Article Text
PDF
Abstract
Introduction Systemic sclerosis (SSc)-related pulmonary arterial hypertension (PAH) has emerged as a major mortality prognostic factor. Mutations of transforming growth factor beta (TGFβ) receptor genes strongly contribute to idiopathic and familial PAH.
Objective To explore the genetic bases of SSc–PAH, we combined direct sequencing and genotyping of candidate genes encoding TGFβ receptor family members.
Materials and methods TGFβ receptor genes, BMPR2, ALK1, TGFR2 and ENG, were sequenced in 10 SSc–PAH patients, nine SSc and seven controls. In addition, 22 single-nucleotide polymorphisms (SNP) of these four candidate genes were tested for association in a first set of 824 French Caucasian SSc patients (including 54 SSc–PAH) and 939 controls. The replication set consisted of 1516 European SSc (including 219 SSc–PAH) and 3129 controls from the European League Against Rheumatism Scleroderma Trials and Research group network.
Results No mutation was identified by direct sequencing. However, two repertoried SNP, ENG rs35400405 and ALK1 rs2277382, were found in SSc–PAH patients only. The genotyping of 22 SNP including the latter showed that only rs2277382 was associated with SSc–PAH (p=0.0066, OR 2.13, 95% CI 1.24 to 3.65). Nevertheless, this was not replicated with the following result in combined analysis: p=0.123, OR 0.79, 95% CI 0.59 to 1.07.
Conclusions This study demonstrates the lack of association between these TGFβ receptor gene polymorphisms and SSc–PAH using both sequencing and genotyping methods.
- Systemic Sclerosis
- Gene Polymorphism
- Autoimmune Diseases
Statistics from Altmetric.com
Introduction
Systemic sclerosis (SSc) is characterised by major vascular involvement. Pulmonary arterial hypertension (PAH) is currently an important challenge in SSc and given the severity of this condition and the poor understanding of its risk factors and pathogenesis, there is an urgent need to identify novel risk factors for the development of SSc–PAH.1 The identification of mutations in the BMPR2 gene, and also in other transforming growth factor beta (TGFβ) receptor genes in idiopathic PAH and familial PAH has been an important step forward. Indeed, mutations in the BMPR2 gene, which encodes a type II bone morphogenetic protein receptor of the TGFβ cell signalling superfamily, underlie the majority of hereditary PAH cases2 but have also been identified in other disease subtypes including idiopathic PAH and PAH associated with other disorders.3 ,4 Mutations in two further receptor members of the TGFβ signalling superfamily have been identified as uncommon causes of hereditary PAH. Indeed, hereditary haemorrhagic telangiectasia (HHT) is an autosomal dominant vascular disease, caused by heterozygous mutations of either TGFβ type I receptor activin-like kinase-type 1 (ALK1) or of the endoglin gene (ENG).5 A small proportion of HHT patients have PAH that is clinically and histopathologically indistinguishable from other heritable forms of PAH. In rare cases, mutations of ALK-1 appeared to cause idiopathic PAH and hereditary PAH without HHT.4 Therefore, BMPR2, ALK1 and ENG genes, belonging to the TGFβ superfamily, represent good candidates for the study of genetic susceptibility to SSc–PAH.
Few studies have attempted to identify SSc–PAH genetic risk factors. Despite some relevant preliminary results, a lack of appropriate cohorts (because of imperfect phenotype and/or insufficient statistical power) has precluded definite conclusions.6–9
The aim of this study was to investigate a specific genetic basis favouring the occurrence of PAH in SSc, using a synergistic strategy combining direct sequencing together with genotyping of common variants of candidate genes encoding four TGFβ receptors: BMPR2, ALK1, ENG and TGFBR2.
Patients and methods
Study population
All SSc patients and controls were of European Caucasian origin and were provided through the European League Against Rheumatism Scleroderma Trials and Research group (EUSTAR) centres. The discovery set consisted of 824 SSc patients, including 54 SSc–PAH patients and 939 controls from French centres. The replication set consisted of cohorts from other French centres (175 SSc patients including 75 SSc–PAH patients and 438 controls), northern/central Europe (455 SSc, 59 SSc–PAH and 1823 controls), Italy (542 SSc, 33 PAH–SSc and 479 controls) and eastern Europe (344 SSc, 52 SSc–PAH and 389 controls). The characteristics of the study population have been detailed previously.10 Clinical data collected included age, sex, disease duration and cutaneous SSc subtypes according to the definition of Leroy et al.11 Precapillary PAH diagnosis was established following right heart catheterisation as recommended.12 All local institutional review boards approved the study, and written informed consent was obtained from all study subjects.
Direct sequencing
As a first approach, 26 French Caucasian individuals (10 SSc–PAH patients, nine SSc patients without PAH and seven healthy controls) were sequenced for the candidate genes encoding four TGFβ receptors: BMPR2, ENG, ALK1 and TGFBR2. Genomic DNA was extracted from blood samples (Qiagen, Courtaboeuf, France). PCR primers were designed using Primer 3 to amplify segments of genomic DNA that included the exons and intron–exon boundaries to detect splice-site variants. PCR products were purified using ExoSAP-IT (USB Corp., Cleveland, Ohio, USA) and sequenced using the BigDye Terminator V.3.1 cycle sequencing kit (Applied Biosystems, Foster City, California, USA).
Genotyping
As a second approach, any single-nucleotide variation detected by direct sequencing in SSc–PAH patients and not present either in controls or PAH-free SSc patients, was tested for association in the genotyping cohort. In addition, Tag single-nucleotide polymorphisms (SNP) with a minor allele frequency (MAF) greater than 5% were genotyped for each of the four TGFβ receptor genes using the KASpar genotyping system (KBioscience, Hoddesdon, UK) as previously described.13 Six SNP of the BMPR2 gene (rs7600694, rs1061157, rs1048829, rs6747756, rs1980153, rs16839127), seven SNP of the TGFBR2 gene (rs377626, rs1841528, rs2372092, rs3773661, rs9867701, rs11466531, rs11466536), four SNP of the ALK1 gene (rs706815, rs772003, rs2277382, rs3782479) and five SNP of the ENG gene (rs35400405, rs1998923, rs1330684, rs10987746, rs17557600) were chosen according to linkage disequilibrium structure. The average genotype completeness for these variants was above 97% for both the SSc and the control samples.
Statistical analyses
Statistical analyses were performed as previously described.13 The Bonferroni correction was applied for all tests performed for SNP marker association with the disease (p value multiplied by n SNP). The analysis of combined data was performed by calculation of the pooled OR under a fixed-effects model (Mantel–Haenszel meta-analysis). No power calculation can be provided for mutation investigations, but regarding common SNP (MAF >5%) and for ALK1 rs2277382 in particular, the combined sample provides a power of 99.9% to detect an association with SSc and of 52.3% for the SSc–PAH subset, with an OR of 1.5.
Results
Sequencing of TGFβR genes in cases and controls
No mutation was identified through the sequencing of 38 SSc and 14 control chromosomes (table 1). We identified 17 polymorphisms: 13 SNP listed in public databases and four variants not yet repertoried, none of which were mutations as they were found both in patients and controls. Two variants emerged as interesting candidates for further study. Indeed, the SNP located at codon 14 of exon 1 in the ENG gene, known as rs35400405, introduces an in-frame ATG sequence downstream of the usual initiating codon that might result in the loss of the first amino acids. This variation was detected in SSc–PAH patients only and none of the controls. The rs2277382 SNP of the ALK1 gene was found only in three SSc–PAH patients, none of the PAH-free patients and none of the controls. This SNP is located in the promoter of the gene, and has previously been identified as a single-nucleotide variation associated with HHT.14
Variants identified by direct sequencing of the ENG, ALK1, TGBR2 and BMPR2 genes
Association testing of identified variants and common tag SNP of the TGFβR genes
Discovery set
We investigated the possible association of polymorphisms in the ALK1, TGFBR2, BMPR2 and EGN genes with SSc and the SSc–PAH subtype by genotyping rs35400405, rs2277382 and 20 tag SNP distributed throughout these genes (table 2). No association was found with SSc in the discovery set. Genetic association was solely observed between the SSc–PAH subset and the ALK1 rs2277382 SNP (OR 2.13, 95% CI 1.24 to 3.65, padj = 0.0066).
Association study of ALK1 rs2277382 and ENG rs35400405 SNP in the French discovery cohort
Replication set
Following the results obtained in the discovery set, we selected the ALK1 rs2277382 SNP to be investigated in the EUSTAR replication cohort (table 3). Genotype frequencies of the rs2277382 variant were in Hardy–Weinberg equilibrium in all control populations. However, we did not observe any association between the rs2277382T allele and either SSc–PAH or the SSc subset in these replication sets.
Association study of ALK1 rs2277382 with SSc and PAH–SSc in the second set of European Caucasian populations and combined analysis including the discovery and replication cohorts
Meta-analysis in the European Caucasian population
Meta-analysis of the combined discovery and replication populations (French, northern European, Italian and eastern European) including a total of 2340 SSc patients, 273 SSc–PAH and 4068 controls did not provide evidence for an association between ALK1 rs2277382 and neither SSc–PAH nor SSc.
Discussion
Genes encoding TGFβ receptors have been identified as major susceptibility genes in familial and idiopathic forms of PAH. Understanding the genetic differences between idiopathic PAH and SSc–PAH, and also between patients with SSc who do and do not develop PAH, may improve our ability to develop genetic biomarkers of SSc–PAH. This may help to identify these patients earlier in the disease course and to risk stratify patients in order to optimise the management of this devastating condition.
So far, preliminary studies investigating BMPR2 and ALK1 have failed to identify variants associated with SSc–PAH by a direct sequencing strategy. However, they were limited by small sample size and heterogeneous definition of PAH.7–9 Furthermore, an insertion in intron 7 of the ENG gene (6bINS) was reported to be negatively associated with the occurrence of SSc–PAH in a previous work from our group in a small cohort of 280 SSc patients including 29 with PAH and 140 controls.15 However, until now this result has not been replicated in a larger cohort.
In this study, the ALK1 rs2277382 and ENG rs35400405 SNP were of particular interest because they were detected only in SSc–PAH patients by direct sequencing, the hypothesis being that their minor alleles could be associated with the development of PAH in our cohorts. However, no association was found between these polymorphisms and both the complication that is SSc–PAH and also SSc. This does not rule out the possible implication of other TGFβ signalling pathway genes. Indeed, mutations in the SMAD genes have recently been identified in PAH patients and could represent another potential candidate to take into account in the genetics of SSc–PAH in further studies.16 Furthermore, another limitation may come from the fact that some SSc patients may develop PAH later during the course of the disease.
In conclusion, this study was conducted using a synergistic strategy combining direct sequencing for the identification of potential mutations or rare variants, and genotyping of common variants in a large sample including a replication step. These analyses demonstrate the lack of association between these TGFβ receptor gene polymorphisms and SSc–PAH.
Acknowledgments
The authors thank the European League Against Rheumatism Scleroderma Trials and Research group (EUSTAR) for facilitating the DNA collection and supporting the project, the KORA S4 study and HYPERGENE consortium for providing data, respectively, from German and Italian controls, the French members of the GENESYS Consortium (Patrick Carpentier (Grenoble), Jean Sibilia (Strasbourg), Jean Cabane (Paris), Luc Mouthon (Paris), Camille Frances (Paris), Zahir Amoura (Paris), Anne Cosnes (Créteil)). The authors also thank Dr J Benessiano and Prof B Grandchamp (Centre de Ressources Biologiques, Hôpital Bichat, Etablissement Français du Sang (Paris), for their assistance in setting up the French Caucasian control sample.
References
Footnotes
-
Contributors All authors contributed substantially to the conception and design, to the drafting of the article and the final approval of the submitted manuscript.
-
Funding This study was supported by the Association des Sclérodermiques de France, INSERM, Agence Nationale pour la Recherche (grant R07094KS). YA is the recipient of an investigator-initiated research grant from Pfizer.
-
Competing interests None.
-
Patient consent Obtained.
-
Ethics approval All local institutional review boards approved the study.
-
Provenance and peer review Not commissioned; externally peer reviewed.