Arrival of Paleo-Indians to the southern cone of South America: new clues from mitogenomes

Abstract

With analyses of entire mitogenomes, studies of Native American mitochondrial DNA (MTDNA) variation have entered the final phase of phylogenetic refinement: the dissection of the founding haplogroups into clades that arose in America during and after human arrival and spread. Ages and geographic distributions of these clades could provide novel clues on the colonization processes of the different regions of the double continent. As for the Southern Cone of South America, this approach has recently allowed the identification of two local clades (D1g and D1j) whose age estimates agree with the dating of the earliest archaeological sites in South America, indicating that Paleo-Indians might have reached that region from Beringia in less than 2000 years. In this study, we sequenced 46 mitogenomes belonging to two additional clades, termed B2i2 (former B2l) and C1b13, which were recently identified on the basis of mtDNA control-region data and whose geographical distributions appear to be restricted to Chile and Argentina. We confirm that their mutational motifs most likely arose in the Southern Cone region. However, the age estimate for B2i2 and C1b13 (11-13,000 years) appears to be younger than those of other local clades. The difference could reflect the different evolutionary origins of the distinct South American-specific sub-haplogroups, with some being already present, at different times and locations, at the very front of the expansion wave in South America, and others originating later in situ, when the tribalization process had already begun. A delayed origin of a few thousand years in one of the locally derived populations, possibly in the central part of Chile, would have limited the geographical and ethnic diffusion of B2i2 and explain the present-day occurrence that appears to be mainly confined to the Tehuelche and Araucanian-speaking groups.

Figure 1. Detailed maximum parsimony tree of 46 novel complete Native American mtDNA sequences belonging to the novel haplogroups B2i2 and C1b13.

These are the first completely sequenced mitogenomes for both B2i2 and C1b13. This tree also includes two previously published sequences of Kayapó individuals from Brazil classified as belonging to sub-clade B2i1. Mutations relative to the L3 node are shown on the branches; they are transitions unless a base is explicitly indicated. The prefix @ indicates reversions while suffixes indicate: transversions (to A, G, C, or T), indels (.1, d), gene locus (∼r, rRNA; ∼t, tRNA), synonymous or non-synonymous changes (s or ns), and non-coding sites outside the control region (nc). The mutations marked by a red @ are reverted only relative to the Revised Sapiens Reference Sequence (RSRS) , all other mutations are relative to both rCRS and RSRS. Recurrent mutations within the phylogeny are underlined. The variation in number of cytosines around nps 309 and 16193 was not included in the tree. Additional information regarding each mtDNA is available in Table 1. Coalescence times shown for B2i2 and C1b13 are Maximum-Likelihood (ML) estimates, and have been obtained by including all sequence changes (except 16182C, 16183C, and at np 16519) from the respective root according to Soares et al. .