Haplotypes of the Caspase-1 Gene, Plasma Caspase-1 Levels, and Cardiovascular Risk
http://www.100md.com
Stefan Blankenberg, Tiphaine Godefroy, O
参见附件。
the Departments of Medicine II (S. Blankenberg, H.J.R., C.B., R.S., T.M.) and Clinical Chemistry and Laboratory Medicine (K.J.L.), Johannes Gutenberg-University Mainz, Germany
INSERM U525 (T.G., O.P., S. Barbaux, V.N., F.C., L.T.), Université Pierre et Marie Curie/Faculté de Médecine Pitié-Salpêtrière, Paris, France
Department of Epidemiology and Public Health (F.K., A.E.), Belfast, UK
Greater Glasgow Health Board (C.M.), Glasgow, UK.
Abstract
Caspase-1 processes the interleukin (IL)-1 and IL-18 inactive precursors to the biologically active cytokines that are known to have proatherogenic effects. The present study investigated the genetic variability of the CASP1 gene and plasma levels of caspase-1 in relation to cardiovascular risk. In Europeans, 3 tag SNPs captured 4 common haplotypes of the CASP1 gene. Among these, the Ain6 allele of the G+7/in6A polymorphism was less frequent in 246 cases with myocardial infarction and a parental history of disease than in 253 controls free of familial history of disease (0.13±0.02 versus 0.20±0.02; P=0.005). However, in a larger case/control study (n=1774), these effects are borderline restricted to the UK population. In a prospective cohort of 1168 patients with coronary artery disease followed up during a median period of 6.0 years, the Ain6 allele exhibited a borderline association with future cardiovascular death (hazard ratio [HR]: 0.64, 0.41 to 1.01; P=0.053) and was associated with lower serum IL-18 levels (P=0.014). Baseline caspase-1 levels in the top quartile of the distribution were predictive of cardiovascular deaths (HR=3.62, 1.81 to 7.27; P=0.0003 compared with the bottom quartile). Finally, in vitro assays of allelic imbalance showed that the CASP1 haplotype carrying the Ain6 allele was associated with a lower mRNA expression. These results indicate that caspase-1 levels are predictive of future cardiovascular death in patients with coronary artery disease. The role of CASP1 genetic variations in the susceptibility to myocardial infarction requires further investigation.
Key Words: caspase-1 inflammation prognosis coronary disease genetics
Introduction
Caspases are a family of intracellular cysteine proteases that include at least 14 members that have been identified in mammals to date.1 Caspases play an essential role in apoptosis, but a subfamily comprising caspase-1, -4, and -5 are primarily involved in the regulation of inflammatory processes.2 Caspase-1, the best characterized of inflammatory caspases, was originally identified on the basis of its proteolytic activity for cleaving the inactive interleukin (IL)-1 precursor into the mature cytokine.3,4 This role was further confirmed by the impaired production of mature IL-1 in caspase-1–deficient mice.5 Caspase-1 is also responsible for the processing of IL-18, a cytokine that shares several structural features with the IL-1 family. The production of mature IL-18 requires cleavage of its precursor pro–IL-18 at a site similar to the cleavage site of pro–IL-1.6,7
Increasing evidence has accumulated from animal, experimental, and epidemiological data that IL-1 and IL-18 are involved in both development of atherosclerosis and plaque vulnerability as well as myocardial dysfunction as a consequence of ischemia/reperfusion injury.8–13 Given the critical role of caspase-1 in the production of these 2 cytokines, this enzyme may play a key role in the proatherogenic effects mediated by IL-1 and IL-18. Several lines of evidence support such a role. Caspase-1–deficient mice have a reduced production of interferon-, a proatherogenic factor whose induction by T cells and macrophages is stimulated by IL-18.5,14 Caspase-1 has been colocalized with apoptotic macrophages in sites of plaque rupture.15,16 Finally, inhibition of caspase-1 has been shown to reduce myocardial ischemic dysfunction in models of ischemia/ reperfusion via an inhibition of IL-18 and IL-1.17,18
As previously shown, circulating IL-18 levels were strongly predictive of future cardiovascular events.11,12 Furthermore, polymorphisms in the IL18 gene might contribute to the predisposition to complications of coronary artery disease (CAD).19 Given the impact of caspase-1 in the production of IL-18, the present study aimed to investigate whether baseline caspase-1 levels and polymorphisms of the CASP1 gene might influence cardiovascular risk. We sequenced the CASP1 gene and genotyped all single nucleotide polymorphisms (SNPs) in the SIPLAC study, a case/control study of myocardial infarction (MI). Three tag SNPs were then studied in relation to future cardiovascular death in the prospective AtheroGene cohort of CAD patients, and the most interesting tag SNP was further genotyped in the ECTIM study, a large case/control study of MI. Finally, functional assays were performed to detect potential differences in allelic expression of the CASP1 gene.
Materials and Methods
Study Populations
AtheroGene
The AtheroGene study is based on a prospective cohort of CAD patients recruited at the occasion of a diagnostic coronary angiography (Figure 1).20 A priori inclusion criterion was the presence of a diameter stenosis of >30% in at least 1 major coronary artery. The initial cohort was recruited between November 1996 and June 2000 and followed up for a median period of 6.0 (maximum 7.6) years. Stable (SAP) and unstable (UAP) angina pectoris were coded by Braunwald classification. Baseline IL-18 measurements and DNA were available for 1178 patients, of whom 106 died from a cardiovascular cause. Information about the cause of death was obtained by hospital or general practitioner. Extension of recruitment until February 2004 resulted in the inclusion of an additional 1294 patients, of whom 26 died from a cardiovascular cause. DNA, but not IL-18, was available in this extension cohort.
Because plasma samples of the early recruitment period tended to have been exhausted, caspase-1 was measured in patients recruited from June 1999 (n=1894). They were followed up for a median period of 2.5 (maximum 5.0) years and 71 died from a cardiovascular cause. There was an overlap of 450 patients in whom both caspase-1 and IL-18 were determined.
Healthy control subjects (n=468) were recruited either from the offices of general practitioners in the course of a routine check-up or by newspaper announcement. Caspase-1 was measured in 60% of controls for whom plasma was available.
Study participants had German nationality, were inhabitants of the Rhein-Main area, and were of European origin. Each participant gave written informed consent.
ECTIM and SIPLAC
The ECTIM study is a case/control study of MI conducted in the United Kingdom and France.21,22 Patients (35% females in the UK, only males in France) aged 25 to 64 years (for males) and 25 to 69 years (for females) were recruited from the WHO MONICA Project registers23 in Belfast and Glasgow (UK) and Lille, Strasbourg, and Toulouse (France). Age- and sex-matched controls were recruited from the populations of the areas covered by the registers. DNA was available for 1143 cases and 1133 controls.
The SIPLAC study is a substudy of ECTIM including highly contrasted samples of cases with MI and a parental history of MI (n=246) and controls without CAD and without parental history of MI (n=253) from the United Kingdom. This study provides us with a tool for selecting, among all SNPs discovered by molecular screening, the subset of those that will be further genotyped in larger studies, after exclusion of rare (0.02) and redundant polymorphisms. It is also intended to rapidly detect MI-associated polymorphisms that will require further replication.
Laboratory Methods in the AtheroGene Study
Blood was drawn under standardized conditions before coronary angiography was performed, and samples were stored at –80°C. Plasma caspase-1 (R&D Systems) and serum IL-18 (MBL Co Ltd) were measured using the ELISA technique according to the instructions of the manufacturer. C-reactive protein (CRP) was determined by a highly sensitive (hs), latex particle-enhanced immunoassay (Roche Diagnostics, Mannheim, Germany). Serum lipid levels were determined immediately by routine methods.
SNP Discovery in the CASP1 Gene
We sequenced all exons, up to 100 bp, of exon/intron junctions and up to 1 kb of 3'- and 5'-flanking sequences. Sequencing was performed in 31 pools of 2 MI patients (62 individuals) using a 16-capillary ABI-PRISM 3100 system and the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems [ABI], Warrington, UK). Sequences were compared using the ABI-PRISM SeqScape software version 2.0.
Genotyping
SNPs detected by sequencing were genotyped in the SIPLAC study by allele-specific oligonucleotide hybridization. Two tag SNPs (G+7/in6A and G+1034C) were selected for genotyping in the AtheroGene cohort plus a third SNP located in intron 8 (T+132/in8C), which, after consultation of the HapMap database (http://www.hapmap.org), appeared necessary to capture all common haplotypes of the CASP1 gene. Finally, the G+7/in6A SNP was genotyped in the ECTIM study. Genotyping in AtheroGene and ECTIM was performed using the 5' nuclease assay with TaqMan probes (see the expanded Materials and Methods in the online data supplement, available at http://circres.ahajournals.org). Primer and probe sequences, as well as amplification conditions for genotyping, can be found at our website, GeneCanvas (http://www.genecanvas.org).
Statistical Analysis
Departure from Hardy–Weinberg equilibrium was tested by 2 with 1 df. Association of polymorphisms with case/control status was tested by 2 analysis and with prospective outcome by Cox proportional-hazards regression analysis. Different models were successively fitted, first adjusted for age and sex, then additionally adjusted for cardiovascular risk factors (body mass index, diabetes, hypertension, smoking, HDL-cholesterol, triglycerides, acute coronary syndrome, number of diseased vessels, history of MI, statin, and -blocker intake), and finally adjusted for IL-18 levels. In all prospective analyses, the primary end point was cardiovascular death, and patients who died from other causes were censored at the time of death. Association of baseline caspase-1 levels, in quartiles, with cardiovascular mortality was tested by Cox analysis adjusted in the same way as for analyses with genotype. Association of polymorphisms with IL-18 and caspase-1 levels was tested by the GLM procedure adjusted for age, sex, and case/control status, and homogeneity of association between cases and controls was tested by introducing an interaction term in the model. IL-18 and caspase-1 were log-transformed to remove positive skewness and geometric means (95% confidence interval [CI]) are given.
Linkage disequilibrium (LD) and haplotype analyses were performed using the THESIAS software,24 available at http://www.genecanvas.org. Haplotypic effects, expressed as mean quantitative effects or as odds ratios (ORs), were estimated by comparison to the most frequent haplotype. A global test of association between haplotypes and the phenotype was performed by a likelihood ratio test. The homogeneity of effects associated with haplotypes carrying the Ain6 allele was tested by introducing a constraint on the parameters (2 with 1 df). Except for haplotype analyses, all analyses were performed with the SAS software version 8.01 (SAS Institute Inc, Cary, NC). P<0.05 was considered to be significant.
Allelic Imbalance Assays
Differential allelic expression of the CASP1 gene was tested by the method of allelic imbalance, which compares the ratio of cDNAs to that of genomic DNAs in individuals heterozygous for a transcribed polymorphism.25 cDNA was synthesized from RNA extracted from Epstein–Barr virus–transformed lymphoblastoid cell lines (LCLs) (expanded Materials and Methods). cDNA and genomic DNA from 11 individuals heterozygous for the L256L (C/T) polymorphism were subjected to the same reaction of allelic discrimination using the TaqMan technology. Each sample was submitted to 2 different RT-PCR, and the resulting cDNAs were amplified in duplicates. The regression line equation from the genomic DNA samples was calculated, and, for each cDNA sample, the deviation from this theoretical line was assessed. For each individual, the log (ratio) of the observed to the expected value was averaged over the 4 repetitions. The mean of the 11 values was compared by t test to the expected value of 0 under the null hypothesis of no allelic imbalance.
Real-Time RT-PCR Assays
mRNA expression levels in LCLs were compared between individuals with different genotypes for the L256L (C/T) polymorphism by real-time PCR quantification (expanded Materials and Methods). Twenty-two homozygotes for the C allele, 9 heterozygotes, and 2 homozygotes for the T allele were analyzed in duplicate, and the mean of the 2 measurements was taken. Rare homozygotes were pooled with heterozygotes for final comparison. The mean mRNA levels between CC subjects and T carriers were compared by t test.
Results
Sequencing of the CASP1 Gene and Genotyping in the SIPLAC Study
Nineteen polymorphisms were detected by sequencing (Figure 2), which were all genotyped in the SIPLAC study, except 3 of them, which were very rare (A+109/in5G, C+92/in7T, and –43in7I/D, each observed in 1 of 124 chromosomes). There was a block of 14 SNPs in complete association covering the whole gene sequence. Only 2 SNPs were outside of this block, 1 nonsynonymous SNP (R240Q) that was present in only 2 subjects and not further studied and 1 SNP located in the 3' region (G+1034C). The G+7/in6A SNP was selected for representing the block in complete association.
In the SIPLAC study, the G+7/in6A and G+1034C SNPs were in almost complete LD (D'0.97). The Ain6 allele was represented less in MI cases than in controls (0.13±0.02 versus 0.20±0.03 in Belfast, 0.12±0.02 versus 0.19±0.03 in Glasgow; P=0.005 for the case/control difference adjusted for center). The G+1034C SNP was not associated with MI risk (supplemental Table I).
CASP1 Polymorphisms and CAD Risk in the AtheroGene Study
Baseline characteristics of cases and controls in the initial AtheroGene study are shown in Table 1. Cases had increased levels of CRP, triglycerides, and a higher ratio of LDL- to HDL-cholesterol. The lower levels of LDL in patients was likely attributable to the high prevalence of lipid-lowering treatment. IL-18 levels were comparable in controls and CAD patients as a whole. However, there was a trend toward increased IL-18 levels in acute MI by comparison to UAP and SAP (66.0 versus 60.9 and 59.7 pg/mL; P=0.11). In the subset of subjects having both IL-18 and caspase-1 measurements, caspase-1 levels were higher in cases than in controls (Table 1).
The G+7/in6A, T+132/in8C, and G+1034C SNPs were genotyped in the initial AtheroGene study. Because of strong LD, they generated only 4 common haplotypes, Gin6Tin8G+1034, Gin6Cin8G+1034, Ain6Tin8G+1034, and Ain6Tin8C+1034, with frequency in controls of 0.54, 0.28, 0.06, and 0.12, respectively.
The Ain6 allele frequency was similar in healthy German controls to that observed in controls from the United Kingdom (0.18±0.01 versus 0.19±0.02; P=0.45). There was no difference of the Ain6 allele frequency between CAD patients as a whole and controls (0.17±0.01 versus 0.18±0.01; P=0.75). However, the Ain6 allele frequency was significantly decreased in patients with acute MI (0.11±0.02; P=0.042), whereas it did not differ in patients with SAP (0.17±0.01; P=0.79) or UAP (0.19±0.02; P=0.60) from the frequency observed in controls (supplemental Table II). The T+132/in8C and G+1034C SNPs were not associated with clinical status (supplemental Table II). Haplotype frequencies did not differ between cases as a whole and controls (P=0.87).
CASP1 Polymorphisms and Circulating IL-18 Levels in the AtheroGene Study
The G+7/in6A SNP was associated with serum IL-18 levels, the Ain6 allele being associated with a lowering effect in a fairly additive fashion (Table 2). The T+132/in8C and G+1034C SNPs were not associated with IL-18 levels. Haplotype analysis was performed in cases and controls pooled because there was no significant heterogeneity between groups. By reference to the most frequent haplotype (Gin6Tin8G+1034), haplotype Gin6Cin8G+1034 had no significant effect on IL-18 levels (P=0.97), whereas haplotypes Ain6Tin8G+1034 and Ain6Tin8C+1034 had similar lowering effects (P=0.48 for the homogeneity of effects), suggesting that neither the T+132/in8C nor the G+1034C SNP had an individual effect and that only the G+7/in6A SNP was associated with IL-18 levels (Figure 3).
CASP1 Polymorphisms and Cardiovascular Mortality in the AtheroGene Study
In the initial AtheroGene cohort, there was a trend toward a lower risk of cardiovascular deaths during follow-up in carriers of the Ain6 allele than in Gin6Gin6 homozygotes, although the association did not reach significance (age- and sex-adjusted hazard ratio [HR]: 0.64, 0.41 to 1.01; P=0.053). Kaplan–Meier curves of survival according to genotype are shown in Figure 4. Adjustment for cardiovascular risk factors did not modify the relationship (HR=0.65, 0.41 to 1.03; P=0.07), nor did additional adjustment for IL-18 (HR=0.68, 0.43 to 1.08; P=0.10). No significant association was observed with the T+132/in8C or the G+1034C SNP (supplemental Table III).
To increase the power of the survival analysis, the G+7/in6A SNP was analyzed in the extension cohort. When combining both cohorts, the HR associated with carriage of the Ain6 allele, before and after adjustment for cardiovascular risk factors, was 0.66 (0.44 to 0.99; P=0.04) and 0.68 (0.45 to 1.03; P=0.07), respectively.
Plasma Caspase-1 Levels and Cardiovascular Mortality in the AtheroGene Study
When considering all subjects having a caspase-1 determination (Figure 1), higher levels were observed in CAD patients than in healthy controls, but this difference was mainly attributable to an increase in MI patients (age- and sex-adjusted mean levels: 59.3, 52.8, 51.9, and 51.4 pg/mL in MI, UAP, SAP, and control groups, respectively; P<0.0001). Caspase-1 correlated positively with CRP (r=0.19; P<0.001), but the association with IL-18 was rather weak (r=0.14; P<0.001) despite their biological link.
Figure 5 displays the Kaplan–Meier curves for survival from cardiovascular death according to caspase-1 quartiles. HRs estimated from different Cox models are given in supplemental Table IV. All models indicated an increased risk in patients in the top quartile of caspase-1 levels compared with those in the bottom quartile, whereas patients in the intermediate quartiles were not at higher risk. The age- and sex-adjusted HR (95% CI) in the top quartile was 3.62 (1.81 to 7.27; P=0.0003). Adjustment for classical risk factors hardly modified the association (HR=2.89, 1.42 to 5.88; P=0.003). Further adjustment for CRP levels slightly weakened the association (HR=2.32, 1.13 to 4.79; P=0.03).
CASP1 Polymorphisms and Plasma Caspase-1 Levels in the AtheroGene Study
There was a borderline association of the G+7/in6A and the G+1034C SNPS with plasma caspase-1 levels, the minor allele of each SNP being associated with a lowering effect (P=0.066 for the significance of each polymorphism, Table 2). By reference to the Gin6G+1034 haplotype, haplotypes Ain6G+1034 and Ain6C+1034 had similar lowering effects (P=0.65 for the homogeneity of effects), indicating that the G+1034C SNP had no effect by itself. The T+132/in8C SNP was genotyped only in a subset of subjects having a caspase-1 measurement and there was no significant association.
CASP1/G+7in6A Polymorphism and MI Risk in the ECTIM Study
Because of the suggestive evidence of an association of the G+7/in6A SNP with MI risk both in SIPLAC and AtheroGene, we further genotyped this SNP in the whole ECTIM study (Figure 1). In the whole population from the United Kingdom, encompassing SIPLAC, the association was weakened, the Ain6 allele frequency being 0.146±0.009 in cases and 0.170±0.010 in controls, respectively (P=0.07). By contrast, there was no case/control difference of allele frequency in France (0.175±0.014 versus 0.161±0.013; P=0.46) (test of homogeneity between countries; P=0.10). When stratifying according to parental history of MI, there was no significant interaction in either country (supplemental Table V).
Functional Assays
We further examined whether the 2 haplotypes defined by the block containing the G+7/in6A SNP were associated with a difference in gene expression levels. The exonic polymorphism taken for allelic imbalance assays was the coding SNP L256L (C/T), which belongs to the block (Figure 2). Expression of allelic transcripts in 11 heterozygotes are shown in Figure 6. The regression line equation predicted from the genomic DNA samples was associated with an R2 of 98%, compared with 100%, for a perfect equimolar ratio between the 2 alleles. Most values for the cDNA samples were situated above the regression line, suggesting a higher expression level associated with the C allele than with the T allele (mean difference of the observed to expected values : 0.31±0.12; P=0.028).
We confirmed these results by the quantification of mRNA levels by real-time RT-PCR. Higher expression levels were found in individuals homozygous for the C allele (1.91±0.34, n=22) than in carriers of the T allele (0.53±0.48, n=11) (P=0.02).
Discussion
Caspase-1 belongs to the subfamily of inflammatory caspases, also comprising caspase-4 and -5, that share similar biological functions in the regulation of inflammatory processes.2 In particular, all 3 caspases can cleave pro–IL-18 to generate the biologically active IL-18, although caspase-1 is the most efficient in this process.6,7 The genes coding for these evolutionary-related caspases are located in a cluster on chromosome 11q22.2 to 22.3 and present more than 50% of sequence homology.1 Besides the CASP4 and CASP5 genes located downstream, we identified several regions located upstream and presenting more than 70% of homology with the CASP1 gene sequence. These sequences are presumably those that were previously identified and considered to be pseudogenes, because none of these encoded a functional cysteine protease.26
The CASP1 gene exhibited a pattern of reduced haplotype diversity, with only 4 haplotypes captured by 3 tag SNPs. We investigated whether this block might extend beyond the CASP1 gene itself by examining the HapMap data. In Europeans, the block stretched over 32 kb but did not include the CASP4 and CASP5 genes, therefore excluding the possibility that polymorphisms in these genes might be responsible for the observed associations.
Our study presented a comprehensive investigation of the role of caspase-1 in susceptibility to CAD and its complications. However, it failed to provide definite conclusions. Our initial results provided several arguments supporting an etiologic role of caspase-1 : (1) the Ain6 allele of the CASP1 gene was associated with a lower risk of MI in 2 different case/control studies, SIPLAC, and AtheroGene; (2) the same allele exhibited a borderline association with prospective cardiovascular death in AtheroGene; (3) plasma caspase-1 was elevated in patients with acute MI and appeared as a biomarker predictive of cardiovascular mortality in CAD patients; (4) circulating IL-18 levels were decreased in individuals carrying the Ain6 allele; and (5) the CASP1 haplotype carrying the Ain6 allele was associated with a lower gene expression, supporting the existence of a functional polymorphism within the CASP1 gene.
Cases in SIPLAC were young survivors of MI with a parental history of disease recruited in the United Kingdom. They were compared with controls selected on the absence of familial and personal histories of MI. Moreover, populations of Northern Ireland and Scotland are among those having the highest prevalence of CAD in Europe.23 When stratified according population in the entire ECTIM study, the effect was observed in the United Kingdom and was absent in those recruited in France. It is then likely that the combination of these 2 criteria—family history and populations—has led to the selection of individuals genetically more susceptible than the other subjects included in the ECTIM study. However, it cannot be ruled out that, given the multiple hypotheses tested and the lack of replication, the initial associations were attributable to chance. Further studies are required to address this issue.
The genetic effects observed in SIPLAC and AtheroGene were modest and can be viewed only as suggestive of a role of caspase-1 in the etiology and clinical consequences of MI. By linking up the epidemiological and experimental results, it might be speculated that the haplotype carrying the Ain6 allele, which is associated in vitro with a lower CASP1 gene expression, results in a reduced rate of cleavage of pro–IL-18 into the biologically active IL-18, which in turn would confer a protection against the proinflammatory effects of IL-18. Given the complexity of the haplotype block, including at least 14 polymorphisms, it is difficult to predict which among them might be functional. Two of the SNPs are synonymous polymorphisms (T54T and L256L), which are unlikely to have a functional role. Six SNPs are located in exon/intron junctions and might cause alternative splicing that would result in the production of proteins of different sizes. However, we did not detect any difference in total cDNA size among individuals with different genotypes (data not shown). One SNP is located in the 5' region and might affect gene expression, whereas 5 SNPs are in the 3' region and might be involved in mRNA stability.
Circulating caspase-1 levels were independently related to future cardiovascular death in CAD patients. This result supports the hypothesis that activation of the caspase-1/IL-18 pathway is involved in the clinical complications of preexisting atherosclerotic disease. In animal models, inhibition of capsase-1 has been shown to reduce ischemia/reperfusion injury.17,18 Therefore, increased levels of caspase-1 might reflect individuals at high risk of post-MI complications, rather than early development of atherosclerosis itself. In the same vein, caspase-1 and IL-18 levels were increased in patients with acute MI but not UAP. The discrepancy might also corroborate the hypothesis that the caspase-1/IL-18 pathway is predominantly activated after MI has occurred.
As an important limitation with regard to the results of the genetic variations, we could not correct for multiple testing. Therefore, these results have to be considered as hypotheses generating and need replication in large population studies.
Paradoxically, circulating levels of caspase-1 weakly correlated with those of IL-18, and the CASP1 gene effects were stronger on IL-18 than on caspase-1 itself. This might be explained by the fact that the caspase-1 assay determined the mass and not the activity itself, whereas the IL-18 assay measured a compound of the active form and the IL-18/IL-18-binding protein complex.
In conclusion, the present results indicated that caspase-1 levels were predictive of future cardiovascular death in CAD patients. They also suggested an etiologic role of CASP1 genetic variations in the susceptibility to MI, although this requires further replication.
Acknowledgments
We thank Biosite Inc for determination of caspase-1 levels.
Sources of Funding.
The ECTIM study was supported by a grant of the British Heart Foundation (PG/93045). The AtheroGene study was supported by a grant from the "Stiftung Rheinland-Pfalz für Innovation," Ministry for Science and Education (AZ 15202-386261/545), Mainz, Germany; by the MAIFOR grant 2001 from the Johannes Gutenberg-University Mainz, Germany; and by grants from the Fondation de France (no. 2002004994), the French Ministry of Research (ACI IMPBIO 032619), and the Institut National de la Santé et de la Recherche Médicale (Programme National de Recherches sur les Maladies Cardiovasculaires A04052DS).
Disclosures
None.
Footnotes
Original received December 30, 2005; revision received May 17, 2006; accepted June 2, 2006.
References
Chang HY, Yang X. Proteases for cell suicide: functions and regulation of caspases. Microbiol Mol Biol Rev. 2000; 64: 821–846.
Martinon F, Tschopp J. Inflammatory caspases: linking an intracellular innate immune system to autoinflammatory diseases. Cell. 2004; 117: 561–574. [Order article via Infotrieve]
Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, Kostura MJ, Miller DK, Molineaux SM, Weidner JR, Aunins J, Elliston KO, Ayala JM, Casano FJ, Chin J, Ding GJ-F, Egger LA, Gaffney EP, Limjuco G, Palyha OC, Raju SM, Rolando AM, Salley JP, Yamin T-T, Lee TD, Shively JE, MacCross M, Mumford RA, Schmidt JA, Tocci MJ. A novel heterodimeric cysteine protease is required for interleukin-1 beta processing in monocytes. Nature. 1992; 356: 768–774. [Order article via Infotrieve]
Cerretti DP, Kozlosky CJ, Mosley B, Nelson N, Van Ness K, Greenstreet TA, March CJ, Kronheim SR, Druck T, Cannizzaro LA, Huebner K, Black RA. Molecular cloning of the interleukin-1 beta converting enzyme. Science. 1992; 256: 97–100.
Li P, Allen H, Banerjee S, Franklin S, Herzog L, Johnston C, McDowell J, Paskind M, Rodman L, Salfeld J, Towne E, Tracey D, Wardwell S, Wei FY, Wong W, Kamen R, Seshadri T. Mice deficient in IL-1 beta-converting enzyme are defective in production of mature IL-1 beta and resistant to endotoxic shock. Cell. 1995; 80: 401–411. [Order article via Infotrieve]
Gu Y, Kuida K, Tsutsui H, Ku G, Hsiao K, Fleming MA, Hayashi N, Higashino K, Okamura H, Nakanishi K, Kurimoto M, Tanimoto T, Flavell RA, Sato V, Harding MW, Livingston DJ, Su MS. Activation of interferon-gamma inducing factor mediated by interleukin-1beta converting enzyme. Science. 1997; 275: 206–209.
Ghayur T, Banerjee S, Hugunin M, Butler D, Herzog L, Carter A, Quintal L, Sekut L, Talanian R, Paskind M, Wong W, Kamen R, Tracey D, Allen H. Caspase-1 processes IFN-gamma-inducing factor and regulates LPS-induced IFN-gamma production. Nature. 1997; 386: 619–623. [Order article via Infotrieve]
Waehre T, Yndestad A, Smith C, Haug T, Tunheim SH, Gullestad L, Froland SS, Semb AG, Aukrust P, Damas JK. Increased expression of interleukin-1 in coronary artery disease with downregulatory effects of HMG-CoA reductase inhibitors. Circulation. 2004; 109: 1966–1972.
Isoda K, Sawada S, Ishigami N, Matsuki T, Miyazaki K, Kusuhara M, Iwakura Y, Ohsuzu F. Lack of interleukin-1 receptor antagonist modulates plaque composition in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol. 2004; 24: 1068–1073.
Mallat Z, Corbaz A, Scoazec A, Besnard S, Leseche G, Chvatchko Y, Tedgui A. Expression of interleukin-18 in human atherosclerotic plaques and relation to plaque instability. Circulation. 2001; 104: 1598–1603.
Blankenberg S, Tiret L, Bickel C, Peetz D, Cambien F, Meyer J, Rupprecht H, for the AtheroGene Investigators. Interleukin 18 is a strong predictor of cardiovascular death in stable and unstable angina. Circulation. 2002; 106: 24–30.
Blankenberg S, Luc G, Ducimetière P, Arveiler D, Ferrières J, Amouyel P, Evans A, Cambien F, Tiret L, on behalf of the PRIME Study Group. Interleukin-18 and the risk of coronary heart disease in European men: the Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation. 2003; 108: 2453–2459.
de Nooijer R, von der Thusen JH, Verkleij CJ, Kuiper J, Jukema JW, van der Wall EE, van Berkel JC, Biessen EA. Overexpression of IL-18 decreases intimal collagen content and promotes a vulnerable plaque phenotype in apolipoprotein-E-deficient mice. Arterioscler Thromb Vasc Biol. 2004; 24: 2313–2319.
Kuida K, Lippke JA, Ku G, Harding MW, Livingston DJ, Su MS, Flavell RA. Altered cytokine export and apoptosis in mice deficient in interleukin-1 beta converting enzyme. Science. 1995; 267: 2000–2003.
Geng YJ, Libby P. Evidence for apoptosis in advanced human atheroma. Colocalization with interleukin-1 beta-converting enzyme. Am J Pathol. 1995; 147: 251–266.
Kolodgie FD, Narula J, Burke AP, Haider N, Farb A, Hui-Liang Y, Smialek J, Virmani R. Localization of apoptotic macrophages at the site of plaque rupture in sudden coronary death. Am J Pathol. 2000; 157: 1259–1268.
Pomerantz BJ, Reznikov LL, Harken AH, Dinarello CA. Inhibition of caspase 1 reduces human myocardial ischemic dysfunction via inhibition of IL-18 and IL-1beta. Proc Natl Acad Sci U S A. 2001; 98: 2871–2876.
Wang TD, Chen WJ, Mau TJ, Lin JW, Lin WW, Lee YT. Attenuation of increased myocardial ischaemia-reperfusion injury conferred by hypercholesterolaemia through pharmacological inhibition of the caspase-1 cascade. Br J Pharmacol. 2003; 138: 291–300. [Order article via Infotrieve]
Tiret L, Godefroy T, Lubos E, Nicaud V, Tregouet D, Barbaux S, Schnabel R, Bickel C, Espinola-Klein C, Poirier O, Perret C, Münzel T, Rupprecht HJ, Lackner K, Cambien F, Blankenberg S. Genetic analysis of the interleukin-18 system highlights the role of the interleukin-18 gene in cardiovascular disease. Circulation. 2005; 112: 643–650.
Rupprecht H, Blankenberg S, Bickel C, Rippin G, Hafner G, Prellwitz W, Schlumberger W, Meyer J, for the AtheroGene Investigators. Impact of viral and bacterial infectious burden on long-term prognosis in patients with coronary artery disease. Circulation. 2001; 104: 25–31.
Parra HJ, Arveiler D, Evans AE, Cambou JP, Amouyel P, Bingham A, McMaster D, Schaffer P, Douste-Blazy P, Luc G, Richard JL, Ducimetiere P, Fruchart JC, Cambien F. A case-control study of lipoprotein particles in two populations at contrasting risk for coronary heart disease. The ECTIM Study. Arterioscler Thromb. 1992; 12: 701–707.
Tregouet D, Ricard S, Nicaud V, Soubigou S, Duverger N, Tahri-Daizadeh N, Poirier O, Kee F, Morrison C, Tiret L, Evans A, Deleuze J, Cambien F. In depth haplotype analysis of ABCA1 gene polymorphisms in relation to plasma ApoA1 levels and myocardial infarction. Arterioscler Thromb Vasc Biol. 2004; 24: 775–781.
Tunstall-Pedoe H, Kuulasmaa K, Mahonen M, Tolonen H, Ruokokoski E, Amouyel P. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet. 1999; 353: 1547–1557. [Order article via Infotrieve]
Tregouet DA, Escolano S, Tiret L, Mallet A, Golmard JL. A new algorithm for haplotype-based association analysis: the Stochastic-EM algorithm. Ann Hum Genet. 2004; 68: 165–177. [Order article via Infotrieve]
Liljedahl U, Fredriksson M, Dahlgren A, Syvanen AC. Detecting imbalanced expression of SNP alleles by minisequencing on microarrays. BMC Biotechnol. 2004; 4: 24. [Order article via Infotrieve]
Rocher C, Faucheu C, Blanchet AM, Claudon M, Herve F, Durand L, Harnois M, Diu-Hercend A, Lalanne JL. Identification of five new genes, closely related to the interleukin-1beta converting enzyme gene, that do not encode functional proteases. Eur J Biochem. 1997; 246: 394–400.
the Departments of Medicine II (S. Blankenberg, H.J.R., C.B., R.S., T.M.) and Clinical Chemistry and Laboratory Medicine (K.J.L.), Johannes Gutenberg-University Mainz, Germany
INSERM U525 (T.G., O.P., S. Barbaux, V.N., F.C., L.T.), Université Pierre et Marie Curie/Faculté de Médecine Pitié-Salpêtrière, Paris, France
Department of Epidemiology and Public Health (F.K., A.E.), Belfast, UK
Greater Glasgow Health Board (C.M.), Glasgow, UK.
Abstract
Caspase-1 processes the interleukin (IL)-1 and IL-18 inactive precursors to the biologically active cytokines that are known to have proatherogenic effects. The present study investigated the genetic variability of the CASP1 gene and plasma levels of caspase-1 in relation to cardiovascular risk. In Europeans, 3 tag SNPs captured 4 common haplotypes of the CASP1 gene. Among these, the Ain6 allele of the G+7/in6A polymorphism was less frequent in 246 cases with myocardial infarction and a parental history of disease than in 253 controls free of familial history of disease (0.13±0.02 versus 0.20±0.02; P=0.005). However, in a larger case/control study (n=1774), these effects are borderline restricted to the UK population. In a prospective cohort of 1168 patients with coronary artery disease followed up during a median period of 6.0 years, the Ain6 allele exhibited a borderline association with future cardiovascular death (hazard ratio [HR]: 0.64, 0.41 to 1.01; P=0.053) and was associated with lower serum IL-18 levels (P=0.014). Baseline caspase-1 levels in the top quartile of the distribution were predictive of cardiovascular deaths (HR=3.62, 1.81 to 7.27; P=0.0003 compared with the bottom quartile). Finally, in vitro assays of allelic imbalance showed that the CASP1 haplotype carrying the Ain6 allele was associated with a lower mRNA expression. These results indicate that caspase-1 levels are predictive of future cardiovascular death in patients with coronary artery disease. The role of CASP1 genetic variations in the susceptibility to myocardial infarction requires further investigation.
Key Words: caspase-1 inflammation prognosis coronary disease genetics
Introduction
Caspases are a family of intracellular cysteine proteases that include at least 14 members that have been identified in mammals to date.1 Caspases play an essential role in apoptosis, but a subfamily comprising caspase-1, -4, and -5 are primarily involved in the regulation of inflammatory processes.2 Caspase-1, the best characterized of inflammatory caspases, was originally identified on the basis of its proteolytic activity for cleaving the inactive interleukin (IL)-1 precursor into the mature cytokine.3,4 This role was further confirmed by the impaired production of mature IL-1 in caspase-1–deficient mice.5 Caspase-1 is also responsible for the processing of IL-18, a cytokine that shares several structural features with the IL-1 family. The production of mature IL-18 requires cleavage of its precursor pro–IL-18 at a site similar to the cleavage site of pro–IL-1.6,7
Increasing evidence has accumulated from animal, experimental, and epidemiological data that IL-1 and IL-18 are involved in both development of atherosclerosis and plaque vulnerability as well as myocardial dysfunction as a consequence of ischemia/reperfusion injury.8–13 Given the critical role of caspase-1 in the production of these 2 cytokines, this enzyme may play a key role in the proatherogenic effects mediated by IL-1 and IL-18. Several lines of evidence support such a role. Caspase-1–deficient mice have a reduced production of interferon-, a proatherogenic factor whose induction by T cells and macrophages is stimulated by IL-18.5,14 Caspase-1 has been colocalized with apoptotic macrophages in sites of plaque rupture.15,16 Finally, inhibition of caspase-1 has been shown to reduce myocardial ischemic dysfunction in models of ischemia/ reperfusion via an inhibition of IL-18 and IL-1.17,18
As previously shown, circulating IL-18 levels were strongly predictive of future cardiovascular events.11,12 Furthermore, polymorphisms in the IL18 gene might contribute to the predisposition to complications of coronary artery disease (CAD).19 Given the impact of caspase-1 in the production of IL-18, the present study aimed to investigate whether baseline caspase-1 levels and polymorphisms of the CASP1 gene might influence cardiovascular risk. We sequenced the CASP1 gene and genotyped all single nucleotide polymorphisms (SNPs) in the SIPLAC study, a case/control study of myocardial infarction (MI). Three tag SNPs were then studied in relation to future cardiovascular death in the prospective AtheroGene cohort of CAD patients, and the most interesting tag SNP was further genotyped in the ECTIM study, a large case/control study of MI. Finally, functional assays were performed to detect potential differences in allelic expression of the CASP1 gene.
Materials and Methods
Study Populations
AtheroGene
The AtheroGene study is based on a prospective cohort of CAD patients recruited at the occasion of a diagnostic coronary angiography (Figure 1).20 A priori inclusion criterion was the presence of a diameter stenosis of >30% in at least 1 major coronary artery. The initial cohort was recruited between November 1996 and June 2000 and followed up for a median period of 6.0 (maximum 7.6) years. Stable (SAP) and unstable (UAP) angina pectoris were coded by Braunwald classification. Baseline IL-18 measurements and DNA were available for 1178 patients, of whom 106 died from a cardiovascular cause. Information about the cause of death was obtained by hospital or general practitioner. Extension of recruitment until February 2004 resulted in the inclusion of an additional 1294 patients, of whom 26 died from a cardiovascular cause. DNA, but not IL-18, was available in this extension cohort.
Because plasma samples of the early recruitment period tended to have been exhausted, caspase-1 was measured in patients recruited from June 1999 (n=1894). They were followed up for a median period of 2.5 (maximum 5.0) years and 71 died from a cardiovascular cause. There was an overlap of 450 patients in whom both caspase-1 and IL-18 were determined.
Healthy control subjects (n=468) were recruited either from the offices of general practitioners in the course of a routine check-up or by newspaper announcement. Caspase-1 was measured in 60% of controls for whom plasma was available.
Study participants had German nationality, were inhabitants of the Rhein-Main area, and were of European origin. Each participant gave written informed consent.
ECTIM and SIPLAC
The ECTIM study is a case/control study of MI conducted in the United Kingdom and France.21,22 Patients (35% females in the UK, only males in France) aged 25 to 64 years (for males) and 25 to 69 years (for females) were recruited from the WHO MONICA Project registers23 in Belfast and Glasgow (UK) and Lille, Strasbourg, and Toulouse (France). Age- and sex-matched controls were recruited from the populations of the areas covered by the registers. DNA was available for 1143 cases and 1133 controls.
The SIPLAC study is a substudy of ECTIM including highly contrasted samples of cases with MI and a parental history of MI (n=246) and controls without CAD and without parental history of MI (n=253) from the United Kingdom. This study provides us with a tool for selecting, among all SNPs discovered by molecular screening, the subset of those that will be further genotyped in larger studies, after exclusion of rare (0.02) and redundant polymorphisms. It is also intended to rapidly detect MI-associated polymorphisms that will require further replication.
Laboratory Methods in the AtheroGene Study
Blood was drawn under standardized conditions before coronary angiography was performed, and samples were stored at –80°C. Plasma caspase-1 (R&D Systems) and serum IL-18 (MBL Co Ltd) were measured using the ELISA technique according to the instructions of the manufacturer. C-reactive protein (CRP) was determined by a highly sensitive (hs), latex particle-enhanced immunoassay (Roche Diagnostics, Mannheim, Germany). Serum lipid levels were determined immediately by routine methods.
SNP Discovery in the CASP1 Gene
We sequenced all exons, up to 100 bp, of exon/intron junctions and up to 1 kb of 3'- and 5'-flanking sequences. Sequencing was performed in 31 pools of 2 MI patients (62 individuals) using a 16-capillary ABI-PRISM 3100 system and the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems [ABI], Warrington, UK). Sequences were compared using the ABI-PRISM SeqScape software version 2.0.
Genotyping
SNPs detected by sequencing were genotyped in the SIPLAC study by allele-specific oligonucleotide hybridization. Two tag SNPs (G+7/in6A and G+1034C) were selected for genotyping in the AtheroGene cohort plus a third SNP located in intron 8 (T+132/in8C), which, after consultation of the HapMap database (http://www.hapmap.org), appeared necessary to capture all common haplotypes of the CASP1 gene. Finally, the G+7/in6A SNP was genotyped in the ECTIM study. Genotyping in AtheroGene and ECTIM was performed using the 5' nuclease assay with TaqMan probes (see the expanded Materials and Methods in the online data supplement, available at http://circres.ahajournals.org). Primer and probe sequences, as well as amplification conditions for genotyping, can be found at our website, GeneCanvas (http://www.genecanvas.org).
Statistical Analysis
Departure from Hardy–Weinberg equilibrium was tested by 2 with 1 df. Association of polymorphisms with case/control status was tested by 2 analysis and with prospective outcome by Cox proportional-hazards regression analysis. Different models were successively fitted, first adjusted for age and sex, then additionally adjusted for cardiovascular risk factors (body mass index, diabetes, hypertension, smoking, HDL-cholesterol, triglycerides, acute coronary syndrome, number of diseased vessels, history of MI, statin, and -blocker intake), and finally adjusted for IL-18 levels. In all prospective analyses, the primary end point was cardiovascular death, and patients who died from other causes were censored at the time of death. Association of baseline caspase-1 levels, in quartiles, with cardiovascular mortality was tested by Cox analysis adjusted in the same way as for analyses with genotype. Association of polymorphisms with IL-18 and caspase-1 levels was tested by the GLM procedure adjusted for age, sex, and case/control status, and homogeneity of association between cases and controls was tested by introducing an interaction term in the model. IL-18 and caspase-1 were log-transformed to remove positive skewness and geometric means (95% confidence interval [CI]) are given.
Linkage disequilibrium (LD) and haplotype analyses were performed using the THESIAS software,24 available at http://www.genecanvas.org. Haplotypic effects, expressed as mean quantitative effects or as odds ratios (ORs), were estimated by comparison to the most frequent haplotype. A global test of association between haplotypes and the phenotype was performed by a likelihood ratio test. The homogeneity of effects associated with haplotypes carrying the Ain6 allele was tested by introducing a constraint on the parameters (2 with 1 df). Except for haplotype analyses, all analyses were performed with the SAS software version 8.01 (SAS Institute Inc, Cary, NC). P<0.05 was considered to be significant.
Allelic Imbalance Assays
Differential allelic expression of the CASP1 gene was tested by the method of allelic imbalance, which compares the ratio of cDNAs to that of genomic DNAs in individuals heterozygous for a transcribed polymorphism.25 cDNA was synthesized from RNA extracted from Epstein–Barr virus–transformed lymphoblastoid cell lines (LCLs) (expanded Materials and Methods). cDNA and genomic DNA from 11 individuals heterozygous for the L256L (C/T) polymorphism were subjected to the same reaction of allelic discrimination using the TaqMan technology. Each sample was submitted to 2 different RT-PCR, and the resulting cDNAs were amplified in duplicates. The regression line equation from the genomic DNA samples was calculated, and, for each cDNA sample, the deviation from this theoretical line was assessed. For each individual, the log (ratio) of the observed to the expected value was averaged over the 4 repetitions. The mean of the 11 values was compared by t test to the expected value of 0 under the null hypothesis of no allelic imbalance.
Real-Time RT-PCR Assays
mRNA expression levels in LCLs were compared between individuals with different genotypes for the L256L (C/T) polymorphism by real-time PCR quantification (expanded Materials and Methods). Twenty-two homozygotes for the C allele, 9 heterozygotes, and 2 homozygotes for the T allele were analyzed in duplicate, and the mean of the 2 measurements was taken. Rare homozygotes were pooled with heterozygotes for final comparison. The mean mRNA levels between CC subjects and T carriers were compared by t test.
Results
Sequencing of the CASP1 Gene and Genotyping in the SIPLAC Study
Nineteen polymorphisms were detected by sequencing (Figure 2), which were all genotyped in the SIPLAC study, except 3 of them, which were very rare (A+109/in5G, C+92/in7T, and –43in7I/D, each observed in 1 of 124 chromosomes). There was a block of 14 SNPs in complete association covering the whole gene sequence. Only 2 SNPs were outside of this block, 1 nonsynonymous SNP (R240Q) that was present in only 2 subjects and not further studied and 1 SNP located in the 3' region (G+1034C). The G+7/in6A SNP was selected for representing the block in complete association.
In the SIPLAC study, the G+7/in6A and G+1034C SNPs were in almost complete LD (D'0.97). The Ain6 allele was represented less in MI cases than in controls (0.13±0.02 versus 0.20±0.03 in Belfast, 0.12±0.02 versus 0.19±0.03 in Glasgow; P=0.005 for the case/control difference adjusted for center). The G+1034C SNP was not associated with MI risk (supplemental Table I).
CASP1 Polymorphisms and CAD Risk in the AtheroGene Study
Baseline characteristics of cases and controls in the initial AtheroGene study are shown in Table 1. Cases had increased levels of CRP, triglycerides, and a higher ratio of LDL- to HDL-cholesterol. The lower levels of LDL in patients was likely attributable to the high prevalence of lipid-lowering treatment. IL-18 levels were comparable in controls and CAD patients as a whole. However, there was a trend toward increased IL-18 levels in acute MI by comparison to UAP and SAP (66.0 versus 60.9 and 59.7 pg/mL; P=0.11). In the subset of subjects having both IL-18 and caspase-1 measurements, caspase-1 levels were higher in cases than in controls (Table 1).
The G+7/in6A, T+132/in8C, and G+1034C SNPs were genotyped in the initial AtheroGene study. Because of strong LD, they generated only 4 common haplotypes, Gin6Tin8G+1034, Gin6Cin8G+1034, Ain6Tin8G+1034, and Ain6Tin8C+1034, with frequency in controls of 0.54, 0.28, 0.06, and 0.12, respectively.
The Ain6 allele frequency was similar in healthy German controls to that observed in controls from the United Kingdom (0.18±0.01 versus 0.19±0.02; P=0.45). There was no difference of the Ain6 allele frequency between CAD patients as a whole and controls (0.17±0.01 versus 0.18±0.01; P=0.75). However, the Ain6 allele frequency was significantly decreased in patients with acute MI (0.11±0.02; P=0.042), whereas it did not differ in patients with SAP (0.17±0.01; P=0.79) or UAP (0.19±0.02; P=0.60) from the frequency observed in controls (supplemental Table II). The T+132/in8C and G+1034C SNPs were not associated with clinical status (supplemental Table II). Haplotype frequencies did not differ between cases as a whole and controls (P=0.87).
CASP1 Polymorphisms and Circulating IL-18 Levels in the AtheroGene Study
The G+7/in6A SNP was associated with serum IL-18 levels, the Ain6 allele being associated with a lowering effect in a fairly additive fashion (Table 2). The T+132/in8C and G+1034C SNPs were not associated with IL-18 levels. Haplotype analysis was performed in cases and controls pooled because there was no significant heterogeneity between groups. By reference to the most frequent haplotype (Gin6Tin8G+1034), haplotype Gin6Cin8G+1034 had no significant effect on IL-18 levels (P=0.97), whereas haplotypes Ain6Tin8G+1034 and Ain6Tin8C+1034 had similar lowering effects (P=0.48 for the homogeneity of effects), suggesting that neither the T+132/in8C nor the G+1034C SNP had an individual effect and that only the G+7/in6A SNP was associated with IL-18 levels (Figure 3).
CASP1 Polymorphisms and Cardiovascular Mortality in the AtheroGene Study
In the initial AtheroGene cohort, there was a trend toward a lower risk of cardiovascular deaths during follow-up in carriers of the Ain6 allele than in Gin6Gin6 homozygotes, although the association did not reach significance (age- and sex-adjusted hazard ratio [HR]: 0.64, 0.41 to 1.01; P=0.053). Kaplan–Meier curves of survival according to genotype are shown in Figure 4. Adjustment for cardiovascular risk factors did not modify the relationship (HR=0.65, 0.41 to 1.03; P=0.07), nor did additional adjustment for IL-18 (HR=0.68, 0.43 to 1.08; P=0.10). No significant association was observed with the T+132/in8C or the G+1034C SNP (supplemental Table III).
To increase the power of the survival analysis, the G+7/in6A SNP was analyzed in the extension cohort. When combining both cohorts, the HR associated with carriage of the Ain6 allele, before and after adjustment for cardiovascular risk factors, was 0.66 (0.44 to 0.99; P=0.04) and 0.68 (0.45 to 1.03; P=0.07), respectively.
Plasma Caspase-1 Levels and Cardiovascular Mortality in the AtheroGene Study
When considering all subjects having a caspase-1 determination (Figure 1), higher levels were observed in CAD patients than in healthy controls, but this difference was mainly attributable to an increase in MI patients (age- and sex-adjusted mean levels: 59.3, 52.8, 51.9, and 51.4 pg/mL in MI, UAP, SAP, and control groups, respectively; P<0.0001). Caspase-1 correlated positively with CRP (r=0.19; P<0.001), but the association with IL-18 was rather weak (r=0.14; P<0.001) despite their biological link.
Figure 5 displays the Kaplan–Meier curves for survival from cardiovascular death according to caspase-1 quartiles. HRs estimated from different Cox models are given in supplemental Table IV. All models indicated an increased risk in patients in the top quartile of caspase-1 levels compared with those in the bottom quartile, whereas patients in the intermediate quartiles were not at higher risk. The age- and sex-adjusted HR (95% CI) in the top quartile was 3.62 (1.81 to 7.27; P=0.0003). Adjustment for classical risk factors hardly modified the association (HR=2.89, 1.42 to 5.88; P=0.003). Further adjustment for CRP levels slightly weakened the association (HR=2.32, 1.13 to 4.79; P=0.03).
CASP1 Polymorphisms and Plasma Caspase-1 Levels in the AtheroGene Study
There was a borderline association of the G+7/in6A and the G+1034C SNPS with plasma caspase-1 levels, the minor allele of each SNP being associated with a lowering effect (P=0.066 for the significance of each polymorphism, Table 2). By reference to the Gin6G+1034 haplotype, haplotypes Ain6G+1034 and Ain6C+1034 had similar lowering effects (P=0.65 for the homogeneity of effects), indicating that the G+1034C SNP had no effect by itself. The T+132/in8C SNP was genotyped only in a subset of subjects having a caspase-1 measurement and there was no significant association.
CASP1/G+7in6A Polymorphism and MI Risk in the ECTIM Study
Because of the suggestive evidence of an association of the G+7/in6A SNP with MI risk both in SIPLAC and AtheroGene, we further genotyped this SNP in the whole ECTIM study (Figure 1). In the whole population from the United Kingdom, encompassing SIPLAC, the association was weakened, the Ain6 allele frequency being 0.146±0.009 in cases and 0.170±0.010 in controls, respectively (P=0.07). By contrast, there was no case/control difference of allele frequency in France (0.175±0.014 versus 0.161±0.013; P=0.46) (test of homogeneity between countries; P=0.10). When stratifying according to parental history of MI, there was no significant interaction in either country (supplemental Table V).
Functional Assays
We further examined whether the 2 haplotypes defined by the block containing the G+7/in6A SNP were associated with a difference in gene expression levels. The exonic polymorphism taken for allelic imbalance assays was the coding SNP L256L (C/T), which belongs to the block (Figure 2). Expression of allelic transcripts in 11 heterozygotes are shown in Figure 6. The regression line equation predicted from the genomic DNA samples was associated with an R2 of 98%, compared with 100%, for a perfect equimolar ratio between the 2 alleles. Most values for the cDNA samples were situated above the regression line, suggesting a higher expression level associated with the C allele than with the T allele (mean difference of the observed to expected values : 0.31±0.12; P=0.028).
We confirmed these results by the quantification of mRNA levels by real-time RT-PCR. Higher expression levels were found in individuals homozygous for the C allele (1.91±0.34, n=22) than in carriers of the T allele (0.53±0.48, n=11) (P=0.02).
Discussion
Caspase-1 belongs to the subfamily of inflammatory caspases, also comprising caspase-4 and -5, that share similar biological functions in the regulation of inflammatory processes.2 In particular, all 3 caspases can cleave pro–IL-18 to generate the biologically active IL-18, although caspase-1 is the most efficient in this process.6,7 The genes coding for these evolutionary-related caspases are located in a cluster on chromosome 11q22.2 to 22.3 and present more than 50% of sequence homology.1 Besides the CASP4 and CASP5 genes located downstream, we identified several regions located upstream and presenting more than 70% of homology with the CASP1 gene sequence. These sequences are presumably those that were previously identified and considered to be pseudogenes, because none of these encoded a functional cysteine protease.26
The CASP1 gene exhibited a pattern of reduced haplotype diversity, with only 4 haplotypes captured by 3 tag SNPs. We investigated whether this block might extend beyond the CASP1 gene itself by examining the HapMap data. In Europeans, the block stretched over 32 kb but did not include the CASP4 and CASP5 genes, therefore excluding the possibility that polymorphisms in these genes might be responsible for the observed associations.
Our study presented a comprehensive investigation of the role of caspase-1 in susceptibility to CAD and its complications. However, it failed to provide definite conclusions. Our initial results provided several arguments supporting an etiologic role of caspase-1 : (1) the Ain6 allele of the CASP1 gene was associated with a lower risk of MI in 2 different case/control studies, SIPLAC, and AtheroGene; (2) the same allele exhibited a borderline association with prospective cardiovascular death in AtheroGene; (3) plasma caspase-1 was elevated in patients with acute MI and appeared as a biomarker predictive of cardiovascular mortality in CAD patients; (4) circulating IL-18 levels were decreased in individuals carrying the Ain6 allele; and (5) the CASP1 haplotype carrying the Ain6 allele was associated with a lower gene expression, supporting the existence of a functional polymorphism within the CASP1 gene.
Cases in SIPLAC were young survivors of MI with a parental history of disease recruited in the United Kingdom. They were compared with controls selected on the absence of familial and personal histories of MI. Moreover, populations of Northern Ireland and Scotland are among those having the highest prevalence of CAD in Europe.23 When stratified according population in the entire ECTIM study, the effect was observed in the United Kingdom and was absent in those recruited in France. It is then likely that the combination of these 2 criteria—family history and populations—has led to the selection of individuals genetically more susceptible than the other subjects included in the ECTIM study. However, it cannot be ruled out that, given the multiple hypotheses tested and the lack of replication, the initial associations were attributable to chance. Further studies are required to address this issue.
The genetic effects observed in SIPLAC and AtheroGene were modest and can be viewed only as suggestive of a role of caspase-1 in the etiology and clinical consequences of MI. By linking up the epidemiological and experimental results, it might be speculated that the haplotype carrying the Ain6 allele, which is associated in vitro with a lower CASP1 gene expression, results in a reduced rate of cleavage of pro–IL-18 into the biologically active IL-18, which in turn would confer a protection against the proinflammatory effects of IL-18. Given the complexity of the haplotype block, including at least 14 polymorphisms, it is difficult to predict which among them might be functional. Two of the SNPs are synonymous polymorphisms (T54T and L256L), which are unlikely to have a functional role. Six SNPs are located in exon/intron junctions and might cause alternative splicing that would result in the production of proteins of different sizes. However, we did not detect any difference in total cDNA size among individuals with different genotypes (data not shown). One SNP is located in the 5' region and might affect gene expression, whereas 5 SNPs are in the 3' region and might be involved in mRNA stability.
Circulating caspase-1 levels were independently related to future cardiovascular death in CAD patients. This result supports the hypothesis that activation of the caspase-1/IL-18 pathway is involved in the clinical complications of preexisting atherosclerotic disease. In animal models, inhibition of capsase-1 has been shown to reduce ischemia/reperfusion injury.17,18 Therefore, increased levels of caspase-1 might reflect individuals at high risk of post-MI complications, rather than early development of atherosclerosis itself. In the same vein, caspase-1 and IL-18 levels were increased in patients with acute MI but not UAP. The discrepancy might also corroborate the hypothesis that the caspase-1/IL-18 pathway is predominantly activated after MI has occurred.
As an important limitation with regard to the results of the genetic variations, we could not correct for multiple testing. Therefore, these results have to be considered as hypotheses generating and need replication in large population studies.
Paradoxically, circulating levels of caspase-1 weakly correlated with those of IL-18, and the CASP1 gene effects were stronger on IL-18 than on caspase-1 itself. This might be explained by the fact that the caspase-1 assay determined the mass and not the activity itself, whereas the IL-18 assay measured a compound of the active form and the IL-18/IL-18-binding protein complex.
In conclusion, the present results indicated that caspase-1 levels were predictive of future cardiovascular death in CAD patients. They also suggested an etiologic role of CASP1 genetic variations in the susceptibility to MI, although this requires further replication.
Acknowledgments
We thank Biosite Inc for determination of caspase-1 levels.
Sources of Funding.
The ECTIM study was supported by a grant of the British Heart Foundation (PG/93045). The AtheroGene study was supported by a grant from the "Stiftung Rheinland-Pfalz für Innovation," Ministry for Science and Education (AZ 15202-386261/545), Mainz, Germany; by the MAIFOR grant 2001 from the Johannes Gutenberg-University Mainz, Germany; and by grants from the Fondation de France (no. 2002004994), the French Ministry of Research (ACI IMPBIO 032619), and the Institut National de la Santé et de la Recherche Médicale (Programme National de Recherches sur les Maladies Cardiovasculaires A04052DS).
Disclosures
None.
Footnotes
Original received December 30, 2005; revision received May 17, 2006; accepted June 2, 2006.
References
Chang HY, Yang X. Proteases for cell suicide: functions and regulation of caspases. Microbiol Mol Biol Rev. 2000; 64: 821–846.
Martinon F, Tschopp J. Inflammatory caspases: linking an intracellular innate immune system to autoinflammatory diseases. Cell. 2004; 117: 561–574. [Order article via Infotrieve]
Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, Kostura MJ, Miller DK, Molineaux SM, Weidner JR, Aunins J, Elliston KO, Ayala JM, Casano FJ, Chin J, Ding GJ-F, Egger LA, Gaffney EP, Limjuco G, Palyha OC, Raju SM, Rolando AM, Salley JP, Yamin T-T, Lee TD, Shively JE, MacCross M, Mumford RA, Schmidt JA, Tocci MJ. A novel heterodimeric cysteine protease is required for interleukin-1 beta processing in monocytes. Nature. 1992; 356: 768–774. [Order article via Infotrieve]
Cerretti DP, Kozlosky CJ, Mosley B, Nelson N, Van Ness K, Greenstreet TA, March CJ, Kronheim SR, Druck T, Cannizzaro LA, Huebner K, Black RA. Molecular cloning of the interleukin-1 beta converting enzyme. Science. 1992; 256: 97–100.
Li P, Allen H, Banerjee S, Franklin S, Herzog L, Johnston C, McDowell J, Paskind M, Rodman L, Salfeld J, Towne E, Tracey D, Wardwell S, Wei FY, Wong W, Kamen R, Seshadri T. Mice deficient in IL-1 beta-converting enzyme are defective in production of mature IL-1 beta and resistant to endotoxic shock. Cell. 1995; 80: 401–411. [Order article via Infotrieve]
Gu Y, Kuida K, Tsutsui H, Ku G, Hsiao K, Fleming MA, Hayashi N, Higashino K, Okamura H, Nakanishi K, Kurimoto M, Tanimoto T, Flavell RA, Sato V, Harding MW, Livingston DJ, Su MS. Activation of interferon-gamma inducing factor mediated by interleukin-1beta converting enzyme. Science. 1997; 275: 206–209.
Ghayur T, Banerjee S, Hugunin M, Butler D, Herzog L, Carter A, Quintal L, Sekut L, Talanian R, Paskind M, Wong W, Kamen R, Tracey D, Allen H. Caspase-1 processes IFN-gamma-inducing factor and regulates LPS-induced IFN-gamma production. Nature. 1997; 386: 619–623. [Order article via Infotrieve]
Waehre T, Yndestad A, Smith C, Haug T, Tunheim SH, Gullestad L, Froland SS, Semb AG, Aukrust P, Damas JK. Increased expression of interleukin-1 in coronary artery disease with downregulatory effects of HMG-CoA reductase inhibitors. Circulation. 2004; 109: 1966–1972.
Isoda K, Sawada S, Ishigami N, Matsuki T, Miyazaki K, Kusuhara M, Iwakura Y, Ohsuzu F. Lack of interleukin-1 receptor antagonist modulates plaque composition in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol. 2004; 24: 1068–1073.
Mallat Z, Corbaz A, Scoazec A, Besnard S, Leseche G, Chvatchko Y, Tedgui A. Expression of interleukin-18 in human atherosclerotic plaques and relation to plaque instability. Circulation. 2001; 104: 1598–1603.
Blankenberg S, Tiret L, Bickel C, Peetz D, Cambien F, Meyer J, Rupprecht H, for the AtheroGene Investigators. Interleukin 18 is a strong predictor of cardiovascular death in stable and unstable angina. Circulation. 2002; 106: 24–30.
Blankenberg S, Luc G, Ducimetière P, Arveiler D, Ferrières J, Amouyel P, Evans A, Cambien F, Tiret L, on behalf of the PRIME Study Group. Interleukin-18 and the risk of coronary heart disease in European men: the Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation. 2003; 108: 2453–2459.
de Nooijer R, von der Thusen JH, Verkleij CJ, Kuiper J, Jukema JW, van der Wall EE, van Berkel JC, Biessen EA. Overexpression of IL-18 decreases intimal collagen content and promotes a vulnerable plaque phenotype in apolipoprotein-E-deficient mice. Arterioscler Thromb Vasc Biol. 2004; 24: 2313–2319.
Kuida K, Lippke JA, Ku G, Harding MW, Livingston DJ, Su MS, Flavell RA. Altered cytokine export and apoptosis in mice deficient in interleukin-1 beta converting enzyme. Science. 1995; 267: 2000–2003.
Geng YJ, Libby P. Evidence for apoptosis in advanced human atheroma. Colocalization with interleukin-1 beta-converting enzyme. Am J Pathol. 1995; 147: 251–266.
Kolodgie FD, Narula J, Burke AP, Haider N, Farb A, Hui-Liang Y, Smialek J, Virmani R. Localization of apoptotic macrophages at the site of plaque rupture in sudden coronary death. Am J Pathol. 2000; 157: 1259–1268.
Pomerantz BJ, Reznikov LL, Harken AH, Dinarello CA. Inhibition of caspase 1 reduces human myocardial ischemic dysfunction via inhibition of IL-18 and IL-1beta. Proc Natl Acad Sci U S A. 2001; 98: 2871–2876.
Wang TD, Chen WJ, Mau TJ, Lin JW, Lin WW, Lee YT. Attenuation of increased myocardial ischaemia-reperfusion injury conferred by hypercholesterolaemia through pharmacological inhibition of the caspase-1 cascade. Br J Pharmacol. 2003; 138: 291–300. [Order article via Infotrieve]
Tiret L, Godefroy T, Lubos E, Nicaud V, Tregouet D, Barbaux S, Schnabel R, Bickel C, Espinola-Klein C, Poirier O, Perret C, Münzel T, Rupprecht HJ, Lackner K, Cambien F, Blankenberg S. Genetic analysis of the interleukin-18 system highlights the role of the interleukin-18 gene in cardiovascular disease. Circulation. 2005; 112: 643–650.
Rupprecht H, Blankenberg S, Bickel C, Rippin G, Hafner G, Prellwitz W, Schlumberger W, Meyer J, for the AtheroGene Investigators. Impact of viral and bacterial infectious burden on long-term prognosis in patients with coronary artery disease. Circulation. 2001; 104: 25–31.
Parra HJ, Arveiler D, Evans AE, Cambou JP, Amouyel P, Bingham A, McMaster D, Schaffer P, Douste-Blazy P, Luc G, Richard JL, Ducimetiere P, Fruchart JC, Cambien F. A case-control study of lipoprotein particles in two populations at contrasting risk for coronary heart disease. The ECTIM Study. Arterioscler Thromb. 1992; 12: 701–707.
Tregouet D, Ricard S, Nicaud V, Soubigou S, Duverger N, Tahri-Daizadeh N, Poirier O, Kee F, Morrison C, Tiret L, Evans A, Deleuze J, Cambien F. In depth haplotype analysis of ABCA1 gene polymorphisms in relation to plasma ApoA1 levels and myocardial infarction. Arterioscler Thromb Vasc Biol. 2004; 24: 775–781.
Tunstall-Pedoe H, Kuulasmaa K, Mahonen M, Tolonen H, Ruokokoski E, Amouyel P. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet. 1999; 353: 1547–1557. [Order article via Infotrieve]
Tregouet DA, Escolano S, Tiret L, Mallet A, Golmard JL. A new algorithm for haplotype-based association analysis: the Stochastic-EM algorithm. Ann Hum Genet. 2004; 68: 165–177. [Order article via Infotrieve]
Liljedahl U, Fredriksson M, Dahlgren A, Syvanen AC. Detecting imbalanced expression of SNP alleles by minisequencing on microarrays. BMC Biotechnol. 2004; 4: 24. [Order article via Infotrieve]
Rocher C, Faucheu C, Blanchet AM, Claudon M, Herve F, Durand L, Harnois M, Diu-Hercend A, Lalanne JL. Identification of five new genes, closely related to the interleukin-1beta converting enzyme gene, that do not encode functional proteases. Eur J Biochem. 1997; 246: 394–400.
您现在查看是摘要介绍页，详见ORG附件。