The impact of group fissions on genetic structure in Native South America and implications for human evolution.

In a series of publications beginning in the 1960s, Neel and colleagues suggested that genetically nonrandom, or "lineal", population fissions contributed to genetic structure in ancient human groups. The authors reached this conclusion by studying the genetic consequences of village fissions among the Yanomamo, a Native South American group thought to have been relatively unaffected by European contact and, therefore, representative of the human past. On the basis of ethnographic accounts and pedigree data, they further concluded that patrilineal relationships were particularly important in shaping the genetic structure of villages following fissions. This study reexamines the genetic consequences of village fissions using autosomal STRs, Y-chromosome STRs, and mitochondrial DNA sequences collected from large samples of individuals from multiple Yanomamo villages. Our analyses of the autosomal STRs replicate the previous finding that village fissions have produced substantial genetic structure among the Yanomamo. However, our analyses of Y-chromosome STRs and mtDNA d-loop polymorphisms suggest that other population processes, including village movements, inter-village migration, and polygynous marriage, affect genetic structure in ways not predicted by a simple model of patrilineal fissions. We discuss the broader implications of population fissions for human evolution and the suitability of using the Yanomamo as a model for the human past.

A formal test of linguistic and genetic coevolution in native Central and South America.

This paper investigates a mechanism of linguistic and genetic coevolution in Native Central and South America. This mechanism proposes that a process of population fissions, expansions into new territories, and isolation of ancestral and descendant groups will produce congruent language and gene trees. To evaluate this population fissions mechanism, we collected published mtDNA sequences for 1,381 individuals from 17 Native Central and South American populations. We then tested the hypothesis that three well-known language classifications also represented the genetic structure of these populations. We rejected the hypothesis for each language classification. Our tests revealed linguistic and genetic correspondence in several shallow branches common to each classification, but no linguistic and genetic correspondence in the deeper branches contained in two of the language classifications. We discuss the possible causes for the lack of congruence between linguistic and genetic structure in the region, and describe alternative mechanisms of linguistic and genetic correspondence and their predictions.

A cautionary tale on ancient migration detection: mitochondrial DNA variation in Santa Cruz Islands, Solomon Islands.

Over the past decade, the origin of the first Malayo-Polynesian settlers of the island Pacific has become a contentious issue in molecular anthropology as well as in archaeology and historical linguistics. Whether the descendants of the ancestral Malayo-Polynesian speakers moved rapidly through Indonesia and Island Melanesia in a few hundred years, or whether they were the product of considerable intermingling within the more westerly part of the latter region, it is widely accepted that they were the first humans to colonize the distant Pacific islands beyond the central Solomon Islands approximately 3,000 years ago. The Santa Cruz Islands in the Eastern Solomons would have most likely been the first in Remote Oceania to be colonized by them. Archaeologically, the first Oceanic Austronesian settlement of this region appears to have been overlain by various later influences from groups farther west in a complex manner. Molecular anthropologists have tended to equate the spread of various Austronesian-speaking groups with a particular mitochondrial variant (a 9-base-pair [bp] deletion with specific D-loop variants). We have shown before that this is an oversimplified picture, and assumed that the Santa Cruz situation, with its series of intrusions, would be informative as to the power of mitochondrial DNA haplotype interpretations. In the Santa Cruz Islands, the 9-bp deletion is associated with a small number of very closely related hypervariable D-loop haplotypes resulting in a star-shaped Bandelt median network, suggesting a recent population expansion. This network is similar to Polynesian median networks. In a pairwise mismatch comparison, the Santa Cruz haplotypes have a bimodal distribution, with the first cluster being composed almost entirely of the 9-bp-deleted haplotypes-again attesting to their recent origins. Conversely, the nondeleted haplogroups bear signatures of more ancient origins within the general region. Therefore, while the profiles of the two sets of haplotypes indicate very distinctive origins in different populations with divergent expansion histories, the sequence of their introduction into the Santa Cruz Islands clearly does not follow simply.

Mitochondrial DNA variation is an indicator of austronesian influence in Island Melanesia.

Past studies have shown a consistent association of a specific set of mitochondrial DNA 9 base pair (bp) deletion haplotypes with Polynesians and their Austronesian-speaking relatives, and the total lack of the deletion in a short series of New Guinea Highlanders. Utilizing plasma and DNA samples from various old laboratory collections, we have extended population screening for the 9-bp deletion into "Island Melanesia," an area notorious for its extreme population variation. While the 9-bp deletion is present in all Austronesian, and many non-Austronesian-speaking groups, it is absent in the more remote non-Austronesian populations in Bougainville and New Britain. These results are consistent with the hypothesis that this deletion was first introduced to this region about 3,500 years ago with the arrival of Austronesian-speaking peoples from the west, but has not yet diffused through all populations there. The pattern cannot be reconciled with the competing hypothesis of a primarily indigenous Melanesian origin for the ancestors of the Polynesians. Although selection clearly has operated on some other genetic systems in this region, both migration and random genetic drift primarily account for the remarkable degree of biological diversity in these small Southwest Pacific populations.

Freezer anthropology: new uses for old blood.

Archived blood fractions (plasma, settled red cells, white cells) have proved to be a rich and valuable source of DNA for human genetic studies. Large numbers of such samples were collected between 1960 and the present for protein and blood group studies, many of which are languishing in freezers or have already been discarded. More are discarded each year because the usefulness of these samples is not widely understood. Data from DNA derived from 10-35-year-old blood samples have been used to address the peopling of the New World and of the Pacific. Mitochondrial DNA haplotypes from studies using this source DNA support a single wave of migration into the New World (or a single source population for the New World), and that Mongolia was the likely source of the founding population. Data from Melanesia have shown that Polynesians are recent immigrants into the Pacific and did not arise from Melanesia.

Mitochondrial versus nuclear admixture estimates demonstrate a past history of directional mating.

Six blood group antigens (ABO, RH, MNS, KK, KP, FY) and five plasma proteins (HP, GC, APOA4, FXIIIB, C1R) were typed in 790 individuals, and 12 mtDNA RFLP and deletion polymorphisms were typed in 657 individuals from the San Luis Valley, Colorado. The 790 nuclear typings were conducted on 399 Anglos and 391 Hispanics, while the 657 mitochondrial haplotypes were generated from 207 Anglos and 450 Hispanics. Chakraborty's ADMIX2 FORTRAN program was used to estimate the average Amerindian admixture using all nuclear loci simultaneously. Since there is no recombination in mtDNA, the sum of the frequencies of the Amerindian/Asian-specific mitochondrial haplotypes represents the level of Amerindian admixture. The nuclear estimates of Amerindian admixture were 33.15 +/- 2.41% for the Hispanics and 9.72 +/- 1.90% for the Anglos, while the strictly maternally inherited mtDNA estimates of Amerindian admixture were 85.11% for the Hispanics and 0.97% for the Anglos. This dramatic difference in estimated levels of admixture between the biparentally derived nuclear estimates and the uniparentally derived mtDNA estimates is indicative of past directional matings between Hispanic males and Amerindian females.

mtDNA variation indicates Mongolia may have been the source for the founding population for the New World.

mtDNA RFLP variation was analyzed in 42 Mongolians from Ulan Bator. All four founding lineage types (A [4.76%], B [2.38%], C [11.9%], and D [19.04%]) identified by Torroni and colleagues were detected. Seven of the nine founding lineage types proposed by Bailliet and colleagues and Merriwether and Ferrell were detected (A2 [4.76%], B [2.38%], C1 [11.9%], D1 [7.14%], D2 [11.9%], X6 [16.7%], and X7 [9.5%]). Sixty-four percent of these 42 individuals had "Amerindian founding lineage" haplotypes. A survey of 24 restriction sites yielded 16 polymorphic sites and 21 different haplotypes. The presence of all four of the founding lineages identified by the Torroni group (and seven of Merriwether and Ferrell's nine founding lineages), combined with Mongolia's location with respect to the Bering Strait, indicates that Mongolia is a potential location for the origin of the founders of the New World. Since lineage B, which is widely distributed in the New World, is absent in Siberia, we conclude that Mongolia or a geographic location common to both contemporary Mongolians and American aboriginals is the more likely origin of the founders of the New World.

mtDNA variation in the Yanomami: evidence for additional New World founding lineages.

Native Americans have been classified into four founding haplogroups with as many as seven founding lineages based on mtDNA RFLPs and DNA sequence data. mtDNA analysis was completed for 83 Yanomami from eight villages in the Surucucu and Catrimani Plateau regions of Roraima in northwestern Brazil. Samples were typed for 15 polymorphic mtDNA sites (14 RFLP sites and 1 deletion site), and a subset was sequenced for both hypervariable regions of the mitochondrial D-loop. Substantial mitochondrial diversity was detected among the Yanomami, five of seven accepted founding haplotypes and three others were observed. Of the 83 samples, 4 (4.8%) were lineage B1, 1 (1.2%) was lineage B2, 31 (37.4%) were lineage C1, 29 (34.9%) were lineage C2, 2 (2.4%) were lineage D1, 6 (7.2%) were lineage D2, 7 (8.4%) were a haplotype we designated "X6," and 3 (3.6%) were a haplotype we designated "X7." Sequence analysis found 43 haplotypes in 50 samples. B2, X6, and X7 are previously unrecognized mitochondrial founding lineage types of Native Americans. The widespread distribution of these haplotypes in the New World and Asia provides support for declaring these lineages to be New World founding types.

The four founding lineage hypothesis for the New World: a critical reevaluation.

It has been proposed that all native American mitochondrial DNA variation in the New World can be attributed to divergence from four "founding lineages" which entered the New World in three waves of migration from across the Bering Strait (T.G. Schurr et al., 1990, Am. J. Hum. Genet. 46: 613-623; A. Torroni and D. C. Wallace, 1995, Am. J. Hum. Genet. 56: 1234-1236; A. Torroni et al., 1994, Am. J. Hum. Genet. 54: 303-318; A. Torroni et al., 1994, Proc. Natl. Acad. Sci. USA 91: 1158-1162; A. Torroni et al., 1994, Am. J. Phys. Anthropol. 93: 189-199; A. Torroni et al., 1993a, Am. J. Hum. Genet. 53: 563-590; A. Torroni et al., 1993b, Am. J. Hum. Genet. 53: 591-608; Wallace and A. Torroni, 1992, Hum. Biol. 64(3): 403-416). Torroni et al. (1993a) believe that only one haplotype from each of these four founding lineages arrived in the New World via migration, and all the additional variation arose in the New World. Any other types were attributed to Caucasian admixture. G. Bailliet et al. (1994, Am. J. Hum. Genet. 55: 27-33), N. O. Bianchi and F. Rothhammer (1995, Am. J. Hum. Genet. 56: 1236-1238), and D. A. Merriwether (1994, Experientia 50: 592-601; 1995, Am. J. Phys. Anthropol. 98(4): 411-430) suggest that multiple variants of each lineage entered the New World, and that additional unrelated lineages also entered. We present the distribution of multiple variants of the four founding lineages, plus two additional lineages which we call X6 and X7, throughout the New World, Siberia, and Asia. These distributions are strong evidence that at least nine different founding lineage haplotypes entered the New World. Further, we find these distributions among Native Americans best fit a single wave of migration into the New World.

Distribution of the four founding lineage haplotypes in Native Americans suggests a single wave of migration for the New World.

The distribution of the four founding lineage haplogroups in Native Americans from North, Central, and South America shows a north to south increase in the frequency of lineage B and a North to South decrease in the frequency of lineage A. All four founding lineage haplogroups were detected in North, Central, and South America, and in Greenberg et al.'s ([1986] Curr. Anthropol. 27:477-497) three major linguistic groups (Amerind, NaDene, and Eskaleut), with all four haplogroups often found within a single population. Lineage A was the most common lineage in North America, regardless of language group. This overall distribution is most parsimonious with a single wave of migration into the New World which included multiple variants of all four founding lineage types. Torroni et al.'s ([1993a] Am. J. Hum. Genet. 53:563-590) report that lineage B has a more recent divergence time than the other three lineages can best be explained by multiple variants of lineages A, C, and D, and fewer variants of lineage B entering the New World. Alternatively, there could have been multiple waves of migration from a single parent population in Asia/Siberia which repeatedly reintroduced the same lineages to the New World.

Genetic variation in the New World: ancient teeth, bone, and tissue as sources of DNA.

Examination of ancient and contemporary Native American mtDNA variation via diagnostic restriction sites and the 9-bp Region V deletion suggests a single wave of migration into the New World. This is in contrast to data from Torroni et al. which suggested two waves of migration into the New World (the NaDene and Amerind). All four founding lineage types are present in populations in North, Central, and South America suggesting that all four lineages came over together and spread throughout the New World. Ancient Native American DNA shows that all four lineages were present before European contact in North America, and at least two were present in South America. The presence of all four lineages in the NaDene and the Amerinds argues against separate migrations founding these two groups, although admixture between the groups is still a viable explanation for the presence of all four types in the NaDene.

The structure of human mitochondrial DNA variation.

Restriction analysis of mitochondrial DNA (mtDNA) of 3065 humans from 62 geographic samples identified 149 haplotypes and 81 polymorphic sites. These data were used to test several aspects of the evolutionary past of the human species. A dendrogram depicting the genetic relatedness of all haplotypes shows that the native African populations have the greatest diversity and, consistent with evidence from a variety of sources, suggests an African origin for our species. The data also indicate that two individuals drawn at random from the entire sample will differ at approximately 0.4% of their mtDNA nucleotide sites, which is somewhat higher than previous estimates. Human mtDNA also exhibits more interpopulation heterogeneity (GST = 0.351 +/- 0.025) than does nuclear DNA (GST = 0.12). Moreover, the virtual absence of intermediate levels of linkage disequilibrium between pairs of sites is consistent with the absence of genetic recombination and places constraints on the rate of mutation. Tests of the selective neutrality of mtDNA variation, including the Ewens-Watterson and Tajima tests, indicate a departure in the direction consistent with purifying selection, but this departure is more likely due to the rapid growth of the human population and the geographic heterogeneity of the variation. The lack of a good fit to neutrality poses problems for the estimation of times of coalescence from human mtDNA data.

Direct and correlated responses to artificial selection on lipid and glycogen contents in Drosophila melanogaster.

A large outbred population of Drosophila melanogaster was subjected to artificial selection on lipid and glycogen storage. In three separate experiments, two replicates underwent sib selection for both increased and decreased storage. In the first study, flies were selected on the basis of total triacylglycerol for ten generations. This experiment resulted in no significant direct response, but there was a significant change in total body weight, underscoring the importance of concern for the allometric relationship between body weight and lipid content. In the second study, selection was performed for 15 generations on the percentage of body composition that was triacylglycerol. A significant direct response was obtained, and the two replicates revealed heritability estimates of 0.40 and 0.43. The third study selected glycogen content for 15 generations, and produced a significant response with heritabilities of 0.25 and 0.31. A series of 12 biochemical and enzyme kinetic traits was examined at five generation intervals in all experiments, and a number of correlated responses were detected. The results are interpreted with respect to the evolutionary constraints on energy storage evolution and the genetic basis of the allometric relationship between body weight and fat content.

Amerindian mitochondrial DNAs have rare Asian mutations at high frequencies, suggesting they derived from four primary maternal lineages.

The mitochondrial DNA (mtDNA) sequence variation of the South American Ticuna, the Central American Maya, and the North American Pima was analyzed by restriction-endonuclease digestion and oligonucleotide hybridization. The analysis revealed that Amerindian populations have high frequencies of mtDNAs containing the rare Asian RFLP HincII morph 6, a rare HaeIII site gain, and a unique AluI site gain. In addition, the Asian-specific deletion between the cytochrome c oxidase subunit II (COII) and tRNA(Lys) genes was also prevalent in both the Pima and the Maya. These data suggest that Amerindian mtDNAs derived from at least four primary maternal lineages, that new tribal-specific variants accumulated as these mtDNAs became distributed throughout the Americas, and that some genetic variation may have been lost when the progenitors of the Ticuna separated from the North and Central American populations.