Frequency of the Q 192 R and L 55 M polymorphisms of the human serum paraoxonase gene ( PON 1 ) in ten Amazonian Amerindian tribes

Human serum paraoxonase (PON1) is an esterase associated with high density lipoproteins (HDLs) in the plasma and may confer protection against coronary artery disease. Serum PON1 levels and activity vary widely among individuals and populations of different ethnic groups, such variations appearing to be related to two coding region polymorphisms (L55M and Q192R). Several independent studies have indicated that the polymorphism at codon 192 (the R form) is a significant risk factor for cardiovascular disease in some populations, although this association has not been confirmed in other populations. Given the possible associations of these mutations with heart diseases and the fact that little or nothing is known of their prevalence in Amerindian populations, we investigated the variability of both polymorphisms in ten Amazonian Indian tribes and compared the variation found with that of other Asian populations in which both polymorphisms have been investigated. The results show that the LR haplotype is the most frequent and the MR haplotype is absent in all Amerindians and Asian populations. We also found that South America Amerindians present the highest frequency of the PON1192*R allele (considered a significant risk factor for heart diseases in some populations) of all the Amerindian and Asian populations so far studied.


Paraoxonase polymorphisms in Amazonian Amerindians
Human serum paraoxonase (PON1) is an esterase associated with serum high-density lipoproteins (HDLs) and is coded by a gene located on the 7q21.3-22.1 region of chromosome 7 in a cluster with two similar genes, PON2 and PON3 (Primo-Parmo et al., 1996).The PON1 protein has been found to be involved in the detoxification of organophosphate insecticides such as parathion and the neural agents soman and sarin (Davies et al., 1996) but PON1 may also confer protection against coronary artery disease by destroying pro-inflammatory oxidized lipids present in oxidized low density lipoproteins (Mackness et al., 1998).
The PON1 coding region contains two common polymorphisms, a leucine (L) to methionine (M) substitution at codon 55 (L55M) and a glutamine (Q) to arginine (R) substitution at codon 192 (Q192R).These polymorphisms are only 8.6 Kb apart from each other (Gen Bank accession number AC004022).By the direct sequencing of the PON1 gene in orangutans and chimpanzees, Koda et al. (2004) showed that the allele which codifies arginine in the 192 codon (PON1192*R) is the ancestor allele in human populations.
Serum PON1 levels and activity varies widely among individuals and populations of different ethnic groups and seems to be related to the Q192R and L55M polymorphisms, with individuals bearing the 192R isoform having higher PON1 activity than those with the 192Q isoform (Garin et al., 1997).
Several independent studies have indicated that the polymorphism at codon 192 (the R form) is a significant risk factor for cardiovascular disease in some populations (Serrato and Marian, 1995;Sanghera et al., 1998a;Durrington et al., 2001;Aynacioglu and Kepekci 2000;Imai et al., 2000;Gnasso et al,. 2002), although this association has not been confirmed in other populations (Antikainen et al., 1996;Herrmann et al., 1996;Suehiro et al., 1996;Garin et al., 1997;Ombres et al., 1998).A possible explanation for this inconsistent association is that the codon 192 polymorphism is related to risk of cardiovascular disease only when it is associated with other mutation(s) in a particular haplotype (Koda et al., 2004).
Given the possible associations of these mutations with cardiovascular diseases and the fact that little or nothing is known of the prevalence of these mutations in Amer-indian populations we investigated the frequency of the Q192R and L55M polymorphisms in ten Amazonian Indian tribes in which heart disease is rare, possibly because of their non-urban lifestyle, and compared the variation found with that of other Asian populations to estimate if the allele frequencies differ significantly.
Between 1983 and 1990 our team studied 259 putatively unrelated individuals from ten tribes with distinct linguistic and cultural backgrounds (Poturujara; Mapuera; Wayampi; Arara; Parakanã; Assurini of Koatinemo; Awá-Guajá; Yanomami; Urubu-Kaapor and Kayapo) which were distributed over a large geographic area of the Brazilian Amazon.Additional data on the genetic and linguistic characterization of theses groups have been published previously (Silva Jr. et al., 2002;Ribeiro-dos-Santos et al., 2001;Vallinoto et al., 1999;Santos et al., 1998).This research was approved by the ethical committee of the Federal University of Pará, Brazil.
Genomic DNA was isolated from whole blood by the salting out method of Miller et al. (1998) and the method described by Motti et al. (2001), with some modifications, was used to simultaneously determine the three most common polymorphisms of the PON cluster (PON1-192, PON1-55 and PON2-311) using a multiplex polymerase chain reaction (PCR) DNA assay with mismatch primers to introduce a unique recognition site for the Hinf I endonuclease in the PCR products when the PON1*192R or PON1*55L alleles are present.Our study did not involve the PON2 polymorphisms.
The multiplex-PCR was performed in a 15 µL reaction mixture containing 100 ng of the template DNA amplified using an initial step of 94 °C for 2 min followed by 30 cycles for 30 s at 94 °C, 45 s at 60 °C and 45 s at 72 °C, with a final extension of 5 min at 72 °C.The total product of the Multiplex-PCR amplification (15 µL) was digested overnight at 37 °C with 5 U of Hinf I (Boehringer Mannheim, Monza, Italy) in a total volume of 25 µL, the digestion products being separated by polyacrylamide gel (9% w/v) electrophoresis and stained with silver.
Allele and haplotype frequencies were computed using an expectation-maximization (EM) algorithm (Excoffier and Slatkin, 1995) and the Hardy-Weinberg equilibrium and linkage disequilibrium evaluated by the exact test using the Markov chain method (Slatkin and Excoffier, 1996).Comparisons between populations were performed using the exact test for sample differentiation based on haplotype frequencies (Raymond and Rousset, 1995).All evaluations were performed using the ARLEQUIN computer program (Schneider et al., 2000).
Table 1 shows the combined genotype and allele frequencies of the two PON1 gene polymorphisms (L55M and Q192R) in the ten tribes.In the Poturujara tribe only the PON1*55L and PON1*192R alleles were found but In the majority of the tribes (Mapuera, Waiampi, Arara, Awa-Guajá, Assurini, Urubu-Kaapor) only the PON1*55L allele was observed.The PON1*55M allele is present in the heterozygous form in individuals from the Parakanã, Yanomami and Kaiapo tribes at frequencies of between 6% and 21%.Considering all subjects, the observed frequencies were PON1*55L = 0.97 and PON1*55M = 0.03.In the majority of the tribes, the frequency of the PON1*192R allele is high, varying from 0.53 to 1.000, except in the Yanomami, which had the highest PON1*192Q frequency (0.605).Overall, the observed frequencies were PON1*192R = 0.73 and PON1*192Q = 0.27.
Only five of the nine possible genotypic combinations of the two polymorphisms were recorded, the most common being the LL/RR double homozygote (56%) followed by the LL/QR (31%), with the LM/QR double-heterozygote being rare and found only in individuals from the Parakanã, Yanomami and Kayapo tribes.The genotype frequencies for the total Amerindian population were distributed according to a Hardy-Weinberg equilibrium for both the L55M (χ 2 = 0.19, p = 0.98) and Q192R (χ 2 = 3.25, p = 0.211) polymorphisms.
Table 2 presents the haplotype and allele frequencies observed in general for the Amerindian tribes studied by us and three Asian populations (Chinese (Sanghera et al., 1998b), Korean (Hong et al., 2001) and Japanese (Yamada et al., 2003)) investigated for both polymorphisms.The dis-  Sanghera et al., 1998b; 2 Hong et al., 2001; 3 Yamada et al., 2003.tribution of the PON1*55L allele is relatively homogeneous, with frequencies ranging from 93% for the Japanese population to 97% for the Amerindian population, while the PON1*192R allele has a broader distribution, ranging from 57% in Chinese to 73% in Amerindians.
The MR haplotype is absent in Amerindian and Asian populations, while the LR haplotype occurs at a relatively high frequency in Asian populations, ranging from 57% in Chinese to 73% in Amerindians, and the MQ haplotype has a very low prevalence of less than 7% among Asian populations.The LQ haplotype has the broadest variation of all the haplotypes, ranging from 24% in Amerindians to 39% in Chinese.
There are few published data on Amerindian populations with which to compare our results but studies carried out on two Canadian tribes (the Oji-Cree and the Inuit) and the Ecuadorian Cayapa tribe have noted variability in the frequency of the Q192R polymorphism (Hegele, 1999;Hegele et al., 2001;Scacci et al., 2003) with the frequency of the PON1192*R allele being 0.770 for the Oji-Cree, 0.700 for the Inuit and 0.789 for the Cayapa, values similar to the mean value of 0.730 found by us for the Amazonian Amerindians.
The data presented in Table 2 were used to test the linkage disequilibrium (Slatkin and Excoffier, 1996) between the Q192R and L55M loci and to compare all populations by means of the exact test of sample differentiation based on haplotype frequencies (Raymond and Rousset 1995).The tests show that all the loci studied were in linkage disequilibrium (p < 0.001) in all populations, such a result being predictable because these two polymorphisms are only 8.6 Kb apart.
Our comparison shows that the Amerindian population studied by us is different from all other populations (p < 0.001, in all comparisons) and that the Japanese population is different from the Korean (p = 0.023) and the Chinese (p < 0.001) populations Only the Chinese and Korean populations did not differ statistically (p = 0.211).
Our data show that the distribution of the L55M and Q192R haplotypes of the Paraoxonase 1 gene of the Amazonian Amerindian tribes investigated by us are different from those observed among Asian populations and that Amazonian Amerindians present the highest frequency of the PON1192*R allele, considered a significant risk factor for heart diseases in some, populations of all the Asian populations so far studied.

Table 1 -
Combined frequencies of the L55M and Q192R polymorphisms and the allele frequencies observed in ten Amazonian Amerindian tribes.