Specificity and complexity in genetic determinants of gene expression revealed by analysis of different primary immune cell types

A major challenge in taking forward the results of recent genome-wide association studies of common disease has been how to identify specific functionally important genetic variants underlying observed disease associations. Where variants disrupt the coding DNA sequence and alter the function of a specific protein effects may be clear but increasingly scientists recognise that variants may have regulatory effects by altering levels of gene expression that can be equally important.

A new study led by Oxford University scientists has shed further light on this question while also revealing how complex these regulatory effects may be. The team used an approach called expression quantitative trait mapping to look for association between levels of expression of a particular gene and possession of specific genetic variants.

Importantly, this is one of the first studies to use the approach in highly purified primary cell populations and specifically to use paired samples of B cells and monocytes from peripheral blood samples.

The results published in Nature Genetics today are striking in a number of ways, notably the extent of cell specificity in the observed associations. This can go to an extreme whereby the same genetic variant may be associated with increased gene expression in one cell type but reduced expression in the other. This highlights how interpretation of the results of a disease association study will be dependent on investigators considering the disease relevant cell or tissue type.

The investigators defined local effects whereby variants may be acting in ‘cis’, together with the rarer situation of variants acting at distance (so called ‘trans’ effects) which may involve variants being associated with expression of a gene present on a different chromosome. For example, genetic modulation of expression of a transcription factor KLF4 was associated with differential expression of a network of other genes while variants altering expression of lysozyme were found to involve a large panel of downstream genes.

The study sheds new light on a number of previously identified genetic variants associated with diseases ranging from type I diabetes to rheumatoid arthritis and HIV infection. In particular, the team resolved how genetic variation at the highly polymorphic HLA locus was associated with trans effects and how this was related to specific disease associated HLA alleles, providing new insights into how such associations may arise.

The work was supported by the Wellcome Trust, the European Research Council and the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre.

For more information see the paper itself: http://dx.doi.org/10.1038/ng.2205

Dr Julian Knight's Research Group