AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 148:395–405 (2012)

Y-Chromosome Diversity in Native Reveals Continental Transition of Genetic Structure in the Karla Sandoval,1,2 Andres Moreno-Estrada,2 Isabel Mendizabal,1 Peter A. Underhill,2 Maria Lopez-Valenzuela,1 Rosenda Pen˜ aloza-Espinosa,1,3 Marisol Lopez-Lopez,1,3 Leonor Buentello-Malo,4 Heriberto Avelino,5 Francesc Calafell,1 and David Comas1*

1Institut de Biologia Evolutiva (UPF-CSIC), CEXS-UPF-PRBB, Barcelona, Catalonia, Spain 2Department of Genetics, Stanford University, Stanford, CA 94305 3Departamento de Sistemas Biolo´gicos, Universidad Auto´noma Metropolitana, City, Mexico 4Instituto de Investigaciones Antropolo´gicas, Universidad Nacional Auto´noma de Me´xico, , Mexico 5Department of Linguistics, Max Planck Institute for Evolutionary Anthropology, , Germany

KEY WORDS Y-chromosome ; native Mexican populations; ; genetic diversity; male lineages

ABSTRACT The genetic characterization of Native paternal lineages reveal different narratives of their Mexicans is important to understand multiethnic based population history, with sex-biased continental contribu- features influencing the medical genetics of present tions and different admixture proportions. At a conti- Mexican populations, as well as to the reconstruct the nental scale, we found that populations and the peopling of the Americas. We describe the Y-chromosome northernmost groups from cluster to- genetic diversity of 197 Native Mexicans from 11 popu- gether, but we did not find a clear differentiation within lations and 1,044 individuals from 44 Native American Mesoamerica and the rest of the , which populations after combining with publicly available coupled with the fact that the majority of individuals data. We found extensive heterogeneity among Native from Central and South American samples are re- Mexican populations and ample segregation of Q-M242* stricted to the Q-M3 branch, supports the notion that (46%) and Q-M3 (54%) haplogroups within Mexico. The most Native Americans from Mesoamerica southwards northernmost sampled populations falling outside Meso- are descendants from a single wave of migration. This america (Pima and Tarahumara) showed a clear differ- observation is compatible with the idea that present day entiation with respect to the other populations, which is Mexico might have constituted an area of transition in in agreement with previous results from mtDNA line- the diversification of paternal lineages during the coloni- ages. However, our results point toward a complex zation of the Americas. Am J Phys Anthropol 148:395– genetic makeup of Native Mexicans whose maternal and 405, 2012. VC 2012 Wiley Periodicals, Inc.

The ability to reconstruct human population history well as Y-chromosome markers (Rangel-Villalobos et al., using uniparental systems has motivated a deluge of 2000, 2001a,b, 2008; Tarazona-Santos et al., 2001; Luna- genetic studies during the last decades. However, several Vazquez et al., 2003, 2005, 2008; Hernandez-Gutierrez et geographic remain underrepresented in compre- hensive surveys of human genetic variation. One such is Mesoamerica, a wide area covering central- Additional Supporting Information may be found in the online southern Mexico and part of (including version of this article. regions from Guatemala, Belize, El Salvador, Honduras, Nicaragua, and Costa Rica). Despite different civiliza- Grant sponsor: Agencia Espan˜ ola de Cooperacio´n Internacional tions arising in this region, Mesoamerica comprised an (AECI, Programa de Cooperacio´n Interuniversitaria e Investigacio´n interrelated cultural area that constituted one of the two Cientı´fica entre Espan˜a e Iberoame´rica); Grant number: A/7694/07. Grant sponsor: Direccio´n General de Investigacio´n, Ministerio de centers of the civilization (Carmack et al., Ciencia e Innovacio´n, Spain; Grant number: CGL2010-14944/BOS. 1996; Lo´pez-Austin and Lo´pez-Luja´n, 2001). The impact Grant sponsor: Direccio´ General de Recerca, Generalitat de Catalu- of post-Columbian European and African admixture with nya; Grant number: 2009SGR1101. Grant sponsor: Basque Govern- Native Mexicans is of great interest to understand dis- ment (Hezkuntza, Unibertsitate eta Ikerketa Saila, Eusko Jaurlar- ease susceptibility in Mexican populations (Bryc et al., itza); Grant number: BFI107.4. 2010), as well as to reconstruct the history of Mesoamer- ica and the peopling of the Americas (Mulligan et al., *Correspondence to: David Comas. E-mail: [email protected] 2008). Numerous studies on this region have focused on Mes- Received 26 July 2011; accepted 29 February 2012 tizo populations using autosomal markers (Rangel-Villa- lobos et al., 1999; Cerda-Flores et al., 1987, 2002a,b; DOI 10.1002/ajpa.22062 Cerda-Flores and Garza-Chapa, 1989; Felix-Lopez et al., Published online 11 May 2012 in Wiley Online Library 2006; Wang et al., 2008; Silva-Zolezzi et al., 2009) as (wileyonlinelibrary.com).

VC 2012 WILEY PERIODICALS, INC. 396 K. SANDOVAL ET AL. al., 2005; Licea-Cadena et al., 2006; Torres-Rodriguez et individuals from 44 Native American populations to al., 2006; Gorostiza et al., 2007; Gutierrez-Alarcon et al., determine the distribution of paternal lineages within 2007; Rubi-Castellanos et al., 2009a,b; Salazar-Flores et Mexico in contrast with the continental landscape and al., 2010). These studies have shown a substantial con- evaluated its correlation to geographical, cultural, and tribution of Native American ancestry to extant Mexican linguistic factors. In addition, we contrasted our results Mestizo populations, supported by additional studies with the female counterpart for the same populations as based on mtDNA sequences (Torroni et al., 1994; Green revealed by mtDNA analysis (Sandoval et al., 2009) and et al., 2000; Gonza´lez-Oliver et al., 2001; Pen˜ aloza-Espi- found a remarkable differential contribution between nosa et al., 2007; Sandoval et al., 2009; Kemp et al., maternal and paternal lineages in Mesoamerica. 2010). Other studies have also described the paternal genetic diversity in Mexican indigenous groups (Rangel- MATERIALS AND METHODS Villalobos et al., 2000; Barrot et al., 2005; Felix-Lopez et al., 2006; Paez-Riberos et al., 2006; Kemp et al., 2010); Populations studied however, a better understanding could be gained by put- A total of 197 Native Mexican male individuals were ting this characterization into context with the continen- analyzed: 22 , 18 Tarahumaras, 2 , 7 Oto- tal landscape. Previous studies on continental genetic mis, 22 from Xochimilco, 22 Nahuas from structure identified by autosomal markers (Wang et al., Zitlala, 9 Nahuas from San Pedro Atocpan, 17 Nahuas 2007; Yang et al., 2010), have reported a North-to-South from Santo Domingo Ocotitlan, 8 Purepechas, 19 Yucatec reduction of diversity in the Americas, consistent with Mayas, and 51 Pimas. An additional sample of 11 Que- early studies based on Y-chromosome diversity showing chuas from were also analyzed and used for inter patterns of low haplotype diversity in Native American population comparisons. DNAs from Pima, Yucatec Maya populations (Ruiz-Linares et al., 1999; Underhill et al., and Quechua were provided by the laboratory of Judith 2001; Salzano, 2002). Concerning Y-chromosome varia- R. Kidd and Kenneth K. Kidd (Yale University, New Ha- tion, components of both Q1a3 and C3 subhaplogroups ven, Conn) where lymphoblastoid cell lines are main- are viewed as prehistoric Native paternal lineages in the tained. All samples were collected with the appropriate Americas while all other haplogroups are considered to informed consent and all individuals were native speak- reflect the consequences of post-Columbian contact. The ers with local ancestors; their geographic location is nearest molecular ancestors to Q and C subhaplogroups shown in Figure 1. in Native Americans are rare elsewhere, with the excep- tion of northern , where they reach frequencies of 28 Y chromosome genotyping and 18%, respectively (Mulligan et al., 2004). With the identification of the Q defining M242 SNP DNA concentration of all samples was normalized to 1 (Seielstad et al., 2003), the age of Q haplotypes in the ng/ll using the QuantifilerTM Human DNA Quantifica- Americas was estimated at 15,000–18,000 ybp (Bortolini tion Kit (Applied Biosystems). Three SNPs (M45, which et al., 2003). Although haplogroup Q is present in 76% of defines the major haplogroup P; M242, which defines Native Americans (Zegura et al., 2004), the Native haplogroup Q; and M207, which defines haplogroup R) American C3-M217 associated haplogroups have only were genotyped in all individuals in 5ll reactions. The been detected among the Na-Dene-speaking Tanana, reactions contained 1 ll of DNA template, 0.25 llof203 Navajo, Chipewyan, and the Amerindian Cheyenne SNP Genotyping Assay, 1.25 ll of sterile-filtered water ( et al., 1999; Karafet et al., 1999; Bortolini et al., and 2.5 ll of TaqMan1 Universal PCR Master Mix. Each 2002, 2003; Zegura et al., 2004). The higher haplotypic individual was amplified under standard conditions with diversity of C-M130 lineages in eastern Siberians and TaqMan1 probes (Applied Biosystems). Those individu- Asians (Lell et al., 2002) and an estimated coalescent for als not assigned within the major haplogroup P, were C-M130 of 25,000–30,000 ybp (Karafet et al., 1999; genotyped for marker M130 (which defines haplogroup Underhill et al., 2000), is consistent with an Asian origin C), and a subset of 69 Native Mexicans and 7 Quechua for C-M130, This inference is bolstered by other age esti- individuals belonging to haplogroup Q were genotyped mates for the entire C lineage and the Native American- for marker M3 (Underhill et al., 1996), which defines specific C3-P39 as 27,500 6 10,100 and 2,550 6 1,910 haplogroup Q1a3a within clade Q (hereafter referred to ybp, respectively (Hammer and Zegura, 2002; Karafet et as Q-M3). After the PCR amplification, allelic discrimi- al., 2002). nation was done with SDS Software TM v2.3 (Applied Here, we report the haplogroup composition based on Biosystems) to determine the haplogroups defined by the Y-chromosome STR and SNP data for 197 unrelated indi- SNPs previously described. viduals from eleven native populations from Mexico A total of 17 Y-STR loci (DYS19, DYS385a/b, DYS389I, belonging to two independent cultural and geographical DYS389II, DYS389II, DYS390, DYS391, DYS392, areas: Tarahumara and Pima from (out- DYS393, DYS437, DYS438, DYS439, DYS448, DYS456, side Mesoamerica); and , , , Xochi- DYS458, DYS635, and GATA-H4) were amplified with milco Nahua, Zitlala Nahua, Atocpan Nahua, Ocotitlan the AmpFlSTR1 YfilerTM PCR Amplification Kit Nahua, Purepecha, and Yucatec Maya, all from Central- (Applied Biosystems). PCR products were discriminated South Mexico and within the Mesoamerican area. These in an ABI 3130 3 l genetic analyzer and the allele sizes populations belong to four different linguistic families were determined with GeneMapper v3.2. Software. (Campbell, 1986, 1997; Mithun, 1990, 1999), Uto-Aztecan including Tarahumara, Pima, and the Nahua languages; Statistical analysis Otomanguean including Triqui, Mixtec, and Otomi; Mayan including Yucatec Maya, and finally Tarascan STR-haplotype data was used to estimate intra popu- including the isolate Purepecha. lation genetic diversity parameters such as: gene diver- We combined our data with additional continental pop- sity, mean pairwise differences and theta values using ulations to analyze the Y-chromosome diversity in 1,044 Arlequin program v3.11 (Schneider et al., 2005). To com-

American Journal of Physical Anthropology Y-CHROMOSOME DIVERSITY OF NATIVE MEXICAN POPULATIONS 397

Fig. 1. Geographical location of Native Mexican populations sampled and the additional populations included in the continental database. Population codes as shown in Supporting Information Table 1. pare the Mexican data with other Native American pop- southwards) for a total of 1,044 individuals analyzed (see ulations, haplotype information of seven Y-STRs (DYS19, Supporting Information Table S1). DYS389I, DYS389II, DYS390, DYS391, DYS392, and Population genetic structure was tested in Mexican DYS393) were collected from the literature. After com- and Native American populations with the analysis of bining with our Native Mexicans and filtering popula- molecular variance (AMOVA; Excoffier et al., 1992) using tions with extreme low sample sizes (N \ 9) a total of 44 the Arlequin program v3.11 (Excoffier et al., 2005). populations were included in the database and used for Genetic relationships among Mexican and continental further analyses (Supporting Information Table S1). populations were analyzed with two approaches. The Individuals belonging to non-Native American hap- first approach was based on genetic distances (using logroups according to (Bortolini et al., 2003) were RST) calculated with Arlequin program v3.11 (Excoffier excluded for all estimations except for the Correspon- et al., 2005) and represented in a multidimensional scal- dence Analysis (CoA; see below). For some analyses, pop- ing plot (MDS). The second approach was based on hap- ulation samples were grouped into major subcontinental logroup absolute frequencies to construct a CoA plot. areas: North (675 individuals from 28 populations from Both analyses were performed with STATISTICA 7 pack- to Mexico), Central (79 individuals from 4 popu- age (http://www.statsoft.com). lations from Guatemala to Panama), and Median joining networks were constructed using Net- (290 individuals from 12 populations from Colombia work 4.5.1.0 package (Bandelt et al., 1999) for the Native

American Journal of Physical Anthropology 398 K. SANDOVAL ET AL.

TABLE 1. Y-chromosome haplogroups found in 11 Native Mexican samples and one Quechua population Haplogroup SNPs genotyped frequencies (%) M3 genotyped Population n M242a M207b M45c M130d QRNon PC n M242e M3f M242 (%) M3 (%) Triqui 22 22 – – – 100 – – – 9 9 – 100 – Tarahumara 18 13 1 4 – 72.2 5.6 22.2 – 0 – – – – Mixtec 2 2 – – – 100 – – – 0 – – – – Otomi 7 4 1 2 – 57.1 14.3 28.6 – 0 – – – – Nahua Xochimilco 22 15 – 7 – 68.1 – 31.8 – 7 – 7 – 100 Nahua-Zitlala 22 19 – 3 – 86.3 – 13.7 – 6 3 3 50 50 Nahua San Pedro 9 9 – – – 100 – – – 5 – 5 – 100 Nahua Santo Domingo 17 15 – 2 – 88.2 – 11.8 – 15 8 7 53.3 46.7 Purepecha 8 6 – 2 – 75 – 25 – 4 1 3 25 75 Yucatec Maya 19 14 2 3 – 73.7 10.5 15.8 – 13 2 11 15.4 84.7 Pima 51 49 – 2 – 96.0 – 3.9 – 10 9 1 90 10 Total 197 168 4 25 – 69 32 37 46 54 SNPs genotyped Haplogroup frequencies M3 genotyped (%) Quechua 8 8 – – – 100 – – – 7 1 6 14.3 85.8 a Derived allele: haplogroup Q. b Derived allele: haplogroup R. c Ancestral allele: individuals not belonging to major haplogroup P. d Derived allele: haplogroup C. e Ancestral allele for M3. f Derived allele: haplogroup Q1a3a.

American haplogroup Q (Bortolini et al., 2003). The time Native origin were found in our sample set, including to the most recent common ancestor was estimated for haplogroups outside clade P (12.7%), and R (2.03%), the Q lineage. Y-STRs weights were determined accord- which is the most common haplogroup in European pop- ing to Hurles et al. (2002). The mutation rate considered ulations (Karafet et al., 2008; see Table 1). The presence was one mutation per 571 years, calibrated on the evolu- of these non-Native lineages in Native Mexicans is most tionary rate of each locus reported in the Y-chromosome plausibly the result of recent admixture since colonial haplotype reference database (YHRD, www.yhrd.org) as times. In almost all Mexican populations the admixture of March 23, 2010. proportion was similar except for the Xochimilco Nahuas, where significant differences were observed when compared with Triqui (exact test of differentiation, RESULTS P 5 0.003) and Pima (P 5 0.0007) samples. These last Paternal lineage composition in native Mexicans two populations showed a high percentage of Q native haplogroup (100 and 96%, respectively). Few SNPs are needed to define the major haplogroups Of the 176 individuals belonging to haplogroup Q, 106 found in the Americas since only two major haplogroups different haplotypes (Supporting Information Table S2) are associated with Native American populations: hap- were observed. Table 2 summarizes the results for the logroup Q; and at lower frequency, haplogroup C (Zegura 17 STRs analyzed on 168 Native Mexican samples and et al., 2004). After genotyping four Y-chromosome SNPs, the 8 Quechua samples used for comparison. Interest- the 197 Native Mexican individuals were classified into ingly, Pima and Tarahumara, the northernmost samples, major haplogroups (Table 1). The most frequent hap- exhibited the lowest diversity values (average genetic di- logroup in Native Mexican populations was haplogroup versity and pairwise differences), which are the only Q (85.3%), that is defined by the M242 mutation and non-Mesoamerican populations. constitutes the major lineage among Native Americans, An analysis of molecular variance (AMOVA) based on but is also widely distributed in North and some Rst genetic distances was performed to define the popula- Siberian populations (Karafet et al., 2008). In contrast, tion structure of Mexican populations according to sample within clade Q, haplogroup Q-M3 (Q1a3a) is almost com- size, geographic and linguistic criteria (Table 3). To avoid pletely restricted to the Americas (Zegura et al., 2004). biases due to admixture of different non-Native hap- In a subset of 69 Native Mexicans additionally geno- logroups, AMOVA estimations were performed only with typed for Q-M3, we observed that 54% belong to Q1a3a Native American haplotypes. To determine whether the (Table 1), whereas a much higher proportion (86%) was four Nahua groups had to be considered as a single group, found in a subset of Quechua individuals, indicating that an AMOVA was performed among the Nahua samples. the ancestral lineages of Native Mexicans are less re- Certain heterogeneity (P \ 0.05) among Nahuas was stricted to Q-M3 bearers compared with some South found (10.18%). To corroborate this result, an exact test of American populations. Interestingly, the population population differentiation was performed (P \ 0.05). within Mexico with the largest count of ancestral states Since both results showed a clear differentiation within for Q-M3 was the northernmost group, namely the Pima. the Nahua populations, for most of the analyses they Conversely, the Maya group showed the largest count of were considered as four separate groups, except for those derived states for Q-M3. Haplogroup C, the other major analyses in which sample size was relevant. Particularly, lineage found in Native Americans (so far observed Xochimilco and San Pedro Nahuas were treated as a sin- exclusively in North America), was not observed in our gle group, since they do not represent separate dialects, Native Mexican sample. Other haplogroups with non- are close geographically, and no statistical differences

American Journal of Physical Anthropology Y-CHROMOSOME DIVERSITY OF NATIVE MEXICAN POPULATIONS 399

TABLE 2. Diversity parameters for 11 Native Mexican groups and one Quechua population based on 17 Y-STRs from haplogroup Q individuals Population NKSHaplotype diversity Average GD Mean PW Thetaa Triqui 22 15 16 0.951/20.02 0.471/20.25 8.121/23.92 20.43 (14.54) Tarahumara 13 10 14 0.921/20.06 0.431/20.24 7.381/23.69 10.56 (11.28) Mixtec 2 1 0 0.001/20.00 0.001/20.00 0.001/20.00 UC Otomı´ 4 4 12 1.001/20.17 0.501/20.35 8.501/24.98 UC Xochimilco Nahua 15 10 14 0.951/20.03 0.481/20.26 8.161/24.03 20.08 (19.26) Zitlala Nahua 19 17 17 0.981/20.02 0.551/20.29 9.401/24.52 82.59 (153) San Pedro Nahua 9 7 15 1.001/20.07 0.501/20.30 8.571/24.51 UC Santo Domingo Nahua 15 8 12 0.911/20.04 0.451/20.24 7.651/23.78 9.28 (5.52) Pure´pecha 6 6 16 1.001/20.09 0.551/20.34 9.461/25.07 UC Yucatec Maya 14 13 17 0.981/20.03 0.551/20.30 9.381/24.58 88.08 (260) Pima 49 16 13 0.891/20.02 0.321/20.17 5.601/22.73 7.08 (2.25) Total 168 Quechua 8 7 15 0.961/20.07 0.441/20.26 7.501/23.92 25.27 (59.97)

N, number of individuals; K, number of different haplotypes; S, number of polymorphic loci; Average GD, average gene diversity; Mean PW, mean pairwise differences; UC, Unable to compute theta (Hom) when all gene copies are different.a Theta, in parenthesis standard deviation.

TABLE 3. Analysis of molecular variance (AMOVA) using different grouping structures of Native American populations from Mexico and the America Among Among populations Within Groups groups within groups populations Native Mexicansa 11 populations 16.30* 83.70* 4 linguistic families 22.17 ns 17.64* 84.53* North vs Meso America 8.43* 11.03* 80.55* Native Mexicansb 7 populations 16.80* 83.20* 3 linguistic families 23.09 ns 18.53* 84.56* North vs Meso America 8.22* 11.38* 80.40* America databasec Single group 20.47* 79.53* Geographic regionsd (North, Central, South) 1.22 ns 19.73* 79.05* North Americae vs. all other 18.51* 12.95* 68.54* 6 Cultural regions 6.63* 14.76* 78.61* 22 Linguistic families 11.60* 9.42* 78.98*

* P \ 0.05; ns: nonsignificant.All AMOVAS were based on RST distances a Eleven populations were considered with none sample size restriction. b Seven populations were considered with more than 10 individuals. c Data from 44 Native American populations based on haplotype information from 7 Y-STRs (see Supporting Information Table S1). d Geographic grouping structure as in Supporting Information Table S1. e In this case North America includes Aleut, Choctaw, Creek, Chippewa, Cherokee, Sioux, Dogrib, and Tanana (based on cultural groups as in Fig. 2B). were observed when an AMOVA was performed between considered, significant values were found (8.22%: P \ these two populations (3.58%; P 5 0.2). 0.05). When the linguistic classification (three families The AMOVA showed that when all Mexican popula- represented) was considered, again nonsignificant values tions were considered as a single group (11 populations), were found (23.09%; P 5 0.8). 16.30% (P \ 0.05) of the genetic variance was found In the MDS calculated with Mexican populations that between populations, showing a notable genetic hetero- only belong to haplogroup Q based on RST genetic dis- geneity among Native Mexican populations. When two tances (obtained from the full set of 17 Y-STR loci), Tara- groups (northern samples versus Mesoamerican samples) humara and Pima were located apart from the rest in were compared, the AMOVA showed significant differen- the first dimension, whereas the rest of the populations ces between the two groups: 8.43% (P \ 0.05). Nonsigni- did not show a clear clustering (Fig. 2A). Additionally, ficant proportions of the variance (22.17%; P 5 0.7) for the Correspondence Analysis (CoA) performed with were explained by the fact of belonging to one of the four Mexican populations (Supporting Information Fig. S1), language families in the sample; this suggests a lack of only samples with more than 10 individuals (only 7 correlation between Y-chromosome STR haplotype diver- groups) were considered (Mixtec, Otomi, and Purepecha sity and linguistic affiliation. When only populations were excluded). The first dimension, which explains with at least 10 individuals (7 groups) were considered 88.19% of the variation, separates Q-M242 and Q-M3 (Mixtec, Purepechas, and Otomi were discarded due to haplogroups. Pimas and Mayas were projected closer to their low sample size, and Nahuas from Xochimilco and Q-M242 and Q-M3, respectively. The second dimension San Pedro were considered together as a single group), was driven by the presence of non-Native (R and non-P) 16.80% (P \ 0.05) of the genetic variance was found lineages in some populations, mainly Tarahumara, between all populations, showing again notable genetic Nahua from Xochimilco, and Maya, denoting higher pro- heterogeneity among Native Mexican populations. When portions of European admixture in these populations cultural groups (northern versus Mesoamericans) were (Supporting Information Fig. S1).

American Journal of Physical Anthropology 400 K. SANDOVAL ET AL.

Fig. 3. Two-dimension CoA based on absolute frequencies of the three major native haplogroups (Q-M242, Q-M3, and C) reported in all available datasets. Twenty-seven Native Ameri- can populations included. Mexican populations were classified separately to contrast their haplogroup composition against North, Central, and South American populations. Selected popu- lations are labeled within each group to ease interpretation.

M242 and Q-M3. It is noteworthy that Native Mexicans are the only regional group with populations represented next to both the Q-M242 cluster and the Q-M3 cluster. In contrast, all Central and South American samples were grouped significantly closer to the Q-M3 hap- logroup (Fig. 3). A broader picture was obtained when MDS analyses were performed at a continental scale using RST genetic distances based on 7 STRs from the full panel of 44 pop- Fig. 2. Genetic relationships among Mexican and continen- ulations (Fig. 2B). In the two-dimension plot (stress tal populations. A: Two-dimensional MDS plot of Native Mexi- value 5 0.142) it is clear that the first dimension is cans based on RST genetic distances using 17 Y-STR loci (stress 5 driven by a strong differentiation of the northernmost value 0.090). Labels in bold indicate non-Mesoamerican popu- populations from the rest: Dogrib and Tanana, both lations. B: Two-dimensional MDS plot of Native American popu- belonging to the same linguistic family (Athapaskan), lations from the continental database based on RST genetic dis- tances using 7 Y-STR loci reported in all available datasets Aleuts (), Chippewa (North- (stress value 5 0.142). A total of 44 populations were grouped east USA), and Sisseton Wahpeton-Sioux (Northeast into 6 major cultural areas. USA). The second dimension appears to be driven by highly isolated tribes, such as the two Yanomami groups from the Amazon, and Bari from Venezuela. No clear Mexican populations within the American clusters were observed among the rest of populations. genetic landscape To formally test whether Native American samples clustered, an AMOVA considering the American data- To compare the genetic diversity found in Mexico with base was performed (Table 3). When samples were con- the rest of Native American populations at a continental sidered as a single group, a high degree of genetic heter- scale, a comprehensive database of previously reported ogeneity between populations 20.47% was found (P \ Y-STR and SNP data was constructed and combined 0.05). To ascertain whether this heterogeneity was with our data. A total of 44 populations with available caused by geographical factors, three continental groups data for a common subset of 7 STRs were included in (North, Central, and South) were considered. The the database. Out of those, 27 populations (7 from North AMOVA showed that nonsignificant 1.22% (P [ 0.1) of America, 12 from Mexico, 4 from Central America, and 4 the genetic variance was due to differences between from South America) had all three major native hap- these three subcontinental groups, whereas the genetic logroups reported (C-M130, Q-M242, and Q-M3), and variance within the three groups remained highly signif- were thus used to perform a CoA based on absolute hap- icant (19.73%; P \ 0.01), suggesting an important logroup frequencies (Fig. 3). The first dimension of the genetic heterogeneity within the subcontinental areas CoA (60.53%) separates Q-M3 from the rest, and the sec- and a lack of continental structure. One plausible expla- ond dimension (39.47%) C-M130 from the rest. In agree- nation for this lack of structure is that when considering ment with the known distribution of haplogroup C, we all territories from Mexico northwards as a part of the observed that the two northernmost populations of this same group, the existing differences between the north- panel (Chippewa and Sioux) cluster next to C-M130 and ern populations and the rest is homogenized by the the rest of populations show varying proportions of Q- inclusion of Mesoamerican populations. In fact, when

American Journal of Physical Anthropology Y-CHROMOSOME DIVERSITY OF NATIVE MEXICAN POPULATIONS 401

Fig. 4. Network of haplogroup Q in Native Mexican samples for 17 Y-STRs. Color code: Triqui (red), Tarahumara (yellow), Mix- tec (black), Otomi (aqua), Nahua (blue), Purepechas (pink), Yucatec Maya (orange), and Pima (green). Area of the network in pale blue indicates a branch restricted to Pima and Tarahumara individuals and was dated separately (see main text for details). continental populations were grouped into only two cate- were included. A star-like pattern without population gories (North America versus all other), highly signifi- clustering was observed, with the exception of one clus- cant differences were obtained (18.51%; P \ 0.01), in ter mainly formed by Pima individuals (Fig. 4). The esti- agreement with the observed clustering pattern in the mated age for haplogroup Q in Native Mexicans was MDS analysis (Fig. 2B). Furthermore, when populations 6,436 (6913) ybp. This cluster of mainly Pima individu- were grouped into six major cultural areas (North Amer- als (shaded area in Fig. 4) comprised 10 different haplo- ica, , Mesoamerica, Central America, Ama- types, from which 9 were found in Pimas (30 individuals) zon, and ), the differences among groups (6.63%) and one in Tarahumaras (4 individuals). It is noteworthy remained significant (P \ 0.05). that these two populations are the northernmost sam- To test the correlation between paternal lineages and ples in the analysis and that they do not belong to what continental linguistic groups, an additional AMOVA was is strictly defined as Mesoamerica. This cluster was performed classifying the 44 Native American popula- characterized by one haplotype shared between 12 Pimas tions into 22 different linguistic families (see Supporting (13, 13, 17, 12, 16, 23, 10, 14, 13, 14, 11, 12, 19, 17, 15, Information Table S1). It is noteworthy that significant 22, and 11, same STR order as the one described in Y differences between linguistic groups were found: 11.60% chromosome genotyping section). The age estimation for (P \ 0.01), suggesting a genetic differentiation of linguis- this cluster was 2,486 (6 785) ybp. tic groups in the Americas when a broader representa- tion of continental linguistic families are considered, but not within linguistic Mexican groups. DISCUSSION We have explored the Y-chromosome diversity of Analysis of haplogroup Q in native Mexicans Native Mexican populations and contrasted to continen- tal patterns of variation and distribution across the To establish the genetic relationships between the Y- Americas. Due to the presence of a substantial propor- STRs haplotypes observed in our sample set, a median tion of non-Native American (mainly European) male joining network was constructed for haplogroup Q, in lineages in our samples (see Table 1), the Y-chromosome which only Native Mexican haplotypes with 17 STRs haplogroup composition of Native Mexican populations

American Journal of Physical Anthropology 402 K. SANDOVAL ET AL. shows a notable contrast with previous results from the northern populations from the Arctic and North Amer- mtDNA counterpart for the same populations (\0.5%, ica. Likewise, we did not find a correlation between lin- Sandoval et al., 2009). These results suggest a sexual guistic affiliation and Y-chromosome diversity within bias in the introgression of non-Native American line- Mexico, which is also consistent with previous observa- ages in native populations, consistent with previous tions of mtDNA diversity in Native Mexicans (Sandoval studies performed in different Latin-American popula- et al., 2009). However, some correlation between linguis- tions (Dipierri et al., 1998; Sans et al., 2002; Carvajal- tic and genetic diversity is found for the paternal line- Carmona et al., 2003; Gonzalez-Andrade et al., 2007; ages at a continental scale, suggesting a genetic link Mendizabal et al., 2008). In addition, there is variation between linguistically close populations regardless their in the Y-chromosome admixture proportions among geographical location. Most of the genetic and linguistic Native Mexican populations, with the highest proportion correlations found in human populations have been of non-Native haplotypes (32%) in the urban cosmopoli- attributed to their correlation with geography; that is, tan sample of Xochimilco Nahuas, whereas Triqui and genes and languages are correlated because of geography Pima (with the highest proportion of Native American (Comas et al., 2008). In the analysis of male lineages in haplotypes) are two indigenous groups well known to be Native Americans, this correlation is primarily captured characterized by their geographic and cultural isolation. at the very broad continental scale and when North Interestingly, Pima and Tarahumara, the northern- American populations are contrasted with the rest, prob- most samples, exhibit the lowest diversity values, based ably as a result of distinct migration waves leaving a on haplogroup Q composition (average genetic diversity deep genetic signature in these major groups. Most and pairwise differences, see Table 2). This observation likely, many more markers (even hundreds of thousands is in total agreement with our previous results based on and genome-wide distributed) are needed to have enough mtDNA diversity of the same set of samples, where we power to explore fine-scale patterns of genetic structure found a clear differentiation of the two northern popula- within sub-continental regions of the Americas. tions from the rest (Sandoval et al., 2009). In addition, Finally, we conducted a phylogenetic analysis on the they present significant differences in Y-chromosome native haplotypes to perform some historical inferences variation compared to the Mesoamerican samples as in Mexican populations. Haplogroup Q (defined by M242 shown in the AMOVAs, and some of their lineages clus- mutation) is one of the two main branches of the major tered differentially in the haplogroup Q network (see the Y-chromosome lineage P, which originated 26,600– highlight cluster in Fig. 4). Pima and Tarahumara are 41,400 ybp (Karafet et al., 2008). It has been established the only sampled populations outside the cultural, lin- that the bearers of M242 mutation migrated eastward guistic, and geographic region of Mesoamerica. Different across until they reached the north-eastern point geographic and climate conditions between North Amer- of Asia through the Bering ice corridor about 15,000– ica and Mesoamerica might have created over millennia 20,000 years ago (Cavalli-Sforza et al., 1994; Torroni et this genetic discontinuity between both geographical al., 1994; Ray et al., 2010). Haplogroup Q represents a areas. recent paternal founder for the Native American popula- An additional evidence of differentiation was obtained tions into the peopling of the Americas (Bortolini et al., by directly genotyping the M3 mutation that defines the 2003; Seielstad et al., 2003). The diversity of haplogroup Q1a3a haplogroup within clade Q. We found ample segre- Q in Native Mexicans is around 6,436 (1913) ybp, that gation of both Q-M242* and Q-M3 haplogroups within is, well within the aforementioned ranges, pointing to a Mexico, with Pima and Triqui showing the lowest and recent origin of these haplotypes. It is noteworthy the Maya the highest counts of Q-M3 individuals. This sce- common and recent origin of a cluster in Pima and Tara- nario differs drastically with other regions at a continen- humara, which could be related to a common expansion tal scale. The fact that we observed the vast majority of in the area located northern of the Mesoamerica border. individuals from Central and South American popula- Overall, our results offer a detailed picture of the tions to be restricted to the Q-M3 branch, is compatible male-specific genetic composition of a dense panel of with the notion that most Native Americans from Mesoa- Native Mexican populations and reveal that Mesoameri- merica southwards are descendants from a single wave of can populations played a crucial role in patterning the migration. In contrast, paternal lineages in Mexico trace subsequent diversification of paternal lineages during back their origin not only to the Q-M3 branch but also to the colonization of the Americas. A north to south gradi- other branches outside Q-M3 within haplogroup Q. This ent of decreasing diversity has been well documented observation supports the idea that present day Mexico before, yet here we are able to locate a region in which might have constituted an area of transition in the diver- one of the major breakpoints of diversity took place at a sification of paternal lineages during the colonization of continental scale as revealed by Y-chromosome markers. the Americas, with a larger proportion of individuals Further studies at larger, genome-wide, scales integrated being more remotely related to the Q-M3 primogenitor, in with further linguistic and archaeological knowledge are particular those from Northern Mexico such as the Pima necessary to extend the characterization of Native Amer- or extremely isolated groups such as the Triqui. ican genomes to a much finer scale. Overall, we found extensive heterogeneity among Native Mexican populations where the strongest signal of differentiation is driven by the northern populations ACKNOWLEDGMENTS outside Mesoamerica, both within Mexico and at the con- tinental scale. With the modest resolution provided by a The authors would like to thank Mo`nica Valle´s (Unitat limited number of markers, we found a lack of genetic de Biologia Evolutiva, UPF), Ste´phanie Plaza and Roger structure within Mesoamerica and the rest of the conti- Anglada (Servei Geno`mica, UPF) for their technical sup- nent, which is likely the result of a shared recent ances- port; and Kenneth Kidd and Judith Kidd (Yale Univer- try from Mesoamerica southwards resulting in lower lev- sity) for kindly providing Pima, Maya and Quechua els of differentiation compared with the more divergent samples.

American Journal of Physical Anthropology Y-CHROMOSOME DIVERSITY OF NATIVE MEXICAN POPULATIONS 403 LITERATURE CITED lotypes: application to human mitochondrial DNA restriction data. Genetics 131:479–491. Bandelt HJ, Forster P, Rohl A. 1999. Median-joining networks Felix-Lopez XA, Arguello-Garcia R, Cerda-Flores RM, Penaloza- for inferring intraspecific phylogenies. Mol Biol Evol 16:37– Espinoza RI, Buentello-Malo L, Estrada-Mena FJ, Ramos- 48. Kuri M, Gomez FS, Arenas-Aranda DJ. 2006. FMR1 CGG Barrot C, Sa´nchez C, Ortega M, Gonzalez-Martin A, Brand- repeat distribution and linked microsatellite-SNP haplotypes Casadevall C, Gorostiza A, Huguet E, Corbella J, Gene M. in normal Mexican Mestizo and indigenous populations. Hum 2005. Characterisation of three Amerindian populations from Biol 78:579–598. State (Mexico) by 15 STR-PCR polymorphisms. Int J Gonzalez-Andrade F, Sanchez D, Gonzalez-Solorzano J, Gascon Legal Med 119:111–115. S, Martinez-Jarreta B. 2007. Sex-specific genetic admixture of Bergen AW, Wang CY, Tsai J, Jefferson K, Dey C, Smith KD, Mestizos, Amerindian Kichwas, and Afro-Ecuadorans from Park SC, Tsai SJ, Goldman D. 1999. An Asian-Native Ameri- Ecuador. Hum Biol 79:51–77. can paternal lineage identified by RPS4Y resequencing and Gonza´lez-Oliver A, Ma´rquez-Morfin L, Jimenez JC, Torre- by microsatellite haplotyping. Ann Hum Genet 63:63–80. Blanco A. 2001. Founding Amerindian mitochondrial DNA Bortolini MC, Salzano FM, Bau CH, Layrisse Z, Petzl-Erler ML, lineages in ancient Maya from Xcaret, Quintana Roo. Am J Tsuneto LT, Hill K, Hurtado AM, Castro-De-Guerra D, Bed- Phys Anthropol 116:230–235. oya G, Ruiz-Linares A. 2002. Y-chromosome biallelic polymor- Gorostiza A, Gonzalez-Martin A, Ramirez CL, Sanchez C, Bar- phisms and Native American population structure. Ann Hum rot C, Ortega M, Huguet E, Corbella J, Gene M. 2007. Allele Genet 66:255–259. frequencies of the 15 AmpF/Str Identifiler loci in the popula- Bortolini MC, Salzano FM, Thomas MG, Stuart S, Nasanen SP, tion of Metztitlan (Estado de Hidalgo), Mexico. Forensic Sci Bau CH, Hutz MH, Layrisse Z, Petzl-Erler ML, Tsuneto LT, Int 166:230–232. Hill K, Hurtado AM, Castro-de-Guerra D, Torres MM, Groot Green LD, Derr JN, Knight A. 2000. mtDNA affinities of the peo- H, Michalski R, Nymadawa P, Bedoya G, Bradman N, Labuda ples of north-central Mexico. Am J Hum Genet 66:989–998. D, Ruiz-Linares A. 2003. Y-chromosome evidence for differing Gutierrez-Alarcon AB, Moguel-Torres M, Leon-Jimenez AK, ancient demographic histories in the Americas. Am J Hum Cuellar-Nevarez GE, Rangel-Villalobos H. 2007. Allele and Genet 73:524–539. haplotype distribution for 16 Y-STRs (AmpFlSTR Y-filer kit) Bryc K, Velez C, Karafet T, Moreno-Estrada A, Reynolds A, in the state of at North Center of Mexico. Legal Auton A, Hammer M, Bustamante CD, Ostrer H. 2010. Collo- Med (Tokyo) 9:154–157. quium paper: genome-wide patterns of population structure Hammer MF, Zegura SL. 2002. The human Y chromosome hap- and admixture among Hispanic/Latino populations. Proc Natl logroup tree: nomenclature and phylogeography of its major Acad Sci USA 107Suppl 2:8954–8961. divisions. Annu Rev Anthropol 31:303–321. Campbell L. 1997. American Indian languages: the historical Hernandez-Gutierrez S, Hernandez-Franco P, Martinez-Tripp S, linguistics of Native America. New York: Oxford University Ramos-Kuri M, Rangel-Villalobos H. 2005. STR data for 15 Press. loci in a population sample from the central region of Mexico. Campbell L, Kaufman T, Smith-Stark T. 1986. Meso-America as Forensic Sci Int 151:97–100. a linguistic area. Language 62:530–558. Hurles ME, Nicholson J, Bosch E, Renfrew C, Sykes BC, Carmack RM, Janine LG, Gary HG. 1996. Legacy of Mesoamer- Jobling MA. 2002. Y chromosomal evidence for the origins of ica: the history and culture of a Native American civilization. oceanic-speaking peoples. Genetics 160:289–303. Upper Saddle River, NJ: Prentice Hall. Karafet TM, Mendez FL, Meilerman MB, Underhill PA, Zegura Carvajal-Carmona LG, Ophoff R, Service S, Hartiala J, Molina SL, Hammer MF. 2008. New binary polymorphisms reshape J, Leon P, Ospina J, Bedoya G, Freimer N, Ruiz-Linares A. and increase resolution of the human Y chromosomal hap- 2003. Genetic demography of Antioquia (Colombia) and the logroup tree. Genome Res 18:830–838. Central Valley of Costa Rica. Hum Genet 112:534–541. Karafet TM, Osipova LP, Gubina MA, Posukh OL, Zegura SL, Cavalli-Sforza L, Menozzi P, Piazza A. 1994. The history and ge- Hammer MF. 2002. High levels of Y-chromosome differentiation ography of human genes. Princeton: Princeton University among native Siberian populations and the genetic signature of Press. a boreal hunter-gatherer way of life. Hum Biol 74:761–789. Cerda-Flores RM, Budowle B, Jin L, Barton SA, Deka R, Chak- Karafet TM, Zegura SL, Posukh O, Osipova L, Bergen A, Long raborty R. 2002a. Maximum likelihood estimates of admixture J, Goldman D, Klitz W, Harihara S, de Knijff P, Wiebe V, Grif- in northeastern Mexico using 13 short tandem repeat loci. Am fiths RC, Templeton AR, Hammer MF. 1999. Ancestral Asian J Hum Biol 14:429–439. source(s) of new world Y-chromosome founder haplotypes. Am Cerda-Flores RM, Garza-Chapa R. 1989. Variation in the gene J Hum Genet 64:817–831. frequencies of three generations of humans from Monterrey, Kemp BM, Gonza´lez-Oliver A, Malhi RS, Monroe C, Schroeder Nuevo Leon, Mexico. Hum Biol 61:249–261. KB, McDonough J, Rhett G, Resende´z A, Pen˜aloza-Espinosa Cerda-Flores RM, Ramirez-Fernandez E, Garza-Chapa R. 1987. RI, Buentello-Malo L, Gorodesky C, Smith DG. 2010. Evaluat- Genetic admixture and distances between populations from ing the Farming/Language Dispersal Hypothesis with genetic Monterrey, Nuevo Leon, Mexico, and their putative ancestral variation exhibited by populations in the Southwest and Mes- populations. Hum Biol 59:31–49. oamerica. Proc Natl Acad Sci USA 107:6759–6764. Cerda-Flores RM, Villalobos-Torres MC, Barrera-Saldana HA, Lell JT, Sukernik RI, Starikovskaya YB, Su B, Jin L, Schurr Cortes-Prieto LM, Barajas LO, Rivas F, Carracedo A, Zhong TG, Underhill PA, Wallace DC. 2002. The dual origin and Si- Y, Barton SA, Chakraborty R. 2002b. Genetic admixture in berian affinities of Native American Y chromosomes. Am J three Mexican Mestizo populations based on D1S80 and HLA- Hum Genet 70:192–206. DQA1 loci. Am J Hum Biol 14:257–263. Licea-Cadena RA, Rizzo-Juarez RA, Muniz-Lozano E, Paez- Comas D, Bosch E, Calafell F. 2008. Human genetics and lan- Riberos LA, Rangel-Villalobos H. 2006. Population data of guages. Encyclopedia of life sciences. Chichester: Wiley.http:// nine STRs of Mexican-mestizos from (Central South- wwwelsnet/ [DOI: 101002/9780470015902a0020810]. Eastern, Mexico). Legal Med (Tokyo) 8:251–252. Dipierri JE, Alfaro E, Martinez-Marignac VL, Bailliet G, Bravi Lo´pez-Austin A, Lo´pez-Luja´n L. 2001. El pasado indı´gena. His- CM, Cejas S, Bianchi NO. 1998. Paternal directional mating toria FdHdlAS, editor. Me´xic\o D.F.: Fondo de Cultura Eco- in two Amerindian subpopulations located at different alti- no´mica y El Colegio de Me´xico. tudes in northwestern Argentina. Hum Biol 70:1001–1010. Luna-Vazquez A, Vilchis-Dorantes G, Aguilar-Ruiz MO, Bau- Excoffier L, Laval G, Schneider S. 2005. Arlequin (version 3.0): tista-Rivas A, Perez-Garcia A, Orea-Ochoa R, Villanueva-Her- an integrated software package for population genetics data nandez D, Munoz-Valle JF, Rangel-Villalobos H. 2008. Haplo- analysis. Evol Bioinform Online 1:47–50. type frequencies of the PowerPlex Y system in a Mexican- Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecu- Mestizo population sample from Mexico City. Forensic Sci Int lar variance inferred from metric distances among DNA hap- Genet 2:e11–13.

American Journal of Physical Anthropology 404 K. SANDOVAL ET AL.

Luna-Vazquez A, Vilchis-Dorantes G, Aguilar-Ruiz MO, Bautista- Bertranpetit J, Feldman MW, Goldstein DB. 1999. Microsatel- Rivas A, Rojo-Nava AL, Rios-Barrios E, Rangel-Villalobos H. lites provide evidence for Y chromosome diversity among the 2005. Population data for 15 loci (Identifiler Kit) in a sample founders of the New World. Proc Natl Acad Sci USA 96:6312– from the . Legal Med (Tokyo) 7:331–333. 6317. Luna-Vazquez A, Vilchis-Dorantes G, Paez-Riberos LA, Munoz- Salazar-Flores J, Dondiego-Aldape R, Rubi-Castellanos R, Valle F, Gonzalez-Martin A, Rangel-Villalobos H. 2003. Popu- Anaya-Palafox M, Nun˜ o-Arana I, Canseco-Avila LM, Flores- lation data of nine STRs of Mexican-Mestizos from Mexico Flores G, Morales-Vallejo ME, Barojas-Pe´rez N, Mun˜ oz-Valle City. Forensic Sci Int 136:96–98. JF, Campos-Gutie´rrez R, Rangel-Villalobos H. 2010. Popula- Mendizabal I, Sandoval K, Berniell-Lee G, Calafell F, Salas A, tion structure and paternal admixture landscape on present- Martinez-Fuentes A, Comas D. 2008. Genetic origin, admix- day Mexican-Mestizos revealed by Y-STR haplotypes. Am J ture, and asymmetry in maternal and paternal human line- Hum Biol 22:401–409. ages in Cuba. BMC Evol Biol 8:213. Salzano FM. 2002. Molecular variability in Amerindians: wide- Mithun M. 1990. Studies of North American Indian languages. spread but uneven information. An Acad Bras Cienc 74:223–263. Ann Rev Anthropol 19:309–330. Sandoval K, Buentello-Malo L, Penaloza-Espinosa R, Avelino H, Mithun M. 1999. The languages of Native North America. Cam- Salas A, Calafell F, Comas D. 2009. Linguistic and maternal bridge, UK: Cambridge University Press. genetic diversity are not correlated in Native Mexicans. Hum Mulligan CJ, Hunley K, Cole S, Long JC. 2004. Population Genet 126:521–531. genetics, history, and health patterns in Native Americans. Sans M, Weimer TA, Franco MH, Salzano FM, Bentancor N, Annu Rev Genomics Hum Genet 5:295–315. Alvarez I, Bianchi NO, Chakraborty R. 2002. Unequal contri- Mulligan CJ, Kitchen A, Miyamoto MM. 2008. Updated three- butions of male and female gene pools from parental popula- stage model for the peopling of the Americas. PLoS One tions in the African descendants of the city of Melo, . 3:e3199. Am J Phys Anthropol 118:33–44. Paez-Riberos LA, Munoz-Valle JF, Figuera LE, Nuno-Arana I, Schneider S, Excoffier L, Laval G. 2005. Arlequin ver. 3.0: an Sandoval-Ramirez L, Gonzalez-Martin A, Ibarra B, Rangel- integrated software package for population genetics data anal- Villalobos H. 2006. Y-linked haplotypes in Amerindian chro- ysis. Evol Bioinform Online 1:47–50. mosomes from Mexican populations: genetic evidence to the Seielstad M, Yuldasheva N, Singh N, Underhill P, Oefner P, dual origin of the tribe. Legal Med (Tokyo) 8:220–225. Shen P, Wells RS. 2003. A novel Y-chromosome variant puts Pen˜ aloza-Espinosa RI, Arenas-Aranda D, Cerda-Flores RM, an upper limit on the timing of first entry into the Americas. Buentello-Malo L, Gonza´lez-Valencia G, Torres J, Alvarez B, Am J Hum Genet 73:700–705. Mendoza I, Flores M, Sandoval L, Loeza F, Ramos I, Mun˜ oz Silva-Zolezzi I, Hidalgo-Miranda A, Estrada-Gil J, Fernandez- L, Salamanca F. 2007. Characterization of mtDNA hap- Lopez JC, Uribe-Figueroa L, Contreras A, Balam-Ortiz E, del logroups in 14 Mexican indigenous populations. Hum Biol Bosque-Plata L, Velazquez-Fernandez D, Lara C, Goya R, 79:313–320. Hernandez-Lemus E, Davila C, Barrientos E, March S, Jime- Rangel-Villalobos H, Jaloma-Cruz AR, Cerda-Aguilar L, Rios- nez-Sanchez G. 2009. Analysis of genomic diversity in Mexi- Angulo CD, Mendoza-Carrera F, Patino-Garcia B, Sandoval- can Mestizo populations to develop genomic medicine in Mex- Ramirez L, Figuera-Villanueva LE. 2001a. [The genetic DNA ico. Proc Natl Acad Sci USA 106:8611–8616. trace in men: chromosome Y haplotypes in a Mexican popula- Tarazona-Santos E, Carvalho-Silva DR, Pettener D, Luiselli D, tion, analyzing 5 STRs]. Rev Invest Clin 53:401–406. De Stefano GF, Labarga CM, Rickards O, Tyler-Smith C, Rangel-Villalobos H, Jaloma-Cruz AR, Sandoval-Ramirez L, Pena SD, Santos FR. 2001. Genetic differentiation in South Velarde-Felix JS, Gallegos-Arreola MP, Figuera LE. 2001b. Y- Amerindians is related to environmental and cultural diver- chromosome haplotypes for six short tandem repeats (STRs) sity: evidence from the Y chromosome. Am J Hum Genet in a Mexican population. Archaeol Med Res 32:232–237. 68:1485–1496. Rangel-Villalobos H, Munoz-Valle JF, Gonzalez-Martin A, Goros- Torres-Rodriguez M, Martinez-Cortes G, Paez-Riberos LA, San- tiza A, Magana MT, Paez-Riberos LA. 2008. Genetic admix- doval L, Munoz-Valle JF, Ceballos-Quintal JM, Pinto-Esca- ture, relatedness, and structure patterns among Mexican pop- lante D, Rangel-Villalobos H. 2006. Forensic potential of the ulations revealed by the Y-chromosome. Am J Phys Anthropol STR DXYS156 in Mexican populations: inference of X-linked 135:448–461. allele null. Legal Med (Tokyo) 8:52–54. Rangel-Villalobos H, Rivas F, Sandoval L, Ibarra B, Garcia-Car- Torroni A, Chen YS, Semino O, Santachiara-Beneceretti AS, vajal ZY, Cantu JM, Figuera LE. 2000. Genetic variation Scott CR, Lott MT, Winter M, Wallace DC. 1994. mtDNA and among four Mexican populations (Huichol, Purepecha, Tara- Y-chromosome polymorphisms in four Native American popu- humara, and Mestizo) revealed by two VNTRs and four STRs. lations from southern Mexico. Am J Hum Genet 54:303–318. Hum Biol 72:983–995. Underhill PA, Jin L, Zemans R, Oefner PJ, Cavalli-Sforza LL. Rangel-Villalobos H, Rivas F, Torres-Rodriguez M, Jaloma-Cruz 1996. A pre-Columbian Y chromosome-specific transition and AR, Gallegos-Arreola MP, Lo´pez-Satow J, Cantu JM, Figuera its implications for human evolutionary history. Proc Natl LE. 1999. Allele frequency distributions of six Amp-FLPS Acad Sci USA 93:196–200. (D1S80, APO-B, VWA, TH01, CSF1PO and HPRTB) in a Mex- Underhill PA, Passarino G, Lin AA, Shen P, Mirazon Lahr M, ican population. Forensic Sci Int 105:125–129. Foley RA, Oefner PJ, Cavalli-Sforza LL. 2001. The phylogeog- Ray N, Wegmann D, Fagundes NJ, Wang S, Ruiz-Linares A, raphy of Y chromosome binary haplotypes and the origins of Excoffier L. 2010. A statistical evaluation of models for the modern human populations. Ann Hum Genet 65:43–62. initial settlement of the American continent emphasizes the Underhill PA, Shen P, Lin AA, Jin L, Passarino G, Yang WH, importance of gene flow with Asia. Mol Biol Evol 27:337–345. Kauffman E, Bonne´-Tamir B, Bertranpetit J, Francalacci P, Rubi-Castellanos R, Anaya-Palafox M, Mena-Rojas E, Bautista- Ibrahim M, Jenkins T, Kidd JR, Mehdi SQ, Seielstad MT, Espana D, Munoz-Valle JF, Rangel-Villalobos H. 2009a. Wells RS, Piazza A, Davis RW, Feldman MW, Cavalli-Sforza Genetic data of 15 autosomal STRs (Identifiler kit) of three LL, Oefner PJ. 2000. Y chromosome sequence variation and Mexican Mestizo population samples from the States of the history of human populations. Nat Genet 26:358–361. (West), (Center), and Yucatan (Southeast). For- Wang S, Lewis CM, Jakobsson M, Ramachandran S, Ray N, ensic Sci Int Genet 3:e71–76. Bedoya G, Rojas W, Parra MV, Molina JA, Gallo C, Mazzotti Rubi-Castellanos R, Martinez-Cortes G, Munoz-Valle JF, Gonza- G, Poletti G, Hill K, Hurtado AM, Labuda D, Klitz W, Bar- lez-Martin A, Cerda-Flores RM, Anaya-Palafox M, Rangel-Vil- rantes R, Bortolini MC, Salzano FM, Petzl-Erler ML, Tsuneto lalobos H. 2009b. Pre-Hispanic Mesoamerican demography LT, Llop E, Rothhammer F, Excoffier L, Feldman MW, Rosen- approximates the present-day ancestry of Mestizos throughout berg NA, Ruiz-Linares A. 2007. Genetic variation and popula- the territory of Mexico. Am J Phys Anthropol 139:284–294. tion structure in Native Americans. PLoS Genet 3:e185. Ruiz-Linares A, Ortiz-Barrientos D, Figueroa M, Mesa N, Wang S, Ray N, Rojas W, Parra MV, Bedoya G, Gallo C, Poletti Munera JG, Bedoya G, Velez ID, Garcia LF, Perez-Lezaun A, G, Mazzotti G, Hill K, Hurtado AM, Camrena B, Nicolini H,

American Journal of Physical Anthropology Y-CHROMOSOME DIVERSITY OF NATIVE MEXICAN POPULATIONS 405

Klitz W, Barrantes R, Molina JA, Freimer NB, Bortolini MC, W, Barrantes R, Llop E, Rothhammer F, Labuda D, Salzano Salzano FM, Petzl-Erler ML, Tsuneto LT, Dipierri JE, Alfaro FM, Bortolini MC, Excoffier L, Dugoujon JM, Ruiz-Linares EL, Bailliet G, Bianchi NO, Llop E, Rothhammer F, Excoffier A. 2010. Contrasting patterns of nuclear and mtDNA diver- L, Ruiz-Linares A. 2008. Geographic patterns of genome sity in Native American populations. Ann Hum Genet admixture in Latin American Mestizos. PLoS Genet 74:525–538. 4:e1000037. Zegura SL, Karafet TM, Zhivotovsky LA, Hammer MF. 2004. Yang NN, Mazie`res S, Bravi C, Ray N, Wang S, Burley MW, High-resolution SNPs and microsatellite haplotypes point to a Bedoya G, Rojas W, Parra MV, Molina JA, Gallo C, Poletti single, recent entry of Native American Y chromosomes into G, Hill K, Hurtado AM, Petzl-Erler ML, Tsuneto LT, Klitz the Americas. Mol Biol Evol 21:164–175.

American Journal of Physical Anthropology