Earlier this summer I read a book called Evolutionary Medicine and Health: New Perspectives, which was published in 2008 by Oxford University Press. I found it both interesting and enjoyable. It’s definitely one of the best books on Darwinian/evolutionary medicine I’ve read. I didn’t find all of the chapters equally captivating though. One chapter in particular stood out to me as being more interesting than the rest. The chapter I’m talking about is about developmental plasticity and the Developmental Origins of Health and Disease (DOHaD) and is written by Christopher Kuzawa (PhD), a professor working in the Department of Anthropology at Northwestern University.
Perhaps needless to say, I had heard and read about developmental/epigenetic plasticity prior to the time when I came across the chapter by Kuzawa. However, I’d never come across anything on this topic that I found to be as well-written and intriguing as that chapter. I was blown away by Kuzawa’s writing, and that’s saying something, seeing as I’m not easily swept off my feet.
The thing that separated chapter 18 – the chapter written by Kuzawa – of Evolutionary Medicine and Health: New Perspectives from other writings on developmental plasticity I had come across in the past is that it provided a more logical and complete explanation of the DOHaD concept. Unlike many other scientists who write about epigenetics and developmental plasticity, Kuzawa appears to be knowledgeable about the evolutionary mismatch concept and many other concepts and theories that are integral to Darwinian/evolutionary medicine. This ensures that his DOHaD writings are robust and evolutionary sound.
In chapter 18 of Evolutionary Medicine and Health: New Perspectives, he moves beyond basic theories regarding how mismatches between ancient human genes and modern environments trigger disease and ill-health. He argues that in order to really make sense of how organisms evolve and adapt to their environments, we can’t simply look at how their genes change over time as a result of natural selection, mutations, etc., we also have to look into how epigenetic characteristics are passed down generations and the role environmental stimuli play in shaping the developmental phase of organisms’ life cycle. I concur.
As I was reading through the chapter Kuzawa penned for Evolutionary Medicine and Health: New Perspectives, I got the idea of doing an interview with him that I could publish here on the site. I had several nagging questions I wanted to ask him. Moreover, I felt that his ideas deserved more attention than they were currently getting. So, I got in contact with Dr. Kuzawa, who told me he was happy to do an interview. I then proceeded to devour some of his most recent papers on DOHaD and epigenetics, which can be found on PubMed, before I penned a list of questions for him.
Here’s the end-product of this process. I hope you like the interview and learn a thing or two.
1. Who are you? What’s your profession and educational background?
I am a biological anthropologist with training in public health (epidemiology). I received both my PhD in Anthropology and my MSPH in Epidemiology from Emory University, in Atlanta, in 2001.
2. How did you get interested in evolutionary biology and Darwinian/evolutionary medicine?
My home discipline of biological anthropology shares much in common with medicine and public health, because we study issues like human growth and development, endocrinology, immunology, and the emergence of chronic disease in relation to culture and lifestyle change. Where anthropology differs from the applied health fields is in focus – because biological anthropology is a field built upon a foundation of evolutionary biology, we tend to study human biology and health through the lens of evolutionary theory. So in many ways our discipline is the home of evolutionary medicine, and some of the earliest examples come from our field. Frank Livingstone, who was instrumental in linking human farming to malaria and sickle cell in West Africa, was an anthropologist. James Neel, the geneticist who proposed the thrifty genotype hypothesis, was an anthropologist. Mel Konner and colleagues, who proposed the “Paleolithic prescription”, were mostly anthropologists. So in our field there has been a natural marriage between evolutionary principles and the study of health.
3. What’s the main focus of your research?
I am broadly interested in using principles from developmental and evolutionary biology to gain insights into the causes of major public health issues like cardiovascular disease. For much of my work I collaborate with an incredible group of researchers in Cebu, the Philippines, along with collaborators in the US, to study the long-term impacts of one’s early life experiences. In addition, I have interests in other topics, including the evolution of the human brain, the hormonal changes that help human males transition into fathering roles and the social origins of health disparities related to race and class.
4. The expression of our genes change constantly in response to environmental inputs (e.g., food, toxins). There are a lot of different definitions of “epigenetics” floating around. On one end of the spectrum, there are those who seem to consider everything that has to do with gene expression to fall under the epigenetics umbrella, whereas on the other side, you’ll find some people who perceive epigenetics as strictly involving heritable changes in gene expression. How do you define epigenetics? What do you feel is the best and most useful definition of this term?
These semantic issues are important and can lead to a lot of confusion. To me, “epigenetic” refers to chemical changes in chromosomes that alter the shape and winding of chromatin fibers, which in turn have the potential to impact patterns of gene expression. Epigenetic changes do not alter the DNA itself, so these are best understood as a second layer of biological memory. Some of these changes may be stable and perpetuated to both daughter cells after a cell divides, showing that there is mitotic stability. I would not describe these stable cellular-level patterns as “inheritance” to avoid confusion with the more widely used meaning of the word as inter-generational transmission. It is important to underscore the word “potential” – epigenetic changes, such as alterations in methylation in the vicinity of transcription factors, have the potential to influence gene transcription, but this is by no means certain. The study of gene transcripts shows this quite clearly. All of this said, most of my work focuses on what geneticists would call “phenotypes” – meaning flesh and blood traits like hormone levels, growth patterns, or blood pressure levels. We relate these and other outcomes back to an individual’s experiences, including early life variables that reflect factors like how well-nourished one’s mother was while pregnant with them, their birth weight, or their exposure to or symptoms of infectious disease during infancy. We think that epigenetic changes are part of the system of “memory” that helps link these early experiences with biology and health later in life. However, our group has only begun to directly explore these mechanisms, primarily in the form of measures of methylation levels measured in blood or placental samples.
5. As you point out in your papers, one shouldn’t automatically assume that developmental phenotypic plasticity is a process that involves adaption to predicted future environments. E.g., the reason why children who are born to undernourished mothers are born small may simply be that they don’t get the nutrients they need to grow big bodies. With that said, in your papers, you bring up a lot of evidence suggesting that at least some organisms have evolved ways of regulating the growth and development of various bodily systems early in life in such a way that these systems “match” predicted future environments (e.g., a child that receives nutritional cues from its mother which indicate that the environment in which it will live is energy-rich, may adjust its metabolic systems to match such an environment). But, the most convincing evidence in this area comes from animal studies. Is it proven beyond a doubt that humans also adapt to perceived future environments via these mechanisms? If so, could you briefly mention how we know this to be true?
The idea that human use cues from the mother to anticipate features of the postnatal environment is certainly still a hypothesis. There has been no definitive demonstration that this principle applies to us, and the literature on this topic generally remains quite speculative. That said, there are certain observations that are intriguing in my view, and that at least hold open the possibility that something along these lines might operate in us. My thinking on this has been heavily influenced by evidence that a wide array of biological systems and health outcomes are influenced by the nutrition that we receive as a fetus, such as that restricting fetal nutrition can increase one’s future risk for diseases like hypertension, diabetes and heart attacks. This shows that one’s own early life experience of nutrition influences one’s future risk for nutritional diseases. Now of course one may rightly question whether these long-term effects are simply deleterious side effects of being under nourished in utero. Might they simply reflect a form of developmental “damage” or “impairment”? That might hold for some outcomes. For instance, smaller birth weight babies tend to have smaller kidneys with fewer nephrons, which leaves them at higher risk for hypertension and kidney disease later in life. This certainly could be a simple example of “impairment”.
But there are several factors that complicate this story. For one, other biological changes initiated by prenatal undernutrition are less easy to explain away as impairment. For instance, lower birth weight individuals tend to have a reduced risk of gaining excess weight and becoming obese. They have higher metabolic disease risk, in part, because they tend to preferentially deposit any excess weight in central, visceral fat depots (a so-called “apple” shaped body type), rather than in peripheral or subcutaneous fat depots. This is important because the fat in our bellies, that surrounds our organs, is innervated with sympathetic nerve fibers. When we experience stress those nerve fibers release adrenaline directly into those fat cells, which then mobilize the fats into the circulation to provide the body with a quick source of energy. What is fascinating is that lower birth weight individuals are not only predisposed to deposit fat in this metabolically active depot, but their fat cells are also on more of a hair trigger and mobilize fat more rapidly in response to adrenaline. It is difficult for me to see this pattern of changes as simple impairment – after all, no organ has had its growth impaired as was the case with the kidneys. Instead, it looks like the body is adjusting its biological settings and priorities in response to these prenatal nutritional cues – preferentially building up an energy store that is useful during stressors, while also increasing the response of those fat cells to stress.
What we know about the causes of the maternal supply of nutrients to the fetus, which is ultimately what shapes “fetal nutrition”, also leads me to speculate that some of these early life responses made by the fetus, and after birth by the breastfeeding infant, could serve an adaptive role. Given evidence that fetal nutrition has a range of long-term biological impacts – including the changes in fat metabolism and stress response described above – a natural next question is what shapes the flow of nutrients from the mother to the fetus. A naïve approach would assume that changing the mother’s diet during pregnancy would lead to corresponding changes in the nutrients that the fetus receives. But we should not expect evolution to have built such a crucial system to be so vulnerable. Instead, the fetus is embedded within the mother’s body and its nutrient supply is buffered and governed by her own homeostatic metabolic physiology. Glucose is a key nutrient for fetal development. After the mother eats, her glucose goes up and is delivered to the fetus. But once she begins her post-meal fast, her glucose will drop to a point, but not go below that. Instead, glycogen stores are mobilized to maintain glucose within normal operating limits. After glycogen is used up, stored fats, and later, proteins, are used to maintain fasting glucose. The point being that glucose supply to the fetus is not tightly linked to the mother’s diet, because her body has multiple redundant systems capable of maintaining a stable glucose supply during even prolonged fasts. In light of this, it should not come as a surprise that nutritional interventions targeting pregnant women typically yield modest or even negligible changes in a baby’s birth weight.
If fetal nutrition – and thus all of the down-stream biological and health changes – are a product of the mother’s metabolism, rather than her concurrent diet, then this leads to the question of what, if not her diet, might shape the mother’s delivery of nutrients to the fetus. We have evidence that her nutrition prior to pregnancy – around the time of conception, across adulthood, and even during her own early development, are important. As an example, a study in Guatemala supplemented the diets of growing kids. The researchers found that the kids who received the highest quality supplement grew better and ended up as larger adults, compared to the individuals who received the less favorable supplement. Independent of this effect on the mother’s size, their babies were about 4 ounces heavier – an effect much larger than the impact of most large nutritional supplementations provided during pregnancy. This and other studies suggest that one way to improve the fetal nutrition of future generations, potentially putting them on a more favorable health trajectory, is to invest in the healthy development of future parents (although the evidence for this is strongest for females, who will become future mothers, there is also fascinating, although still quite contentious, evidence that a male’s life experiences might also be passed on to influence offspring biology, operating through epigenetic changes transmitted in sperm).
Putting all of these pieces together: we know that fetal nutrition influences biological settings in the offspring that seem functional rather than simply pathological. And we see that the mother’s delivery of nutrients to the fetus, which drives those processes, is likely anchored to something like a cumulative impact of her nutrition across development. This latter point is particularly important in my mind, because it suggests that there could be a relatively stable signal conveyed across generations, reflecting a running average of her experiences. This could increase the reliability of nutrition as a cue of typical nutritional conditions in the local environment.
This latter process I call intergenerational “phenotype inertia” – it is an intergenerational signal that is based in the phenotype (e.g. nutrients or hormones) rather than genes. And it shifts in response to the environment, but not in a rapidly-responsive fashion. Instead, it appears to adjust more slowly to chronic conditions experienced across the lifecycle. To me, this hints at a system of communication that could allow offspring to adjust development in response to reliable maternal cues reflecting her typical experiences of the local ecology.
6. You point out in your paper that both high and low birthweight are associated with elevated type-2 diabetes risk. Why?
Studies find that risk for diabetes is reduced as birth weight increases – but only up to a point. At the heaviest birth weights, risk increases again. Thus, the relationship is U-shaped, with highest risk found among both the lowest and highest birth weight individuals. This seems to reflect two quite distinct pathways. At the low end of the birth weight distribution are the low nutrition processes that I discussed above. The high end of the birth weight distribution reflects a situation in which mothers with uncontrolled high glucose levels overfeed their baby. This leads to increased fat deposition prior to birth, increasing birth weight, but also modifies their metabolism in a way that predisposes to higher risk for obesity and diabetes later in life. The pathways are quite distinct.
7. More and more people are becoming obese. Moreover, a lot of people find it difficult to lose weight. Could this partly be explained by epigenetic developmental processes? Would you expect that a child who’s exposed to an energy-rich environment in utero (as well as receives signals from the mother which indicate that the environment in which it will live is energy-rich) and early in life will find it more difficult to lose weight than a child that is not exposed to such an environment early in life, given that it may adjust bodily systems and adipose cell count to “match” an energy-rich environment? Are we epigenetically adapting to our novel conditions by growing bigger? If so, is this making it more difficult for us to lose weight?
This is an interesting question, and one that we can only speculate on. In general, when considering obesity it is important to keep in mind that this is, generally speaking, a state characterized by dysregulation of biological systems. Our bodies maintain stability through time through the use of homeostatic systems that operate through negative feedback. For instance, we typically maintain a (relatively) stable body weight because our appetite increases after we restrict our intake and lose weight, while it decreases if we’ve over-eaten and managed to put on extra weight. A condition like obesity involves a modest break down of this type of self-regulating, stable system, so we are dealing with dysregulation.
Of course it is challenging to reconstruct what adaptive function a particular biological system may have evolved to serve, if any. In this case, I would be inclined to think that the effects of being exposed to high maternal glucose, and thus being predisposed to gaining excess weight as a result, does not likely have adaptive roots. There are several factors that support this interpretation. First, the experience of being in utero and exposed to very high maternal glucose levels was I assume a relatively rare occurrence in most ancestral hominid populations. Our distant ancestors were nomadic foragers who were extremely active and subsisted off of foods that could be gathered or hunted. I think it’s safe to assume that, in the absence of a rare genetic variant, overweight or obesity were likely rare in these settings. As such, gestational diabetes, or unusually elevated glucose during pregnancy, was also likely very rare. What this means is that there wouldn’t have been much of an opportunity for natural selection to shape the nature of this response, nor would there have been an incentive to.
The second reason that I think this interpretation unlikely is the nature of the phenotype itself. As an exercise in “reverse engineering” I would expect signals of an abundant, rich environment to allow an organism to not waste its calories by building up and carrying around large energy reserves. Large fat stores are a burden unless they will be needed. I can see an increased imperative to build up such reserves under scarce conditions rather than abundant conditions.
8. I’m very interested to hear how the developmental theories of adult health and function fit in with Darwin’s theory of natural selection. As I see it, natural selection is what “drove” the evolution of the adaptive epigenetic mechanisms discussed so far. Is this how you see things as well?
As I outline in a few examples above, I see some of the developmental responses made by the fetus as having characteristics consistent with a possible adaptive function. That implicitly presumes that natural selection has shaped aspects of the mother’s signal, the fetal response, or both. It is important to keep in mind that maternal passage of nutrients and hormones is already happening, as a necessary and unavoidable driver of development. What is at question is whether those cues embody useful, reliable information about the local ecology, and if so, what options might have been available for the fetus to modify its biology and development in response to them. So long as there is reliability in the cues, it seems likely that natural selection would have harnessed them, especially as they are already the key inputs driving variation in growth and development.
9. What do you consider to be the most important things Darwinian/evolutionary medicine has to contribute to our current medical system?
There is a growing awareness that evolutionary principles need to assume a more central role in medical education, clinical practice, public health and policy. Some examples are well established already, such as the focus on evolutionary dynamics in the emergence of antibiotic resistance or cancer development. In my view, there is a particular need to incorporate evolutionary principles more seamlessly into basic research and bench science. It is in these settings that evolutionary theory can complement a focus on mechanism, and help researchers avoid “trial and error” approaches. Understandings of theoretical principles and an appreciation for the value of comparative biology can lead to testable hypotheses that are more targeted and thus can act as short cuts. As a simple example, evolutionary biologists tend to view energy use in the body as governed by trade-offs, because organisms are faced with allocating a finite metabolic expenditure across the body’s various competing needs. As such, some hormones that stimulate one function may suppress others in tandem to keep things in balance. The principle of trade-offs is crucial to understanding this, but is not, as far as I am aware, a core principle in biological research, akin to e.g. homeostasis. In addition, it is almost always illuminating to view a biological system and its variation across a wide swath of distantly related species. This can reveal how the trait varies across species, how it relates to other aspects of biology, and which species characteristics seem to influence or drive this variation. This type broad, comparative approach is fundamentally complementary to the mechanistic focus of much current research. Together, they are a powerful duo, and yet, I see very little emphasis on placing biological systems within this broader comparative framework.
10. What would you do if you were put in charge of improving public health? What would you focus on?
I think that an understanding of the evolutionary origins of our biology points to reducing inequality as the biggest priority that we face as a society. An understanding of the responsiveness of our biological systems, whether through developmental or epigenetic change, or through wear and tear across our adult lives, makes it abundantly clear that stressors, undernutrition and other factors related to poverty have toxic effects on individuals and on society as a whole. In my lifetime, and especially under the current regime in Washington, there has been a concerted promotion of policies that will make these problems worse, rather than alleviating them. It is distressing, and a true shame. But that is obviously not a problem to be tackled by public health, but that is the best answer that I have to this question.