- Email: [email protected]
Combined effects of apoE-CI–CII cluster and LDL-R gene polymorphisms on chromosome 19 and coronary artery disease risk
ARTICLE IN PRESS
Int. J. Hyg. Environ.-Health 209 (2006) 265–273 www.elsevier.de/ijheh
Combined effects of apoE-CI–CII cluster and LDL-R gene polymorphisms on chromosome 19 and coronary artery disease risk Chunhong Wanga,, Xin Zhoub, Shuiqing Yec, Dingfen Hana, Xiaodong Tana, Fang Zhengb, Qun Shia a
Department of Hygiene, School of Medicine, Wuhan University, DongHu Road 115, Wuhan 430071, P.R. China Genetic Diagnosis Center, Zhongnan Hospital, Wuhan University, Wuhan, P.R.China c Lipid Research Atherosclerosis Division, Johns Hopkins University School of Medicine, Baltimore, MD, USA b
Received 9 March 2005; received in revised form 12 December 2005; accepted 25 December 2005
Abstract Objective: To investigate associations of gene polymorphisms of the apoE-CI–CII gene cluster and the LDL-R gene on coronary artery disease (CAD) and their interactions with alcohol drinking and smoking in the Chinese Han population. Methods: A questionnaire survey of the behaviors of smoking and drinking, dietary patterns and anamnesis was conducted among 203 patients of CAD, aged 65.0711.1 years, and 365 controls, aged 63.6712.0 years. Peripheral blood samples were colleted and the total DNA was extracted. The apoE genotypes were identified by multiplex amplification refractory mutation system (multi-AMRS), the apoCI promoter polymorphisms and AvaII polymorphisms of the apoCII and LDL-R gene were detected by using PCR-RFLP. Pairwise linkage disequilibrium coefficients (D, D0 ) were estimated by the LINKAGE program. The interactions between genes with alcohol drinking and smoking were analyzed by using multivariate logistic regression models. Results: The differences of systolic/diastolic blood pressure, total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) concentrations, smoking and drinking were significant between subjects with CAD and controls. The frequencies of apoE gene e 3/4 genotype (25.9%) and e 4 (13.9%) in CAD were significantly higher than those in controls (12.5% and 6.9%, respectively, po0:05). A significant difference was also found for the apoCI locus, the frequencies of H2 allele were 20.5% in the CAD and 11.3% in the control. Linkage disequilibrium coefficient D0 was 0.672 (po0:01) between apoE and apoCI genes. Significant differences for a deficit of e 3-H1-T1 and excess of e 4-H2T1 was found in CAD by estimation of the haplotype frequencies. After control for possible confounding factors, the multivariate logistic analysis showed that e 4, H2 allele, smoking and drinking were risk factors of CAD. A significant interaction among e 4, H2 and smoking was observed (OR 18.3, 95% CI: 2.35–150.81, po0:05), it was a multiplicative model. An additive model was shown among e 4, H2 and drinking (OR12.7, 95% CI: 2.8–58.6, po0:05). Conclusion: The results suggested that both apoE and apoCI on chromosome 19 were the susceptibility locus for CAD, their linkage disequilibrium should be responsible for the development of CAD. Drinking and smoking enhance the genetic predisposition to CAD. r 2006 Elsevier GmbH. All rights reserved. Keywords: Coronary artery disease; Gene polymorphism; Drinking and smoking Corresponding author. Tel./fax: +86 27 68758648.
E-mail address: [email protected] (C. Wang). 1438-4639/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijheh.2005.12.005
ARTICLE IN PRESS 266
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Introduction Coronary artery disease (CAD) is a multifactor disorder that is thought to result from an interaction between genetic background and environmental factors such as diet, smoking, and physical activity. In the human genome, between 20,930 to 27,160 genes are estimated to be expressed in the human cardiovascular system (Dempsey et al., 2001). To date, a large number of genes and gene products are suggested to contribute to the development and promotion of CAD. Some of them are involved in lipid transport and metabolism. Genetic epidemiological studies have suggested that certain genetic polymorphisms in apolipoprotein (apo)E (Day and Wilson, 2001; Hirashiki et al., 2003; Shimokata et al., 2004), apoCI (Klasen et al., 1987) and apoCII (Graaf et al., 2000) genes are associated with an increased prevalence of CAD. In human genome structure, the apoE gene lies in a cluster with the apoCI and apoCII at the long arm of chromosome 19 (19q13.2). The apo ECI–CII cluster gene spans approximately 48 kb (Smit et al., 1988). The human low-density lipoprotein receptor (LDLR) gene has also been mapped to chromosome 19p13.2 (Yamamoto et al., 1984; Lindgren et al., 1985). This gene is approximately 45 kb long and contains 18 exons separated by 17 introns. The Ava II polymorphisms in exon 13 and Nco I polymorphisms in exon 18 have been found to affect total cholesterol (TC) and LDL-C levels in the general population (Ahn et al., 1994). ApoE, C1 and C2 are protein constituents of chylomicrons, very low-density lipoproteins (VLDL), and some subtractions of high-density lipoproteins (HDL). The apoE serves as a ligand for low-density lipoprotein (LDL) and apoE receptors (Jong et al., 1999). The common apoE e 2/e 3/e 4 polymorphism explains part of the variation in plasma cholesterol levels (Viiri et al., 2005). Among environmental factors, drinking and smoking have been largely related to lipid metabolism. One report (Humphries et al., 2001) suggested that smoking increased the risk of CAD in men of all apoE genotypes, particularly in men carrying the apoE gene e 4 allele. In this study, we describe the genetic association between the genetic polymorphism in the apoE-CI-CII cluster and LDL-R gene locus with CAD in a Chinese Han population in Wuhan City, China. We also found that the interplay between alcohol drinking, smoking and genetic polymorphisms in this population is the effect modifier for risk factors of CAD.
Hospital of Wuhan University, China. Their mean age was 65.0711.1, ranging from 41 to 88 years old. Three hundred and sixty-five controls were matched with age and gender to CAD patients from the same hospital without clinical signs of CAD or other cardiovascular diseases, as confirmed by anamnesis, physical examination, and investigation by electrocardiogram. Individuals who suffered from severe systemic disease which potentially influence the cardiovascular risk factor profile were excluded, the mean age was 63.6712.0, ranging from 40 to 87 years old. All subjects belonged to the Han nation. All articipants gave their written informed consent according to the study protocol, which was approved by the University Ethics Committee. The preliminary data had been presented in a previous paper in Chinese (Wang et al., 2005).
Questionnaire A thorough, structured questionnaire was completed by all subjects. Three experienced nurses (one interview, two collections) were assigned to obtain total information by face to face interview in the hospital ward or subject’s home. Information mainly included gender, date of birth, nationality, type of education, drinking and smoking habits, diet pattern, health problems, family history etc. Alcohol consumption was carefully evaluated by a set of 15 questions and further categorized as one of 3 categories: frequent drinkers (those who had drunk at least 25 g alcohol daily on average for six months or more), occasional drinkers (those who had drunk less than 25 g alcohol a day), or non-drinkers. Smoking was also recorded as frequent smokers (those who had smoked at least one cigarette daily on average for a year or more), occasional smokers (those who had smoked less than one cigarette a day), and non-smokers.
Determination of lipid levels Participants were instructed to fast for at least 12 h before the morning examination. Venous blood samples were collected during the medical check-up into EDTA-containing glass tubes. Plasma was separated from blood cells by centrifugation (1600g; 10 min; 4 1C) and was used immediately for lipid measurements. Plasma total triglycerides (TG) and total cholesterol (TC) were measured by enzymatic methods, higher density cholesterol (HDL-C) and lower density cholesterol (LDL-C) were determined by homogeneous methods, using an automatic chemistry analyzer (Aeroset, USA). Coefficients of variation for TG, TC, HDL-C, LDL-C measurements were each less than 5%. Anthropometrical measurements were taken using standard techniques: weight with light clothing by digital scales and height without shoes by fixed stadiometer. Body mass index (BMI) was calculated as weight (kg)/height (m2).
Materials and methods DNA extraction and genotype assay Study population From August, 2002 to January, 2003, 203 subjects with CAD were recruited from Zhongnan Hospital and Rengmin
Genomic DNA was isolated from leukocytes using the NaI method (Loparev et al., 1990) and stored at 20 1C until genotype analysis.
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
ApoE genotypes were determined by multi-ARMS (Zheng et al., 2000; Wenham et al., 1991). The upstream internal primers were P1 (50 -ATG CCG ATG ACC TGC AGA ATT30 ) and P2 (50 -ATG CCG ATG ACC TGC AGA ATC-30 ), the upstream external primers were P3 (50 -CGC GGA CAT GGA GGA CGT TT-30 ) and P4 (50 -CGC GGA CAT GGA GGA CGT TC-30 ), the downstream primer was P5 (50 -GTT CAG TGA TTG TCG CTG GGC A-30 ). The amplification reactions (25 ml) were done in duplicate using two tubes: A and B, each tube contained 200 ng genomic DNA, 20 pmol P5, 1.5 mmol/l MgCl2, 200 mmol/l dNTPs, 10% (v/v) dimethyl sulfoxide, 1.0 units of Taq DNA polymerase. Beside, tube A contained 15 pmol P1 and 10 pmol P3, tube B contained 15 pmol P2 and 10 pmol P4. The amplification (GeneAmp PCR system 9600, Perkin–Elmer, Foster City, CA) was performed for 35 cycles at 95 1C for 45 s, at 62 1C for 45 s and 72 1C for 45 s. A 451 bp fragment of the apoE gene covering the codons for amino acids 112 and 158 was amplified by PCR using the primers P1/ P2 and P5, another 588 bp fragment of the apoE gene was amplified using the P3/P4 and P5. In addition, a 228 bp fragment of the LDL-R exon 13 was amplified to act as inner reference. The fragments were separated on 2% (w/v) agarose gels. The apoCI promoter polymorphism was determined by PCR amplification followed by digestion with Hpa I as described previously (Nillesen et al., 1990). Since the Hpa I (Promega CO) restriction site is produced by a CGTT insertion, a 226 bp PCR product was digested into fragments of 160 and 66 bp only in the presence of the insertion allele (H2). An Ava II polymorphism in apoCII gene was carried out as described previously (Geisel et al., 1996). The undigested PCR product had a length of 530 bp. In the case of the naturally occurring Ava II restriction site, a 361 bp and a 169 bp fragment were observed. The LDL-R polymorphism in exon 13 was identified as previously described (Ahn et al., 1994). Since the Ava II restriction site is produced by a T-C mutation in the third base of codon 632 (valine), a 228 bp PCR product was digested into fragments of 141 and 87 bp.
Statistical analyses All data were introduced into a database built with the software Foxpro6.0. The analyses were performed with the SAS V8.01 and SPSS 10.0 software package. Differences in the characteristics between cases and control individuals were analyzed by t-tests for continuous variables and w2 tests for categorical variables. Allele frequencies for individual loci were calculated by the direct gene-count. w2 or Fisher’s exact test was used to compare differences in genotype and allele frequencies between subjects with CAD and controls. Pairwise linkage disequilibrium coefficients (D,D0 ) were estimated by the LINKAGE program. Estimation of haplotype frequencies was performed by the EH program provided by Terwilliger and Ott (Terwilliger and Ott, 1994; Xie and Ott, 1993). The associations of genotypes with alcohol drinking and smoking on CAD were assessed by a multivariate logistic regression model with dummy variables (for categorical and interaction terms). Results were presented as odds ratios (OR) and 95%
267
confidence interval (95% CI). To evaluate for gene-environment interaction (Khoury and Flanders, 1999), if ORge ¼ ORg+ORe1, we designated it as a additive model; if ORgeXORg ORg, it was a multiplicative model. ORge means joint OR for the exposure and the genotype, ORg means odds ratios for the effects of gene alone, ORe for the effects of exposure alone. p represents probability, p less than 0.05 is significant.
Results Demographic, anthropometric and plasma lipid data of the study subjects are summarized in Table 1. Besides, in subjects with CAD, all variables, except TC concentration, smoking and drinking, didn’t differ significantly between men and women. In controls, the differences of diastolic blood pressure, TC and LDL-C concentrations, smoking and drinking, were significant between men and women (Wang et al., 2005).
Genotype and allele frequencies Two hundred and one subjects with CAD and 360 controls (201/360) were genotyped for apoE gene polymorphism (allele designated e 2, e 3 and e 4), 190/ 357 for the promoter polymorphism in the apoCI gene (allele H1 and H2), 192/355 for the Ava II polymorphism in the intron 3 of apoCII gene (allele T1 and T2), and 202/358 for the Ava II polymorphism in the exon 13 of LDL-R gene (allele C1 and C2). Their electrophoresis fragments are shown in Figs. 1–4. Table 2 shows genotypes and allele frequencies for the observed alleles. The observed distributions of apoE, CI and LDL-R gene fit the Hardy–Weinberg equilibrium (p40:05). For apoE locus, the frequency of both e 4 allele and e 3/4 genotype were significantly higher in CAD than those in controls. The significant difference was found in the apoCI locus, the subjects with CAD showed higher frequency both for the H2 allele and genotypes carrying this allele. For LDL-R and apoCII locus, no differences of genotype distribution were found between subjects with CAD and controls.
Linkage disequilibrium analysis Linkage disequilibrium between the apoE, apoCI and apoCII locus may exist because the apo E-CI-CII cluster gene spans approximately 48 kb. So genotype analysis among the apoE, apoCI and apoCII polymorphisms was performed. Linkage disequilibrium coefficient (D,D’) and the haplotype frequencies are shown in Table 3 and 4. The results show that only linkage disequilibrium between the apoE and apoCI was significant. Significant differences for a deficit of e
ARTICLE IN PRESS 268
Table 1.
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Demographic, biochemical and life-style characteristics of all study subjects Subject with CAD
Controls
p
Men/women (number) Age (years) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Family history with CAD (number (%))
121/82 65.0711.1 23.273.1 134.5720.3 81.5711.8 4.4271.00 1.6170.86 1.1670.35 2.7770.86 21 (10.3)
204/161 63.6712.0 22.772.8 122.9717.1 77.3711.3 4.2471.11 1.3570.80 1.1970.35 2.5670.81 22 (6.0)
0.391 0.168 0.063 o 0.001 o 0.001 0.055 0.001 0.302 0.005 0.062
Smoking (number (%)) Non-smokers Occasional smokers Frequent smokers
102 (50.2) 58 (28.8) 43 (21.1)
250 (68.5) 89 (24.4) 26 (7.1)
o 0.001
Drinking (number (%)) Non-drinkers Occasional drinkers Frequent drinkers
92 (45.3) 45 (22.2) 66 (32.5)
199 (54.5) 114 (31.2) 52 (14.3)
o 0.001
Fig. 1. ApoE genotypes by the method of multi-ARMS (2% agarose gels, EB stain) Genotypes: lane A1+B1: e 2/2, lane A2+B2: e 3/3, lane A3+B3: e 4/4; lane A4+B4: e 2/3, lane A5+B5: e 3/4, lane A6+B6: e 2/4; M: pUC19/Msp I marker.
3-H1-T1 and excess of e 4-H2-T1 was found in the CAD by estimation of the haplotype frequencies.
Effect of apoE and apoCI allele type, and their interaction with smoking and drinking on CAD After control for possible confounding factors, the results of multivariate logistic regression models of the association between apoE and apoCI allele and their interactions with smoking and drinking on CAD are summarized in Table 5. e 4, H2 allele, smoking and frequent drinking were risk factors of CAD. A highly
significant interaction between e 4 and H2 allele was observed, it was a multiplicative model: OR: ¼ 5.623; 95% CI: 2.8–11.1. Between smoking and apoE allele type, there was a gradient of CAD risk from nonsmokers to occasional smokers and frequent smokers. This effect was much greater in the e 4 group than in others combined. It was also a multiplicative model between e 4 with frequent smokers (OR ¼ 11.1; 95% CI: 3.1–40.1; po0:01). Moreover, it was more associable when the subjects carried the apoCI gene H2 allele (OR ¼ 18.8; 95% CI: 2.4–150.8; po0:01). The interaction between apoE allele and drinking was also significant. If the subjects with e 4 allele had a larger
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Fig. 2. The LDL-R polymorphism in exon 13 by PCR-RFLP (2% agarose gels, EB stain); lane 1: PCR produce, lane 2: C1/1 genotype; lane 3: C1/2 genotype, lane 4: C2/2 genotype; M: pUC19/Msp I marker.
Fig. 3. The apoCI promoter polymorphism by PCR-RFLP; lane 1: H1/2 genotype, lane 2: H2/2 genotype; lane 3: H1/1 genotype, lane 4: PCR produce; M:PGEM-7Zf (+) /HaeIII marker; (8% polyacrylamide gel, silver stain).
alcohol consumption, they would have higher risk suffering from CAD than others (OR ¼ 3.0; 95% CI: 1.2–7.3; po0:05). An additive model was shown among e 4*H2* frequent drinkers (both OR ¼ 12.7; 95% CI: 2.8–58.6; po0:01).
Discussion This study confirmed that the apoE gene e 3 allele represents the most common allelic form of the polymorphism as in other populations (Hallman et al.,
269
Fig. 4. The apoCII polymorphism in intron 3 by PCR-RFLP; lane 1: T2/2 genotype, lane 2: T1/1 genotype; lane 3: T1/2 genotype, lane 4: PCR produce; M: FX174DNA/BSUri (HaeIII) Marker 9 (8% polyacrylamide gel, silver stain).
1991) and could be considered as a ‘‘non-pathological’’ apoE allele, whereas the frequency of both e 4 allele and e 3/4 genotype were significantly higher in CAD than that in the control. The results of in vivo studies have also shown that e 4 carriers have low plasma concentrations of apoE, and apoE knockout mice are susceptible to atherosclerosis (Linski et al., 1994; Reddick et al., 1994). The most interesting result was the association between apoCI polymorphism and CAD. We found an overrepresentation of apoCI gene H2 allele in the subjects with CAD, moreover, the linkage disequilibrium between the apoE and apoCI gene was significant in subjects with CAD. Using haplotype analyses, significant differences for a deficit of e 3-H1 and excess of e 4-H2 were found in the CAD. No associations between apoCII or LDL-R polymorphisms with CAD were found in this study population. ApoCI is a component of VLDL, IDL, and HDL. Analysis of homozygous apoE-CI knockout mice has shown that RNA expression levels of apoE, CI, CII in liver are genetically influenced by each other (van Ree et al., 1995). ApoCI can displace apoE from triglyceriderich emulsions and lipoproteins, and thereby it interferes indirectly with lipoprotein clearance (Jong et al., 1999). Effect of the apoCI allele on TG level was mediated by serum apoCI level (Shachter et al., 2005). Xu et al. (1999) found that a significantly higher (1.5fold) increase in apoCI gene transcription was observed in a reporter gene assay. This polymorphim is due to the 4 bp insertion that creates the HpaI site at 317 relative to the start of apoCI gene transcription (the H2 allele). Furthermore, a moderately overexpressing human
ARTICLE IN PRESS 270
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Table 2. Genotype distribution and allele frequencies of the apoE-CI-CII cluster and LDL-R gene polymorphisms in subjects with CAD and controls Subjects with CAD n (%)
Controls n (%)
pc
2/2 2/3 3/3 2/4 3/4 4/4
0 28 (13.9) 118 (58.7) 2 (1.0) 52 (25.9) 1 (0.5)
3 (0.9) 46 (12.8) 263 (73.1) 1 (0.3) 45 (12.5) 2 (0.5)
0.5562 0.6989 o 0.001 0.2628 o 0.001 1.00
apoC1
H1/1 H1/2 H2/2
121 (63.7) 60 (31.6) 9 ( 4.8)
283 (79.3) 67 (18.7) 7 (1.7)
o 0.001 o 0.001 0.1061
apoCII
T1/T1 T1/T2 T2/T2
105 (54.7) 51(26.6) 36 (18.7)
197 (55.5) 105 (29.6) 53 (14.9)
0.8461 0.3066 0.8670
LDL-R
C1/C1 C1/C2 C2/C2
133 (65.8) 63 (31.2) 6 (3.0)
254 (70.9) 99 (27.7) 5 (1.4)
0.3034 0.3992 0.2823
apoE alleleb
e2 e3 e4
30 (7.5) 316 (78.6) 56 (13.9)
53 (7.4) 617 (85.7) 50 (6.9)
0.9503 0.024 o 0.001
apoCI allele
H1 H2
302 (79.5) 78 (20.5)
633 (88.7) 81(11.3)
o 0.001
T1 T2
261 (68.0) 123 (32.0)
499 (70.3) 211 (29.7)
0.7032
C1 C2
329 (81.4) 62 (16.7)
6071 (84.8) 109 (15.2)
0.1473
Locus
Genotype
apoEa
e e e e e e
apoCII allele LDL-R allele
a For genotype distribution–apoE: e 2/2+e 2/3 vs. e 3/3 vs. e 3/4+e 4/4, w2 ¼ 16.7718, p ¼ 0.0002; apoC1: w2 ¼ 16.1127, p ¼ 0.0003; H1/1 vs. H1/ 2+H2/2, w2 ¼ 15.6043, po0.0001; apoCII: w2 ¼ 1.5396, p ¼ 0.4654; LDL-R: w2 ¼ 2.6733, p ¼ 0.2627; C1/1 vs. C1/2+C2/2, w2 ¼ 1.5783, p ¼ 0.2090. b For allele frequencies–apoE: w2 ¼ 14.8879, p ¼ 0.0006; apoCI: w2 ¼ 16.8323, po0.0001; apoCII: w2 ¼ 0.6286, p ¼ 0.7032; LDL-R: w2 ¼ 2.0995, p ¼ 0.1473. c p means differences between subjects with CAD and controls at the same genotype or allele.
Table 3. The apoE-CI-CII linkage disequilibrium in subjects with CAD and controls
Table 4. The apoE -CI–CII haplotype distribution in subjects with CAD and controls
Variants
Haplotype
apoE–apoCI apoE–apoCII apoCI–apoCII
Subjects with CAD
Controls
D
D0
D
D0
0.115 0.033 0.031
0.672* 0.129 0.134
0.089 0.041 0.032
0.320 0.278 0.156
*po0.01.
apoCI transgenic mouse showed that inhibition of hepatic lipase may be an important mechanism of the decrease in lipoprotein clearance mediated by apoCI (Conde–Knape et al., 2002).
e e e e e e e e e e e e
2-H1-T1 2-H1-T2 2-H2-T1 2-H2-T2 3-H1-T1 3-H1-T2 3-H2-T1 3-H2-T2 4-H1-T1 4-H1-T2 4-H2-T1 4-H2-T2
*po0.05; **po0.01.
Subjects with CAD
Controls
0.027078 0.028783 0.014946 0.001773 0.439746* 0.238090 0.065550 0.038872 0.052036 0.009965 0.053869** 0.029291
0.038073 0.018794 0.014039 0.000522 0.544985 0.230900 0.068403 0.014284 0.036347 0.020901 0.005296 0.007456
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
271
Table 5. Effect of apoE and apoCI allele type, and their interaction with smoking, drinking on CAD: multivariate logistic regression analysisa B7SE
OR
Reference 0.19070.277 0.81470.328
0.827 2.257
0.480–1.423 1.188–4.288
0.492 0.013
apoCI alleleb H1 H2
Reference 0.79170.204
2.206
1.478–3.292
o 0.001
Alcohol drinking Non-drinkers Occasional drinkers Frequent drinkers
Reference 0.02870.277 1.31370.304
1.029 3.719
0.598–1.771 2.050–6.749
0.918 o 0.001
Smoking Non-smokers Occasional smokers Frequent smokers
Reference 0.73570.243 1.52170.310
2.085 4.575
1.296–3.356 2.493–8.397
0.002 o 0.001
Interaction terms e 2*H1 e 3*H2 e 4*H2 e 2* Non-drinkers e 3* Occasional drinkers e 4* Occasional drinkers e 3* Frequent drinkers e 4* Frequent drinkers e 2* Non-smokers e 3* Occasional smokers e 4* Occasional smokers e 3* Frequent smokers e 4* Frequent smokers e 2*H1* never drinker e 3*H2* occasional drinkers e 4*H2* occasional drinkers e 3*H2* frequent drinkers e 4*H2* frequent drinkers e 2*H1* non-smokers e 3*H2* occasional smokers e 4*H2* occasional smokers e 3*H2* frequent smokers e 4*H2* frequent smokers
Reference 0.23570.273 1.72770.348 Reference 0.27270.287 0.65870.460 0.92370.307 1.09870.453 Reference 0.09170.280 1.32770.435 0.96170.356 2.40470.657 Reference 0.40270.585 1.39270.716 1.06070.483 2.54270.780 Reference 0.04970.512 2.14270.657 2.31671.133 2.93571.061
1.265 5.623
0.742–2.159 2.844–11.116
0.388 o 0.001
0.766 1.930 2.518 2.998
0.434–1.338 0.784–4.753 1.378–4.599 1.233–7.288
1.095 3.769 2.615 11.063
10.632–1.897 1.607–8.838 1.302–5.250 3.053–40.090
0.746 0.002 0.007 o 0.001
0.669 4.023 2.886 12.706
0.212–2.108 0.989–16.360 1.119–7.440 2.757–58.551
0.492 0.052 0.028 0.001
1.050 8.514 10.135 18.831
0.385–2.866 2.351–30.835 1.100–93.346 2.351–150.805
0.924 0.001 0.041 0.006
apoE allele e2 e3 e4
95% CI
p
b
0.344 0.153 0.003 0.015
a
Models were additionally adjusted by age, gender, BMI, family history with CAD, systolic blood pressure, diastolic blood pressure and concentration of TG, TC LDL-C and HDL-C in blood. b For analytical purpose, genotypes e 2/2, and e 2/3 were grouped as e 2, genotypes e 3/4 and e 4/4 as e 4, and genotype e 3/3 as e 3. Subjects with the e 2/4 genotype were excluded from analysis; genotype H1/H1 was grouped as H1, genotypes H/2 and H2/2 as H2.
This study also showed a significant interaction between smoking and apoE allele type, and there was a gradient of CAD risk from non-smokers to occasional smokers and frequent smokers. This effect was much greater in the e 4 group than in others combined. Moreover, it was more associable when the subjects carried the apoCI gene H2 allele. This supported the concept that the effects of gene and environment interaction increase the susceptibility to CAD.
The mechanisms for e 4 * and smoking effecting CAD risk are not all clear. Some reports suggested smoking increases the rate of oxidation of lipoprotein particles (Fickl et al., 1996). The results of studies with recombinant apoE have shown that the protection against oxidation in vitro is e 24e 34e 4 (Smith et al., 1998), e 4 carriers who smoke might therefore have a higher propensity to LDL oxidation and subsequent risk of atherosclerosis than smokers with other apoE
ARTICLE IN PRESS 272
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
genotypes or e 4 carriers who do not smoke. This effect might be due to the fact that e 2 has two free sulphydryl groups, e 3 has one, and e 4 none. Since e 4 carriers have a preponderance of small dense LDL (Haffer et al., 1996), which are themselves more prone to oxidation, this might be the reason of the increased risk of coronary heart disease in e 4 smokers. From the results we knew that the significant interaction between apoE allele and drinking was also shown between e 4* frequent drinkers and e 4*H2* frequent drinkers, they were additive. The effect of alcohol intake on CAD has been the subject of extensive research. A meta-analysis showed that alcohol intake was causally related to lower risk of CAD because of changes in lipid and hemostatic factors (Rimm et al., 1999). The Framingham Offspring study (Corella et al., 2001) observed a significant interaction between alcohol intake and apoE allele type in men. We are going to increase the size of sample and study and analysis the relation between apoE gene and alcohol intake by gender. There are several limitations in our study: (1) The numbers of individuals in each group were relatively small. (2) Although we selected controls from individuals with no history of CAD who exhibited normal electrocadiograms, we could not exclude the possibility that some of these subjects were affected by CAD. (3) Given that the subjects with CAD in the present study were survivors of CAD, they are likely not representative of all CAD patients. In spite of these limitations, the results suggested that both apoE and apoCI on chromosome 19 were the susceptibility locus for CAD, their linkage disequilibrium should be responsible for the development of CAD. Drinking and smoking enhance the genetic predisposition to CAD.
Acknowledgements The authors are grateful for the postgraduates in our laboratory who diligently worked throughout the samples collection. The study was carried out with the financial support of Hubei Educational Administration (Proj no.: 2000B03012) and Outstanding Youth Fund (Proj no.: 2003) of the School of Medicine, Wuhan University, P.R. of China.
References Ahn, Y., Kamboh, M., Aston, C., 1994. Role of common genetic polymorphisms in the LDL receptor gene in affecting plasma cholesterol levels in the general population. Arterioscler. Thromb. 14, 663–670.
Conde-Knape, K., Bensadoun, A., Sobel, J., 2002. Overexpression of apoC-I in apoE-null mice: severe hypertriglyceridemia due to inhibition of hepatic lipase. J. Lipid Res 43, 2136–2145. Corella, D., Tucker, K., Lahoz, C., 2001. Alcohol drinking determines the effect of the APOE locus on LDLcholesterol concentrations in men: the Framingham Offspring study. Am. J. Clin. Nutr. 73, 736–745. Day, I., Wilson, D., 2001. Science, medicine and the future: Genetics and cardiovascular risk. Br. Med. J. 323, 1409–1412. Dempsey, A., Dzau, V., Liew, C., 2001. Cardiovascular genomics: estimating the total number of genes expressed in the human cardiovascular system. J. Mol. Cell. Cardiol. 33, 1879–1886. Fickl, H., Van Antwerpen, V., Richards, G., 1996. Increased levels of autoantibodies to cardiolipin and oxidised low density lipoprotein are inversely associated with plasma vitamin C status in cigarette smokers. Athersclerosis 124, 75–81. Geisel, J., Weihaar, C., Oette, K., 1996. An AvaII polymorphism in the human apolipoprotein C-II gene. Clin. Genet. 49, 163. Graaf, J., Hoffer, M., Stuyt, P., 2000. Familial chylomicronemia caused by a novel type of mutation in the apoE-CICIV-CII gene cluster encompassing both the apoCII gene and the first apoCIV gene mutation: apoCII-CIVNijmegen. Biochem. Biophys. Res. Comm. 273, 1084–1087. Haffer, S., Stern, M., Miettinen, H., 1996. Apolipoprptein E polymorphism and LDL size in a biethnic population: multilocus genetic determinants of LDL particle size in coronary artery disease families. Arterioscler Thromb. Vasc. Biol. 58, 585–594. Hallman, D., Boerwinkle, E., Saha, N., 1991. The apolipoprotein E polymorphisms: a comparison of allele frequencies and effects in nine populations. Am. J. Hum. Genet. 49, 338–349. Hirashiki, A., Yamada, Y., Murase, Y., 2003. Association of gene polymorphisms with coronary artery disease in low-or hight-risk subjects defined by conventional risk factors. J. Am. Coll. Cardiol. 42, 1429–1437. Humphries, S., Talmud, P., Hawe, E., 2001. Apolipoprotein E4 and coronary heart disease in middle-aged men who smoke: a prospective study. Lancet 358, 115–119. Jong, M., Marten, H., Louis, M., 1999. Role of ApoCs in metabolism functional differences between ApoC1, ApoC2, and ApoC3. Arterioscler. Thromb. Vasc. Biol. 19, 472–484. Khoury, M., Flanders, W., 1999. Nontraditional epidemiologic approaches in the analysis of gene-environment interaction: case-control studies with no controls. Am. J. Epidemiol. 144, 207–213. Klasen, E., Talmud, P., Havekes, L., 1987. A common restriction fragment length polymorphism of the human apolipoprotein E gene and its relationship to type III hyperlipidema. Hum. Genet. 75, 244–247. Lindgren, V., Luskey, K., Russell, D., 1985. Human genes involved in cholesterol metabolism: chromosomal mapping of the loci for the low density lipoprotein receptor and 3hydroxy-3-methylglutaryl-coenzyme A reductase with cDNA probes. Proc. Natl. Acad. Sci. USA 82, 8567–8571.
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Linski, W., Ord, V., Plump, A., 1994. ApoE-deficient mice are a model of lipoprotein oxidation in atherogenesis. Demonstration of oxidative –specific epitopes in lesions and high titers of autoantibodies to malondialdehyde-lysine in serum. Arterioscler. Thromb. 14, 605–616. Loparev, V., Cantas, M., Monken, C., 1990. An efficient and simple method of DNA extraction from whole blood and cell line to identify infectious agents. J Virol. Method. 34, 105–106. Nillesen, W., Smeets, H., van Oost, B., 1990. Human ApoCI HpaI restriction site polymorphism revealed by the polymerase chain reaction. Nucl. Acid. Res. 18, 3428. Reddick, R., Zhang, S., Maeda, N., 1994. Atherosclerosis in mice lacking apoE: evaluation of lesional development and progression. Arterioscler. Thromb. 14, 141–147. Rimm, E., Williams, P., Criqui, M., 1999. Moderate alcohol intake and lower risk of coronary heart disease: metaanalysis of effects on lipids and haemostatic factors. Br. Med. J. 319, 1532–1538. Shachter, N., Rabinowitz, D., Stohl, S., Conde-Knape, K., Cohn, J., Deckelbaum, R., Berglund, L., Shea, S., 2005. The common insertional polymorphism in the APOC1 promoter is associated with serum apolipoprotein C-I levels in Hispanic children. Atherosclerosis 179, 387–393. Shimokata, K., Yamada, Y., Kondo, T., Ichihara, S., Izawa, H., Nagata, K., Murohara, T., Ohno, M., Yokota, M., 2004. Association of gene polymorphisms with coronary artery disease in individuals with or without nonfamilial hypercholesterolemia. Atherosclerosis 172, 167–173. Smit, M., Ellien, K., Rune, R., 1988. Apolipoprotein gene cluster on chromosome 19. Hum. Genet. 78, 90–93. Smith, J., Miyata, M., Poulin, S., 1998. The relationship between apolipoprotein E and serum oxidation-related variables is apolipoprotein E phenotype dependent. Int. J. Clin. Lab. Res. 28, 116–121.
273
Terwilliger, J., Ott, J., 1994. Linkage disequilibrium and disease loci. In: Handbook for Human Genetic Linkage. Johns Hopkins University Press, Baltimore, pp. 199–210. van Ree, J., van den Broek, W., van der Zee, A., 1995. Inactivation of apoE and apoCI by two consecutive rounds of gene targeting: effects on mRNA expression levels of gene cluster members. Hum. Mol. Genet. 4, 1403–1409. Viiri, L., Loimaala, A., Nenonen, A., Islam, S., Vuori, I., Karhunen, P., Lehtimaki, T., 2005. The association of the apolipoprotein E gene promoter polymorphisms and haplotypes with serum lipid and lipoprotein concentrations. Atherosclerosis 179, 161–167. Wang, C., Zhou, X., Zheng, F., Hang, D., Shi, Q., Liu, F., 2005. The apolipoprotein E-CI-CII gene cluster polymorphisms and coronary artery disease. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 22, 164–168 (in Chinese). Wenham, P., Newton, C., Price, W., 1991. Analysis of apolipoprotein E genotypes by the amplification refractory mutation system. Clin. Chem. 37, 241–244. Xie, X., Ott, J., 1993. Testing linkage disequilibrium between a disease gene and marker loci. Am. J. Hum. Genet. 53, 1107. Xu, Y., Berglund, L., Ramakrishnan, R., 1999. A commom HpaI RFLP of apolipoprotein C-I increases gene transcription and exhibits an ethnically distinct pattern of linkage disequilibrium with the allete of apolipoprotein E. J. Lipid Res. 40, 50–58. Yamamoto, T., Davis, C., Brown, M., 1984. The human LDL receptor: a cystcine-rich protein with multiple Alu sequences in its mRNA. Cell 39, 27–28. Zheng, F., Zhou, X., Ye, G., 2000. Identify apolipoprotein genotypes in 85 subjects by the multiplex amplification refractory mutation system PCR. Basic. Med. Sci. Clin. 20, 88–90.
Int. J. Hyg. Environ.-Health 209 (2006) 265–273 www.elsevier.de/ijheh
Combined effects of apoE-CI–CII cluster and LDL-R gene polymorphisms on chromosome 19 and coronary artery disease risk Chunhong Wanga,, Xin Zhoub, Shuiqing Yec, Dingfen Hana, Xiaodong Tana, Fang Zhengb, Qun Shia a
Department of Hygiene, School of Medicine, Wuhan University, DongHu Road 115, Wuhan 430071, P.R. China Genetic Diagnosis Center, Zhongnan Hospital, Wuhan University, Wuhan, P.R.China c Lipid Research Atherosclerosis Division, Johns Hopkins University School of Medicine, Baltimore, MD, USA b
Received 9 March 2005; received in revised form 12 December 2005; accepted 25 December 2005
Abstract Objective: To investigate associations of gene polymorphisms of the apoE-CI–CII gene cluster and the LDL-R gene on coronary artery disease (CAD) and their interactions with alcohol drinking and smoking in the Chinese Han population. Methods: A questionnaire survey of the behaviors of smoking and drinking, dietary patterns and anamnesis was conducted among 203 patients of CAD, aged 65.0711.1 years, and 365 controls, aged 63.6712.0 years. Peripheral blood samples were colleted and the total DNA was extracted. The apoE genotypes were identified by multiplex amplification refractory mutation system (multi-AMRS), the apoCI promoter polymorphisms and AvaII polymorphisms of the apoCII and LDL-R gene were detected by using PCR-RFLP. Pairwise linkage disequilibrium coefficients (D, D0 ) were estimated by the LINKAGE program. The interactions between genes with alcohol drinking and smoking were analyzed by using multivariate logistic regression models. Results: The differences of systolic/diastolic blood pressure, total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) concentrations, smoking and drinking were significant between subjects with CAD and controls. The frequencies of apoE gene e 3/4 genotype (25.9%) and e 4 (13.9%) in CAD were significantly higher than those in controls (12.5% and 6.9%, respectively, po0:05). A significant difference was also found for the apoCI locus, the frequencies of H2 allele were 20.5% in the CAD and 11.3% in the control. Linkage disequilibrium coefficient D0 was 0.672 (po0:01) between apoE and apoCI genes. Significant differences for a deficit of e 3-H1-T1 and excess of e 4-H2T1 was found in CAD by estimation of the haplotype frequencies. After control for possible confounding factors, the multivariate logistic analysis showed that e 4, H2 allele, smoking and drinking were risk factors of CAD. A significant interaction among e 4, H2 and smoking was observed (OR 18.3, 95% CI: 2.35–150.81, po0:05), it was a multiplicative model. An additive model was shown among e 4, H2 and drinking (OR12.7, 95% CI: 2.8–58.6, po0:05). Conclusion: The results suggested that both apoE and apoCI on chromosome 19 were the susceptibility locus for CAD, their linkage disequilibrium should be responsible for the development of CAD. Drinking and smoking enhance the genetic predisposition to CAD. r 2006 Elsevier GmbH. All rights reserved. Keywords: Coronary artery disease; Gene polymorphism; Drinking and smoking Corresponding author. Tel./fax: +86 27 68758648.
E-mail address: [email protected] (C. Wang). 1438-4639/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijheh.2005.12.005
ARTICLE IN PRESS 266
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Introduction Coronary artery disease (CAD) is a multifactor disorder that is thought to result from an interaction between genetic background and environmental factors such as diet, smoking, and physical activity. In the human genome, between 20,930 to 27,160 genes are estimated to be expressed in the human cardiovascular system (Dempsey et al., 2001). To date, a large number of genes and gene products are suggested to contribute to the development and promotion of CAD. Some of them are involved in lipid transport and metabolism. Genetic epidemiological studies have suggested that certain genetic polymorphisms in apolipoprotein (apo)E (Day and Wilson, 2001; Hirashiki et al., 2003; Shimokata et al., 2004), apoCI (Klasen et al., 1987) and apoCII (Graaf et al., 2000) genes are associated with an increased prevalence of CAD. In human genome structure, the apoE gene lies in a cluster with the apoCI and apoCII at the long arm of chromosome 19 (19q13.2). The apo ECI–CII cluster gene spans approximately 48 kb (Smit et al., 1988). The human low-density lipoprotein receptor (LDLR) gene has also been mapped to chromosome 19p13.2 (Yamamoto et al., 1984; Lindgren et al., 1985). This gene is approximately 45 kb long and contains 18 exons separated by 17 introns. The Ava II polymorphisms in exon 13 and Nco I polymorphisms in exon 18 have been found to affect total cholesterol (TC) and LDL-C levels in the general population (Ahn et al., 1994). ApoE, C1 and C2 are protein constituents of chylomicrons, very low-density lipoproteins (VLDL), and some subtractions of high-density lipoproteins (HDL). The apoE serves as a ligand for low-density lipoprotein (LDL) and apoE receptors (Jong et al., 1999). The common apoE e 2/e 3/e 4 polymorphism explains part of the variation in plasma cholesterol levels (Viiri et al., 2005). Among environmental factors, drinking and smoking have been largely related to lipid metabolism. One report (Humphries et al., 2001) suggested that smoking increased the risk of CAD in men of all apoE genotypes, particularly in men carrying the apoE gene e 4 allele. In this study, we describe the genetic association between the genetic polymorphism in the apoE-CI-CII cluster and LDL-R gene locus with CAD in a Chinese Han population in Wuhan City, China. We also found that the interplay between alcohol drinking, smoking and genetic polymorphisms in this population is the effect modifier for risk factors of CAD.
Hospital of Wuhan University, China. Their mean age was 65.0711.1, ranging from 41 to 88 years old. Three hundred and sixty-five controls were matched with age and gender to CAD patients from the same hospital without clinical signs of CAD or other cardiovascular diseases, as confirmed by anamnesis, physical examination, and investigation by electrocardiogram. Individuals who suffered from severe systemic disease which potentially influence the cardiovascular risk factor profile were excluded, the mean age was 63.6712.0, ranging from 40 to 87 years old. All subjects belonged to the Han nation. All articipants gave their written informed consent according to the study protocol, which was approved by the University Ethics Committee. The preliminary data had been presented in a previous paper in Chinese (Wang et al., 2005).
Questionnaire A thorough, structured questionnaire was completed by all subjects. Three experienced nurses (one interview, two collections) were assigned to obtain total information by face to face interview in the hospital ward or subject’s home. Information mainly included gender, date of birth, nationality, type of education, drinking and smoking habits, diet pattern, health problems, family history etc. Alcohol consumption was carefully evaluated by a set of 15 questions and further categorized as one of 3 categories: frequent drinkers (those who had drunk at least 25 g alcohol daily on average for six months or more), occasional drinkers (those who had drunk less than 25 g alcohol a day), or non-drinkers. Smoking was also recorded as frequent smokers (those who had smoked at least one cigarette daily on average for a year or more), occasional smokers (those who had smoked less than one cigarette a day), and non-smokers.
Determination of lipid levels Participants were instructed to fast for at least 12 h before the morning examination. Venous blood samples were collected during the medical check-up into EDTA-containing glass tubes. Plasma was separated from blood cells by centrifugation (1600g; 10 min; 4 1C) and was used immediately for lipid measurements. Plasma total triglycerides (TG) and total cholesterol (TC) were measured by enzymatic methods, higher density cholesterol (HDL-C) and lower density cholesterol (LDL-C) were determined by homogeneous methods, using an automatic chemistry analyzer (Aeroset, USA). Coefficients of variation for TG, TC, HDL-C, LDL-C measurements were each less than 5%. Anthropometrical measurements were taken using standard techniques: weight with light clothing by digital scales and height without shoes by fixed stadiometer. Body mass index (BMI) was calculated as weight (kg)/height (m2).
Materials and methods DNA extraction and genotype assay Study population From August, 2002 to January, 2003, 203 subjects with CAD were recruited from Zhongnan Hospital and Rengmin
Genomic DNA was isolated from leukocytes using the NaI method (Loparev et al., 1990) and stored at 20 1C until genotype analysis.
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
ApoE genotypes were determined by multi-ARMS (Zheng et al., 2000; Wenham et al., 1991). The upstream internal primers were P1 (50 -ATG CCG ATG ACC TGC AGA ATT30 ) and P2 (50 -ATG CCG ATG ACC TGC AGA ATC-30 ), the upstream external primers were P3 (50 -CGC GGA CAT GGA GGA CGT TT-30 ) and P4 (50 -CGC GGA CAT GGA GGA CGT TC-30 ), the downstream primer was P5 (50 -GTT CAG TGA TTG TCG CTG GGC A-30 ). The amplification reactions (25 ml) were done in duplicate using two tubes: A and B, each tube contained 200 ng genomic DNA, 20 pmol P5, 1.5 mmol/l MgCl2, 200 mmol/l dNTPs, 10% (v/v) dimethyl sulfoxide, 1.0 units of Taq DNA polymerase. Beside, tube A contained 15 pmol P1 and 10 pmol P3, tube B contained 15 pmol P2 and 10 pmol P4. The amplification (GeneAmp PCR system 9600, Perkin–Elmer, Foster City, CA) was performed for 35 cycles at 95 1C for 45 s, at 62 1C for 45 s and 72 1C for 45 s. A 451 bp fragment of the apoE gene covering the codons for amino acids 112 and 158 was amplified by PCR using the primers P1/ P2 and P5, another 588 bp fragment of the apoE gene was amplified using the P3/P4 and P5. In addition, a 228 bp fragment of the LDL-R exon 13 was amplified to act as inner reference. The fragments were separated on 2% (w/v) agarose gels. The apoCI promoter polymorphism was determined by PCR amplification followed by digestion with Hpa I as described previously (Nillesen et al., 1990). Since the Hpa I (Promega CO) restriction site is produced by a CGTT insertion, a 226 bp PCR product was digested into fragments of 160 and 66 bp only in the presence of the insertion allele (H2). An Ava II polymorphism in apoCII gene was carried out as described previously (Geisel et al., 1996). The undigested PCR product had a length of 530 bp. In the case of the naturally occurring Ava II restriction site, a 361 bp and a 169 bp fragment were observed. The LDL-R polymorphism in exon 13 was identified as previously described (Ahn et al., 1994). Since the Ava II restriction site is produced by a T-C mutation in the third base of codon 632 (valine), a 228 bp PCR product was digested into fragments of 141 and 87 bp.
Statistical analyses All data were introduced into a database built with the software Foxpro6.0. The analyses were performed with the SAS V8.01 and SPSS 10.0 software package. Differences in the characteristics between cases and control individuals were analyzed by t-tests for continuous variables and w2 tests for categorical variables. Allele frequencies for individual loci were calculated by the direct gene-count. w2 or Fisher’s exact test was used to compare differences in genotype and allele frequencies between subjects with CAD and controls. Pairwise linkage disequilibrium coefficients (D,D0 ) were estimated by the LINKAGE program. Estimation of haplotype frequencies was performed by the EH program provided by Terwilliger and Ott (Terwilliger and Ott, 1994; Xie and Ott, 1993). The associations of genotypes with alcohol drinking and smoking on CAD were assessed by a multivariate logistic regression model with dummy variables (for categorical and interaction terms). Results were presented as odds ratios (OR) and 95%
267
confidence interval (95% CI). To evaluate for gene-environment interaction (Khoury and Flanders, 1999), if ORge ¼ ORg+ORe1, we designated it as a additive model; if ORgeXORg ORg, it was a multiplicative model. ORge means joint OR for the exposure and the genotype, ORg means odds ratios for the effects of gene alone, ORe for the effects of exposure alone. p represents probability, p less than 0.05 is significant.
Results Demographic, anthropometric and plasma lipid data of the study subjects are summarized in Table 1. Besides, in subjects with CAD, all variables, except TC concentration, smoking and drinking, didn’t differ significantly between men and women. In controls, the differences of diastolic blood pressure, TC and LDL-C concentrations, smoking and drinking, were significant between men and women (Wang et al., 2005).
Genotype and allele frequencies Two hundred and one subjects with CAD and 360 controls (201/360) were genotyped for apoE gene polymorphism (allele designated e 2, e 3 and e 4), 190/ 357 for the promoter polymorphism in the apoCI gene (allele H1 and H2), 192/355 for the Ava II polymorphism in the intron 3 of apoCII gene (allele T1 and T2), and 202/358 for the Ava II polymorphism in the exon 13 of LDL-R gene (allele C1 and C2). Their electrophoresis fragments are shown in Figs. 1–4. Table 2 shows genotypes and allele frequencies for the observed alleles. The observed distributions of apoE, CI and LDL-R gene fit the Hardy–Weinberg equilibrium (p40:05). For apoE locus, the frequency of both e 4 allele and e 3/4 genotype were significantly higher in CAD than those in controls. The significant difference was found in the apoCI locus, the subjects with CAD showed higher frequency both for the H2 allele and genotypes carrying this allele. For LDL-R and apoCII locus, no differences of genotype distribution were found between subjects with CAD and controls.
Linkage disequilibrium analysis Linkage disequilibrium between the apoE, apoCI and apoCII locus may exist because the apo E-CI-CII cluster gene spans approximately 48 kb. So genotype analysis among the apoE, apoCI and apoCII polymorphisms was performed. Linkage disequilibrium coefficient (D,D’) and the haplotype frequencies are shown in Table 3 and 4. The results show that only linkage disequilibrium between the apoE and apoCI was significant. Significant differences for a deficit of e
ARTICLE IN PRESS 268
Table 1.
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Demographic, biochemical and life-style characteristics of all study subjects Subject with CAD
Controls
p
Men/women (number) Age (years) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) TC (mmol/l) TG (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Family history with CAD (number (%))
121/82 65.0711.1 23.273.1 134.5720.3 81.5711.8 4.4271.00 1.6170.86 1.1670.35 2.7770.86 21 (10.3)
204/161 63.6712.0 22.772.8 122.9717.1 77.3711.3 4.2471.11 1.3570.80 1.1970.35 2.5670.81 22 (6.0)
0.391 0.168 0.063 o 0.001 o 0.001 0.055 0.001 0.302 0.005 0.062
Smoking (number (%)) Non-smokers Occasional smokers Frequent smokers
102 (50.2) 58 (28.8) 43 (21.1)
250 (68.5) 89 (24.4) 26 (7.1)
o 0.001
Drinking (number (%)) Non-drinkers Occasional drinkers Frequent drinkers
92 (45.3) 45 (22.2) 66 (32.5)
199 (54.5) 114 (31.2) 52 (14.3)
o 0.001
Fig. 1. ApoE genotypes by the method of multi-ARMS (2% agarose gels, EB stain) Genotypes: lane A1+B1: e 2/2, lane A2+B2: e 3/3, lane A3+B3: e 4/4; lane A4+B4: e 2/3, lane A5+B5: e 3/4, lane A6+B6: e 2/4; M: pUC19/Msp I marker.
3-H1-T1 and excess of e 4-H2-T1 was found in the CAD by estimation of the haplotype frequencies.
Effect of apoE and apoCI allele type, and their interaction with smoking and drinking on CAD After control for possible confounding factors, the results of multivariate logistic regression models of the association between apoE and apoCI allele and their interactions with smoking and drinking on CAD are summarized in Table 5. e 4, H2 allele, smoking and frequent drinking were risk factors of CAD. A highly
significant interaction between e 4 and H2 allele was observed, it was a multiplicative model: OR: ¼ 5.623; 95% CI: 2.8–11.1. Between smoking and apoE allele type, there was a gradient of CAD risk from nonsmokers to occasional smokers and frequent smokers. This effect was much greater in the e 4 group than in others combined. It was also a multiplicative model between e 4 with frequent smokers (OR ¼ 11.1; 95% CI: 3.1–40.1; po0:01). Moreover, it was more associable when the subjects carried the apoCI gene H2 allele (OR ¼ 18.8; 95% CI: 2.4–150.8; po0:01). The interaction between apoE allele and drinking was also significant. If the subjects with e 4 allele had a larger
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Fig. 2. The LDL-R polymorphism in exon 13 by PCR-RFLP (2% agarose gels, EB stain); lane 1: PCR produce, lane 2: C1/1 genotype; lane 3: C1/2 genotype, lane 4: C2/2 genotype; M: pUC19/Msp I marker.
Fig. 3. The apoCI promoter polymorphism by PCR-RFLP; lane 1: H1/2 genotype, lane 2: H2/2 genotype; lane 3: H1/1 genotype, lane 4: PCR produce; M:PGEM-7Zf (+) /HaeIII marker; (8% polyacrylamide gel, silver stain).
alcohol consumption, they would have higher risk suffering from CAD than others (OR ¼ 3.0; 95% CI: 1.2–7.3; po0:05). An additive model was shown among e 4*H2* frequent drinkers (both OR ¼ 12.7; 95% CI: 2.8–58.6; po0:01).
Discussion This study confirmed that the apoE gene e 3 allele represents the most common allelic form of the polymorphism as in other populations (Hallman et al.,
269
Fig. 4. The apoCII polymorphism in intron 3 by PCR-RFLP; lane 1: T2/2 genotype, lane 2: T1/1 genotype; lane 3: T1/2 genotype, lane 4: PCR produce; M: FX174DNA/BSUri (HaeIII) Marker 9 (8% polyacrylamide gel, silver stain).
1991) and could be considered as a ‘‘non-pathological’’ apoE allele, whereas the frequency of both e 4 allele and e 3/4 genotype were significantly higher in CAD than that in the control. The results of in vivo studies have also shown that e 4 carriers have low plasma concentrations of apoE, and apoE knockout mice are susceptible to atherosclerosis (Linski et al., 1994; Reddick et al., 1994). The most interesting result was the association between apoCI polymorphism and CAD. We found an overrepresentation of apoCI gene H2 allele in the subjects with CAD, moreover, the linkage disequilibrium between the apoE and apoCI gene was significant in subjects with CAD. Using haplotype analyses, significant differences for a deficit of e 3-H1 and excess of e 4-H2 were found in the CAD. No associations between apoCII or LDL-R polymorphisms with CAD were found in this study population. ApoCI is a component of VLDL, IDL, and HDL. Analysis of homozygous apoE-CI knockout mice has shown that RNA expression levels of apoE, CI, CII in liver are genetically influenced by each other (van Ree et al., 1995). ApoCI can displace apoE from triglyceriderich emulsions and lipoproteins, and thereby it interferes indirectly with lipoprotein clearance (Jong et al., 1999). Effect of the apoCI allele on TG level was mediated by serum apoCI level (Shachter et al., 2005). Xu et al. (1999) found that a significantly higher (1.5fold) increase in apoCI gene transcription was observed in a reporter gene assay. This polymorphim is due to the 4 bp insertion that creates the HpaI site at 317 relative to the start of apoCI gene transcription (the H2 allele). Furthermore, a moderately overexpressing human
ARTICLE IN PRESS 270
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Table 2. Genotype distribution and allele frequencies of the apoE-CI-CII cluster and LDL-R gene polymorphisms in subjects with CAD and controls Subjects with CAD n (%)
Controls n (%)
pc
2/2 2/3 3/3 2/4 3/4 4/4
0 28 (13.9) 118 (58.7) 2 (1.0) 52 (25.9) 1 (0.5)
3 (0.9) 46 (12.8) 263 (73.1) 1 (0.3) 45 (12.5) 2 (0.5)
0.5562 0.6989 o 0.001 0.2628 o 0.001 1.00
apoC1
H1/1 H1/2 H2/2
121 (63.7) 60 (31.6) 9 ( 4.8)
283 (79.3) 67 (18.7) 7 (1.7)
o 0.001 o 0.001 0.1061
apoCII
T1/T1 T1/T2 T2/T2
105 (54.7) 51(26.6) 36 (18.7)
197 (55.5) 105 (29.6) 53 (14.9)
0.8461 0.3066 0.8670
LDL-R
C1/C1 C1/C2 C2/C2
133 (65.8) 63 (31.2) 6 (3.0)
254 (70.9) 99 (27.7) 5 (1.4)
0.3034 0.3992 0.2823
apoE alleleb
e2 e3 e4
30 (7.5) 316 (78.6) 56 (13.9)
53 (7.4) 617 (85.7) 50 (6.9)
0.9503 0.024 o 0.001
apoCI allele
H1 H2
302 (79.5) 78 (20.5)
633 (88.7) 81(11.3)
o 0.001
T1 T2
261 (68.0) 123 (32.0)
499 (70.3) 211 (29.7)
0.7032
C1 C2
329 (81.4) 62 (16.7)
6071 (84.8) 109 (15.2)
0.1473
Locus
Genotype
apoEa
e e e e e e
apoCII allele LDL-R allele
a For genotype distribution–apoE: e 2/2+e 2/3 vs. e 3/3 vs. e 3/4+e 4/4, w2 ¼ 16.7718, p ¼ 0.0002; apoC1: w2 ¼ 16.1127, p ¼ 0.0003; H1/1 vs. H1/ 2+H2/2, w2 ¼ 15.6043, po0.0001; apoCII: w2 ¼ 1.5396, p ¼ 0.4654; LDL-R: w2 ¼ 2.6733, p ¼ 0.2627; C1/1 vs. C1/2+C2/2, w2 ¼ 1.5783, p ¼ 0.2090. b For allele frequencies–apoE: w2 ¼ 14.8879, p ¼ 0.0006; apoCI: w2 ¼ 16.8323, po0.0001; apoCII: w2 ¼ 0.6286, p ¼ 0.7032; LDL-R: w2 ¼ 2.0995, p ¼ 0.1473. c p means differences between subjects with CAD and controls at the same genotype or allele.
Table 3. The apoE-CI-CII linkage disequilibrium in subjects with CAD and controls
Table 4. The apoE -CI–CII haplotype distribution in subjects with CAD and controls
Variants
Haplotype
apoE–apoCI apoE–apoCII apoCI–apoCII
Subjects with CAD
Controls
D
D0
D
D0
0.115 0.033 0.031
0.672* 0.129 0.134
0.089 0.041 0.032
0.320 0.278 0.156
*po0.01.
apoCI transgenic mouse showed that inhibition of hepatic lipase may be an important mechanism of the decrease in lipoprotein clearance mediated by apoCI (Conde–Knape et al., 2002).
e e e e e e e e e e e e
2-H1-T1 2-H1-T2 2-H2-T1 2-H2-T2 3-H1-T1 3-H1-T2 3-H2-T1 3-H2-T2 4-H1-T1 4-H1-T2 4-H2-T1 4-H2-T2
*po0.05; **po0.01.
Subjects with CAD
Controls
0.027078 0.028783 0.014946 0.001773 0.439746* 0.238090 0.065550 0.038872 0.052036 0.009965 0.053869** 0.029291
0.038073 0.018794 0.014039 0.000522 0.544985 0.230900 0.068403 0.014284 0.036347 0.020901 0.005296 0.007456
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
271
Table 5. Effect of apoE and apoCI allele type, and their interaction with smoking, drinking on CAD: multivariate logistic regression analysisa B7SE
OR
Reference 0.19070.277 0.81470.328
0.827 2.257
0.480–1.423 1.188–4.288
0.492 0.013
apoCI alleleb H1 H2
Reference 0.79170.204
2.206
1.478–3.292
o 0.001
Alcohol drinking Non-drinkers Occasional drinkers Frequent drinkers
Reference 0.02870.277 1.31370.304
1.029 3.719
0.598–1.771 2.050–6.749
0.918 o 0.001
Smoking Non-smokers Occasional smokers Frequent smokers
Reference 0.73570.243 1.52170.310
2.085 4.575
1.296–3.356 2.493–8.397
0.002 o 0.001
Interaction terms e 2*H1 e 3*H2 e 4*H2 e 2* Non-drinkers e 3* Occasional drinkers e 4* Occasional drinkers e 3* Frequent drinkers e 4* Frequent drinkers e 2* Non-smokers e 3* Occasional smokers e 4* Occasional smokers e 3* Frequent smokers e 4* Frequent smokers e 2*H1* never drinker e 3*H2* occasional drinkers e 4*H2* occasional drinkers e 3*H2* frequent drinkers e 4*H2* frequent drinkers e 2*H1* non-smokers e 3*H2* occasional smokers e 4*H2* occasional smokers e 3*H2* frequent smokers e 4*H2* frequent smokers
Reference 0.23570.273 1.72770.348 Reference 0.27270.287 0.65870.460 0.92370.307 1.09870.453 Reference 0.09170.280 1.32770.435 0.96170.356 2.40470.657 Reference 0.40270.585 1.39270.716 1.06070.483 2.54270.780 Reference 0.04970.512 2.14270.657 2.31671.133 2.93571.061
1.265 5.623
0.742–2.159 2.844–11.116
0.388 o 0.001
0.766 1.930 2.518 2.998
0.434–1.338 0.784–4.753 1.378–4.599 1.233–7.288
1.095 3.769 2.615 11.063
10.632–1.897 1.607–8.838 1.302–5.250 3.053–40.090
0.746 0.002 0.007 o 0.001
0.669 4.023 2.886 12.706
0.212–2.108 0.989–16.360 1.119–7.440 2.757–58.551
0.492 0.052 0.028 0.001
1.050 8.514 10.135 18.831
0.385–2.866 2.351–30.835 1.100–93.346 2.351–150.805
0.924 0.001 0.041 0.006
apoE allele e2 e3 e4
95% CI
p
b
0.344 0.153 0.003 0.015
a
Models were additionally adjusted by age, gender, BMI, family history with CAD, systolic blood pressure, diastolic blood pressure and concentration of TG, TC LDL-C and HDL-C in blood. b For analytical purpose, genotypes e 2/2, and e 2/3 were grouped as e 2, genotypes e 3/4 and e 4/4 as e 4, and genotype e 3/3 as e 3. Subjects with the e 2/4 genotype were excluded from analysis; genotype H1/H1 was grouped as H1, genotypes H/2 and H2/2 as H2.
This study also showed a significant interaction between smoking and apoE allele type, and there was a gradient of CAD risk from non-smokers to occasional smokers and frequent smokers. This effect was much greater in the e 4 group than in others combined. Moreover, it was more associable when the subjects carried the apoCI gene H2 allele. This supported the concept that the effects of gene and environment interaction increase the susceptibility to CAD.
The mechanisms for e 4 * and smoking effecting CAD risk are not all clear. Some reports suggested smoking increases the rate of oxidation of lipoprotein particles (Fickl et al., 1996). The results of studies with recombinant apoE have shown that the protection against oxidation in vitro is e 24e 34e 4 (Smith et al., 1998), e 4 carriers who smoke might therefore have a higher propensity to LDL oxidation and subsequent risk of atherosclerosis than smokers with other apoE
ARTICLE IN PRESS 272
C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
genotypes or e 4 carriers who do not smoke. This effect might be due to the fact that e 2 has two free sulphydryl groups, e 3 has one, and e 4 none. Since e 4 carriers have a preponderance of small dense LDL (Haffer et al., 1996), which are themselves more prone to oxidation, this might be the reason of the increased risk of coronary heart disease in e 4 smokers. From the results we knew that the significant interaction between apoE allele and drinking was also shown between e 4* frequent drinkers and e 4*H2* frequent drinkers, they were additive. The effect of alcohol intake on CAD has been the subject of extensive research. A meta-analysis showed that alcohol intake was causally related to lower risk of CAD because of changes in lipid and hemostatic factors (Rimm et al., 1999). The Framingham Offspring study (Corella et al., 2001) observed a significant interaction between alcohol intake and apoE allele type in men. We are going to increase the size of sample and study and analysis the relation between apoE gene and alcohol intake by gender. There are several limitations in our study: (1) The numbers of individuals in each group were relatively small. (2) Although we selected controls from individuals with no history of CAD who exhibited normal electrocadiograms, we could not exclude the possibility that some of these subjects were affected by CAD. (3) Given that the subjects with CAD in the present study were survivors of CAD, they are likely not representative of all CAD patients. In spite of these limitations, the results suggested that both apoE and apoCI on chromosome 19 were the susceptibility locus for CAD, their linkage disequilibrium should be responsible for the development of CAD. Drinking and smoking enhance the genetic predisposition to CAD.
Acknowledgements The authors are grateful for the postgraduates in our laboratory who diligently worked throughout the samples collection. The study was carried out with the financial support of Hubei Educational Administration (Proj no.: 2000B03012) and Outstanding Youth Fund (Proj no.: 2003) of the School of Medicine, Wuhan University, P.R. of China.
References Ahn, Y., Kamboh, M., Aston, C., 1994. Role of common genetic polymorphisms in the LDL receptor gene in affecting plasma cholesterol levels in the general population. Arterioscler. Thromb. 14, 663–670.
Conde-Knape, K., Bensadoun, A., Sobel, J., 2002. Overexpression of apoC-I in apoE-null mice: severe hypertriglyceridemia due to inhibition of hepatic lipase. J. Lipid Res 43, 2136–2145. Corella, D., Tucker, K., Lahoz, C., 2001. Alcohol drinking determines the effect of the APOE locus on LDLcholesterol concentrations in men: the Framingham Offspring study. Am. J. Clin. Nutr. 73, 736–745. Day, I., Wilson, D., 2001. Science, medicine and the future: Genetics and cardiovascular risk. Br. Med. J. 323, 1409–1412. Dempsey, A., Dzau, V., Liew, C., 2001. Cardiovascular genomics: estimating the total number of genes expressed in the human cardiovascular system. J. Mol. Cell. Cardiol. 33, 1879–1886. Fickl, H., Van Antwerpen, V., Richards, G., 1996. Increased levels of autoantibodies to cardiolipin and oxidised low density lipoprotein are inversely associated with plasma vitamin C status in cigarette smokers. Athersclerosis 124, 75–81. Geisel, J., Weihaar, C., Oette, K., 1996. An AvaII polymorphism in the human apolipoprotein C-II gene. Clin. Genet. 49, 163. Graaf, J., Hoffer, M., Stuyt, P., 2000. Familial chylomicronemia caused by a novel type of mutation in the apoE-CICIV-CII gene cluster encompassing both the apoCII gene and the first apoCIV gene mutation: apoCII-CIVNijmegen. Biochem. Biophys. Res. Comm. 273, 1084–1087. Haffer, S., Stern, M., Miettinen, H., 1996. Apolipoprptein E polymorphism and LDL size in a biethnic population: multilocus genetic determinants of LDL particle size in coronary artery disease families. Arterioscler Thromb. Vasc. Biol. 58, 585–594. Hallman, D., Boerwinkle, E., Saha, N., 1991. The apolipoprotein E polymorphisms: a comparison of allele frequencies and effects in nine populations. Am. J. Hum. Genet. 49, 338–349. Hirashiki, A., Yamada, Y., Murase, Y., 2003. Association of gene polymorphisms with coronary artery disease in low-or hight-risk subjects defined by conventional risk factors. J. Am. Coll. Cardiol. 42, 1429–1437. Humphries, S., Talmud, P., Hawe, E., 2001. Apolipoprotein E4 and coronary heart disease in middle-aged men who smoke: a prospective study. Lancet 358, 115–119. Jong, M., Marten, H., Louis, M., 1999. Role of ApoCs in metabolism functional differences between ApoC1, ApoC2, and ApoC3. Arterioscler. Thromb. Vasc. Biol. 19, 472–484. Khoury, M., Flanders, W., 1999. Nontraditional epidemiologic approaches in the analysis of gene-environment interaction: case-control studies with no controls. Am. J. Epidemiol. 144, 207–213. Klasen, E., Talmud, P., Havekes, L., 1987. A common restriction fragment length polymorphism of the human apolipoprotein E gene and its relationship to type III hyperlipidema. Hum. Genet. 75, 244–247. Lindgren, V., Luskey, K., Russell, D., 1985. Human genes involved in cholesterol metabolism: chromosomal mapping of the loci for the low density lipoprotein receptor and 3hydroxy-3-methylglutaryl-coenzyme A reductase with cDNA probes. Proc. Natl. Acad. Sci. USA 82, 8567–8571.
ARTICLE IN PRESS C. Wang et al. / Int. J. Hyg. Environ.-Health 209 (2006) 265–273
Linski, W., Ord, V., Plump, A., 1994. ApoE-deficient mice are a model of lipoprotein oxidation in atherogenesis. Demonstration of oxidative –specific epitopes in lesions and high titers of autoantibodies to malondialdehyde-lysine in serum. Arterioscler. Thromb. 14, 605–616. Loparev, V., Cantas, M., Monken, C., 1990. An efficient and simple method of DNA extraction from whole blood and cell line to identify infectious agents. J Virol. Method. 34, 105–106. Nillesen, W., Smeets, H., van Oost, B., 1990. Human ApoCI HpaI restriction site polymorphism revealed by the polymerase chain reaction. Nucl. Acid. Res. 18, 3428. Reddick, R., Zhang, S., Maeda, N., 1994. Atherosclerosis in mice lacking apoE: evaluation of lesional development and progression. Arterioscler. Thromb. 14, 141–147. Rimm, E., Williams, P., Criqui, M., 1999. Moderate alcohol intake and lower risk of coronary heart disease: metaanalysis of effects on lipids and haemostatic factors. Br. Med. J. 319, 1532–1538. Shachter, N., Rabinowitz, D., Stohl, S., Conde-Knape, K., Cohn, J., Deckelbaum, R., Berglund, L., Shea, S., 2005. The common insertional polymorphism in the APOC1 promoter is associated with serum apolipoprotein C-I levels in Hispanic children. Atherosclerosis 179, 387–393. Shimokata, K., Yamada, Y., Kondo, T., Ichihara, S., Izawa, H., Nagata, K., Murohara, T., Ohno, M., Yokota, M., 2004. Association of gene polymorphisms with coronary artery disease in individuals with or without nonfamilial hypercholesterolemia. Atherosclerosis 172, 167–173. Smit, M., Ellien, K., Rune, R., 1988. Apolipoprotein gene cluster on chromosome 19. Hum. Genet. 78, 90–93. Smith, J., Miyata, M., Poulin, S., 1998. The relationship between apolipoprotein E and serum oxidation-related variables is apolipoprotein E phenotype dependent. Int. J. Clin. Lab. Res. 28, 116–121.
273
Terwilliger, J., Ott, J., 1994. Linkage disequilibrium and disease loci. In: Handbook for Human Genetic Linkage. Johns Hopkins University Press, Baltimore, pp. 199–210. van Ree, J., van den Broek, W., van der Zee, A., 1995. Inactivation of apoE and apoCI by two consecutive rounds of gene targeting: effects on mRNA expression levels of gene cluster members. Hum. Mol. Genet. 4, 1403–1409. Viiri, L., Loimaala, A., Nenonen, A., Islam, S., Vuori, I., Karhunen, P., Lehtimaki, T., 2005. The association of the apolipoprotein E gene promoter polymorphisms and haplotypes with serum lipid and lipoprotein concentrations. Atherosclerosis 179, 161–167. Wang, C., Zhou, X., Zheng, F., Hang, D., Shi, Q., Liu, F., 2005. The apolipoprotein E-CI-CII gene cluster polymorphisms and coronary artery disease. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 22, 164–168 (in Chinese). Wenham, P., Newton, C., Price, W., 1991. Analysis of apolipoprotein E genotypes by the amplification refractory mutation system. Clin. Chem. 37, 241–244. Xie, X., Ott, J., 1993. Testing linkage disequilibrium between a disease gene and marker loci. Am. J. Hum. Genet. 53, 1107. Xu, Y., Berglund, L., Ramakrishnan, R., 1999. A commom HpaI RFLP of apolipoprotein C-I increases gene transcription and exhibits an ethnically distinct pattern of linkage disequilibrium with the allete of apolipoprotein E. J. Lipid Res. 40, 50–58. Yamamoto, T., Davis, C., Brown, M., 1984. The human LDL receptor: a cystcine-rich protein with multiple Alu sequences in its mRNA. Cell 39, 27–28. Zheng, F., Zhou, X., Ye, G., 2000. Identify apolipoprotein genotypes in 85 subjects by the multiplex amplification refractory mutation system PCR. Basic. Med. Sci. Clin. 20, 88–90.