Research article

Homozygosity and risk of childhood death due to invasive bacterial disease

Emily J Lyons^1,4* , William Amos^2* , James A Berkley³ , Isaiah Mwangi² , Mohammed Shafi³ , Thomas N Williams³ , Charles R Newton³ , Norbert Peshu³ , Kevin Marsh³ , J Anthony G Scott³ and Adrian VS Hill¹

¹  The Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford OX3 7BN, UK

²  Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK

³  Centre for Geographic Medicine Research (Coast), Kilifi, Kenya

⁴  Current address: MRC Centre for Outbreak Analysis and Modeling, Department of Infectious Disease Epidemiology, Imperial College, St. Mary's Campus, Norfolk Place, London W2 1PG, UK

author email corresponding author email* Contributed equally

BMC Medical Genetics 2009, 10:55doi:10.1186/1471-2350-10-55

Published:	12 June 2009

Abstract

Background

Genetic heterozygosity is increasingly being shown to be a key predictor of fitness in natural populations, both through inbreeding depression, inbred individuals having low heterozygosity, and also through chance linkage between a marker and a gene under balancing selection. One important component of fitness that is often highlighted is resistance to parasites and other pathogens. However, the significance of equivalent loci in human populations remains unclear. Consequently, we performed a case-control study of fatal invasive bacterial disease in Kenyan children using a genome-wide screen with microsatellite markers.

Methods

148 cases, comprising children aged <13 years who died of invasive bacterial disease, (variously, bacteraemia, bacterial meningitis or neonatal sepsis) and 137 age-matched, healthy children were sampled in a prospective study conducted at Kilifi District Hospital, Kenya. Samples were genotyped for 134 microsatellite markers using the ABI LD20 marker set and analysed for an association between homozygosity and mortality.

Results

At five markers homozygosity was strongly associated with mortality (odds ratio range 4.7 – 12.2) with evidence of interactions between some markers. Mortality was associated with different non-overlapping marker groups in Gram positive and Gram negative bacterial disease. Homozygosity at susceptibility markers was common (prevalence 19–49%) and, with the large effect sizes, this suggests that bacterial disease mortality may be strongly genetically determined.

Conclusion

Balanced polymorphisms appear to be more widespread in humans than previously appreciated and play a critical role in modulating susceptibility to infectious disease. The effect sizes we report, coupled with the stochasticity of exposure to pathogens suggests that infection and mortality are far from random due to a strong genetic basis.