The importance of understanding the distribution of GSTM1 and GSTT1 genotypes and haplotypes in a region with intense agriculture activity

All 602 individuals are born and raised in Goiás. 68.6% are males, and 31.4% are females. The sample group presented a mean age of 38 years ± 13.5 for males and 36 years ± 12.3 for females. We also verified that 51% of males consume alcohol; in contrast, 66.1% of females do not drink alcohol. Finally, 13.6% and 10.6, men and women, respectively, were smokers ( Table 1 ). Table 1 Social habits of the sample group distributed by sex. Table 1 Men (413) Women (189) Total (602) Age (Mean ± SD ∗ ) 38 ± 13.5 36 ± 12.3 37 ± 13.2 Alcohol intake Alcoholic drinker 51% (211) 33.9% (64) 45.6% (275) Non alcoholic drinker 49% (202) 66.1% (125) 54.3% (327) Smoke habit Non smoker 86.4% (357) 89.4% (169) 87.4% (526) Smoker 13.6% (56) 10.6% (20) 12.6% (76) ∗ SD = Standard deviation. Social habits of the sample group distributed by sex. SD = Standard deviation. The frequency of the GSTT1 null genotype was 38.2%, and for the municipalities with more than 20 individuals sampled ( Table 2 ), it can be seen in Fig. 1 A. Therefore, the frequency of the GSTM 1 null genotype for Goias was 50.3%, and for the municipalities with more than 20 individuals sampled are in Fig. 1 B. Finally, the frequencies of GSTT1 and GSTM1 null genotypes (-/-) was 27.24%, −/+ 11%, +/- 23% and +/+ 39% for Goias ( Fig. 1 C). Fig. 2 illustrated the haplotypes frequencies in municipalities with more than 20 individuals sampled. Table 2 Absolute number and haplotype composition by Goiás state municipalities. Table 2 Municipality GST T1/M1 +/+ GST T1/M1 −/+ GST T1/M1 −/− GST T1/M1 +/- Total number Abadiânia 2 1 0 0 3 Anápolis 1 1 1 0 3 Aparecida de Goiânia 79 32 23 31 165 Aragoiânia 0 0 1 0 1 Bela Vista 15 5 1 14 35 Bomfinópolis 1 0 0 0 1 Brasabrantes 1 1 1 0 3 Caldazinha 0 0 2 0 2 Goianira 0 0 0 1 1 Goiânia 28 6 112 19 165 Itapuranga 14 3 7 12 36 Itumbiara 0 0 1 0 1 Leopoldo de Bulhões 4 0 0 11 15 Montividiu 29 9 3 19 60 Nerópolis 4 1 1 3 9 Ouro Verde 1 0 0 2 3 Piracanjuba 7 2 0 1 10 Rio Verde 1 0 0 3 4 Rozalândia 5 0 0 0 5 Santo António de Goiás 1 0 2 0 3 Santa Terezinha 2 0 0 0 2 Senador Canedo 0 0 2 0 2 Silvânia 26 5 3 14 48 Trindade 0 0 1 1 2 Turvânia 12 0 3 8 23 Goiás state municipalities (Total) 233 66 164 139 602 +/+: Genotype GSTT1 wild type and genotype GSTM1 wild type. −/+: Genotype GSTT1 null and genotype GSTM1 wild type. +/-: Genotype GSTT1 wild type and genotype GSTM1 null. −/−: Genotype GSTT1 null and genotype GSTM1 null. In bold = municipalities with more than 20 individuals. Fig. 1 Distribution of GSTT1 and GSTM1 null genotypes among Goias state. A. Distribution of GSTT1 null genotype in municipalites with more than 20 individuals sampled. B. Distribution of GSTM1 null genotype for Goias state in municipalities with more than 20 individuals sampled. C. GSTT1 and GSTM1 null genotypes (-/-) for the Goias state. Fig. 1 Fig. 2 Map illustrating the distribution of GSTT1 and GSTM1 haplotype frequencies in Goias state municipalities with more than 20 individuals sampled. Fig. 2 Absolute number and haplotype composition by Goiás state municipalities. +/+: Genotype GSTT1 wild type and genotype GSTM1 wild type. −/+: Genotype GSTT1 null and genotype GSTM1 wild type. +/-: Genotype GSTT1 wild type and genotype GSTM1 null. −/−: Genotype GSTT1 null and genotype GSTM1 null. In bold = municipalities with more than 20 individuals. Distribution of GSTT1 and GSTM1 null genotypes among Goias state. A. Distribution of GSTT1 null genotype in municipalites with more than 20 individuals sampled. B. Distribution of GSTM1 null genotype for Goias state in municipalities with more than 20 individuals sampled. C. GSTT1 and GSTM1 null genotypes (-/-) for the Goias state. Map illustrating the distribution of GSTT1 and GSTM1 haplotype frequencies in Goias state municipalities with more than 20 individuals sampled. The frequencies for the null GSTT1 were 43.2% (207) in the central region and 18.7% (23) in the south region, and for the null GSTM1 we found frequencies of 54% (259) in the central region and 35.8% (44) in the south region. The distribution of GSTT1 and GSTM 1 haplotypes by region was as follows: central region +/- 35.5% (170), -/+ 10.4% (50), -/- 32.8% (157), +/- 21.3 % (102) and in the south +/+ 51.2% (63), -/+ 13% (16), -/- 5.7% (7), +/- 30% (37) ( Table 3 ). Table 3 Frequency of genotypes GST T1 and GST M1 in the south and central regions of Goias state. Table 3 Region Genotype GSTT1 - GSTT1 + GSTM1 - GSTM1 + Central 43.2% (207) 56.8% (272) 54.% (259) 46% (220) South 18.7% (23) 81.3% (100) 35.8% (44) 64.3% (79) +: Genotype GSTT1 wild type and genotype GSTM1 wild type. -: Genotype GSTT1 null and genotype GSTM1 null. Frequency of genotypes GST T1 and GST M1 in the south and central regions of Goias state. +: Genotype GSTT1 wild type and genotype GSTM1 wild type. -: Genotype GSTT1 null and genotype GSTM1 null.

This research was approved by the Research Ethics Committee of the Pontifical Catholic University of Goiás, under the protocol number 1978/2011. To determine the power of our sample group, we used the following formula: z 2 x p ( 1 − p ) e 2 1 + ( z 2 x p ( 1 − p ) e 2 N ) where N = population size; e = p-value ; z = score z (1.96, considering a confidence interval of 95%). The population size of Goias state is 6,000000 individuals, according to the Brazilian Institute of Geography and Statistics ( IBGE, 2017 ). In this context, the minimal of individuals to be sample should be 384. To this study, we analyzed 602 unrelated individuals, born in Goias, distributed in 25 municipalities of two regions of the state, central (N = 479) and south (N = 123). The central municipalities of Goias are Abadiania, Anapolis, Aparecida de Goiania, Aragoiania, Bela Vista, Bonfinopolis, Brazabrantes, Caldazinha, Goiania, Goianira, Itapuranga, Leopoldo de Bulhões, Neropolis, Ouro Verde, Roselandia, Santa Terezinha, Santo Antônio de Goiás, Senador Canedo, Trindade, and Turvania. The south municipalities are Itumbiara, Montividiu, Piracanjuba, Rio Verde, Silvania. The criteria for inclusion in the survey were only those individuals born and live in the State of Goias. All the municipalities presented tomato, corn, sorghum, and soybean crops (the central plants of the state of Goias). We obtained blood samples voluntarily according to the Informed Consent Form (TCLE) Data such as age, sex, and social habits were obtained by a lifestyle questionnaire. We sampled 10 ml of whole blood, stored on ice, and immediately sent it to the Mutagenesis Laboratory at the Federal University of Goias. Genomic DNA samples were obtained from peripheral blood lymphocytes using the Illustra Genomic Blood MS® (GE, USA) extraction kit, according to the manufacturer's protocol. DNA samples were then quantified using the NanoVue Plus TM (GE, USA) equipment, following the manufacturer's recommendations. After quantification, DNA samples were diluted to a final concentration of 10 ng/μL. The samples were then stored in a freezer at -20 °C, until use. We performed the analyses of GSTM1 and GSTT1 genes deletion using the real-time polymerase chain reaction (PCR), with SYBR® Green PCR Master Mix (Applied Biosystems®, USA), and we also evaluated the co-amplification of the gene RH92600 as an internal control. Marin et al. (2010) previously suggested the primers and the PCR cycling conditions. For the PCR reaction, the final volume of 25 μL containing 12.5 μL of master mix, 0.5 μL of magnesium chloride (MgCl2), 2.4 μL of forward and reverse primers of the GSTT 1 gene, 3.2 μL of forward and reverse primers of the GSTM1 gene, 4μL of forward and reverse primers of the RH92600 gene, and 1μL of DNA at 10 ng/μL. We used the Step One Plus thermal cycler Real-Time PCR Systems (Applied Biosystems®, USA). The sociodemographic characterization data of the sample group was stored in an Excel 2013 worksheet (Microsoft®), along with the data of the polymorphisms, for further evaluation of Statistica 7 software (StartSoft Inc). We estimated frequencies, genotypes, and haplotypes by direct counting. Differences in allele frequencies of GSTT 1 and GSTM 1 polymorphisms were calculated using the X 2 test, with a 95% confidence interval (p < 0.05).

In all municipalities of Goias, the frequency of the GSTT1 null genotype was 38.2%, and of the GSTM1 null genotype 50.3%. In the Northeast of India, some authors ( Thoudam et al., 2010 ) demonstrated a frequency of 32.7% for null GSTT1 and 41.9% for the null GSTM1 genotype. Klautau-Guimarães et al. (2005) evaluated 120 individuals, distributed in two Amerindian tribes of Brazil, the Munduruku tribe of the village of Missão Cururu, and the tribe of Kayabi. The frequency was 27% for null GSTT1 and 0% for null GSTM1 in Munduruku samples. In the Kayabi tribe, those authors detected frequencies of 29% for GSTT1 and 27% for GSTM1 null genotypes. Hiragi et al. (2007) analyzed 91 samples from the Federal District, Brazil, and 273 samples of Brazilian Quilombolas (African offspring) from the states of Bahia, Sergipe, and Goias. Those authors observed frequencies for GSTT1 null genotype in the Federal District of 22% for GSTM1 null genotype of 34% and GSTT1/M1 null genotypes of 11%. Chiurillo et al. (2013) analyzed 120 urban samples from a Venezuelan population and 188 Amerindian samples, showing a null GSTT1 genotype frequency of 11% and a null GSTM1 genotype of 51% in the urban population. The Amerindian samples were divided into five populations. They presented the following frequencies for the null GSTT1 and GSTM1 null genotypes, respectively: Bari with 11.4% and 54.3%, Panare with 6.5% and 15.2%, Pemon with 0% and 40%, Warao with 0% and 51.7% and Wayuu with 7.9% and 44.7%. Magno et al. (2009) found lower frequencies for null GSTM1 genotypes (33.2%) and the null GSTT1 genotype (30.2%) in 203 samples from the city of Ilheus, Bahia state, Brazil. Berthou et al. (2001) analyzed 174 samples from Santiago, Chile, and found frequencies for the GSTM1 null genotype of 24%. Rossini et al. (2002) observed frequencies of 42.1% of the GSTM1 null genotype and frequencies of 25.4% of the GSTT1 genotype in 591 individuals from the state of Rio de Janeiro. However, no significant differences were observed between the genotypes GSTT1 and GSTM1 concerning the population of Goias and the populations of Kalunga (Goias, Brazil), suggesting a genetic similarity between the communities ( Table 4 ). Table 4 GSTT1 and GSTM1 genotypes distribution in worldwide populations compared to our population. Table 4 GSTT1 GSTM1 Country References p-value * p-value ** Mexican (Mejia-Sanchez et al., 2017) < 0.0001 0.5838 North of Indian ( Thoudam. et al., 2010 ) 0.08172 0.009758 Riacho de Sacutiaba (Bahia, Brazil) ( Hiragi et al., 2007 ) 0.2333 < 0.0001 Rio das Rãs (Bahia, Brazil) ( Hiragi et al., 2007 ) 0.002535 < 0.0001 Mocambo (Sergipe, Brazil) ( Hiragi et al., 2007 ) 0.3621 0.08033 Kalunga (Goiás, Brazil) ( Hiragi et al., 2007 ) 0.1281 0.07782 Distrito Federal (Brazil) ( Hiragi et al., 2007 ) 0.01328 0.02819 Kayabu (Mato Grosso, Brazil) ( Klautau-Guimarães et al., 2005 ) < 0.0001 0.07798 Munduruku (Pará, Brazil) ( Klautau-Guimarães et al., 2005 ) 0.5679 < 0.0001 Venezuela ( Chiurillo et al., 2013 ) < 0.0001 0.9998 Bari (Venezuela) ( Chiurillo et al., 2013 ) 0.002576 0.7788 Panare (Venezuela) ( Chiurillo et al., 2013 ) < 0.0001 < 0.0001 Pernon (Venezuela) ( Chiurillo et al., 2013 ) < 0.0001 0.2703 Warao (Venezuela) ( Chiurillo et al., 2013 ) < 0.0001 1 Wayuu (Venezuela) ( Chiurillo et al., 2013 ) 0.000328 0.6158 Ilhéus (Bahia, Brazil) ( Magno et al., 2009 ) 0.04471 < 0.0001 Rio de Janeiro (Brazil) ( Rossini et al., 2002 ) < 0.0001 0.005401 Bold values are significant values demonstrating differences in the populations compared to our sample group. ∗ p-values associated to the x 2 .test referred to the GSTT1 polymorphism distribution. ∗∗ p-values associated to the x 2 .test referred to the GSTM1 polymorphism distribution. GSTT1 and GSTM1 genotypes distribution in worldwide populations compared to our population. Bold values are significant values demonstrating differences in the populations compared to our sample group. p-values associated to the x 2 .test referred to the GSTT1 polymorphism distribution. p-values associated to the x 2 .test referred to the GSTM1 polymorphism distribution. We observed that the city of Goiania has a different distribution of GSTM 1 and GSTT1 frequencies (p < 0.01) regarding all the municipalities and the other groups and countries ( Table 4 ). Goiania is only like the populations of Riacho de Sucutiaba (Bahia, Brazil) ( Hiragi et al., 2007 ) (p = 0.891) and Kayabu (Mato Grosso, Brazil) ( Klautau-Guimarães et al., 2005 ) (p = 0.103) ( Table 5 ). Considering the distribution of the null GSTM1 genotype, the other municipalities of our study do not present a significant difference from the Kalunga population and the Federal District, except Goiania. Only the cities of Aparecida de Goiania and Goiania showed substantial differences in the distribution of the null GSTM1 genotype when compared to Rio de Janeiro ( Table 6 ). The municipality of Goiania also differed from all other cities of this study (Aparecida de Goiania, Bela Vista de Goiás, Itapuranga, Montividiu, Silvania, Turvania). Table 5 Pairwise comparison of GSTT1 frequencies between our population and other authors worldwide. Table 5 APA BEL GYN ITA MON SIL TUR MEX 0.000 0.000 0.253 0.000 0.000 0.000 0.000 NEI 0.961 0.086 0.000 0.674 0.066 0.035 0.081 RSB 0.100 0.008 0.025 0.096 0.005 0.003 0.010 RRB 0.083 0.608 0.000 0.724 0.783 0.481 0.427 MSE 0.145 0.011 0.001 0.140 0.007 0.004 0.014 KGO 0.043 0.004 0.001 0.066 0.001 0.001 0.006 DIS 0.166 0.541 0.000 0.846 0.688 0.420 0.384 KYM 0.000 0.000 1 0.000 0.000 0.000 0.000 MUP 1 0.104 0.000 0.641 0.099 0.053 0.090 VEN 0.000 0.479 0.000 0.025 0.148 0.439 1 BAV 0.017 0.733 0.000 0.152 0.428 0.723 1 PAV 0.001 0.250 0.000 0.021 0.091 0.227 0.650 PEV 0.000 0.021 0.000 0.001 0.007 0.020 0.084 WAV 0.001 0.056 0.000 0.006 0.024 0.053 0.160 WUV 0.003 0.399 0.000 0.052 0.182 0.376 0.833 ILB 0.574 0.172 0.000 0.939 0.173 0.091 0.141 RIO 0.053 0.372 0.000 0.902 0.446 0.241 0.273 APA 1 0.092 0.000 0.653 0.077 0.041 0.083 BEL 0.092 1 0.000 0.431 0.943 1 0.959 GYN 0.000 0.000 1 0.000 0.000 0.000 0.000 ITA 0.653 0.431 0.000 1 0.531 0.337 0.313 MON 0.077 0.943 0.000 0.531 1 0.846 0.676 SIL 0.041 1 0.000 0.337 0.846 1 0.965 TUR 0.083 0.959 0.000 0.313 0.676 0.965 1 MEX = Mexican; NEI = North of Indian; RSB = Riacho de Sacutiaba (Bahia, Brazil); RRB = Rio das Rãs (Bahia, Brazil); MSE = Mocambo (Sergipe, Brazil); KGO = Kalunga (Goiás, Brazil); DIS = Distrito Federal (Brazil); KYM = Kayabu (Mato Grosso, Brazil); MUP = Munduruku (Pará, Brazil); VEN = Venezuela; BAV = Bari (Venezuela); PAV = Panare (Venezuela); PEV = Pernon (Venezuela); WAV = Warao (Venezuela); WUV = Wayuu (Venezuela); ILB = Ilhéus (Bahia, Brazil); RIO = Rio de Janeiro (Brazil); APA = Aparecida de Goiânia; BEL = Bela Vista de Goiás; GYN = Goiânia; ITA = Itapuranga; MON = Montividiu; SIL = Silvânia; TUR = Turvânia. Table 6 Pairwise comparison of GSTM1 frequencies between our population and other authors worldwide. Table 6 APA BEL GYN ITA MON SIL TUR MEX 0.009 0.755 0.000 0.698 0.198 0.190 1 NEI 0.050 1 0.000 0.278 0.524 0.474 0.734 RSB 0.000 0.000 0.891 0.007 0.000 0.000 0.007 RRB 0.191 0.063 0.000 0.003 0.140 0.233 0.047 MSE 0.000 0.086 0.030 0.416 0.007 0.008 0.297 KGO 0.514 0.809 0.000 0.224 1 0.909 0.573 DIS 0.542 0.717 0.000 0.165 1 0.967 0.498 KYM 0.000 0.075 0.103 0.350 0.007 0.008 0.253 MUP 0.000 0.000 0.000 0.000 0.000 0.000 0.000 VEN 0.003 0.523 0.000 0.988 0.101 0.101 0.971 BAV 0.027 0.473 0.004 1 0.145 0.137 0.831 PAV 0.033 0.012 0.000 0.001 0.025 0.045 0.009 PEV 0.494 0.988 0.000 0.376 0.900 0.825 0.734 WAV 0.078 0.648 0.003 1 0.262 0.243 1 WUV 0.226 1 0.000 0.646 0.560 0.512 1 ILB 1 0.347 0.000 0.037 0.710 0.882 0.236 RIO 0.037 1 0.000 0.280 0.496 0.450 0.744 APA 1 0.342 0.000 0.038 0.694 0.862 0.233 BEL 0.342 1 0.000 0.549 0.705 0.646 0.919 GYN 0.000 0.000 1 0.002 0.000 0.000 0.002 ITA 0.038 0.549 0.002 1 0.183 0.171 0.917 MON 0.694 0.705 0.000 0.183 1 1 0.497 SIL 0.862 0.646 0.000 0.171 1 1 0.458 TUR 0.233 0.919 0.002 0.917 0.497 0.458 1 MEX = Mexican; NEI = North of Indian; RSB = Riacho de Sacutiaba (Bahia, Brazil); RRB = Rio das Rãs (Bahia, Brazil); MSE = Mocambo (Sergipe, Brazil); KGO = Kalunga (Goiás, Brazil); DIS = Distrito Federal (Brazil); KYM = Kayabu (Mato Grosso, Brazil); MUP = Munduruku (Pará, Brazil); VEN = Venezuela; BAV = Bari (Venezuela); PAV = Panare (Venezuela); PEV = Pernon (Venezuela); WAV = Warao (Venezuela); WUV = Wayuu (Venezuela); ILB = Ilhéus (Bahia, Brazil); RIO = Rio de Janeiro (Brazil); APA = Aparecida de Goiânia; BEL = Bela Vista de Goiás; GYN = Goiânia; ITA = Itapuranga; MON = Montividiu; SIL = Silvânia; TUR = Turvânia. Pairwise comparison of GSTT1 frequencies between our population and other authors worldwide. MEX = Mexican; NEI = North of Indian; RSB = Riacho de Sacutiaba (Bahia, Brazil); RRB = Rio das Rãs (Bahia, Brazil); MSE = Mocambo (Sergipe, Brazil); KGO = Kalunga (Goiás, Brazil); DIS = Distrito Federal (Brazil); KYM = Kayabu (Mato Grosso, Brazil); MUP = Munduruku (Pará, Brazil); VEN = Venezuela; BAV = Bari (Venezuela); PAV = Panare (Venezuela); PEV = Pernon (Venezuela); WAV = Warao (Venezuela); WUV = Wayuu (Venezuela); ILB = Ilhéus (Bahia, Brazil); RIO = Rio de Janeiro (Brazil); APA = Aparecida de Goiânia; BEL = Bela Vista de Goiás; GYN = Goiânia; ITA = Itapuranga; MON = Montividiu; SIL = Silvânia; TUR = Turvânia. Pairwise comparison of GSTM1 frequencies between our population and other authors worldwide. MEX = Mexican; NEI = North of Indian; RSB = Riacho de Sacutiaba (Bahia, Brazil); RRB = Rio das Rãs (Bahia, Brazil); MSE = Mocambo (Sergipe, Brazil); KGO = Kalunga (Goiás, Brazil); DIS = Distrito Federal (Brazil); KYM = Kayabu (Mato Grosso, Brazil); MUP = Munduruku (Pará, Brazil); VEN = Venezuela; BAV = Bari (Venezuela); PAV = Panare (Venezuela); PEV = Pernon (Venezuela); WAV = Warao (Venezuela); WUV = Wayuu (Venezuela); ILB = Ilhéus (Bahia, Brazil); RIO = Rio de Janeiro (Brazil); APA = Aparecida de Goiânia; BEL = Bela Vista de Goiás; GYN = Goiânia; ITA = Itapuranga; MON = Montividiu; SIL = Silvânia; TUR = Turvânia. The high prevalence of GSTT 1 and GSTM1 null genotypes in the municipality of Goiania may reflect a founder effect of non-representative colonizing ancestors. However, as other enzymes are part of the xenobiotic detoxification pathway ( Nebert et al., 2004 ; Nebert and Russell, 2002 ; Norppa, 1997 ; Singh et al., 2011 ), the ancestral population adapts to the environment with the frequencies observed in Goiania. However, as in the present day there is excessive use of pesticides, the community becomes susceptible to the harmful effects of exposure to such xenobiotics, probably due to the high frequencies of null genotypes. In this context, GSTT1 null and GSTM1 null genotypes, alone or in combination, should identify individuals with some detoxification capacity deficiency and, therefore, more likely to undergo carcinogen formation, DNA adducts, and mutations. Some authors ( Strange et al., 2001 ) found the null GSTM1 associated with increased chromosomal aberrations in smokers. Wang et al., (2010) in a case-control study in the Han Chinese population, found a contribution of GSTT1 null and GSTM1 null for the development of respiratory bronchial dysplasia. Malik et al. (2017) also associated the null genotypes of GSTT 1 and GSTM1 to the development of lung cancer. Tacca et al., 2019 observed that null GSTT1 and GSTM1 were not associated with the development of cervical cancer in smokers but GSTT1 null genotype was significantly associated with worse prognosis of this disease. In another study conducted in Brazil ( Kubiszeski et al., 2015 ), there was a higher prevalence of GSTT1 null with the onset of endometriosis. Singh et al., (2011) found higher values of DNA damage in workers with GSTT1 null exposed to organophosphates. The central and south regions of Goias showed frequencies of 43.2% and 18.7% for GSTT 1 null and 54% and 35.8% for GSTM1 null ( Fig. 3 A). These regions together are responsible for the production of 18,356,659 tons of grain in the state of Goias, which comprises 80.46% of the state's total production ( Instituto Mauro Borges, 2018 ). As Goias is the fourth-largest producer of grains and is also the fourth-largest user of pesticides ( Instituto Mauro Borges, 2018 ), the null genotypes represent a risk to the population due to high agricultural production and the high handling of pesticides in the state. The mean population age (37 ± 13.2) may represent the basal exposure to pesticides of the people of Goias. However, another important fact is the number of grains produced by the state; 22,814,803 tons ( Instituto Mauro Borges, 2018 ) can be representative of the number of rural workers (95,745 by 2017) ( Instituto Mauro Borges, 2018 ). For those who have direct contact with the pesticide, the null genotypes represent an even higher risk. In this case, the south region is the most representative region for grain production (17,572,323 tons) ( Fig. 3 B). Fig. 3 A. Distribution of GSTT1 and GSTM1 in central and south regions of Goias state. B. Amount of grain production in the five regions (central, east, northwest, north and south) of Goias state. Fig. 3 A. Distribution of GSTT1 and GSTM1 in central and south regions of Goias state. B. Amount of grain production in the five regions (central, east, northwest, north and south) of Goias state. Exposure to pesticides can produce abnormal amounts of free radicals, leading to overproduction of oxidants or an inadequate supply of antioxidants ( Corsini et al., 2013 ). In this context, occupational exposure to pesticides increases DNA damage, as Franco et al. (2016) noted in their study of public state agents who handle pesticides to combat the mosquito Aedes aegypti , which is responsible for the transmission of arboviruses. Those authors observed that recently intoxicated agents had an increased frequency of translocation between chromosomes 14 and 18. In addition, Roulland et al. (2004) concluded that the BCL2-IGH translocation (14:18) is linked to non-Hodgkin's lymphoma, since it may be a measure of instability caused by the genotoxic effect of the pesticide in a relevant lymphomagenesis. Ventura et al. (2019) demonstrated that 100-day exposure to chlorpyrifos promoted the development of mammary tumors in addition to relieving a hypo-expression of stromal cell receptors. Therefore, risk factors for a variety of diseases, such as smoking, sun exposure, obesity, alcohol consumption, and lack of physical activity, in addition to pesticide exposure, have all gained widespread repercussions due to their impact on public health ( Khayat et al., 2013 ; Donkor et al., 2016 ; Godoy et al., 2019 ). Ingestion of contaminated water and food is a significant source of human exposure to pesticide residues in the general population ( Sharma et al., 2010 ; Donkor et al., 2016 ), although occupational and para-occupational exposures are the most worrying ( Deziel et al., 2017 ; Godoy et al., 2019 ). Neurological, reproductive, and developmental disorders have already been associated with occupational exposure to pesticides; however, the most recognized factor is the risk of developing specific cancers ( Blair and Freeman, 2009 ; Bolognesi et al., 2011 ; Stojanovic et al., 2018 ; Zeng et al., 2017 ). It has also become essential to determine if people living in areas of high agricultural intensity where pesticides are widely used show a higher prevalence and risk of cancer and, ultimately, to elucidate the mechanisms by which these environmental contaminants relate to the emergence of disease ( Alavanja et al., 2013 ; Parrón et al., 2014 ). Additionally, along with daily pesticide exposures, people living in areas of intense agricultural activity may interact with other products to which the general population is exposed, such as pollution, dyes, preservatives, and various products that, for example, can affect the immune system, where malfunction leads to the susceptibility of multiple diseases. Genetic factors, such as the presence of polymorphisms capable of increasing the risk of cancer, should be added to the relevant aspects. In general, the increasing trend in cancer incidence in recent years can be attributed not only to aging but also to the spread of carcinogenic agents that lead to daily exposure. This makes it very important to study the factors related to the risks to which different populations are exposed ( Tebourbi et al., 2011 ). In Goias, the use of pesticides has grown considerably, with 43,466 kg consumed in 2017 alone ( IBAMA, 2019 ). In this context, knowing the critical role of GST enzymes for detoxification pathways and that haplotypes with null alleles are of high risk for the population, this use reached 61% of the Goias population. Finally, our results indicated that it is essential to know the frequencies of the GSTT1 and GSTM1 genes for the monitoring of risk groups; for example, farmers who have contact with pesticides directly or indirectly, as well as assisting in the development of preventive medicine practices.

The use of glyphosate, 2,4-D, acephate, and other pesticides totaling 43,663 tons is related to the high agricultural productivity of Goias, leaving Goias in four places in underreporting numbers of intoxication in Brazil. However, with the new Brazilian government authorizing the use of banned pesticides worldwide, this situation may increase the number of poisonings, especially in Brazilian regions involved with agriculture. With the high frequency of null genotypes in the south and central regions of Goias, and also with these regions being the first and third in the ranking of grain production respectively, it is of great importance that the government take measures to prevent cases of intoxication, given the risk that the null GSTT1 and GSTM1 genotypes offer for people exposed to pesticides. In this context, GSTT1 and GSTM1 genotypes may be essential biomarkers of susceptibility to poisoning, mainly in populations with pesticide contact, due to the role of GST enzymes. However, it is necessary to associate GSTT1 and GSTM1 genotyping with other biomarkers such as comet assay, micronucleus test, oxidative stress enzyme measurements, for confirmation, quantification of biological damage and better elucidation of a given natural process. This study presented a combined frequency (haplotypes) of GSTT1 and GSTM1 genes in Central Brazil. Future studies to monitor the toxicity of occupationally exposed workers and planting areas in which pesticides are used, as well as non-target animals, are necessary to verify human and environmental health, against the use of pesticides in this Brazilian region, which is characterized by intense agricultural activity.

Alessandro Arruda Alves: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper. Fernanda Craveiro Franco: Performed the experiments; Analyzed and interpreted the data. Fernanda Ribeiro Godoy: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data. Jheneffer Sonara Aguiar Ramos, Hugo Freire Nunes, Thannya Nascimento Soares: Contributed reagents, materials, analysis tools or data; Wrote the paper. Daniela de Melo e Silva: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper. This work was supported by CNPq (Process: 402222/2016-4) and FAPEG /PPSUS (Process: 201810267000111). AAA, FRG, HFN, TNS and DMS, thank CAPES and CNPq for their fellowships. The authors declare no conflict of interest. No additional information is available for this paper.

Brazil has been the largest consumer of pesticides in the world due to intense agricultural activity and government incentives ( Abreu e Alonzo, 2014 ). Also, until October 2019, the Brazilian government authorized the commercialization of 382 pesticides, many of them banned in other parts of the world, making Brazil one of the countries that uses these products most, increasing the risk to human, animal and environmental health ( Ministério da Agricultura, 2019 ). The State of Goias is in Central Brazil and is one of the largest producers of corn, beans, cotton, sorghum, sugarcane, and soybeans in the country, with 75% of its exports composed of agricultural products ( Instituto Mauro Borges, 2018 ). This vast production led to the use of 43,466.3 tons of active pesticide ingredients in 2017. Among the 10 (ten) most used in Goias are: glyphosate (15,486 tons of active ingredient [AI]); 2,4-D (4,026 tons AI); acephate (2,374 tons AI); mineral oil (1,949 tons AI); atrazine (1,919 tons of AI); mancozeb (1,797 tons of AI); cypermethrin (771 tons of AI); paraquat dichloride (718 tons of AI) and carbendazim (606 tons of AI) ( IBAMA, 2019 ). In this case, Goias became the fourth-highest Brazilian state in commercialized pesticide use ( Instituto Mauro Borges, 2018 ). In this context, the use of pesticides has intensified, and there is an increase in inadequate handling and excessive use. These situations lead to the presence of pesticides in ponds, rivers, and many other water sources, polluting the environment and damaging both fauna and humans, mainly due to bioaccumulation in food chains ( Souza et al., 2011 ). This may cause respiratory diseases ( Ye et al., 2013 ), non-Hodgkin's lymphoma ( Schinasi and Leon, 2014 ), lung cancer ( Bonner et al., 2016 ), changes in reproductive hormones ( Miranda-Contreras et al., 2013 ), Parkinson's disease ( Panov et al., 2002 ), and Alzheimer's disease ( Richardson et al., 2014 ). In 2015, it was underreported that there were 31,900 cases of poisoning by pesticides in Goiás, according to the Minister of Health ( Ministério da Saúde, 2016 ). In 2018, there were 12.74 incidences of poisoning notifications in every 100,000 inhabitants, while Brazil reported only 6.24 ( Ministério da Saúde, 2018 ). Pesticides are xenobiotics metabolized in two phases in the liver ( Coles and Kadlubar, 2003 ). The superfamily Glutathione S-transferase (GST) is involved in the conjugation phase, making the compounds less toxic and protecting the cell against oxidative damage and electrolyte substrates ( Da Fonseca et al., 2010 ). The subclasses of the GST superfamily studied most in mammals are the Mu (μ), Pi (п), and Theta genes. The gene GSTM1 (μ) is 4.2 kb, consisting of eight exons and is located on chromosome 1p13.3, with four known alleles. The most frequent is the wild-type allele, and the second one is the null allele characterized by a structural deletion of a region of the gene that causes the loss of the function of the enzyme (θ) ( Rodríguez et al., 2014 ; Strange et al., 2001 ). The gene GSTT1 (θ) is located on chromosome 22p11.2, is composed of six exons, and is flanked by two homologous regions HA3 and HA5. This gene has two alleles known as the most frequent allele (wild) and the null allele. This polymorphism occurs due to the recombination of the homologous regions HA3 and HA5, resulting in the deletion of 5.4 kb and, as a consequence, the loss of function of the enzyme ( Rodríguez et al., 2014 ; Webb et al., 1996 ). In this context, individuals with null GSTM1 genotypes are associated with an increased risk for the development of a variety of diseases, such as arteriosclerosis, Parkinson's disease, and various types of cancer, whereas the null GSTT1 genotype is associated with the colorectal and central nervous systems ( Strange et al., 2001 ; Wahner et al., 2007 ; Wang et al., 2010 ). Thus, the knowledge of the distribution of GSTM1 and GSTT1 polymorphisms in an agricultural population, such as that of Goias, may assist in understanding the frequency of diseases related to poisoning by pesticides or other xenobiotic agents. As Goias is a famous agricultural region, the study of these genes in municipalities of Central Brazil is significant. Therefore, it could be possible to associate the exposure to pesticides, with the null genotypes of GSTT 1 and GSTM1 , considered to be at risk for the development of chronic diseases, such as cancer.