HLA Genes in Mayos Population from Northeast Mexico

HLA class I and class II alleles have been studied in 60 unrelated people belonging to Mayos ethnic group, which lives in the Mexican Pacific Sinaloa State. Mayos HLA profile was compared to other Amerindians and worldwide populations’ profile. A total of 14,896 chromosomes were used for comparisons. Genetic distances between populations, Neigbour-Joining dendrograms and correspondence analyses were performed to determine the genetic relationship among population. The new specific Mayo HLA haplotypes found are: HLA-A*02-B*35-DRB1*1406-DQB1*0301; HLA-A*02-B*48-DRB1*0404-DQB1*0302; HLA-A*24-B*51-DRB1*0407-DQB1*0302 and HLA-A*02-B*08-DRB1*0407-DQB1*0302. However, the typical Meso American HLADRB1*0407 represents a 40% of all DRB1 alleles. While common HLA characteristics are found in Amerindian distant ethnic groups, still new group specific HLA haplotypes are being found, suggesting that a common founder effect (i.e. high DRB1*0407) is noticed. Moreover, new HLA haplotypes are almost certainly appearing along time probably due to specific pathogen (?) selection for diversity. Mayo language is close to the Tarahumara one (another geographically close group); notwithstanding both groups are not genetically close according to our results, showing again the different evolution of genes and languages, which do not correlate. Finally, Sinaloa is one of the Mexican States in which more European genes are found. However, the results presented in this paper, where no European HLA genes are seen in Mayos, should have a bearing in establishing transplant programs and in HLA and disease studies.


INTRODUCTION
The West Mexican Mayos form an isolated ethnic Amerindian group, which lives on a flat half desert and half cultivated area, Southern Sonora and North Sinaloa, see Fig. (1). They speak a Nahua/Aztec type language with some dialects that do not hinder mutual understanding. They call themselves Yoremes and probably migrated to the present day location from the North (Colorado River Basin) about 1,000 years ago [1]. They usually get married within their own group [1] which is important for the analyses of our results. Whether the Nahua type language (Mayo) also a sign of Mexica or Aztec relatedness is unknown, since Mexicas imposed their language to other probably non-genetically related Mexican groups [2]. Later, the Spanish conqueror Hernan Cortes arrived to the Mexicas Empire in 1,519 AD which was sized and destroyed by war and invaders' borndiseases [3].
The present study aims: 1) To relate Mayos group with other Amerindian groups; and 2) To relate Mayos with other worldwide groups, in order to test the uniqueness of all Amerindians [4] with respect to their HLA profiles (different *Address correspondence to this autor at the Departamento de Inmunologia, Facultad de Medicina, Universidad Complutense de Madrid, Avenida Complutense s/n, 28040 Madrid, Spain; Tel: +34 91 301 7354; Fax: +34 91 301 7213; E-mail: aarnaiz@med.ucm.es; htpp://chopo.pntic.mec.es/biolmol # These authors contributed equally to this work and the order of authorship is arbitrary. loci allele frequencies). This is important for establishing transplant programs in rural areas where some ethnic groups still live, i.e. Mayos, and for studying the epidemiology of HLA linked disorders.
In order to perform this work, a sample of sixty unrelated Mayos has been HLA typed and thereafter the genetic relationship with others Amerindians tested by HLA-A, -B, -DRB1 and -DQB1 DNA typing calculating genetic distances, neighbor-joining dendrograms and correspondence analyses.

Population Samples
We studied sixty unrelated healthy individuals from the Mayos ethnic group. They were HLA class I and class II typed. They were living at Capomos town, Municipio del Fuerte, in Sinaloa State, northeast Mexico. In the Mayos communities the inbreeding pedigrees are common but we selected only those unrelated individuals for our study. Each individual was born in the Mayos area, their four grandparents had been born in the same area and spoke their own Mayo language. All participants gave their written consent for being included in the present study.

HLA Typing and DNA Sequencing
HLA class I (A and B) and class II (DRB1 and DQB1) allele families typing was performed by polymerase chain reaction-sequence specific oligonucleotide (PCR-SSO) reverse dot blot hybridization (Amplicor, Hoffmann La Roche, Basel, Switzerland) (according to IMGT/HLA version 2.6, July 2004). HLA-DRB1 and DQB1 high-resolution typing was done by DNA sequencing in a Perkin Elmer 310 automated DNA sequencer (Foster City, CA, USA). The sequences were analyzed using the IMGT/HLA sequence database (http://www.ebi.ac.uk/imgt/hla/align.html).

Statistical Analysis
Statistical analysis was performed with Arlequin v2.0 software kindly provided by Schneider [5]. In summary, this program calculated HLA-A, -B, -DRB1, and -DQB1 allele frequencies, Hardy-Weinberg equilibrium and the linkage disequilibrium (D'; also named LD) [6] and their level of significance (P) for 2X2 comparisons was determined [7] and the 11 th International Histocompatibility Workshop methodology [6]. In addition, the most frequent complete haplotypes were deduced from: (1) the 2, 3, and 4 HLA loci haplotype frequencies [8]; (2) the previously described haplotypes in other populations [9]; and (3) haplotypes if they appeared in two or more individuals and if the alternative haplotype was well defined [9]. In order to compare phenotype and haplotype HLA frequencies with other populations, the reference tables of the 11 th and 12 th International HLA Workshops are used (see also Table 1) [10,11]. Dendograms were constructed with the allelic frequencies using the neighbor-joining (NJ) method [12] with the genetic distances (DA) between populations [13], using DISPAN software comprising the programs GNKDST and TREEVIEW [14,15]. Correspondence analysis in n-dimensions and its bidimensional representation was carried out using the VISTA v5.02 computer program (http://www.forrest.psych.unc.edu) [16]. Correspondence analysis consists of a geometric technique that may be used for displaying a global view of the relationships among populations according to HLA (or other) allele frequencies. This methodology is based on the allelic frequency variance among populations and on the display of a statistical visualization of differences.

Characteristic HLA Allele Frequencies in the Mayos Population Compared to other Populations
The expected and observed genotype frequency values for HLA-A, -B, and -DQB1 loci were in Hardy-Weinberg equilibrium. Table 2 shows the HLA frequencies found in the Mayos population. HLA-A and -B were typed on 60 individuals. Ten different HLA-A and fifteen different HLA-B alleles were found in the Mayos ethnic group. The most frequent HLA-A alleles were A*02 (allele frequency=0.44) and A*24 (allele frequency=0.36) whereas the most frequent HLA-B alleles were B*35 (allele frequency=0.25), B*39 (allele frequency=0.17) and B*51 (allele frequency= 0.17).
Two types of analysis were done in order to compare Mayos HLA frequencies with other world population's frequencies: 1) with pooled DRB1 and DQB1 data; and 2) with DRB1 only. It was not possible to carry out a study comparing frequencies of HLA class I and II conjointly due to the lack of class I studies in many Amerindian and other populations. The single DRB1 study was carried out in order to compare the American Indian HLA population frequencies with those of Polynesians, Melanesians and Micronesians, who lacked DQB1 analyses, see Table 1, Table 3 and Fig. (2). HLA-DRB1, DQB1 neighbor-joining tree grouped together the Amerindians and separated them from Na-Dene and Eskimo Native American groups and also from the Orientals and Caucasoids, Fig. (3). When the Polynesians, Melanesians and Micronesians were included (see Table 1) the topology of the tree (DRB1 alone) does not change (data not shown), it is also seen in the correspondence analysis, Fig. (2). This effect is more evident when genetic distances are considered ( Table 3).

HLA-A, -B, -DRB1 and -DQB1 Linkage Disequilibrium Analysis in Mayos Compared to other Populations
Associations between different HLA loci were estimated. The most probable two loci haplotype combinations (A-B, B-DRB1 and DRB1-DQB1) were calculated (data not shown).
The ten most frequent extended haplotypes are depicted in Table 4. Mayos extended HLA haplotypes have been obtained for the first time, allowing their comparison with the previously reported ones in other populations ( Table 4 and its footnote).    Fig. (2). Correspondence analysis showing a global view of the relationship among Amerindian, Na-Dene, Eskimo, Asian, European and Polynesian populations according to HLA-DRB1 allele frequencies in n dimensions (bi-dimensional representation). Grey intensity represents an approximate grouping of populations. 1406-DQB1*0301 was associated with A*2-B*35; DRB1* 0404-DQB1*0302 was associated with A*02-B*48 and DRB1*0802-DQB1*0402 was associated with A*02-B*35.
The most frequent extended HLA haplotypes have Amerindian characteristics and it has been found in other Amerindian populations. The most frequent extended HLA haplotype A*24-B*39-DRB1*0407-DQB1*0302 (8.2%) has also been found only in Mazatecan Amerindian population. Six of the ten most frequent Mayos haplotypes have been found in other Amerindians populations, mainly in other Meso-American populations as Mazatecan, Seri, Mixe, Teeneks, Mixtecos, Mayans, Nahuas and Zapotecans. Four of them have not been found in any other worldwide ethnic group, these haplotypes seem to be specific for this Mexican ethnic group ( Table 4 and its footnote).

Mayos and Mexican Populations
We studied an isolated population from Northeast Mexico (Mayos) which confirmed the limited HLA polymorphism found in other isolated tribes from Meso and South America. We detected only ten HLA-A alleles and fifteen HLA-B different alleles in this population. Only two (A*02 and A*24) HLA-A alleles and three (B*35, B*40 and B*51) HLA-B alleles had frequencies higher than 0'1. With respect to HLA class II genes, only two (DRB1*0407 and DRB1*1406) HLA-DRB1 alleles and three (DQB1*0302, DQB1*0301 and DQB1*0402) HLA-DQB1 alleles had frequencies higher than 0'1. All of these HLA class I and II alleles have already been described in other Mexican populations [2,4,17,18,19].
Neighbour-Joining dendrogram and correspondence analyses correlate Mayos tribe together with Mexican populations like Teeneks and Tarahumaras, Fig. (3), and also with Nahuas ethnic group, Fig. (2). Regarding the genetic distances (a bidimensional parameter), Kogi from Colombia, and Mayans from Guatemala are the populations most related to Mayos ( Table 3).
We can deduce from HLA genetic results (allele frequencies and extended haplotypes) that our Mayos sample has not had any contact with European people, because it would be clearly shown by using these particular HLA genetic markers [20].

Mayos and South American Populations
The results of HLA genetic typing in Mayos population confirm that they belong to Amerindian HLA genetic profile.
The presence of HLA-B*48 allele in this population is remarkable (allele frequency= 0'05, Table 2) as it is lacking in all Mexican populations so far except in Andean populations like Aymaras, Quechuas and Lamas [21][22][23].  Table 1. HLA-B*48 is relatively common in Na-Dene and Asian populations [9,24]. The presence of this allele (together with DRB1*0901) in a few Amerindian populations is the only evidence that could strictly support the three wave model of peopling America including Meso and South America [25] because these two alleles would be shared with southern Asian populations.
All Amerindian ethnic groups bearing HLA-B*48 allele in its HLA pofile are living in the Pacific Coast (Mayos and Andean Populations). This might mean the existence of southern Asian migrations to coastal Meso and South America.

Amerindian Languages
Again, while Mayos seem to be genetically related with South American populations [Kogi, Table 1; Terena and Aymaras, Fig. (3); and Wayu and Quechua, Fig. (2)], languages spoken by these groups are disparate and belong to far related linguistic branches.
It is also remarkable that Tarahumara ethnic group, which speaks a similar language to Mayos and is geographically very close to it, genetically shows relatively distant, Table 2, Fig. (2) and Fig. (3). This is a general phenomenon in Amerindians: genes and languages do not correlate.
Finally, Mayos ethnic group lives in Sinaloa Mexican State and at least in our sample European HLA genes and haplotypes are not found; this occurs in spite of that Sinaloa is one of the Mexican States where more European HLA alleles are found [20].