Traits (or the DNA sequences that code for them) that differ in expression between populations and individuals are called polymorphisms, and they’re the main focus of human variation studies. A genetic trait is polymorphic if the locus that governs it has two or more alleles. (See Chapter 4 for a discussion of the ABO blood group system, which is governed by three alleles at one locus.) A locus can consist of hundreds of nucleotides or just one nucleotide.
Understanding polymorphisms requires evolutionary explanations, and geneticists use polymorphisms as a principal tool to understand evolutionary processes in modern populations. By using these polymorphisms to compare gene frequencies between different populations, we can begin to reconstruct the evolutionary events that link human populations with one another. The ABO system is interesting from an anthropological perspective because the frequencies of the A, B, and O alleles vary tremendously among humans. In most groups, A and B are rarely found in frequencies greater than 50 percent, and usually their frequencies are much lower. Still, most human groups are polymorphic for all three alleles, but there are exceptions For example, in native South American Indians, frequencies of the O allele reach 100 percent. Exceptionally high frequencies of O are also found in northern Australia, and some islands off the Australian coast show frequencies exceeding 90 percent. In these populations, the high frequencies of the O allele are probably due to genetic drift (founder effect), although the influence of natural selection can’t be entirely ruled out.
Besides ABO, there are many other red blood cell phenotypes, each under the control of a different genetic locus. These include the well-known Rh blood group as well as the less familiar Duffy and MN blood groups. Some antigens on white blood cells are also polymorphic. Called human leukocyte antigens (HLAs) in humans, these are crucial to the immune response because they allow the body to recognize and resist potentially dangerous infections. But unlike simple polymorphisms, such as ABO (one locus with three alleles) or MN (one locus with only two alleles), the HLA system is governed by perhaps hundreds of alleles at six different loci. Therefore, the HLA system is by far the most polymorphic genetic system known in humans.
Because there are so many HLA alleles, they’re useful in showing patterns of human population diversity. For example, Lapps, Sardinians, and Basques differ in HLA allele frequencies from other European populations, and these data coincide with allele frequency distributions for ABO, MN, and Rh (Fig. 14-3). Founder effect is the most likely explanation for the distinctive genetic patterning in these smaller, traditionally more isolated groups. Likewise, some of the atypical frequencies of HLA alleles characteristic of certain populations in Australia and New Guinea probably result from founder effect. Natural selection has also influenced the evolution of HLA alleles in humans, especially as related to infectious disease. For example, certain HLA antigens appear to be associated with resistance to malaria and hepatitis B and perhaps to HIV as well. And finally, one physiological and evolutionary influence of HLA concerns male fertility. Data suggest that some HLA antigens are found in higher frequencies in infertile males, suggesting that there may be some influence of two or more HLA loci on sperm production and function (van der Ven et al., 2000).
Another well-studied polymorphism is the ability to taste an artificial substance called phenylthiocarbamide (PTC). While many people perceive PTC as extremely bitter, others don’t taste it at all. The mode of inheritance follows a Mendelian pattern, with two alleles (T and t). The ability to taste PTC is a dominant trait, while the inability to taste it is recessive. So, “nontasters” are homozygous (tt) for the recessive allele. The frequency of PTC tasting varies considerably in human populations, and the evolutionary explanation for the patterns of variation isn’t clear. But it’s possible that perceiving substances as bitter could be advantageous, especially in children, because poisonous plants are often bitter. Thus heightened sensitivity to bitter substances increases the likelihood that toxic substances will be avoided.
Polymorphisms at the DNA Level
As a result of the Human Genome Project, we’ve gained remarkable insights into human variation at the DNA level, and molecular biologists have recently discovered many variations in the human genome. For example, there are thousands of DNA segments called copy number variants (CNVs), where DNA segments are repeated, in some cases just a few times and in other cases hundreds of times. These segments vary tremendously from person to person and, in fact, every person has a unique arrangement that defines his or her distinctive “DNA fingerprint.”
Researchers are expanding their approach to map patterns of variation for individual nucleotides. As you know, point mutations have been recognized for some time. But what’s only recently been appreciated is that single-nucleotide changes also frequently occur in non–protein-coding portions of DNA. These point mutations, together with those in coding regions of DNA, are all referred to as single-nucleotide polymorphisms (SNPs). From years of detailed analyses, about 15 million SNPs have been recognized. These are dispersed throughout the human genome (the majority found in noncoding DNA), and they’re extraordinarily variable (Durbin et al., 2010). SNPs are only one of several recent genetic discoveries and indeed, geneticists have gained access to a vast biological “library” that documents the genetic history of our species.
The field of population genetics is taking advantage of these new discoveries. While traditional polymorphic traits, such as ABO, are still being studied, researchers are directing more and more attention to the remarkably variable DNA polymorphisms. These molecular applications are now being widely used to evaluate human variation at a microevolutionary level, and this information provides far more accurate measures of within- and
between-group variation than was previously possible. Besides that, we can now use the vast amount of new data to more fully understand very recent events in human population history, including the many roles of natural selection, genetic drift, gene flow, and mutation. As an example of how far the study of human variation has moved toward a molecularly based approach, more than 95 percent of papers dealing with population variation presented at a recent anthropology conference made use of DNA polymorphisms obtained from populations from all over the world.
The most recent and most comprehensive population data regarding worldwide patterns of variation come from scans of extremely large portions of DNA, called “whole-genome” analysis. Three recent studies have evaluated molecular information for the entire genome in more than 1,000 individuals. The first two studies each identified and traced the patterning of more than 500,000 SNPs as expressed in a few dozen populations worldwide (Jakobsson et al., 2008; Li et al., 2008). The most recent study, called the 1000 Genomes Project, is a massive collaboration of more than 400 scientists worldwide; its preliminary findings reported on close to 15 million SNPs (as well as other DNA variants such and insertions and deletions); indeed, with more detailed sequencing methods and more sophisticated analyses, the researchers concluded they already had discovered the molecular basis for 95 percent of all fairly common patterns of human variation (Durbin et al., 2010). They have also identified between 50 and 100 gene variants associated with disease. Rather than relying on scans of DNA segments (which locate SNPs), this study also made use of the latest sequencing techniques of human genomes and could thus quite accurately reconstruct the entire genome for 179 individuals (with an ultimate goal of completing whole genome sequences for 2,500 people from all around the world). These more complete data, particularly as they are enhanced further, will provide the basis for the next generation of human population genetics studies.
So far, the results of these new studies are highly significant because they confirm earlier findings from more restricted molecular data and they provide new insights. The higher degree of genetic variation seen in African populations as compared with any other geographical group was once again clearly seen. All human populations outside Africa have much less genetic variation than is seen in Africa. These findings further verify the earlier genetic studies (as well as fossil discoveries) that suggest a fairly recent African origin of all modern humans (as discussed in Chapter 13). Moreover, these new data shed light on the genetic relationships between populations worldwide and the nature of human migrations out of Africa (see “A Closer Look: What DNA Tells Us about Ancient Human Migrations”). They also provide evidence of the role of genetic drift (founder effect) in recent human evolution as successively smaller populations split off from larger ones. Finally, preliminary results suggest that the patterning of human variation at the global level may help scientists identify genetic risk factors that influence how susceptible different populations are to various diseases. Specifically, the relative genetic uniformity in non-African populations (for example, European Americans) as compared with those of more recent
African descent (for example, African Americans) exposes the former to a greater risk of developing disease (Lohmueller et al., 2008). How such information might be put to use, however, is controversial. All these genetic data, including the more traditional polymorphisms (such as blood groups) and the vast new DNA-based evidence, point in the same direction: Genetically, humans differ individually within populations far more than large geographical groups (“races”) differ from each other. Does this mean, as eminent geneticist Richard Lewontin concluded 40 years ago, that there’s no biological value in the further study of geographical populations (Lewontin, 1972)? Even with all our new information, the answer isn’t entirely clear. Some of the recent genetic evidence from patterns of two different types of CNVs (Rosenberg et al., 2002; Bamshad et al., 2003) has pointed to broad genetic correlations that consistently indicate an individual’s geographical ancestry. We must consider some important points here, however. These geographically patterned genetic clusters aren’t “races” as traditionally defined, and so they aren’t closely linked to simple patterns of phenotypic variation (such as skin color). What’s more, the correlations are broad, so not all individuals can be easily classified. In fact, many people would probably be misclassified, even when the best information for dozens of genetic loci is used. This debate isn’t entirely academic, and it really never has been. Just consider the destructive social impact that the misuse of the race concept has caused over the last few centuries. A contemporary continuation of the debate concerns the relationship of ancestry and disease. It’s long been recognized that some disease- causing genes are more common in certain populations than in others (such as the allele that causes sickle-cell anemia). The much more complete data on human DNA patterns have further expanded our knowledge, showing, for example, that some people are more resistant than others to HIV infection (see Chapter 15 for further discussion). Does this mean that a person’s ancestry provides valuable medical information in screening or even treating certain diseases?
Some experts argue that such information is medically helpful (e.g., Rosenberg et al., 2002; Bamshad and Olson, 2003; Burchard et al., 2003). What’s more, official federal guidelines recently issued by the U.S. Food and Drug Administration recommend the collection of ancestry data (“race/ethnic identity”) in all clinical trials testing new drugs. Other researchers disagree and argue that such information is at best tenuous (King and Motulsky, 2002) or that it has no obvious medical use (e.g., Cooper et al., 2003). A major difficulty fueling this controversy has been poor communication between biomedical researchers and anthropologists and other evolutionary biologists. To allow for a more balanced and useful approach, anthropologist Clarence Gravlee has argued for adoption of a “more complex biocultural view of human biology” (2009, p. 54).
Even the general public has weighed in on this issue, defeating a 2003 California ballot measure that would have restricted the collection of “racial” (ethnic) information on medical records. There are no easy answers to the questions we’ve raised, and this is an even stronger argument for an informed public. The subject of race has been contentious, and anthropology and other disciplines have struggled to come to grips with it. Our new genetic tools have allowed us to expand our knowledge at a rate far beyond anything seen previously. But increased information alone doesn’t permit us to fully address all human concerns. How we address diversity, both individually and collectively, must balance the potential scientific benefits against a history of social costs.