Molecular Biology and Evolution 17:2-22 (2000)
© 2000 Society for Molecular Biology and Evolution
Review Article |
Population Bottlenecks and Pleistocene Human Evolution
*Department of Anthropology, University of Utah;
and
Department of Anthropology, University of Michigan;
and
Department of Biosystems Science, Graduate University for Advanced Studies, Kanagawa, Japan
 |
Abstract |
|---|
 TOP Abstract IntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
We review the anatomical and archaeological evidence for an early population bottleneck in humans and bracket the time when it could have occurred. We outline the subsequent demographic changes that the archaeological evidence of range expansions and contractions address, and we examine how inbreeding effective population size provides an alternative view of past population size change. This addresses the question of other, more recent, population size bottlenecks, and we review nonrecombining and recombining genetic systems that may reflect them. We examine how these genetic data constrain the possibility of significant population size bottlenecks (i.e., of sufficiently small size and/or long duration to minimize genetic variation in autosomal and haploid systems) at several different critical times in human history. Different constraints appear in nonrecombining and recombining systems, and among the autosomal loci most are incompatible with any Pleistocene population size expansions. Microsatellite data seem to show Pleistocene population size expansions, but in aggregate they are difficult to interpret because different microsatellite studies do not show the same expansion. The archaeological data are only compatible with a few of these analyses, most prominently with data from Alu elements, and we use these facts to question whether the view of the past from analysis of inbreeding effective population size is valid. Finally, we examine the issue of whether inbreeding effective population size provides any reasonable measure of the actual past size of the human species. We contend that if the evidence of a population size bottleneck early in the evolution of our lineage is accepted, most genetic data either lack the resolution to address subsequent changes in the human population or do not meet the assumptions required to do so validly. It is our conclusion that, at the moment, genetic data cannot disprove a simple model of exponential population growth following a bottleneck 2 MYA at the origin of our lineage and extending through the Pleistocene. Archaeological and paleontological data indicate that this model is too oversimplified to be an accurate reflection of detailed population history, and therefore we find that genetic data lack the resolution to validly reflect many details of Pleistocene human population change. However, there is one detail that these data are sufficient to address. Both genetic and anthropological data are incompatible with the hypothesis of a recent population size bottleneck. Such an event would be expected to leave a significant mark across numerous genetic loci and observable anatomical traits, but while some subsets of data are compatible with a recent population size bottleneck, there is no consistently expressed effect that can be found across the range where it should appear, and this absence disproves the hypothesis.
|
Introduction |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
The paleodemographic history of humanity has classically been studied as a problem of archaeology (Birdsell 1972
; Hassan 1981
; Wobst 1993
), with largely theoretical contributions from mathematical modeling (Yellen and Harpending 1972
; Weiss 1973
; Buikstra and Konigsberg 1985
). However, in the past several years, an increased availability of data on human genetic variation, coupled with advances in theoretical population biology, have allowed us to further examine human demographic history from a genetic perspective. These data have addressed some of the earlier problems but have also created others, because genetic hypotheses about the past are largely a reflection of the paleodemographic models assumed (Brookfield 1997
).
Large-scale genetic studies have assessed the worldwide pattern of variation in human mtDNA (Cann, Stoneking, and Wilson 1987
; Vigilant et al. 1991
; Takahata 1993
; Easteal, Harley, and Betty 1997
), Y chromosomes (Dorit, Akashi, and Gilbert 1995, 1996
; Hammer 1995
; Hammer and Zegura 1996
; Underhill et al. 1997
; Hammer et al. 1998
), ß-globin (Harding et al. 1997
), and HLA alleles (Klein et al. 1993
; Ayala 1995
; Takahata and Satta 1998
). Other studies have analyzed systems comprising multiple loci interspersed throughout the genome, including microsatellites (Di Rienzo et al. 1997
; Jorde et al. 1997
; Kimmel et al. 1997
; Calafell et al. 1998
; Reich and Goldstein 1998
; Stephan and Kim 1998
), single-nucleotide polymorphisms (Mountain et al. 1992
; Mountain and Cavalli-Sforza 1994, 1997
; Wang et al. 1998
), and human-specific Alu insertions (Batzer et al. 1992
; Harpending et al. 1993
; Rogers and Jorde 1995
; Sherry 1996
; Sherry et al. 1997
). When these sources of data have been compared, they have sometimes yielded contradictory results (Ruvolo 1996
; Hey 1997
; Wise et al. 1997
). Methods of population genetic analysis have begun to address these contradictions in the context of testing hypotheses of past demographic change (Jorde et al. 1995
; Hey 1997
).
Contradictions also occur between genetic and other, more traditional, sources of data addressing past human evolution (Bower 1999
; Pennisi 1999
). Unlike indirect methods based on population genetics, archaeological and paleontological sources provide direct evidence about the past that can be independently compared with genetic inferences. Recent debates have pitted such evidence against genetic interpretations (e.g., Thorne and Wolpoff 1992
vs. Wilson and Cann 1992
), simplifying the controversy to dueling slogans about which source of evidence is "better": e.g., "fossils are the only direct evidence of evolution," and "all of the living have ancestors while it may be that no fossils have descendants." This is not a useful way to proceed, and the use of simulations provides the possibility of using such evidence quite differently, to test models of demographic evolution rather than to emphasize the incompatibilities of different data sources.
Here, we provide a broad application of the simulation approach to examine both genetic and nongenetic sources of information concerning the most fundamental demographic issues—the population size of the human lineage in the past, and how it has changed during the last 2 Myr. The different reconciliations of paleoanthropological data and recombining genetic systems, nonrecombining systems, or systems with low rates of recombination all have one thing in common: the expectation that small population size bottlenecks (reductions in population size followed by population size increases) played an important role in the Plio-Pleistocene evolution of our lineage. No evidence contradicts the contention that one of these bottlenecks took place at the time of the speciation at the beginning of our lineage, at the end of the Pliocene some 2 MYA. This early population size bottleneck has great explanatory power and important implications for understanding genetic variation and its relationship with past population size. The question we address is whether this bottleneck is compatible with any other more recent ones.
In this paper, we review the anatomical and archaeological evidence for an early population size bottleneck and bracket the time when it could have occurred. We outline the subsequent demographic changes that the archaeological evidence of range expansions and contractions address, and we examine how estimates of inbreeding effective population size from genetic data may provide an alternative view of past population size change. We discuss the possibility of more recent population size bottlenecks, and we review nonrecombining and recombining genetic systems that may reflect them. We examine the constraints that these genetic data place on the possibility of significant population size bottlenecks (i.e., of sufficiently small size and/or long duration to minimize genetic variation in autosomal and haploid systems) at several different critical times in human history. Different constraints appear in nonrecombining and recombining systems, and among autosomal loci most are incompatible with any Pleistocene population size expansions. Microsatellite data can be construed as showing Pleistocene population size expansions but are difficult to interpret because different microsatellite studies do not show the same expansion. The archaeological data are only compatible with a few of these analyses, most prominently with data from Alu elements, and we use these facts to question whether the view of the past from analysis of inbreeding effective population size is valid. Finally, we examine the conditions under which inbreeding effective population size can be expected to provide any reasonable measure of the actual past size of the human species.
|
A 2-Myr Bottleneck |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
There are many reasons to believe that there may have been a number of severe population size bottlenecks on the lineage leading to living humans, principally because of the many speciation events that must have occurred. The diversity of the Pliocene hominid fossil record, beginning with the large samples from Aramis and Kanapoi 4.0–4.4 MYA (White, Suwa, and Asfaw 1994
; Leakey 1995
; Leakey et al. 1998
), indicates that ours is just the most recent of a wide array of hominid species that once existed. The demographic effects of such speciations can be expected to have been intense, probably involving significant founder effects due to small population sizes, and they eradicated evidence of earlier speciations, such as the chimpanzee-hominid divergence. In turn, we expect that any genetic evidence of these early hominid speciations would have been covered up by the most recent significant bottleneck. We believe this bottleneck could have been the speciation event at the beginning of the lineage leading to living human populations. There are two issues to consider here: what is the paleoanthropological evidence of a Late Pliocene hominid speciation, and what is the evidence that this speciation was cladogenic and involved a small population size bottleneck? In later sections, we will explore the question of whether this was the most recent significant bottleneck.
A hominid speciation is documented with paleoanthropological data at about 2 MYA by significant and simultaneous changes in cranial capacity and both cranial and postcranial characters. This marks the earliest known appearance of our direct ancestors. The new species has been called Homo erectus or Homo ergaster by some authors. Following others (Jelínek 1978
; Aguirre 1994
; Wolpoff et al. 1994
), we call this emerging evolutionary species early Homo sapiens, as it begins an unbroken lineage leading directly to living human populations. The first specimens are humanity’s earliest known direct ancestors.
We, like many others, interpret the anatomical evidence to show that early H. sapiens was significantly and dramatically different from earlier and penecontemporary australopithecines in virtually every element of its skeleton (fig. 1
) and every remnant of its behavior (Gamble 1994
; Wolpoff and Caspari 1997
; Asfaw et al. 1999
; Wood and Collard 1999
). Its appearance reflects a real acceleration of evolutionary change from the more slowly changing pace of australopithecine evolution. For instance, Australopithecus afarensis, Australopithecus africanus, and the earliest H. sapiens sample are three species that are generally thought to be an ancestral-descendent line, although with cladogenesis between them. There certainly is cladogenesis between the last two, as H. sapiens and the penecontemporary habiline species now attributed to Australopithecus (Wolpoff 1999
; Wood and Collard 1999
) must have a recent common ancestor later than A. africanus. These consecutive species samples are about half a million years apart, but the amounts of change between them are quite different. From the earlier to later australopithecine species, cranial capacity (approximate midsex average) goes from 450 cm3 to 475 cm3, while from A. africanus to the earliest African H. sapiens sample the change is much greater: 860 cm3. Supporting this, a newly named 2.5-Myr-old australopithecine species that is argued to be a direct ancestor of H. sapiens, Australopithecus garhi, has a male cranial capacity of 450 cm3 (Asfaw et al. 1999
). The significant change to the cranial size of H. sapiens is greater than could be explained by body size alone (fig. 2
), which also greatly increases as discussed below.
|
|
Yet, brain size is only one of the evolving systems reflected in early H. sapiens anatomy. There are four interrelated complexes of changes at the very beginning of H. sapiens (Wolpoff 1999
): (1) changing brain size (larger, especially longer vault, with a broad frontal bone and an expanded parietal association area; neural canal expansion); (2) changing dental function (more anterior tooth use, greater emphasis on grinding and less on crunching) as reflected in broader faces and larger nuchal areas; (3) development of a cranial buttressing system to strengthen the vault, including vault bone thickening and prominent tori; and (4) dramatic expansion of body height (estimated average weights double) and numerous changes in proportions (fig. 1
). These, and other changes involving the visual and respiratory systems, reflect significant adaptive differences for the new species and give us important insight into the mode of speciation because they seem to happen all together, at the time of its origin. The anatomy of the earliest H. sapiens sample indicates significant modifications of the ancestral genome and is not simply an extension of evolutionary trends in an earlier australopithecine lineage throughout the Pliocene. In fact, its combination of features never appears earlier; some of its characteristics are unique, such as the very large body sizes and long legs described below, while others can be found in isolation in various different Pliocene and penecontemporary hominid species.
A Genetic Revolution
If we assume these earlier australopithecines are a group of very closely related species, for instance, nearer to each other than Pan and Homo, we can expect that they differ much more in allele frequencies than in the presence or absence of specific genes for these features. Therefore, a reshuffling of existing alleles could result in the frequencies of features we observe in early H. sapiens. Thus, our second question is about this reshuffling, whether early H. sapiens is a consequence of rapid speciation with significant founder effect or the result of a long, gradual process of anagenic change. The first explanation, cladogenesis, is suggested by the fact that no gradual series of changes in earlier australopithecine populations clearly leads to the new species, and no australopithecine species is obviously transitional. This may seem to be an unexpected statement, because for 3 decades habiline species have been interpreted as being just such transitional taxa, linking Australopithecus through the habilines to later Homo species. But with a few exceptions, the known habiline specimens are now recognized to be less than 2 Myr old (Feibel, Brown, and McDougall 1989
) and therefore are too recent to be transitional forms leading to H. sapiens.
Our interpretation is that the changes are sudden and interrelated and reflect a bottleneck that was created because of the isolation of a small group from a parent australopithecine species. In this small population, a combination of drift and selection resulted in a radical transformation of allele frequencies, fundamentally shifting the adaptive complex (Wright 1942
); in other words, a genetic revolution (Mayr 1954
; Templeton 1980
).
This interpretation is also supported by the fact that several different adaptive complexes changed significantly (as noted above) and together, and that evidences of these changes is found in the earliest specimens. These earliest remains exemplify the significance and magnitude of the newly evolved differences, although not exhaustively, as not all body parts are represented. The most ancient finds are the KNM-ER 3228 innominate (Rose 1984
) and the KNM-ER 2598 occipital bone (Wood 1991
), dated, respectively, at 1.95 ± 0.05 and 1.89 ± 0.01 Myr (Feibel, Brown, and McDougal 1989
). It was noted in their descriptions, and we found in our comparisons, that each of these bones closely resembles its later (what we refer to as) early H. sapiens counterpart and differs markedly from australopithecines.
KNM-ER 2598 is the upper portion of a big, thick occiput with a broad, vertically tall, backward-projecting and thick nuchal torus (bone thickness is 18 mm at inion) and a flexed occipital angle (108°). Internally, there are large cerebral fossae. None of these features, reflecting complexes 1 and 3 above, are found on earlier hominid occiputs. For instance, bone thickness at inion averages 10.3 mm for A. afarensis and 12.8 mm for A. africanus. The midline vertical height of the ER 2598 nuchal torus is 23.5 mm, compared with an A. afarensis mean of 12.5 mm and an A. africanus mean of 13.9 mm.
KNM-ER 3228 is a very large right male innominate that exceeds the size of the largest male australopithecine bones (Stw 431, SK 50). It differs from them in features such as the relatively large acetabulum and strongly developed iliac pillar (fig. 3
). These reflect complex 4, above. The emerging anatomy shows that there were increases in the hip joint reaction and gluteal abductor forces from the australopithecine condition and is compatible with Ruff’s (1995)
model of postaustralopithecine pelvic changes. As he reconstructs this specimen and the more complete ER 15000 pelvis from 400,000 years later (both males), the transverse breadth of the pelvis was constrained by climatic adaptation to the tropics, while the pelvic aperture’s breadth increased in response to larger head size at birth (with the male condition presumably reflecting the female responses). The more vertical orientation of the iliac blade that resulted from these changes, combined with the very long legs of early H. sapiens, created more bending stress in the ilium and higher joint reaction force on the acetabulum.
|
Ruff (1995)
believes increasing head size at birth during the australopithecine-like birth process (reflecting complex 1, above) controlled the pelvic aperture shape (cf. Tague and Lovejoy 1986
). Birth in the earliest H. sapiens did not involve a second rotation during the trip through the pelvic aperture to bring the baby into the sacrum-facing position of today’s births, because pelvic outlet shape appears to have matched the inlet shape in being transversely broad and anterior-posteriorly narrow. Ruff thereby contends that many of the pelvic features in the early H. sapiens males, including pelvic aperture shape, are related to the birthing problems faced by women. His interpretation of this shape implies a significant change to relatively premature (altricial) births in earlier H. sapiens, because, in spite of markedly greater cranial capacity, fetal heads were too small at birth to influence pelvic aperture dimensions and shape constraints and select for the changes in aperture shape that characterize women today (and in the Late Pleistocene). figure 3 shows the basis for this interpretation, with the earliest known specimen of H. sapiens, the male innominate discussed above. The australopithecine that is illustrated in the comparison is the most complete male innominate of this genus. It is a much smaller, lighter biped with a large pelvis relative to his body size. This reflects the fact that body size itself is a very significant aspect of change (complex 4, above). Early H. sapiens is considerably taller and markedly heavier than earlier australopithecines or penecontemporary habilines. In fact, all large-sized postcranial remains from the Koobi Fora and Olduvai deposits found with diagnostic cranial material are associated with early H. sapiens, and no early H. sapiens crania are associated with anything but the largest postcranial remains. The frequency distribution for femur length (fig. 4 ) shows this quite unequivocally. The distribution appears to have three modes, with the large mode including all femora attributed to early H. sapiens, such as male specimens ER 736 and the estimated adult length for the WT 15000, and ER 1808, which other skeletal evidence shows to be a female.
|
In sum, the earliest H. sapiens remains differ significantly from australopithecines in both size and anatomical details. Insofar as we can tell, the changes were sudden and not gradual.
Behavioral Changes
This section addresses a second reason for suspecting there was a bottleneck and a genetic reorganization at the beginning of H. sapiens evolution. The characteristic early H. sapiens features denote a new adaptive pattern that many describe as the first true hunting, gathering, and scavenging adaptation and that we believe may be uniquely associated with the Oldowan archaeological occurrences. These facts provide insight into what some of the sources of selection promoting the new species might have been.
Body size is a key element in the behavioral changes reflected at the earliest H. sapiens archaeological sites because of the locomotor changes that large body size denotes and the increased metabolic resources it requires. Moreover, the marked increase in brain size for early H. sapiens has significant metabolic consequences, because the human brain, which is 2% of the body weight, uses some 20%–25% of its metabolic energy. Larger brain size evolved in spite of these increased energy requirements, but the additional energy had to come from somewhere, and the answer must certainly lie in meat (Milton 1999
). Larger body size in nonhuman primates is associated with the consumption of increasing amounts of low quality foods, and an increase in the amount of time and energy spent eating. The greater human body mass, and especially the longer legs, reflected a new foraging strategy related to this, in which, as Leonard and Robertson (1996)
note: "large day ranges, increased meat consumption, division of foraging activities, and sharing of resources ... may have both necessitated and allowed for a higher-quality diet." These authors estimate that the body size increase from the australopithecines would require a 40%–45% increase in the total energy expenditure of early H. sapiens. They suggest that if this evolutionary change were associated with a shift to a more human-like foraging strategy, it would mean that the energy expenditure increase may have been even greater, perhaps as much as 85% greater than that for australopithecines, because of the locomotor requirements. The payoff for early H. sapiens populations, and the source of the additional energy, was in the higher-quality diet with its concentrated energy sources and the predictable use of more resources provided by the newly developed hunting, gathering, and scavenging strategy.
These behavioral changes are far more massive and sudden than any earlier changes known for hominids. They combine with the anatomical evidence to suggest significant genetic reorganization at the origin of H. sapiens, and from this genetic reorganization, we deduce that H. sapiens evolved from a small isolated australopithecine population and that small population size played a significant role in this evolution.
Population History After the Bottleneck
We have no way of directly estimating with any certainty the size of the human species immediately after the bottleneck at its origin. Archaeological sites from this time are widely scattered, but their sampling is too incomplete for a direct assessment. The problem is that significant range expansion out of Africa occurred a half million years or more later than the first H. sapiens. Population size before then may have remained small, and this is not an insubstantial time span, being one quarter of the time H. sapiens has existed. An important date in behavioral evolution is 1.5 MYA because it is marked by the earliest appearance of the Acheulean (Asfaw et al. 1992
), the ubiquitous hand-axe industry of the Early and Middle Pleistocene. The appearance of the Acheulean involves dramatic behavioral changes. The earliest-dated Acheulean site is also the earliest site with significant butcher marks on the limb bones of megafauna and occurs just before the time of significant human colonization of the Old World tropics and semitropics. Before this time, humanity was limited to Africa and immediately adjacent sections of Asia such as the Levant. These are major changes in human paleoecology and paleodemography, and it is possible that in the half million or more years between the origin of H. sapiens and these changes, the human population was quite small and restricted to only a narrow ecological and geographic range.
Following these first significant range expansions, population size estimates are increasingly accurate for more recent times (cf. Birdsell 1972
; Weiss 1984
). Today, the human species numbers approximately 6 billion individuals, although as recently as the Early Holocene there may have been as few as 6 million (Coale 1974
; Weiss 1984
; Eldredge 1998
). The pattern of population size change across the Pleistocene has come to be of critical interest, linking paleodemography with population genetics, paleoecology, and paleoanthropology.
Exponential expansion of the human species has certainly been ongoing since the inventions of agriculture and domestication early in the Holocene (Pennington 1996
). It seems likely that this expansion began even earlier, as reflected by increasing site densities and complexity of material culture during the Late Pleistocene (Birdsell 1972; Gamble 1987
; Klein 1989
). Humans became a colonizing species early in the Pleistocene; humanity was first restricted to some parts of Africa, but by 1 MYA, populations had spread widely and occupied the tropics and some temperate regions of the Old World. The archaeological record shows that these range expansions have continued since (Butzer 1971
; Ward and Weiss 1976
; Soffer 1987
; Gamble 1994
; Lahr and Foley 1994
). In spite of oscillating population sizes across the temperate zones everywhere, perhaps corresponding to the glaciations and their effects (Gamble 1987
; Jochim 1987
; Roebroeks, Conrad, and van Kolfschoten 1992
; Mussi and Roebroeks 1996
), the archaeological record reflects increased habitat specialization and continually larger population numbers worldwide. However, the oscillations were significant. For instance, both central/western Europe and southern Africa were largely depopulated in the Late Pleistocene, Europe several times, according to Klein (1989, 1994)
.
Because of the pattern of population increase suggested by the distribution of dated archaeological sites, traditional estimates of past population size have been based on assumptions of long-term exponential growth (Keyfitz 1966
; Coale 1974
; Biraben 1979
). Weiss (1984), in his modeling of past population parameters, postulates that the often-observed hunter-gatherer population density of 0.28 per km2 (Tindale 1940
; Birdsell 1958
; Hassan 1981
) can be applied to estimating population size from the areas of habitation in the Pleistocene. From this, and the distribution of archaeological sites, his interpretation of Pleistocene paleodemography implies that peoples who inhabited the Paleolithic world lived in small groups with low population densities and a slow average rate of growth, an interpretation that has been continually confirmed (e.g., Stiner et al. 1999
). Weiss (1984)
estimates a population of about 0.5 million between a half million and a million years ago, and about 1.3 million in the Middle Paleolithic. However, all of these estimates have high probable errors (Petersen 1975
), not only because of the difficulties in applying archaeological information to demographic questions, but also because of the evidence of significant population size fluctuations.
|
Effective Population Size |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
Analysis of genetic evidence taken from large numbers of individuals may provide a different avenue of information about human paleodemography. Under selective neutrality and mutation-drift equilibrium, we might expect the genetic diversity within a population to be related to the size of the population. Ideally, we might be able to interpret what the population size has been in the past from the level of current genetic variation. In reality, however, many different factors can affect the relationship between genetic variation and population size, so populations of the same actual, or census, size (Nc) may have very different levels of genetic variation. Humans are nonideal in many ways, including overlapping generations, population subdivisions with sizes that vary both over time and across space, local population extinctions and recolonizations, and different reproductive patterns between sexes. These factors combine to make the relationship between the level of genetic variation and the census population size very complex.
To account for the many factors other than population size that affect genetic variation, population geneticists replace census population size with a surrogate they can calculate, the effective population size (Ne). Ne is the number of individuals in an ideally behaving, random-mating population that has the same magnitude of genetic drift as the actual population of interest (Wright 1938
; Crow and Kimura 1970
; Hartl and Clark 1997
). Its calculation always assumes that the genes concerned are neutral and unlinked to genes that may be perturbed by selection (Caballero 1994
). Unfortunately, there is no single effective population size (Chesser et al. 1993
; Templeton and Read 1994
). Rather, its definition varies in accordance with the kind of diversity of interest and the factors thought to disturb it. For studies of past population, the measure of magnitude of genetic drift that we are interested in is the change in average inbreeding coefficient, which is itself the probability of identity by descent of two randomly chosen alleles (Crow and Dennison 1988
). Using this measure of genetic drift in calculations of Ne yields the inbreeding effective size. It is the inbreeding effective size that is addressed throughout this paper.
|
Finding Other Population Size Bottlenecks |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
It is unreasonable to assume that the human population either has been constant or has changed in a smooth, continuous manner throughout the past 2 Myr. The issue we examine is whether these oscillations and variations in the size of the human population ever again attained a magnitude sufficient to be observable as a bottleneck. Genetic drift reduces variation at a rate proportional to the effective size of the population, such that for a population size bottleneck to be observable, its duration must be long relative to the effective size of the population. Therefore, only in the extreme case, in which the number of generations of reduced size during the bottleneck approaches the number of individuals at the reduced size, does the long-term effective population size approach the bottleneck size.
If a population size bottleneck is followed by a population expansion, we might expect to see evidence of this in the pattern of genetic diversity (Tajima 1989
). The principal effect of a postbottleneck expansion in population size is to increase the number of low-frequency genetic variants, since individuals are much more likely to share common ancestors during the bottleneck than at times after the bottleneck. This effect will be more pronounced after longer and more severe bottlenecks, which leave less ancient genetic variation in the population. Tajima’s D statistic, which may be significantly negative for severe past bottlenecks, may be used to detect this effect. In contrast to selection, which may affect one genetic locus independently of others, these population size changes are expected to affect all genetic loci.
Thus, traditional methods of interpreting patterns of past population size changes based on paleontology and archaeology have been joined by new methods of interpreting current patterns of genetic diversity. The use of genetic methods requires that certain assumptions are met, and the accuracy of these methods reflects the extent to which the required assumptions have been the case. As Brookfield (1997)
notes, interpretations of ancient demographic events based on genetic evidence are sensitive to, and in many cases stem from, our assumptions about the characteristics of those events. Because we must base our interpretations on the present pattern of genetic diversity, which is a product of multiple competing demographic and selective forces, our choices about which factors are important will influence our conclusions and may render them inaccurate at best or meaningless at worst. When examining genetic data for evidence of ancient population size and structure, then, it is important to aim for consistency with other sources of evidence, including those based on more traditional methods.
|
Nonrecombining Haploid Systems |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
The question of whether there have been population size bottlenecks within the past million years was raised by the application of genetic data to human paleodemography, with the finding that human mtDNA has little variation relative to the current size of the human population (Cann, Stoneking, and Wilson 1987
; Excoffier 1990
; Vigilant et al. 1991
). Many researchers used mtDNA diversity to estimate the time to the most recent common ancestor, or coalescence time, of human mtDNA. A wide range of estimates was obtained (fig. 5
); 200,000 years is a widely accepted median estimate. This estimate allows an estimate of inbreeding Ne of about 8,800 individuals (table 1
). The effective human population size estimated from mitochondrial diversity is therefore far removed from traditional estimates of the census population size of our species in the past (Weiss 1984
).
|
|
The small effective size of mtDNA led to the hypothesis that the human lineage had undergone a recent bottleneck in population size. It was suggested that at the time of mtDNA coalescence, the entire human species was limited to one thousand to several thousand individuals (Cann, Stoneking, and Wilson 1987
; Vigilant et al. 1991
). To account for the mtDNA data, such a bottleneck would have to have been of sufficient duration to allow the fixation by drift of a single ancestral mtDNA variant. Presumably, this bottleneck was followed by expansion to the current population size. The postbottleneck population expansion could have resulted in a relatively increased number of low-frequency genetic variants. This would explain the departure of mtDNA from neutral mutation-drift equilibrium (Excoffier 1990
; Merriwether et al. 1991
; William, Ballard, and Kreitman 1995
; Nachman 1996
; Hey 1997
; Loewe and Scherer 1997
; Parsons, Muniec, and Sullivan 1997
; Wise, Sraml, and Easteal 1998
). Proceeding from this expectation, several researchers have examined the possibility of recent population expansions, such as those that would follow a bottleneck, using the distribution of pairwise genetic differences in human mtDNA (Harpending et al. 1993
; Sherry et al. 1994
; Rogers and Jorde 1995
). This distribution appears to be consistent with a massive Late Pleistocene population expansion.
The hypothesis of a recent population size bottleneck is also supported by some analyses of the human Y chromosome (Dorit, Akashi, and Gilbert 1995, 1996
; Hammer 1995
; Whitfield, Sulston, and Goodfellow 1995
; Underhill et al. 1997
; Hammer et al. 1998
). For the parts of the Y chromosome with observed variation, coalescence time estimates vary from 37 to 516,000 years (Hammer 1995
; Donnelly et al. 1996
; Fu and Li 1996
; Weiss and von Haeseler 1996
; Hammer et al. 1998
). The antiquity of Y-chromosomal variation is not significantly different from that of mtDNA (Hammer 1995
). As in the case of human mtDNA, estimated Ne for the human Y chromosome is low and is consistent with a recent period of small population size. However, if this is the result of a recent bottleneck, such a bottleneck would have to have been of sufficient duration to cause the fixation of a single Y chromosome variant. As with mtDNA, this bottleneck would be expected to cause a departure from equilibrium in the Y chromosome data. This expectation is apparently met by the frequency spectrum of Y chromosome variants (Harpending et al. 1998
).
The interpretation that the departure from neutral mutation-drift equilibrium reflects population size expansions assumes selective neutrality for these gene systems. However, several geneticists have suggested that selection may influence the distribution of mtDNA and Y chromosome variation in humans (Whitfield, Sulston, and Goodfellow 1995
; Hey 1997
; Templeton 1997
; Wise, Sraml, and Easteal 1998
). This has been a persistent interpretation from studies examining haploid and autosomal variation in the same individuals. Within nonrecombining systems such as mtDNA and parts of the Y chromosome, all the alleles are linked, so selection on any portion reduces variability in the entire genome (Spuhler 1989
; Braverman et al. 1995
; Templeton 1997
; Nachman et al. 1998
). Genetic systems with little or no recombination are consistently biased toward low levels of variation in Drosophila. Selection is the only reasonable explanation for the pattern of interlocus variance in Drosophila (Nurminsky et al. 1998
; McAllister and Charlesworth 1999
), where regions with low rates of recombination retain greater intraspecific diversity than those with higher rates of recombination (Begun and Aquadro 1991, 1992
; Hudson 1994, 1995
; Stephan et al. 1998
). The same pattern of variation is found on the human X chromosome (Nachman et al. 1998
) and may characterize other parts of the human genome.
The suggestion that selection has occurred many times in human evolution is not unexpected, and it is consistent with the pattern of great morphological change in humans during the past 2 Myr. Selection could take several forms. Hitchhiking (Kaplan, Hudson, and Langley 1989
; Johnson 1999
) would help explain the small Ne calculated for these nonrecombining systems because of their linkage. Background selection (Charlesworth, Morgan, and Charlesworth 1993
) is an alternative explanation for reduced variation that is related to selective sweeps, since hitchhiking during a selective sweep could be followed by background selection (Nachman et al. 1998
). An explanation for low Ne based on selection is more compatible with the lack of ancient genetic variation in these systems than a short-duration (<2,000 generations) bottleneck of very small population size. The possibility that there has been selection in these nonrecombining systems (Hudson 1994, 1995
; Stephan et al. 1998
; Whitehead 1998
) points to the necessity of considering autosomal diversity in humans for further evidence of whether the hypothesis of a severe recent bottleneck that some interpretations of haploid variation suggest can be refuted.
|
Autosomal Loci |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
Data are available from a number of autosomal gene systems to address the effective population size of the human lineage and the possibility of bottlenecks as explanations of genetic diversity. It is imperative to compare these with data from haploid systems (Hey 1997
; Wise et al. 1997
). Autosomal systems studied include those interspersed throughout the genome: microsatellites, Alu insertions, and single-nucleotide polymorphisms (SNPs), as well as single genetic loci, including ß-globin, dystrophin, and ZFX.
When considered together (table 1
), these gene systems provide substantial evidence with regard to the limits of ancient population size changes. All of the autosomal systems examined to date are consistent with a long-term average Ne on the order of 104 to 105 for the human species (table 1
). Moreover, they are all consistent with earlier analyses of protein polymorphisms (Nei and Graur 1984
) and of nucleotide polymorphisms (Li and Sadler 1991
; Takahata, Satta, and Klein 1995
) that estimated the long-term effective human population size as on the order of 104.
The striking agreement of all autosomal sources of data on a relatively small Ne for the human lineage is inconsistent with the hypothesis that a recent short population size bottleneck explains it. Such a bottleneck, even if very severe, would leave ancient variation in many gene systems from the prebottleneck period of large population size. Such ancient variation is not observed. This lack of ancient variation also cannot be explained by recurrent ancient bottlenecks that also limited variation. If these had occurred, we expect they would have left some signs of the population expansions between them. While some gene systems are compatible with the interpretation of such expansions, others reject them (Harris and Hey 1999
, and see below). These observations indicate that there is no recent severe bottleneck in human prehistory. Our results agree with those of others who have examined both nuclear and mitochondrial genetic evidence (Ayala 1995
; Jorde et al. 1995
; Whitfield, Sulston, and Goodfellow 1995
; Hey 1997
; Wise et al. 1997
; Hammer et al. 1998
; Harpending et al. 1998
; Wise, Sraml, and Easteal 1998
; Harris and Hey 1999
).
Certain estimates of autosomal Ne values and values determined for mtDNA can potentially be reconciled, because under neutrality, the autosomal Ne is expected to be four times the haploid value (Takahata 1993
). However, this reconciliation is not compatible with the explanation of low diversity in the haploid genes based on a recent bottleneck. Such a bottleneck would have predictable effects on the combined pattern of nuclear and mitochondrial DNA diversity, effects that have not been observed. The mtDNA coalescent is not expected to be one fourth that of nuclear genes following a small population size bottleneck that eradicates variation in both. Instead, because of its smaller effective population size, mtDNA should return to mutation-drift equilibrium more rapidly after a small population size bottleneck than would nuclear DNA, since drift is stronger in a smaller population. However, what we actually observe is mitochondrial DNA that is relatively invariant and out of equilibrium (Excoffier 1990
; Merriwether et al. 1991
; William, Ballard, and Kreitman 1995
; Nachman et al. 1996
; Templeton 1996
; Hey 1997
; Parsons, Muniec, and Sullivan 1997
; Wise, Sraml, and Easteal 1998
), while equilibrium cannot be disproved for most nuclear systems. This is clearly inconsistent with a severe recent population size bottleneck.
|
Long-Term Small Effective Size: the Long-Necked Bottle |
|---|
|
TOPAbstractIntroductionA 2-Myr BottleneckEffective Population SizeFinding Other Population Size...Nonrecombining Haploid SystemsAutosomal LociLong-Term Small Effective Size:...Population ExpansionsProblems with Population...MicrosatellitesEffective and Census Population...Discussion: Bottlenecks in Human...ConclusionsAcknowledgementsReferences |
|---|
An alternative reconciliation of these contradictions is found in the hypothesis that current human genetic variation is the product of a very long history of small population size in equilibrium (Takahata 1993
; Donnelly et al. 1996
; Weiss and von Haeseler 1996
; Fu and Li 1997
; Harding et al. 1997, 1998
; Hammer et al. 1998
; Harpending et al. 1998
; Zietkiewicz et al. 1998
). In this long-necked-bottle model, either Ne remained constantly small, or it oscillated frequently to low levels due to periodic events such as glaciations. This hypothesis differs from a single, short-term bottleneck explanation in that the population size is posited to be small for a long enough period for an equilibrium to be reached in most, if not all, neutral gene systems. This could account for differences in coalescence between recombining autosomal and haploid genetic systems. The fact that Ne in haploid systems is expected to be one quarter of that in recombining autosomal systems predicts that we will calculate a fourfold difference in coalescence times if small ancestral population size, and not a single population size bottleneck, is the cause of the variation (Takahata 1993
).
The long-necked bottle model was developed by Harpending et al. (1998)
as part of their analysis of Alu variation. It addresses the implications of an Ne value on the order of 104–105 for a long period in humans. If the Ne calculated from Alu variation (17,500 according to Harpending et al. 1998
) is a significant fraction of the number of breeding adults in the human species (as Harpending et al. [1998]
assume, following Wood [1987]
), there must have been far too few people to occupy all of the continents inhabited during the Pleistocene, or even to inhabit a significant part of one continent. Such a population spread around the world would have a density so low that there would be only about 22 breeding couples in Germany and 35 in France (Takahata and Klein 1998
). To account for this problem, Harpending et al. (1998, p. 1967)
conclude that a population on the order of 104 could not have occupied the entire Old World, but lived for a million years or more "in an African area the size of Rhode Island or Swaziland" as a separate species. This species would presumably be the direct ancestor of modern humans, H. sapiens.
If correct, this would mean that the vast majority of known archaeological sites represent the remains of the activities of extinct human species. These sites are direct evidence of somebody’s behavior, and they show that expansions of the geographic range of humans from Africa to the rest of the Old World may have begun shortly after the appearance of significant changes in human mobility. These changes are suggested by the much larger size, particularly the longer legs, of our earliest direct ancestors some 2 MYA. An early range expansion, with the implication of increasing population size, is indicated by Late Pliocene/Early Pleistocene dates published for the first Indonesian hominids (Swisher et al. 1994
) and by the early dates variously suggested for the Yuanmou incisors from China (Qing 1985
) and the Dmanisi mandible from Georgia (Gabunia and Vekua 1995