Well, viruses are the smallest kinds of parasites or microbes that can infect other living things. There are many kinds of viruses. Viruses cause disease in humans; other kinds infect animals, plants. There are even viruses that infect bacteria. They work by getting inside the living cells of their hosts and making more of themselves inside those cells.

There are different classes of viruses. We're particularly interested in the retrovirus.

Well, the most notorious retrovirus in the retrovirus family is HIV, which causes AIDS. . . They're called retroviruses because they go backwards, so to speak, in biological information flow. The central dogma of molecular biology that holds up, by and large, is that the genes of almost all living organisms, in the form of DNA, can make RNA, and RNA makes protein.

Viruses can be different. You can have viruses with DNA genes or viruses with RNA genes. Examples of viruses with DNA genes are smallpox or herpes viruses. There are many others. There are also many viruses that carry their genes in the form of RNA. The viruses that cause human diseases such as measles and mumps, rabies, and influenza have RNA genes.

Retroviruses carry their genetic information in the virus particle that goes from one person to another, or from one animal to another . . . as RNA. But when it gets inside the living cell, it makes a DNA copy of the RNA. That's where it's gone into reverse; that's why it's retro, because normally, DNA can make RNA, but RNA cannot make DNA. . .

Is there any sense of how long they've been around, and how they've evolved?

Well, how ancient are viruses? The kinds of viruses we know about today could not have been around in the world before true living organisms evolved, because they're all parasites. Many evolutionists accept that an RNA world came before a DNA world, and they think that some RNA viruses may be relics from those days. But we just don't know.

And obviously, there's no fossil record for something that's so small; you can't even see it with a microscope. But we can look in the living organisms, in the hosts that they infect, and say, well, how many different species harbor similar viruses? What evidence is there that they've jumped around? How far back do they go in evolution? And almost certainly, the major families of viruses have been around for hundreds of millions of years, including retroviruses.

Do viruses and retroviruses always cause harm in their host?

I don't think all viruses necessarily cause disease in the hosts that they naturally infect. Some viruses can be deadly, like smallpox, and others may be almost harmless, just getting a free ride without causing much harm. There are other viruses that do no harm, most of the time; but in certain conditions, they may come up and cause disease. In the case of humans and animals, that's usually because our immune system is pretty efficient at controlling viruses. So we may get an initial disease and then clear the virus out. Or we may get an initial disease that's very minor and put the virus to bed, or the virus may put itself to bed.

Let's take herpes simplex virus as an example, which almost all of us pick up, probably as infants. It may give us a cold sore and nothing much else, but we never really get rid of it. Once it's given you a little sore on the lip, it also travels up the sensory nerves and gets embedded in the nerve ganglia, and there it can stay for the rest of our lives. Normally, it will cause no harm. But if we are particularly stressed out, either because our immune system isn't working properly -- as in AIDS patients or transplant patients, or due to psychological stress -- you can break out in cold sores. And if you're seriously immune-deficient, like advanced in AIDS, then something as innocent as this cold sore virus can spread throughout the body and kill you.

Are there any viruses that have, in a sense, become evolved to permanently establish themselves as part of the host animals they seem to infect?

I think there are some viruses that are so well adapted to the hosts that they infect that they live with them, and co-evolve with them, over evolutionary time. The herpes viruses I was just talking about are like that. Some of the retroviruses have gone even one step smarter. On occasion during evolution -- at least during the evolution of vertebrates birds and mammals and amphibians, and so on -- [the retroviruses] managed to insert their genes into the DNA of the chromosomes in the host. And in fact, they have to do this in order to replicate. It's an obligatory step in the reproductive cycle of the virus, that once it's made its DNA copy [it] must integrate into the DNA of the cell.

Occasionally, these viruses have infected the cells . . . that are destined to become the eggs and the sperm of the host species. And that's a clever thing for a virus to do, in a sense, because if it's integrated latent, keeping quiet there, it gets a free ride to the next generation. In this way, the virus becomes inherited as a genetic trait, just as we might inherit blue eyes or some other feature.

So these are a class of virus that is impossible to remove? They are part and parcel of the very being of that animal?

Yes. This particular subset of retroviruses we call endogenous viruses, because they are part and parcel of our own DNA. They are, therefore, difficult to eradicate or eliminate. We have them on board ourselves as humans. . . .

Do endogenous retroviruses exist in the pig?

We think that most animal species amongst mammals, if not all of them, contain some classes of endogenous retroviruses -- these retroviruses where the genes are in the chromosomes. Most of these are not infectious, so they can't come out again as infectious viruses. They've been there so long that they've become defective. Chunks of the viral genes have come out, or they've mutated in some way, so that they can't be regenerated as infectious virus. For example, of all these many retrovirus sequences in human DNA, none have been shown to be infectious, to come out again.

But in some species, they can become activated and emerge as potentially infectious viruses. We first found this in chickens over 30 years ago. The domestic chicken has a virus like that, which you can recover in infectious form. Interestingly, it can't re-infect chickens, but it can infect turkeys and quail and other species of birds. The same is true of mice and of cats, and it turns out to be true of pigs. And that's why we're concerned about this type of retrovirus in xenotransplantation.

Why did you first become interested in the pig retrovirus?

I first became interested in the question or the problem of pig retroviruses really out of curiosity, because I'm a retrovirus aficionado. I've studied retroviruses for most of my career. They're my pets, if you like. And there were research reports published back in the early 1970s, over 25 years ago, that gave some preliminary evidence -- as good as it was with the technology of those days -- that pigs contained endogenous retroviruses, and even that they could be recovered in infectious form.

The one experiment that tested whether they could infect human cells gave a negative result. But it was only one cell type, and it wasn't continued for very long. So we knew from these old reports that pig retroviruses existed, and could be recovered as infectious virus particles. With the debate about the use of pigs in xenotransplantation, of course this question popped up again. . .

Did you and other experimenters set out to discover how many of these endogenous retroviruses were actually in the pig DNA?

. . . I was able to persuade a very able student in our lab . . . to take on a small project, to re-examine these pig endogenous retroviruses, using the kind of modern technology for studying viruses that we have available to us today. The upshot was that we found more than one strain. We know of three infectious strains, and that two of those, there, at least, could infect human cells in culture.

And we published that also in Nature Medicine in 1997, and that had an immediate impact in the field. We also looked to see how many copies of these porcine or pig endogenous retroviruses (PERV) were embedded in the chromosomes of the pig. And we reckon there are roughly 50 copies in the average domestic pig, if you add the three different strains of retrovirus together. There's perhaps 30 of one strain, 15 of another, and 12 to 15 of the third strain, although that third strain is more variable. . .

What experiment was actually involved to show that PERV could infect human cells?

We went about this investigation first by looking at the handful of research reports that had been published 25 years earlier. We obtained some pig cells that grow permanently in culture from a repository . . . which had been reported to release virus with reverse transcriptase activity -- that's the marker enzyme for retroviruses. And we were able to confirm that, indeed, these cells that were growing in flasks in the incubator were producing lots of virus particles.

We then exposed human cells in culture to the semi-purified virus particles. And we found only one human cell type, amongst about 15 or 20 that we tested, which could readily become infected with this virus. However, if we took the pig cells that were releasing this virus and cultured them in the same flask as human cells, then we found that about half the different human cells we tested were susceptible to infection. This means that human cells are not very highly sensitive to infection. To put it another way, the virus wasn't a ripping hot virus for human cells. But if you allowed the pig cells and the human cells to be next to each other, given enough time, the virus would get in.

Think of the xenotransplantation setting, where human blood cells will be coursing through the pig organ, where human cells that line blood vessels will begin to invade and line the blood vessels of the organ. Then, of course, you can see that that this co-cultivation experiment in culture was perhaps the more significant one. And that really did show us that this virus had the potential to infect humans.

What was the worldwide scientific reaction when news of that discovery was published?

Well, when our first report was published, showing that these pig viruses could infect human cells, it really had an even bigger impact than I expected. It raised the debate about possible animal infections in xenotransplantation onto the front line. Prior to that, the FDA in America had really decided to . . . leave decisions about whether to go ahead to local ethical committees in the hospitals and research medical schools concerned.

Our report had an immediate effect. The FDA in America called their xenotransplant committee together. They organized workshops to discuss this virus more seriously. Other laboratories, including the FDA's own research labs, became involved, as did the CDC. And much more effort was put into studying these viruses. I think this is a welcome thing that the medical community and the regulatory authorities responsible for these things were able to respond fairly rapidly to new evidence, like the evidence we published. . .

Is it possible to have any kind of xenotransplant -- xenograft, cellular, whole organ, regular or transgenic -- without being exposed to the pig endogenous retrovirus?

The problem of possible infection through xenotransplantation is quite complex. We've got to weigh the benefit of the transplant and the risk of infection to the individual patients. But we also must consider if an animal virus transfers to the patient, whether it might take off and spread in the human population. We have to weigh that possibility, too. So it's quite difficult to come to a sensible decision about this, because these are unlikely events, but not impossible events.

By and large, pigs that are going to be used as sources of tissues or organs for transplantation are going to be very carefully screened that they are free and clean of the known pig viruses. So they will be derived from special herds of pigs that are kept behind barriers with filtered air intake. The ancestors to these herds will probably be born by cesarean section, so that no virus on the surface of the mother pig can get across to them. And in this way, one can exclude the majority of viruses and other microbes that might be there in the average farm pig, that almost certainly are there.

The problem with endogenous retroviruses is that they are sitting there in the DNA of the pig chromosomes, so it doesn't matter how clean the pigs are; the viruses are sitting there right inside them. And it's not going to be easy to get rid of them. . .

Do you feel that the cellular transplants that are going on at the moment pose a risk, in terms of possible infections from endogenous pig viruses?

There's a question of whether all forms of xenotransplantation are equally dangerous, in respect to endogenous retroviruses and other microbes. In a sense, because the virus is sitting in the DNA of all pig cells, it doesn't matter which organ you take or which tissue you take; they're going to be there. In terms of saying, are they going to come out as infectious viruses, there may be differences between one tissue and another. So they might perhaps pop out less regularly from, say, nerve cells in the brain that are being used for Parkinson's disease, than, perhaps, cells in the pancreas that could be used for diabetes, or perhaps from putting a whole kidney into a transplant recipient.

But at the moment, we don't have enough detail about this. We cannot say for sure. I think I could say for sure that it would be rash, that it would be unwise, to say that any particular tissue or organ is safe.

The committees that are trying to decide about relative risk and dangers tend to say that implanting whole organs that are going to last for years in people is likely to be more dangerous than being exposed to a few pig cells, or taking the human blood circulation outside the body and perfusing it over pig tissues. . . . I think that's right, in the common sense sort of way. A whole organ lasting a long time gives you a higher potential dose of virus. But they might all be over the threshold. It might only require a few minutes' exposure to do the trick. We simply don't know yet.

Regardless of circumstances, does exposure to the pig retrovirus always inevitably lead to infection? Do we know what will happen, in that sense?

Exposure to an infectious agent doesn't necessarily lead to infection. You could have sexual intercourse with an HIV-positive person, and you might well get away with it. The calculations are that you do not get transmission except in a small percentage of such sexual encounters. But it sure does happen, or we wouldn't have the worldwide epidemic of AIDS in front of us. So mere exposure doesn't guarantee infection.

I mentioned HIV because it's also a retrovirus. It's a distant relative of the pig retrovirus, a second cousin, so to speak. It looks to me, from the knowledge we have at this time, that the pig retrovirus is going to be substantially less contagious for humans. So I think it's going to struggle to infect xenotransplant patients, and I don't think it's that readily going to adapt to pass on from person to person.

But life is full of surprises. It might well adapt inside the first patient, as, presumably, HIV had to adapt when it got into the first human, and then take off. So we're looking at quite a long time horizon before we can decide whether such viruses are safe to have on board, or might evolve to become less safe. These are all unknowns at the present time.

Given the class of virus that PERV is, do you have any idea what kind of illnesses it would cause if it did become a hot virus in humans?

We could ask whether PERVs are likely to cause any disease at all. The most sensible way to consider that is to look at the closest animal relatives to the pig virus, which are viruses of mice and dogs, and also of a species of ape . . . and say, "Well, when they get a hot virus infection, what do they go down with?" And, by and large, the most common disease that occurs is a type of cancer, usually a lymphoid cancer. They develop lymphoma or leukemia.

For this to happen, the animal has to have quite a high dose of virus on board. The virus that gets introduced may be a low dose, but it then propagates within the infected animal to produce lots of virus. And the disease only appears after quite a long latent period. So if we extrapolate that to the human condition, if we took on board the pig virus, we might expect it to be incubating as a silent or inapparent infection, perhaps for years before it came out and caused the disease.

Cancer isn't the only disease that this type of retrovirus can cause. It can cause a form of immune deficiency. Related viruses can cause neuro-degenerative diseases, paralysis and encephalopathy -- brain disease. Related viruses in chickens have caused bone disease. It's quite extraordinary what a broad range of disease these retroviruses cause. But most of the time, in most of the animals they infect, they're causing no disease at all.

So, in a sense, for a terminally ill transplant patient, even if it was provable that some of these pig retroviruses could cause disease, that might not be unacceptable to someone who is facing a death sentence if their organ was failing?

I think the risk of infection and the risk of disease in a transplant patient is much higher than just getting bitten by a pig, or even a healthy person being inoculated with the pig virus. By necessity, so that the transplant patient does not reject his animal organ or animal tissues, he or she is going to be immunosuppressed. And we know that immunosuppressed people are more susceptible to unusual infections from animals, as well as to human infections than the healthy person. So there is an added risk here.

You might say that, at the expense of the transplant patient, we can evaluate earlier whether these viruses are actually dangerous, because you are doing, so to speak, just the right experiment to find out. You're putting whole living pig tissues into people, and you're immunosuppressing the people so that it's not rejected. If you wanted to design an experiment to test whether such viruses will get across and will cause disease, you could hardly do it better.

On the other hand, from the point of view of the individual transplant patient, if they are so severely ill with a life-threatening disease that they need a pig organ or pig tissue to survive at all, from that individual's point of view, the risk may well be worth taking. I think if I was in that position, I'd say, "Go ahead, I'll take the risk, because I'm a goner otherwise, in any case."

That's the individual perspective. But what's the worst-case scenario of this from a public health point of view?

We virologists and microbiologists are enjoined to not only think of the potential benefit to the transplant individual or to the few dozens of patients that will be treated in the first trials, though that might grow to many hundreds of thousands if they're successful. We have to think of public health, as well. And the worst-case scenario is that we could be setting off a new pandemic that would spread across the world, just like HIV has done. That is the worst-case scenario. I don't expect it to happen. But I can't say that it's impossible.

What would the PERV virus have to do to create a pandemic? . . .

. . . Well, it's not inconceivable. But you would be asking the virus to change and adapt to growing in humans to quite a degree. So the virus, when it gets out of pigs and infects humans, won't become a genetic trait in humans, at least not for thousands of years. It will insert its DNA into the chromosomes of human cells. But as an ongoing infection, it's much less likely to do so into the germ cells. And in any case, the first transplant patients are going to be strongly advised not to have children, though quite how you control that stringently is a more difficult question.

So the virus is now going to be in human tissues, and the demands on it are to propagate itself, and to spread throughout that individual infected person. The virus can go through millions of replicative cycles, producing its own progeny, its own daughter viruses, again and again and again. So it's got plenty of chance to evolve fast. And we've seen that by looking at the evolution of HIV, which is a recent introduction into mankind. I think PERV, being the kind of retrovirus it is, won't change quite as rapidly as HIV. But the potential for it to change is there.

How would it then learn to become transmitted from the first infected patient to others? Probably, in the first place, by very intimate contact, by sexual fluids, by contact, by kissing. Different retroviruses in different animals can be transmitted in many different ways. We don't know of any natural transmission from animal to animal via aerosols. But there are some retroviruses that even get transmitted from one mammal to another via biting insects, so they can get around. They can learn new tricks.

But I would expect that the most probable way of becoming transmitted would be through sexual contact and through kissing, through the intimate association of fluids and of mucus membranes. Obviously, the first generation of transplant patients will not be allowed to donate blood, which could be a very efficient way of passing it on to others. So it's going to be important to monitor not only the first patients, but their intimate contacts, their husbands and wives, their children, household contacts, to see whether there's any evidence, not only of infection in the patient, but of infection in those they're living with and among.

Does the fact that the pig retrovirus is dormant in the pig guarantee that it would remain so, if it traveled to a new human host via a transplant?

We could argue that the retrovirus in the pig is dormant, just sitting quietly in the chromosomes. Actually, it's turned out to be less quiet than we thought. Dr. Walid Heneine and his colleagues at the CDC in Atlanta have shown that the average farm pig has quite active virus production. You can interpret that in two ways. One is that for the last 6,000 or 7,000 years that we've lived with domestic pigs, we haven't become infected, although pigs have the virus around. Or, we could say, "My goodness, if those pig cells or organs get inside us, they're readily producing virus, and they're not so dormant."

But once a virus gets into a new host, it's very difficult to judge whether it will grow towards latency, dormancy, or whether it will hot up, whether it's likely to become more active. I would guess -- but it is pure guesswork -- that with this kind of virus, it will go the second way, and could well become more active. Because it's in a new host where there's no age-old host- parasite-evolutionary controls in play. And so the virus really has virgin territory in which to do what it likes.

On the other hand, from the limited investigations we've done with the infection of human cells in culture, these viruses grow in a rather wimpy way, as retroviruses go. In human cells, they have not grown to high levels. And we're wondering why they grow so poorly, when retroviruses quite closely related to them can grow to very much higher levels in human cells.

However, one of our colleagues, Caroline Wilson at the FDA labs, has shown that the level of virus increases as you pass them through human cells after two or three passages of putting the viruses in human cells, taking the progeny virus, and putting it on new human cells. So there is possibly a mechanism for these viruses to adapt to growing in humans and getting on with their business rather faster than they otherwise might have done.

We have to be careful about this, because xenotransplant as a treatment is aimed not for the few, but for the many, isn't it? We could find ourselves a great many opportunities for the virus to evolve within one person. But actually this is being touted as a possible solution to the worldwide donor organ shortage. Does that put an extra facet of caution in your thinking on this?

. . . The biotechnology and pharmaceutical industries have not been slow to see those opportunities and have, collectively, invested billions of dollars into developing xenotransplantation possibilities. So if these viruses are not highly contagious and, therefore, if the first handful of xenotransplant recipients do not become infected, then the pressure will be to do ever-increasing numbers.

To my mind, that's when the problems actually become more complex. In some ways, we may be reassured if people don't become infected. But we don't know whether one in a thousand or one in a million people will, and what will then happen to that virus once it's adapted to grow in humans. I would say to be forewarned is to be forearmed.

And we do know quite a lot about retroviruses these days. For example, one study that we have been involved in collaboration with the CDC is to say that these viruses, as I've mentioned, are second cousins to HIV. And we're asking if there any anti-HIV drugs already licensed for use in humans that might also prevent the replication, the growth of pig retrovirus. And indeed, one of them is quite good at doing this. Most of the ones that are tailored to knock HIV on the head don't touch PERV. But the fact that we've already got one is something to be saved up. If one of the early xenotransplant patients proved to have a roaring infection of PERV, we already have a drug licensed to use in people. . . .

In the history of microbiology, have you ever offered a virus an opportunity like this?

We have known about animal-to-human infections -- zoonoses -- since ancient times. I know of an old cuneiform tablet from ancient Babylonia that talks about the fine if you let your dog bite a human and the human dies from rabies. Rabies is an interesting example. If you do catch rabies from a bite, you die. It's almost invariably fatal. But there are very few cases of human-to-human transmission of rabies. So long as you can prevent the rabid person from biting the people who are caring for him, then that infection stops dead with the first human death.

It's the viruses that perhaps are not so acute, that you can't detect so easily and have time to spread before symptoms appear, that turn out to be the major difficulties. And HIV is an obvious example. Some others have crossed over from animals to man and have set off epidemics more rapidly. New strains of influenza virus, of flu, probably come from birds, but pigs may be the so-called mixing vessel. They may adapt to become flu-like and to be transmissible through coughing and sneezing as they pass through a pig, and then get more readily picked up by people. So we do have a number of examples of that.

And new ones occur as well. Just because we've lived together with pigs for so many thousands of years doesn't mean that there aren't new viruses. This could be a genuinely new virus, like the outbreak in Malaysia that started in 1998. More than 250 people have died from this pig virus. Well, in fact, it looks as if that virus was new to pigs, as well as new to people. And it's believed that the natural reservoir is in fruit bats living in the tropical forest. It may have started through the massive lumbering, cutting down the tropical forest when the new international airport outside Kuala Lumpur began to be built. The bats that can fly took flight from the falling trees and the trees that were still standing were those around the farmsteads. That's how it got from the bats to the pigs. This is speculation, but it's quite reasonable speculation. So we don't know when a new virus might get into pigs.

. . . There are other viruses that might have been there all the time, but we simply didn't know about them, like the pig retrovirus, or the newly discovered pig virus that's related to human Hepatitis E. That was only discovered in 1997 in America. It turns out that this virus is extraordinarily widespread in pigs. And there's some evidence that it does infect people, and that people who work in abattoirs and who are pig farmers are more likely to have signs of infection than others are.

So there might be viruses that are there all the time [but] not yet known to medical or veterinary knowledge that could still cause a problem. All of these are problems that can be worked out, but they need to be studied very thoroughly. . . .

. . . In xenotransplantation . . . the pig organs and the pig cells will be "humanized" by the insertion of human DNA to try and overcome the rejection process. Is that significant, in terms of what viruses could potentially do in that situation?

In considering the use of pig organs and tissues for xenotransplantation, I've been concerned that the very medical developments that may allow those pig tissues to be accepted by the transplant recipient could increase the risk of transmission of viruses. On the one hand, the type of husbandry, keeping them in contained units, will tend to exclude most of the infectious viruses that travel from pig to pig. But the fact that the pigs are humanized, that they are made transgenic to express certain human proteins on the surface of the pig cells, could allow some pig viruses to more readily adapt to cross into humans.

This is for two quite distinct reasons. First of all, it seems to have escaped the notice of the scientists developing these transgenic pigs that the very human genes they were breeding into them can code for proteins that act as virus receptors. After all, they're not virologists; they weren't reading that type of research report. But it's odd that that is the case.

The second reason is that those human genes have been put into pigs to protect against hyperacute rejection, which is a type of rejection that's mediated by factors in the human blood, including complement. Many viruses, including the pig retrovirus, mature from cells by budding through the cell membrane. So they have the same outer envelope as cells do. And they can also be susceptible to inactivation by these blood factors, such as complement.

If you breed the human genes into pigs that protect the cell membrane from being destroyed in hyper-acute rejection, any virus that has a cell membrane -- like outer envelope-like retroviruses, like influenza viruses, like measles viruses -- may well also be protected. And we have to ask why humans have this hyper-rejection phenomenon against foreign tissues. It's probably evolved to protect us against animal viruses. So we are deliberately breeding pigs where that type of barrier is going to be broken down. And that is another area that I think needs to be looked at.

What I am saying is speculation. But it's speculation that can lead to the design of investigations that could bring solid answers and say whether I'm just a scaremonger, or whether, indeed, I've got an important point to make. ...

Do you think we should be looking to do that urgently?

I think this is an important question, and one of our students is currently setting up to investigate this problem.

Have there been any recent studies showing that pig retrovirus can cross over, for example, in mice?

After our initial report that the pig retroviruses could grow in human cells, there was very much concern, even alarm, about proceeding with xenotransplantation. Several other groups were able to confirm our results and extend them to other cell types and other conditions quite rapidly.

However, surveys of people who had already been exposed to living pig cells for one reason or another showed no evidence that any of them had picked up a pig retrovirus infection so far as we could tell. And we used really quite sensitive techniques. Our characterization of the virus allowed very sensitive detection methods to be developed. So we thought, well, maybe this virus isn't so dangerous after all. We don't know whether it causes disease, and the 160-odd people who have been investigated, who have been exposed to living pig tissue through experimental xenotransplantation -- through having pig skin to protect from third-degree burns; through a rather odd practice in Russia of allowing patients' blood to be pumped through pigs' spleens -- none of these people seem to have picked up infection. That looked pretty good. . .

The significance of this new report from the USA and the UK is that the infection was seen not just for cells and culture, but in the living animal, in the mouse. And the mouse was not deliberately injected with purified virus. It was simply xenotransplanted with pig pancreas cells. So it's significant really, in both ways, that the cells on xenotransplantation really do start to release virus, and that that virus can take in the living animal. There is also a report that guinea pigs inoculated with the virus can become infected. So these are significant. It means xenotransplantation can allow cross-species virus transfer.

Actually, we've known this for about 25 years, when we do xenotransplantation in reverse. A very useful method in cancer research has been to grow human cancer cells as xenotransplants in the same kind of immunodeficient mouse. And we know that mouse retroviruses can then invade the human cancer cells in the mouse and infect them. So the viruses can go in both directions in situations of xenotransplantation.

In this latest mouse study, are you describing the pig retrovirus taking the first steps towards becoming a hotter virus? . . .

The recent report showed that mice can become infected, but the infection was still at a very low level. We don't yet know -- because these are very early steps made by these investigators -- whether the virus can hot up. We'll have to wait to know the answer to that. . . .

In public health terms, can we make xenotransplantation risk-free from the PERV virus?

I don't think we can make xenotransplantation risk-free in public health terms. Any new medical procedure carries a certain element of risk, but so does crossing the street. So we should not be demanding an entirely risk-free medical environment. It's not going to happen. More specifically, can we eradicate the risk of PERV transmission to humans? I don't think we can eradicate it, eliminate it; not yet. It may become possible in the future, though it will take many years to cut out the PERV genes that give rise to infectious virus, one by one, getting rid of them from the pig chromosomes. That's going to be a big job. But that doesn't mean it shouldn't be tackled. And we don't know whether it would be successful. I think we should make a start now. But I cannot say for certain if or when that will be accomplished.

There are some breeds of pigs -- I'm thinking of the miniature swine -- that seem to offer at least some hope of having naturally getting rid of some of the PERV virus. What do you make of that as a possible help for the future?

. . . Some of the miniature swine lack one of the three groups of PERV, the one that doesn't infect human cells readily, so they still have the other two that are the real problem. . . . These viruses have been in the pig chromosomes for a long time, probably about 30 million years. We're not going to get rid of them that easily.

Do you think that we can only move forward with this and learn by engaging in some kind of very carefully controlled human clinical trial involving xenotransplant?

. . . The potential benefits are enormous, and I think the likelihood of the worst-case scenario setting off a serious epidemic that can then not be stopped is very, very remote. And, therefore, I am not against the notion of going forward step by step, cautiously, but thinking about what we're doing.

I'm not sure that the pioneering transplant surgeons are always good at thinking ahead about infection. If they were, perhaps those baboon livers would not have been transplanted into patients and then the patients analyzed eight years after they died. Only then did they say, "Oh, let's look at this and that virus and see if they became infected, because their tissues have been preserved in a freezer since then." I think the right way is to think ahead. But certain things cannot be found out, cannot be clarified, without doing limited clinical xenotransplantation.

Is it possible to move forward in this area of science without animal experiments? And do you think the public needs to understand and appreciate that they are done in the best possible way?

Can we make medical advances and save lives without using animals experimentally? In some small areas of medical research, we can. But it's my opinion, as is the case for most medical researchers, that some use of animals has been absolutely essential for the progress we've made to date. There is no drug that's licensed for use that hasn't been tested for toxicity in animals, and, quite frankly, I prefer them to be tested in mice or rats before use than tested in my grandchildren.

There are no vaccines today that haven't been tested in animals. We would not have eradicated smallpox and could not now be eradicating poliomyelitis without the use of animals. Even today, every batch of polio vaccine that is used, that is released to give to babies, must be tested on animals first, to check whether it has reverted to so-called neuro-virulence -- whether it could become paralytic. Animals are absolutely essential if we are not going to revert back to Stone Age medicine.

. . . Having said that some use of animals is essential for medical progress, I do think we need very stringent regulations to make sure we don't abuse animals any more than is necessary. Not all the procedures are pleasant for animals. Many of them have to end in the death of the animal. And all this must be done in a proper manner, and the animals must be treated as humanely as possible. Cruelty has absolutely no place in medical research.