Regulatory polymorphisms underlying complex disease traits

Abstract

There is growing evidence that genetic variation plays an important role in the determination of individual susceptibility to complex disease traits. In contrast to coding sequence polymorphisms, where the consequences of non-synonymous variation may be resolved at the level of the protein phenotype, defining specific functional regulatory polymorphisms has proved problematic. This has arisen for a number of reasons, including difficulties with fine mapping due to linkage disequilibrium, together with a paucity of experimental tools to resolve the effects of non-coding sequence variation on gene expression. Recent studies have shown that variation in gene expression is heritable and can be mapped as a quantitative trait. Allele-specific effects on gene expression appear relatively common, typically of modest magnitude and context specific. The role of regulatory polymorphisms in determining susceptibility to a number of complex disease traits is discussed, including variation at the VNTR of INS, encoding insulin, in type 1 diabetes and polymorphism of CTLA4, encoding cytotoxic T lymphocyte antigen, in autoimmune disease. Examples where regulatory polymorphisms have been found to play a role in mongenic traits such as factor VII deficiency are discussed, and contrasted with those polymorphisms associated with ischaemic heart disease at the same gene locus. Molecular mechanisms operating in an allele-specific manner at the level of transcription are illustrated, with examples including the role of Duffy binding protein in malaria. The difficulty of resolving specific functional regulatory variants arising from linkage disequilibrium is demonstrated using a number of examples including polymorphism of CCR5, encoding CC chemokine receptor 5, and HIV-1 infection. The importance of understanding haplotypic structure to the design and interpretation of functional assays of putative regulatory variation is highlighted, together with discussion of the strategic use of experimental tools to resolve regulatory polymorphisms at a transcriptional level. A number of examples are discussed including work on the TNF locus which demonstrate biological and experimental context specificity. Regulatory variation may also operate at other levels of control of gene expression and the modulation of splicing at PTPRC, encoding protein tyrosine phosphatase receptor-type C, and of translational efficiency at F12, encoding factor XII, are discussed.

Fig. 1

Investigation of regulatory polymorphism: allele-specific analysis of mRNA. Strategy: This approach uses the presence of an exonic transcribed SNP (shown here as G/A) to resolve the allelic origin of transcribed mRNA [18]. By analysis of mRNA from cells derived from an individual heterozygous for the marker SNP, an internally controlled system is established in which relative allele-specific gene expression can be estimated. Uses: Allele-specific quantification of mRNA is a useful in vivo approach to resolving functionally important haplotypes using transcribed marker SNPs. It provides a direct assessment of the relative abundance of allele-specific transcript in a natural chromosomal context in which the normal regulatory machinery and chromatin environment are operating. Limitations: The assay requires accurate and sensitive quantification of relative transcript abundance, typically based on primer extension methods. A major limitation is that for many genes and for the majority of haplotypes no exonic marker is present to allow resolution of transcript origin. Some information may be achieved by using intronic SNPs to study relative expression of unspliced RNA; a further approach is to use a different indirect measure of gene expression, namely phosphorylated Pol II loading by haploChIP in living cells [80]. The allele-specific density of Pol II loading can be used in the same way as transcript abundance except that as the Pol II is being measured in situ by crosslinking it to DNA, any SNP marker can be used within 2 kb 5′ or 3′ to the gene including promoter and 3′UTR SNPs which considerably expands the number of haplotypes that can be interrogated

Fig. 2

Investigation of regulatory polymorphism: assays of secreted protein. Strategy: An intuitive approach is to compare levels of protein produced from the gene of interest in individuals of differing genotype, for example, homozygous AA or BB or heterozygous AB. Uses: This approach can be highly informative where sufficient numbers of individuals are assayed on multiple occasions using appropriate controls to minimise confounding by environmental or experimental factors. Limitations: The approach is potentially confounded at many levels including environmental factors and other variables affecting the levels of expression of a gene between individuals such as differences in receptor-ligand interaction and signal transduction as well as translation and post-translational effects. Genetic variation on the compared haplotypes may confound interpretation of the differences seen with the chosen marker SNP

Fig. 3

Investigation of regulatory polymorphism: the reporter gene assay. Strategy: Reporter gene assays are a powerful approach to resolve the effects of DNA sequences on gene expression. The DNA sequence of interest, for example a promoter region spanning an SNP, is placed upstream of a reporter gene whose expression can be measured. The reporter gene construct is then inserted (transfected) into a cell and expression assayed. Uses: This robust approach allows the effects of polymorphisms to be assayed, for example, comparing the expression of reporter gene constructs differing only by the nucleotide(s) of interest. The assay is highly sensitive and with appropriate controls is reproducible and specific. Limitations: The assay is an in vitro approach as the transfected DNA sequences lack the native chromatin configuration which may be essential to accurate interpretation of the consequences of genetic polymorphism. The design of reporter constructs is critical, notably the choice of which portions of the naturally occurring regulatory regions of the gene locus within which the polymorphism is found to include in the reporter gene. Any functional effect of an SNP may also be dependent on its naturally occurring haplotype: many early studies assayed SNPs in isolation rather dissecting their naturally occurring coinherited combinations. Results of reporter gene assays are highly context specific with respect to choice of cell type to transfect; stimulus used for induction of gene expression; the mode and efficiency of transfection; and the DNA plasmid design and preparation

Fig. 4

Investigation of regulatory polymorphism: the gel shift assay of protein-DNA interactions. Strategy: The ‘gel shift’ or electrophoreitic mobility shift assay investigates protein-DNA binding. Short DNA probes corresponding to a genomic DNA sequence of interest are synthesised, annealed as a duplex and radiolabelled. These are then usually incubated with a crude nuclear extract or a recombinant protein of interest. On binding by protein to DNA the mobility of the probe on electrophoresis is retarded (‘shifted’). The nature and specificity of these complexes can then be resolved. Uses: A highly specific and sensitive assay to investigate relative binding affinities of proteins to the two allelic forms of an SNP. Specificity can be resolved using unlabelled competitor probes. The nature of retarded complexes can be investigated by UV crosslinking experiments and using antibodies to abolish or ‘supershift’ bound complexes. Limitations: This is an in vitro assay, typically used as a screening tool using crude nuclear lysates for hypothesis generation which does not define whether binding actually occurs in vivo. The DNA probes used are short and lack a native chromatin structure. The absence of flanking sequences and conformational effects may lead to discrepancies in observed binding