Gln-Arg192 polymorphism of paraoxonase and coronary heart disease in type 2 diabetes

Gln-Arg192 polymorphism of disease in type 2 diabetes

paraoxonase and coronary heart

Introduction

Summary is

high-density-lipoprotein-associated enzyme capable of hydrolysing lipid peroxides. Thus it might protect lipoproteins from oxidation. It has two isoforms, which arise from a glutamine (A isoform) to arginine (B isoform) interchange at position 192. The relevance of this polymorphism to coronary heart disease (CHD) in non-insulin-dependent diabetic patients was investigated in case-control study. Of the 434 patients, 171 had confirmed coronary artery disease; the other 263 had no history of such disease. The Paraoxonase

a

B allele and AB+BB genotypes were associated with an increased risk of coronary heart disease. Compared with subjects homozygous for the A allele (AA genotype), the odds ratio of CHD for subjects homozygous for the B allele was 2·5 (95% Cl 1·2-5·3) and that for those heterozygous for the B allele was 1·6 (95% CI 1·1-2·4), suggesting a codominant effect on cardiovascular risk. When subjected to multivariate analysis, the B allele remained significantly associated with CHD (odds ratio 1·94, p=0·03). The paraoxonase gene polymorphism is thus an independent cardiovascular risk factor in non-insulindependent diabetic patients. A possible explanation for this finding is that activity of the paraoxonase B isotype

does not protect well

against lipid oxidation,

a

major

atherogenic pathway.

mortality and morbidity in nondiabetic insulin-dependent (NIDDM) patients are heart disease coronary (CHD) and other vascular diseases.’ The high risk of CHD is partly explained by hyperglycaemia and by the frequent association of NIDDM with other cardiovascular risk factors, including arterial hypertension and dyslipidaemia.2 All of these cardiovascular risk factors have strong genetic which contribute to macrovascular as components,3 may well as microvascular complications of diabetes mellitus. For example, recent reports indicate an important role for the gene for the angiotensin-converting enzyme in the risk of nephropathy in insulin-dependent-diabetic patients4 and of CHD in NIDDM patients.5 Genetic determinants may accelerate diabetes-induced atherosclerosis. Oxidation of low-density lipoproteins (LDL) has been implicated in the development of coronary heart disease.6 Oxidised lipoproteins are atherogenic and initiate foam-cell chronic Furthermore, development.’7 " the of and advanced production hyperglycaemia 7

The main

causes

of

glycosylation end-products9 seem to predispose oxidation. This predisposition may contribute

to

lipid

the risk for atherosclerotic disease in diabetes. Paraoxonase is an enzyme exclusively bound to highdensity lipoproteins (HDL) in human serum.’" Although its natural substrate is still unknown, there is growing evidence that the enzyme is able to hydrolyse lipid peroxides. 11,12 This property may partly explain the intriguing observation that HDL protects LDL from excess oxidation." Human paraoxonase exhibits wide variation in its activity on certain exogenous substrates. On the basis of relative enzyme activities, two isotypes, A and B, have been defined. The isotype difference has been linked to the presence of a single aminoacid to

excess

polymorphism (Gln-Argl92) of the paraoxonase protein. 14 The physiological relevance of this polymorphism is unknown. A case-control study in a group of French NIDDM patients was thus conducted to address this question. Patients and methods Patients Division d’Epidémiologie Clinique, Geneva University Hospital, Switzerland (J Ruiz MD, A Morabia MD); CEPH, Fondation Jean Dausset, Paris, France (H Blanche PhD); Division d’Endocrinologie et Diabétologie, Geneva University Hospital, Switzerland (R W James PhD, M-C B Garin PhD); Hôpital de Corbeil, Corbeil, France (G Charpentier MD); Service de Diabétologie, Saint-Louis Hospital, Paris, France (Prof Ph Passa MD); and CNRS EP 10, Institut Pasteur de Lille, France (C Vaisse PhD, J Ruiz, Ph Froguel MD)

Correspondence to: Dr R W James, Division d’Epidémologie Diabétologie, 25 rue Micheli-du-Crest, 1211 Genève 14, Switzerland

et

The subjects were 434 unrelated, Caucasian, NIDDM patients who fulfilled the World Health Organisation criteria for diabetes mellitus.171 of these were classified as having CHD (CHD+ group). They were patients from Saint-Louis Hospital (Paris) and Corbeil Hospital (Corbeil), recruited between June, 1992, and September, 1994. 126 were in hospital for confirmed transmural myocardial infarction and the other 45 had CHD confirmed by coronary angiography (ie, they had stenoses with greater than 70% narrowing in the cross-sectional area of one of the major arteries or greater than 50% of the left main artery). The control group (CHD- group) was selected from a register of NIDDM families from all over France established at the Centre d’Etude du Polymorphisme Humain (CEPH) during

869

AA genotypes, n=207; AB+BB

genotypes, n=227. Findings given

(SD). Table 2: Lipid and apolipoprotein concentrations associated with paraoxonase genotypes

cycles with annealing at 61°C and a secondary amplification of 16 cycles with annealing at 63°C. PCR products were digested with Alwl, separated by non-denaturing acrylamide gel (10%) electrophoresis and visualised by use of ethidium bromide. Allele A (glutamine) corresponded to a 99 base-pair fragment and allele B (arginine) to 65 and 34 base-pair fragments. of

*For

waist-to-hip ratio, H, high; N, normal. tfindings given as mean (SD). Table 1: Clinical and biological characteristics of NIDDM patients with (CHD+) or without (CHD-) coronary heart disease 1993 and from the Corbeil Hospital (France) from 1993 June, 1994. Subjects in the CHD- group had no history of angina pectoris and had a normal resting electrocardiogram. An additional control group of non-diabetic, healthy subjects with a normal glucose tolerance test (n=125, mean age=52’2 [SD 12-8] yr, sex ratio M/F=49/76) was selected from the CEPH register. They were unrelated to the diabetic patients, were free

1990

as mean

to

30

Statistical analysis

to

Case-control differences in clinical and biological continuous variables were analysed by use of either one-way analysis of variance or the Mann-Whitney U test, depending on the shapes of the distribution curves. Categorical variables were compared between groups by use of X2 test and crude odds ratio. Allele frequencies were estimated by the gene-counting method and

of angina pectoris, and had a normal resting electrocardiogram. The study protocol was approved by the ethics committee of the Centre Hospitalo-Universitaire Lariboisiere-Saint-Louis (Paris). Written, informed consent was obtained from patients who agreed to participate in the study.

Hardy-Weinberg’s equilibrium was tested by the X2 test. For paraoxonase polymorphism, statistical analyses were based on the calculation of odds ratios to provide an estimate of the relative

Clinical data Clinical and biological data were extracted from the databases of the Saint-Louis Hospital, CEPH, and Corbeil Hospital. Arterial blood pressure was recorded with a mercury sphygmomanometer after 10 min at rest. Smokers were defined as subjects who regularly smoked more than one cigarette per day in the 3 months before examination, and ex-smokers as those having regularly smoked more than one cigarette per day but who had stopped at least 3 months before examination. All other subjects were classified as non-smokers. Patients were divided into two groups according to the waist-to-hip ratio-men, normal <0,95, high 0-95; women, normal <0-85, high =s0-85.

Laboratory measurements Venous blood samples were drawn after an overnight fast. Serum cholesterol, HDL-cholesterol (after precipitation of lower density lipoproteins with phosphotungstate/MgC12), and triglycerides were measured by automated enzymatic methods with reagents from Boehringer-Mannheim GmbH (Mannheim, Germany). Apo AI and apo B titres in whole plasma were measured by use of commercial antibodies (Behring, Germany). Paraoxonase enzyme

activity

risk of CHD associated with paraoxonase AB and BB genotypes. Test of linear trend was done with paraoxonase genotypes coded as ordinal variable (0, 1, 2). Multivariate analyses were done with a logistic regression model adjusted for all variables.

Results

expected, classic cardiovascular risk factors were overrepresented in the CHD+ group. CHD+ patients were older, more commonly male and smokers or exsmokers, and had higher values for waist-to-hip ratio, plasma triglycerides, systolic and diastolic blood pressure,

As

but lower values for HDL-cholesterol and apo A-1

(table 1). Table 2 shows that there were no significant differences between carriers of the B allele and AA homozygotes with respect to plasma cholesterol, triglyceride, or apo B concentrations, whereas HDL-cholesterol and apo A-I were higher in the B allele carriers. There were no significant differences in clinical characteristics between the groups (data not shown). When CHD+ and CHDgroups were compared, there were no significant differences in enzyme activities between B allele carriers, or between AA homozygotes. In the combined NIDDM groups, with paraoxon as substrate, the B allele carriers showed the expected higher activity in the presence of

Enzyme activity was measured with paraoxonlo or phenylacetate16 as substrate. For phenylacetate, serum (2 jjbL) was added to phenylacetate (4-0 mmol/L) in "tris"-acetate buffer (2-5 mL, 50 mmol/L, pH 7-8, containing CaCl2, 20 mmol/L). The reaction was

monitored

at

270

nm

(Kontron

Uvikon

spectrophotometer) and results expressed as AOD min-1 (arbitrary units AU). Paraoxonase serum protein levels measured by a competitive ELISA assay."

810 mL-l were

I?

DNA

analysis

Paraoxonase was genotyped by restriction isotyping; the 192 nucleotide substitution corresponding to position were an Alwl restriction site. creates Lymphocytes (GIn—Arg) isolated from blood and DNA extracted using standard procedures." A 99 base-pair fragment covering the region containing the mutation was amplified by PCR, with primers described by Humbert et al."It involved a primary amplification

870

Figure: Genotype frequencies (%) in NIDDM patients with (CHD+) and without (CHD-) coronary heart disease Trend test,

p=O.OO3.

necessary

to

confirm

the

role

of the

paraoxonase

polymorphism. Moreover, it is possible that early from CHD in NIDDM patients could lead to an underestimation of the cardiovascular consequences of this gene in our study. Second, our study refers to the association between the polymorphism in the paraoxonase gene and CHD specifically in French Caucasian NIDDM patients. The finding should be confirmed by studies including other populations, and the epidemiological relevance of this polymorphism should be investigated in the general population. How paraoxonase polymorphism could influence the susceptibility to CHD is speculative at present. A comparison between B allele carriers and non-carriers2O

mortality

*Y=yes; N=no. Table 3: Logistic regression coronary heart disease

analysis of determinants of

significant association between paraoxonase polymorphism and blood lipids, where the B allele carriers showed a less favourable lipid profile. Our study did not show such an association. In fact, in our study, the B allele carriers had higher concentrations of the cardioprotective factors HDL-cholesterol and apo A1. Differences between the two studies might explain the apparent disparities; these differences include the type of population (NIDDM patients in ours and a healthy, genetically homogenous population in the other) and the number of subjects recruited. Moreover, the other report concluded that paraoxonase polymorphism accounted for only a 1 % variation in total cholesterol and related lipoprotein traits; this percentage would seem too low to explain the association with CHD that we observed. An alternative explanation is that paraoxonase polymorphism may influence the putative role of the enzyme in the hydrolysis of oxidised fatty acyl groups.12,21 Such an explanation would offer a biologically plausible mechanism for the observed association with CHD, since

has 1 mmol/L salt (B allele carriers vs AA homozygotes, 0-79 [0-06] vs 0-29 [0-01] AU/mL; p<0-0001). By contrast no differences in activity were observed with phenylacetate as substrate (B allele carriers vs AA homozygotes, 11-6 vs 11-7 [0-26] AU/mL). However, serum concentrations of paraoxonase were significantly higher in the B allele carriers (B allele carriers vs AA homozyotes, 87-9 [2°2] vs 78-8 [2-4] )JLg/mL; p<0-0001), as previously observed in a study of non-diabetic subjects. 17 The genotype frequencies (figure 1) in the CHD+ and CHD- groups were in Hardy-Weinberg equilibrium. Genotype analyses revealed an association of the B allele and CHD in NIDDM patients. A stronger association was observed in homozygotes (OR 2-5 [95% CI 1-2-5-3]) than in heterozyotes (OR 1-6 [95% CI 1-1-2-4]), with a

significant trend test (p=0-003), suggesting a co-dominant effect of the B allele on cardiovascular risk. Thus, the prevalence of the Gln-Argl92 polymorphism of the paraoxonase gene (B allele) was significantly higher in the CHD+ group than in the CHD- group (0-35 vs 0-26) associated with vascular disease in univariate analyses (odds ratio 1-5 [95% CI 1-1-2-0], p
and

was

0-03, not significant). Discussion The findings indicate that the presence of the

Gln-Argl92

is of gene polymorphism paraoxonase in at least this French pathophysiological significance, Caucasian NIDDM population. The association between the presence of the polymorphism in the heterozygous or homozygous states (corresponding to the AB and BB genotypes, respectively) and coronary heart disease is highly significant and independent. Moreover, the trend test suggests a gene-dose response of the Gln-Argl92 polymorphism, since the BB homozygotes exhibit a in

the

stronger association with vascular disease than do heterozygotes. On the assumption that the odds ratio is a good approximation of the relative risk, 30% of cardiovascular cases could be attributed to the AB+BB genotypes in this population.’9 Several points should be borne in mind when interpreting these results. First, the study was case-control in design, so cases were not recruited prospectively. A survival bias cannot be avoided in a disease-association study and prospective and family studies will therefore be

indicated

a

several of increased there have been reports concentrations of oxidised serum lipids in diabetes.22,23 These higher concentrations may reflect a tendency to oxidative stress24,25 which, together with a modified capacity to eliminate lipid peroxides, could place diabetic patients at risk of CHD, especially when compared with a non-diabetic population. Unfortunately, conventional studies of paraoxonase enzyme activities cannot shed light on this possibility since they employ exogenous substrates whose relevance to the natural substrate is unknown. Some clinical studies have shown that certain subgroups at high risk of CHD (insulin-dependent diabetic and familial patients heterozygous hypercholesterolaemic patients) have low paraoxonase enzyme activities,26 but they did not show whether this finding was due to lower concentrations of paraoxonase protein or modified enzyme activity. Neither did the studies provide any evidence of a link with disease. Our study shows an independent association of the enzyme polymorphism with cardiovascular disease. This association is independent of the classic cardiovascular risk factors and in particular HDL. The proposed involvement of paraoxonase in the hydrolysis of lipid peroxides suggests a pathophysiological mechanism whereby the polymorphism may modulate the protective influence of the enzyme against oxidative stress. NIDDM patients seem particularly exposed to such conditions, given that diabetes predisposes to oxidative modifications of lipoproteins. We thank the Assistance Publique-Hopitaux de Paris, Bayer, France, Lilly France, Ministere Francais de l’Education et de la Recherche, and the

871

Swiss National Research Foundation (to RWJ, no 32-40292.94 and to JR, no 3.07.00) for grants. Prof Philippe and Dr M Mackness for helpful comments; and Ms Nathalie Dechamps and Ms Severine Clauin for technical assistance. RWJ and MCBG are members of the Geneva Diabetes Group.

References Stamler J, Vaccaro O, Neaton JD, Wentworth D, for the Multiple Risk Factor Intervention Trial Research Group. Diabetes, other risk factors, and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 1993; 16: 434-44. 2 Assmann G, Schulte H. The Prospective Cardiovascular Munster (PROCAM) Study: prevalence of hyperlipidemia in persons with hypertension and/or diabetes mellitus and the relationship to coronary heart disease. Am Heart J 1988; 116: 1713-24. 3 Tas S, Abdella NA. Blood pressure, coronary artery disease, and glycaemic control in type 2 diabetes mellitus: relation to apolipoprotein-CIII gene polymorphism. Lancet 1994; 343: 1194-95. 4 Marre M, Bernadet P, Gallois Y, et al. Relationships between 1

angiotensin 1 converting enzyme gene polymorphism, plasma levels, and diabetic retinal and renal complications. Diabetes 1994; 43: 5

384-88. Ruiz J, Blanché H, Cohen N,

et al. Insertion/deletion polymorphism of angiotensin converting enzyme gene is strongly associated with coronary heart disease in non-insulin-dependent diabetes mellitus. Proc

Natl Acad Sci USA 1994; 91: 3662-65. Steinberg D, Witzum JL. Lipoproteins and atherogenesis. JAMA 1990; 264: 3047-52. 7 Chisolm GM, Irwin KC, Penn MS. Lipoprotein oxidation and lipoprotein-induced cell injury in diabetes. Diabetes 1992; 41 (suppl 2): 61-66. 8 Hunt JV, Smith CCT, Wolff SP. Auto-oxidative glycosylation and possible involvement of peroxides and free radicals in LDL modification by glucose. Diabetes 1990; 39: 1420-24. 9 Bucula R, Makita Z, Koschinsky T, Cerami A, Vlassara H. Lipid advanced glycosylation: pathway for lipid oxidation in vitro. Proc Natl Acad Sci USA 1993; 90: 6434-38. 10 Blatter M-C, James RW, Messmer S, Barja F, Pometta D. Identification of a distinct high-density lipoprotein subspecies defined by a lipoprotein-associated protein, K-45. Identity of K-45 with paraoxonase. Eur J Biochem 1993; 211: 871-79. 11 Mackness MI, Arrol S, Durrington PN. Paraoxonase prevents

accumulation

in

low-density lipoprotein. FEBS Lett

12 Watson AD, Navah M, Hough GP, et al. Biologically active phospholipids in MM-LDL are transferred to HDL and are hydrolysed by HDL-associated esterases. Circulation 1994; 90: 1-250. 13 Parthasarathy S, Barnett J, Fong L. High density lipoprotein inhibits the oxidative modification of low density lipoprotein. Biochim Biophys Acta 1990; 1044: 275-83. 14 Humbert R, Adler DA, Disteche CM, Hassett C, Omiecinski CJ, Furlong CE. The molecular basis of the human serum paraoxonase activity polymorphism. Nat Genet 1993; 3: 73-76.

Organization: Diabetes Mellitus: Report of the WHO Study Group, World Health Org. WHO Tech Rep Ser 1985 no 727. 16 Lorentz K, Flatter B, Augustin E. Arylesterase in serum: elaboration and clinical application of a fixed-incubation method. Clin Chem 1979;

15 World Health

25: 1714-20. 17 Blatter-Garin MC, Abbott C, Messmer S, et al. Quantification of human serum paraoxonase by enzyme-linked immunoassay: population differences in protein concentrations. Biochem J 1994; 304: 549-54. 18 Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. New York: Cold Spring Harbor Laboratory Press, 1989. 19 Walter SD. Calculation of attributable risks from epidemiological data. Int J Epidemiol 1978; 7: 175-82. 20 Hegele RA, Brunt JH, Connelly PW. A polymorphism of the paraoxonase gene associated with variation in plasma lipoproteins in a genetic isolate. Arterioscler Thromb Vasc Biol 1995; 15: 89-95. 21 Mackness MI, Arrol S, Abbott C, Durrington PN. Protection of low-

6

872

of lipoperoxides

1991; 286: 152-54.

22

density lipoprotein against oxidative modification by high-density lipoprotein associated paraoxonase. Atherosclerosis 1993; 104: 129-35. Nishigaki I, Hagihara M, Tsunekawa HT, Maseki M, Yagi K. Lipid peroxide levels of serum lipoprotein fractions of diabetic patients. Bioch Med 1981; 25: 373-78.

23 Bellomo G, Maggi E, Poli M, Agosta FG, Bollati P, Finardi G. Autoantibodies against oxidatively modified low-density lipoproteins in NIDDM. Diabetes 1995; 44: 60-66.

Oxidised low density lipoproteins: a role in the pathogenesis of atherosclerosis in diabetes? Diabetic Med 1991; 8: 411-19. 25 Haffner SM, Agil A, Mykkanan L, Stern MP, Jialal I. Plasma oxidizability in subjects with normal glucose tolerance, impaired glucose tolerance and NIDDM. Diabetes Care 1995; 18: 646-53.

24

Lyons T.

26 Mackness MI, Harty D, Bhatnagar D, et al. Serum paraoxonase activity in familial hypercholesterolaemia and insulin-dependent diabetes mellitus. Atherosclerosis 1991; 86: 193-99.