PLoS Genetics (Jan 2012)

A genome-wide association scan on the levels of markers of inflammation in Sardinians reveals associations that underpin its complex regulation.

  • Silvia Naitza,
  • Eleonora Porcu,
  • Maristella Steri,
  • Dennis D Taub,
  • Antonella Mulas,
  • Xiang Xiao,
  • James Strait,
  • Mariano Dei,
  • Sandra Lai,
  • Fabio Busonero,
  • Andrea Maschio,
  • Gianluca Usala,
  • Magdalena Zoledziewska,
  • Carlo Sidore,
  • Ilenia Zara,
  • Maristella Pitzalis,
  • Alessia Loi,
  • Francesca Virdis,
  • Roberta Piras,
  • Francesca Deidda,
  • Michael B Whalen,
  • Laura Crisponi,
  • Antonio Concas,
  • Carlo Podda,
  • Sergio Uzzau,
  • Paul Scheet,
  • Dan L Longo,
  • Edward Lakatta,
  • Gonçalo R Abecasis,
  • Antonio Cao,
  • David Schlessinger,
  • Manuela Uda,
  • Serena Sanna,
  • Francesco Cucca

DOI
https://doi.org/10.1371/journal.pgen.1002480
Journal volume & issue
Vol. 8, no. 1
p. e1002480

Abstract

Read online

Identifying the genes that influence levels of pro-inflammatory molecules can help to elucidate the mechanisms underlying this process. We first conducted a two-stage genome-wide association scan (GWAS) for the key inflammatory biomarkers Interleukin-6 (IL-6), the general measure of inflammation erythrocyte sedimentation rate (ESR), monocyte chemotactic protein-1 (MCP-1), and high-sensitivity C-reactive protein (hsCRP) in a large cohort of individuals from the founder population of Sardinia. By analysing 731,213 autosomal or X chromosome SNPs and an additional ∼1.9 million imputed variants in 4,694 individuals, we identified several SNPs associated with the selected quantitative trait loci (QTLs) and replicated all the top signals in an independent sample of 1,392 individuals from the same population. Next, to increase power to detect and resolve associations, we further genotyped the whole cohort (6,145 individuals) for 293,875 variants included on the ImmunoChip and MetaboChip custom arrays. Overall, our combined approach led to the identification of 9 genome-wide significant novel independent signals-5 of which were identified only with the custom arrays-and provided confirmatory evidence for an additional 7. Novel signals include: for IL-6, in the ABO gene (rs657152, p = 2.13×10(-29)); for ESR, at the HBB (rs4910472, p = 2.31×10(-11)) and UCN119B/SPPL3 (rs11829037, p = 8.91×10(-10)) loci; for MCP-1, near its receptor CCR2 (rs17141006, p = 7.53×10(-13)) and in CADM3 (rs3026968, p = 7.63×10(-13)); for hsCRP, within the CRP gene (rs3093077, p = 5.73×10(-21)), near DARC (rs3845624, p = 1.43×10(-10)), UNC119B/SPPL3 (rs11829037, p = 1.50×10(-14)), and ICOSLG/AIRE (rs113459440, p = 1.54×10(-08)) loci. Confirmatory evidence was found for IL-6 in the IL-6R gene (rs4129267); for ESR at CR1 (rs12567990) and TMEM57 (rs10903129); for MCP-1 at DARC (rs12075); and for hsCRP at CRP (rs1205), HNF1A (rs225918), and APOC-I (rs4420638). Our results improve the current knowledge of genetic variants underlying inflammation and provide novel clues for the understanding of the molecular mechanisms regulating this complex process.