Collagen type III alpha I is a gastro-oesophageal reflux disease susceptibility gene and a male risk factor for hiatus hernia

Bengt Åsling, Johan Jirholt, Paul Hammond, Mikael Knutsson, Anna Walentinsson, G. Davidson, Lars Agréus, A. Lehmann, Maria Lagerström‐Fermér

Gut · 2009 · 67 citations · 47 references

DOIFull text

Open access

Concepts

Abstract

BACKGROUND AND OBJECTIVES: Gastro-oesophageal reflux disease (GORD) is a common gastrointestinal disorder with a genetic component. Our aim was to identify genetic factors associated with GORD. PATIENTS AND METHODS: Four separate patient cohorts were analysed using a step-wise approach. (1) Whole genome linkage analysis was performed in 36 families. (2) Candidate genes were tested for GORD association in a trio cohort. (3) Genetic association was replicated in a case-control cohort. We also investigated genetic association to hiatus hernia (HH). (4) Protein expression was analysed in oesophageal biopsies. RESULTS: A region on chromosome 2, containing collagen type III alpha 1 (COL3A1), was identified (LOD = 3.3) in families with dominant transmission of GORD, stratified for hiatus hernia (HH). COL3A1 showed significant association with GORD in an independent paediatric trio cohort (p(corr) = 0.003). The association was male specific (p(corr) = 0.018). The COL3A1 association was replicated in an independent adult case control cohort (p(corr) = 0.022). Moreover, male specific association to HH (p(corr) = 0.019) was found for a SNP not associated to GORD. Collagen type III protein was more abundant in oesophageal biopsies from male patients with GORD (p = 0.03). CONCLUSION: COL3A1 is a disease-associated gene in both paediatric and adult GORD. Furthermore, we show that COL3A1 is genetically associated with HH in adult males. The GORD- and HH-associated alleles are different, indicating two separate mechanisms leading to disease. Our data provides new insight into GORD aetiology, identifying a connective tissue component and indicating a tissue remodelling mechanism in GORD. Our results implicate gender differences in the genetic risk for both for GORD and HH.

References

47