Blue eyes

Scientists are now able to demonstrate how a mutation in a single base pair (one of 3.2 billion) has resulted in a change in a known trait, such as blue eyes. A mutation in a single base pair is called a single nucleotide polymorphism (SNP). In 2008, Eiberg et al. were able to show how the mutation of a single base pair (from A to G) within the HERC2 regulatory gene inhibits the activity of the nearby OCA2 gene. Essentially, the mutation turns off the biosynthesis of melanin (the brown pigment that results in brown eyes) in the iris. People who have two copies of the mutation (the rs12913832*G allele) have blue eyes with no brown pigmentation. People who have two copies of the ancestral allele (the rs12913832*A allele) have brown eyes. The team’s research suggests that this is a “founder mutation,” meaning that everyone who has the *G allele at this position on chromosome 15q can trace their ancestry back to a single person in whom the mutation first occurred, who most likely lived in the northwest part of the Black Sea region 6,000 to 10,000 years ago (Eiberg et al., 2008). This type of work is pioneering, because it follows a specific single-point mutation (A to G), through its mechanisms of action, to show how it produces a known human trait. This is an example of the type of random mutation that has recently introduced genetic diversity into a population. However, scientists believe that the rapid increase in blue eyes in Europeans was due to sexual selection, rather than natural selection. They suggest that blue eyes may have been seen as novel, and therefore desirable, in mate selection (Eiberg et al., 2008). In contrast, let us look at how mutations affect traits that appear to have been beneficial to successful reproduction, and therefore were subject to natural selection.

Hemoglobin S

Hemoglobin S is an abnormal red blood cell which causes sickle-cell anemia. Sickle-cell anemia, a serious, often fatal disease, is also caused by a single-point mutation, from an A to a T. To get sickle-cell anemia, a person must inherit two copies of the mutation (one from each parent). The mutation alters one of the amino acids in the hemoglobin protein which results in abnormal red blood cells, that, under certain conditions, change to a long, sickled shape which causes anemia, pain, and tissue damage (see DNA Learning Center, Cold Spring Harbor Laboratory, n.d.). Prior to modern medicine, sickle-cell anemia was invariably fatal. This should have resulted in people with sickle-cell anemia dying before they could reproduce, thus selecting against the mutation over the generations.

However, instead of being selected against, the sickle-cell mutation is very common in areas with malaria, such as India, the Mediterranean, and sub-Saharan Africa, where 10% to 40% of the population carries it. The reason is that for people who have only one copy of the sickle-cell mutation, the mutation confers tolerance to malaria—people still get malaria, but they do not die from it (Ferreira et al., 2011). Studies of the distribution of Hemoglobin S and the distribution of malaria show that the higher the prevalence of malaria, the higher the percentage of people with hemoglobin S. Where there is no malaria, 0% of the population has Hemoglobin S. Brace (1996)concluded that the frequency of the sickle-cell mutation in populations “is controlled by the intensity of infestation with falciparum malaria”—the parasite that causes malaria (p. 113). This is a classic example of natural selection. Over time, geography and the environment (areas conducive to malaria) interacted with a random mutation (Hemoglobin S) that conferred a reproductive advantage in people who carried one copy of the gene.

Researchers believe that the Hemoglobin S mutation occurred independently three times in Africa and once in either the Arabian Peninsula or central India between 70,000 and 150,000 years ago (Desai & Dhanani, 2004). While such mutations are rare, it is likely that over the last 100,000 years or so, this mutation also occurred a few times in areas of the world where there was no malaria. However, as the sickle-cell mutation conferred no advantage in the interaction between people and their (nonmalarial) environment, it died out.

Lactase persistence

All human infants are able to digest lactose, the primary carbohydrate in milk, because their intestines secrete lactase. In most human populations, the ability to secrete this enzyme goes away after weaning. This is the ancestral condition for humans (and all mammals; Itan, Powell, Beaumont, Burger, & Thomas, 2009). Accordingly, in East Asian and Southeast Asian populations, indigenous North and South American populations, and in large parts of Africa, 0% of the adult population has the lactase persistence trait. This makes sense, because, as long as human cultures were based on hunting and gathering, lactase persistence was unnecessary. However, as animals began to be domesticated, nonhuman milk (goat, cow, horse, camel, reindeer) became readily available as a source of food to pastoralists following their herds. Consequently, there was strong natural selection for lactase persistence in herding societies, and so dairying and lactase persistence co-evolved.

In Europeans, lactase persistence is due to a single C to T mutation at −13910*T, upstream from the lactase gene (Itan et al., 2009). The 13910*T allele accounts for virtually all the variance in lactase persistence frequency in Europe. It ranges in frequency from 73% to 95% in the British Isles and Scandinavia, 56% to 67% in Central and Western Europe, and 6% to 36% in Eastern and Southern Europe. Itan et al. (2009) date the origin of dairying and the co-evolving −13910*T lactase persistence mutation to about 7,500 years ago. Moreover, they conclude that this is a “recent mutation event” (Ingram, Mulcare, Itan, Thomas, & Swallow, 2009, p. 586); Bersaglieri et al. (2004) concluded that the signals of recent selection for the −13910*T lactase persistence mutation they observed were among the “strongest yet seen for any gene in the genome” (p. 1111).

However, genotype frequency comparisons have shown that the European mutation is not found in African and Middle Eastern dairying populations. Instead, different lactase persistence-associated mutations occurring in the same region of the genome have been identified. For example, Bedouins in Jordan and Saudi Arabia have the −13915*G mutation, while pastoralists in Kenya and Tanzania carry the −13907*G variant and the −14010*C variant (Ingram et al., 2009; Tishkoff et al., 2008). This is an example of convergent evolution. It is likely that over the last 10,000 years, mutations in the control area upstream of the Lactase (LCT) gene also occurred a few times in areas of the world where there was no dairying. However, as those mutations conferred no advantage to nondairying populations, they died out.

Skin color

Geography and environment also influenced the genetic structures that resulted in varied skin colors among different human populations. Unlike blue eyes or Hemoglobin S, skin color is a polygenic trait—one whose phenotype is influenced by multiple genes, each one having a relatively small effect (Sturm, 2009). Observable skin color is determined by the amount and distribution of two types of melanin pigment granules within skin cells—a red-yellow form known as pheomelanin and a black-brown form known as eumelanin (Barsh, 2003; McEvoy, Beleza, & Shriver, 2006). For example, Beaumont et al. (2007) showed that in Europeans, red hair and fair skin with freckles are associated with loss-of-function alleles of the Melanocortin Receptor (MC1R) gene. These alleles have been shown to decrease the ability of the skin to produce eumelanin (the black-brown melanin) in in vitro functional studies (Beaumont et al., 2007). Skin color is measured with a hand-held instrument using skin reflectance spectroscopy (Shriver & Parra, 2000).

Anthropologists assume that the “earliest members of the hominid lineage probably had a mostly unpigmented or lightly pigmented integument [skin] covered with dark black hair, similar to that of the modern chimpanzee” (Jablonski & Chaplin, 2000, p. 57). They also assume that after our lineage diverged from the lineage of modern chimpanzees, “coincident with the loss of fur . . . there was strong selection for skin darkening” to protect the skin from the ultraviolet rays of the sun (McEvoy et al., 2006, p. R176). Both of these changes occurred roughly 2 to 5 million years ago (Fuentes, 2012). Therefore, anthropologists assume that the process of differentiation of skin color started from a population of Homo sapiens with dark skin.

On a global level, skin color in indigenous populations in the Old World tends to be darkest in the tropics. The farther north or south of the tropics, the lighter skin color becomes. The reason appears to be related to the amount of sun light, which is much greater at the equator than at the poles. Melanin absorbs and scatters light (Jablonski & Chaplin, 2000). Jablonski and Chaplin (2013)found that “86% (r = 0.927) of the variation in human skin reflectance” can be accounted for by autumn levels of ultraviolet radiation (p. 672). The reason is that humans need enough ultraviolet light to make Vitamin D, but too much ultraviolet light results in sunburn and the destruction of folate, which is necessary for healthy reproduction. The amount of pigment in human skin enables populations to balance these two needs. Therefore, on one hand, in the high latitudes like Sweden or Siberia, there was “positive directional selection” for less melanin in the skin so that people in the high latitudes could make more Vitamin D (Jablonski & Chaplin, 2000, p. 673). On the other hand, near the equator, in places like Papua New Guinea, Southern India, or the Congo, there was positive selection for more melanin in the skin, so that people could be protected from too much sun and the destruction of folate (Jablonski & Chaplin, 2000).

Selective sweeps

With regard to how much effect natural selection has had on Homo sapiens in the last 50,000 years or so, at the turn of the 21st century, there were still two camps. Those in the first camp argued that cultural evolution, such as adapting to extreme cold by learning to make thick clothes out of fur skins, had lessened the effects of natural selection. Those in the second camp argued that Homo sapiens continued to benefit from positive natural selection during the last 50,000 years (Hawks, Wang, Cochran, Harpending, & Moyzis, 2007). Rogers (2011) states that “as recently as the year 2000, it was possible for Stephen Jay Gould to argue that ‘natural selection has almost become irrelevant in human evolution. There’s been no biological change in humans in 40,000 or 50,000 years’” (p. 93). Gould (1980) had previously argued that “cultural evolution is our primary innovation. It works by the transmission of skills, knowledge and behavior through learning—a cultural inheritance of acquired characters” (p. 137). Rogers states that in 2000, when Gould gave this interview, there was no way to tell which position was correct. However, had Gould lived 10 years longer, advances in genetics would have made Gould’s position “untenable,” and Rogers believes Gould would have changed his mind. Today, just 15 years later, there is plenty of evidence that natural selection is still affecting humans. Indeed, Hawks et al. (2007) argue that the rapidly expanding world population since the domestication of plants and animals (resulting in shifts from hunter-gatherer cultures to pastoralism and agriculture, about 10,000 years ago) has resulted in a 100-fold increase in mutations, thus increasing the amount of positive natural selection that is occurring.

The way scientists evaluate genomes for evidence of positive natural selection is to look for selective sweeps. Long (2003) states thata favorable allele may occasionally rise rapidly and sweep through the population replacing all other alleles at that locus. This mode of selection is often referred to as positive selection. One consequence of a selective sweep is that the level of background genetic variation in the vicinity of the favored allele is reduced. (p. 7)

Another way to explain this is that DNA tends to be inherited in chunks or blocks of base pairs grouped around a functional SNP on a chromosome. Functional mutations (SNPs) that occurred earlier in human history tend to have shorter blocks or chunks of unchanged DNA surrounding them, because they had many more generations for recombination to shorten the length of the unchanged stretches of DNA attached to them. However, functional mutations (SNPs) that have occurred relatively recently in human history have much longer chunks of unchanged DNA surrounding them. Several new statistical tests have been developed to evaluate the presence of selective sweeps and their population frequency (Huff, Harpending, & Rogers, 2010; Pickrell et al., 2009; Sabeti et al., 2002; Voight, Kudaravalli, Wen, & Pritchard, 2006). These tests are particularly effective at identifying selective sweeps at moderate frequency (~50%–80%), and at high frequency (>80% to fixation) within a population (Pickrell et al., 2009). For example, McEvoy et al. (2006) identified “a 150 kb region surrounding the SLC24A5 gene” that “shows a large drop in heterozygosity in Europeans.” This means that the 150 kb region around the gene is unchanged and thus shows evidence of recent selection (p. R177).

McEvoy et al. (2006) were looking for genes within regional populations that affected skin pigmentation. The team studied DNA from three geographically distinct populations from West Africa, East Asia, and Northern Europe. Based on differences among these three groups, they developed a four-step evolutionary model for skin pigmentation. First, they found evidence of mutations in two genes (MITF and EDN3) that occurred in the ancestral human population, prior to any splits between the groups. Second, they found evidence of a split between the African group and the ancestors of Asians and Europeans, who carried the ASIP and BNC2 genes. Third, they identified five genes that showed positive selection only in the East Asian population. Fourth, they identified five different genes that showed positive selection only in the Northern European population. All the genes identified were involved in the pathway for the biosynthesis of melanin. Both Northern Europe and East Asia are at much higher latitudes than Africa, so pale skin with less melanin benefitted humans in both environments. But since mutations are random, nature could only positively select the mutations related to the biosynthesis of melanin that actually occurred in the Northern European and East Asian populations. Thus, the genetic mechanisms that resulted in the pale skin color of East Asians and Europeans are different; this is an example of convergent evolution (McEvoy et al., 2006). The McEvoy et al. (2006) study shows clear evidence for the operation of natural selection with regard to skin color. As a result of studies like this one, molecular biologists have concluded that “skin colour as a selectable trait has likely occurred multiple times at diverse geographical sites around the globe” (Sturm, 2009, p. 13).

MtDNA

In 2001, Sykes summarized 25 years of work on mitochondrial DNA (mtDNA), which has the advantage of being passed intact, directly from mother to child. As mtDNA is found within the mitochondria, small “energy factories” within cells, it is not involved in sexual reproduction. When a mutation, such as a change from a G to an A, occurs in the DNA of a mitochrondrion in a woman’s germ cell, if that germ cell is passed on to a daughter, the mutation will continue in the daughter’s DNA. In other words, your mtDNA is an exact copy of your mother’s mtDNA, which is an exact copy of her mother’s DNA, which is an exact copy of her mother’s DNA, and so on. A mutation occurs in mtDNA about once every 10,000 years, which enables scientists to estimate how long ago two lines shared a common maternal ancestor (Sykes, 2001).

After analyzing thousands of samples from all over Europe, Sykes (2001) found that nearly all native Europeans could trace their maternal ancestry to one of just seven different women, who lived between 45,000 and 10,000 years ago. Remarkably, 47% of Europeans have mtDNA that traces back to one woman, who lived about 20,000 years ago, most likely in the area of the Rhone River in France. Sykes explained that each of these women was certainly not the only woman living in her hunting and gathering band at the time. But if a woman has no children or only boys, her mtDNA line dies out.

Using mtDNA sequences from all over the world, Sykes (2001) identified 33 women who are the mothers of all people living today, and 13 of them originated in Africa. His findings suggest that a small group from one of the African mtDNA clans left Africa to populate the rest of the world. The 20 women whose mtDNA clans are found in the rest of the world are all descendants of that single woman whose descendants left Africa. Sykes writes,Incredibly, even though the African clans are easily the most ancient in the world, we are still able to reconstruct the genetic relationships among them . . . the ancestors of the ancestors . . . One by one the clans converge until there is only one ancestor, the mother of all of Africa and of the rest of the world . . . “Mitochondrial Eve.” (Sykes, 2001, p. 276)

Sykes (2001) concluded that “we are all direct maternal descendants” of Mitochondrial Eve, who lived about 150,000 years ago in Africa (pp. 276-277).

Y-chromosome DNA

In 2003, Wells released a book on his research team’s findings using DNA from the Y-chromosome. The Y-chromosome is passed down from fathers to sons just as mtDNA is passed down from mothers to daughters. As with maternal lineages, if a man has no children or no sons, his Y-chromosome lineage is lost. The Wells (2003) team studied a “worldwide sample of men, from dozens of populations on every continent” (p. 53).

Wells (2003) concluded that all modern humans originated in Africa. He also concluded that all men alive today can trace their paternal lineage back to one man—“Y-chromosome Adam”—who lived in Africa about 60,000 years ago. While there were other men alive then, their lineages have died out. He noted that the difference between the estimated age of “Mitochondrial Eve” and “Y-chromosome Adam” (who were real people) means that more female lineages survived, unbroken, into the present day. As a result, scientists could trace the maternal line much further back into prehistory (150,000 years) than they could trace the paternal line of all people living today. Wells explained that the greater the diversity, as shown by the accumulation of additional genetic markers, the older the lineage. As a result, he said that “you are more likely to sample extremely divergent genetic lineages within a single African village than you are in the whole of the rest of the world” (Wells, 2003, p. 39). He emphasized that the real surprise is that as all modern humans lived in Africa until at least 60,000 years ago, our species has only had 60,000 years to populate the globe (Wells, 2003).

Wells (2003) divided the descendants of men who left Africa into a genealogical tree with 11 lineages. Each genetic marker represents a single-point mutation (SNP) at a specific place in the genome. First, genetic evidence suggests that a small band with the marker M168 migrated out of Africa along the coasts of the Arabian Peninsula and India, through Indonesia, and reached Australia very early, between 60,000 and 50,000 years ago. This very early migration into Australia is also supported by Rasmussen et al. (2011). Second, a group bearing the marker M89 moved out of northeastern Africa into the Middle East 45,000 years ago. From there, the M89 group split into two groups. One group that developed the marker M9 went into Asia about 40,000 years ago. The Asian (M9) group split three ways: into Central Asia (M45), 35,000 years ago; into India (M20), 30,000 years ago; and into China (M122), 10,000 years ago. The Central Asian (M45) group split into two groups: toward Europe (M173), 30,000 years ago and toward Siberia (M242), 20,000 years ago. Finally, the Siberian group (M242) went on to populate North and South America (M3), about 10,000 years ago.

Whole genomes

While the earlier studies compared frequencies of gene alleles for blood proteins, or genetic markers in parts of the mtDNA or the Y-chromosome, second generation sequencing now enables teams to compare whole genomes. For example, Li et al. (2008) analyzed 650,000 base pairs that had been previously identified as having mutations (SNPs), in the genomes of 1,064 individuals from 51 populations. The genomes of members of 51 populations were compared with each other and with chimpanzees. (The chimpanzee genomes were assumed to represent “ancestral allele frequencies”—the frequencies of alleles before humans and chimpanzees began to diverge from a common ancestor.) Li et al. constructed a phylogenetic tree diagram showing the relationships among the populations. The least different from the ancestral allele frequencies were the San (Bushmen) and other sub-Saharan African populations, followed by North African populations, European, Middle Eastern, Central Asian, East Asian, Amerindian, and Oceanian populations. (Australian aboriginal populations were not studied.) They concluded that their findings were consistent with a “serial founder effect, a scenario in which population expansion involves successive migration of a small fraction of individuals out of the previous location, starting from a single origin in sub-Saharan Africa” (Li et al., 2008, p. 1103). Their study fully supports the recent African origin model.

Second-generation population history studies

Recent work with second generation sequencing and new statistical modeling methods adds much finer detail on migration and admixture of populations. For example, in Europe, Stewart and Stringer (2012) show a cyclical pattern of interglacial expansion of Homo sapiens out of Africa and retreat back to Africa or warmer refuges during ice ages. They showed that populations of Homo sapiens moved out of Africa during warm interglacial periods, then retreated from Europe and Asia into southern areas like Spain, Italy, the Balkans, and the Caspian Sea during the last ice age (about 27,000-11,700 years ago), and then expanded again to repopulate Europe after the climate warmed. Behar et al. (2006) supported these findings by showing that the Basques are a remnant of the Paleolithic hunter-gatherer population that retreated into the area around the Bay of Biscayne during the last ice age. The hunter gatherers then expanded to resettle Central and Western Europe after the end of the last ice age. Using genome-wide data from 94 ancient Europeans who lived 8,000 to 3,000 years ago, Haak et al. (2015) modeled the genetic admixture among the indigenous hunter gatherers, Neolithic farmers expanding into Europe from Anatolia (starting about 8,000 years ago), and the Yamnaya or kurgan culture—a pastoralist culture from the steppes north of the Black and Caspian Seas. The Yamnaya culture had horses, wheeled vehicles, and carried the gene for lactase persistence. Haak et al.’s study showed a massive migration of the Yamnaya people into central Europe about 4,500 years ago, one that could have spread an early form of the Indo-European language, which is the basis of most of today’s European languages.

Much of northern India also speaks Indo-European languages. In the first major work on population history in India, Reich, Thangaraj, Patterson, Price, and Singh (2009) used second generation sequencing technology and statistical modeling to compare the genomes of 25 diverse groups from India to the genomes of people from Western Asia, Europe, and the Andaman Islands (who are thought to be closest to the ancestral populations of southern India). They found that the “Ancestral North Indians” (ANI) form a clade with Europeans (meaning that they are closely related), but are genetically distinct from the “Ancestral South Indians” (ASI), who form a clade with the Andaman Islanders (Reich et al., 2009). Using admixture modeling, they found that different Indian groups inherited different proportions of ancestry from the ANI who are related to western Eurasians, and the ASI, who are related to the Andaman Islanders (p. 491). This parallels the language data—Northern Indian groups speak Indo-European languages, while Southern Indian groups speak Dravidian languages. However, Reich’s team called for further investigation to date the beginning of the admixture of Northern Indian and Southern Indian populations.

Thus, the work of teams examining blood group allele frequencies, mtDNA, Y-chromosome DNA, and whole genomes, all shows the same pattern. By comparing the DNA of people from around the world, scientists can determine where their ancestors came from and approximately when migrations took place. And in doing so, they have found that the ancestors of all humans living today came from Africa (Veeramah & Hammer, 2014). We are all Africans, no matter the color of our skin.