A Closer Look at Type 1 Diabetes Causal Variants

By Allison Proffitt

September 7, 2021 | Stephen Rich’s group at the University of Virginia along with John Todd’s team at the University of Oxford, have been tracing the genes responsible for type 1 diabetes for years. In 2009 the team conducted the largest GWAS metanalysis of type 1 diabetes, identifying 60 loci that likely influenced incidence. Since then, they’ve used increasingly sophisticated technologies to better understand the genetic underpinnings of this complex disease.

But a big problem in their work, Rich said, was not the technology but the sample set. The samples used in the GWAS and other studies have been predominantly of Northern European ancestry. We needed to diversity the ancestry of our sample set, Rich explained.

In the group’s latest study—led by Catherine Robertson at the University of Virginia and Jamie Inshaw at Oxford and published in Nature Genetics in June (DOI: 10.1038/s41588-021-00880-5)—researchers looked at genetic data from 61,427 participants. “We really used a multi-ethnic strategy,” Rich said. The majority were still of Northern European ancestry, but the group also included African, Finnish, East Asian, and other ancestries. The group included T1D cases, trios (an affected child and parents), and controls.

Rich said the team was purposeful in diversifying the dataset. For the earlier studies, samples were collected through the Type 1 Diabetes Genetics Consortium (T1DGC), a part of the National Institute of Diabetes and Digestive and Kidney Diseases. For this larger, more diverse sample set, the team sought samples of Mexican and African ancestry from within the databases. They also developed international relationships with collaborators from around the world. Even those who could only share a few cases were helpful, Rich said.

The team gathered DNA as well as phenotypic data: age of diagnosis and duration of disease. The team also established B cell lines and collected additional serum and plasma. With the ImmunoChip, the team was able to use the genetic data in a much more refined way, Rich said. And the approach yielded some findings.

Expanded Regions of Interest

The team identified 64 T1D-associated regions outside the major histocompatibility complex, including 24 new regions associated with T1D at genome-wide significance, the authors write. The vast majority of T1D mutations are enriched in DNA regulatory regions, Rich explained. In total, the group identified 78 regions independently associated with T1D. Compared to previous GWAS studies, that represents 36 new regions.

The researchers also used Open Targets from EMBL-EBI to “cross-tabulate” the regions of interest with known drug targets. They found a number of variants associated with T1D risk that are also associated genetically with increased or decreased risks for other autoimmune diseases, Rich explains. For example, he said, T1D is most similar to rheumatoid arthritis and juvenile diabetes. This suggests that drugs developed to treat those diseases might work on the same gene products.

But the goal is not just treating the disease. Rich cast a vision of early identification of who is at risk for developing T1D—not just children, but adult-onset illness as well. Already there is some evidence that a monoclonal antibody can delay the onset of T1D. Better stratifying who is at risk will let researchers more strategically test therapies for delayed onset.

Rich and his academic teammates will leave the drug identification to others who are developing therapeutic targets. Instead, he said his collaborators plan to continue to collect diverse T1D samples and focus on whole genome sequencing for about 3,500 individuals of diverse ancestry. They also plan to continue performing ATAC-seq to develop high-resolution maps of accessible chromatin. And new collaborations with the Children’s Hospital of Philadelphia and Johns Hopkins University plan to explore spatial genomics and its impact on T1D.