Hunting Down Culprit Genes

By Amanda Goh 

 

June 7, 2013 | SINGAPORE—At the 2013 Bio-IT World Asia conference, Professor Paul Tam described his efforts to investigate the genetic causes of rare congenital diseases, because, “any further improvement on existing medical treatment is hampered by the lack of basic understanding of how diseases arise.” 

 

Advances in our understanding of basic biology would allow us to devise new strategies for disease management, the pro-vice chancellor and the vice president of the University of Hong Kong asserted. “Until and unless we can find the real cause, we might be mistreating or misdirecting our efforts.”

 

Hirschsprung’s disease (HSCR) provides an example of such misguided treatment. These patients have no neurons in the lower part of their intestine, resulting in the constant contraction of muscles in that region. The resulting constriction causes the dilation of the upper part of the digestive tract, which was mistakenly the primary focus of previous treatment strategies.   

 

As the Chair of Pediatric Surgery, Tam focuses on HSCR, biliary atresia and anorectal malformation, all of which arise from birth defects in the digestive system. Although these illnesses are relatively rare, they provide models of disease that enable us to better understand the underlying basic science. 

 

 

Genetic disorders may arise from alterations in single genes or in a combination of multiple genes. They may also be caused by gross chromosomal changes leading to copy number variations (CNVs). These different chromosomal abnormalities can all be identified by genome-wide association studies (GWAS). This method thus casts a broader net than linkage analysis or a candidate gene approach. 

 

GWAS analyzes the frequency of common genetic markers (small nucleotide polymorphisms, or SNPs). The markers occurring more frequently in patients than controls are said to be disease-associated and indicate the “guilty gene(s)”. GWAS was validated by its successful association of HSCR with the RET gene, which had been previously identified using a candidate gene approach. 

 

To verify that a gene is a new susceptibility locus, the association between that gene and the disease must be observed in a replication study using a second, independent population.  

 

The next step is to ask whether there is a plausible role for the gene in pathogenesis, which would be supported by its expression pattern or biochemical function being consistent with the disease pathology. Further validation experiments may be performed with mouse models of the disease or by analyzing human patient information, including their genetic data. 

 

To determine how the variants affect gene function, fine mapping can be used to specify their precise locations. For instance, GWAS linked HSCR to the NRG1 gene, and further analysis put the variants in the region of the gene controlling its expression. Comparative genomics verified the conservation of that region across different species and thus its importance. 

 

Such validation studies are crucial to correctly identify the disease-associated gene. For example, GWAS of biliary atresia implicated the ADD3 and XPNPEP1 genes, which fine-mapping and gene expression analysis subsequently narrowed down to ADD3. 

 

 

GWAS requires the recruitment of large numbers of patients and generates massive amounts of data. The bioinformatic and analytical support required for data interpretation poses a major challenge to the use of this technology. 

 

Furthermore, the genetic analysis for GWAS is performed using chips that do not cover the entire genome. The disease-causing mutation or variant may be in an omitted region or a noncoding region. 

 

Technologies that may serve as alternatives or complements to GWAS are whole genome sequencing or whole exome sequencing. Whole genome sequencing can reveal chromosomal differences even in noncoding regions, and it requires the participation of fewer patients. However, whole exome sequencing, which focuses only on the coding regions, may be more pragmatic and is in fact currently underway for HSCR.   

 

The genetic heterogeneity of the patient population may mask interesting results. For example, GWAS data of anorectal malformation did not identify any disease-linked variants, although several genes were implicated by CNV and validated by data from gene expression analysis and mouse models. Genetic heterogeneity may be overcome by increasing the number of patients in the study. Given the rarity of these diseases, collaborations that pool their patients and expertise, such as the Hirschsprung Disease Consortium, are thus vital for research to progress.  

 

“The association of a genetic marker with a disease will stimulate new ideas about the causes and biology of the condition under study,” said Tam, “but how do we make the best use of the data we have? How do we make people with different expertise work towards a common goal?” It is only through collaboration by geneticists, bioinformaticians and physicians that the culprit genes will be found and appropriate therapies developed.