Skip to main content

Analysis of tridimensional changes caused by type 2 Diabetes-Associated varianTs

Periodic Reporting for period 1 - 3D-ADAPT (Analysis of tridimensional changes caused by type 2 Diabetes-Associated varianTs)

Reporting period: 2015-06-01 to 2017-05-31

Type 2 diabetes (T2D) is a metabolic disorder that affects more than 400 million people worldwide. The prevalence of diabetes has doubled in the last 30 years becoming a serious global concern; however, the molecular mechanisms underlying this disease remain largely unknown. In practice, this means that patients with T2D are treated with drugs that try to lower their blood glucose, but do not affect the subjacent mechanisms, failing to affect the progression of the disease. Lack of molecular insight also prevents personalization of preventive or therapeutic interventions.

T2D results from the interplay between genetic, environmental and behavioral factors, such as obesity and a sedentary lifestyle. Genetic studies carried out using large datasets of T2D patients and healthy control individuals have been able to identify genetic variants that predispose to T2D risk. Interestingly, most of these susceptibility sequences do not lie on genes (the DNA regions that code for proteins), but on non-coding segments. Non-coding regions comprise around 98% of the total of our DNA, and even though they were initially dismissed as “junk DNA”, we currently know that non-coding regions contain regulatory sequences that are responsible for the activation of genes in the right tissue and the right stage during our lifetime. Not only they are essential for the normal functioning of our body, but alterations in them have also been associated with several human disorders.

Among non-coding regions, “enhancers” are regulatory elements that concentrate the highest frequency of variants that predispose to T2D. Some of these enhancers activate the expression of genes that are important for the function of insulin-producing beta cells in the pancreatic islets. Importantly, studies have shown that the presence of T2D-associated genetic variants within these enhancers can alter their function. However, in most cases, the variant-containing enhancers are very far away from their target genes in the DNA molecule, which hinders the identification of the gene that will ultimately be affecting beta-cell function.

Recent advances in the field of 3D structure have shown that chromosomes are highly compacted and organized within the nucleus of each cell, which means that regions that might be very distant in the linear space are close in 3D via the formation of DNA loops. 3D methods will thus help us understand what genes the enhancers are contacting with.

There is very limited information about 3D interactions of regulatory elements occurring in human pancreatic islets in health and in a disease condition such as T2D, therefore this project aimed at investigating the 3D chromatin structure of human islets and to use these 3D maps to identify gene targets of T2D-associated genetic variants.
During this project, high-resolution genome-wide interaction maps of human pancreatic islets using Promoter-Capture HiC (PCHiC) were produced, identifying almost 200.000 tridimensional contacts stemming from the promoters of all genes in the genome.

We verified that the vast majority of 3D interactions happen within the same chromosome, are connecting distant elements in the linear space, and connect gene promoters with enhancers. Interestingly, PCHiC recapitulates interactions identified by another chromosome conformation method called 4C-Seq.

Analysis of a manually-curated list of genetic variants associated to T2D revealed that they are significantly enriched in islet enhancers, in DNA binding sites of proteins that help mediate the 3D loops, as well as in interacting regions. We also interrogated the utility of 3D maps to identify the target genes of T2D-associated variants. Using this strategy we were able to assign target genes to 80% all regions that were associated to T2D or fasting glycaemia in the literature. Surprisingly, PCHIC assigns many of these signals to distant genes that were not reported before, highlighting the importance of using 3D maps to uncover new candidate genes that can influence T2D risk.

In conclusion, we have identified a network of enhancer-gene interactions happening in the 3D space in pancreatic islets, which highlights the complexity of the regulatory mechanisms controlling gene expression. We showed that T2D risk genetic variants are enriched in features mediating 3D loops, suggesting a potential mechanistic role of these variants in the 3D organization. We also demonstrate the utility of 3D maps in the disease-relevant tissue to identify novel transcriptional targets of disease-associated variants, providing new insight into mechanisms underlying T2D susceptibility.
So far preventive and therapeutic strategies for T2D include lifestyle recommendations, as well as a variety of drugs, none of which are known to affect the molecular mechanisms of the disease. The main consequence is that no pharmacologic therapy is known to influence the progression of T2D.

The tridimensional maps of human islets generated during this project have provided experimental evidence of existing physical links that connect genetic variants predisposing to T2D with their endogenous target genes. Some of the genes contacted by T2D variants are very far away in the linear DNA space and therefore were never predicted to be influenced by them. This project thus provides novel insights into the molecular mechanisms underpinning the disease. Furthermore, the identification of new putative causal genes of T2D susceptibility will assist efforts to develop refined therapies that involve modulation of transcription of target genes.