Vivienne Parry is a writer and broadcaster. A scientist by training, she is a former presenter of BBC TV's Tomorrow's World.
IS VCJD caused by BSE?
The only way to prove that variant CJD is caused by BSE is to inject infected material from a cow into a human and sit back and wait to see whether, in ten, twenty or even forty years time, that person develops vCJD. Curiously, there is a conspicuous shortage of volunteers for this crucial experiment. But without it, there is still no direct proof that vCJD is caused by BSE.
What causes BSE? Metals, organophosphates, tiny viruses called virions have all at one time been implicated. But the front runner is rogue prions.
Prions are elaborately folded proteins found in all species of animals - each species has their own sort. Researchers speculate that the job of normal prions might be that of an anti-oxidant in the brain. What an abnormal prion does is try and recruit all the normal prions it meets to the dark side, making them flip in shape to become like it. There is then a cascade effect with abnormal prions collecting in the brain and causing the symptoms that are now so familiar. Prion diseases include kuru, CJD and scrapie.
Stanley Prusiner, a Californian professor, won a Nobel Prize in 1997 for his prion hypothesis. At first his work was greeted with scepticism. After all, an infectious agent that contained no information in the form of DNA or RNA was hard to credit.
Personally I have no doubt that prions are the mechanism of these diseases. First, mice genetically engineered to have too much normal prion, develop a non-infectious type of brain disease. And if you subject infected tissue from a cow to extreme temperatures and pressures, of the sort that would crucify any self respecting virus or bacteria, it will still cause infection in a recipient animal. So that's it then. Prions and nothing else. Sorry, I don't buy it.
Levels of metals in the brain for instance are known to be important. There is a high incidence of CJD amongst South American manganese miners for instance. And your genetic make up may be the deciding factor in determining how early your symptoms will appear.
But here's the bit that foxes me. If you have a mouse, genetically engineered to have no prions, they don't catch any disease when injected with infected tissue. Fine. That's what you would expect. But take tissue from these non diseased animals and guess what, they can cause infection in normal mice, even though they have no prions. OK Professor Prusiner, get out of that one.
The latest theory, released after the programme was recorded, by the European Molecular Biology Laboratory in Germany is that there IS some other infectious agent. And it's a parasitic gene called an Alu. It interferes with the way proteins are produced, and eating meat from an infected animal may be enough to tip the balance and start a runaway acceleration of proteins - causing the symptoms of vCJD.
It's an interesting theory - and it could be right. Personally I believe that we are a long way from understanding the causes of BSE and until we know that, vCJD will remain an enigmatic mystery.
Professor Roy Anderson is an epidemiologist at Imperial College School of Medicine, and an expert on the vCJD epidemic.
Could you explain the different kinds of CJD and what causes them?
The groups of agents which CJD belongs to are called the TSE, transmissible spongiform encephalopathies. These are very severe diseases which are almost a challenge to one of the central dogmas in biological science, which is that a living organism (an agent) has its informational coding, which is its genetic code, in one or other of 2 molecules called DNA or RNA. The TSE are diseases caused by an agent called a prion protein which is essentially a replicating protein and it has no informational coding molecules such as DNA or RNA. So it's an extremely unusual infectious agent, almost invariably fatal as far as we know.
There are quite a number of these TSE: there are agents which cause disease in animals, such as cows with BSE; elk in North America with a wasting syndrome disease; and a thing called scrapie which is in sheep flocks. Then the human ones are various: there is a thing called kuru, which occurred only in Papua New Guinea in one small tribal grouping and there is CJD, classical CJD, which occurs in the elderly. The new form that we're very concerned with at the moment is vCJD, which is very tightly, or closely related to the BSE agent.
So it's a broad class of lethal diseases, typically very rare in the world's population, and their primary pathology is to create a spongiform appearance in the brain and essentially the destruction of the brain and motorneuron function. Mortality is virtually inevitable in the current state of therapies and diagnostic methods.
What is the story of vCJD in Britain?
BSE, or Mad Cow Disease, is one of the class of these TSE. So the concern when Mad Cow Disease first arose in the cattle herd in GB was that this infectious agent might transmit from some animal product into humans. The government body SEAC (Spongiform Encephalopathy Advisory Committee) detected some cases of a disease called CJD in people much, much younger than was typical, in the age range of 15 -35 and a suspicion arose that they may indeed have been the transmission of mad cow disease to humans. Subsequently a variety of scientific approaches strengthened that hypothesis in the spring of 1996 the then Minister of Health Stephen Dorrell made an announcement to the House of Commons that these new cases of vCJD were probably associated with exposure to the Mad Cow Disease agent, the bovine songiform encephalopathy. That was the beginning of the story which then became in the public and government's attention.
Could you describe the experimental work that has suggested BSE and vCJD are the same thing?
The scientific evidence that BSE is the causative agent of vCJD in humans is diverse. First of all, there is the epidemiological inference that humans were exposed to a lot of this Mad Cow Disease agent, particularly prior to 1989. All these diseases have one characteristic which is a very long incubation period, which is the time from infection to when symptoms of disease appear. And therefore if you had a lot of exposure prior to 1989, you would expect quite a number of years before you detected the first cases in humans.
Then there's a more important body of evidence which is experimental science. Now, in an ideal world what you'd do to prove the hypothesis would be to give the agent to the host species in a controlled and defined way and then monitor what happened in the particular exposed host. Well, clearly you can't do that with humans, so you use a variety of other methods.
One of them is to use a primate model such a marmoset or a monkey, and either inject or feed the host species, experimental species, with the bovine agent and then monitor the behaviour of the infected group and the control group to see what disease emerges. Now, the trouble with that is that it requires a very long time period, so if you expose a marmoset or a primate, a monkey to the agent, you might have to wait 5-7 years before you see disease. So an alternative approach is to use mice, which have a much shorter life expectancy, and we know from a variety of scientific studies that some of these TSEs, such as scrapie, do actually as it were, establish in mice and cause a disease which is very similar to scrapie. And also you can nowadays use a particular method which is called transgenic biology, where you insert a gene into a mouse which mimics or is an exact copy of the human gene. So knowing that there's an abnormal prion protein, you can put the human prion gene into the mouse, expose the mouse to BSE agent and then look at the type of pathology or disease that develops.
And what transpired was that the major classifying approach to the disease to start with was on straight pathology, that was if you take a cross-section of the brain of the diseased animal, you look at the pattern of the spongiform appearance of the brain and each of the TSEs induces a slightly different pattern. Subsequently there are more advanced methods which are called immino chemistry, where you're staining to look for the presence of the abnormal protein, and that is the method that is very much in favour at the moment.
Do you believe the infectious agent in BSE/ vCJD is a viral or a protein-only agent?
There's been much discussion over many, many years about what the spongiform encephalopathies are. When I was a student they were referred to as the slow virus diseases because it was thought that the destruction was due to a virus which hadn't been detected at present and it took a very long time to cause this pathology. Stan Prousner in the US was the main scientist, along with a body of other scientists who pointed out that these abnormal proteins could in a classical experimental sense transmit disease. Yet there still persisted views that something else was involved - a virus, bacterium, some abnormal chemical or whatever.
Well, experimentation, increasingly refined over the years, has demonstrated that if you take an inoculating sort of quantity of an abnormal protein, you treat it in ways that would kill all known living organisms that are based on DNA or RNA, such as viruses and bacteria, treat it in a way you kill all of that, you're left with this extraordinarily resistant protein only and when you give that via the oral route, or indeed inject it directly into the brain, it causes the disease that we see in patients and indeed in animals.
What is the likely extent of a vCJD epidemic?
When first encouraged to take an interest into what the future might hold for the vCJD epidemic, my initial reaction was that this was really an impossible scientific problem because there were too many unknowns. But we adopted an approach where we took the early number of cases, which were very small, we looked at their age distribution, particularly they were distributed to younger people, we took what we knew about past exposure to the BSE epidemic, which you could call the risk function over time, and we simulated millions and millions of scenarios, changing the values of the unknowns, drawing them off a distribution in a big computer and then created a cloud of possibilities into the future. As expected, the range of possibilities is from a few hundred to over 100 000 cases and developing over a very, very long time period. If the incubation period is short in humans, say it was 8 years, then we may have seen the worst of the epidemic in the next few years. If it's much longer, it could even be 20-30 years, because one characteristic of these agent is when they cross species barriers the incubation period tends to lengthen, and this is cross from cows to humans. If it's much longer, say 20-30 years, then the epidemic could be very large.
Now, just recently an important hint - it's not hard evidence - has emerged to give us at least some minimum bound on the incubation period. This was a cluster of vCJD cases in Leicestershire, for which, given detailed investigations of when individuals were exposed to a common source of food from a set of butchers, the minimum bound on the incubation period may be somewhere in the region of say 12-15 years. Now, they may be the early incubators, so that says to me at this current point in time, although the evidence is weak, that the incubation period is likely to be quite long, and so that's not particularly good news at the moment.
What do you think the government has learnt from this experience?
There are good and bad things that have come out of this experience. The bad things are well documented - exposure of the human community to a very dangerous agent. The good things are perhaps less widely experienced. One of the most important, actually, is I think government has learnt the lesson that openness in scientific advice to the government is an important component. So scientific advisory committees are now encouraged to have public meetings, they are encouraged to have representatives who are not scientists, but are representing the public or consumer interests. Secondly, the government is more organised in the way it receives scientific advice, and it's also very conscious of the fact that absence of evidence of risk is not absence of risk, and that's really important.
Do you think a treatment for the diseases will be found?
Progress in research is always slow, particularly towards therapies, but there's been good progress recently on diagnostic methods and good progress in developing ideas of possible therapeutic lines of attack, in other words, thinking about ways that the agent causes the cascade of conversion of normal protein to abnormal - are there scientific ways of stopping that, or slowing it down? So I'm an optimist in the longer term that we will have some therapeutic options. They may not be cures, but they could be slowing progression to disease or alleviating symptoms. But, as always in science, it's going to be a long time and science is very unpredictable process, you can't say progress is linear in a particular direction, it goes in bumps and starts, depending on people, opportunities and odd ideas that emerge.
Dr Rosalind Ridley and Dr Harry Baker - scientists at the Dept of Psychology at the University of Cambridge, and authors of 'The Fatal Protein.' They are experts on prion diseases and the history of research in this area.
Could you explain the history of prions, widely believed to be the agent of BSE and vCJD?
HB: It's a kind of complex story, but it starts way back in the 60s when CJD and a rather obscure disease called kuru which affects the cannibals of Papua New Guinea, both neurodegenerative brain diseases were shown to be transmissible experimentally into animals. What this involved was taking bits of brain, injecting them into the brains of monkeys and waiting for up to 2 years before these animals got sick.
Now, this is an awful long time for a viral disease, and in any case people were unable to find viruses: using all the techniques available for identifying viruses, including denaturation studies in which you try to destroy the viruses, all those things that normally destroyed viruses didn't seem to destroy the infectious agents in these brains.
Some little while later Stanley Prousner and his team in the States were looking at the brains of infected animals, experimenting on infected animals and normal animals and trying to see what differences they were between them. And they isolated a protein from the infectious animals they called a 'proteinatious infectious material', which they contracted into the term 'prion'. It was later discovered that normal animals make prion proteins but those proteins are destroyed in the extraction process of denaturation, so it seemed as though the normal animals did not have them, whereas in the infected animals, the prion protein is not destroyed to the same extent and could be detected.
They then used the techniques of reverse genetics to find out where this prion protein was coming from - was it some kind of external agent was being expressed, say by a virus, or was it something that the animal itself was producing? They identified a piece of DNA which was producing prion protein in both the normal and infected animals, but in the case of the infected animals, the prion proteins had undergone a change in its structure, it was a different shape, and therefore this conferred on it a sort of resistance to denaturation.
Do we know how the normal prion protein is converted to the abnormal, malignant form?
RR: It took quite a while for this idea of the prion protein to be accepted because it at first sight appeared to challenge some of the most fundamental ideas about molecular biology. Of course, we're used to the idea that infections are caused by bacteria, which are large and can be seen, and viruses, which can be sometimes rather more obscure. Now, here we had a situation where we had an infection but apparently no virus, and viruses contain nucleic acid and this was just a protein. And I think people probably misunderstood in the first instance, and thought that a protein was making more copies of itself, whereas the fundamental dogma of molecular biology says that the information required to make proteins is contained in nucleic acid which then has to be converted into the production of proteins.
Of course, it later became obvious that the prion hypothesis doesn't actually challenge the fundamental laws of molecular biology at all, because the DNA that is responsible for making the protein is in the host - the animal that was infected - and indeed it's in all animals and all people. What happens is the abnormal protein directs the conversion which produces more abnormal protein from normal protein, but not from nothing, it doesn't make it itself, it still has to be made originally under the instructions of nucleic acid. It's just the conversion of its shape which is directed by the protein to another protein molecule. It wasn't as heretical as some people first thought, but it is wholly new, because it's a new way in which information can be transferred from one molecule to another without going through nucleic acid, which is the normal repository of information in biology.
HB:The mechanism by which normal form of the prion protein acts as a template for the conversion of the normal to the abnormal is not yet clearly understood. We do have quite a lot of information about this interaction and it goes back, funnily enough, to some of the human cases of CJD. There are some cases where CJD runs in families and having established that prion protein is involved in this process it did not take long before we were able to examine the prion gene in the families in which CJD was inherited. And we found mutations in that gene, they were coding for very slightly different prion proteins from the normal ones.
How easy is it for prion disease to transfer from one animal to another?
HB: There's a phenomenon which we ought to be aware of, the phenomenon of the species barrier. If you take an animal infected with prion disease and you inject some of its brain into an animal of a different species, that animal may or may not get prion disease, it may take a long time, it may take a short time. If it takes a longish time but gets the disease and then you transmit to another animal of the same species, that time gets shorter. The first transmission is known as going across the species barrier.
We know, for example, that if you have mice which carry hamster genes and you infect the mice with hamster infectious prion protein, then the animals will get sick. But a normal mouse infected with hamster prions takes an awfully long time to get sick. So we clearly know now that an infectious (abnormal) prion protein from one species will convert normal prion protein from that species much more readily than it will the prion protein from another species, where the difference might lie in small amounts of the primary structure.
What is the story of the outbreak of BSE and of vCJD in humans?
RR: What seems to have happened in the case of BSE and vCJD is that BSE started in GB some time in the early 1980s. It's origin is not entirely clear, but it certainly transmits from cow to cow via eating with considerable ease. And one of the ways in which cattle are fed is to take rendered material from a wide variety of species and make pelleted cattle cake out of it, and this was fed back to cattle and other species as well. So once there were a few animals with BSE in the country, then the uneaten remains of those would be rendered down and made into cattle cake and fed back to cattle. And because the species barrier when you are going from one animal to another animal of the same species isn't there, then transmission was very easy from cow to cow via that mechanism.
Of course, it's a very long incubation period for BSE, about 5 or 6 years, and so this process was going on very quietly in the background without any cattle actually becoming sick, because they're usually killed and eaten at an earlier time. So Britain was then hit, as it were, by a large number of infected animals and really before we knew about the existence of the disease at all a fair amount of beef and beef products had been eaten from animals which were incubating that disease. It wasn't possible to know in the first instance whether BSE would transmit across the species barrier to humans.
As far as we know, scrapie, the similar disease which occurs in sheep, does not transmit across species. But the prion protein from different animals has a very slightly different shape and depending on the compatibility between the prion protein in cows and people or sheep or experimental animals such as mice or hamsters, you will get different lengths of incubation period. In some cases the species barrier is so big that it doesn't transmit at all. Unfortunately that wasn't the case for BSE and vCJD.
What evidence is there that vCJD is caused by BSE?
RR: When I heard about the first few cases of vCJD I was very worried that people were jumping to conclusions that just because they had occurred in GB at the time of the BSE epidemic and shortly after it that we might actually be missing some other, and therefore potentially very important source of infection. So I was rather sceptical originally. But then I think a very key experiment was done that convinced me that vCJD did come from BSE.
If you take BSE from cows or from animals to which it has also been transmitted - because BSE did transmit to some cats and to some antelopes at the zoo and things like that - and you then inject those various sources of BSE into one particular strain of mouse then you get a particular pathology. If you inject material from someone with CJD or from a sheep with scrapie, you get a different pattern of pathology. So that was a key background against which the crucial experiment was done, which was to inject material from the brain of a patient with vCJD into that type of mouse, and to look at the pathology and to compare it with the cluster of those which were of BSE origin and the cluster of those which were of other origin. And it came out looking like BSE, and that was the moment that I was convinced that we didn't have to look for another source of infection for vCJD, it came from BSE in some form or another.
Is the evidence you have concrete proof of the link between BSE and vCJD?
HB: It's actually very difficult to get definitive proof - you'd have to do an experiment, you'd have to inject humans with BSE infected material from cows and prove that they got the same disease as that which we recognise now as vCJD. The problem is, when a scientist does an experiment and he's pretty certain that the outcome says something about BSE, the government then says to the scientist, 'is this the case or is this not the case'. And scientists can't say 'it is the case', what scientists would say is, 'well we think it is, with a probability of such and such'. But this isn't enough to satisfy the public, they want to know whether it's safe to eat meat or not safe.
RR: There's also a difference between doing an experiment and demonstrating an historical event, they're really very, very different things. So even if you were to inject a person with brain from a BSE infected cow and they got sick, that would prove that they got sick, but it wouldn't actually prove what happened in history. A similar problem arises with our understanding of how kuru was transmitted amongst the people of Papua New Guinea. That they contaminated each other I think is not something which anybody would argue about, precisely how they came to contaminate each other is not clear. It is believed that they were cannibals and that they ate the brains of their relatives, including their sick relatives when they had died and that they transmitted it in that way. Whether it was through the brain going through the alimentary canal or whether it was through going through cuts in the skin or holes in their teeth or something is not something that we can now prove.
HB: But it doesn't in a sense matter very much now that we can't prove conclusively that kuru was transmitted by cannibalism. What we know now is enough about the agent that we wouldn't allow cannibalism, because we would think that was a risky procedure. And even though we can't prove definitively that vCJD arose from BSE, we think the evidence is strong enough to make sure we get rid of BSE from the herd, in order to minimise the risk of vCJD.
Is BSE transferred to humans through eating infected meat?
RR: The precise route by which it got through is a problem that is actually very, very difficult to answer. People have tended to blame meat, meat products, all sorts of things, and certainly it must have come from cows in some form or other. We can be fairly sure that it's not some means of transmission, like milk, we are confident that milk is not infected. We're confident that muscle is a very, very low source of infection, so eating what is obviously a piece of meat is probably not one of the higher risk factors. We do know that brain is very infectious, and brain is incorporated into food not as much as people imagine, but it certainly is to some extent - precisely which products are responsible for causing the infection is not at all clear. People have made many assumptions about how much beef they eat and really it's not possibly to say precisely how it was transferred across.
To what extent do the experiments you have done test transmission through eating infected material rather than injected transmission?
HB: For scientific experiments, most of the transmission is done by injections directly into the brain of brain tissue from infected animals, in order to maximise the infection rate and to minimise incubation period. But as far as the BSE / vCJD story is concerned, it's unlikely that the victims of vCJD got it from brain injections. The question is, if these people are getting their disease from BSE, how is it getting in? And the obvious candidate is by the oral route, they're eating it.
Now, there have been some experiments on feeding infected material to experimental animals and it's known that the infectivity rate is much lower, by the order of perhaps 100 000 times less efficient. So relatively few experiments have been done using the oral transmission route. BSE has been transmitted orally from infected cow brain to experimental calves, and these animals have become sick after a very long incubation period and a very low rate. Transmission of disease from non-brain tissue, for example muscle meat, has not been achieved, even though muscle meat does contain nerve endings. But we must assume, we do assume, that vCJD probably arose from eating meat, and it's probably meat which has been combined with brain tissue and other tissues from peripheral nervous areas and stuff like that.
RR: The problem is where you have very, very low levels of infectivity, you would have to inject into an enormous number of animals in order to show that there was no infectivity. In fact, logically you can never do that, and of course in the BSE epidemic the recipients unfortunately were potentially 50 million people and there's no way you can do an experiment that has 50 million recipients to show that something won't happen.
Dr David Brown works at the department of Biochemistry at the University of Cambridge. Dr Brown has researched the way in which metals bind to prions, and believes that the balance of copper and manganese in the environment can contribute to the onset of both BSE and CJD. He believes that metal balances can cause prions to 'flip' in cows and humans independently, and that BSE does not necessarily cause CJD.
Could you summarise the BSE/ vCJD story?
The discovery of the disease scrapie and its parallels with vCJD suggested that perhaps scrapie was transmitted to humans by a virus, and so there was a lot of work to try and find a virus that could transmit a disease between sheep and humans. However, after many years of research nobody was able to find a virus and it was hypothesised by numerous people that perhaps there was no virus and there was no DNA associated with the transmissible agent. And so Stanley Prousner suggested that it could be in fact the protein that transmits the disease, and so much work was spent looking to see how a protein could cause these diseases. And it all boiled down to conversion of a normal protein into an abnormal protein, and this abnormal protein then could then lead to the disease.
Do you believe vCJD is caused by BSE? If not, why not?
I don't believe that vCJD is caused by BSE. There is good evidence that these diseases are very similar, but it is based on research involving infecting mice with extracts carrying the BSE agent or the vCJD agent and then looking at pathology in the mice. And this evidence suggests that the diseases are very similar, but doesn't actually prove one caused the other. One example of evidence of transmission is that of kuru in Papua New Guinea where humans ate the brains of other humans and then developed a disease from that. But that seems to be the exception and there's no proof that BSE or vCJD occur that way. Any evidence is based on a lot of circumstantial statements, such as that the outbreak of BSE preceded vCJD by some period of ten years, which although is true is a bit misleading because the incubation period - the time at which a person has the disease but doesn't show any symptoms - is much, much longer in humans than it is in cows. This implies that probably the first people with vCJD were developing the disease at the same time the first cows had BSE.
Are there experiments that would prove the link?
Unfortunately it's a very difficult thing to prove that BSE caused vCJD because you basically have to infect humans. Even if you could do that it wouldn't necessarily prove it, it would simply prove that you can infect humans with BSE. We would have to see once we had eradicated BSE that vCJD disappeared, and that's going to take 40 or 50 years, even if we eradicated BSE now.
Could you explain your theory that transition metals cause the diseases?
Our work has focused on the ability of metals to bind to the prion protein. The prion protein exists in 2 forms - it exists in the form that occurs normally in the brain and is therefore good and a healthy thing to have, and the abnormal form which is associated with the disease. We've been investigating the differences between these forms and we've found that the normal form binds copper that appears to be necessary for its normal function. However, we know the abnormal protein isn't broken down as readily as the normal protein, and builds up to high levels. So studying the brains of animals with the diesease we have high levels of the abnormal protein but low levels of copper, implying that it doesn't bind copper. What we've done in the lab is to investigate this further by looking for other metals which bind to the prion protein.
We've found that manganese can also bind and it will cause the normal protein to flip into a different structure which is more like that of the full found disease. So we are hypothesising that manganese can substitute for copper and this will lead to conversion of the protein into the abnormal form found in the disease.
It's uncertain yet as to whether the metals directly cause the disease or whether it's the imbalance of the metals. What is important is that anything which will increase the base levels of prion protein expression will make you more susceptible to the disease, because the central event in the disease is conversion of the normal prion protein to the abnormal form, so if you add more normal protein then there's a higher chance you'll generate the abnormal form. And if you have more protein you would also be more likely to incorporate manganese, possibly there would be insufficient copper to occupy all the sites on the high levels of prion protein that you have. So, it could be that another factor somehow is also necessary for the manganese to lead to disease.
Does this mean there is not necessarily a link between the animal and human forms of the disease?
The implication of this is that our hypothesis provides a model by which both BSE and vCJD would be caused by exposure to manganese, and something else possibly, that would lead to the change in humans and cows. Therefore there would not need to be a link between the two. One other thing about this is that if there's a common cause to both diseases then they would have a common pathology, which is what is found. It's theoretically possible that if you eat infected meat then if you have a disbalance in metals that this could lead to disease. However, the problem remains that we still have no evidence that meat will carry the infectious agent or that people when they eat infected meant or BSE-carrying meat that they will develop the disease from this.
So the advice to stop eating beef was a mistake?
I think the government's actions in regards meat and bringing in restrictions was a sensible move at the time because we had no evidence one way or the other what was causing the diseases, and it was far better to be safe than to be sorry. So if it really turns out to be that beef if the cause - which I greatly doubt at the moment, but certainly the possibility remains - I think it's appropriate that the government did make efforts to improve the standard of meat that people eat. Regardless of whether it's infected or not you don't want to eat beef from sick animals.
What evidence is there for this transmission metal hypothesis?
The work of Mark Purdy is of great interest because he's gone to various places in the world to look at regions where where there is high incidence of sporadic prion diseases such as scrapie or sporadic CJD. In these regions he has looked at the soil and he has found that there is a decrease in the levels of copper and very high levels of manganese. This is important because it can influence metals in foliage and food derived in these regions and when animals graze in these regions they would have an intake of high manganese and low copper.
Could you explain why there is confusion over what causes the diseases, whether it's virus or prion-only?
The problem is that when we think about a disease we have to find where it comes from, and in most diseases which are infectious it comes from somewhere like a bacterial virus that's carried by some other animal. However, most prion diseases - or Transmissible Spongiform Encephalopathies, as they have been called - actually occur sporadically with no known infectious event occurring: something just happens in the animal's brain and the protein flips into this abnormal form and then you have disease. And most animals there is no transmission to another animal, that animal becomes sick and dies and that's the end of the story.
Do you believe vCJD to be a new disease?
My hypothesis suggests that vCJD is also a form of sporadic CJD, but whatever is the cause of this disease is new. In other words that possibly there is a slow form of disease, which is what we call the sporadic form of CJD, but there's also a new form which for some reason the disease process is accelerated and people come down with the disease at a much younger age.
Dr Moira Bruce is a scientist at the neuropathogenesis unit in Edinburgh. Dr Bruce is an expert in the behaviour of the different strains of encephalopathies. She believes that there may be some kind of viral-type, information carrying agent involved in the disease transmission process, as well as prions.
Could you outline evidence that suggests vCJD is caused by BSE?
When vCJD was recognised there was a suspicion that it might be a transmission of BSE to humans because it was a new disease, it had very different neuropathology and clinical presentation to sporadic CJD. So we tested that by taking a piece of tissue from a vCJD patient and injecting it into mice. We'd previously shown that if you did the same with BSE it produces a very characteristic disease pattern in the mice, based on the length of time that the mice develop disease and the sorts of changes that you see in their brains. And when we did this with vCJD we found exactly the same pattern of disease in the mice as we'd seen with BSE.
Could you explain the prion hypothesis?
We know that in animals that are infected with TSEs (transmissible spongiform encephalopathies) we see accumulations of a modified host protein, PRP, particularly in the brain and sometimes also in the tissues of the immune system, which has changed its shape and aggregated into forms that are really very difficult to remove. The protein-only hypothesis suggests that when introduced into a new host this changed protein with its abnormal shape interacts with the normal protein from the new host and in some way converts it into the abnormal form. This newly converted PRP goes on to change more and more PRP molecules in the host and this causes an amplification of the abnormal form of the protein, which eventually causes disease.
What do you believe to be the problem with the prion-only hypothesis?
If the prion-only hypothesis were true, this would be a new form of infectious agent, quite unlike conventional viruses or bacteria or any other types of micro-organism. We know that there can be many different strains of TSE agent and we've established this over very many years from studies in mice. These strains produce very different types of disease in the mice when they're injected, particularly in terms of the amount of time it takes for the mice to develop the disease and the patterns of damage that you see in the brains of these mice. The conclusions of the studies is that TSE agents contain some form of information which specifies their strain characteristics: we've shown that this information can be retained when, for instance, these strains are passed through mice with different PRP types or even under some circumstances through different host species. Thus, for example, the BSE strain retained its characteristics after being passed through cattle, sheep, goats, cats and humans.
If these agents contain just abnormal PRP and have no other component, then the existence of different strains and the requirement for an informational component that specifies strains means that the protein itself has to carry that information. It's very difficult to envisage how it could do that. Those who support this hypothesis suggest that there are multiple different confirmations (shapes) of the abnormal protein which can be passed on faithfully to new PRP molecules. But this would involve a very faithful reproduction of that shape over very many cycles of reproduction and in that case there would have to be as many confirmations as there are different strains and we've identified 20 different TSE strains in rodents.
What do you believe the nature of the infectious agent to be?
For conventional viruses, where you also see strain variation, the nucleic acid of the virus determines this property. Many years ago Alan Dickinson put forward the vireno theory that suggested that the agent was a hybrid structure consisting of an infection-specific informational molecule, which might be a nucleic acid, which is very closely associated with and protected by a host component, which would be the host PRP. In my opinion the presence of a small nucleic acid within the agent structure should not at the moment be totally ruled out.
Isn't it a problem for the vireno hypothesis that there isn't immune reaction?
If there is a small nucleic acid involved in the agent structure it doesn't necessarily code for a protein and you would only expect to see an immune response if the nucleic acid was producing foreign protein which triggered a response. So I think it's still possible that there's a nucleic acid there, even though there's no immune response to any form of protein.
What are the implications of all this in terms of the disease?
I think there's very convincing evidence that vCJD is caused by the same strain as BSE, but we really don't know exactly what the infectious agents in this whole group of diseases consist of and I think a lot more work has to be done to establish the molecular nature of these agents.
Is the research being done going to lead to treatment?
These diseases are going to be very difficult to treat because they have a very long incubation period during which time infection is accumulating in tissues without any outward sign of infection. By the time symptoms appear there are very widespread degenerative changes in the brain. So it's very difficult to envisage how a treatment might work after a patient is showing symptoms. There are a number of treatments which do reduce susceptibility, but they're most effective around about the time of initial exposure.
What do you think we should learn from this episode?
I think the whole BSE episode and now vCJD has taught us that we have to take TSE seriously. I think for many years we were happy to tolerate scrapie in our sheep, but I think we've come to the stage where really we should make every effort to eradicate scrapie from our sheep flock.
Professor Peter Smith is the acting chair of the Spongiform Encephalopathy Advisory Committee (SEAC), based at the London School of Hygiene and Tropical Medicine.
What are the different kinds of encephalopathies?
There are a number of different forms of these diseases which affect different animal species. There's a disease called scrapie which affects sheep and has affected sheep for several hundred years. As far as we know we've been eating scrapie infected sheep without harm for all of that period.
There's a disease in humans called CJD which is essentially a disease of old people, a rare disease, affects about one in a million people. We don't know what causes the disease, it may just arise spontaneously through some mutation associated with ageing, but again it's been around for a long time. There are some genetic forms of the disease which are inherited in more or less a Mendelian fashion, which are associated with mutations of the prion protein gene, and that is transmitted within families. And there have been a small number of cases of CJD which have been transmitted through medical procedures, the largest number being through the use of contaminated human growth hormone.
The new diseases in this group are BSE, which was first recognised in the cattle population in 1986, and then variant CJD (vCJD), which we believe is caused by the BSE agent, which was recognised in the human population for the first time in 1995/6. The distinctive feature of vCJD as against CJD is that it affects, so far, predominantly young people.
What is the story of the link between the development of BSE and humans suffering from vCJD?
We don't know where the first case of BSE arose from. There are 2 theories: it could have either arisen spontaneously in a cow or it could have been the scrapie agent in sheep mutating in some way and becoming adapted to cows. That wouldn't have been a problem in itself, but at that time there was a practice of recycling waste parts of cattle and sheep - so-called rendering them - and feeding them back to sheep and cattle as a high protein supplement feed. And in that way an epidemic was propagated before we knew it was going on, because of the long incubation period. So by the time we saw the first case of BSE, in about 1986, there was quite a large epidemic of BSE in the cattle population, affecting probably around a million cattle. Not all of those animals developed BSE, only about 180 000 or so, the remaining number were killed before they showed any signs of BSE and were eaten by the human population. But consumption is the most plausible route for the BSE agent having got into the human population, which, again after an incubation period, manifested itself as this new disease, vCJD.
What evidence is there of the link between BSE and vCJD?
When vCJD was first recognised in the human population in 1996, the evidence actually directly linking it to BSE was circumstantial - although I think many people believed that this was likely to be due to BSE, the evidence for that was more that this was a new disease of the same form as BSE, a BSE-type disease in the human population. It was really in the following year that the evidence that they were caused by the same agent became much firmer, and that was through essentially strain typing studies, where the strain of the agent that was affecting cows was compared with the strain of the agent that was affecting people, and they were found to be identical. And that was really the clinching evidence that these two diseases were caused by agents which were indistinguishable and probably the same.
How do you respond to people who argue that BSE does not directly cause vCJD?
I think there are two questions that people have raised issues about. One is whether BSE was due to the recycling of waste parts of sheep and cattle back to sheep and cattle. The other is whether in fact this disease that we've seen in humans is anything to do with BSE - the alternative is that they've got some other common cause but they're not themselves directly related.
I think the evidence really doesn't support these interpretations. First of all, with respect to the BSE epidemic itself, there was a very sharp fall in the number of cases as soon as controls on feeding were put in place, so I think the evidence that feeding was the main cause of the BSE epidemic is fairly convincing. With regard to this new disease, vCJD, as to whether that's caused by BSE, again I think to my mind the evidence is pretty persuasive that these are caused by the same agent. It's true that we don't know the route by which the agent got into people - the most plausible route is it was through eating contaminated beef products, but we don't have conclusive proof as of yet. However, the biological agents are identical and the temporal sequence the vCJD epidemic following on about ten years later from the BSE epidemic is the sort of thing that we'd expect with this kind of disease.
All of that I think makes a pretty persuasive case that the main determinant of the BSE epidemic was the recycling of waste parts of cattle back to cattle, and the main determinant of the human epidemic is exposure in some way to the BSE agent from infected cattle.
Do we know the exact nature of the infectious agent involved in the diseases?
There is still debate about the nature of the infectious agent. When these were first postulated, unlike bacteria and viruses, it was supposed that these were an infectious agent without any nucleic acid and that sort of went against the tenets of basic biology. As time has gone on I think an increasing number of scientists have become persuaded that indeed this is a protein-only infectious agent, the prion protein, something which every individual produces.
We don't know what the purpose of that normal prion protein is, in fact in experimental studies people have produced mice without a prion protein gene and it doesn't seem to affect those mice very much. But when an abnormal form of the protein is introduced into somebody, it seems that this has a deforming effect on normal prion protein and transforms it into this abnormal form and there's a cascade effect which eventually leads to the build up of this abnormal form in the brain, which leads to disease and shortly thereafter to death. However, there are others who believe there must be some nucleic acid involved somewhere in the process, and that debate continues.
What is the likely scale of a vCJD epidemic?
When the disease was first recognised the first ten cases were reported we had really no idea just how large this epidemic was going to be. We knew that getting on for a million infected cattle had gone into the food chain, so there was the possibility that very large numbers of us had been exposed to this agent. The hope was that there was what is called a large species barrier between cattle and man, such that it would be very difficult for man to become infected with an agent coming directly from cattle. We don't know still exactly how large that species barrier is, and we don't know how large the epidemic is going to be. Since those initial ten cases there have now been a further 80 or so, so the current total is 99 cases. It could be that the eventual epidemic size will be just a few hundred cases, but it could be many thousands.
Our uncertainty is due to two things. One, we don't know what the length of the incubation period is in people, that is the period between infection and when disease occurs. If that's very long, then we could just be seeing the first few cases of what's going to be a very long epidemic. The other uncertainty we have is that we have no test for who's infected, unlike with the AIDS virus where at a very early stage of infection we can test somebody's blood, find out they're infected and they are likely to develop AIDS some time within 10-20 years or so. With this agent we can only really identify those infected either when or just before they develop disease, which restricts us greatly in being able to assess just how many of people in the population are affected.
What are the issues surrounding policy recommendations and taking precautions against transmitting disease?
A number of measures have had to be put in place on a precautionary basis in the face of really enormous uncertainty. For example, there is the worry that once this agent has entered the human population there is no longer a species barrier in terms of transmitting from person to person. So we now lucodeplete, that is, remove the white cells from blood, because of the possibility that there may be individuals who are donating blood who unknowingly are infected with the vCJD agent, and when their blood is transfused to somebody else, they will actually transmit the agent and that person will develop disease. Because we don't know how many people there are who are incubating the disease we just don't know what the magnitude of the problem is. We know it could be very large, or alternatively it could be quite small, and making public health decisions in the face of that enormous uncertainty really does pose many challenges.
How close are we to finding treatment or a cure for vCJD?
I think one has to be fairly pessimistic in the near future about a cure being developed. Firstly, there is a relatively small number of cases of the disease looked at in pharmaceutical company terms, so it's unlikely the pharmaceutical industry is going to invest a lot of research resources, which is where most of our new drug therapies come from, so any research on treatment is mostly being government sponsored.
Also, I think it's perhaps being over-optimistic to imagine that by the time the disease has developed and already quite a lot of damage has been done in the brain that really one's going to find an intervention, a treatment that's going to reverse that damage. I mean, hopefully that may come one day, but that does seem quite a long way off at the moment. Perhaps a bit more hope lies in being able to interfere with the process of the disease at an early stage. There are some animal experiments which have been done in which interventions have been made, giving a drug shortly after we know infection occurs. That has modified the course of that infection, has lengthened it, and that I think is an area where there will be continuing research and will perhaps result in some significant progress.