GWA studies, gene regulation and disease; pointers for the future of medical science?

Analyses of existing genome wide association (GWA) studies have shown that 88% of disease associated polymorphisms are contained in the human non-coding genome. However, our understanding of the role of the non-coding human genome in human health, and how variation in this component can lead to disease, has seriously lagged behind our understanding of the role of genes.

Thanks to the availability of the genome sequence of multiple vertebrate species and our ability to identify functional non-coding regions by virtue of their conservation, this picture is now changing and we are gradually learning the language o f the non-coding regulatory genome. What is now required is a co-ordinated effort by computational biologists, molecular biologists, medical geneticists and pharmacologists to better understand the effects of variation on this murky, but critical part of the human genome. By combining the skills of these different disciplines it is highly likely that major contributions will be made in our ability to predict susceptibility to disease and develop novel interventions.

Diseases such as heart disease, arthritis, asthma, depression and obesity have been shown to have a strong heritable component. However, despite many decades of genetic analysis, the majority of the genetic lesions which confer susceptibility to these diseases have eluded detection. Many of these diseases are referred to as “complex” diseases because it is thought that they are caused by a combination of relatively harmless gene coding changes that collectively contribute to increased disease susceptibility. Whilst this hypothesis may hold true in many cases, evidence for the role of the non-coding genome in the progression of these diseases has been more closely considered in recent years.

The comparatively recent sequencing of the human genome has allowed the development of a non-biased genome wide association (GWA) method of screening for genetic disease associations. GWA studies promised much and it was expected that the majority of polymorphisms that associate with disease would occur within the coding sequence of genes whose functionality could be easily studied using a wide spectrum of well established biochemical techniques. However, this has not proven to be the case and as many as 88% of disease associated polymorphisms occur in regions of the genome which do not encode proteins. Because of our comparative lack of knowledge of the non-coding genome the results of these analyses have caused understandable consternation in many genetics labs.

By contrast, for those of us with a passion for understanding gene regulation these observations do not come as a great surprise but rather as further evidence to support hypotheses proposed for over 20 years. Up until very recently understanding the role of gene regulation in the development of disease was considered to be esoteric with little application to biomedical science. The main reason for this was a lack of evidence for gene regulation in the progression of disease and our relative inability to identify the major components of the regulatory systems that ensured the appropriate expression of genes. To rub salt into our wounds a number of studies on genes such as Sonic hedgehog and the DACH demonstrated that the non-coding sequences required for regulating tissue specific gene expression are often spread over many hundreds or thousand of kilobases. Thus, the task seemed impossible. However, in the middle of the present decade a number of genome sequences were made available on the internet together with computer programmes that allowed for their comparative analysis. Thus was born comparative genomics for the masses.

Since that time a number of labs, including my own, have used comparative genomics, in combination with transgenic analysis, to identify the key tissue specific regulatory regions for genes that include substance-P, galanin, NPY, CGRP, BDNF and cannabinnoid receptor 1. What is most surprising about these regulatory regions was how far they often lie from the genes they control, how well conserved they are over hundreds of millions of years and, despite their conservation, how polymorphic they can be within the human population. Identifying these regulatory elements has also allowed us to experiment on their functional relationships with one another. Thus, we now know that many promoter sequences can only respond to specific signal transduction cues in the presence of remote regulatory regions (Shanley et al 2010). In addition, these relationships are surprisingly specific which limits our ability to use generic promoters to assay the activity of novel regulatory regions and endogenous promoters must be used instead. We can also identify the signal transduction, ligand-receptor and protein-DNA interactions that modulate the activity of regulatory regions and can determine the effects of human polymorphisms and drug therapies on these relationships and modulations.

These experiments, and parallel studies being carried out in a number of similar minded laboratories, are allowing us to glimpse into the dark depths of the human non-coding regulatory genome. It is now time for different disciplines including computational biologists, molecular biologists, medical geneticists and pharmacologists to make the most of this opportunity to fully appreciate and understand the contributions of regulatory polymorphic variation to human disease susceptibility. Indeed the University of Aberdeen have made the first tentative steps in this direction by establishing a multidiscipline Centre for Gene Regulation and Disease and we gladly welcome input and collaboration from all interested scientists (http://www.abdn.ac.uk/regulatory-variation-disease/). In conclusion, we believe that GWA studies have shown us the future direction in biology where understanding how genes are regulated, and how this regulation is compromised by polymorphisms in disease, will become a major priority in terms of research and funding.

Information about Alasdair MacKenzie and his research can be found here.

Published: Tuesday, 16th November 2010 by Alasdair MacKenzie