21 September 2012
Manhattan plot and list of genome-wide significant PBC risk loci across Immunochip. Novel risk loci are highlighted in blue. Loci with more than one independent signal are highlighted in red. [DOI: 10.1038/ng.2395]
By Jimmy Liu
Over the last five years, genetic studies of complex diseases have found that the same genetic variant will often have a role in susceptibility across multiple diseases. Variants in genes that regulate the cell proliferation, for example, are associated in multiple types of cancers. This genetic overlap is especially evident in autoimmune diseases, which occur when the immune system falsely recognises its own cells as a foreign pathogen, and then sets about to destroy them. Well-known examples of such diseases include type 1 diabetes, inflammatory bowel disease and multiple sclerosis. By learning more about both the shared and disease-specific genetics behind these diseases, we hope to gain a greater understanding of disease etiology, and ultimately, better treatment options for patients.
An international consortium of clinicians and geneticists was formed two years ago to take advantage of the genetic overlap between autoimmune diseases, and developed the Immunochip DNA chip. The chip focuses on 186 genomic regions that have been implicated in least one of twelve immune-related diseases. These regions were first identified through individual genome-wide association studies (GWAS) of these diseases, which tested for association between single letter changes in our DNA sequence, or single nucleotide polymorphisms (SNPs), spread out across the entire genome in diseased and healthy populations.
Unlike the DNA chips used in GWAS, where, for practical purposes, only a fraction of known SNPs are tested, the Immunochip contains all the known SNPs within the 186 regions. This enables the refinement of disease associated genomic regions through both the detection of additional independent associated SNPs as well allowing us to hone in on the biologically functional SNP. The chip is also extremely cost-effective. At only $40 each (five-to-ten times cheaper than a typical GWAS chip), the Immunochip allows for studies to be conducted in up to tens of thousands of individuals, which in turn provides a powerful means to detect association at rare variants as well as enable powerful cross-disease comparisons.
Our study aimed at discovering and refining genetic regions linked to susceptibility for primary biliary cirrhosis, or PBC, an immune-mediated disease that causes the progressive destruction of liver bile ducts (Nature Genetics 2012). PBC affects approximately 1 in every 3-4,000 individuals, and there is currently no known cure. Treatment options for PBC typically involve slowing disease progression, and in extreme cases, a liver transplant may be required.
Using the Immunochip, we identified three novel genetic variants associated with PBC susceptibility, and better defined potential biologically relevant variants across these and 15 previously known genetic regions. Two of the novel SNPs are found in genes involved in the regulation of cytokines, a molecule that signals immune cells when foreign agents in the body are detected. The same variants in these genes are also implicated with susceptibility to other autoimmune diseases such as type-1 diabetes, Crohn’s disease, multiple sclerosis and Celiac’s disease, and demonstrate part of the shared etiology across these diseases.
Using dense genotyping technology (and eventually whole-genome sequencing) means that, unlike GWAS, the true functional variant that increases disease risk is likely to be among those we tested. While it is difficult to isolate this SNP among a set of highly correlated SNPs without the use of functional experiments, we can still use the data to investigate the distribution of certain functional elements among the set of SNPs that may be relevant.
To do this, we developed a method that takes advantage of the wealth of data generated from the Encyclopedia of DNA Elements project, or ENCODE. Part of this project involved identifying the regions of DNA that certain proteins, known as transcription factors, attach themselves to in order to regulate the amount of expression of certain genes. This differential regulation of genes across different cell types is one of the reasons why a single DNA sequence can produce such a variety of cells.
Our method looked at whether the SNPs we identified are more often found in such gene regulatory regions than you would expect by chance. Of the17 cell types we tested, we found that our PBC associated variants show an almost 1.5-fold enrichment for these regions in B-cells, a type of blood cell that plays a central role in the immune system, when compared to the other cell types. This suggests that gene regulation in these cells play a major role in PBC. While this result is not in itself surprising, and does not exclude the role of cells that were not tested, our method provides a guide for future experimental studies and can be readily applied to other complex diseases such as schizophrenia, where the relevant tissue of interest is still unclear.
Overall, our findings pave the way for future functional work to better elucidate the biology behind PBC susceptibility, with the hope that this will lead to the development of more effective treatment options for this and other complex diseases.
Jimmy Liu is a PhD student in the Statistical Genetics team at the Wellcome Trust Sanger Institute.more...
Jimmy Z. Liu, Mohamed A. Almarri et al (2012). 'Dense fine-mapping study identifies novel disease loci and implicates coding and non-coding variation in primary biliary cirrhosis risk'
Nature Genetics DOI: 10.1038/ng.2395