Trends in Genetics
ReviewRevealing the architecture of gene regulation: the promise of eQTL studies
Section snippets
The study of gene expression phenotypes
Variation in gene expression is abundant in all organisms studied to date 1, 2, 3. It has been suggested repeatedly that modifications in gene regulation are responsible for much of the observed phenotypic variation in natural populations. Indeed, like substitutions at the protein level, changes in gene regulation have been found to underlie numerous adaptive phenotypes in a variety of organisms, from beak morphology in Darwin finches [4], bristle number, wing pigmentation and trichome patterns
Variation in gene expression is widespread in human populations
Gene expression levels measured by microarrays can be affected by many nongenetic factors, including environmental variation, epigenetic modifications and random fluctuations in expression, as well as by experimental issues including measurement error, staging and (for many of the studies) variation that arises in transformed cell lines [21]. For that reason, it was not clear initially how heritable these measured expression levels would be, and early eQTL studies spent considerable effort
Cis and trans regulation of gene expression
A common observation, from a variety of both linkage and association eQTL studies, is that numerous genes have proximal eQTLs, likely in cis regulatory elements (see Box 2 for a discussion of the definitions of cis and trans). For example, Stranger et al. [28] measured gene expression in the transformed lymphoblastoid cell lines that were prepared by the International HapMap Project [29]. Using 2.2 million common SNPs genotyped by the HapMap to test for association in 210 of these cell lines
Mapping in model organisms: evidence for eQTL hotspots
Genome-scale eQTL mapping studies in nonhuman organisms have predominantly focused on three objectives: (i) to identify QTLs associated with variation in transcript abundances in defined mapping populations and categorize them as proximal or distal to the locus of the transcript they affect, (ii) to determine the numbers, genomic distributions and magnitudes of eQTL effects on transcript levels and (iii) to evaluate whether eQTLs interact additively to control transcript levels. Despite
Beyond eQTL screens in model organisms?
One of the long-term goals of eQTL studies in model organisms is to delineate which eQTLs regulate metabolic and developmental pathways [49] – not an easy task, given that gene expression is often a polygenic trait with genes of both major and minor effects. Furthermore, transcriptional networks can be plastic 50, 51 and, as a consequence, eQTLs can be extremely context dependent, differing because of a large number of factors, such as temperature [41], sex [38], developmental stage and tissues
Concluding remarks and future perspectives
The considerable advances in expression quantitative trait loci (eQTL) studies notwithstanding, there are still open questions about the biology and applications of eQTL mapping. First, there are important technical questions about the extent to which eQTLs are replicated across independent samples and independent platforms for measuring gene expression (see Box 3 for a discussion on different approaches to compare results across studies). Second, most of the human eQTL studies to date have
Acknowledgements
The authors thank J. Borevitz for helpful discussions and Sridhar Kudaravalli for helping to generate Figure 1. Y.G. is supported by NIH Grant GM077959, J.K.P. is supported by Grant HG002772 and S.A.R. is supported by NIH fellowship 5F32GM080966.
Glossary
- Expression quantitative trait loci (eQTL) hotspot
- a locus in which genetic variation is associated with the expression variation of many genes. Because the resolution of the mapping depends on the density of markers, an eQTL hotspot may reflect the presence of a single influential regulator (such as a transcription factor) or several linked loci that affect transcript levels of different genes.
- Genetic architecture of a quantitative trait
- a description of the association between variation at the
References (73)
Natural selection on gene expression
Trends Genet.
(2006)- et al.
Long-range control of gene expression: emerging mechanisms and disruption in disease
Am. J. Hum. Genet.
(2005) - et al.
The quantitative genetics of transcription
Trends Genet.
(2005) - et al.
Using DNA microarrays to study natural variation
Curr. Opin. Genet. Dev.
(2006) - et al.
Genetical genomics: the added value from segregation
Trends Genet.
(2001) - et al.
Use of unlinked genetic markers to detect population stratification in association studies
Am. J. Hum. Genet.
(1999) Variation in gene expression within and among natural populations
Nat. Genet.
(2002)Cis and Trans Regulatory effects contribute to natural variation in transcriptome of Drosophila
Mol. Biol. Evol.
(2008)Bmp4 and morphological variation of beaks in Darwin's finches
Science
(2004)Morphological evolution through multiple cis-regulatory mutations at a single gene
Nature
(2007)