Prions

Your body is full of tiny machines called proteins. Each one is a long chain that folds up into a particular shape, and the shape is what lets it do its job. Fold it wrong and it usually just stops working, and the body clears it away.

A prion is different, and that is what makes it frightening. It is a protein that has folded into a wrong shape which happens to be catching. When a wrongly folded prion bumps into a normal one of the same kind, it forces the normal one to twist into the wrong shape too. Now there are two bad ones. Each of those can convert more. One becomes two, two become four, and the wrong shape spreads like a rumour.

No DNA is involved. No virus, no bacterium, nothing alive. It is just a shape making copies of itself out of the proteins already sitting in your cells. That idea was so strange that when a scientist first suggested it, most people did not believe him.

The misfolded proteins are sticky, so they clump together into lumps the body cannot break down. In the brain those lumps slowly kill nerve cells and leave the tissue full of tiny holes, like a sponge. The diseases this causes are rare and always fatal, and there is no cure yet. Mad cow disease is one of them, and people caught a human version of it by eating infected beef, which is why it became such a scare in the 1990s.

Proteins are chains of amino acids that fold into a precise three-dimensional shape, and the shape is the function. The prion protein, PrP, is a normal protein you make throughout life, most of it on the surface of nerve cells. In its healthy form, written \(\mathrm{PrP^{C}}\), it is rich in the corkscrew coils called alpha-helices.

A prion is that same protein, same sequence of amino acids, folded into a different and stable shape, written \(\mathrm{PrP^{Sc}}\). In this form much of the coil has flattened into beta-sheet, a stacked, pleated structure that loves to stick to more copies of itself. The wrong shape is not a broken protein that falls apart. It is an alternative fold that happens to be a trap.

Here is the crucial part. \(\mathrm{PrP^{Sc}}\) acts as a template. When it meets a healthy \(\mathrm{PrP^{C}}\), it lowers the barrier for that molecule to refold into the same misfolded shape. So the count of bad copies doubles, and doubles again, an exponential chain reaction that needs no genetic material at all. That is what made the idea so radical when Stanley Prusiner proposed it in 1982; he received the Nobel Prize for it in 1997.

The beta-sheet copies stack edge to edge into long fibres called amyloid, and those pile up into deposits the cell cannot clear. In the brain the deposits kill neurons and riddle the tissue with microscopic holes, which is why these are called spongiform diseases. Because the infectious agent is only a protein, it shrugs off the things that kill germs. Ordinary cooking, radiation and standard hospital sterilisation do not reliably destroy it.

And because it is a shape rather than an organism, it can pass from animal to animal, and even between species when their prion proteins are similar enough. That is the story of bovine spongiform encephalopathy, mad cow disease, which crossed into people who ate infected tissue and appeared as variant Creutzfeldt-Jakob disease in the 1990s. The same protein, misfolded, sits behind scrapie in sheep and kuru, a disease spread through funerary cannibalism in Papua New Guinea.

The protein-only hypothesis. The heretical claim is that the infectious agent carries no nucleic acid: the information that propagates the disease lives in the conformation of a protein, not in a genome. \(\mathrm{PrP^{C}}\) and \(\mathrm{PrP^{Sc}}\) share an identical amino-acid sequence and differ only in fold, alpha-helical and monomeric for the cellular form, beta-sheet rich and aggregation-prone for the scrapie form. The particle resists nucleases and UV at doses that would inactivate any virus, yet is knocked down by agents that denature protein. Mice engineered to lack the \(\textit{PRNP}\) gene make no PrP, and they cannot be infected, which nails the substrate: the prion converts the host's own protein.

Templated conversion and nucleation kinetics. Propagation is a seeded, autocatalytic process. A \(\mathrm{PrP^{Sc}}\) aggregate recruits \(\mathrm{PrP^{C}}\) onto its growing end and templates its refolding, so the fibril elongates; fragmentation then snaps one long fibril into many shorter ones, multiplying the number of active ends. Elongation plus fragmentation is what turns linear growth into exponential growth of infectious units, and it explains the long, silent incubation: a lag phase while seeds are scarce, then a steep rise once the seed population has amplified. It is the same nucleation-polymerisation physics seen in other amyloids.

Strains in a protein. One protein sequence supports several distinct, self-propagating \(\mathrm{PrP^{Sc}}\) conformations, and these behave as strains: they give reproducible incubation times, brain-lesion patterns and biochemical signatures, and each faithfully templates its own shape. Heritable information is being carried by structure rather than by sequence, which is the deepest and strangest feature of the whole system.

The species barrier. Transmission between species is inefficient when donor and recipient PrP sequences differ, because a template converts a near-matching substrate far better than a mismatched one. The barrier is not absolute; bovine prions crossed to humans as variant CJD, and the strain's conformation, not just the sequence match, sets how readily it jumps.

Three routes to the same disease. Human prion disease is sporadic, when a \(\mathrm{PrP^{C}}\) molecule spontaneously misfolds, the commonest form of classic Creutzfeldt-Jakob disease; inherited, from dominant mutations in \(\textit{PRNP}\) that destabilise the fold, as in familial CJD, Gerstmann-Straussler-Scheinker syndrome and fatal familial insomnia; or acquired, through infected tissue, as in kuru, iatrogenic transmission by grafts and instruments, and dietary variant CJD. One misfolding mechanism, reached by chance, by genetics, or by infection.

Why it matters beyond the rare diseases. Prion-like templating is turning out to be a general motif. Yeast carry benign, heritable prions such as [PSI+] that switch traits without any DNA change, and the self-propagating spread of misfolded aggregates through tissue is now a working model for how tau and alpha-synuclein march through the brain in Alzheimer's and Parkinson's. Those proteins are not known to be infectious between people the way \(\mathrm{PrP^{Sc}}\) is, but the underlying trick, a bad shape seeding more of itself, looks like something biology reuses. There is still no cure for a classic prion disease, and its resistance to sterilisation keeps it a genuine hazard in surgery and in the food chain.