Paraoxonase 1
Description
The paraoxonase (PON) gene family includes 3 genes, PON1, PON2 (602447), and PON3 (602720), aligned next to each other on chromosome 7. PON1 (EC 3.1.1.2) hydrolyzes the toxic oxon metabolites of several organophosphorous insecticides, including parathion, diazinon, and chlorpyrifos, as well as nerve agents, such as sarin and soman. PON1 also hydrolyzes aromatic esters, preferably those of acetic acid. In addition, PON1 hydrolyzes a variety of aromatic and aliphatic lactones, and it also catalyzes the reverse reaction, lactonization, of gamma- and delta-hydroxycarboxylic acids. Human PON1 is synthesized in liver and secreted into blood, where it is associated exclusively with high density lipoproteins (HDLs) and may protect against development of atherosclerosis (Draganov et al., 2005).
Cloning and ExpressionHassett et al. (1991) isolated a full-length PON1 cDNA from a human liver cDNA library using rabbit Pon1 as a hybridization probe. The deduced PON1 protein contains 355 amino acids and is more than 85% similar to the rabbit protein. N-terminal sequences derived from purified rabbit and human PON1 proteins suggested that the PON1 signal sequence is retained, except for the initiator methionine. Characterization of the rabbit and human PON1 cDNAs confirmed that the signal sequences are not processed, except for the N-terminal methionine.
Using SDS-PAGE, Draganov et al. (2005) found that PON1 appeared as a doublet of about 39 and 42 kD. However, using nondenaturing PAGE, they observed human serum PON1 and recombinant PON1 at apparent molecular masses of 91.9 and 95.6 kD, respectively, suggesting that PON1 forms dimers. Glycosidase treatment of human serum PON1 suggested that the secreted form of PON1 contains complex carbohydrates.
Lu et al. (2006) stated that human PON1, PON2, and PON3 have 3 conserved cysteines. Cys41 and cys351 are predicted to form an intramolecular disulfide bond, and cys283 is predicted to be involved in antioxidant activity.
Gene StructureClendenning et al. (1996) characterized a 28-kb contig encompassing 300 bp of 5-prime sequence, the entire coding region, and 2 kb of 3-prime flanking sequence of the PON1 gene. The structural portion of the paraoxonase protein is encoded by 9 exons that form the primary transcript through the use of typical splice donor and acceptor sites.
Sorenson et al. (1995) showed that the Pon1 gene in mice contains 9 exons spanning approximately 25 to 26 kb.
MappingEiberg and Mohr (1979) presented linkage data. No linkage with any of 19 markers was found by Mueller et al. (1983). Eiberg et al. (1985) showed that cystic fibrosis (219700) and PON are linked on chromosome 7 (maximum lod 3.70 at theta = 0.07 in males and 0.00 in females)--the first step in the cloning of the CF gene in 1989. Tsui et al. (1985) confirmed the PON-CF linkage by finding linkage of PON to a DNA marker that is also linked to CF. Schmiegelow et al. (1986) found the PON and CF loci linked with lod score of 3.46 at recombination fraction 0.07 in males and 0.13 in females. By in situ hybridization, Humbert et al. (1993) demonstrated that the PON gene maps to chromosome 7q21-q22. Mochizuki et al. (1998) pointed out that the PON1, PON2, and PON3 genes are physically linked on chromosome 7q21.3.
Sorenson et al. (1995) mapped the mouse Pon1 gene to the proximal end of chromosome 6 by interspecific backcross analysis. Li et al. (1997) likewise mapped the gene to mouse chromosome 6.
Gene FunctionSimpson (1971) found a unimodal distribution of serum arylesterase activity in 176 individuals. There was no difference in enzyme activity between sexes, but the level of activity gradually increased with age. From a study of twins, heritability of arylesterase activity was estimated to be 74%. Data from parent pairs suggested that, in addition to genetic and age factors, unknown nongenetic factors substantially affected enzyme activity.
Eckerson et al. (1983) concluded that arylesterase activity, measured with phenylacetate as substrate, and paraoxonase activity are determined by the same gene. They used the designation esterase A for the paraoxonase/arylesterase enzyme (see HISTORY for information on the identification and classification of esterases). Furlong et al. (1991) also demonstrated that both arylesterase and paraoxonase activities are expressed by a single enzyme.
Furlong et al. (1988) studied hydrolysis of an insecticide metabolite, chlorpyrifos oxon, by PON1.
The physiologic role of paraoxonase in detoxication and in intermediary metabolism is uncertain (La Du, 1988). However, animal studies, including examination of quantitative adequacy of PON and protection against paraoxon toxicity, correlation of LD50 values with PON levels, and demonstration that intravenously injected PON provides protection against paraoxon toxicity, indicate that serum PON is protective against organophosphate poisoning (reviewed by Humbert et al., 1993).
In a series of animal experiments, Navab et al. (1997) demonstrated that the ratio of Apoj (185430) to Pon was increased in fatty streak-susceptible mice fed an atherogenic diet, in Apoe knockout mice on a chow diet, in LDL receptor (LDLR; 606945) knockout mice on a cholesterol-enriched diet, and in fatty streak-susceptible mice injected with mildly oxidized LDL fed a chow diet. Human studies showed that the APOJ/PON ratio was significantly higher than that of controls in 14 normolipidemic patients with coronary artery disease in whom the cholesterol/HDL ratio did not differ significantly from that of controls.
Draganov et al. (2005) found that glycosylation of recombinant PON1 with high-mannose-type sugars did not alter its enzymatic activity, but it may have affected protein stability. They found that PON1, PON2, and PON3, whether expressed in insect or HEK293 cells, metabolized oxidized forms of arachidonic acid and docosahexaenoic acid. Otherwise, the PONs showed distinctive substrate specificities. PON1, but not other PONs, specifically hydrolyzed organophosphates. About 60% of total arylesterase and lactonase activity of PON1-transfected HEK293 cells was secreted into the culture medium. Draganov et al. (2005) found that recombinant PONs did not protect human LDL against Cu(2+)-induced oxidation in vitro, and no antioxidant activity copurified with any of the PONs. They stated that they had previously copurified antioxidant activity with human serum PON1, but that it was attributable to a low molecular mass contaminant and to the detergent in the preparation.
Molecular GeneticsVariation in PON1 Enzyme Activity
Geldmacher-von Mallinck et al. (1973) first found polymorphism of paraoxonase activity.
Playfer et al. (1976) found bimodality for plasma paraoxonase activity in British and Indian persons, defining low and high activity phenotypes. Study of 40 British families confirmed this genetic polymorphism. Two phenotypes controlled by 2 alleles at 1 autosomal locus were defined. The frequency of the low activity phenotype was lower in the Indian population than in the British population. Malay, Chinese, and African populations failed to show clear bimodality. Possibly multiple alleles are present in these populations and result in a continuous distribution.
Mueller et al. (1983) described a test based on the differential inhibition of EDTA of plasma paraoxonase from persons with the high activity allele. With this test, trimodality of activity levels was suggested by population studies. The gene frequency of the low activity allele in 531 Seattle blood donors of European origin was 0.72. Family studies were consistent with codominant autosomal inheritance of 2 alleles encoding products with low and high activity levels.
Eckerson et al. (1983) could clearly distinguish heterozygotes from both homozygous phenotypes on the basis of the ratio of paraoxonase to arylesterase activities.
Ortigoza-Ferado et al. (1984) concluded that albumin has paraoxonase activity and proposed that an optimal assay of polymorphic paraoxonase activity should be based on activity of the nonalbumin fraction.
Nielsen et al. (1986) reexamined extensive family data and reaffirmed that segregation into high and low paraoxonase activity is largely or exclusively due to a 1-locus system.
Humbert et al. (1993) found that arginine at position 192 of PON1 specifies high-activity plasma paraoxonase, whereas glutamine at this position specifies a low-activity variant (Q192R; 168820.0001). This polymorphism is also referred to as gln191 to arg. Adkins et al. (1993) demonstrated that glutamine or arginine at amino acid 191 determines the A and B allozymes, respectively, of PON1.
In a study of 376 white individuals, Brophy et al. (2001) determined the genotypes of 3 regulatory region polymorphisms and examined their effect on plasma PON1 levels as indicated by rates of phenylacetate hydrolysis. The -108C-T polymorphism (168820.0003) had a significant effect on PON1 activity level, whereas a polymorphism at position -162 had a lesser effect. A polymorphism at position -909, which is in linkage disequilibrium with the other sites, appeared to have little or no independent effect on PON1 activity level in vivo. Brophy et al. (2001) presented evidence that the effect of the L55M (168820.0002) polymorphism on lowered paraoxonase activity is not due primarily to the amino acid change itself but to linkage disequilibrium with the -108C-T regulatory region polymorphism. The L55M polymorphism marginally appeared to account for 15.3% of the variance in PON1 activity, but this dropped to 5% after adjustments for the effects of the -108C-T and Q192R polymorphisms were made. The -108C-T polymorphism accounted for 22.8% of the observed variability in PON1 expression levels, which was much greater than that attributable to other PON1 polymorphisms.
Using a validated microsomal expression system of metabolizing enzymes, Bouman et al. (2011) identified PON1 as the crucial enzyme for the bioactivation of the antiplatelet drug clopidogrel, with the common Q192R polymorphism determining the rate of active metabolite formation. Analysis of patients with coronary artery disease who underwent stent implantation and received clopidogrel therapy revealed that Q192 homozygotes were more likely to undergo stent thrombosis than patients with the RR192 or QR192 genotypes (odds ratio, 3.6; p = 0.0003).
Susceptibility to Organophosphate Poisoning
PON1 hydrolyzes diazinonoxon, the active metabolite of diazinon, which is an organophosphate used in sheep dip. Cherry et al. (2002) studied PON1 polymorphisms in 175 farmers with ill health that they attributed to sheep dip and 234 other farmers nominated by the ill farmers and thought to be in good health despite having also dipped sheep. They calculated odds ratios for the Q192R (168820.0001) and L55M (168820.0002) polymorphisms, and for PON1 activity with diazinonoxon as substrate. Cases were more likely than referents to have at least 1 R allele at position 192 (odds ratio 1.93), both alleles of type LL (odds ratio 1.70) at position 55, and to have diazoxonase activity below normal median (odds ratio 1.77). The results supported the hypothesis that organophosphates contribute to the reported ill health of people who dip sheep.
Susceptibility to Coronary Artery Disease
Serrato and Marian (1995), who referred to the gln192-to-arg (Q192R; 168820.0001) polymorphism as the A/G polymorphism or the A/B polymorphism, demonstrated a relationship to coronary artery disease. The A and G alleles code for glutamine (A genotype) and arginine (B genotype), respectively. Individuals with the A genotype have a lower enzymatic activity than those with the B genotype. Serrato and Marian (1995) determined the genotypes in 223 patients with angiographically documented coronary artery disease and in 247 individuals in the general population. The distribution of genotypes was in Hardy-Weinberg equilibrium in both groups. Genotypes A and B were present in 49% and 11% of control individuals and in 30% and 18% of patients with coronary artery disease, respectively (p = 0.0003). The frequency of the A allele was 0.69 in controls and 0.56 in patients (p = 0.0001). There was no difference in the distribution of PON genotypes in the subgroups of patients with restenosis, myocardial infarction, or any of the conventional risk factors for coronary artery disease as compared with corresponding subgroups.
The L55M and Q192R polymorphisms in the PON1 gene and the ser311-to-cys (S311C; 602447.0001) polymorphism in the PON2 gene are associated with the risk of coronary artery disease in several European or European-derived populations. Chen et al. (2003) examined the association between these 3 markers and the severity of coronary artery disease as determined by the number of diseased coronary artery vessels in 711 women. No significant association was found between the PON polymorphisms and stenosis severity in either white or black women. However, among white women, when data were stratified by the number of diseased vessels, the frequency of the PON codon 192 arg/arg genotype was significantly higher in the group with 3-vessel disease than in the other groups (those with 1-vessel and 2-vessel disease) combined. Similarly, the frequency of the PON2 codon 311 cys/cys genotype was significantly higher in the group with 3-vessel disease than in the other groups combined. The adjusted odds ratio for the development of 3-vessel disease was 2.80 for PON1 codon 192 arg/arg and 3.68 for PON2 codon 311 cys/cys. The data indicated that the severity of coronary artery disease, in terms of the number of diseased vessels, may be affected by common genetic variation in the PON gene cluster on chromosome 7.
Garin et al. (1997) identified homozygosity for the leu54 allele of PON1 (168820.0002), which is associated with high paraoxonase activity, as an independent risk factor for cardiovascular disease. A linkage disequilibrium was apparent between the polymorphisms giving rise to leu54 and arg191. Garin et al. (1997) stated that their study underlined the fact that susceptibility to cardiovascular disease correlated with high-activity paraoxonase alleles. Linkage disequilibrium could explain the association between both the leu54 and the arg191 polymorphisms and cardiovascular disease.
Susceptibility to Coronary Artery Spasm
Ito et al. (2002) found that the incidence of the PON1 192R allele was significantly higher in a cohort of 214 Japanese patients with coronary spasm than in 212 control subjects. They speculated that the high frequency of the PON1 arg192 allele may be related to the higher prevalence of coronary spasm among Japanese than among Caucasians.
Susceptibility to Microvascular Complications of Diabetes 5
Kao et al. (1998) found an association between the L55M polymorphism in the PON1 gene (168820.0002) and diabetic retinopathy (MVCD5; 603933) in patients with type 1 diabetes (222100).
Kao et al. (2002) confirmed the association between L55M and diabetic retinopathy, finding increased susceptibility for retinopathy with the leu/leu genotype (odds ratio 3.34; p less than 0.0001).
Other Associations
Ikeda et al. (2001) found that the distribution of the Q192R and L55M (168820.0002) polymorphisms in the PON1 gene was significantly different between Japanese patients with exudative age-related macular degeneration (ARMD; see 153800) and age- and sex-matched controls. The BB genotype at position 192 and the LL genotype at position 55 occurred at higher frequency in patients with ARMD compared to controls (p = 0.0127 and p = 0.0090, respectively). The mean oxidized LDL level in patients was significantly higher than in controls (p less than 0.01). Ikeda et al. (2001) concluded that the PON1 gene polymorphisms might represent a genetic risk factor for ARMD and that increased plasma oxidized LDL might be involved in the pathogenesis of ARMD.
Data on gene frequencies of allelic variants were tabulated by Roychoudhury and Nei (1988).
Genotype/Phenotype CorrelationsDavies et al. (1996) analyzed the paraoxonase catalytic activity against the toxic oxon forms which result from the bioactivation of the organophosphorus insecticides parathion, chlorpyrifos, and diazinon in the P450 system. They also analyzed the hydrolytic activity of PON1 against the nerve agents soman and sarin. Davies et al. (1996) reported a simple enzyme analysis that provided a clear resolution of PON1 genotypes and phenotypes. The plot of diazoxon hydrolysis versus paraoxon hydrolysis clearly resolved all 3 genotypes (Q192Q192, Q191R192, R192R192; see 168820.0001) and at the same time provided important information about the level of enzyme activity in an individual. They observed the reversal of the effect of PON1 polymorphisms for diazoxon hydrolysis relative to paraoxon hydrolysis. RR homozygotes (high paraoxonase activity) had lower diazoxonase activity than the QQ homozygotes (low paraoxonase activity). They reported that the effect was also reversed for the nerve gases soman and sarin (sarin was the nerve gas released in the Tokyo subway in March 1995). The mean value of sarin hydrolysis was only 38 U per liter for the R192 homozygotes compared with 355 U per liter for the Q192 homozygotes. Davies et al. (1996) observed an increased frequency for the R192 allele (0.41) in the Hispanic population compared with a frequency of 0.31 for populations of northern European origin. These frequencies result in approximately 16% of individuals of Hispanic origin being homozygous for the R192 PON1 isoform compared with 9% of individuals of northern European origin. They noted that newborns have very low activities of PON1, leading them to predict that newborns would be significantly more sensitive to organophosphorus compounds than adults. The authors cited studies showing that injected PON1 protects against organophosphorus poisoning in rodents (Li et al., 1995).
Phuntuwate et al. (2005) studied the activity of 4 PON1 polymorphisms towards paraoxon, phenylacetate, and diazoxon. They found that the L55M, Q192R, and -909G-C polymorphisms significantly and variably affected serum PON1 activity towards the substrates, whereas the -108C-T polymorphism had no significant effect on serum PON1 activity towards any substrate. Phuntuwate et al. (2005) suggested that the physiologic relevance of the PON1 polymorphisms is that they are associated with significant differences in serum PON1 activity that are substrate dependent.
Mackness et al. (1998) examined the effects of the 2 common polymorphisms in PON1 on the ability of HDL to protect LDL from oxidative modification. HDL protected LDL from oxidative modification, whatever the combination of PON1 alloenzymes present in it. However, HDL from QQ/MM homozygotes was most effective in protecting LDL, while HDL from RR/LL homozygotes was least effective. Thus after 6 hours of coincubation of HDL and LDL with Cu(2+), PON1-QQ HDL retained 57 +/- 6.3% of its original ability to protect LDL from oxidative modification, while PON1-QR HDL retained less at 25.1 +/- 4.5% and PON1-RR HDL retained only 0.75 +/- 0.40%. In similar experiments, HDL from LL and LM genotypes retained 21.8 +/- 7.5% and 29.5 +/- 6.6%, respectively, of their protective ability, whereas PON1-MM HDL maintained 49.5 +/- 5.3%. PON1 polymorphisms may affect the ability of HDL to impede the development of atherosclerosis and to prevent inflammation.
EvolutionMeyer et al. (2018) used phylogenetic methods to identify convergent functional losses across independent marine mammal lineages. They determined that PON1 has accrued lesions in all marine lineages while remaining intact in all terrestrial mammals. These lesions coincide with PON1 enzymatic activity loss in marine species' blood plasma. This convergent loss is likely explained by parallel shifts in marine ancestors' lipid metabolism and/or bloodstream oxidative environment affecting PON1's role in fatty acid oxidation. PON1 loss also eliminates marine mammals' main defense against neurotoxicity from specific man-made organophosphorus compounds, implying potential risks in modern environments.
Animal ModelTo study the role of PON1 in vivo, Shih et al. (1998) created Pon1-knockout mice by gene targeting. Compared with their wildtype littermates, Pon1-deficient mice were extremely sensitive to the toxic effects of chlorpyrifos oxon, the activated form of chlorpyrifos, and were more sensitive to chlorpyrifos itself. HDLs isolated from Pon1-deficient mice were unable to prevent LDL oxidation in a cocultured cell model of the artery wall, and both HDLs and LDLs isolated from Pon1-knockout mice were more susceptible to oxidation by cocultured cells than were lipoproteins from wildtype littermates. When fed on a high-fat, high-cholesterol diet, Pon1-null mice were more susceptible to atherosclerosis than were their wildtype littermates.
Watson et al. (2001) identified a serum paraoxonase polymorphism in rabbit with functional characteristics similar to those of human Q192R. They suggested that the rabbit may serve as a model in examining the effect of human PON1 polymorphisms in disease development.
HistoryIdentification and Classification of Esterases
Using azo dye coupling techniques and electrophoresis, Tashian (1965) defined several different esterases in human red cells. Three main groups, differing as to electrophoretic properties, substrate specificities and inhibition characteristics, were A, B, and C esterases. Variants of esterase A were reported by Tashian and Shaw (1962) and Tashian (1965).
Using starch-gel electrophoresis, Coates et al. (1975) identified multiple esterase isozymes in every human tissue, and they characterized the isozymes in terms of electrophoretic mobility, tissue distribution, developmental changes in utero, substrate specificity, inhibition properties, and molecular weight. On these criteria, 13 sets of esterase isozymes were identified, in addition to the esterase isozymes due to cholinesterase and carbonic anhydrase. The data suggested that the 13 sets of isozymes are determined by at least 9 different genes. The acetylesterases, which prefer acetate esters as substrates, were divided into 9 sets of isozymes, designated ESA1 to ESA7, ESC (133270), and ESD (133280). Coates et al. (1975) divided the butyrylesterases, which prefer butyrate esters as substrates, into 4 sets of isozymes, designated ESB1 to ESB4.