21. Evolutionary Medicine

Prof: Well today we're going to talk about evolutionary medicine. And there are some resources that you can use, if you get interested in this. There's a book, got twenty-three chapters in it. There is a website called Evo-Med Symposia, that you can go to, and you can have talks on the evolution of ***, antibiotic resistance, etcetera; and that's in Streaming Digital video. So each one of those talks lasts about an hour. So if you get interested in any of that, you can actually see--and by the way, it's not just the PowerPoints, you see the people giving the talks. So if you want to actually more or less meet these people and see what they're like, that's a place that you can do it. Now the range of issues in evolutionary medicine is really quite large. And I often get asked--you know, people haven't heard the term before, they don't know what evolutionary medicine is. So here is a description. Okay? Part of it is that we contain traces of our evolutionary history and they bias our responses in significant medical issues. So there's the hygiene hypothesis about autoimmune disease. There is our genetic variation for resistance and drug response. There are traces of the selection that illnesses, that diseases, have written on our genome. Then there are issues in reproductive medicine. And the human life history is really quite special. If you contrast us to chimpanzees or bonobos, human females are capable of pumping out children about twice as fast as a female chimpanzee, and the only way they can do it is by having help. So it indicates that we have been highly social for a long time, and our life history has responded to that. You know something about genetic conflicts, imprinting and mental disease, because I talked about that earlier. Then there are the issues of ovacytic atresia and selective abortions and mate choice, which are an interesting part of reproductive medicine. A big part of evolutionary medicine has to do with the evolution and ecology of disease. And diseases have adaptive strategies. They have their own agendas. They have, many of them, have developed ways of avoiding our immune responses, of manipulating hosts. Some of them manipulate us; that's what coughing and sneezing is about. That's also what making us extremely tired and lying down is about, in malaria. Their virulence evolves, they evolve drug resistance rapidly, and those are very significant medical issues. Then there's all the information that's coming in now from evolutionary genetics and genomics about where viruses originated. So, for example, the detective work necessary to determine that the sooty mangabey was the ancestor of ***-2 is done with molecular phylogenetics, and that the chimpanzee is the ancestor of ***-1; that the SIV living in the chimpanzee is the ancestor of ***-1 is done that way. Then there are very significant differences between different kinds of bacteria, in terms of their genetics and their population biology, particularly in how readily they can do horizontal gene transfer. So that if a bacterium in one species evolves resistance to a certain drug, how likely is it that that resistance gene will get into another species? Okay? That's obviously a critical question, and it depends on the particular kind of evolutionary genetics that that bacterium has; and they vary in this respect. Okay? And then there are all of the issues about under what conditions do new diseases emerge? And that itself is quite a growing field. Then there's all about the degenerative diseases. Okay? How did aging evolve? And given that we have an evolutionary theory of aging, what can we expect to be the characteristics of the aging organism? Are they going to be simple or complex; and if we fix one thing, will another thing break? That kind of issue. We can view cancer as an evolutionary process. Every cancer is its own little microevolutionary process. A population of cancer cells is a genetically heterogeneous ball of growing cells; that has important implications. And then there are links here in degenerative disease--in heart disease, obesity and diabetes--back to traces of our evolutionary history. So that's the scale of the issues. If we think about the--oh let me just run through that quickly-- if we think about traces of our history, we usually think about hunter-gatherers and the kinds of groups that they lived in. If we think about the evolutionary biology of diseases, we think about things like Ebola and *** and malaria. And if we think about degenerative diseases, we think about this process. So that's what evolutionary medicine is about; it's about a lot of different stuff. Okay? So I can't talk about all of that. I've just described the course that I gave last fall in Copenhagen. Okay? It took two months. > So I could give you some important classical themes. I could give you some surprising new insights. I could give you some overarching general messages, such as our bodies are compromises that impose indirect costs; or that evolution takes time; or pathogens have their own agendas. I could present research, stuff I've worked on myself. I've done a fair amount on the evolution of aging, and I'm currently working on how natural selection is operating in contemporary human populations. Or I could give you messages primarily aimed right at practicing doctors; so practical applications in clinic and public health. And this is what I chose: I chose Mismatches to Modernity. So I'm going to talk a little about thrifty phenotypes, and parasites and autoimmune disease; and then I'm going to talk about how pathogens have their own agendas and evolve rapidly. Okay? So I hope you've got the picture. This is a small portion of the subject matter of evolutionary medicine. But these are arguably important themes. So the point about thrifty phenotypes is this: Early life events are failing to predict late life environments. Perhaps they used to be good predictors, or perhaps those early life events were correlated well with the environment in the Pleistocene, for ten or fifteen years, something like that. What we do know is this: if you nutritionally stress a mother and infant, the fetuses and infants will have increased risk of obesity, diabetes and cardiovascular disease fifty or sixty years later. And the initial data that demonstrated this came from the Dutch Hungry Winter. The idea is that stress early in life is switching the individual into a physiology that's very effective at conserving energy, but it is inappropriate if there's an adequate diet. So the muscle cells become insulin resistant, fat becomes concentrated in special depots. And we now have a lot of data indicating that this is the case in humans. So they come from the Dutch Hungry Winter of '44/'45, when the Nazis basically cut off the food supply to Amsterdam, and actually to much of Holland. But you can also see this when there have been historical famines in Scandinavia. In the late nineteenth century there was a famine in Finland; and more recently, in the U.K. and the Philippines. And you can reproduce this in rats and sheep. By the way, the fact that you can reproduce it in a model system is quite important, because it means that for whatever reason that thing evolved, that kind of reason must also have been there for something as short-lived as a rat. Now if we look around the world, about 20% of American adults are obese. Interestingly, in rural Mexico, 60 to 70% are obese. That's not something you'd necessarily expect; go into rural Mexico, you don't necessarily think that those people have a lot of money to eat a lot of food, but they are obese. The incidence of diabetes is exploding; so late-onset diabetes is exploding. As you might expect, most of them are in India and China, simply because the populations of India and China are so large. And this is becoming a really significant portion of the world health budget. So these are significant issues. And if you look at percent obese across many countries, the least obese nation is Japan, and a lot of the European countries kind of have low levels. But the ones that have very high levels of obesity are the U.S., U.K., and Germany, Australia. These are not necessarily the ones in which this kind of nutritional stress early in life would be very frequent. It is much more likely that as countries like India and China, and countries in Africa and Mexico, go through the demographic transition, and go through the economic transition into developing countries, so that they have a parental generation that was more food stressed, and an offspring generation which is more well fed, and more exposed to junk food, that you will get this kind of a response. So when we look at this kind of data, there's a lot of this that probably really isn't due to the thrifty phenotype hypothesis. I would guess that of the total amount of obesity that you see in the world, the part which is really due to a developmental switch being thrown early in life, and then setting that phenotype up to respond inappropriately to a rich diet late in life, thereby developing heart disease and obesity and so forth, at about the age of fifty or sixty, is probably somewhere down around 5%. So not all of it, but probably a significant component. And the argument is that that was something that was adaptive in the Pleistocene environment, because if you could switch the offspring into a thrifty phenotype, it would have a higher probability of surviving the dangerous childhood years and making it perhaps to its first reproductive event. And, in that environment, what was going on at age fifty or sixty was probably irrelevant because most of the population was dead by then anyway. So that's the kind of evolutionary argument that gets at it. I don't think we actually know what's selected for this. I think that we have a plausible evolutionary story to tell about it, but the fact on the ground is that it really happens. So it's important to know about, and it might be for the evolutionary reasons that I just mentioned, but we don't know. Okay, now here's one where we are a little bit more certain. And again, this is a hypothesis that is in the category of things where humans are mismatched to modernity. So they are experiencing a disease which is caused, in part, by our historical shift into a civilized state. It runs like this: our immune system coevolved with worms and bacteria. So it more or less evolved on the assumption that we would always have worms and bacteria in our bodies. And when modern hygiene--so basically good clean water systems-- and antibiotics take out the worms and bacteria, our immune systems respond inappropriately. We can see that autoimmune diseases are actually exploding. So asthma, allergy, Type-1 diabetes, multiple sclerosis, other auto--Crohn's disease--other autoimmune diseases are increasing very rapidly. And as the infectious diseases have gone down, the autoimmune diseases have gone up. So there are some spatial correlations that are suggestive. I'll show you some data that are tighter than this. But if you look across the planet, you can see that where diabetes, Type-1 diabetes is common. Type-1 is an autoimmune disease, okay? You see Type-1 diabetes being common basically in Europe and in Australia, and it's also fairly common in Saudi Arabia. And if you look at where worms and leprosy are common, where countries that have a fairly high incidence of these different worm infections, those are pretty much across the Tropics. The countries where there's no data basically, are in white. So this is a partial plot. And if you look at Type-1 diabetes against tuberculosis, you see where there's a lot of Type-1 diabetes there's not very much tuberculosis, and where there's a lot of tuberculosis there's not very much Type-1 diabetes. Okay? So that's a negative spatial correlation. There's more data than that. In Germany, and in other European countries, farm children have fewer allergies than city children. If you go to Gabon and you go around testing by just nicking people on their arms--which is a very easy test; you just apply a little bit of dust mite egg to somebody's arm and see whether they have a reaction-- the kids with schistosomiasis don't have so many allergies, and they don't have a reaction to dust mites. And if you look in these countries, adults with less asthma are more likely to be infected with nematodes. And just let me comment before I take that one off, that if you are a doctor working in the Tropics, you almost never see autoimmune disease. So if you go into Médecins Sans Frontières, and you go to Gabon, or you go to the Congo, you'll see a lot of infectious disease, and you will see a lot of worms, but you will not see very much autoimmune disease. That's the take-home message from this summary. Now how might this work? Well worms are big, multicellular parasites, and they have to live in our bodies a long time to reproduce successfully. When they send their eggs out, to get into another host, those eggs are going into an extremely risky environment, and it's not very likely that any single individual egg is going to make it. So the worms have evolved ways of living in our bodies, for a long time, without being knocked out by our immune systems. This has been going on for hundreds of millions of years. They're very good at it. They are interfering with signaling pathways that also happen to be the pathways that elicit allergies and asthma. Now think about it from our point of view. We got these worms in our system, and they got to be really good at living in our bodies for a long time, but we have an immune system that wants to react to them with a big inflammatory response, but it's not going to be able to get rid of them, because the worms have out-foxed us. So we have to make the best of a bad deal. What we have to do is we have to down-regulate our inflammatory response, in the presence of worms, so that we don't damage ourselves; because inflammatory responses turn out to be one of the most damaging parts of degenerative disease. That's what's going on in arteriosclerosis. That's what's going on in rheumatoid arthritis; you know, there's just a lot of inflammatory response, damage can happen to your body. So our immune system basically down-regulated, in the presence of worms. Now that means both sides of this co-evolutionary interaction have evolved. So the causes really are rather complex. The parasites have been removed, that actively down-regulate the immune response. That leaves inappropriate responses of our anti-worm machinery, and that anti-worm machinery lacks proper targets and is fooled by inappropriate targets. There is ongoing research right now to see whether or not this is in part the basis for nut allergies, which--things like peanut allergies-- which have really exploded. It appears to be possibly part of it, but probably not the whole story. And then, of course, we have changed our inflammatory response. And another interesting part of this--and again this is open research--imagine your body having come to evolutional equilibrium with worm infections. So the worms are down-regulating your immune system, and your immune system is just--it has a lot of other things to deal with besides worms, so it's cranking along, it's producing a range of cells that can react to different kinds of invaders. And it has a screening apparatus, which is in your spleen and in your thymus glands, to screen out any molecule or any population of cells that is recruited by your immune system to attack your own tissue. And it's screening along at that level. Then you pull the worms out. The immune system is no longer down-regulating because of the presence of worms; the immune system cranks up, and it throws a lot of stuff at that screening apparatus. But the screening apparatus didn't evolve to deal with that much stuff. So it's kind of leaky. So it is letting through more cells that might react with your own tissue. Okay? That's a hypothesis; that's not a demonstrated fact. But what I'm trying to do is I'm trying to indicate to you that this issue of autoimmune diseases arises logically, either at the points where the worms had been manipulating signaling in the immune system, and then that has been withdrawn, or it is operating on the screening mechanisms that are built in for the immune system; both could be going on. Now, what kind of data have we got? Well here--this is kind of small, but basically what you've got here is a knockout mouse that simulates Type-1 diabetes. Okay? So it's a model mouse; people have genetically constructed a model mouse, to make it like Type-1 diabetes in humans. And then they have infected it with various kinds of worms to see whether or not it is changing the T-cell bias in a way that would be plausible to basically down-regulate autoimmune disease. And these are things that prevent Type-1 diabetes in knockout mice. So Schistosoma will do it, Heligmosomoides will do it, Trichinella will do it. Mycobacterium--that's TB and TB's relatives--will do. Salmonella will do it. Basically infectious agents are antagonists of Type-1 diabetes in model mice. And if you ask a little bit more widely, if you have an animal model for another kind of a disease, what can we treat it with? Well we've got Schistosoma, we've got Trichinella, Trichuris and so forth. These things will prevent colitis, inflammatory bowel disease, collagen-inducted arthritis, Graves' thyroiditis, and so forth, in model systems. So there's some evidence in animal model systems that this works. So if you decided that you wanted to do therapy on humans, using these nasty worms, which have a big yuck factor, which one would you choose? Well you would want to have a worm that doesn't really cause much pathogenic problem itself in a human. You wouldn't want it to multiply in the human. You'd want to be able to regulate the dose. You wouldn't want the infection to get away from you, in treating a human. You wouldn't want it to be spread. You wouldn't want it to alter the behavior in patients that have depressed immunity. You wouldn't want to be affected by common medications like aspirin and stuff like that. Okay? Well, which one will do that? It turns out this pig whipworm has these characteristics. And what you can do is you can breed these things in the lab--I've seen them in Rick Maizels' lab in Edinburgh, growing in a little vial; they're whipping around in the little vial; they look like little pieces of thread--and basically you use their eggs. Now here's some data. Patients with Crohn's disease and ulcerative colitis improved after ingesting 2500 pig whipworm eggs. I mean, you guys all have issues with what they're serving you in the dining hall. > How about a little pig whipworm egg? People with Crohn's disease who got a fairly prolonged treatment with this stuff responded well. Patients with ulcerative colitis, in a double-blind, placebo controlled trial--which is another step up in rigor-- did better on worm eggs than they did on placebos. But this is the one that really gets me, and it's about multiple sclerosis. Okay? This is a very, very nasty disease, and multiple sclerosis is an autoimmune disease that attacks the sheaths on the axons in your brains, and it does so in a slightly different way in each individual. So the symptoms start developing in different ways, but basically what's happening is that you're losing your brain. And these are some of the symptoms: numbness, tingling, pins and needles, weakness, spasm, spasticity, cramps, pain, blindness, blurred vision, incontinence, urinary urgency, constipation, slurred speech, loss of sex function, loss of balance, nausea, disabling fatigue, depression, short-term memory problems. People with multiple sclerosis often go to Switzerland to commit suicide; I think about 60 of them have, because they're faced with something which is a very painful way for life to end. Well there was a case control study done recently in Argentina that showed that the progress of multiple sclerosis is a lot slower in the patients that are infected with parasitic worms. And that was convincing enough--this was a case control study; so for clinical medicine that's sort of the gold standard. You take a bunch of people and you match them with cases and controls, and then you see what happens differently in the two populations. So the data there was convincing enough to persuade the NIH to begin a clinical trial in Iowa in which MS patients are being treated with the eggs of pig whipworms. Now this is the data from Argentina, and the X--by the way, the four panels are four different ways of measuring the progress of multiple sclerosis, and all four panels have a five-year time axis on the X-axis, and then they have some measure of multiple sclerosis on the Y-axis. And in all four panels the uninfected patients-- they were matched at the start, infected and uninfected by worms, and at the same stage of multiple sclerosis-- the uninfected patients got worse, and the infected patients did not get worse. Very clear. When I first got in contact with evolutionary medicine, this hypothesis wasn't really out there yet, or wasn't very prominent. It came to my attention ten years ago. I didn't believe it at the time, and I'm actually rather astonished that this is the part of evolutionary medicine that is actually resulting in an important clinical result that could change treatment and save a lot of agony. I hadn't expected that. So humans evolve more slowly than their cultures, and therefore we are mismatched to modern life. This is important in both our diet and in our cleanliness and our hygiene. And it appears, certainly for the hygiene, and quite possibly for certainly people who are born very food stressed and then encounter junk food, that that causes serious medical problems. So one of the visions of evolutionary medicine is that we evolved to a diet and an ecology and a social life and a degree of cleanliness that was characteristic of a Pleistocene hunter-gatherer group, and that that's now changed radically and we haven't caught up yet; our bodies have not yet adjusted. The other thing that I want to tell you about basically is about how pathogens evolve. And they evolve very rapidly in response to things that we do to them, both to antibiotics and to vaccines. So the antibiotic resistance story is in large part a story about hospitals, because that's where most intense use of antibiotics is. Virulence also evolves, and there are lots of interesting stories about how virulence evolved. For example, plague in Europe, from 1348 to 1350, getting less virulent as it goes northward; or a new strain of syphilis coming into Europe from the New World and getting into Naples in about 1500 and preventing the French army from taking over Italy at that time, and then decreasing rapidly in virulence as it spread. There are lots of stories like that in history, and they're interesting. But the issue that confronts us today actually I think is most tightly focused on what vaccines will do, because we are now contemplating vaccines for a new kind of disease, not a childhood disease. We're not looking at vaccines that basically sterilize a population. We're looking at imperfect vaccines, and the issue is will they cause virulence to increase? So let's look at these. So a little bit about antibiotics first. Okay? Almost all of the bacterial genes that allow them to process the drugs that we use, and deal with those drugs, that provide them with resistance, evolved before the human drug industry existed. And that's because bacteria have been engaged in warfare, chemical warfare, with each other and with fungi, for hundreds of millions of years. And they are biochemical maestros. They have developed a large spectrum of synthetic capacity, and it's out there naturally in nature. There's about a ton of bacteria per acre in a cropland; that's about 10^(17th )bacteria. That's an enormous number. There's a lot of info that can be stored in 10^(17th) bacteria. Here's a little bit of data. Drug resistance evolves in the soil and in wild animals. So if you go out and just take out samples of spore-forming bacteria from soil, that's not near a hospital, that's out there, every single one of 480 strains of bacteria was multiply resistant, and there was no existing class of drug that was effective against all strains. That's just natural variation that's out there. Okay? That is the downside of biodiversity. There's a lot of evolutionary potential in natural bacteria. If you go around the outback, in Australia, and you sample enteric bacteria, that is gut bacteria, from various Australian mammals--you do this essentially by collecting feces-- what you find is that they have multiply resistant strains of bacteria; and they have never been close to a city, or to human beings that are taking antibiotics. So that's on the one hand; that's what's out there naturally. Now what are we doing to it? Well the agricultural use of antibiotics is quite important. I'm going to talk a bit about hospitals in a minute. But the reason that farmers use antibiotics is that by reducing the amount of energy that their pigs, cattle and chickens have to put into resisting disease, their pigs, cattle and chickens grow more rapidly. So it pays them. If they use antibiotics, they increase their production. So one antibiotic that's actually quite critical is vancomycin. Vancomycin has been the last line of defense against multiply resistant staphylococcus aureus for about twenty years. You don't want resistance to evolve to vancomycin. If it does evolve to vancomycin, it becomes very hard to do surgery in hospitals. Well Danish farmers were using vancomycin, and the Danish government noticed that and banned it. So we have a before/after comparison of how frequently do you pick up vancomycin resistant enterococci bacteria in Copenhagen, in the city? Well it dropped from 12% to 3%. There was a 9% rate drop in the rate at which doctors picked up vancomycin resistant bacteria in the city, when they stopped using it out there on the farms. That is a measure of how dirty the meat processing plants are, on the one hand. Okay? There's crap getting into the meat. There's a movie about that, by the way, about McDonald's; it will really turn your stomach. But it also indicates just how important the widespread agricultural use is. Now the other place where there's really a lot of antibiotic use is in the hospital. Okay? So the Center for Disease Control estimated-- I think this is in 2003--that there were 90,000 residents of the United States that went into the hospital for some other reason, picked up a resistant bacterium, and died of a bacterial infection that they didn't have when they went into the hospital. And when the cynical researchers checked the claims to the health insurance companies, they discovered that the actual number was probably ten times higher than that. So this is just for comparison. AIDS was killing 17,000 a year in the U.S. at the time; flu about 37,000; breast cancer about 40,000. So there are actually more people who were dying of bacterial infections that they acquired in hospitals than of all of these leading killers combined. Now the bacteria that live in hospitals are almost all either resistant or multiply resistant, because that's where so many antibiotics are used. And it's a good thing to use antibiotics in hospitals. Okay? When you bring somebody into the Emergency Room, or if they're in Intensive Care, and they are possibly just a few hours away from having to have an operation, you don't want them to be in a susceptible state. You want them to be clean, when they go into that operating theater. So you're going to use antibiotics on them to increase their probability of survival, if they have to have a major operation. But the consequence of that, which is of benefit for the individual, is a cost for the population. And resistant strains are much more expensive to cure. The cost of curing one case of TB, if it's not resistant, is about 15 to $20,000.00, and the cost of curing one case of multiply resistant tuberculosis is about a quarter of a million dollars. So it's about ten times higher. So the economic burden for the U.S. was about 80 billion annually, for resistance, and the economic burden for the planet is probably about a trillion. It's a big problem. So basically I'm going to just put--I'm not going to read all the way through this. Okay? Basically what this says is people move back and forth between hospitals and nursing homes, and when they move, they move the bacteria with them. And so however you're managing it in the hospital, you have to deal with a situation where it could be coming back in. And I can tell you that if you operate a nursing home, you're just deathly afraid that one of your patients in the nursing home is going to come up with a multiply resistant strain of bacterium, because in old people that can go through and just wipe them out. You'll get incurable pneumonia very quickly occurring. This idea--well let me just go back here. In this context of the ecology of hospitals and nursing homes, there's been some fairly sophisticated thought given to how should we manage the use of antibiotics. The kind of simple-minded way, which has often been used, is that well we'll just cycle the antibiotics. We'll use Antibiotic A for three weeks in the hospital, and then we'll replace it with Antibiotic B; and that way every time they start to evolve resistance to Antibiotic A, they get hit with Antibiotic B, and so forth. It turns out that produces a selection regime which is extremely effective at causing the rapid evolution of multiple resistance; happens again and again and again. Turns out the best way to really screw up the bacteria is to assign antibiotics at random, to individual patients within the hospital, and change them about every two days. Well that would drive the nursing staff crazy; I mean, that's just hard to manage. Right? But that's the most effective method. Well if we apply that to chemotherapy, what we notice when we look at the community of oncologists is that many of them aren't aware that a cancer is a genetically heterogeneous population of cells. I mean, the whole thing that gets a cancer going is an optimum mutation rate, and those cells continue to mutate; so they become quite genetically heterogeneous. It takes seven to nine mutations to turn a stably differentiated cell into a cancer cell, one after the other. And those cells then--and by the way, the mutations that do it are often mutations to the DNA repair apparatus. So cancer cells tend to have a pretty elevated mutation rate, and they become- a cancer becomes very genetically heterogeneous. So if you start prescribing one chemotherapy, and wait until it fails, and then start another one, you are applying a selection pressure that very effectively selects for resistance to chemotherapy. So if a more sophisticated strategy were used, it's been calculated that the lifespan of cancer patients might be prolonged by well several times; it all depends on the cancer. But say take something like breast cancer, instead of perhaps having a ten or twenty year potential survival, you might be able to manipulate the chemotherapy to have a thirty to forty year potential survival; which for many women would get it to their normal lifespan. So this is a place where evolutionary models can actually really help to better manage the use of antibiotics. Okay, virulence. Now I've used Ebola, *** and malaria to symbolize the three different stages in the evolution of virulence when a disease emerges and moves into the human population, and then starts to become adapted to it. So the first phase, which would be Ebola, Lyme disease, bird flu, SARS, rabies, it's accidental; it's an accidental infection. It's coming in from another species, it's not adapted to us yet, and sometimes these things are just incredibly virulent. By the way, they aren't always. We probably don't even notice the thousands that come into us and never take root and die off quickly, because they simply pass without having caused any major disease. But the point, the reason that some of them, some perhaps small proportion, are highly virulent is that they've never had any evolutionary experience in humans, and they're not adapted to the level of virulence that's best for them. They kill us too quick; they kill us so quickly they can't get out. Ebola is essentially a self-snuffing disease. It won't spread out of one village, because everybody's dead too quickly for it to transmit. Phase Two would be one in which the parasite's been established, but it's still far away from its optimal virulence. Okay? So this is probably the case with ***. The virulence of *** is probably still evolving. It's been in humans, we think, about seventy, eighty years, something like that. And the Myxoma virus that was used on rabbits in Australia. So it evolved its virulence downward in Australia, because it was killing rabbits too fast. Then in Phase Three you're dealing with a parasite that's well established, it's been in that host for a very long time. It's probably at its optimal level of virulence. Okay? So yes, it will kill some people, but it doesn't kill them too fast. It kills them at a rate where most of it can still get out and get into another individual, before the first host dies. And that's probably the case with malaria and tuberculosis. So let's take something which is in Phase Two, and put it to work. So here's where virulence evolution actually becomes part of a medical technology. Microbiologists have been using serial passage to produce attenuated vaccines for a long time. And what an attenuated vaccine is, is a pathogen that would cause a serious disease, but it's been evolutionarily changed, so that it's attenuated. It will infect you but it won't make you sick, and it will therefore elicit a very strong immune response, which is also effective against the unattenuated relatives. And that's been used to produce the Sabin oral polio vaccine; the measles, mumps, rubella, yellow fever and chickenpox vaccines; one flu vaccine; and a TB vaccine and a typhoid vaccine. So this is actually showing you that rapid evolution of virulence is a medical technology, and has been now for fifty years. The reason it works is that pathogens evolve rapidly. And the results demonstrate that there really are widespread tradeoffs in performance on different hosts. This tradeoff right here, that you do well on one host and poorly on another--that a jack-off-all trades is a master of none; the master of one doesn't do well on another--limits host range and constrains the emergence of new diseases. So these kinds of data, which basically were directly technically related to the production of vaccines, are indirectly telling us a lot about pathogen evolution and ecology. Here's the way it works. What you do is you get a nice genetically homogenous mouse, which is not going to be any kind of a genetic challenge to the parasite. So you give it a sitting duck--except it's a sitting mouse--and you inject it with parasite; parasite grows exponentially, and while it's still in exponential growth phase, you take some of it out. You remove its transmission costs. You take away any tradeoff it might have had with transmission. Okay? So it's going to become really bad at transmission, but boy does it get good at growing in this thing. You extract it, you re-inject it, and you let it go through exponential phase--you just keep it in exponential phase the whole time. You're killing mice like crazy. This is what happens. This is a passage through mice. This is the percentage of dead mice. This is salmonella. So you start it in a new host, and it gets more and more virulent in that host. As it specializes on its new host, it gets really good at growing in that host. This is what happens through passages in cell culture. And this is the number of monkeys being killed for polio virus. And this is actually Sabin's original data. Okay? So he's passaging polio through cell culture. So it's really good at living in cell culture. It's getting really lousy at living in monkeys, and the longer it lives in cell culture, the fewer monkeys it kills, until after 50 passages in cell culture it isn't deadly at all, in monkeys; and at that point they began a clinical trial and put it into humans. Okay, so the point of that--I mean, there are a number of points in that whole story about manipulating virulence. One is, virulence can evolve really quick. Virulence has been manipulated by medical technology, for the last fifty years, to produce some of the most successful vaccines on the planet. That itself is an impressive confirmation of this hypothesis, that I've just put up there, which is that in order to do really well on one host, you have to give up the ability to infect others. So if you want to produce a vaccine that's a live, attenuated vaccine, that infects a human, you take it out of the human, you put it into something else; you make it really good at killing that other thing; it becomes lousy at killing humans, and when it gets lousy enough at killing humans, you can use it as a live vaccine. Now there's one more thing I want to tell you about evolutionary medicine, and that's about whether virulence will evolve in response to vaccines. So I've already introduced you to the virulence transmission tradeoff. If you're too virulent, you won't transmit, because you will have killed your host before you can get out. Okay? This is supposed to be the most fundamental tradeoff shaping virulence evolution. It's thought to be widespread, and it really is thought to drive virulence to an intermediate level. There's quite a bit of evidence indicating that this is, broadly speaking, true. Okay? Now what happens when you make an imperfect vaccine? It does pretty well, but it doesn't kill all of the pathogens in all of the hosts. Okay? That's why we call it imperfect. Well that imperfect vaccine will reduce the cost of virulence by making likely that some hosts will survive in the presence of virulent strains. So you're getting a partial immune response. The pathogen can persist in the body, a longer period of time; because, after all, the vaccine is working a bit. But then if the virulent strains are the more competitive ones, and you've got multiple infection, then the virulent strains are the ones that are going to be surviving the longest in the bodies of people that have an imperfect response to the vaccine. Okay? So it turns out that this actually happens in mice with malaria; you can demonstrate with mouse malaria that this is the case. And the Gates Foundation and WHO would like to vaccinate 500 million humans against malaria. All of the malaria vaccines are imperfect; as a matter of fact, there isn't one that's really very good at all yet, but it looks like all the malaria vaccines will be imperfect. And that really creates an ethical or public health dilemma, which is rather similar to antibiotic resistance. It's going to be really good for the individual human being to be vaccinated against malaria. Hundreds of millions of lives would probably be saved. But, as an unfortunate byproduct of this wonderful thing, we are probably going to have a situation in which the surviving disease becomes more virulent, and a few people are then hit by a really nasty strain of malaria. So, as with antibiotic resistance, it's probably a good thing to know, that this might happen, so that you can start getting ready for it. It's not a recommendation that you don't vaccinate, it's a recommendation that you understand the consequences of vaccination, which are evolutionary, and be prepared to deal with them. So if you're interested in this particular thing, I've listed authors that you can search on. So the take-home on evolutionary medicine basically is that evolutionary thinking actually provides some interesting new illuminations of problems in both medical research and practice. But it certainly doesn't eliminate, or replace, all of that other important insight that we've gotten from molecular medicine, and basically from evidence-based scientific medicine, up to this point. There's just a tremendous amount in physiology and genetics and biochemistry which is absolutely essential to know. This, however, also is something that is important to know. Okay, see you tonight, if you're coming.