Mitochondrial DNA (mtDNA) haplogroup L2 originated in Western Africa but is nowadays spread across the entire continent. L2 movements were previously postulated to be related to the Bantu expansion, but L2 expansions eastwards probably occurred much earlier. By reconstructing the phylogeny of L2 (44 new complete sequences) we provide insights on the complex net of within-African migrations in the last 60 thousand years (ka). Results show that lineages in Southern Africa cluster with Western/Central African lineages at a recent time scale, whereas, eastern lineages seem to be substantially more ancient. Three moments of expansion from a Central African source are associated to L2: (1) one migration at 70–50 ka into Eastern or Southern Africa, (2) postglacial movements (15–10 ka) into Eastern Africa; and (3) the southward Bantu Expansion in the last 5 ka. The complementary population and L0a phylogeography analyses indicate no strong evidence of mtDNA gene flow between eastern and southern populations during the later movement, suggesting low admixture between Eastern African populations and the Bantu migrants. This implies that, at least in the early stages, the Bantu expansion was mainly a demic diffusion with little incorporation of local populations.
Introduction Africa has been considered the cradle of mankind for a long time. Both genetic data (uniparental genetic markers and genome-wide diversity) and fossil evidence suggest that anatomically modern humans originated in this continent1, spreading later all over the globe. However, there is still a vigorous debate not only on the specific region within Africa where modern humans appeared, but also regarding the initial migrations within this continent2. Despite being geographically restricted to Africa before the Out-of-Africa (OOA) migration, ancestral populations most likely already displayed a strong genetic structure for at least 100 thousand years (ka)2,3,4, highly influenced by episodes of climate oscillation5.
The climate dynamics continued to contribute to African population structure after the OAA, notoriously the African Late Glacial Maximum (LGM; at ~18–16 ka6, later on than the northern hemisphere one at ~22–19 ka7), which contributed to aridity, resulting in the expansion of the Sahara desert several kilometres southwards6. The Pleistocene/Holocene transition (~11.5 ka) was characterized by changes in atmospheric circulation and solar radiation6, improving environmental conditions and leading to major human expansions in southwest Asia8, Europe9, and also in Africa (Saharan areas were recolonized10,11, allowing frequent flow across West/Central and North/South2,12,13,14). The humid conditions peaked at the Holocene climatic optimum (~9–6 ka), when Sahara desert virtually disappeared and the Chad lake was seven times larger than today14. A shift to aridity occurred later in the Sahara, at ~6 ka15.
More recently, the African genetic and cultural landscape was deeply affected by an event known as the Bantu expansion. The expansion of Bantu-speakers is thought to have started in the Grassfields region between southeast Nigeria and western Cameroon and taken two main routes from its starting point: a western route, throughout the west coast of Africa, having arrived to Angola, South Africa and Botswana around 3.5 ka, and an eastern route, towards the Great Lakes in Eastern Africa, reaching the region of Uganda about 2.5 ka, where they remained for a couple thousand years, expanding later into the south, reaching Mozambique by ~1.8 ka16,17,18. The Eastern route is of particular interest to study potential crossings between migrants and local eastern populations (namely Nilotic and Cushitic people), during the period in which the Bantu people were stationed in the Great Lakes region. Linguistic differences between eastern and western Bantu languages seem to mirror the two routes of expansion, but, recent evidence suggests a later split of Eastern and Western Bantu19. Either way, the Bantu expansion probably forced the retreat of contemporary local sub-Saharan populations: the San were further confined to the South towards the Kalahari desert and kept their typical Khoisan languages (with click consonants) and ethnic identity, and the Pygmies, on the other hand, were pushed deeper into the forests and eventually some adopted Bantu languages20.
Recent methodological and technical advances led to the emergence of genome-wide (GW) studies, whose main advantage for demographic inference is allowing us to identify and quantify admixture between populations of distinct ancestries21. However, current GW dating methods are still limited in dissecting between several migration waves, usually leading to the identification of a single event of average/young age (discussed in22). Mitochondrial DNA (mtDNA), on the other hand, is only maternally inherited, but due to its fast mutation rate, accumulates variation fast enough amongst different locations, to make it a suitable molecular marker for the phylogeographic approach. Since reliable mtDNA mutation rates have been calculated, it is possible to frame the various demographic events within distinct time periods23. A lineage-based approach can thus provide insights into the demography of populations and reveal patterns that would otherwise be dismissed, and has proved particularly useful to resolve the old debate regarding the Bantu expansion: the identification of specific lineages suggested that the expansion of Bantu languages was due to the migration of Bantu-speakers, rather than just a cultural diffusion19, as previously thought.
Previous studies based on hypervariable segment I (HVS-I) diversity have shown that haplogroup L2 played a major role in the Bantu migration17,18,24. MtDNA haplogroup L2 is the sister branch of the Eastern African L3′4′6 clade that contains all the OOA diversity within haplogroup L3. While L3′4′6 originated in Eastern Africa22, haplogroup L2 probably originated in Western Africa but is nowadays widespread across the continent; it is highly frequent in many regions, such as in Western/Central and Southeast Africa (probably associated with the Bantu expansion that occurred in the last few millennia) and in Northwest, most likely due to trans-Saharan slave trade18,25. Together with haplogroup L3, it represents ~70% of sub-Saharan mtDNA variation but despite its high frequency and wide distribution, L2 was not involved in the OOA26, since most likely it was not yet arrived in Eastern Africa by that time.
The demographic history of L2 is not yet completely understood, especially concerning the age of the expansion into Eastern Africa, a region that might have acted as a refuge during some severe episodes of climate oscillations over the last hundred thousand years27. One possibility is that the expansion of L2 to the East, most likely as with the expansion to the South, was related with movements of Bantu-speaking populations. However, in the regions of highest frequency of L2 in Eastern Africa (over 30%, in the area of Sudan and Ethiopia)13 there are no records of Bantu groups. Furthermore, recent evidence from HVS-I13 suggests that this haplogroup might have first expanded to Eastern Africa much earlier, possibly due to the improvement of climate conditions during the early Holocene. This signal was also observed with Bayesian analysis of L2 (and L2a) complete sequences28. Moreover, particular clades of L2a and L2c suggest an expansion, possibly along the Sahel corridor, after the LGM18. Migrations at this time frame are also observed in branches of other African haplogroups, such as L0a, L1b and L3f2,12,18,29.
Despite being spread across different regions, most of the haplogroup L2 sequences available in online databases are either from Western or Southern Africans or from African-Americans. We aim to better understand the phylogeographic patterns of L2 by improving its phylogeny based on complete sequence information especially for Eastern Africa, a region poorly characterized for L2 clades. This increased resolution will enable us to ascertain about the intensity of the gene flow from Eastern populations to the Bantu migrants towards south. The L2 complete sequence analysis was complemented by a similar analysis for haplogroup L0a (also present in Central and Eastern Africa by the time of the Bantu expansion2) and a HVS-I population-based approach.
[...]
--Marina Silva, Farida Alshamali, Paula Silva, Carla Carrilho, Flávio Mandlate, Maria Jesus Trovoada, Viktor Černý, Luísa Pereira & Pedro Soares
60,000 years of interactions between Central and Eastern Africa documented by major African mitochondrial haplogroup L2
-------------------- Without data you are just another person with an opinion - Deming Posts: 12143 | From: When you have eliminated the impossible, whatever remains, however improbable | Registered: Jun 2007
| IP: Logged |
UBBFriend: Email this page to someone!
Printer-friendly view of this topic