Evolution 2011: Darwinian medicine

June 20, 2011 • 8:12 am

The meetings so far have gone very smoothly; the organizers have done a terrific job (despite us having to live out in the middle of nowhere), and there have been few glitches.  What a great idea it was, too, to have a free happy hour from 5:30-7:30 every day after the last session, with all the drinks you can swallow and lots of people to talk to.

I want to report on one talk on evolutionary medicine.  If you’ve followed this website, you’ll know that I was once down on the practical uses of evolution: I thought of the discipline more as a way to understand the world than to improve it.  But I’ve changed my mind, largely at the instigation of Dave Hillis at the University of Texas at Austin, who has enlightened me about the real applications of evolution in medicine.

At the meeting yesterday there was an all-day symposium on “Darwinian medicine” (DM). This is the discipline that studies not only how the evolution of pathogens helps us understand disease (antibiotic resistance in bacteria is, of course, the classic example), but also how human evolution affects not only our susceptibility to disease, but explains some of our symptoms (fever, for example, may be an evolved adaptation to kill pathogens, and so you might want to hold off reducing mild fevers).

I didn’t go to all the talks, for there are many interesting talks to see on a given day, but I wanted to briefly summarize one talk on Darwinian medicine, by Randoph Nesse at the University of Michigan. Nesse, along with the late George Williams, has published extensively on Darwinian medicine and, in fact, is largely responsible for founding the discipline.

Nesse’s talk was called “What evolutionary biology and medicine offer to each other, and reflections on George Williams.”  He began simply by recounting some statistics: how few evolutionary biologists there are on medical school faculty: almost none, but that’s not much of a surprise. More surprising is how little evolutionary biology actually gets into med-school curricula, despite its importance for medicine.  Most schools teach things about antibiotic resistance, more or less because they have to, but other aspects of DM aren’t often taught in medical school: things like “adaptive” human symptoms of disease, or things that pathogens do to facilitate their own spread (the fact that malaria makes you prostrate, for instance, may actually be an adaptation of the malaria parasite to facilitate its spread; you’re more likely to be bitten by a mosquito, who transmits the parasite, if you’re laid out flat in bed).

Nesse gave six evolutionary reasons why humans are still susceptible to disease, despite the adaptive advantages of being resistant to disease:

1.  Evolutionary constraints.  We may be unable to evolve resistance to some diseases simply because to do so would entail a greater fitness cost than a fitness benefit, for there are constraints that prevent us from changing one thing without changing others, perhaps at a greater fitness cost.  Although he didn’t give any disease examples, Nesse did use the example of the human eye, which is less than optimally designed because of its evolutionary origin as an everted part of the brain. (Light has to pass through blood vessels and nerves to get to the photosensitive cells, and this reverse-wiring gives us a blind spot that isn’t present in the camera eye of the octopus, which has an independent evolutionary origin). If a reader has an example of a constraint that keeps us from evolving resistance to a disease, let me know.

2.  Mismatch between genes and environment.  Many modern genetically-based diseases may reflect genes that were neutral or even adaptive in our ancestors, but maladaptive in a modern environment to which we haven’t yet adapted.  Heart disease may result from genes that make us crave fats. Such genes may well have been adaptive in our ancestors but are now maladaptive in our fat-rich environment. Such diseases aren’t really purely “genetic diseases,” but reflect an interaction between genes and new environments.

Many “common” genes that cause disease, like those for hypertension or diabetes, may have been either neutral or adaptive in our ancestors, but are not so great to have in a modern environment. Do remember that we’ve only had about two thousand years of “modern” civilization in our 6 million years of living in small groups in Africa: modern life is thus about 0.03% of our total evolutionary history.

3.  Coevolution with pathogen transmission.  Pathogens have evolved in ways that make us susceptible to disease.  Nesse used the example of malaria that I gave above, but this isn’t really coevolution: it’s simply an adaptation in the sporozoan parasite that facilitates its own transmission.  In “coevolution,” two species each undergo evolutionary change in response to each other. I don’t really see any evolution in humans in response to malaria (except for the spread of the sickle-cell allele, which prevents malaria in heterozygotes, but that’s not what Nesse was talking about), so the point he was making here is unclear. If someone heard that talk and can explain why “coevolution” is something that explains our evolutionary susceptilibility to disease, please explain below.

4.  Tradeoffs.  Bilirubin is one product of the metabolic breakdown of hemoglobin. It’s a very toxic compound, and if you have too much of it you get jaundice (it’s responsible for not only the yellow color of jaundice, but also of urine and the yellowish tint around bruises).  About 5% of the population has elevated bilirubin, which causes Gilbert’s syndrome, which in some people has mild deleterious effects but is often asymptomatic.  Why do we get this disease? Because bilirubin, though toxic, is also beneficial in some ways: it acts as a potent antioxidant and thus can prevent the damaging effects of too much oxygen on cells, which can include heart disease.

As Nesse pointed out, people with Gilbert’s syndrome have a significantly reduced incidence of coronary artery disease because of the elevated bilirubin.  His point was that although bilirubin is a toxic compound that can cause mild (or asymptomatic) disease in a substantial number of people, that disease is a byproduct of a greater good—a tradeoff between producing a toxic compound and the net beneficial effects that that compound has on the body.

5.  Reproduction trumps health.  The currency of natural selection is not longevity or bodily well-being, but reproduction.  Nesse argued that, for example, men get sick and die more often than women simply because there’s a tradeoff between their evolved tendency to compete with other men for women, thereby spreading their genes, and the fact that this competition wears men out physiologically, making them die earlier. He cited one surprising statistic: for every 100 Oklahoma women who die at age 20, there are 300 men who die at that same age. And we all know that the tour buses of old folks you see in your city are largely women, for many of the men who would be their contemporaries died long ago.

6.  Some “disease” symptom and defenses against mortality are useful, even if costly.  Pain, fever, and vomiting may be signs of sickness, but can be useful in preventing or mitigating illness.  Pain, of course, is a way of alerting you that something is wrong: people without the ability to feel pain often suffer infected wounds and other injuries, simply because they don’t get the signal that there’s something wrong. (I think this is why sufferers of leprosy—now called Hansen’s disease—often lose noses, ears, and fingers.) Vomiting helps us purge our bodies of toxic substances.

Similarly, many of our anxieties may also reflect adaptation.  As Nesse pointed out, if our ancestors got anxious and ran away from a noise that could be a leopard, maybe 99 times out of 100 it would be nothing, and that energy and anxiety would be wasted, but it’s that one time in 100 when it’s really a leopard that the anxiety and flight reaction really pay off. It will often be useful to be fearful and anxious even without much cause, if those reactions can save your life on the rare occasion when something really is around to harm you.

It was interesting to contemplate these issues, which of course are only one part of DM, for they concern only susceptibility to disease in humans, not other aspects of human disease or evolution in the pathogens themselves.

I have only one small quibble about Nesse’s talk. He gave one example of what he considered an untenable “just-so story” about human disease: the speculation that juvenile (type I) diabetes was an adaptation for the ice ages 10,000-20,000 years ago.  Those individuals with high levels of blood glucose, so the story goes, were better able to avoid freezing to death, and so the genes producing higher glucose, i.e., those that now give us juvenile diabetes, were adaptive in our recent evolutionary past.

This does seem to be a bit of a just-so story, and Nesse is right to be wary of it. For one thing, juvenile diabetes has severe side effects that can kill you when you’re young, before you can reproduce.  The curious thing is, though, that Nesse included clinical depression on his list of one of the “diseases” that probably were adaptive in our ancestors.  Yet the evidence that depression is adaptive is even less convincing than for type I diabetes  (see my posts about this herehere, and here).

While I now think that Darwinian medicine is a useful and intriguing discipline, its practitioners must be careful not to fall into the same trap that’s snared many evolutionary psychologists: uncritical and untestable storytelling.  So far, many advocates of DM, including Nesse, seem to be largely avoiding that trap.

22 thoughts on “Evolution 2011: Darwinian medicine

  1. I think the application of evolutionary principles to the treatment of HIV has been a major success story in our time. The use of combination antiviral therapy means that the virus finds it much more difficult to mutate to a resistant version. This has resulted in the disease becoming a chronic and treatable condition rather than a death sentence.
    What’s more, this particular success is highlighting a way that cancer – the other disease that relies on evolution to evade current drugs – may be tackled. The use of novel combination therapies in cancer will be the thing that keeps many of us and our relatives alive in the coming decades if we end up with the sorts of cancer that are currently only treatable with limited effect.

      1. In two ways. First, 1-15% of the population (the proportion varies geographically) is resistant to AIDS due to the presence of a naturally occurring genetic variation. Second, and this is what I think Sigmund had in mind, combination drug therapies can sufficiently inhibit HIV so that in a substantial fraction of patients HIV can be managed as a chronic infection with high long term survival. Magic Johnson and Andrew Sullivan are two well known examples of the latter course of infection.

        1. Correct. If you have change a disease from one that takes a couple of years to kill you into one that takes several decades it not only extends the lifespan of the sufferer but it provides a window of opportunity for new treatments to be uncovereds that will add to this lifespan. A disease that IS a ‘death sentence’ is one that cannot be managed in this way and which will remove decades from your expected lifespan. A disease that can be managed for decades up until the patient reaches old age is not in this category and it is the development of new treatments for HIV that have led to this wonderful result.
          That is not to say that HIV is non trivial or that the treatments are available everywhere – in many parts of the world (sub Saharan Africa for instance) HIV is still a death sentence.

  2. An interesting post indeed, but…

    Some of these ‘just-so’ stories do seem a bit naive. To consider Type 1 Diabetes as having any adaptive advantages is just plain silly. Hyperglycemia does not help you live longer under cold stress – it just kills you prematurely of ketoacidosis. (The pathogenesis of Type 1 DM may be secondary to viral-induced islet cell death. Genetic susceptibility notwithstanding, in the days before exogenous insulin therapy, it was uniformly fatal and cannot regarded as having any evolutionary advantage.

  3. things that pathogens do to facilitate their own spread

    The cold virus primarily spreads through sneezing. The prime symptom of a cold infection is sneezing.

    If a reader has an example of a constraint that keeps us from evolving resistance to a disease, let me know.

    I’m not aware of any pathogens that could survive immersion in bleach. On the other hand, I can’t imagine any evolutionary pathway that could develop to use bleach against pathogens.

    Thanks for the report! Sounds like you’re having a blast, and I wish I could be there.



    1. Bleach is an oxidizer. Other compounds produced by animal cells are also oxidizers, and could potentially be deployed against pathogens. The constraint here, I think, would be the large risk of damage to other tissues by unleashing such chemicals within an animals body.

      I saw Dr. Nesse give a presentation about 5 or 6 years ago, and I wasn’t very impressed, to be honest. He made some excellent points about the benefits of some evolutionary biology approaches to understanding and treating diseases, but so much of his talk was hand-waving just-so-stories and so little was hard data that I came away feeling a bit disappointed. I’m hopeful that in the intervening years he’s been able to accumulate more evidence than that one old study on lizards.

  4. The evolutionary constraint on cervical vertebrae number in mammals may have pathological implications. Almost all mammals (exceptions-manatees and sloths) have 7 cervical vertebra, while this number is very variable in birds. In humans, having small ribs (partial thoracic character) on the 7th cervical vertebra, and thus a departure from the canonical anatomy, is observed in about 30% of miscarriages, but only about 1% of adults.

  5. If a reader has an example of a constraint that keeps us from evolving resistance to a disease, let me know.

    First, I’m not an evolutionary biologist or a physician. But what pops into my head first is that it may be difficult for the body to differentiate between helpful and harmful colonizers, and that, perhaps, there is less harm getting sick occasionally than there would be in eliminating helpful organisms. So a conservative immune response might be favored to allow the helpful colonizers (like intestinal flora) to persist.

    Taking it a step farther, maybe the risk of autoimmune diseases–in which “self” cells are attacked as foreign objects–has also resulted in a conservative immune system. Lupus and type-I diabetes are deadlier than many pathogens, after all.

    Of course, this all assumes that an immune system is the only method of evolving resistance to a disease.

  6. If a reader has an example of a constraint that keeps us from evolving resistance to a disease, let me know.

    I think you may have answered your own question when talking about fevers. Assuming that fevers are an adaptation that helps to kill pathogens, a constraint right there is how hot we are able to actually run. Surely, if we could safely elevate our body temperature to 212F in order to kill pathogens, we would never have a pathogen-borne illness get out of control, because we’d just “boil it away”. But we can only get so hot without screwing up our own body chemistry in addition to that of the pathogen.

  7. re 3.: “If someone heard that talk and can explain why “coevolution” is something that explains our evolutionary susceptilibility to disease, please explain below.” (could someone tell me how to make blockquotes? 😛 )

    I haven’t heard the talk but it sounds like he talks about the Red Queen Hypothesis or host/parasite arms races. I think the term Red Queen was coined here: http://rspb.royalsocietypublishing.org/content/205/1161/489.short

    I don’t know what his particular example was regarding malaria, but Plasmodium does have mechanisms that actively influence/help avoid the immune systems of vertebrate and mosquito hosts. That would be one example how parasites evolve to overcome host defences.

    If he used the whole malaria/lieing down thing I’m not sure how that is a good example either 😀
    There is a bit of discussion going on among parasitologists about parasite-mediated changes that supposedly help in transmission (macroparasites have some striking examples there) – namely about which of them are adaptive and which are simply consequences of pathology. http://www.sciencedirect.com/science/article/pii/S1471492210000620 (sorry, paywall)

    1. <blockquote>Silly christian comment here</blockquote>

      To look like this after posting:

      Silly christian comment here

  8. For a variety of reasons I’ve gotten to this post late so I am afraid that my comment and question will go unnoticed.

    I was interested in the comment about the differential age-sex mortality rate in Oklahoma. While I think it is pretty well known that there is a big difference between male and female mortality rates at that age, the speaker’s framing was forceful. As I understand it, that difference is pretty much true for the whole of the United States and it is my understanding that accidents and homicides account for the difference.

    My question is, “What does this mortality difference have to do with evolution?”

    The answer that springs to mind is that males of that age engage in reckless behavior in order to attract females for breeding purposes–hence the point that reproduction is more important than health. Is that the correct explanation?

    Further, I am interested in cultural comparisions of that mortality rate. What is the rate for the Yanomamo where sexual dimorphism is important? For traditional rice cultures as in southern India of Thailand where sexual dimorphism is of less importance? Perhaps differences show up in the US between, say, urban and agrarian.

  9. Would you consider the lethality of homozygous for the HS form of hemoglobin a constraint which prevents resistance to malaria becoming fixed in a population an example of the type you requested?

  10. Thanks for  letting people know about this area of work. My talk at SSE had two theses. The first was that George Williams was one of the great biologists of the 20 th century and his passing is a great loss. The second was that evolution has a lot to offer to medicine; the approaches George and I suggested are only parts of a much larger enterprise. See “The Great Opportunity'” article with Steve Stearns, and the Jan 2010 special issue of PNAS on evolutionary medicine for reviews of the field, and my recent article on “Ten questions for evolutionary studies of disease vulnerability” for strategies to avoid errors. 

    NESCent has just funded an initiative to develop teaching resources, and Mt Desert Island Biological Labs will offer a week long course in August, so things are coming along. 

  11. > “While I now think that Darwinian medicine is a useful and intriguing discipline, its practitioners must be careful not to fall into the same trap that’s snared many evolutionary psychologists: uncritical and untestable storytelling. So far, many advocates of DM, including Nesse, seem to be largely avoiding that trap.”

    No, I think he still falls into the trap. Over and over again. However, by declaring that he’s on the lookout for these traps, he can feel confident that he hasn’t fallen into them. But he has. Not so much when he states his theories about bacterial-level changes, but on the bigger levels.

Leave a Reply