A low-frequency variant in a lincRNA doubles the risk of pneumococcal bacteraemia in Kenyan children. A. Rautanen1, M. Pirinen1, T. C. Mills1, S. J. Chapman1, V. Naranbhai1, J. A. Scott2,3,4, T. N. Williams2,3,4, P. Donnelly1,5, A. V. S. Hill1,6, C. A. Spencer1, The Wellcome Trust Case Control Consortium 2 1) WTCHG, University of Oxford, Oxford, United Kingdom; 2) Kenya Medical Research Institute, Wellcome Trust Research Programme, Kilifi, Kenya; 3) Nuffield Department of Clinical Medicine, University of Oxford, Oxford, United Kingdom; 4) INDEPTH Network, Accra, Ghana; 5) Department of Statistics, University of Oxford, Oxford, United Kingdom; 6) The Jenner Institute, University of Oxford, Old Road Campus Research Building, Oxford, United Kingdom.

   Bacteraemia (bacterial bloodstream infection) is a major cause of illness and death globally, and especially in sub-Saharan Africa, but little is known about the genetic basis of individual susceptibility. We conducted a genome-wide association study of bacteraemia susceptibility using 893,635 genotyped autosomal markers in 1536 children with blood culture-proven bacteraemia and 2677 healthy infants living in Kilifi District in Kenya. We performed whole-genome imputation resulting in nearly 11 million SNPs that passed our QC filters. These were analysed both with bacteraemia overall and with the largest subphenotype, bacteraemia caused by Streptococcus pneumoniae (pneumococcus). We verified the genotypes at the most associating SNPs by direct genotyping, and replicated the results in a further 434 cases and 1366 controls. This confirmed a previously reported strong association in the HBB locus including the sickle cell disease-causing polymorphism rs334 (P discovery = 4.09x10-11; P replication = 7.7x10-4); the association was driven by a strong homozygote risk (P combined = 2.66x10-12; OR = 4.93, 95% CI = 3.15-7.70) and relatively modest heterozygote protection (P combined = 4.67x10-3; OR = 0.77, 95% CI = 0.65-0.92). More importantly, we identified a novel association in a lincRNA gene (P discovery = 3.58x10-7; P replication = 1.16x10-3; P combined = 1.69x10-9; OR combined = 2.47, 95% CI = 1.84-3.31) with pneumococcal bacteraemia. Although the function of this gene is unknown, it has been reported to be expressed solely in placenta and white blood cells. We further studied its mRNA expression in different white blood cell types, and detected expression exclusively in neutrophils that are known to be important in killing the invading pneumococcus. Interestingly the susceptibility allele is derived rather than ancestral, occurs at low frequency (2.7% in controls and 6.4% in cases), and shows evidence for variation in the susceptibility it confers to different bacterial infections. Understanding the molecular mechanisms leading to the doubled risk of pneumococcal bacteraemia associated with this allele could provide new clues in the pressing search for new therapeutic targets. The importance of lincRNAs as key regulators of gene expression has only recently been recognised and our finding will further focus research on the role of lincRNAs in human disease.

You may contact the first author (during and after the meeting) at