What Would Thomas Henry Huxley Have Made of Prion Diseases?


"Science is nothing but trained and organized common sense, differing from the latter only as a veteran may differ from a raw recruit. "a Prion disease is a disease of the second half of the twentieth century, but the scientific method that has elucidated this fascinating group of diseases is much older. As an illustration of this, this topic considers the way in which a nineteenth century scientist might have reacted to the challenge that prion disease has presented. T. H. Huxley (1825-1895) was an ardent naturalist, who traveled around the world collecting specimens, and who peered down the microscope (1). He amassed vast amounts of data, and could work prodigiously hard. His approach to science can be judged from some of things that he said. He was a confrontational character, and would undoubtedly have joined in the arguments that led to the concept of prion disease, if he had lived a century later.

Paradigm Shift and Paradigm Drift

"The great tragedy of science - the slaying of a beautiful hypothesis by an ugly fact."b

It has been my privilege to work in an area that has undergone a major ‘paradigm shift’ (2) in a period of a few years, a shift exemplified by the change in name, from "transmissible spongiform encephalopathy" (TSE) to "prion disease." This change is critical, because it moves the central, defining feature of this type of disease away from clinical features (etiology and neuropathology) toward the recognition of the central role of a particular protein in pathogen-esis. This paradigm shift has been exciting, not just because of the impact it has had on understanding these diseases, but also because it casts into sharp relief the process of evolution of ideas and perceptions, which constitutes scientific development. An important feature of this process is scientific consensus, which rests as much on psychological factors, such as perspective, persuasion, and comprehension, as it does on the production of factual data. Kuhn (2) put forward a view of the process of science as consisting of periods of slow accumulation of experimental results, much of which elaborates current theories, but some of which produces data of such knotty contradiction to the prevailing view that eventually the theoretical edifice falls apart. This opens up the possibility of a move to a new theoretical paradigm, and rapid changes in scientific understanding ensue. All this is true, but in its simplest form it does not take account of the difference between facts and beliefs, the difference between events in the outside world and ideas in the collective mind of the scientific community. What goes on in the latter produces "paradigm drift," in which concepts change their meaning to such an extent that they fail to resemble the ideas of their originators.

Paradigm drift is as important as paradigm shift in the progress of science, but it needs to be recognized for what it is: an essential development of ideas, not a change in the facts of experimental data. Charles Darwin is the most influential of all biologists, and arguably is among the handful of the most important of all scientists. But his influence lies in the very fact that his ideas have undergone radical development in subsequent years. Asked what "Darwinism" means today, many people will mention something about Survival of The Fittest as opposed to The Inheritance of Acquired Characteristics, a counterview attributed solely to the pre-Darwinian biologist, Jean-Baptiste de Lamarck. Something may also be mentioned about "Genetics and Mutation," ideas more appropriately attributable to Mendel, Fisher, and Haldane, although Darwin did not know of Mendel’s work, lived before Fisher and Haldane, attributed the phrase Survival of The Fittest to Herbert Spencerc and believed that the "variability," upon which natural selection worked, arose "from the indirect and direct actions of the conditions of life, and from use and disuse." d This clearly means that he thought that changes of form, which had occurred as a consequence of interaction with the environment, were inherited by the offspring of these well-adapted and reproductively successful animals. But so deeply embedded is the notion that Darwinism stands for the opposite of the inheritance of acquired characteristics that the Oxford Dictionary of Quotations excludes the phrase "from the indirect and direct actions of the conditions of life, and from use and disuse " from the paragraph that it chose to quote in order to exemplify what Darwin believed in. In effect, Darwin is being censored for not being sufficiently Darwinian. What Darwin did was to propose a reductionist explanation of evolution, onto which the subsequent study of genetics could be neatly mapped. The fact that Darwinism has undergone so much drift is why Darwin is so important.

Paradigm drift occurs in all science and a brief consideration of the concepts of Creutzfeldt-Jakob disease (CJD) and TSE illustrates the ways in which these terms evolved before they became enveloped in the concept of prion disease. CJD did not appear as an entry in the International Classification of Diseases until 1979, although Creutzfeldt had described his original case in 1920 (3,4) and Jakob described four cases in 1921 (5,6). In the intervening period, many authors have described cases that were thought to resemble these early cases, and a disease entity began to emerge by the process of consensus. Many of these cases were collated by Kirschbaum in a monograph entitled Jakob-Creutzfeldt disease, published in 1968 (7). (The process by which Jakob and Creutzfeldt changed places in this appellation is, in itself, an example of the evolution of an idea in the corporate mind). However, the picture had been very confused, and Kirschbaum says that he undertook his review because his colleagues had questioned ”whether the syndrome is more than a convenient dumping ground for otherwise unclassifiable dementias with interesting cross relations to certain systemic degenerations”. Kirschbaum’s remarkable topic documents many of the possible presentations of prion disease, including cases of rapid onset and progression, ataxic forms, and cases that resemble fatal insomnia. Modern molecular techniques have now rediscovered these forms as part of the spectrum of prion disease, e.g., acute presentation (8), the ataxic form (9), and familial and sporadic fatal insomnia (10,11), although some of these cases fall outside the strict rubric of TSE. Nonetheless, by the neuro-pathological criteria that have evolved since the 1920s, Creutzfeldt’s first case, and two of Jakob’s first four cases, did not have CJD (12). Without wishing to denigrate the contribution of the eponymous authors, it must be acknowledged that many other, less easily identifiable, scientists participated in the emergence of CJD as a disease entity. Thus, Creutzfeldt and Jakob instigated a scientific concensus from which, ultimately, their original cases were largely disqualified.

Another way in which the consensus surrounding TSE has been subject to paradigm drift is the perceived centrality of spongiform encephalopathy to these diseases. Spongiform encephalopathy was not regarded as being a major feature of the original cases of this group of diseases. Hadlow’s perspicacious letter to the Lancet in 1959 (13), which pointed out the clinical and neuro- pathological similarities between scrapie and kuru, mentions ”large single or multilocular soap bubble vacuoles in the cytoplasm” as a feature of either disease, only as the last of all the similarities. Klatzo et al. (14) did not comment on spongiform change in 12 early cases of kuru that came to postmortem, and Beck and Daniel’s early work found that spongiform change, if present at all, was a minor feature (15). Subsequently, spongiform encephalopathy came to be regarded as the defining feature of this type of disease, hence the term ”transmissible spongiform encephalopathy”. But, later still, the diagnostic use of prion protein immunostaining (16), prion protein immunochemistry (17), and prion gene analysis (18) indicated that spongiform encephalopathy was not an obligatory feature of cases which were clearly diagnosable as prion disease by other criteria.

The experimental transmissibility of kuru and CJD was demonstrated in the late 1960s (19,20). This crucial scientific discovery, together with the practical development of the production of mouse-adapted strains of the transmissible agent (21), opened up the possibility of much experimentation and the production of a great deal of intriguing data, which is still important. But the etiology that transmissibility was thought to imply produced a profound conceptual segregation of these diseases from all the other human neurodegenerative diseases, which delayed the comparison of the clinical variables across the neurodegenerative diseases until a much later time (22). This, together with the molecular diagnostic techniques referred to above, which identified cases in which transmission was either not attempted or did not succeed, and in which spongiform encephalopathy was not seen, led to the somewhat comical concept of ”nontransmissible, nonspongiform TSE.” This situation inevitably gave way to the adoption of the term ‘prion disease.”

Myth and Misunderstanding

”Irrationally held truths may be more harmful than reasoned errors.’"

An important aspect of the interpretation of data— the conversion of facts into knowledge—is an understanding of the circumstances in which the data have been collected. Until the development of mouse-adapted strains of agent in the 1960s, work on scrapie was slow, and required field experiments as well as laboratory analysis. Some of these experiments were limited, and were sometimes subsequently subject to a degree of overinterpretation, to the consternation of the original author. For example, Hadlow, who conducted a few experiments on infectivity in the peripheral tissues of sheep (23) was distressed to find that these data, collected in the context of general interest, were later used as the definitive data on which to base decisions about the possible infectivity of peripheral tissues of bovine spongiform encephalopa- thy (BSE)-infected cattle, prior to the completion of experiments in cattle (24). Hadlow said of his own data ”rather than treating them as tentative findings, they are accepted as established facts about the disease; they become part of the scrapie dogma. But sometimes they do not deserve that distinction” (25).

One of the most frequently cited pieces of evidence in favor of maternal transmission, as an important factor in the epidemiology of natural scrapie are the reports by Pattison (26,27) which said that a number of sheep, fed on pla-cental membranes taken from dams with scrapie, subsequently developed scrapie. But Parry (a friend of Pattison) claims that Pattison believed that this route of infection, if it did occur in the field, could account for no more than 5% of field cases of scrapie (28). In reviewing the role of placental infection in the epidemiology of natural scrapie, Hadlow later remarked of some people, ”For them it is one of the facts about scrapie” (25). It remains so, despite the fact that embryo transfer experiments in sheep (29,30), epidemiological surveys of natural scrapie (28,31,32) and experimental studies in primates (33) have failed to detect maternal transmission. Cohort studies (34) and epidemio-logical surveys (35) in cattle have failed to distinguish between common exposure, genetic predisposition, and maternal transmission as an explanation of the excess occurrence of BSE in the offspring of cows that subsequently developed BSE. Hoinville et al. (36) calculated that, even if this excess BSE were cause by maternal transmission, it would have had a negligible impact on the BSE epidemic in Britain. Kuru was not passed to the children of kuru victims, except by contamination during funerary practices (37) and the familial occurrence of other forms of prion disease is entirely accounted for by mutations within the prion gene.

Nonetheless, infectivity has been reported in rodent transmission studies from human blood and placenta in various laboratories, which has fueled the view that maternal transmission as a cause of sporadic CJD is both a risk and a fact. But, as Baron et al. (38) have pointed out, the lack of difference in incubation time between assays using brain, compared to other tissues or bodily fluids, lack of experimental replication, and high levels of unexplained deaths in experimental and control groups, casts doubt on these reports. Failure to transmit from human blood to primates, following intracerebral inoculation (the most sensitive bioassay), would seem to be a more robust finding (39). Infec-tivity has been found in blood of experimentally infected rodents (e.g., refs. 40-42), but this occurrence in artificial situations may not have epidemiological implications. HIV is carried in blood, and mosquitoes transfer blood from person to person in sufficient quantity to transmit diseases such as malaria, but AIDS is not transmitted by mosquitoes.

"The chess-board is the world; the pieces are the phenomena of the universe; the rules of the game are what we call the laws of nature. The player on the other side is hidden from us. We know that his play is always fair, just, and patient. But we also know, to our cost, that he never overlooks a mistake, or makes the smallest allowance for ignorance." 

Luck was never going to be on the side of those whose job it was to cope with BSE. It made little difference in terms of planning how to cope with BSE, whether, in the mid-1980s, one accepted the relatively new prion hypothesis or clung to the ”unconventional virus” view of the TSEs. BSE was a new disease that may not behave entirely like other TSEs. It was clear that doing the experiments necessary to establish the transmissibility, incubation period, species barrier, and tissue distribution of infectivity of this new disease would take at least 5 years.

Meanwhile, attempts had to be made to establish the source of infection (assuming that contamination, rather than inbreeding, was the source of the early cases), and to remove it. But it would not be known for 4-5 yr whether such measures had been successful, and each further attempt to reduce the spread of infectivity between animals would also take a further 4-5 yr to evaluate. In addition to this the agent of prion disease is difficult to destroy, and is infectious at very low doses. Transmission to other animals is still the most sensitive method of detecting infectivity. Despite advances in in vitro tests for prion protein (17), it is still not possible to demonstrate that tissue, food, or medicinal products contain so little infectivity that no disease will occur when several million cows or people are exposed to it. The number of people infected with new variant CJD cannot be accurately assessed at present (43). If the first cases of new variant CJD result from exposure early in the BSE epidemic, then the minimum incubation period is about 10 yr (44). The maximum incubation for kuru amongst the cannibals of Papua New Guinea is in excess of 40 yr (45) and the same may apply for new variant CJD. Most of the scientists who witnessed the onset of the BSE epidemic in Britain will not know the full extent of the disaster, because they will have died of old age before it can be certain that there will be no more cases of new variant CJD.

Development of The Prion Hypothesis

"It is the customary fate of new truths to begin as heresies and to end as superstitions."

In 1960, Palmer published a paper in which he acknowledged the wholly unusual nature of the scrapie agent, and suggested that it ”may be a non-protein moiety, perhaps a carbohydrate, which on introduction to the body forms a template for the subsequent reduplication of the agent… If the nature of the agent causing scrapie can be finally determined the results may lead to spectacular changes in the present-day concept of the genesis of disease” (46).

Palmer was incorrect to dismiss the possibility that the infectious agent could be a protein, but his idea that the agent could act as a template for the formation of more of itself, is central to the current theory of prion replication.

Other authors, notably Pattison and Jones (47), Griffith (48), and Lewin (49) saw that the remarkable resistance of the scrapie agent to physiochemical inactivation (50,51) implied that it may not contain nucleic acid and proposed that the information-containing and replicating part of the agent may be a protein. Gibbons and Hunter (52) and Hunter et al. (53) made the same sort of arguments, but proposed that the agent was a replicating polysaccharide. None of these authors was able to suggest how this replication might take place, although Pattison suggested that, perhaps, ”the scrapie agent is present in an inhibited form in normal tissue and in a released form in scrapie tissue.” Griffith’s contribution (48) was truly prescient, in that he discussed various possible mechanisms by which infectious disease could arise spontaneously, and information enciphered (to use the modern term [54]) in protein structure could be transferred to other protein molecules. First, he suggested that a disease-producing gene may normally be silent, but be expressed during disease. This allowed strain variation, and variation in host susceptibility, to reflect different polymorphisms in the host gene, with concomitant differences in predilection for gene de-repression. Second, he suggested that the protein may take up different conformations, one of which was envisaged as being disease-related, without changes in primary structure. Such a conformational change was an unknown phenomenon at that time. Third he recognized that the ability of proteins to form polymers was another way in which proteins of the same primary structure may have different biological properties.

In 1962, Parry published an article claiming that, despite being experimentally transmissible, natural scrapie was wholly genetic in origin (55). Dickinson et al. (56) replied with the more conventional idea that the pattern of disease was consistent with genetic susceptibility to an environmental agent or maternal transmission of the infectious agent. Parry persisted, and, in 1973 submitted an article to Nature, the last sentence of which read ”the hypothesis most consistent with present evidence is that the scrapie TSEPA [transmissible encephalopathy agent] is formed de novo in each affected animal by the metabolic activity of the natural recessive gene” (28). His cover letter to the editor said, ”in view of … Dr Gadjusek’s Nobel Prize Oration last year, it seems important to place on record facts regarding scrapie in sheep which are generally overlooked in the scramble to establish a primary infectious aetiology for this groups of disorders” (28). The paper was rejected on the grounds that the conclusions were erroneous. What the word ”erroneous” meant in this case was not ”incompatible with the evidence,” but rather ”incompatible with the prevailing view” a confusion between facts and beliefs. Parry was vilified for his views by many of the virologists who were working on scrapie (see the Foreword by Alpers in ref. 28) but he had friends among the shepherds with whom he had worked all his life, and who knew that scrapie was associated with excessive in-breeding (57).

These early workers knew that they were dealing with something that was outside the revolutionary developments in molecular genetics in the 1960s. Between them, they had all the essential bits of the jigsaw, but lacked the experimental protocols that were later to allow the ”prion hypothesis” to be proposed (58). The prion hypothesis is not heretical to the central dogma of molecular biology—that the information necessary to manufacture proteins is encoded in the nucleotide sequence of nucleic acid—because it does not claim that proteins replicate. Rather, it claims that there is a source of information within protein molecules that contributes to their biological function, and that this information can be passed on to other molecules. But the protein molecules are still manufactured according to the instructions contained in nucleic acid. The additional information source is the conformation or shape of the protein molecule. The conversion of prion protein from the normal cellular form to the disease-associated form involves a conformational change (59). Furthermore, the abnormal form of prion protein can have one of several different conformations, and these differences explain the existence of the many strains of agent (54) which for so long were regarded as the main evidence in favor of a nucleic acid based informational system within the infectious agent.

”I am too much of a sceptic to deny the possibility of anything."

Are prions alive? They contain information enciphered in the shape of the prion protein molecule, and that information is transmissible from molecule to molecule. The information encoded in DNA is transferred in the replication process to the two DNA strands that are manufactured from the unfolding of the one parent DNA molecule. In prion replication, there is no manufacture of new prion molecules, but the principle of information transfer, and therefore information replication, persists. The precise mechanism by which this occurs is still elusive, but it seems to involve the partial unfolding and subsequent refolding of abnormal prion protein molecules, so that contiguous normal prion protein molecules also assume the abnormal conformation. Are there other examples of such self-replicating information? Although computer viruses were invented by computer terrorists, their defining feature is that they contain enough information to direct the computer in which they reside to recreate more of the viral information-containing sequences, and so they behave as self-propagating machine infections. Another example of self-replicating information is that of the spread of ideas within an intra communicating population, i.e., within a culture. The autonomy of ideas as replicable information, whether they be pieces of factual information or new ways of looking at things, is emphasised in the concept of ”memes” (60). By mechanisms that are not entirely understood, the brain modifies its fine structure to store information from outside, and such information can bring about this change in as many brains as it has direct or indirect contact with. From this point of view, prions are not alive like conventional organisms, but they belong to a group of interesting phenomena that comprises not only living organisms, but also other forms of replicating information systems including the propagation of ideas.

Contemporary prions are parasitic on the prion protein manufactured by the host cell, but the mechanism by which the information contained in the shape of the prion protein is imparted to other prion protein molecules does not depend on cellular mechanisms (61). This mechanism of replication could, therefore, have evolved prior to the evolution of cellular systems, and, because it also does not depend on DNA, could have been at work in the ”primeval soup” of small polypeptides, which is presumed to have preceded the evolution of life itself. The demonstration that heritable conformational changes can also occur in certain proteins found in yeasts and fungi (62,63) and possibly widely throughout biological systems suggests that this form of replication may be ancient.

In addition to the evolution of DNA-based replication, another problem of the change from primeval soup to organisms is the change from liquid to solid life forms. It is not enough that the genes contain all the information necessary to make an organism: That organism must also be capable of developmental self-assembly. The process begins with individual molecules that must stick together. The abnormal form of prion protein belongs to that class of proteins capable of forming amyloids (64). These orderly aggregations of molecules are formed by self-assembly, which often occurs under artificial, as well as natural, conditions. Where these aggregations cause disease, the disease may be regarded as a disorder of molecular self-assembly, a process which is inevitable, given that biology is fallible and self-assembly is obligatory. Abnormal amyloid formation by at least 18 different proteins is associated with disease (65).

There is simplicity in wishing to confine the term ”prion disease” to those diseases in which the abnormal form of prion protein can be detected, but, if prions are defined as ”elements that impart and propagate conformational variability” (66) then prions have been found in other biological systems, notably yeasts and fungi (62,63). These discoveries suggest that protein conformational variability may be a widespread component of non-Mendelian inheritance, which would have important biological functions, as well as disease potential. Like mitochondria, which pursue their own genetic destiny within the cell, and ”junk” DNA, which quietly replicates itself within the genome, an archaic replication mechanism of the primeval soup may be working on its own agenda within the cytoplasm of the cells of other organisms. This multiplicity of information replication systems is further exemplified at the level of the whole organism. What is quaintly regarded as one organism with one genome carries within itself many obligatory parasites that are essential for the survival of the main organism, and, within the ecological system, all organisms play a role in the survival, as well as the destruction, of other organisms. The biomass is itself composed of many information-replicating systems, but none of them is independent. The concept of "one genome, one organism" is beginning to look less clear cut.

Prions are on the borderline between biology and chemistry. Because of their disease-causing, infectious nature, they have been regarded as a biological problem, and, for many years, they were studied as though they were viruses, or at least unconventional viruses. But their ability to persist outside living organisms, seemingly indefinitely (67), and their resistance to chemical and physical inactivation by methods that include ashing at 600oC (68), means that they can also be regarded as environmental pollutants. Prions polluted cattle feed in the 1980s and led to the BSE epidemic in Britain in the 1980s and 1990s. Cattle, cattle-feed, and the ingredients of cattle-feed were exported to Europe and beyond, and BSE is now emerging as a serious problem in Europe. BSE is likely to behave like other new diseases, whether caused by infection or pollution: a high incidence, but geographically confined, effect eventually gives way to a widespread, but low-incidence disease occurrence. Although the early effects of a major disease epidemic may be very dramatic, the widespread and potentially permanent endemic stage of a disease may ultimately claim more lives. The cost of destroying a wide but thinly spread hazard may be much greater than the cost of containing a small but high level of contaminant. BSE is currently confined to countries capable of dealing with it, given the necessary political will. If BSE were to escape to developing countries it would be quite impossible to eradicate it even though the conditions that lead to large outbreaks of disease may not occur in those countries.


"If a little knowledge is dangerous, where is the man who has so much as to be out of danger?

The TSEs have produced two Nobel Laureates, Carlton Gadjusek and Stanley Prusiner, both within the decades that saw, in a different arena, the unraveling of the genetic code and its control of cell function. Lewis Thomas referred to scrapie as ”the greatest puzzle in all biology” (69), and, from the point of view of the main thrust of molecular biology during that time, TSEs, and subsequently prion disease, were always eccentric. Their very peculiarity attracted the maverick who could see that these diseases indicated a secret important to understanding all biology. Worrying about the bits of data that do not fit the picture is as important as understanding the way the other bits hang together to produce a coherent whole. Caution as well as audacity is required to get it all right. As Prusiner himself argues, ”In prion research as well as in many other areas of scientific investigation, a single hypothesis has all too often been championed at the expense of a reasoned approach that involves continuing to entertain a series of complex arguments until one or more can be discarded on the basis of experimental data" (66).

Thomas Huxley was part of the biggest paradigm shift that there has ever been: the battle for the acceptance of evolution as the origin of species. He knew nothing of prion disease, and is unlikely to have heard of scrapie, because the introduction of many cross-breeds of sheep in the nineteenth century had produced a dramatic decline in this disease (28). He had visited Papua New Guinea as a naturalist-explorer, but had limited access to the island, because the ship’s captain was reluctant to land, fearing that the indigenous population were cannibals (1). The kuru epidemic that decimated some highland tribes in Papua New Guinea less than a century later, and which was almost certainly maintained by cannibalism (70), suggests that the explorers’ fears may have been justified. Huxley was a vehement supporter of Darwin and of the atheistic, bottom-up explanation of our existence that evolution implied. Darwin seemed to win, although a kind of compromise arose in the early part of the twentieth century between scientists and theologians, so that it appeared that evolution and religion were not incompatible. But the elucidation of the precise mechanisms of DNA replication and genetic determinism in the second part of the twentieth century rekindled the row between the bottom-up evolutionary biologists, evolutionary psychologists, and sociobiologists and the top-down theologians, philosophers, and academics of the arts and humanities. The row is still about where the information comes from that drives the structure and behavior of the biological world, including man. In his latest work, Consilience, E. O. Wilson is striving to push the domain of the bottom-up explanation of the world beyond individual psychology and into the area of population dynamics: sociology, economics, and ecology (71). The "selfish genes" (60) have had great impact in this debate but prion disease has shown that they are not the only replicating information system that can have a bottom up influence. The conformational changes of prion protein have led through cellular dysfunction and fatal disease, to the political, economic, and ecological disasters of BSE, and the personal and social disasters of kuru and new variant CJD. T. H. Huxley would have been in his element in these debates.

Next post:

Previous post: