GSTT1 and GSTM1 null variants in Mestizo and Amerindian populations from northwestern Mexico and a literature review

Abstract The GSTT1 and GSTM1 genes are key molecules in cellular detoxification. Null variants in these genes are associated with increase susceptibility to developing different types of cancers. The aim of this study was to determine the prevalence of GSTT1 and GSTM1 null genotypes in Mestizo and Amerindian individuals from the Northwestern region of Mexico, and to compare them with those reported worldwide. GSTT1 and GSTM1 null variants were genotyped by multiplex PCR in 211 Mestizos and 211 Amerindian individuals. Studies reporting on frequency of GSTT1 and GSTM1 null variants worldwide were identified by a PubMed search and their geographic distribution were analyzed. We found no significant differences in the frequency of the null genotype for GSTT1 and GSM1 genes between Mestizo and Amerindian individuals. Worldwide frequencies of the GSTT1 and GSTM1 null genotypes ranges from 0.10 to 0.51, and from 0.11 to 0.67, respectively. Interestingly, in most countries the frequency of the GSTT1 null genotype is common or frequent (76%), whereas the frequency of the GSMT1 null genotype is very frequent or extremely frequent (86%). Thus, ethnic-dependent differences in the prevalence of GSTT1 and GSTM1 null variants may influence the effect of environmental carcinogens in cancer risk.


Introduction
The family of the glutathione S-transferases (GSTs) is composed of enzymes that play an essential role in the cellular protection against a wide range of hazardous molecules, such as reactive oxygen species (ROS), xenobiotics and electrophilic compounds. The mammalian GSTs can be classified into three groups: cytosolic, mitochondrial and membrane-associated proteins in eicosanoid and glutathione metabolism (MAPEG). Cytoplasmic enzymes are further subdivided into seven groups: Alpha (GSTA), Mu (GSTM), Omega (GSTO), Pi (GSTP), Sigma (GSTS), Theta (GSTT), and Zeta (GSTZ) (Tew and Townsend, 2012). Since the individual GSTs proteins can share ligands, functional redundancy is a common event in the GST-mediated biotransformation of toxic compounds (Luo et al., 2011).
GSTs catalyze the conjugation of reduced glutathione (GSH), the major antioxidant molecule in the cell, to a myriad of hazardous molecules, including carcinogens, drugs and xenobiotics. GSH-conjugated substrates are then transported out of the cell mainly via the ABC (ATP-binding cassette) efflux pumps. Additionally, GSTs are able to detoxify noxious products of the cellular metabolism, such as reactive oxygen and nitrogen species through their glutathione peroxidase activity (Board and Menon, 2013;Galal et al., 2015). These enzymes are involved in cellular processes others than detoxification, including chaperone activities, regulation of kinase-mediated signal transduction and S-glutathionylation cycle (Pajaud et al., 2012;Klaus et al., 2013;Zhang et al., 2014a).
Early studies highlight the presence of deletion variants (null variants) in the GSTM1 and GSTT1 genes, which are located at chromosomal positions 1p13.3 and 22q11.23, respectively. Individuals with the homozygous genotype for the deletion variants (null/null) in GSTM1 or GSTT1 genes showed the total loss of enzymatic activity of the respective protein (Pemble et al., 1994;Xu et al., 1998). In accordance with their detoxification properties, the deficiency of GSTM1 and GSTT1, either individually or in combination, greatly increases the susceptibility of developing cancer in different organs, including liver, lung and colon (Csejtei et al., 2008;Sui et al., 2014;Zhang et al., 2014b).
The prevalence of GSTM1 and GSTT1 null alleles shows strong variation among different ethnic groups. For instance, the frequency of the GSTM1 null allele was as low as 0.23 in South Africa, but up to 0.42 in Spain and 0.67 in Singapore (Masimirembwa et al., 1998;Chan et al., 2011;Ruano-Ravina et al., 2014). With regard to GSTT1, the frequency of the null genotype among Greek individuals was 0.10, whereas in England and Japan the frequency was 0.21 and 0.50, respectively (Garte et al., 2001;Dialyna et al., 2003;Hishida et al., 2005). These differences could modulate the risk to different types of tumors in populations of different ethnic ancestry. For instance, Japan, one of the countries with the highest frequency of the null genotype for both GSTM1 and GSTT1 genes, has a high incidence of colorectal, stomach, esophagus and prostate cancer (WHO, 2012). Although studies about the distribution of GSTM1 and GSTT1 null genotypes in a Mexican-Mestizo population have been performed previously (Pérez-Morales et al., 2008;Pérez-Morales et al., 2011;Sánchez-Guerra et al., 2012;Gutiérrez-Amavizca et al., 2013;Sandoval-Carrillo et al., 2014;García-González et al., 2015;Jaramillo-Rangel et al., 2015) no reports of the prevalence of these variants in Mexican Amerindian individuals are available. Thus, the aim of this study was to determine and compare the frequencies of GSTM1 and GSTT1 null genotypes in Mexican-mestizo and Amerindian individuals (Tarahumara) from the Northwestern part of the country (State of Chihuahua) with those previously found in other regions of Mexico and around the world.

Study population
The sample population was composed of 422 unrelated individuals from the State of Chihuahua, in the Northwest of Mexico: 211 subjects from the Amerindian ethnic group (Tarahumara) and 211 Mexican-mestizo persons. The Tarahumara sample consisted of 138 females and 73 males with ages ranging from seven to 18 years, whereas the Mexican-mestizo group was composed of 88 females and 123 males, with ages ranging from 16 to 30 years. Samples were collect from July 2009 to March 2014. The Tarahumara group consisted of individuals self-recognized as Amerindians, whose two parents and four grandparents were all born in the locality and speak the Tarahumara language. All participants signed a written informed consent, and in the case of underage individuals, the parents signed their informed consent. Local committees of research ethics approved the study following the Declaration of Helsinki.

GSTM1 and GSTT1 genotyping
Genomic DNA was isolated from 300 mL of whole blood samples using the MasterPure DNA Purification kit (Epicentre Biotechnologies, Madison, WI, USA), according to the manufacturer's protocol. DNA integrity was verified by electrophoresis on a 1.2% agarose gel and DNA concentration was evaluated in a Nanodrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA).
Genotyping of null variants in the GSTM1 and GSTT1 genes (GenBank accession number: AP000351 and X68676, respectively) was performed by multiplex PCR, as previously described (Arand et al., 1996). Briefly, we used primers to amplify a fragment of the genes GSTM1 (215 bp), GSTT1 (480 bp) and the housekeeping GAPDH (315 bp), as an internal amplification control, for each sample using a conventional PCR protocol. Also, we used DNA samples with known genotype for GSTM1 and GSTT1 null alleles (GST-T1/M1: wt/wt, wt/null, null/wt and null/null) as positive controls.
The primers used for PCR amplification were: GSTT1 Forward: 5-TTC CTT ACT GGT CCT CAC ATC TC-3 Reverse: 5-TCA CCG GAT CAT G GC CAG CA-3 GSTM1 Forward: 5-GAA CTC CCT GAA AAG CTA AAG C-3 Reverse: 5-GTT GGG CTC AAA TAT ACG GTG G-3 GAPDH Forward: 5-GGA TGA CCT TGC CCA CAG CCT-3 Reverse: 5'-CAT CTC TGC CCC CTC TGC TGA-3' DNA amplification was carried out with an initial denaturing step at 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s. The PCR reactions were performed in a Veriti 96-well thermal cycler (Applied Biosystems). The PCR products were separated by electrophoresis in 2.5% agarose gels stained with ethidium bromide and visualized by ultraviolet light. In addition, 10% of the samples were genotyped twice from the original DNA sample with a 100% concordance.

Literature search for genotype data
To identify studies reporting on frequencies of GSTM1 and GSTT1 null variants worldwide, a PubMed search was conducted. After excluding meta-analyses and review articles, we considered in our study a total of 57 reports. In order to avoid a bias imposed by the frequency of a gene variant in association with a disease, the frequencies of the null and wild-type genotypes of GSTM1 and GSTT1 genes were extracted only from the healthy population reported in each manuscript, but the respective frequencies in the disease-affected population was not considered.

Statistical analysis
GSTM1 and GSTT1 null and wild-type genotypes in Mestizo and Tarahumara populations were reported as frequency. Our findings were compared with those found in other ethnic groups worldwide. The frequencies of the GSTM1 and GSTT1 null genotypes were used to generate maps with their geographic distribution using the QGIS 2.4.0-Chugiak shape file (www.naturalearthdata.com). Statistical analysis was performed using the Fisher's exact test, with p < 0.05 considered statistically significant.

Results
After genotyping the GSTM1 and GSTT1 null polymorphisms, we observed that the GSTM1 null genotype showed a significantly higher frequency than the GSTT1 null genotype in both the Mestizo (0.44 vs. 0.11) and Tarahumara groups (0.47 vs. 0.11). The most common compound genotype in both groups was GST-T1/M1 wt/wt (Mestizo=0.50; Tarahumara=0.47), followed by the GST-T1/M1 wt/null genotype (Mestizo=0.38; Tarahu-mara=0.42). The compound genotypes with lower frequency in both groups were GST-T1/M1 null/wt (Mestizo=0.05; Tarahumara=0.06) and GST-T1/M1 null/null (Mestizo=0.06; Tarahumara=0.05) (Figure 1). We found no significant difference in the frequencies of the wild type or of null genotype for GSTT1 and GSTM1 between Mestizo and Tarahumara individuals. Likewise, the frequency distribution of the compound genotypes showed no significant difference between Mestizo and Tarahumara individuals.
The frequency of the GSTT1 null genotype observed in the Mestizo individuals included in our study was similar to those previously reported in Mexican-Mestizos from the northeastern and central regions of the country (0.11 vs. 0.10-0.13 and 0.12-0.15, respectively), as well as in one population from the Southeast (0.11 vs. 0.09). However, it was significantly higher in comparison with those reported in the western region (0.11 vs. 0.03) ( Table 1). In the case of GSTM1, the frequencies of the null genotype found previ-ously in Mexican-Mestizos from the northeastern and western regions were similar to those of our study (0.44 vs. 0.44-0.48, and 0.43, respectively), but populations from the central and southeastern regions showed a significantly lower frequency (0.44 vs. 0.33-0.37, and 0.31, respectively) ( Table 1). It is worth mentioning that a Mexican-Mestizo population located in the coastal zone of the southeastern region showed the highest frequency of the null genotype for GSTT1 and the lowest for GSTM1 in our country (0.17 and 0.22, respectively). These data show a clear reduction in the frequency of the GSTM1 null genotype from North to South, whereas in the case of the GSTT1 null genotype no apparent tendency was observed.
We also collected from the literature the frequencies of the GSTT1 and GSTM1 null genotypes found in 57 countries around the world. The worldwide frequency of the GSTT1 null genotype ranges from 0.10 to 0.51, whereas that of the GSTM1 null genotypes ranges from 0.11 to 0.67 (Table 2). To further compare the prevalence of GSTT1 and GSTM1 null genotypes worldwide, we classified these frequencies in four groups: common (0.10-0.22), frequent (0.23-0.35), very frequent (0.36-0.48), and extremely frequent (more than 0.48). We observed that the reported frequencies for the GSTT1 null genotype were common in 31 countries (55%), frequent in 12 (21%), very frequent in 11 (19%) and extremely frequent only in three (5%) (Figure 2, GSTT1 and GSTM1 variants 729   upper panel). In sharp contrast, the reported frequencies of the GSTM1 null genotype were common in only two countries (3%), frequent in six (10%), very frequent in 17 (30%) and extremely frequent in 32 (56%) (Figure 2, lower panel). Because of the low number of studies reporting frequencies for the compound genotypes, it was not possible make comparisons.
Regarding the geographical distribution of these null variants, the countries where GSTT1 null genotype frequencies were common or frequent were distributed over the five continents, whereas those where the GSTT1 null genotype was very frequent were concentrated mainly in Africa (Namibia, Gambia, Ethiopia, Tunisia, Somalia and Cameroon) and Asia (China, Taiwan and Thailand). It is worth mentioning that the only three countries with extremely frequent presence of this variant were in East Asia (Singapore, Japan and Korea) ( Figure 3A). In the case of the GSTM1 null genotype, this variant was common in Namibia and Gambia (Africa), and frequent in Syria and India (Asia), and in South Africa, Zimbabwe, Cameroon and Tanzania (Africa), whereas countries with very frequent and extremely frequent frequency were distributed all over the world ( Figure 3B).

Discussion
GST proteins are essential molecules in cellular protection against a myriad of environmental and intracellular compounds. Null variants occurring in the GSTT1 and GSTM1 genes are the most common polymorphisms in GST proteins, and their association with various chronicdegenerative diseases such as hypertension, diabetes, asthma, and different types of cancer including prostate, neck, colorectal, liver and leukemia has been thoroughly studied in different populations (Song et al., 2012;Zhang et al., 2012;Liang et al., 2013;Liu et al., 2013;Eslami and Sahebkar, 2014;He et al., 2014;Rao et al., 2014;Masood et al., 2015). Both the prevalence of the GSTT1 and GSTM1 null genotypes as well as their association with disease phenotypes are highly dependent on ethnic background.
The Mexican-Mestizo population is a complex genetic admixture consisting of Amerindian (56%), Caucasian (41%) and African alleles (3%), with a decreasing Caucasian and an increasing Amerindian ancestry from North to South (Lisker et al., 1986).
In our study, we found no significant difference in the frequencies of the GSTT1 and GSTM1 null genotypes among Mexican-Mestizo and Tarahumara individuals from the northwestern region of the country. In addition, we observed a high variability in the frequency of the null genotypes for GSTT1 and GSTM1 among the different geographic regions of the country, ranging from 0.03 to 0.17 for GSTT1 and from 0.22 to 0.48 for GSMT1 (Pérez- Morales et al., 2008Morales et al., , 2011Sánchez-Guerra et al., 2012;Gutiérrez-Amavizca et al., 2013;Sandoval-Carrillo et al., 2014;García-González et al., 2015;Jaramillo-Rangel et al., 2015). In the case of GSTM1, the frequencies of the null genotypes showed a clear reduction from North to South, whereas the frequency of the GSTT1 null genotypes showed no apparent tendency. The genetic structure of the Mexican population is very complex and is strongly affected by geographical location. For example, populations located in the northern region near to the US border are characterized by an intense admixture with Europeanderived populations. In contrast, more than 90% of the Amerindian populations in Mexico (68 ethnic groups) are located in the southern region of the country. As the frequency of the GSTM1 null genotype is higher in American populations with European ancestry (e.g., USA and Canada) than in Latino American populations (e.g,. Mexico, Chile and Paraguay), it may be possible that the decreasing frequency from North to South of this variant could be caused by the admixture occurring with Caucasian populations.
Regarding the prevalence of the GSTT1 and GSTM1 null genotypes worldwide, the GSTM1 null genotype was very frequent or extremely frequent (0.36 and above) in the majority of the analyzed countries (86%), whereas the GSTT1 null genotype was common or frequent (from 0.10 to 0.35) in most of the countries (76%). Since the GSTM1 null genotype is more frequent than GSTT1 in every country, this indicates that the loss of function of GSTT1 has a more deleterious effect than GSTM1. However, we cannot GSTT1 and GSTM1 variants 731 discard that other GST proteins could replace GSTM1 but not GSTT1 function. The lowest frequencies of the GSTT1 null genotype were found in America (0.11-0.20), with exception of Greenland (0.46), followed by Europe (0.10-0.26) and Africa (0.20-0.47); the highest frequencies were in Asia (0.16-0.51). For GSTM1, the lowest frequencies were observed in Africa (0.11-0.55), followed by Asia (0.23-0.67) and America (0.36-0.52); Europe had the highest frequencies (0.42-0.58). Middle East countries showed lower frequencies for both GSTT1 and GSTM1 null genotypes than those from far East Asia (GSTT1: 0.16-0.23 vs. 0.25-0.51; GSTM1: 0.23-0.49 vs. 0.42-0.69). Moreover, countries such as Japan, Korea, Singapore and Thailand showed extremely high frequencies for both the GSTT1 and GSTM1 null genotypes (Table 2). It is worthy of note that the frequency of the GSTM1 null genotype found in a Mexican-Mestizo population from the southeastern region, which has a high African ancestry, was very similar to the frequency found in several African populations, including Cameroon, Gambia, and Zimbabwe (0.22 vs. 0.28, 0.20 and 0.24, respectively) (Wild et al., 2000;Dandara et al., 2002;Piacentini et al., 2011;Sánchez-Guerra et al., 2012).
In summary, the prevalence of the GSTT1 and GSTM1 null genotypes showed a very high diversity, dependent on ethnic background. 732 Palma-Cano et al.