utah page.
• Genetics fits. Mutations in the prion protein gene cause familial disease (CDC), and gene variants such as the one at codon 129 and the G127V change in Fore people alter susceptibility.

Taken together, these threads converge on a single idea. No one piece settles it, but the combination is why the protein-only model won acceptance, and why Prusiner's Nobel Prize followed in 1997.

Mad cow disease and the human form

Kuru was a confined epidemic. Bovine spongiform encephalopathy (BSE) became a continental one. A 1998 review in Emerging Infectious Diseases by Pattison summarizes the timeline. BSE was first identified in indigenous UK cattle in 1986, and the epidemic peaked in 1992. Through the end of 1997 about 170,000 cases had been reported in the United Kingdom, among roughly one million infected animals. The source of infection was meat and bone meal in concentrated cattle feed, meaning that animals were being fed processed remains of other animals.

Feed bans and why they came late

A ban on ruminant protein in ruminant feed took effect in July 1988. Infections dropped right away, and clinical cases began to decline about five years later, which fits the long incubation. But the ban was not fully effective: almost 36,000 cattle were born after it, some as late as 1994. Further restrictions followed over the next decade, among them a ban on specified bovine offal in human food in November 1989 and a worldwide EU ban on UK beef in March 1996.

The agent did not stay in cattle. Feline spongiform encephalopathy was first diagnosed in 1990, and the Pattison review notes that exotic zoo ruminants and some big cats were also affected. The prion crossed species lines when animals ate contaminated tissue.

Variant CJD

In May 1995, the first adolescent death from CJD in the United Kingdom was reported. By March 1996, a new variant of CJD had been recognized. Pattison's 1998 review, based on mouse transmission studies in which incubation periods and brain lesion patterns matched, states that variant CJD can now be regarded as human BSE. At the time of that article, 26 UK cases and one French case had been diagnosed, and every UK case carried two methionine copies at codon 129 of the prion protein gene. The mechanism proposed for exposure was human consumption of tissues with the highest concentrations of the agent, a working hypothesis at the time.

The UK's national CJD surveillance unit, in its February 2025 update, lists 178 deaths from variant CJD, with the yearly count running from 1995 to a single death in 2016 and none of the patients still alive. Set beside roughly 170,000 reported cattle cases, and an estimated million infected animals, the number shows how uneven the path from exposure to illness can be for a transmissible protein.

The main prion diseases at a glance

Prion diseases are rare. The CDC reports that CJD occurs at about one to two deaths per million people per year worldwide. In the United States, roughly 500 to 600 cases are reported each year. The disease usually kills within a year of symptoms starting, and there is no therapy that can slow or stop it; care focuses on comfort.

Type	How it arises	Notes
Sporadic CJD	Cause unknown, apparently a spontaneous misfolding	About 85 percent of CJD cases (CDC)
Familial (genetic)	Inherited mutation in the prion protein gene	About 5 to 15 percent of CJD cases (CDC)
Acquired	Exposure to prion-contaminated material	Includes kuru, variant CJD, and cases traced to contaminated surgical or medical equipment

The three origins match the three ways a rare event can happen. A protein can misfold on its own by chance, which is the sporadic case. An inherited mutation can make misfolding much more likely. Or a person can be exposed to misfolded protein from outside. The mutation route shows that the gene matters even though the agent itself has no gene.

Why prions are hard to remove

Ordinary sterilization targets nucleic acid, membranes, and the proteins of microbes. Prions are protein aggregates with a very stable fold, and they resist many of those procedures. Prusiner's original paper noted exactly this: the agent survived procedures that modify nucleic acids.

Hospitals that treat reusable surgical instruments after exposure to high-risk tissue have used stricter protocols. WHO infection-control guidance for these diseases has described soaking non-disposable instruments in sodium hydroxide or a strong bleach solution and then autoclaving at a higher temperature than routine, and it notes that ordinary steam cycles that reliably kill bacteria and viruses are not dependable against prions. Infection-control literature treats discarding and incinerating contaminated instruments as the alternative where that is practical. That is a statement about hospital practice, not advice for individuals.

Prion-like spreading in other brain diseases

Alzheimer's, Parkinson's, and related disorders involve other misfolded proteins that also clump in the brain: amyloid-beta and tau in Alzheimer's, alpha-synuclein in Parkinson's. A natural question is whether these spread by the same templating trick. The honest answer is "partly established, partly debated."

A review on PubMed Central that asks whether tau pathology spreads through prion-like propagation finds ample, albeit circumstantial, evidence. In cell culture, tau seeds can be taken up by cells and induce normal tau to aggregate. In animals, tau pathology shows up in brain regions connected to the injection site, which suggests spread along neural connections. Evidence in patients is mostly indirect, and the authors call the overall support ample but circumstantial: the pattern of pathology across brain regions that fits spread can also be explained by some regions being more vulnerable than others.

Prion-like is not the same as infectious

The same review makes a distinction that matters. Bona fide prions show transmission from cell to cell, tissue to tissue, and organism to organism, and they propagate distinct misfolded conformations stably. Tau has not been shown to meet all those steps, so the authors caution against equating prion-like behavior with prion infectivity. In plain terms, the templating mechanism might help explain how Alzheimer's progresses through the brain, but that is a different claim from saying Alzheimer's is contagious. The biology of the disease is covered in the article on Alzheimer's disease, and the broader question of how nerve cells communicate and fail is in how the nervous system works.

Why one sequence can fold two ways

One reason prions fascinate biologists is that they break a comfortable assumption: that a protein's amino acid sequence fixes a single stable fold. Prediction tools such as AlphaFold have done remarkable work by learning how sequences map to shapes, and the problem of how proteins fold is a field in itself. A prion protein shows that the answer can be more than one shape for the same sequence, one of which can pull its neighbors along.

The lesson for how to think about proteins is that the sequence defines which shapes are possible, but which one a protein ends up in can depend on history, on neighbors, and on chance.

Common misunderstandings

"A prion is a kind of virus." It is not. A virus has a genome and uses a host cell to copy it. A prion has no genome, and what it copies is a shape. The category of agents that jump from animals to humans includes both, but the mechanisms differ.

"Prions are alive." Prions do not metabolize, grow, or reproduce in the way living organisms do. They are better described as self-propagating protein conformations.

"All prion diseases come from eating infected meat." Most CJD cases are sporadic or familial, as the CDC figures show: about 85 percent sporadic and 5 to 15 percent familial. Acquired forms are the minority, and diet is only one route within them.

"Germ theory explained everything about infection." Germ theory rests on microbes that carry their own genetic material. Prions extended the concept of an infectious agent, but they did not overturn the evidence that bacteria and viruses cause the diseases they cause.

"Prion diseases are like the deadliest outbreaks." They are lethal but rare, and the documented routes of acquisition are specific ones, such as eating contaminated tissue or exposure to contaminated medical equipment. For comparison, see the discussion of the deadliest viruses.

The takeaway

Prions force a wider definition of what counts as infectious. What gets copied in a prion disease is a shape: one misfolded protein recruits a normal one, and the new copy recruits the next. That explains why an agent that survives nucleic-acid-destroying treatments can still replicate, why disease can lie hidden for decades, and why kuru ended when a funeral custom did. It also explains a surprising feature of the epidemiology, that most cases arise with no outside source at all. The open scientific question is how far the same templating logic extends into other brain diseases, where the evidence is substantial in cells and animals and still incomplete in people. For the surrounding biology, the Complete Biology Guide maps where these ideas fit.

Frequently asked questions

Q: What is a prion in simple terms? A: A prion is a misfolded version of a normal brain protein, PrP, that can convert other normal copies into the same misfolded shape. It contains no DNA or RNA. Because each converted molecule becomes a new template, misfolded protein accumulates and damages brain tissue.

Q: How can a protein be infectious without DNA or RNA? A: What it passes on is a shape, not information stored in a genetic code. The misfolded protein binds the normal form and changes its conformation, rich in beta-sheet, so the new copy can do the same. The mechanism is called templated conversion, and a 2005 test-tube experiment supported the protein-only idea.

Q: What causes kuru? A: Kuru was a prion disease among the Fore people of Papua New Guinea. It spread through funeral rituals in which relatives ate the tissue of the dead, which is why it fell mostly on women and children. After the practice stopped, cases declined, and no one born after 1960 developed it.

Q: What is the connection between mad cow disease and variant CJD? A: Variant CJD is regarded as human BSE. BSE was first identified in UK cattle in 1986, and the first adolescent CJD death was reported in May 1995. Mouse transmission studies with matching incubation periods and lesion patterns supported the link. The UK has recorded 178 variant CJD deaths.

Q: How common are prion diseases? A: Very rare. CJD occurs at about one to two deaths per million people per year worldwide, according to the CDC. Roughly 85 percent of cases are sporadic, and 5 to 15 percent are familial, caused by a mutation in the prion protein gene.

Q: Are Alzheimer's and Parkinson's prion diseases? A: No. Misfolded proteins in these diseases show prion-like spreading between cells in lab and animal studies, but reviewers distinguish that from prion infectivity. Bona fide prions are transmissible from organism to organism, and tau has not been shown to meet that standard.

Q: Is there a treatment for prion diseases? A: The CDC states that no therapy can slow or stop CJD, and that care focuses on keeping the patient comfortable. Researchers are studying possible approaches, but none has established benefit.

Sources

• A history of kuru, Hektoen International (kuru, Gajdusek, Alpers, Prusiner Nobel Prizes, case dates).
• Prions, Genetic Science Learning Center, University of Utah (PrP-sen and PrP-res, 2005 test-tube experiment, kuru).
• Creutzfeldt-Jakob Disease, CDC (incidence, sporadic and familial proportions, no treatment).
• The Emergence of Bovine Spongiform Encephalopathy and Related Diseases, Pattison, Emerging Infectious Diseases, 1998 (BSE timeline, feed ban, variant CJD).
• What is the evidence that tau pathology spreads through prion-like propagation?, PubMed Central (prion-like spread, criteria for bona fide prions).