How the Cows Turned Mad

The mysterious, misfolded form

Do not be put off by the title, which trivializes the text. The molecular biologist Maxime Schwartz is a pillar of the French scientific establishment. Recently he was Director of the Pasteur Institute, the living embodiment of French pride in Louis Pasteur, the nineteenth-century inventor of microbiology. Now he has capped all that with a brilliant piece of science writing, How the Cows Turned Mad.

Schwartz has written about the diseases now known collectively as the TSEs, where "T" stands for "transmissible" and "SE" for "spongiform encephalopathy", meaning little more than "spongy brain". The most familiar version now is BSE, which has ravaged the British cattle industry for twenty years. (The "B" in "BSE" stands for "bovine", not "British".) The under-lying sense of menace, now tragically a reality, is that, if sheep and cattle can be affected, why not people?

Schwartz, who refers to this potentially universal scourge as "The Disease", begins with scrapie in sheep in the eighteenth century. An interesting account by a Lincolnshire parson tells of an outbreak in the 1730s: at onset, the sheep are light-headed and "wild", then they rub themselves compulsively against trees and posts until the wool and even skin come away (hence "scrapie"), finally — after six months or so — they are moribund and die.

For a century and a half, scrapie had to be repeatedly rediscovered because the symptoms varied from flock to flock — and farmers were unwilling to stigmatize their flocks. (That behaviour is not much changed.) Wild theories of causation flourished, from lightning strikes to the excessive ardency of rams. The serious contenders were infection and heredity. Science stepped in only towards the end of the nineteenth century, when Charles Benoit, a professor at a French veterinary college, used the microscope — Pasteur's most valuable tool — to identify telltale nerve cells in the spinal cords of diseased sheep: the cells appeared to contain empty bubbles. That was at least a way of telling whether a sheep had died of scrapie.

It has taken the best part of a century to conclude that scrapie (like other TSEs) is neither hereditary nor an infection, but both. The fact that, even now, Australia's merino sheep do not suffer from scrapie is a sufficient proof of genetic influences. However, there are many others. Of infection, the stumbling block has always been the nature of the infecting organism. Pasteur's disciples were quick to say that it could not be a bacterium: they could not see it under a microscope. But neither was it a virus: affected animals never ran the fever that marks viral infection. What else might it be? And why is the incubation period so long — two years at least for scrapie? Newfangled molecular biology has only incompletely answered these questions.

Between the two World Wars, scrapie continued to kill up to 15 or 20 per cent of Europe's sheep flocks, but little was done to understand it. In an unrelated field — human medicine — two German physicians independently described cases of a previously unknown neurological disease that began with a loss of motor functions and progressed through a loss of reason to death within a few months: that disease is now known as Creutzfeldt-Jacob disease (after the two physicians), or CJD. Then, in 1938, came a report from Vienna and Munich of seven members of the same family who had died from a similar disease (not identified as CJD until 1968). As Schwartz puts it, "it was nothing other than The Disease in a new guise", but nobody realized it at the time.

An important step to understanding was the success, in 1936, of two French scientists in transmitting scrapie from one sheep to others. They ground up parts of the brain and spinal cord from a diseased sheep and injected the mixture into an eye of a healthy animal. (The eye, being a part of the brain, is a convenient route to the central nervous system.) After eighteen months, one out of seven sheep went down with scrapie. The disease was later transmitted to goats.

By the 1950s, transmission experiments were all the rage, especially in Britain, largely because the Australian and United States authorities had banned the import of British sheep because of the high prevalence of scrapie. It was hugely important, when BSE appeared in the 1980s, that Richard Chandler had transmitted scrapie to mice which, with a breeding cycle of six months, are much more convenient experimental animals than sheep. Unsuspected, a large-scale transmission experiment to human beings, no less, was already under way. In the early 1960s, physicians in the US and Britain began treating congenital dwarfism with human growth hormone extracted from pituitary glands retrieved from mortuary cadavers. Twenty years later, on the eve of the outbreak of BSE in Britain, came reports of deaths from CJD by people treated with human growth hormone in their youth. By good luck, synthetic growth hormone was then available, but the full death toll among the 25,000 worldwide believed to have received the old treatment is not yet known.

The next step in Schwartz's detective story seems a diversion — the heroic tale of how, in the latter half of the 1950s, the larger-than-life US paediatrician Carleton Gajdusek and the German physician Vincent Zigas, the latter employed by the Australian administration of Papua New Guinea, described in detail the disease called kuru which was endemic among people of the Fore tribe. Again, the presenting symptoms were an impairment of walking, which steadily deteriorated until patients were moribund (when they lost the faculty of speech) and could not swallow (so that they starved to death). Kuru hit people of all ages, children included. Again the microscope revealed characteristic defects of nerve cells in the brain and spinal cord.

In due course, in 1959, says Schwartz, "the wall came down"; meaning the wall between physicians and their humbler cousins, the vets. When Gajdusek had transmitted kuru to a chimpanzee, scrapie, CJD and kuru were recognized as essentially similar diseases caused by agents that were probably very similar. Later, it became plain that kuru was already on the wane by the time Gajdusek had reached the Fore tribe: the Australian administration had put a stop to the practice of mortuary cannibalism in the early 1950s. But then came BSE: The Disease in yet another guise.

So what is the agent responsible for all these diseases? The answer (such as it is) we owe to the almost paranoid persistence of Stanley Prusiner, now a professor at the University of California, San Francisco. Flying in the face of molecular biology's conventional wisdom that infective agents such as viruses and bacteria require nucleic acids (either DNA or its cousin, RNA) to retain their character over the generations, he argued from the mid-1970s that protein molecules alone could do the trick. It is a perplexing trick.

All mammals have a gene that controls the production of a protein now called "prion protein". Prion molecules sit on the outsides of nerve cells, but their natural function there is entirely obscure. Like other protein molecules in cells, prions are synthesized as long strings of chemical units (roughly 300 in prions) which must be folded into a more compact shape to do what is required of them. The current explanation of all the TSEs is that they are caused by a misfolded form of the prion protein — the infectious agent. Like other protein molecules, they associate with other molecules like themselves, including prions with the natural or "healthy" shape. But what if the misfolded molecule induces its normal partner-molecule to adopt its own misfolded shape? Then all of the natural prions in a cell will be converted to the mis-folded form. To make good the deficiency of natural protein, the cell will synthesize more of it, which will quickly end up in the same blind alley. Eventually the cell will die from the accumulation of useless protein.

That is the best theory yet of the TSEs, from scrapie and kuru to CJD and BSE — and also the rare disease called Familial Fatal Insomnia. It explains, for example, why the immune systems of mammals do not react against infection (causing a fever in the process): the immune system does not react against the body's own proteins, misfolded prions included. What the theory lacks is a clear idea of the structure of the misfolded prions -the difficulty is that they are insoluble and form clumps. The theory also explains why it is more difficult to transmit TSE disease from, say, sheep to goats, which requires misfolded sheep prions to corrupt the shape of dissimilar goat prions, than to transmit the goat disease thus acquired to other goats.

That is a novel mechanism of infection, but where does heredity come in? Thanks to research stimulated by the emergence of BSE, the answer is partly known: like other genes, the prion genes vary in the details of their chemical structure from one individual to another; some of these variants (which are passed from parents to offspring) appear to be more likely than others to misfold.

Maxime Schwartz deals with these complicated questions with consummate clarity, reinforced by his evidently thorough knowledge of the research literature, which commands respect. The construction of his book closely follows the modern paradigm of narrative style: each chapter ends with the question that will be answered in the next. The translation from the French is accurate, but the literal translation of what the English call the "future conditional" — "he would discover" for "later, he discovered" — is irksome.

And what about the long-term menace? Will there ever be a time when some TSE is as great a menace to human populations as scrapie was to sheep in the eighteenth century? Until much more is known of the mechanism of these diseases, doubt will persist. Lacking the understanding, not eating other mammals is the only certain safeguard. It is enough to make vegetarians of us all.

John Maddox's most recent book is What Remains to Be Discovered, 1998. He was Editor of Nature for over 20 years.

• Read more about this book