How did the Europeans become white? - Archaeological Science Ep 12

Have you ever taken a DNA test? If you have, then your report may have estimated what the color of your hair, eyes, and skin probably is based on your genetics. Did you know that we can do this for ancient skeletons too? I'm Madam Archaeologist, your go-to informant on everything archaeology, and in this episode, I'll show you what the latest archaeogenetic evidence suggests about how and when the Now, the traditional story goes that our species originated in East Africa and then migrated out of the continent, and that those anatomically modern humans who migrated to Europe more than 45,000 years ago became lighter in skin color in Europe as an adaptation to regions with lower UV radiation. Pale skin absorbs more UV radiation and so would have helped prevent vitamin D deficiencies It looks like the story isn't entirely correct. I want to know before getting into the details that the picture I am presenting here is based on current DNA evidence. There's always a chance that a new finding will shift our understanding in a new direction, but until then, this is the picture that is best supported by our evidence at hand. Our skin color, at least the color that we are born with before sun exposure, is dictated by our genes, and I won't give you a masterclass on genetics right now, but it's useful to at least know a simple definition of what a gene is. You're probably already familiar with what our nuclear DNA looks like. It's a double-stranded helix. It looks like this because nitrogenous bases are linked to one side of the helix and then pair with the respective complementary base. These bases are called adenine, thymine, guanine, and cytosine, abbreviated A, T, G, and C. A always pairs with T and G always with C. We call these base pairs, and the human genome has 3.2 billion of them. Nuclear DNA is packed into units called chromosomes, which are located in a cell's nucleus. We humans have 23 pairs of chromosomes, 22 pairs of autosomes, and one pair of six chromosomes, and we inherit one pair from each of our parents. A gene is a section of DNA at a particular location on a chromosome that codes for a protein. Codes for means that it contains the instructions or blueprint for building proteins. There are also some non-coding regulatory genes, but for now we'll just focus on the genes that code the proteins. One of the things that these genes do is determine what you look like, which geneticists call your phenotype. Two genes are strongly associated with lighter skin pigmentation amongst modern-day Europeans, SLC24A5, which is located on chromosome 15, and SLC45A2, which is located on chromosome 5. These genes carry alleles associated with lighter skin color. What's an allele? A simple definition would be that it's a version of a gene. And how do alleles arise in a population? One way is through single nucleotide polymorphisms, or SNPs. Remember the four nitrogenous bases? A SNP is a type of mutation, a change in a genetic sequence, where a nitrogenous base at a single base pair location has been substituted for another. So say that base pair location so-and-so originally boasted an A, a mutation may occur where that gets substituted for a G. If G rises in frequency within the population, we then have these two alleles, A and G. In gene SLC24A5, a SNP known as RS1426654 refers to the substitution of the nitrogenous base G for A, while SNP RS16891982 in gene SLC45A2 refers to the substitution of C for G. It's the A allele in the former gene and the G allele in the latter that are associated with lighter skin pigmentation. Using archaeogenetic terminology, when we want to say that someone carries the allele associated with lighter skin color, we say that he or she carries the derived allele. For others, we say that they carry the ancestral allele. It's worth keeping this in mind because I'll be mentioning the words derived and ancestral a lot here. When humans reached Europe more than 45,000 years ago, they had darker skin. In other words, they still carried the ancestral alleles in the two genes. You've probably heard of the famous Cheddar Man. He lived in Britain about 10,000 years ago and his DNA showed that he carried the ancestral alleles in genes SLC24A5 and SLC45A2. This is a reconstruction made after his DNA was tested. It also shows him with blue eyes, though note that the actual prediction was blue or green. This reconstruction has been subjected to some debate. One of the scientists involved in the reconstruction said that this is just his predicted skin color and that there's no way of knowing for sure that it's accurate. And if we take a look at the supplementary materials from the study Ancient Genomes Indicate Population Replacement in Early Neolithic Britain, where Cheddar Man's phenotype was predicted, we see that there was a chance, based on an incomplete profile, that he had very dark skin or intermediate colored skin. Since information on some alleles for other minor genes are missing, his predictions range anywhere from about 15% for intermediate skin and about 85% for dark to black skin, when the researchers assumed that the missing alleles would have been in the ancestral state, to a 75% chance of intermediate skin and a 25% chance of dark to black skin, when these alleles are omitted entirely. Now, Cheddar Man did not carry the derived alleles in genes SLC20485 and SLC45A2, the most important genes involved in lighter skin pigmentation, so I doubt he was pale like it's shown here in this reconstruction done before the DNA test. The chance of him having been darker than the average European today is very high, it's just the exact shade that's hard to establish. Maybe this is what he looked like, but maybe it's not, because when it comes to estimating phenotypic traits, we work with predictions. We say what someone's most likely skin, hair, or eye color is, and sometimes our confidence in our predictions is high, but sometimes we can't establish someone's most likely phenotype with high confidence. The researchers even admit that better coverage of Cheddar Man's genome would help in estimating his exact skin pigmentation better. Still, Cheddar Man's predicted phenotypic traits, darker skin, whatever the exact shade, and light eyes, follows the general trend predicted for the hunter-gatherers of Western, Central, and Southern Europe. A 17,000-year-old baby from Puglia in Italy, referred to as La Mura Juan, the 7,000-year-old La Braña Juan man from Spain, and an individual from Grotte du Bichon in Switzerland from the late Upper Paleolithic period are just a few of the many more examples of the European hunter-gatherers that carry the ancestral alleles in genes SLC24A5 and SLC45A2 and were predicted to have had dark skin. This all changed in Western, Southern, and Central Europe when farmers from Anatolia, which is modern-day Turkey, migrated westward, first reaching the They carried the derived alleles in both genes. So, these farmers brought not only agriculture with them, but also genes for light-colored skin. They settled around the continent and admixed, to varying degrees, with the local hunter-gatherers. With the arrival of Anatolian farmers, the derived allele in gene SLC24A5 increased greatly in frequency. This one study on Anatolian and early European farmers suggests that the frequency of the light-skinned allele in that gene rose to 90% in Europe during the early Neolithic period when it was introduced. You can also see this here in this graph. A value of 1 means that the allele is fixed or present in 100% of the population. We can see the fixation of SLC24A5's derived allele in the Neolithic Anatolian farmers, represented by the blue circle, and a very high frequency in the early European farmers, whose ancestry was mostly Anatolian farmer-derived. However, the other derived allele in gene SLC45A2 seems to have taken some more time to become prominent in Europe. But there's an interesting plot twist. Researchers found alleles for light-colored skin present in hunter-gatherers from Eastern Europe and in hunter-gatherers from Scandinavia. At the time of the Mesolithic, that is the term we use in prehistory for the period before the arrival of farming, hunter-gatherers in Scandinavia were genetically a mix of those hunter-gatherers from Eastern Europe, whom we call Eastern hunter-gatherers, or EHGs, who appear to have had skin colors ranging from light to intermediate, and the hunter-gatherers from Southern, Central, and Western Europe that had darker skin pigmentation. These are called Western hunter-gatherers, or WHGs. It appears to have been migration of EHGs, already carrying the derived alleles, that brought the first wave of lighter skin genes to Scandinavia. According to this paper, which studied the genomes of Scandinavian hunter-gatherers from Motala in Sweden, the derived alleles in genes SLC24A5 and SLC45A2 were present at high frequencies in their sample. Interestingly, this study found the combination of light skin and blue eyes in Scandinavian hunter-gatherers. WHGs had dark skin and many are predicted to have had blue eyes, while the EHGs had light to intermediate colored skin and largely boasted brown eyes. So it looks like as mixing between the WHGs and EHGs took place in Scandinavia, the resulting mixed population took the blue eyes of the WHGs and the light to intermediate skin of the EHGs. One reason these traits were strongly selected for is likely adaptation to the northerly climate. That the derived alleles were found in Eastern hunter-gatherers means that migrations out of West Asia, other than the Neolithic expansion, also came to contribute to the lightening of skin pigmentation in some parts of Europe. Why am I suggesting more migrations out of West Asia? Using data from modern and ancient DNA studies, researchers have proposed that the derived alleles in genes SLC24A5 and SLC45A2 each had a single origin and that that place of origin for both was likely West Asia. The derived allele in gene SLC24A5 is specifically thought to have arisen between 22,000 and 28,000 years ago. Given the single origin of these derived alleles, Eastern European hunter-gatherers must have obtained these alleles following migrations out of West Asia that were not associated with the Neolithic expansion. The story so far is best summed up in this paper. I quote, The SNP RS1426654 within the SLC24A5 gene has the single largest effect on skin lightening of all gene variants identified to date. It is located within a large haplotype block C11 shared by virtually all carriers of this allele, including ancient Scandinavian, Eastern, and Caucasus hunter-gatherers, suggesting that this light skin variant derives from a single carrier who lived 22,000 to 28,000 years ago in the Middle East. Similarly, haplotype analysis of the SNP RS16891982 of the SLC45A2 gene concluded that this skin lightening mutation occurred only once in the ancestry of the Caucasians. Migrations across the Caucasus and Eastern Europe would have brought both alleles to Scandinavia, in contrast to Southern and Central Europe, where they were introduced by farmers from Western Anatolia expanding 8,500 to 5,000 years ago. End quote. This graph from the same study also summarizes the overall trends well visually. We see dark skin in Western hunter-gatherers and light to intermediate skin in Eastern hunter-gatherers, Neolithic Anatolian farmers, and steppe herders, the latter of which were a mixture of Eastern hunter-gatherers and Caucasus hunter-gatherers, and whose migrations from the Pontic-Caspian steppe westward into Europe during the early Bronze Age probably contributed to the increase in the frequency of the derived alleles in genes SLC24A5 and SLC45A2. You can see with the arrows where these populations migrated to and who they mixed with. It's clear that the story of how Europeans became white is more complex than we originally thought, but definitely no less interesting. Based on current evidence, it looks like the important genes for lighter skin pigmentation that we see in modern-day Europeans actually arose in West Asia, and so they would have had to have been brought to Europe through migrations. But the migratory routes and groups bringing these genes were different. Hunter-gatherers in the easternmost parts of Europe appear to have shifted towards lighter skin first. They then introduced these genes to Scandinavia, where they mixed with the darker-skinned Western hunter-gatherers. But for most of the European continent, the first wave of white skin genes came with migrating Anatolian farmers. Now we have new questions to ask. If the shift towards lighter skin pigmentation was initiated in West Asia and not in Europe, why? And even though it doesn't look like the Europeans became white in Europe as an adaptation to regions with lower UV radiation, this doesn't mean that the idea should be outright rejected. After the alleles associated with light skin color were introduced, it's likely that lower UV radiation was one of the reasons why these alleles were strongly selected for amongst European populations, especially in places like Scandinavia, where the lightening of skin pigmentation appears to have taken place sooner than in most other places in Europe. And this is the picture based on the ancient DNA evidence we have at the moment. Of course, there's always a chance that another groundbreaking discovery will be made that will shift the story in another direction. We've already discovered that a late Mesolithic individual from Spain carried one copy of the derived allele in gene SLC24A5, and in a study on Mesolithic France, although most individuals were predicted to have had dark skin, two of the individuals in the sample were predicted to have had pale or intermediate skin. Now, I don't doubt the impact that the Neolithic Anatolian farmers had in shifting the trend towards lighter skin in these parts of Europe, because the derived alleles for genes SLC24A5 and SLC45A2 were fixed or at high frequency in their population and rare in the Western hunter-gatherer populations. But since we are finding the derived alleles at very low frequencies in Western European hunter-gatherers, it's possible that there are regional nuances within the overall picture that we can only come to understand as more samples get tested. Before wrapping this episode up, I want to go over the study that you've probably seen making headlines lately, that dark skin was dominant in Europe until 3,000 years ago. It sounds interesting, but the first thing I want to note is that this is based off of a study that has not been peer-reviewed yet. It was uploaded as a pre-print onto the bioRxiv. This is a pre-print server where some researchers publicly share their work before it gets picked up by a journal. There are some important points that need to be addressed. The study focused entirely on pre-published genomes. They didn't sequence any new ones. And we know from pre-published genomes that the two major alleles responsible for white skin in Europeans were fixed or at very high frequency in the Neolithic Anatolian population. So why did the method used in the new pre-print predict a dark skin phenotype for this population? We have many studies attributing the introduction of the two main genes associated with white skin in many parts of Europe to the Anatolian farmers. Here are just a few examples. This paper tells us, I quote, In admixed Neolithic individuals, we found excess Neolithic farmer ancestry at the pigmentation locus SLC24A5. Previous studies also found evidence of natural selection at SLC24A5 in European populations and showed that the allele was almost fixed in Anatolian farmers and was introduced into Western Europe during the Neolithic, but our study now further demonstrates that subsequent selection during the Middle Neolithic was rapid enough to result in the end quote. Here is a paper I mentioned previously telling us, similarly, the derived allele of SLC24A5 that is the other major determinant of light skin pigmentation in modern Europe appears fixed in the Anatolian Neolithic, suggesting that this rapid increase in frequency around 0.9 in early Neolithic Europe was mostly due to migration, end quote. In the supplementary materials to this study on Scandinavian hunter-gatherers, early Neolithic farmers from Anatolia and Europe are also mentioned, where it states, I quote, While both pigmentation alleles were observed at RS16891982, although the derived allele in much higher proportion, virtually only the light skin allele was observed at RS1426654, end quote. Sure, there may be other genes involved in skin pigmentation, but they are not as important as SLC24A5 and SLC45A2. Carrying the derived alleles in these genes should not result in a dark skin phenotype. Predicting otherwise is a very unexpected result. I actually found a discrepancy in the preprint itself with regards to this. The researchers do say in their discussion that the incoming Anatolian farmers impacted on the phenotypic traits of the Europeans, including skin color, but this doesn't match the predicted skin color they presented in the study. We see the paper state here that people in Turkey during the Neolithic had a dark skin phenotype. We can also see the color they predicted here. But then the paper later states, I quote, Among the episodes of gene flow documented in Western Eurasian prehistory, the spread of early Neolithic farmers from Anatolia is known to have profoundly changed the genetic makeup of populations. Immigrating farmers had two evolutionary advantages over their hunting-gathering counterparts. By farming an animal herding, they increased the amount of available food and had a end quote. So first the researchers predict that Neolithic Anatolian farmers had dark skin, and then they say they had a skin phenotype that's fit for lower levels of UV radiation, which is essentially lighter skin. It's known that lighter skin is adapted to low UV regions because it allows for greater penetration of UV radiation and so prevents vitamin D deficiencies in these areas. It is even stated at the beginning of the preprint that the migrating Neolithic Anatolian farmers played a big role in initiating the shift towards lighter skin in Europe, is also emphasized in the press as one of the main points of the study. But when you actually read the paper, the predicted skin color and conclusion contradict each other. This discrepancy needs to be addressed before publication because one of these statements has to be wrong, and if it's the method that's wrongly predicted the skin color of the Neolithic Anatolians, which seems more likely at the moment given the results of previous studies on the phenotype of Neolithic Anatolians, but now this needs to be looked into, then this could mean that the method is flawed and can ultimately affect the paper's final conclusion. Also some of the regions were represented by very small sample sizes. 348 samples in total might seem like a lot, but it's not when you see how the genomes are distributed by location and period. In this graph you can see how many regions in different periods are represented by just one sample. This is hardly enough for estimating the average phenotype of a population at a specific point in time, let alone tracing changes in phenotypes over thousands of years. I'm a seeker of truth, so if dark skin was dominant in Europe until 3,000 years ago, I want to know. But until more samples get tested and the discrepancy with the Neolithic Anatolian phenotype is investigated, this claim should be seen as a possibility, not a fact. How the genes associated with white skin came to be fixed in Europeans is clearly a complex story and different regions followed their own local trajectories. We've already seen how Scandinavia was different from Southern, Western, and Central Europe, but these trajectories can only be established in greater detail once we have sampled many, many more ancient genomes. That's it for this episode. If you have any thoughts or questions, let me know in the comments. Subscribe for more cool content by your go-to informant on everything archaeology, Madam Archaeologist.