PLoS Genetics (Jan 2012)

Geographic differences in genetic susceptibility to IgA nephropathy: GWAS replication study and geospatial risk analysis.

  • Krzysztof Kiryluk,
  • Yifu Li,
  • Simone Sanna-Cherchi,
  • Mersedeh Rohanizadegan,
  • Hitoshi Suzuki,
  • Frank Eitner,
  • Holly J Snyder,
  • Murim Choi,
  • Ping Hou,
  • Francesco Scolari,
  • Claudia Izzi,
  • Maddalena Gigante,
  • Loreto Gesualdo,
  • Silvana Savoldi,
  • Antonio Amoroso,
  • Daniele Cusi,
  • Pasquale Zamboli,
  • Bruce A Julian,
  • Jan Novak,
  • Robert J Wyatt,
  • Krzysztof Mucha,
  • Markus Perola,
  • Kati Kristiansson,
  • Alexander Viktorin,
  • Patrik K Magnusson,
  • Gudmar Thorleifsson,
  • Unnur Thorsteinsdottir,
  • Kari Stefansson,
  • Anne Boland,
  • Marie Metzger,
  • Lise Thibaudin,
  • Christoph Wanner,
  • Kitty J Jager,
  • Shin Goto,
  • Dita Maixnerova,
  • Hussein H Karnib,
  • Judit Nagy,
  • Ulf Panzer,
  • Jingyuan Xie,
  • Nan Chen,
  • Vladimir Tesar,
  • Ichiei Narita,
  • Francois Berthoux,
  • Jürgen Floege,
  • Benedicte Stengel,
  • Hong Zhang,
  • Richard P Lifton,
  • Ali G Gharavi

DOI
https://doi.org/10.1371/journal.pgen.1002765
Journal volume & issue
Vol. 8, no. 6
p. e1002765

Abstract

Read online

IgA nephropathy (IgAN), major cause of kidney failure worldwide, is common in Asians, moderately prevalent in Europeans, and rare in Africans. It is not known if these differences represent variation in genes, environment, or ascertainment. In a recent GWAS, we localized five IgAN susceptibility loci on Chr.6p21 (HLA-DQB1/DRB1, PSMB9/TAP1, and DPA1/DPB2 loci), Chr.1q32 (CFHR3/R1 locus), and Chr.22q12 (HORMAD2 locus). These IgAN loci are associated with risk of other immune-mediated disorders such as type I diabetes, multiple sclerosis, or inflammatory bowel disease. We tested association of these loci in eight new independent cohorts of Asian, European, and African-American ancestry (N = 4,789), followed by meta-analysis with risk-score modeling in 12 cohorts (N = 10,755) and geospatial analysis in 85 world populations. Four susceptibility loci robustly replicated and all five loci were genome-wide significant in the combined cohort (P = 5×10⁻³²-3×10⁻¹⁰), with heterogeneity detected only at the PSMB9/TAP1 locus (I² = 0.60). Conditional analyses identified two new independent risk alleles within the HLA-DQB1/DRB1 locus, defining multiple risk and protective haplotypes within this interval. We also detected a significant genetic interaction, whereby the odds ratio for the HORMAD2 protective allele was reversed in homozygotes for a CFHR3/R1 deletion (P = 2.5×10⁻⁴). A seven-SNP genetic risk score, which explained 4.7% of overall IgAN risk, increased sharply with Eastward and Northward distance from Africa (r = 0.30, P = 3×10⁻¹²⁸). This model paralleled the known East-West gradient in disease risk. Moreover, the prediction of a South-North axis was confirmed by registry data showing that the prevalence of IgAN-attributable kidney failure is increased in Northern Europe, similar to multiple sclerosis and type I diabetes. Variation at IgAN susceptibility loci correlates with differences in disease prevalence among world populations. These findings inform genetic, biological, and epidemiological investigations of IgAN and permit cross-comparison with other complex traits that share genetic risk loci and geographic patterns with IgAN.