PRION PROTEINS, in the conformation shown, are normal proteins found in all healthy cattle, sheep and humans.

Thousands of people in the United Kingdom ate beef and drank milk from cattle infected with bovine spongiform encephalopathy (BSE), or "mad cow disease," in the late 1980s and early 1990s. No one was overly concerned at the time because it was, after all, a disease that seemed to only affect cows.

Then in the mid-1990s, the first cases of a novel form of a deadly human version of BSE, called new variant Creutzfeldt-Jakob disease (vCJD), surfaced in the U.K. And for the first time, the government seriously considered the possibility that the disease might be transmitted from cattle to humans through the food chain. It has been a lingering fear ever since, and researchers are trying hard to find a cure in case the disease turns into an epidemic.

Both BSE and CJD, as well as scrapie in sheep, are terrible neurodegenerative disorders; behavioral and neuromuscular problems--symptoms that look like madness--arise as the disease drills spongelike holes in the brain. Previously known forms of CJD are not transmitted directly: they are inherited, set off spontaneously, or caused by operations or injections involving infected tissues or instruments.

What has been learned is that CJD, BSE and scrapie are so-called prion diseases, probably solely caused by prions, or "proteinaceous infectious particles" that do not contain nucleic acids like all other infectious agents. Prions consist mostly of an abnormal form of the otherwise ordinary prion protein (PrP). Once abnormal PrP has entered an organism, it most likely spreads by forcing normal PrPs to alter their shape, thus converting them to the disease-causing form.


BRAIN TISSUE of a sheep that died of scrapie shows changes seen in all of the spongiform encephalopathies. Large empty spots, called vacuoles, appear in the cytoplasm of two neurons. It is these cavities that make the tissue resemble a sponge, hence the name "spongiform."

Nobody knows the number of potentially deadly prions that lingered in beef from infected cows, or how many people in the U.K. bear the vCJD agent as a consequence. Nor are scientists aware of how long the incubation period lasts. To date, 56 people in the U.K. and two in France have fallen victim to the new form of the "killer protein." And that may not be the ultimate death toll.

Scientists in the U.K. recently tried to get a better estimate of the problem's scope by testing for the abnormal prion protein in more than 3,000 appendix and tonsil samples, removed from patients in the late 1990s. These tissues were--unlike those collected from vCJD victims--free of the "bad" prion protein. Still, 3,000 is not a large enough number to warrant any great sighs of relief. Some 15,000 further samples are waiting to be assayed, and only a few positives among them would indicate thousands of future cases.

If such cases do arise, scientists hope to be ready. Several teams are currently studying how prion disease progresses in animals to find a way to stop it. In the May 19 issue of Science, researchers from Switzerland and England reported a way to delay the spread of prions in mice from the body to the central nervous system, where they cause scrapie. Scientists had already noticed that when they injected prions into the bellies of mice, the proteins infected primarily the spleen and lymph nodes before they entered the brain. The spleen contains so-called follicular dendritic cells (FDCs), cells of the immune system that seem to be particularly important for nurturing the abnormal prion proteins because they contain a lot of the ordinary PrP that can "turn around."


INFECTIOUS PRION PROTEINS have a different shape, which they impose on normal prion proteins in a chain reaction that ends in sickness and death.

In order for FDCs to develop in the spleen, they need to receive signals that originate from a molecule called lymphotoxin (LT) alpha/beta on the surface of B cells. The researchers blocked this signal by injecting a molecule into the mice that binds to lymphotoxin-alpha/beta. The results were twofold: FDCs disappeared almost completely from the spleen, and there were no prion proteins detectable in the organ.

To gauge the sensitivity of the LT blocker, the researchers gave them to mice both before and after they were infected with prions. Animals who received the blocker for eight weeks, beginning a week after the prion injection, developed scrapie several days later than untreated mice. Because their spleen could still infect other mice, it must have harbored small amounts of the bad prion protein. But most mice who started treatment with the LT blocker a week before the infection developed scrapie weeks after their untreated counterparts. More important, during the time of treatment, their spleens did not make other mice sick, at least not for five months, and so likely did not contain any bad prion protein.

Still, the animals were not entirely protected. Virtually all mice in the experiment succumbed to the disease in the end--probably because both the FDCs and the prion proteins reappeared in the spleen after the end of the treatment period. There is, however, another possibility that the researchers cannot exclude. The prions, once barred from the spleen by the LT blocker, could have taken a different route to get into the central nervous system and brain.

Despite the fate of these poor mice, the results hold significant promise. Prions causing vCJD act like those causing scrapie in mice in that they appear in the lymphoreticular system before they reach the brain. Thus, blocking FDC development in the spleen might be a way to at least delay the onset of the disease in humans. "We observed no ill effects in the mice treated for several weeks with the fusion protein," says Charles Weissmann from Imperial College in London, one of the authors of the Science article. "Whether or not there would be a serious [side] effect on humans upon long-term treatment I cannot predict."

Weissmann stresses that their approach would be useful only if clinicians could readily detect an infection at a very early stage, before it affected the central nervous system and symptoms arose. Barring that possibility, other therapeutic interventions under study would hinder the infection later on in its progress through the body. Several groups are working on ways to prevent abnormal PrPs in the brain from converting normal ones. Alternatively, drugs might ultimately be found to simply reduce in the brain the amount of normal PrP--without which the disease cannot develop. Although this last approach has received the least attention so far, it may turn out that eliminating prions is indeed the best way to stop them from "going mad."