Abstract
Objective. To determine whether angiotensin-converting enzyme (ACE) polymorphisms including I/D and 2 single-nucleotide polymorphisms (SNP) affect susceptibility to systemic sclerosis (SSc) in a large French Caucasian population.
Methods. A case-control study was performed in 494 patients with SSc and 280 healthy controls for I/D polymorphism. Two supplementary exonic SNP of ACE gene (rs4309, rs4362) were genotyped in 659 patients with SSc and 511 matched healthy controls. Among the whole SSc population, 453 (67%) patients with SSc had the limited cutaneous subtype, 47 (7%) had precapillary pulmonary arterial hypertension, 209 (32%) had digital ulcers, and 10 (1.5%) had renal crisis. A combined analysis of the available results for ACE I/D genotypes in Caucasians was also performed.
Results. There was no association between the 3 polymorphic markers and SSc for allelic and genotype frequencies. No association was observed for the different vascular subsets of the disease. Haplotype analyses did not detect any association. The lack of association for ACE I/D was confirmed by the combined analysis.
Conclusion. These results in a large cohort of European Caucasian patients with SSc do not support that the ACE gene is implicated in the pathogenesis of SSc and its vascular damage.
Although the pathogenesis of systemic sclerosis (SSc) remains unclear, microvascular abnormalities have been reported as an early key step of the disease. In SSc renal crisis, a local activation of the renin-angiotensin system (RAS) has been observed1. Moreover, angiotensin-converting enzyme (ACE) inhibitors were found to dramatically improve the outcome of the renal crisis in SSc2. The RAS is closely implicated in macrovascular disease favoring vasoconstriction; in SSc, prevalence of macrovascular involvement remains a matter of debate3,4.
SSc is a multifactorial disease and it is believed that both genetic and environmental factors contribute to disease susceptibility5. ACE, a key enzyme in the RAS, is encoded by the ACE gene, mapping to the 17 (17q 23) region. An insertion/deletion (I/D) polymorphism has been reported associated not only with the level of ACE but also with endothelial disturbances6. In SSc, the question of an association of ACE I/D has been studied in different populations, with conflicting results. Our aim was to assess whether polymorphisms in ACE gene affect susceptibility to SSc in large European Caucasian populations.
MATERIALS AND METHODS
First, we included 774 consecutive unrelated subjects for the ACE I/D genotyping according to a published procedure6, comprising 494 patients with SSc classified according to LeRoy, et al7 and 280 healthy age- and sex-matched controls (mean age 66 ± 16 yrs; 75% were female). Then, 1170 subjects, 659 patients with SSc and 511 healthy matched controls (including those of the first set), were included for the genotyping of 2 supplementary ACE SNP (rs4309 and rs4362) using the KASpar Genotyping system (KBioscience)8.
All subjects were of European Caucasian origin, defined by all 4 grandparents being French Caucasian. The Ethics Committee of Cochin Hospital approved our study and all the individuals gave written informed consent.
We thereafter conducted a combined analysis restricted on Caucasian populations; literature search was performed on articles published and expanded on Medline, EMBASE, and European League Against Rheumatism/American College of Rheumatology congress abstract archives. We obtained 4 independent populations including our study. We extracted the data from contingency tables of Fatini, et al9, Assassi, et al10, and Guiducci, et al11 for ACE I/D genotypes.
Statistical analyses
The Hardy–Weinberg equilibrium was investigated with a chi-squared test with 1 degree of freedom. Power calculations were driven through an asymptotic non-central chi-squared approach and provided a power of 80% to detect the association between SSc for a genotype of 30% frequency with an OR of 1.5 at the 5% significance level. Finally, ACE haplotypes were constructed using the maximum likelihood procedure. Fisher’s exact test was used to compare alleles and genotype frequencies using R software version 2.6.0. Odds ratios (OR) were calculated with the most frequent homozygous genotype or allele as reference.
A combined analysis was conducted via logistic regression models, adjusted for each of the 4 Caucasian populations studied.
RESULTS
The demographic data and disease characteristics of patients with SSc in both sets are detailed in Table 1. All the 3 polymorphisms were in Hardy-Weinberg equilibrium for the control group. Frequencies of alleles and genotypes for the 3 markers are provided in Table 2. The frequency of D allele of insertion polymorphism and the minor allele for, respectively, the 2 SNP (rs4309 and rs4362) was not different in SSc and in controls, with, respectively, 54% versus 56% (p = 0.56, OR 0.94, 95% CI 0.77–1.16) for ACE I/D, 41% versus 39% (p = 0.33, OR 1.09, 95% CI 0.92–1.30) for rs4309, and 45% versus 44% (p = 0.79, OR 1.02, 95% CI 0.86–1.21) for rs4362. The frequency of genotypes carrying at least one D allele for ACE I/D and one minor allele for SNP was similar in SSc patients compared to controls (Table 2). Regarding SSc and subgroup analyses, especially for those with vascular involvement, we did not detect any allelic and genotypic difference. There was neither overall association (p = 0.65) nor association for common haplotypes with disease status (data not shown).
Characteristics of the French Caucasian cohort of patients with systemic sclerosis.
Genotypic and allelic frequencies of the three ACE polymorphisms.
The combined analysis (Figure 1) showed that the frequency of D allele among the SSc cases (n = 684; Table 2, combined populations) did not differ from that found among the controls (n = 563): 56% vs 54%, respectively (p = 0.48, OR 0.89, 95% CI 0.65–1.22). Neither was the frequency of genotypes carrying at least one D allele different among patients and controls (p = 0.89, OR 1.02, 95% CI 0.79–1.29).
Combined analysis of the 4 Caucasian case–control studies of systemic sclerosis (SSc) and the ACE I/D genotypes. Results from our study and those conducted by Fatini, et al9, Assassi, et al10, and Guiducci, et al11 were analyzed by logistic regression. Values are the odds ratio and 95% confidence interval (plotted on a logarithmic scale).
DISCUSSION
In SSc, vascular involvement has a major effect on the prognosis12. Numerous arguments support a role of RAS in SSc: (1) ACE is suspected to play a key role in renal crisis; (2) SSc is characterized by vasospasm and ACE participates in regulation of vascular tone; and (3) genetic association between ACE I/D and SSc has been suggested.
Conflicting data have been reported: an association was detected between D allele of ACE I/D in an Italian cohort9 but this was not replicated within North American10, Greek11, or Korean populations13. These discrepancies may be explained by underpowered studies and by population stratification because of a heterogeneous genetic background. To avoid these biases we have focused on European Caucasians and recruited a large sample size providing a strong power (> 80%). Our results show no association between D allele or genotype carrying at least one D allele (DD and ID) or the 2 SNP investigated and SSc. The validity of our results is further supported by our control group results that are close to those reported in HapMap (NCBI database) and also in a recent series14. Our combined analysis for ACE I/D polymorphism in Caucasians provides further evidence of the absence of involvement of the ACE I/D in SSc. One limitation may be that we did not perform macrovascular assessment6 in our large series, although digital ulcers may somehow represent patients with middlesize/small artery involvement15. Another limitation comes from the small number of patients with renal crisis; thus ACE polymorphisms will have to be investigated in a larger cohort of SSc patients with renal crisis.
The genotyping of 3 ACE gene polymorphisms in a large cohort of European Caucasian patients with SSc did not allow us to detect any allelic, genotypic, or haplotypic associations with the disease or its main vascular complications.
Footnotes
-
Supported by Association des Sclérodermiques de France, Société Française de Rhumatologie, INSERM, and Agence Nationale pour la Recherche (grant R07094KS) and by Groupe Français de Recherche sur la Sclérodermie.
- Accepted for publication September 15, 2008.
