Genetic disruption of serine biosynthesis is a key driver of macular telangiectasia type 2 aetiology and progression.

Publication Type
Journal Article
Year of Publication
2021
Authors
Bonelli, Roberto; Ansell, Brendan R E; Lotta, Luca; Scerri, Thomas; Clemons, Traci E; Leung, Irene; MacTel Consortium; Peto, Tunde; Bird, Alan C; Sallo, Ferenc B; Langenberg, Claudia; Bahlo, Melanie
Secondary
Genome Med
Volume
13
Pagination
39
Date Published
2021 Mar 09
Keywords
Gwas Mendelian randomisation Metabolomics Retinal disease Serine
Abstract

BACKGROUND: Macular telangiectasia type 2 (MacTel) is a rare, heritable and largely untreatable retinal disorder, often comorbid with diabetes. Genetic risk loci subtend retinal vascular calibre and glycine/serine/threonine metabolism genes. Serine deficiency may contribute to MacTel via neurotoxic deoxysphingolipid production; however, an independent vascular contribution is also suspected. Here, we use statistical genetics to dissect the causal mechanisms underpinning this complex disease.

METHODS: We integrated genetic markers for MacTel, vascular and metabolic traits, and applied Mendelian randomisation and conditional and interaction genome-wide association analyses to discover the causal contributors to both disease and spatial retinal imaging sub-phenotypes.

RESULTS: Genetically induced serine deficiency is the primary causal metabolic driver of disease occurrence and progression, with a lesser, but significant, causal contribution of type 2 diabetes genetic risk. Conversely, glycine, threonine and retinal vascular traits are unlikely to be causal for MacTel. Conditional regression analysis identified three novel disease loci independent of endogenous serine biosynthetic capacity. By aggregating spatial retinal phenotypes into endophenotypes, we demonstrate that SNPs constituting independent risk loci act via related endophenotypes.

CONCLUSIONS: Follow-up studies after GWAS integrating publicly available data with deep phenotyping are still rare. Here, we describe such analysis, where we integrated retinal imaging data with MacTel and other traits genomics data to identify biochemical mechanisms likely causing this disorder. Our findings will aid in early diagnosis and accurate prognosis of MacTel and improve prospects for effective therapeutic intervention. Our integrative genetics approach also serves as a useful template for post-GWAS analyses in other disorders.