Genetic ancestry of families of putative Inka descent

Tawantinsuyu was the climax of approximately six millennia of autochthonous cultures in the Central Andes since the Norte Chico culture. The Inkas ruled Tawantinsuyu for approximately 150 years (Marsh et al. 2017) until the arrival of the Europeans in 1531. The first tribal settlement of the Inkas in the Cusco valley in Peru probably dated to the twelfth century, and their empire expanded to a territory of approximately 1,800,000 km² with the largest political system of pre-Columbian Americas. According to a genetic study (Sandoval et al. 2013a), the current population of the Central Andes represents the highest proportion of Native American ancestry, and most Andeans are probably direct descendants of the Tawantinsuyu people. Furthermore, other studies have ratified the cultural and genetic homogeneity among Andean populations in contrast to the highly heterogeneous eastern lowland Amazonian populations (Tarazona-Santos et al. 2001; Sandoval et al. 2016). These and other genetic studies (Llamas et al. 2016; Jota et al. 2016) have shown significant advances in recovering the Native American history; however, no study has focused on the descent of the royal Inka family.

Due to the haploid mode of inheritance, the Y-chromosome and mitochondrial (mt) DNA markers are widely used in reconstructing the genealogical history of populations. Genetic analysis of autochthonous Y-chromosome markers shows most South American populations with haplogroup Q (characterised by different sub-lineages) and other rare haplogroups, such as C-M217 (Jota et al. 2016). Simultaneous analysis of Y-chromosome single-nucleotide polymorphisms (Y-SNPs), which define a haplogroup or sub-lineage, and short tandem repeats (Y-STRs), which are used for paternal testing, familial genealogy and kinship identification, have allowed for the discrimination between sub-lineages among South American natives (Jota et al. 2011, 2016). Using multiple genotyping approaches, we identified informative SNPs exhibiting restricted Q haplogroup sub-lineages in South American populations (Jota et al. 2016). For example, the Q-Z19483 sub-lineage is associated with population movements of the Late Intermediate Period, as it is distributed in Central Andes, including the Altiplano region, and probably connected with the Inka expansion.

With respect to matrilineal inheritance, South American natives exhibit mainly four mtDNA haplogroups (A2, B2, C1 and D1), which comprise several sub-lineages and minor haplogroups, such as D4h3a (Brandini et al. 2017). The coalescence times for each Native American mtDNA haplogroup are estimated between 14 and 20 kya (Llamas et al. 2016; Brandini et al. 2017; Barbieri et al. 2017), suggesting that the first human settlers arrived in Americas by the end of the Pleistocene (Moreno-Mayar et al. 2018). Furthermore, comparisons of mtDNA haplogroup frequencies between ancient and modern South American populations show distinct regional patterns (Fehren-Schmitz et al. 2015; Barbieri et al. 2017). However, a general continuity in northern, central and southern parts of South America is apparent, with the North and Central Andes displaying high intra-population and low inter-population diversities with a high prevalence of haplogroup B2 due to greater effective population size and continuous gene flow (Valverde et al. 2016; Brandini et al. 2017). Coincidentally, the high prevalence area of the B2 lineage corresponds to the Tawantinsuyu region.

Information on lineages of the Inka nobility and rulers of Tawantinsuyu and their ancestral origins is limited to data gathered by the Spanish chroniclers, anthropologists and modern historians and becomes more uncertain and surrounded by myths as we go further back in time (Gamboa 1572; de la Vega 1609; Cobo 1653; Bauer 1991; Espinoza-Soriano 1997). Although the origins of the Inka family before their settlement in Cusco are still unknown, there are two major legends: (1) an older myth that the Inka ancestors originated from a cave in the district of Pacarictampu in the Province of Paruro, 50 km south of Cusco and (2) an imperial tale that the ancestors originated alongside the Sun at the Isla del Sol on the Bolivian side of Lake Titicaca at 380 km southeast of Cusco (Urton 2004). For many scholars, these alternative origin places are not contradictory and represent successive settlements of the original migration of the Inka ancestors (Espinoza-Soriano 1997; Bauer and Covey 2002; Cerrón-Palomino 2013).

While historical data have been published on families of putative Inka descendants (Amado-Gonzales 2009; Dunbar-Temple 2009), this information is available only for the colonial period until 1824. To identify the present-day patrilineal descendants of the Inka rulers, it is necessary to identify families that were recognised as Inka descendants before 1824 and to reconstruct family genealogies till date. At the start of the colonial period, the putative descendants of Inka rulers were recognised as nobles with rights, were not required to pay tributes and were exempt from forced labour. After the rebellion of Tupac Amaru II when the colonial government implemented political changes related to the native nobility, it became important to again prove noble ancestry. Several documents are available during the period 1780–1824 that not only identify noble families, but also contain data on ancestors and lineages.

The complex society of the pre-Columbian Central Andes is based on ‘ayllu’, a kinship system of families and clans that shared the same land and labour division for several generations (Espinoza-Soriano 1997). During the Inka Empire, new ayllus were founded by different rulers for political and administrative management of Tawantinsuyu, as well as for worship in the afterlife for coming generations. Pachacutec, the first Inka Emperor, was responsible for reorganising the state and imperial family. Ten imperial ayllus, or groups of rulers’ descendants, were distributed into two groups: five for lower (hurin) Cusco from former Manco Capac till Capac Yupanqui and five for higher (hanan) Cusco from Inka Roca till Tupac Yupanqui, the son and successor of Pachacutec (Rostworowski 2001; Zuidema 2007). During the Spanish rule around 1570, Viceroy Francisco de Toledo again reorganised the imperial descendants in Houses, and the Pachacutec ayllu was split into two (Supplemental Table 1) (Gamboa 1572; Rostworowski 2001; Amado-Gonzales 2003). By 1572, the descendants of the Inka royal family had properties assigned by the Spanish crown in six so-called parishes of ‘Indians’ around Cusco: Belén, Hospital de Naturales, San Blas, San Cristobal, San Sebastian and San Jerónimo. The parishes of Santiago and Santa Ana were reserved for other groups (Gamboa 1572). During colonial times, uninterrupted written records of the noble Inka descendants (known as Panakas) could be found on census, municipal, parochial and legal registers until 1824 (Amado-Gonzales 2003). With the formation of the Peruvian Republic, privileges were abolished and these descendants were no longer a separate group. Other Inka descendants were also registered in the Lake Titicaca Basin, including Azángaro, Capachica and Copacabana, on the Peruvian and Bolivian sides of Lake Titicaca (de León 1553; Espinoza-Soriano 1972). A few families have been traced during the colonial period (Amado-Gonzales 2003, 2005, 2009; Dunbar-Temple 2009); these may be used as subjects of research in the absence of mummies of the Inka rulers. Most mummies were probably destroyed by Corregidor Polo de Ondegardo in the late sixteenth century (Deza and Barrera 2001), whereas five that were displayed in the Hospital de San Andrés, Lima, disappeared in the first half of the seventeenth century. Moreover, no evidence has been recovered from the Convent of Santo Domingo in Cusco where the remains of the Inkas of Vilcabamba were buried. With the complete absence of genetic material of former rulers, present-day Panakas families are the only DNA source that could help gain insight into the origin of the Inka rulers.

In this study, we focused on the Panakas families to investigate whether they are related to each other by a patrilineal descent, and if they are genetically linked to populations from Lake Titicaca or to those from Pacarictampu. To tackle these questions, we compared Y-chromosome and mtDNA data from the Panakas with individuals from different provinces of Cusco, including the Pacarictampu district, and many Quechua- and Aymara-speaking populations from Peru (including populations from Lake Titicaca), Bolivia and Ecuador.