Proteome-wide Mendelian randomization identifies APOM and TNXB as actionable mediators of steroid-sensitive nephrotic...

Discovery of autoantibodies in steroid-sensitive nephrotic syndrome (SSNS) has transformed our understanding of SSNS as an immune-mediated disease; however, mechanisms underlying autoantibody production are unknown. Current treatments for SSNS are non-targeted and cause serious adverse effects. We sought to identify circulating proteins with a causal relationship to SSNS, which likely reflect underlying immune derangements, using an unbiased two-sample Mendelian randomization (MR) and colocalization approach to inform novel drug targets for disease.

Summary-level data from eight large independent proteome-wide association studies were used to estimate the causal effect of 1540 proteins on SSNS risk by Mendelian randomization (MR) using cis genetic determinants (protein quantitative trait loci, pQTL). Genetic colocalization, gene expression, and affinity binding analyses were performed to further investigate loci identified by MR.

Two proteins, apolipoprotein M (APOM) and Tenascin XB (TNXB), were causally linked to SSNS by MR and colocalization analysis. Increases in APOM and TNXB levels were associated with decreased risk of SSNS (p = 1.37 × 10^-5;[OR] = 0.40;95%[CI] = 0.27-0.61 for APOM, and p = 2.95 × 10^-4;[OR] = 0.49;95%[CI] = 0.33-0.72 for TNXB). Colocalization with SSNS occurred at HLA-DRB1 (98%) and HLA-DQA1 (80%) for APOM and TNXB, respectively. Binding affinity and gene expression analysis showed that APOM peptides have strong affinity for HLA-DRB1, that APOM has a biologically plausible causal relationship with SSNS, and that there are approved drugs targeting the APOM pathway.

We identified APOM and TNXB as novel blood proteins associated with SSNS in children. These proteins are promising targets for the development or repurposing of drugs as novel treatments for SSNS.