Beyond Ötzi: European Evolutionary History and its Relevance to Diet. Part II

In previous posts, I described how Otzi was (at least in large part) a genetic descendant of Middle Eastern agriculturalists, rather than being purely descended from local hunter-gatherers who adopted agriculture in situ.  I also reviewed evidence showing that modern Europeans are a genetic mixture of local European hunter-gatherers, incoming agricultural populations from the Middle East, neanderthals, and perhaps other groups.  In this post, I'll describe the evidence for rapid human evolution since the end of the Paleolithic period, and research indicating that some of these changes are adaptations to the Neolithic (agricultural/horticultural/pastoral) diet.

Humans have Evolved Significantly Since the End of the Paleolithic

Evolution by natural selection leaves a distinct signature in the genome, and geneticists can detect this signature tens of thousands of years after the fact by comparing many genomes to one another.  A landmark paper published in 2007 by Dr. John Hawks and colleagues showed that humans have been undergoing "extraordinarily rapid recent genetic evolution" over the last 40,000 years (1).  Furthermore:

...Neolithic and later periods would have experienced a rate of adaptive evolution [more than] 100 times higher than characterized most of human evolution.

"Adaptive evolution" refers to natural selection (e.g., evolving the ability to digest the milk sugar lactose in adulthood), rather than genetic change by random drift.  From a later paper published in Science (2):

In general, we find that recent adaptation is strikingly pervasive in the human genome, with as much as 10% of the genome affected by linkage to a selective sweep.

In other words, the authors estimate that the rate of adaptive evolution since the development of agriculture has been more than 100 times faster than during most of the Paleolithic period, and that as much as 10 percent of the genome shows evidence of recent evolution in European-Americans, African-Americans and Chinese (although some of this predates agriculture).  This suggests that we may have evolved as much over the last 10,000 years as we did over the previous 1 million year interval!  Just to give you an idea of the significance of that genetic distance, one million years ago the closest thing to a human was Homo heidelbergensis, a burly, thick-browed hominid that used spears to take down large prey (3). 

This rapid rate of genetic change was driven by at least two factors (Cochran and Harpending. The 10,000 Year Explosion. 2009):

1. A major change in environment, and thus a change in selective pressures on the genome.
2. A large increase in population.  The higher the population base, the higher the likelihood that adaptive mutations will arise by chance.

Some of these Adaptations are to Diet

Looking at the archaeological record of early agriculturalists, it's clear that they experienced severe physical stress, including stunting and skeletal abnormalities that indicate nutritional and infectious stress (Cohen and Crane-Kramer. Ancient Health. 2007; Cohen. Health and the Rise of Civilization. 1991).  Looking at the human genome, it's clear that it has changed substantially since the adoption of agriculture.  If we believe that the Neolithic grain-focused diet contributed to the ill health of early agriculturalists, it must have exerted a significant selective pressure on the genome, and therefore it is an inevitable conclusion that some of the genetic changes that have occurred in the last 10,000 years in populations eating a Neolithic diet are adaptations to this diet. 

To give you an idea of how fast genetic adaptations to diet can arise and spread, let's return to the example of lactase persistence (4).  Normally, humans lose the ability to digest the milk sugar lactose after infancy, rendering them lactose intolerant in adulthood.  Certain genetic mutations break the switch that turns lactose production off after childhood, allowing continued lactose digestion in adulthood.  These mutations arose independently multiple times in human history, in Europe, Africa and the Middle East (5).  They appeared shortly after the acquisition of dairy as food.  In each case, a mutation arose in a single individual and rapidly spread throughout the population.  The most common mutation in Europeans arose in one person ~7,500 years ago, and today is present in 80 percent of Europeans and people of European descent.  This illustrates how rapidly evolution and dietary adaptation can occur, although there's an even faster means of evolution that I'll get to later.

Another example is salivary amylase copy number.  Amylase is an enzyme that digests starch into glucose, and salivary amylase is a version of the enzyme that's contained in saliva.  Different people produce different amounts of salivary amylase, and this corresponds to the number of copies of the salivary amylase gene they carry in their genome (6).  Populations that have historically eaten a high-starch diet tend to have more copies, and their genomes show evidence of recent natural selection favoring high copy number due to gene duplications (7).  European-Americans, Japanese, and Hadza hunter-gatherers tend to have high copy number, suggesting adaptation to regular starch consumption, while several traditionally low-starch hunter-gatherers and pastoralists including Biaka, Mbuti, Datog, and Yakut tend to have low copy number.

It's worth noting that there's a lot of variability in the European-American and Japanese samples, with copy number ranging from 2-15 in the European-American sample.  Most people cluster in the 4-10 copy range.  Salivary amylase copy number correlates with glucose tolerance-- more copies is associated with better glucose handling-- but the mechanism remains unknown (8).

Chimpanzees only carry two copies (one on each chromosome), less than any known human population, consistent with the fact that they eat very little starch (despite getting a substantial amount of carbohydrate from fruit sugars).  The increase in salivary amylase copy number presumably occurred after humans diverged from chimpanzees, and probably reflects increasing reliance on starch as a food source during human evolution.

Mutation and selection is one path to adaptation, but there's actually another much faster path.  Each human contains essentially the same set of genes as every other human, however, different people often carry different versions of the same genes.  These different versions are called "alleles" of a gene.  Eye color, skin color, hair color, hair texture and blood type are all common examples of traits where different alleles of the same genes create different physical outcomes.  In any population, there's a pool of common alleles, each present at a different frequency.  Changing the population frequency of pre-existing common alleles is the most rapid form of natural selection because it doesn't rely on new mutations arising spontaneously.  Allele frequencies can change dramatically in as little as one generation if there's a strong selective pressure.  For example, if there were a global epidemic of a deadly virus that only infected people with blood type A (or, more likely, people with a particular immune system-related allele), the frequency of that allele could greatly decline within only a few years.  Mutations within a gene result in a new allele, which can then be subject to natural selection, as in the case of lactase persistence, but waiting for the right mutation to occur takes a lot longer than selecting from a pool of alleles that are already present in a population.

As one would expect, changes in allele frequency (even in the absence of new mutations) are one of the genetic forces that has permitted the rapid adaptation of humans to unique environments throughout the world (9).  For example, there are specific allele patterns related to digestion and metabolism that associate with populations that have ancestral dietary patterns dependent on grains or tubers:

A SNP (rs162036) that is strongly correlated with a diet containing mainly the folate-poor roots and tubers lies within the methionine synthase reductase (MTRR) gene, which activates the folate metabolism enzyme methionine synthase and is implicated in spina bifida (22). Perhaps the most interesting signal comes from a SNP (rs4751995) in pancreatic lipase-related protein 2 (PLRP2) that results in premature truncation of the protein and is strongly correlated with the use of cereals as the main dietary component (Fig. 2). Several lines of evidence support an important role for this protein in a plant-based diet. First, unlike other pancreatic lipases, PLRP2 hydrolyzes galactolipids, the main triglyceride component in plants (23, 24). Second, a comparative analysis found that the PLRP2 protein is found in nonruminant herbivore and omnivore pancreases but not in the pancreases of carnivores or ruminants (25). Our results show that the truncated protein is more common in populations that rely primarily on cereals, consistent with the hypothesis that this variant results in a more active enzyme (26, 27) and represents an adaptation to a specialized diet.

These patterns reveal the traces of rapid changes in allele frequency that presumably underlie dietary adaptations.


Patterns of Genetic Change

I just described several examples of rapid, recent human evolution to a change in diet.  If we take a broader look at the types of genes that have undergone recent selection, they cluster predominantly into several categories (10, 11, 12):

• Immunity
• Skin pigmentation
• Brain development/function
• Food digestion/metabolism
• Sensory perception (including smell)
• Muscle-related genes
• Assorted cell signaling pathways

Some of this is expected, knowing that the late Paleolithic and Neolithic environment brought major changes in pathogen exposure, latitude (in some populations), physical demands, social and cognitive demands, and diet.  Each genetically distinct population (e.g., European, African, Asian) has a unique suite of recent adaptations that helped it thrive in its own ancestral environment, although many of the genetic processes that have occurred are broadly similar between populations. 

We have a broad outline of the kinds of processes that have been subject to recent natural selection in humans, and in some cases the location and function of the selected gene variants are known.  However, the truth is that in most cases where we know natural selection has occurred, we don't know exactly where the variant in question resides, what it does, and often we don't even know what gene it's in.  The point is that the large majority of recent genetic adaptations in the human genome remain totally uncharacterized, and judging by the patterns observed among the mutations we do understand, a number of them are probably adaptations to the Neolithic diet that remain to be explored.

I think it's clear at this point that modern Europeans, and many other populations with long-term ancestral Neolithic diets, carry meaningful genetic adaptations to the Neolithic diet.  However, there's a major caveat here.  The presence of adaptation does not imply that we're completely adapted to the Neolithic diet-- we may only be partially there.  This is a concept I'll explore in the next post.