#  Video: Making Milk: Mongolia’s Unique Role in Dairy’s History 

 



Embed



 



Lecture by Christina Warinner, Professor of Anthropology and Human Evolutionary Biology, Harvard University

Milk is both ancient and enigmatic. First transformed into dairy products over 9,000 years ago in the Near East, its production required the domestication of not only animals, but also microorganisms. Dairy technologies spread across Europe, Africa, and Asia, reaching as far as Mongolia 5,000 years ago. Today, dairy products are produced and consumed worldwide; annual global milk and dairy production exceeds 900 million tons. And yet, the majority of the world’s population is estimated to be lactose intolerant. How did such an unlikely and often indigestible food become a staple of global cuisines? Christina Warinner examines the long and often surprising history of milk in Mongolia, where more than ninety percent of the population should be lactose intolerant—but is not. This talk takes a fresh look at the history of milk in Asia and its unexpected ethnographic and archaeological paradoxes. Far from familiar, milk is an ancient food with a modern scientific mystery at its heart.

About the speaker: Christina Warinner is Professor of Anthropology and Human Evolutionary Biology at Harvard University, and she leads international research groups at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany and the Leibniz Institute for Natural Products Research and Infection Biology in Jena, Germany. Warinner specializes in the analysis of ancient DNA and proteins, and her research focuses on the study of ancient biomolecules to better understand past human diet, health, and the evolution of the human microbiome. She has conducted groundbreaking studies on the evolution and changing ecology of the human oral microbiome, including reconstructing the oldest microbiome to date from a 100,000-year-old Neanderthal, and she has published extensively on prehistoric migrations, the origins and spread of dairy pastoralism, and the biodiversity of the human gut microbiome. She has published two books and more than 70 peer-reviewed articles in journals such as Nature, Science, Cell, and PNAS (Proceedings of the National Academy of Sciences). Warinner is the recipient of the American Anthropological Association’s Exemplary Cross-Fields Award, the Federation of European Microbiological Societies Article Award, and the Shanghai Archaeological Forum Research Award. Her ancient microbiome findings were named among the top 100 scientific discoveries of 2014 by Discover Magazine, and her research on medieval women artists was named one of the top 10 archaeological discoveries of 2019 by Archaeology Magazine. She was named one of the Top 10 Scientists Ready to Transform Their Field in 2017 by Science News, and her research has been featured in more than a hundred news articles and programs. Warinner is passionate about public education and outreach, and she designed the Dairy Cultures exhibit at the Natural History Museum of Mongolia, and she has been featured in documentaries produced by PBS NOVA, Netflix, and the genome sequencing company Illumina. She created the Adventures in Archaeological Science coloring book, now available in more than sixty languages, including many Indigenous and underrepresented languages. She is engaged in the open science movement, and her research group has been actively involved in improving scholarly communication, data sharing, and research transparency.

*Recorded date: March 6, 2025*

## Transcript

**Making Milk: Mongolia’s Unique Role in Dairy’s History**

\[00:00:07.76\] Good evening. My name is Caroline Jean Fernald. I'm the Executive Director of Harvard Museums of Science and Culture, a partnership of four public-serving museums where our mission is to foster curiosity and a spirit of discovery in visitors of all ages, enhancing public understanding of and appreciation for the natural world, science, and human cultures.

\[00:00:30.96\] With this mission in mind, I am delighted to welcome our in-person, as well as our Zoom audience, to tonight's lecture co-sponsored by the Peabody Museum, the Harvard Museums of Science and Culture, Harvard's Initiative for the Science of the Human Past, the Department of Anthropology, and the Department of Human Evolutionary Biology.

\[00:00:50.04\] We are honored to have Professor Christina Warinner with us tonight, who will speak about the long and fascinating history of milk in Mongolia. After the lecture, this is a very important thing to listen to for our in-person audience, we invite you to join us in the galleries of the Peabody Museum for a reception. Our staff--

\[00:01:08.37\] \[LAUGHTER\]

\[00:01:09.51\] Good timing, right? For your appetizers before dinner. Our staff and volunteers will help guide you to the galleries. HMSC's partnership of museums crosses many disciplines, and we host a wide array of events throughout the year at our four museums. On March 13th, we are partnering with Harvard's Brain Science Initiative to present an exciting program that will explore some of the biggest mysteries of the human brain.

\[00:01:32.89\] On March 26, we will host the lecture Murder, Poetry, and Scribes in Ancient Egypt, a very provocative title, in collaboration with the Harvard Museum of the Ancient Near East. And on March 27, we will host free museum hours at the Harvard Museum of the Ancient Near East as part of our Arts Thursdays programming from 5:00 to 9:00 PM. To learn more about additional upcoming museum events, I invite you to visit our website HMSC.harvard.edu, where you can sign up for our newsletter or you can also follow us on social media.

\[00:02:04.39\] We are grateful to all of our members and supporters for making these programs possible. I will now turn it over to Michael McCormick, Francis Goelet, Professor of Medieval History and Chair of the Initiative for the Science of the Human Past, who will introduce our speaker.

\[00:02:19.47\] Thank you, Caroline, for that very, very kind introduction. And thank you, Caroline, and the whole Peabody Museum of Anthropology and Ethnology, and your staff and your sponsors, for sponsoring this Science of the Human Past event, along with the Departments of Anthropology and Human Evolutionary Biology.

\[00:02:40.92\] You'll be able to learn more about Professor Warinner's discoveries and the selected posters that we have assembled for this event during the reception that Caroline has just spoken to you about. And thank you all, dear audience, for having honored us with your presence. And especially thank you to you, our all-important individual supporters and benefactors in the live and online audience. Without you, this would not be happening as you, our supporters, know better than most.

\[00:03:13.52\] Thanks to you, we are able to do our work and to share it with the world because of your steadfast material and moral support. It is a great pleasure, a special pleasure indeed, to introduce this afternoon the Landon T Clay professor of Scientific Archeology, member of the Steering Committee of the Science of the Human Past at Harvard, and Deputy Director of the Max Planck Harvard Research Center for Archeoscience, Professor Christina Warinner.

\[00:03:44.89\] You heard that my name is Mike McCormick, and it's my privilege to Chair the Science of the Human Past at Harvard, and to direct in the United States, the Harvard-Max Planck Institute-- Excuse me, Center for Archeoscience.

\[00:03:58.60\] Professor Warinner and I are among the many, many people who have come to Harvard from the American middle west, from across the Americas, and from around the globe to learn, teach, and discover new things together in this remarkable and remarkably open institution of education and discovery.

\[00:04:23.26\] While I hail from the Western end of the Erie Canal and the rust bowl, Professor Warinner first came to Harvard from the University of Kansas with a BA, listen carefully, in Germanic literatures and languages, and an Honors BA in anthropology. Here at Harvard, she got her PhD with a dissertation on the archeology, archaeogenetics, and isotopic profiles of an early colonial population in Mexico.

\[00:04:53.24\] As a graduate student affiliated with Lowell House, she was appointed, along with a certain Kyle Harper, currently at the University of Oklahoma, to serve as an official scribe of the first meeting at Harvard, dare I say it, 20 years ago, dedicated to bringing together bioarchaeology, DNA, and economic history, sponsored by a Mellon Foundation Distinguished Achievement Award that ultimately would lead to the launching of the Science of the Human Past at Harvard, the Max Planck program, and much else, as you will hear.

\[00:05:29.24\] Professor Warinner went on to work at the University of Zurich and begin her professorial career at the University of Oklahoma, continuing as a group leader in the new Max Planck Institute in Jena and Professor in the University of Jena before returning to Harvard and rising in six years from assistant to full-tenured professor in our university. Hooray for Professor Warinner.

\[00:05:53.75\] \[APPLAUSE\]

\[00:05:55.94\] Professor Warinner has won just about every prize I have ever heard of, and many that I have not heard of. She has published about 90 articles, hard to keep track, across the spectrum of archeological science, characterized each time by probing and rigorous methodological innovation, with a knack for identifying big questions and resolving them whether she's studying pre-Columbian Mexico, the Xiongnu Empire of East Asia, or the pigment from Afghanistan trapped in the dental plaque of an 11th century German nun and artist.

\[00:06:34.25\] She is at home in the field as an archeologist, including as an experimental archeologist, learning how to milk horses in Mongolia and in the laboratory. One set of publications I would like to highlight are Professor Warinner's wonderful coloring books about the adventures of a young female scientist, published initially in English and German, but also translated into languages such as Persian and Arabic, Chinese, and Hebrew, Polish, Spanish, and Nahuatl, among so many others.

\[00:07:11.41\] I could go on for a very long time about all that Professor Christina Warinner has already accomplished, but I cannot defer any longer the pleasure you and I will have in learning how the science of the human past has allowed professor Warner to learn how on Earth human beings figured out how to get digestible food from the milk of fast-moving animals, what that means for the understanding of human history and in general, as she tells us about her latest discoveries and her Peabody Museum exhibition in Making Milk, Mongolia's Unique Role in Dairy's History. And just to help out, here are the slides I failed to show you.

\[00:07:57.43\] \[LAUGHTER\]

\[00:08:06.57\] It's such a pleasure to be here, to see so many old friends and new faces. And I'm really excited today to talk to you about a project that's very near and dear to my heart, which is a project looking at the origins of dairying and trying to understand how we as humans developed milk into one of our oldest prehistoric technology, our dairy products.

\[00:08:27.46\] And it turns out, understanding this question is going to take us on a lot of surprising journeys into many places. And for me, it took me on a multiyear journey through Mongolia, which has been this incredible experience. I've learned so incredibly much.

\[00:08:41.47\] It's been such a fun project for me because it's stretched me in so many different directions. Everything from ethnography, to microbiology, to archeology, genomics, proteomics, to try to answer this really fundamental question. So today I'm going to take you a bit on this journey with me as I try to explore and try to explain milk's great story, which has a surprisingly important role, or Mongolia has a surprisingly important role to play. So today, I'm going to talk to you about Making Milk, Mongolia's Unique Role in Dairy History.

\[00:09:12.89\] And first, though, I just want to acknowledge this project was only made possible by the work of so many dedicated people in Mongolia and around the world who worked so hard to make this possible, contributing all sorts of expertise, knowledge, ideas, time, energy. They really are the kind of engine that has driven this research. And after the talk, when we have the reception, we have a number of posters that features many different aspects of their work that I won't have time to explain today in detail, but you can learn far more about it upstairs at the reception.

\[00:09:47.23\] So I've always been fascinated by food and the foods we eat, because we, as humans-- we as "Homo," are really unusual. We're different than many other animals. We're dietary generalists. People often ask me, what is the food we're supposed to eat?

\[00:10:01.58\] And you can't answer that question exactly, because one of the things that really distinguishes us from other animals is our dietary flexibility. We can survive and thrive on many different foods and many different diets. And we see that reflected in different populations around the world who eat vastly different diets and yet are healthy and thriving.

\[00:10:21.64\] But there are a few foods that do pose a bit of a mystery for us in terms of how they became prominent, because they're such unlikely foods for humans to adopt. And so today, I'm going to focus on one of these, which is milk.

\[00:10:35.95\] Now milk is really important. It's really important in the global economic food system. It's really important economically. Today, global milk production exceeds 970 million tons. And I encourage you to try to imagine what that looks like. And I bet, if you're like me, you have no idea what one ton of milk would look like, let alone 970 million.

\[00:10:59.29\] So I'm going to convert this into some more familiar units. So this is about 398,000 Olympic-sized swimming pools. That's an enormous amount of milk being produced every year. It's enough for every person on Earth to drink a quarter liter of milk every single day. And yet we know that the majority of people on Earth are actually lactose intolerant. So how did we develop this unlikely food into this global phenomenon when it's so indigestible by so many people?

\[00:11:28.27\] So this is a question that really drives me, this kind of milk paradox. How did this food become so widespread? How did it become an important global commodity on nearly every continent and where are its beginnings?

\[00:11:41.97\] Now on the one hand, milk is the most natural food that we as mammals can eat. All mammals begin life consuming mother's milk and milk is incredibly nutritious. It is designed to grow a young infant. It is really well-balanced in macronutrients and micronutrients. It's rich in sugars, and fats, and proteins.

\[00:12:05.10\] And a lot of people don't realize, though, that the composition of milk actually differs quite a lot from species to species. So here's just some milk compositional information for different common dairy taxa.

\[00:12:16.57\] And so the amounts-- the proportions of fats, and proteins, and lactose sugar, which is the dominant carbohydrate in milk, can vary a lot from animal to animal. And that's because the milk of different animals is really attuned for that specific species and the growing infant, or the growing offspring of that species. So it's really formulated to help those young grow as quickly and as healthfully as possible.

\[00:12:40.71\] So milk is an excellent source of nutrition for young children. And for young infants, it's the sole nutrition for many, many months, up to six months or more children are subsisting only on milk. But we are unusual amongst the animal kingdom in that we not only consume our own milk, we consume the milk of other animals.

\[00:13:01.78\] No other animal does that. And another thing that makes humans unusual is that we also consume milk long after infancy into adulthood, and even into senescence. We're the only animals that do that as well. So why is it that milk is so restricted in other animals?

\[00:13:19.08\] So while milk is incredibly beneficial for the growing young, it does have some limitations as well. The biggest problem that people face when consuming milk is this right here. It's the sugar lactose. It's the dominant carbohydrate within milk.

\[00:13:32.53\] It's actually what's called a disaccharide. So it's two sugars linked together. The linkage here is what's called a beta-glycosidic bond, which is actually quite hard to break. And it requires an enzyme to break it.

\[00:13:46.43\] And we produce the enzyme that breaks this into the individual sugars of glucose and galactose. We produce this enzyme. It's actually here encoded on chromosome 2. And the lactase is expressed in the small intestine.

\[00:14:01.57\] And it cleaves this bond between the two sugars. So the enzyme kind of acts as a pair of molecular scissors cutting it apart. And once cut apart, we can absorb those individual sugar molecules.

\[00:14:13.18\] And what's really important is also that the expression of lactose is actually controlled by a small upstream regulatory region in a different gene called MCM6. So you can think of it as a kind of switch that will turn on and off lactase.

\[00:14:26.92\] So what happens is when you're making a lactase in your small intestine and you consume sugar, then what happens is your lactase breaks it down into glucose and galactose. You absorb the sugars into your bloodstream and it encourages growth. So this is how you can derive energy from lactose and this is how infants do it.

\[00:14:46.72\] The problem, however, is that if you don't produce lactase in your small intestine, then it passes undigested through the small intestine until it reaches the colon, the large intestine, where bacterial concentrations increase 100 million fold. And they are very happy to break down that lactose for you. They will ferment it into the component sugars.

\[00:15:08.21\] The problem, however, though, is as a byproduct, it can produce up to something like eight liters of hydrogen gas for a large glass of milk. So you have all these kind of negative associations with the microbial fermentation of lactose.

\[00:15:21.16\] So as I mentioned, turning lactase expression on and off is really important. And it's controlled by this regulatory region in MCM6. It acts as a kind of dimmer switch. So early in life, MCM6 signals for lactase expression to be high, and you have very high amounts of lactase produced in the small intestine.

\[00:15:41.31\] But as a person grows, it slowly turns off. And by the time an individual reaches adulthood, it dramatically decreases in expression, to the point where it's barely measurable. And this is a normal process that happens not only in humans, but in all mammals. And it's a natural part of the weaning process of mammals.

\[00:16:01.31\] But of course, we know that humans consume the milk of animals long into adulthood. So how does this actually work? So in the early 2000s, a lot of the genetics behind this were worked out, and it was a major feat of early genetic sequencing.

\[00:16:15.14\] Because it turns out that this region here-- so this is the regulatory region. This is just the actual sequence part of the MCM6 intron that controls this expression of lactase. It has a number of mutations in it that, when present, will cause the regulation to nonfunction and it essentially leaves the switch on all the time. So it's like you turn the switch on as if you were an infant, but you keep it on for your entire life.

\[00:16:43.07\] And it was noticed that these different mutations that are present within this regulatory region, are found in different populations in the world. So this one here, which is-- if this cytosine turns into a thymine, which you see right there-- this is very hard to control. Bear with me. So the cytosine turns into a thymine. This particular mutation is very common in populations in Europe and Anatolia, in the Iranian plateau, and in Northern South Asia, for example.

\[00:17:11.99\] Other ones, like this one here, this T, if it turns into a G, this is a common in Arabian and Bedouin populations found throughout the Arabian Peninsula and parts of North Africa, and it's often associated with camel milking. We have a couple of other mutations, or allele variants here, that are associated with populations in Tibet. And there's also at least three different mutations that have been found in East African populations that also convey the same ability to produce lactase long into adulthood. And so what effectively these different mutations do is it keeps the lactase expression high so you have continued lactase production and lactose digestion.

\[00:17:55.26\] So this was really exciting when this was worked out in the 2000s. And this was thought to explain why we have variation around the world in the ability to digest lactose as adults. And what was really exciting in the early 2000s when ancient DNA technologies were coming online, there was a lot of excitement about trying to understand the origins of these particular mutations. And particularly the one that was that's found in Europe, because that one is quite geographically widespread. It's the most widespread of all the variants and has the most people that have that genotype.

\[00:18:26.14\] And so there was some really exciting work that was done in the early days of ancient DNA that found that the earliest farmers within Europe derived their ancestry from the ancient Near East. And so this was part of this growing archeological model that Near Eastern farmers that developed the world's earliest farming spread into Europe. And the idea was that they had adapted, they had kind of evolved this ability to turn off this regulatory system, to keep their lactase expression high, and that had allowed them to develop dairy as a technology. And then when they spread into Europe, they carried this with them. And this is how we start to see the beginning of the global spread of this genotype.

\[00:19:07.47\] There was a lot of other work that was supporting this. So this is some of the work by Wolfgang Haak that was supporting this. There was also work showing that domesticated cattle in Europe from these early farmers were related to Near Eastern farmers.

\[00:19:19.33\] So it was building up this model that seemed to really conform with earlier work by Gordon Child and also Andrew Sharrett talking about the Neolithic Revolution and the secondary products revolution of bringing these technologies into Europe. And there was also work looking at ancient pottery that had established that even the earliest pottery that's being made in the ancient Near East has milk in it. So it's a very ancient origin.

\[00:19:47.18\] So we know that milking begins at least by 7,000 BC, so 9,000 years ago. And so this idea is that milking technologies emerge early, the genetic adaptation emerges early, and it spreads with these farmers. And that's how we get our present day distributions.

\[00:20:06.35\] And then later, there was also evidence of milking found in milk fats absorbed into pottery throughout Europe, linking this all together, including this cheese strainer here. There was a lot of modeling to try to integrate genetics into this and to try to predict exactly where and when this was selected for. And a lot of these models suggest that it was heavily selected for as it moved into Europe, all part of this Neolithic package.

\[00:20:32.27\] And a lot of this work culminated in 2013 with this really fantastic article called "The Milk Revolution," that kind of put all these pieces together. It was kind of this singular achievement of archeology of bringing together ancient DNA, bringing together more than 50 years of archeological research, putting all these pieces together for the Neolithic origins and spread of dairying technologies, and the spread of lactase persistence into Europe.

\[00:20:57.44\] This was the model. So it begins around 11,000 to 10,000 years ago. It spreads out of Anatolia and out of the Eastern Mediterranean, up through the Aegean, and up through Italy and then into Europe. And then along the way, we have these different developments.

\[00:21:16.04\] So ta-da. We solved it. We now know how people have adapted to milk.

\[00:21:20.55\] We know why there's variation in populations, why some people can digest it easily, and some can't. And it's related to this lactase persistence genetic variant that is present in different populations around the world.

\[00:21:31.71\] So we're like, we're done. This was amazing. Archeologists had solved a real problem. It was fantastic.

\[00:21:38.78\] And then came 2015 and our model broke. There was a series of a couple of papers that were kind of pre-- there was like a prelead to this in 2007 that gave a hint that maybe not all was right. So this was an article that came out of the University of Mainz, led by Joachim Burger.

\[00:21:56.43\] And they had done the first ancient DNA study looking specifically for the European variant of the lactase persistence allele. And they had looked at a number of early European farmers and they hadn't found it. But they hadn't actually looked at that many individuals.

\[00:22:11.25\] So in 2007, they said, you know, we're probably just underpowered. It was the early days in ancient DNA technology. We didn't have a large enough sample size. Surely, if we just keep looking, we will find lactase persistence.

\[00:22:22.81\] Maybe it's just not a very high frequency yet and we have a biased sample size. And so everyone kind of moved on. But in 2015, three large follow-up studies came out, really bringing with them all the power of the next-generation sequencing revolution, and they looked at very large numbers of individuals spanning the Neolithic through the Bronze Age. And they came to some really startling conclusions.

\[00:22:48.63\] They found that lactase persistence does not arise amongst the early dairy farmers. It is completely absent amongst all of the Neolithic Europeans. Even to this day, we haven't found lactase persistence in the Early Neolithic farmers. They simply didn't have it.

\[00:23:05.34\] We know that Europe's earliest cheese making linearbandkeramik farmers, the earliest farmers, were not lactase-persistent. So we know they're consuming dairy. We have lots of evidence of dairy. We have lots of milk fats on pottery. And yet they didn't have these genetic variants.

\[00:23:20.38\] So we know they were consuming dairy products, and making cheeses, and milking animals. But they didn't have this genetic variation. And we now know that dairying precedes the earliest evidence of lactase persistence by more than 4,000 years.

\[00:23:35.38\] So for 4,000 years, humans invent this prehistoric technology, use it to augment their diets. They spread with it around the world, and yet they have no genetic capacity to digest it. So our model of what makes it possible to digest milk doesn't work for the first 4,000 years of dairying. So it turns out that actually lactase persistence, while it can be really important today, is completely irrelevant for the early history of dairying.

\[00:24:03.55\] So what is happening here? This was so interesting to me when these studies came out, and I thought, this is an amazing question. This is an incredible anthropological question, and this is one I want to set out to try to answer.

\[00:24:16.68\] So if you look around the world at where you find dairying technologies, you actually find different dairying populations all over the world who have traditionally practiced dairying for long periods of time. But one of them in particular really caught my eye and that is in Mongolia.

\[00:24:32.09\] A lot of people don't realize that Mongolia has a dairy-based economy. In fact, dairy is the central focal point of the subsistence economy in Mongolia. Even today, approximately 40% of the population still lives as nomadic herders, managing their livestock and surviving on their dairy products.

\[00:24:50.11\] A lot of people also don't know is that Mongolia milks more livestock species than any other country on Earth. They milk seven different species of livestock, including horses, camels, reindeer, yak, cows, sheep, and goats. And so I began to wonder if perhaps there might be an alternative explanation for how humans have adapted to consuming dairy diets.

\[00:25:16.61\] What if it's not all about adapting the human genome? What if it might also be about adapting the microbes that aid us in our digestion, either the culinary microbes that we use to make those dairy products, or our own microbiome that aids us in digestion, and it also is involved in the symptoms of lactose intolerance.

\[00:25:35.53\] Now this is challenging, because there's a lot of work that needs to be done to establish this and to bring in the ethnographic record to also kind understand how these things might link up. Here are some of the dairy products that are produced in Mongolia. This is a form of yogurt here that you see on the top. It's called tarag. And this down here is the beginning stages of making curds, a kind of-- similar to cheese.

\[00:26:02.35\] And so I began to wonder if these microbes couldn't be involved, somehow, in mediating the effects of lactose intolerance. And if that's happening today in Mongolia, perhaps it could be a model for thinking about how humans first adapted to dairy digestion in the deep past. And so in 2017, I began the dairy cultures project with the idea of trying to understand how these things might link together.

\[00:26:26.76\] And to answer this question, we were going to need to bring lots of different lines of evidence. We were going to have to look at ethnography, microbiology, archeology. We're going to have to use the tools of genomics and proteomics to come up with a plausible model for how ancient and present day Mongolians successfully adapt to a dairy-based diet that continues to the present day.

\[00:26:45.53\] So we worked with different herders who gave us enormous amounts of information and insights into how they produced the dairy products. We generated a lot of microbiome data to try to study and understand how microbial variation is affecting both the production of the products and also their digestion. And we also integrated a lot of archeological work to understand how old dairying was in Mongolia, which when we began the study, was not known. And we integrated the tools of genomics and proteomics to do this.

\[00:27:18.08\] So we had a five-year research grant to do this project and we had so many incredible outputs that we did, integrating all of this. Mike mentioned earlier our coloring book, which is now available in 63 languages.

\[00:27:33.38\] One of my undergraduate students was studying both social anthropology and fashion design, and in order to raise awareness about how climate change is affecting microbial ecology and the subsistence of herders in Mongolia today, she actually put together a whole fashion line featuring microscopic images of dairy bacteria.

\[00:27:55.50\] We did a number of workshops. I took my entire team to Mongolia, my entire field team, and my bioinformaticians, and my microbiologists, and laboratory technicians. We went to Mongolia.

\[00:28:08.55\] We met the herders. They learned how to make cheese. We brought the herders to conferences. We really experimented a lot on trying to learn as much as we possibly could from each other and to build up a really robust project. So I'm going to show you some of the things that we learned over the course of this work.

\[00:28:24.99\] The first question we wanted to know is when did dairying begin in Mongolia? This is an image of a khirigsuur. This is a Bronze Age burial mound. There are thousands, if not tens of thousands of these, across the Mongolian landscape. And there's an incredibly rich archeological record that allows us to ask this question.

\[00:28:44.67\] We ended up partnering with archeologists across Mongolia to have a near-complete coverage of the country and a really great coverage of time. So we were able to look at the last approximately 6,000 years of Mongolian history and we were able to collect human remains samples, specifically dental calculus, and teeth was what we used to study from many different periods throughout Mongolia.

\[00:29:11.46\] You can see here some of the different archeological periods represented from some of the earliest Pastoralist populations associated with the Afanasievo culture through the Early Bronze Age Chemurchek, lots of different Late Bronze Age sites, into the Iron Age and the height of the building of the nomadic empires, on all the way to the Turkic and Uyghur era, and also to the Mongol empire.

\[00:29:37.98\] So to solve this, we wanted to look at two things. We knew we were going to have to use genetics to reconstruct the genomes of these ancient people to look at how they're related to one another and to neighboring populations, and also to genotype them to see if they have evidence of lactase persistence, which wasn't known. We also were going to use proteomics, which allows us to track milk proteins through time so we can identify when and where people are consuming milk.

\[00:29:59.86\] So we know that today that populations in Mongolia are predicted to not have lactase persistence. There's very little work that's been done here. I'll show you some of the work that we've done.

\[00:30:09.29\] So we look through all 6,000 years of time, more than-- we had 214 ancient genomes. And what we found is that the lactase persistence allele has been at extremely low levels across that time. So it gets introduced during the Early Bronze Age with the migration of pastoralists that come from the West, from Western Southern Russia.

\[00:30:30.31\] They bring their herds with them. They're the first group that introduces dairy pastoralism into Mongolia. And we start to see very low levels of lactase persistence alleles. But it never rises above 6% of the population in Mongolia's entire history.

\[00:30:43.70\] And this is amazing, because this locus in Europe is the locus under the highest selection in the entire human genome. It's under higher selection than, for example, resistance to malaria. And yet it has not undergone any selection in Mongolia, despite the fact that people today have a dairy-based diet.

\[00:31:05.95\] We were also able to reconstruct the population history of Mongolia through different time slices and look to see how the population was moving and interacting with neighboring populations and modeling their contacts with all of their neighbors across the 6,000-year period.

\[00:31:22.75\] When did dairying begin? We needed to be able to detect milk directly. And so there's a couple of different methods that have been used for this. So early on, people had the idea of trying to detect milk proteins directly in pots.

\[00:31:35.36\] But it turns out that proteins do not survive well at all in pottery. So that's really not a very good option for us. One of the other problems that we have is sometimes the milk fats can be absorbed, but then you're reliant on those pots, and we don't always have those in our archeological context in Mongolia.

\[00:31:52.61\] And one of the problems we have in the ancient Near East is that the very earliest pots contain dairy fats. So we know that dairy predates the invention of pottery, but because we were relying on pottery to identify it, we could never determine the earliest date. So this got me and my team thinking, how could we think about this in a different way? And how could we utilize the fact that Mongolia has such an incredibly rich archeological and bioarcheological record?

\[00:32:14.22\] So one of the things we started focusing on is dental calculus, which is the calcified dental plaque on the surface of your teeth. You pay good money for people to scrape this away, probably, twice a year. But it turns out it is a goldmine of information about the life that you live. It acts as a kind of sink that traps all sorts of information about you.

\[00:32:33.30\] It contains your DNA. It contains your oral microbiome. It contains seasonal pollen that you're exposed to. It contains microscopic, tiny feather particles, if you have feather pillows. It has all sorts of food debris that become entrapped.

\[00:32:47.67\] And importantly, it entraps dietary proteins very well, and especially milk proteins. And so we began a study of dental calculus. So just scraping off a bit of this tooth tartar, or dental calculus, much like your dental hygienist would do. And we extract proteins from it and look for milk proteins.

\[00:33:06.78\] So it turns out milk has about 10 abundant proteins. And we can look for these to be able to track milk through time. There's one in particular called beta lactoglobulin I, which is a really robust protein. It's not the most abundant protein.

\[00:33:22.75\] It's about the fifth most abundant protein in milk. But it's a protein that sticks around for a really, really long time. These other proteins are actually quite easy to digest. And in fact, most cheeses are made on the basis of curdling those different proteins in order to make all sorts of delicious foods.

\[00:33:39.46\] But beta lactoglobulin is really resistant. In fact, people have gone so far as to try to genetically engineer cows to not make beta lactoglobulin because it can be so hard to eliminate from dairy systems. But those very processes that make it very hard to digest and make it hard to remove mean that it persists very deeply in the archeological record.

\[00:33:58.86\] And so we can look for this protein. This is what it looks like. This is if you were to unroll it, if you were to fold out that protein. This is this amino acid sequence here. And you can see a protein model of it over there.

\[00:34:11.79\] In my laboratory here at Harvard University, we use a tandem mass spectrometry to be able to identify it. And you see some people in the lab working on it right now. This is what the data looks like.

\[00:34:23.98\] Some people always wonder, what is your data actually look like? What does mass spectrometry output look like? So this is an example of dental calculus that you see here from some of these samples. And we're looking at an enormous number of proteins that we're identifying in it.

\[00:34:36.90\] We can then individually look at individual ones and identify specific proteins that can be highly diagnostic for milk. It turns out beta lactoglobulin is not found in any other tissue in the body. It's extremely specific to milk.

\[00:34:48.52\] So when we find it, we can be very certain milk is present. And not only that, the sequence varies between different animals. So we can say not only milk is present, but we can tell you, oh, it's sheep milk, or it's cattle milk, or it's goat milk, or its horse milk. And now we can start to look at how these different ancient dairy economies were constituted.

\[00:35:05.65\] So this is some of the output we had from our study. So we're looking at all these different milk peptides that we've been able to identify and that we've been able to taxonomically classify. And for this particular study here, we identified 139 different dairy peptides deriving from beta lactoglobulin and alpha I casein that ranged in age from 1,000 to about 5,000 years ago.

\[00:35:30.22\] And so we can use this, then, to reconstruct these different animals. So if we go back to our model where we were analyzing the genomes of these individuals, we also analyzed their dental calculus proteomes looking for the milk. And we were able to place it in time.

\[00:35:43.73\] So we were able to show that the earliest evidence of dairying, the earliest evidence of milk consumption, begins with the Afanasievo period. So these Afanasievo pastoralists that arrive at this time, they originate from the North Caucasus region between the Black Sea and the Caspian Sea. They're the ones who introduced dairy pastoralism, the first groups that we see it in. And they bring with them primarily sheep, but also goats and cattle.

\[00:36:09.19\] We next see horses come in the late Bronze Age as the earliest evidence we have of horse milking. And then later on, we see camel milking being folded into this dairy system. So we can see the-- we can track, actually, trajectory of the growth of Mongolia's dairy economy to become what it is today. And it formed over many thousands of years, but begins at roughly 3,000 BC. And we also get yaks that come in also around this period.

\[00:36:36.52\] And so this has allowed us to build up this understanding of ancient Mongolia and how it developed the dairy-based system that we see today. And this was a painting that we had made on the basis of the genetic evidence that we had reconstructed from these ancient people.

\[00:36:52.99\] But how did it begin in Mongolia? So I've kind of hinted at this. We know that dairying begins at least by 7,000 BCE, probably earlier in the Near East. The reason we don't know how early is because that's when pottery begins and the method that's been used to identify it is based on isolating milk fats from pottery.

\[00:37:10.50\] And we know that it spreads to Europe by 5,000 BC. It goes into Africa by 4,000 BC, and it spreads up to this North Caucasus region by 4,000 BC. So people are very interested in spreading. And this is all prior to any of these people having the lactase persistence allele.

\[00:37:27.98\] We know that in this North Caucasus region, it becomes really important for laying the groundwork of the Bronze Age. We call this area the crucible of the Bronze Age. This is the place where the beginning of the Great migrations are going to come. And this is a really critical area for the spread of dairying.

\[00:37:42.18\] Although dairying is invented during the Neolithic,. the kind of dairy global economy we have today doesn't really have much to do with that Early Neolithic history. Much of what the global milk economy is today actually derives from this Bronze Age period and the developments that are happening here in the North Caucasus region.

\[00:37:59.97\] It is here where the first fully mobile pastoralism is developed. It's here where wagons are invented and long-distance mobility that ends up spreading dairying far and wide across the entire Eurasian continent. We know, then, because of our evidence in Mongolia that arrived here by 3,000 BC.

\[00:38:15.98\] And so we did a study a couple of years ago, led by my postdoc, Ashley Scott, where we investigated different individuals from across this region to try to understand how their dairy economies worked. We were able to reconstruct the origins of dairying in this region to a little bit before 4,000 BC, and we can show we have dairying continuously.

\[00:38:36.39\] But one thing that's very interesting is that the earliest dairying that we see is all sheep dairying. So during this really critical early period, they seem to be entirely focused on sheep dairying. And only in the early third Millennium BC do we start to see a switch to diversification to milking more cattle and goats, which is probably a response to a degrading climate. And it's out of this population here, this sort of sheep-focused population, that we get a migration of people, the Afanasievo that introduced dairy technologies to Mongolia 5,000 years ago, at about 3,000 BC.

\[00:39:13.86\] As I said, we can actually look at this genetically. So here are these individuals that we find in Mongolia. They fall right within the genetic variation that we would expect for West Eurasians, and in particular, Yamnaya-like populations in the North Caucasus region. And then we can show that we have subsequent migrations that bring new people and more animals into Mongolia afterwards.

\[00:39:36.54\] But why dairying? So you might wonder, like, why? Why would you do this? It seems unnecessary. Why would something like this be really attractive?

\[00:39:45.97\] Because also what's really interesting in Mongolia is that although these individuals brought dairying, the dairying went much further than these migrants. We see that hunter-gatherers throughout all of Mongolia, over the short span of a couple thousand years, all adopt dairy pastoralism.

\[00:40:00.70\] So dairy pastoralism becomes the dominant, the main, almost the only subsistence strategy that's practiced relatively quickly. What led all of these people to adopt the production of a food that they didn't have the genetic capacity to digest?

\[00:40:17.16\] And I think one of the things that's really important to note in this region is that dairy is a solution to the steppe. The Eurasian steppe is vast, but it's actually very hard for humans to live in, because there's very few foods that we can grow to survive. It has a very short frost-free season.

\[00:40:33.58\] There's almost no crops that can be grown. Millet is one of the only crops that can be marginally grown in this region, because it simply doesn't have a long enough growing season. In many places, the growing season is only three months long.

\[00:40:44.80\] And so if you don't have access to something like a lactating bovid or equid, you're really limited to riverine corridors, and so much of the steppe, you can't access or utilize. But as soon as you have a lactating cow, or sheep, or a goat, or a horse, suddenly those animals are able to convert all of that grass into nutrition that you can consume, and the entire steppe opens up for habitation.

\[00:41:11.89\] Dairying is also an incredible solution to mountain areas and also deserts. It's thought that the reason that we see one of the lactase persistence alleles at such high levels in Bedouin populations is because of their association with camel milking. Camels have a much higher tolerance for salty water than humans do.

\[00:41:30.65\] So a lactating camel acts as a kind of desalinization plant for you and allows you to go on longer journeys deeper into the desert, and also access salt water oases that wouldn't be suitable for humans. So there's many reasons why having a lactating animal and utilizing their milk can be really advantageous.

\[00:41:49.74\] So we wanted to dig in now more into the microbiology of this. What's the link? How are people actually making this work? And to do this, we worked with a lot of herders to try to understand how they produce their dairy products.

\[00:42:01.26\] We took lots of samples of dairy products. We also conducted a large-scale gut microbiome study. The herders we worked with were extremely generous in allowing us to interview them, document this process. I love this photo.

\[00:42:12.75\] This woman, her name is \[? Geryl. ?\] She's showing us how to use a still. This is a 100-year-old still from her family. And she's making alcoholic yak yogurt that she's going to distill into something called shimiin arkhi, which is a type of yak vodka.

\[00:42:28.16\] And so Mongolia is one of those places where they also make a wide range of alcohols, three different major types of alcohols, from milk. So they make shimiin arkhi, which is this fermented yak or cattle milk that's distilled. They also make something called airog, which is fermented horse milk, and they also make something called khoormog which is a fermented camel's milk.

\[00:42:50.54\] We did a lot of ethnographic comparisons of this to strategies that are used, for example, in Alpine regions. There's many similarities, but also differences. And we were interested in understanding how these differences might affect the microbiology of the products and the different digestive properties that then has.

\[00:43:07.58\] And we also mapped out many of the different dairy products that are produced. There are dozens and dozens of different dairy products that are all produced from milk in Mongolian households across the countryside. We partnered with Soninkhishig Tsolmon, who is the Director of the Mongolian Dietetic Association, to conduct a large-scale nutrition study across Mongolia and also a gut microbiome study.

\[00:43:30.99\] And for our work, we focused on four different field sites. So we worked with herders in the far North who are mainly cattle and yak herders. We worked with herders that live in Central Mongolia and Bulgan, which is a region that's really famed for its horses. We worked with herders that live in the far South, who raise camels. And then we also worked with individuals who lived in the capital city, in Ulaanbaatar.

\[00:43:55.05\] We collected a lot of biometric information, a lot of nutritional information. We also conducted something called a lactose challenge test, where we gave individuals a 25-gram dose of lactose, and then we measured the breath hydrogen that they produce. So breath hydrogen is an indication that you have colonic fermentation.

\[00:44:13.18\] So if you're not digesting the lactose in your small intestine and it passes to your large intestine, you can then measure the products of that bacterial fermentation from that hydrogen gas that I mentioned that is produced as a product of bacterial fermentation. And so we conducted a lactose challenge test. And we collected gut microbiome samples as well.

\[00:44:33.82\] So I just want to show you a few results because I think they're pretty exciting. And also, this is how real research works, which is it's very tantalizing and very exciting, but also contains a lot of puzzles that we still haven't answered. So one of the things that's really interesting is that if you follow kind of the medical model of lactose intolerance, then these individuals are lactase nonpersistent.

\[00:44:56.71\] They should not be able to digest milk without producing large amounts of hydrogen. And yet we find that across the individuals in our study, which is 100 individuals, 32% of them actually will digest the lactose and not produce any raised amount of hydrogen. So they have no hydrogen response. They have no colonic fermentation response to a dose of pure lactose.

\[00:45:19.89\] Now what's really interesting is we can break that down, because it turns out it's actually different between individuals living in the city and living in the countryside. So in the city, that number actually falls to about 24%. About a quarter of individuals show no response. But in the countryside, people who are consuming dairy products all the time and making it, 40% of those individuals showed no hydrogen response. So no lactose intolerance response to a lactose dose.

\[00:45:48.19\] We can then break this down by region. And we see some kind of interesting patterns here. So we actually see that the highest level of nonresponse is in northern Mongolia in Khovsgol amongst the cattle and yak herders. And then we see kind of an intermediate-- or we see an intermediate response among Bulgan. And we actually see the lowest response in Gobi, which I think is something that's really interesting, and we're not sure we totally understand.

\[00:46:08.60\] But this is a latitude gradient. And also it's very different in climate, with it being much wetter in the North and much more arid in the South. But we can also test to see, do these individuals have lactase persistence?

\[00:46:21.56\] We saw from ancient DNA that from over the last 6,000 years, going up to the Mongol era, we never saw levels of lactase persistence in the population greater than 6%. But it turns out before our study, no one had ever done a lactase persistence study of present-day Mongolians. And so we conducted the first study.

\[00:46:38.30\] So what I can do is we can say, OK. Well, what percentage, or what fraction of this nonrepsonse can actually be explained by their genotype, by lactase persistence? And this was really fascinating to me. So it turns out in the North, we actually have a fair number of individuals who are lactase persistent, who do have that European variant that has probably persisted since it was introduced thousands of years ago.

\[00:47:02.33\] But we still have about a quarter of individuals that are lactase nonpersistent and still showing a nonresponse. That's similar to kind of what the level is like in Ulaanbaatar. And we don't see any lactase persistent individuals, either in the Gobi or in Bulgan.

\[00:47:17.42\] But what really struck me is Bulgan. So these horse herders have this really amazing phenotype, in that none of the horse herders that we looked at, or that we worked with had lactase persistence. And yet 45% of them show no response to lactose.

\[00:47:31.52\] And that's really incredible, because in the summertime, during the height of the milking season, people in Bulgan consume extremely high levels of lactose. Adult men will consume more than 200 grams of lactose per day, that's like an entire cake's worth of lactose per day, without showing obvious symptoms. So how is this possible?

\[00:47:52.41\] So we proceeded to try to understand the gut microbiome data that we had collected, the samples that we collected. We ended up sequencing more than 17 billion read pairs from our data set, with an average of about 175 million DNA sequences per sample. From that, we have computationally reconstructed these into bacterial genomes.

\[00:48:13.21\] We now have 9,000 microbial genomes reconstructed that we're now combing through and trying to understand their function. And one of the things we found is that many of these are actually-- appeared to be unique to Mongolia, or at least they've never been described in other places. And we're very interested to know if they might have some interesting functions.

\[00:48:29.77\] One of the things we can do is we can look at the composition. So beyond looking at individual genomes, we can also look at the proportions of different species that are present. And one of the things that's really interesting is we see that-- this plot just shows each point there is a different individual.

\[00:48:44.64\] And points that are close together are more similar in their microbial composition. And points that are further apart are more different. And we can see that there's a distinction between the herders and the people who live in the city, who live in Ulaanbaatar. They have distinct microbiomes. There's something going on there. And it appears that the herders may have an adapted microbiome, although the details of that, we're still trying to work out.

\[00:49:04.73\] We can visualize this in a different way. We can use a heat map to try to understand what are the specific bacteria that are overrepresented or underrepresented in different groups. And we see there's actually a large number of bacteria that are overrepresented in the herder population.

\[00:49:19.80\] So all of these bacteria here are overrepresented-- This is very hard to use-- are overrepresented in the herding populations, whereas these down here are the ones that are overrepresented in the city population. And what's really interesting is the city population, these are the microbes that are in our guts.

\[00:49:37.03\] My gut is full of these. Your gut is full of these. These are very typical urban population gut microbes. But what's really interesting is the ones that we see elevated in the herders are not typical, but include many bacteria that are really abundant in dairy products. So it appears that many of these lactic acid bacteria may be assisting in digestion in these herders, and that may be to help to explain how they're having a reduced response.

\[00:50:02.49\] So in particular, one of the things we see is there's a bacteria called Lactobacillus helveticus. It's named helveticus after Switzerland, because it's a major bacteria in Swiss cheese. But it turns out it's really abundant in horse milk and horse milk products. We saw that at very high levels in the gut microbiome of people in Bulgan.

\[00:50:18.48\] And we also see Lactobacillus delbrueckii at very high levels of individuals in the far North and in the far South. This is a bacteria that is very typical of yogurt. So if you go to the store, this is going to be the dominant bacteria that's going to be present in yogurt. But it's very high levels in these Mongolian population gut microbiomes.

\[00:50:36.18\] We also saw that Bifidobacterium angulatum-- this is actually a gut microbe itself. It's not coming from the diet, but it belongs to a group of bacteria that are really special. And they're highly milk adapted and are very abundant in infants. We saw that these are also really enriched in some of these populations.

\[00:50:55.32\] This, I think, is really exciting to explore. But we need to do more work. So one of the things that has been known for a long time is that these lactic acid bacteria, when consumed in live form, can actually ferment, can break down up to 50% of the lactose in the small intestine before it ever reaches the large intestine. This can really mitigate the symptoms of lactose intolerance.

\[00:51:14.80\] But one of the problems that has always dogged studies of probiotics is that there's enormous unexplained variability between species and strains. So just knowing the species isn't enough to help you predict how it's actually going to function in your body, which is why so many probiotics that have been brought to market are disappointing. Oftentimes, they're not being brought because of their proven efficacy, but because they're already approved for sale, or it's easy to market them, or easy to grow them. But if we could do more targeted work to actually find the specific strains that have potential benefit, this could be really important.

\[00:51:47.83\] And so I'm working with a team of microbiologists. We've been characterizing the dairy products. This is just showing some of the many dairy-- the bacteria that we're isolating from the dairy products. And we are culturing them. We have now more than 400 isolates in culture, which we're now planning to study in an expanded version of our project to try to look at their function and to understand if we can link specific strains and their genetic functions to protective effects against lactose intolerance.

\[00:52:18.04\] So I hope you've learned a little bit of something about dairy products that you didn't know before. One of the things that I think is so interesting about humans is that we often find more than one way to solve a problem. And it appears in the case of milk and dairy, and the problem of lactose, we may have adapted two very distinct approaches to solving this problem and creating a food.

\[00:52:39.95\] In one track, there's some populations in the world changed their genome to be able to continue to digest lactose long into adulthood, to be able to utilize dairy foods. But perhaps the earliest form of adaptation was not the genomic adaptation, but rather a microbiome adaptation. And that may have been the original one that functioned deep in human prehistory and still persists today among certain populations around the world, like Mongolia.

\[00:53:07.44\] So thank you for your time and attention. And if you're interested in learning more about any of the work that we've done, I'd be happy to tell you about it. And I want to be sure to acknowledge the many people who contributed to this project. Thank you so much.

\[00:53:21.08\] \[APPLAUSE\]



 



 

 See also:- [ Asia ](/region/asia)
- [ Public Lectures ](/video-type/public-lectures)
- [ History ](/video-type/history)