Association Study of Serotonin Transporter Gene (SLC6A4) in Systemic Sclerosis in European Caucasian Populations

MATERIALS AND METHODS

A total of 1170 unrelated individuals were included: 667 SSc patients (56 having precapillary PAH diagnosed by catheterization, 207 having a history of digital ulcerations, and 17 with concomitant PAH and digital ulcerations) and 511 healthy matched controls, all of European Caucasian origin. The whole population has been genotyped for 3 SLC6A4 Tag single-nucleotide polymorphisms (SNP: rs1042173, rs2066713, rs6354) using the KASpar Genotyping system (Kbioscience, Hoddesdon, UK)^11,12. These 3 TagSNP account for the common genetic diversity according to HapMap data. In a second study, using classical direct genotyping by polymerase chain reaction⁴, a randomly selected subsample of 364 SSc patients (34 with PAH and 138 with a history of digital ulcerations) and 218 healthy controls were genotyped for the SLC6A4 L/S polymorphism.

The ethics committee of Cochin Hospital approved the study and all subjects gave written informed consent.

The screening and definition of patients with PAH were performed as reported¹³. Patients identified by (1) an echocardiographically estimated systolic pulmonary artery pressure (PAP) > 40 mm Hg; (2) DLCO < 50% of predicted in the absence of pulmonary fibrosis; or (3) unexplained dyspnea, were offered right-heart catheterization. Confirmed PAH was defined as a resting mean PAP ≥ 25 mm Hg with pulmonary capillary wedge pressure ≤ 15 mm Hg.

Statistical analysis

Hardy–Weinberg equilibrium was investigated with a chi-square test with 1 degree of freedom. Power calculations were driven through an asymptotic non-central chi-square approach and provided a power of 81% to detect association for a genotype of 38% frequency with an OR of 1.5 at the 5% significance level. Fisher’s exact test was used to compare allele and genotype frequencies.

RESULTS

All demographic data and disease characteristics of SSc patients are given in Table 1. Two of the 3 SNP studied (rs1042173 and rs2066713) were in Hardy-Weinberg equilibrium for the control group, but the rs6354 variant deviated and was excluded from subsequent analyses. Regarding the 2 TagSNP, no difference was found between allelic and genotypic frequencies of patients and controls (Table 2). The sub-phenotype analyses, especially those related to vascular involvement, did not detect any SLC6A4 rs1042173 and rs2066713 allelic or genotypic association (Table 2). Further, the frequency of the S allele was not different in SSc patients compared to controls (Table 3). In the SSc-PAH and digital ulcerations subgroups, the allelic and genotypic frequencies of the L/S polymorphism were similar in patients and in controls (Table 3). Intra-cohort analyses did not reveal any difference between SSc patients with and those without the vascular trait for any of the 3 variants.

DISCUSSION

SSc-related PAH has emerged as a leading and lethal complication of the disease. Identification of PAH risk factors is a key issue to improve outcomes in SSc. Evidence of a key role of the serotonin transporter in PAH has been provided recently: (1) 5-HTT expression was shown to be increased in lungs of patients with idiopathic PAH¹⁴; (2) serotonin induces greater activation of PA-SMC proliferation in patients with PAH⁴; (3) mice disrupted for the SLC6A4 gene developed less severe hypoxic PAH than wild-type controls¹⁵; and (4) an association between the functional L/S polymorphism and PAH was reported^4–6.

Our results show that the SLC6A4 polymorphisms we studied do not contribute to susceptibility to SSc and its main microvascular consequences. The large sample size of our cohort provides strong power (> 80%) for global analysis of SSc, and the focus on European Caucasian individuals avoids a heterogeneity bias. In idiopathic PAH, the LL genotype of the SLC6A4 promoter polymorphism was present in homozygous form in 65% of patients, but in only 27% of controls⁴. Results from our control group are close to these and support the validity of our findings. Conversely, the discrepancies we observed between the SSc-PAH findings and previous results in PAH studies may be explained by specificities of SSc-PAH compared to familial or idiopathic PAH, including differences in genetic background. Despite inclusion of a large number of patients, and due to the scarcity of PAH, the statistical power was lower for this subset and this may lead to a type 2 error. In addition, as the PAH phenotype can change during the disease course, a prospective followup of the SSc cohort is needed to clarify the phenotype and improve sub-phenotype analyses. However, our cohort was the largest used to date for an association study related to vascular sub-phenotype; in addition PAH was diagnosed for our study using the “gold standard,” that is, catheterization.

To conclude, common variants of the SLC6A4 gene, including the L/S polymorphism, were found not to be associated with overall SSc or with SSc-related PAH and/or digital ulcerations. Rare variants and other serotonin pathways should now be investigated.