HLA Allele and Haplotype Frequencies in Three Urban Mexican Populations: Genetic Diversity for the Approach of Genomic Medicine

Genetic variability defends us against pathogen-driven antigens; human leucocyte antigens (HLA) is the immunological system in charge of this work. The Mexican mestizo population arises mainly from the mixture of three founder populations; Amerindian, Spaniards, and a smaller proportion of the African population. We describe allele and haplotype frequencies of HLA class I (-A and -B) and class II (-DRB1 and -DQB1), which were analyzed by PCR-SSP in Mexican mestizo from three urban populations of Mexico: Chihuahua-Chihuahua City (n = 88), Mexico City-Tlalpan (n = 330), and Veracruz-Xalapa (n = 84). The variability of the allele HLA class I and class II among the three regions of Mexico are in four alleles: HLA-A*24:02 (36.39%), -B*35:01 (16.04%), -DRB1*04:07 (17.33%), and -DQB1*03:02 (31.47%), these alleles have been previously described in some indigenous populations. We identified 5 haplotypes with a frequency >1%: HLA-A*02:01-B*35:01-DRB1*08:02-DQB1*04:02, A*68:01-B*39:01-DRB1*08:02-DQB1*04:02, A*02:01-B*35:01-DRB1*04:07-DQB1*03:02, A*68:01-B*39:01-DRB1*04:07-DQB1*03:02, and A*01:01-B*08:01-DRB1*03:01-DQB1*02:01. Also, the haplotype A*02:01-B*35:01-DRB1*08:02-DQB1*04:02 was identified in Tlalpan and Xalapa regions. Haplotype A*01:01-B*08:01-DRB1*03:01-DQB1*02:01 was found only in Tlalpan and Chihuahua. In the Xalapa region, the most frequent haplotype was A*24:02-B*35:01-DRB1*04:07-DQB1*03:02. These alleles and haplotypes have been described in Amerindian populations. Our data are consistent with previous studies and contribute to the analysis of the variability in the Mexican population.


Introduction
The human leukocyte antigen (HLA; known as MHC in other vertebrates) plays a central role in the recognition and presentation of antigens to the immune system and represents the most polymorphic gene cluster in the human genome [1]. This extensive polymorphism of the HLA genes among world populations results from selective pressures, including functional adaptations, particularly of bacteria, viruses, and parasites [2,3], which are particularly important in the understanding of human population variability. The HLA system in different populations is important in disease association, transplantation,

The Sample
The participants were collected as controls for disease association studies. The study protocol was approved (approbation codes: B20-08, B05-10 and B20-15) by the Institutional Committee for Science and Ethics of the Instituto Nacional de Enfermedades Respiratorias Ismael Cosío Villegas (INER). After having been informed of the purpose of the research, all the volunteers signed a letter of consent and were provided an assurance-of-personal-data document.

HLA Typing
Genomic DNA was extracted from peripheral blood using the BDTrack DNA isolation kit (Maxim Biotech, San Francisco, CA, USA). Genotyping for HLA class I (-A and -B) and class II (-DRB1 and -DQB1) was performed using PCR by Sequence-Specific Primers (PCR-SSP, (One Lambda Micro SSP™, Hannover Germany). Nomenclature for HLA genes was according to official WHO Nomenclature [20]. The two-fields resolution was performed by two steps. Firstly, typing was performed using a low-resolution technique (One Lambda Micro SSP™ Generic Trays, Hannover Germany), based on IMGT/HLA 3.23.0, which included HLA-A, -B, -DRB1 and -DQB1 specificities which can be from 8 to 48 independent well reactions, depending on the locus. Then, the allele discrimination of two fields and ambiguities resolution were done employing sets of high-resolution

The Sample
The participants were collected as controls for disease association studies. The study protocol was approved (approbation codes: B20-08, B05-10 and B20-15) by the Institutional Committee for Science and Ethics of the Instituto Nacional de Enfermedades Respiratorias Ismael Cosío Villegas (INER). After having been informed of the purpose of the research, all the volunteers signed a letter of consent and were provided an assurance-of-personal-data document.

HLA Typing
Genomic DNA was extracted from peripheral blood using the BDTrack DNA isolation kit (Maxim Biotech, San Francisco, CA, USA). Genotyping for HLA class I (-A and -B) and class II (-DRB1 and -DQB1) was performed using PCR by Sequence-Specific Primers (PCR-SSP, (One Lambda Micro SSP™, Hannover Germany). Nomenclature for HLA genes was according to official WHO Nomenclature [20]. The two-fields resolution was performed by two steps. Firstly, typing was performed using a low-resolution technique (One Lambda Micro SSP™ Generic Trays, Hannover Germany), based on IMGT/HLA 3.23.0, which included HLA-A, -B, -DRB1 and -DQB1 specificities which can be from 8 to 48 independent well reactions, depending on the locus. Then, the allele discrimination of two fields and ambiguities resolution were done employing sets of high-resolution primers (One Lambda Micro SSP™ High-Resolution Trays, Hannover Germany), based on IMGT/HLA 3.23.0 with independent panels, oscillating from 22 to 48 primer-pairs, depending on the variability of each locus.
PCR-SSP methodology is based on the principle that oligonucleotide primers are used efficiently to amplify a target sequence [21]. The total number of primers used must amplify all known alleles (positive result), the PCR-SSP employed were low and high resolution, in order to solve the ambiguities generated by low resolution genotyping. This technique requires a pair of internal control primers (β-globin gene) for the entire amplification process, which serves to verify PCR reaction integrity. The pairs of primers were designed to have a perfect match with only one allele or group of alleles. In each well of the plate, we added DNA (150 ng/uL) to dried primers. Next, we added recombinant Taq polymerase (Thermo Scientific, Wilmington, DE, USA) and dNTP buffer mixture (Micro SSP D-mix). The amplification was carried out with Verity 96-Well thermal cycler (Applied Biosystems/Thermo Fisher Scientific Inc., Singapore) with a standardized amplification program [22]. After the PCR process, electrophoresis was performed to amplify DNA fragments on a 2.0% agarose gel and visualized by staining with 1.0% Ethidium Bromide (Sigma-Aldrich, St. Louis, MO, USA) with exposure to UV light transilluminator (UVP Inc. Upland, CA, USA) [23]. Finally, the interpretation of the PCR-SSP results was based on the presence or absence of a specific amplified DNA fragment that, using HLA Fusion™ 3.0 Software ((One Lambda, Inc. Canoga Park, CA, USA), identifies the alleles.

Statical Analysis and Data Visualization
Allele and haplotype frequencies were determined by Maximum-likelihood estimation (MLE) [24] using the software Arlequin v. 3.1 ( L. Excoffier, CMPG University of Berne, Berne, Switzerland) [25] and the Expectation-Maximization (EM) algorithm function in the total population and in the three regions. The observed versus expected heterozygosity (for each locus) was analyzed to determine the Hardy-Weinberg equilibrium (HWE). The linkage disequilibrium (LD) coefficient standardized D (∆ ) was calculated according to Lewontin (1964) [26]. Absolute D values of 1 indicate complete LD; 0 corresponds to no LD [27]. Frequencies were compared using χ2 analysis in 2 × 2 contingency tables, as well as with Fisher's exact test when appropriate; we consider p values that were ≤0.05 as statistically significant. The analysis was performed through Epi-Info v.7.2.2.6. R Studio v. 3.6.1 (R Core Team, Vienna, Austria) was used to create the Venn diagram and geographical map.

Results
In the whole-population analysis (Mexican mestizo, MM, n = 502), we identified 26 alleles for HLA-A, 54 alleles in HLA-B, 46 alleles in HLA-DRB1, and 16 in HLA-DQB1. We observed the highest diversity of alleles in the -B and -DRB1 loci in our study population.
The number of alleles in each locus in the three populations, and their distribution in the studied population, are shown in Figure   In the analysis of independent populations, the HLA-A and -B loci for the region Tlalpan differ in HWE (p < 0.05); while, for the regions Chihuahua City and Xalapa, the four loci did not differ significantly in HWE; the results from each region are shown in Table 1.

HLA-B
For the HLA-B locus in the Tlalpan mayoralty, we reported 47 alleles, while for Veracruz-Xalapa we reported 31 alleles, and 33 alleles for the Chihuahua region. We found that HLA-B*35:01, B*39:01, and B*40:02 were present in Tlalpan at~36%; in the Chihuahua region B*51:01, B*39:01, and B*35:01 alleles were found in a great part of the population, with a total frequency of 34%. In the Xalapa municipality, the alleles with higher frequency were similar to those described previously in Tlalpan, but the top-three almost reach 45%. According to the frequencies shown in Table 3

Haplotype Frequency
In the whole population, 761 haplotypes were detected, shared by the four loci. The Tlalpan population was where most haplotypes were identified (514), followed by Chihuahua (166) Table 6 shows the haplotype frequencies for those that have a haplotype frequency (HF) >1.0%. Table 6. HLA-A-B-DRB1-DQB1 haplotypes from MM.

Discussion
Mexico's population is mostly composed of Mestizos, as with other Latin American populations, which are a recently admixed population composed of Amerindian, European and, to a lesser extent, African and Asian ancestries. In this matter, an important role of ethnicity in the susceptibility to different inflammatory/autoimmune and infectious diseases has been attributable to the inclusion of HLA alleles by miscegenation with Caucasian, Asian, and African populations. Nevertheless, studies of the genetics of diseases are difficult to replicate due to the complex nature of the environmental factors and the degree of genetic variability among human populations.
Comparative analyses between Mexicans and other neighboring populations reveal significant differences in genetic diversity [28]. The HLA allele-distribution varies between distinct populations; in our study the alleles HLA-A*02:01, B*35:01, DRB1*04:07, and DQB1*03:02 were found at the highest frequencies in the three studied regions. These alleles have been described in various Amerindian (Native Americans) groups, such as the Nahuas [29]. In the Tarahumara indigenous population, HLA-A*24:02:01, B*40:02, DRB1*08:02:01, DQB*04:02 were described with a frequency greater than 10% [30], and these alleles have also been found in our study at a frequency > 5% in each region; interestingly, the HLA-B*40 has been reported in other Amerindian populations [31]. For HLA class II, DRB1*04:07 and DQB1*03:02 alleles have been reported mainly in the Amerindian population; in Mexico, it has been reported more frequently in the Mayos population to the northeast of the country [32], and these alleles have been found to be the most frequent for the three regions analyzed in our study.
For the Xalapa population, the haplotype with the highest frequency was A*24:02-B*35:01-DRB1*04:07-DQB1*03:02, followed by A*02:01-B*35:01-DRB1*04:07-DQB1*03:02 and A*02:01-B*35:01-DRB1*08:02-DQB1*04:02. Other studies have described these haplotypes as frequent in native Amerindians [33,36]. These results showed that most of the haplotypes found in the Tlalpan and Xalapa population were of Amerindian origin, while one haplotype of Caucasian origin was found at a higher frequency in Chihuahua compared with the other two populations, probably as a product of contact with the Spaniards conquers, who were attracted by the discovery of large deposits of mineral resources [8].
In some populations, genetic associations have been described with the presence of HLA alleles with different clinical phenotypes/diseases. The HLA-B*35 has been associated with increased risk of developing pulmonary arterial hypertension in systemic sclerosis patients [37], and we found a high frequency of this allele in our Mexican mestizo population; remarkably, it has been reported at a greater proportion in the Xalapa region compared to Tlalpan and Chihuahua. In our study population, we found the HLA class II alleles, DRB1*04:04 and DRB1*04:05, which have been described as associated with susceptibility to rheumatoid arthritis [38][39][40], while the DRB1*03:01 allele with Systemic Lupus Erythematosus [41,42]. Another allele associated with autoimmune diseases is HLA-A*01:01, which has been associated with the development of psoriatic arthritis in the Chinese population [43]. In a pilot study in the Mexican population, HLA-DQB1*05 was associated with susceptibility of reinfection with human papillomavirus [44]. Two alleles (HLA-DRB1*03:01 and DRB1*04:04) that we found in greater proportions in our study have been conferred with an increased risk of Addison's disease [45]. Diabetes mellitus type 1 is one of the most complex diseases with the highest incidence in Mexico; the DQB1*05:01 and DQB1*02:01 alleles that occur in linkage disequilibrium with the DRB1*03:01 [46] allele, are presented with a frequency >1%. Interestingly, the DRB1*03: 01-DQB1* 02:01 haplotype was found as the most frequent haplotype in the Chihuahua region.
It is important to know the relationship between HLA alleles and the development of diseases, as some alleles are distributed in higher proportions in different regions of Mexico, which may act as a form of epidemiological monitoring. Our results are important for future comparative genetic studies in different Latin American ethnic groups, particularly Mexican Mestizos and Amerindians.

Conclusions
The alleles HLA-A*02:01, -B*35:01, -DRB1*04:07, and -DQB1*03:02, as well as the A*02:01-B*35:01-DRB1*08:02-DQB1*04:02 and A*68:01-B*39:01-DRB1*04:07-DQB1*03:02 haplotypes were found with higher frequencies in the whole Mexican mestizo population studied. Our results show the existence of alleles and haplotypes that have been reported with an increased frequency in Amerindian populations as well as others from Caucasian populations. The current data contribute to the understanding of genetic diversity in Mexico and serve to extend our knowledge of genetic variants of critical relevance for the development of genomic medicine in Mexico.