An international team led by researchers from the University of Massachusetts Amherst and Helmholtz Munich has identified hundreds of genetic drivers of type 2 diabetes found in specific tissues rather than in blood, showing that mechanisms of the disease are highly tissue-specific and often shared across different populations. The study, published in Nature Metabolism, analyzed genetic data of more than 2.5 million people worldwide and found causal evidence implicating 676 genes across seven diabetes-relevant tissues, many of which would not have been detected using blood samples alone.
“We’ve known for some time now that tissue context is important to consider when trying to understand the mechanisms underlying the development of type 2 diabetes,” said co-senior author Cassandra Spracklen, PhD, an associate professor of epidemiology at UMass Amherst. “But this work demonstrates just how important that context truly is.”
Type 2 diabetes develops from dysfunction across multiple organs, including adipose tissue, the liver, skeletal muscle and the insulin-producing cells of the pancreas. Blood is the most accessible tissue for molecular studies, but in this study the researchers found that blood alone captures only a fraction of gene implicated in the disease. From the seven different tissues examined, only 18% of genes with a causal effect in a primary diabetes tissue were detected in a blood sample, while 85% of genes detected in diabetes-relevant tissues were absent from blood samples.
For their work, the researchers used genome-wide association data generated by the Type 2 Diabetes Global Genomics Initiative, a research consortium that has assembled one of the largest multi-ancestry genetic datasets for diabetes, which includes data from more than 700,000 people of non-European ancestry. Using these data, the team tested how genetic variants influence gene expression and protein levels, and whether those genetically predicted changes causally affect diabetes risk.
When the team examined only blood tests for diabetes, they identified causal effects of 335 genes and 46 proteins on type 2 diabetes risk. When the analysis was broadened to tissue-specific gene expression it identified causal links between the expression of 676 genes and diabetes risk. Some genes showed consistent effects across ancestry groups, while came to light only when the data from historically underrepresented populations were included.
The researchers noted their investigation was necessary to both unlock suspected hidden drivers of diabetes and to broaden the research to other populations. Genome-wide association studies (GWAS) and causal analyses have largely relied on blood samples taken from people of European ancestry, even though gene regulation varies by tissue and diabetes prevalence and presentation differ across populations. “Previous causal relationships investigated in blood might not represent biological mechanisms directly linked to the disease and might miss causal mechanisms relevant to complex diseases,” the researchers noted.
Some of the newly identified tissue-specific genes found in this analysis include BAK1, a gene involved in cell death, and newly identified genes such as CPXM1 and HIBCH, providing ripe areas for future research. Chi “Josh” Zhao, a doctoral student at UMass Amherst and co-first author said: “By revealing both shared and tissue-specific mechanisms, our findings move us closer to improving strategies for type 2 diabetes prevention and treatments that may be more effective across global populations.”
Next steps for the researchers will be to validate their findings in real-world studies, as they noted their work used a solely computational approach using existing data. To confirm the biological effects the team will look to expand tissue and protein data beyond European ancestry groups and increase resolution of their findings through single-cell analyses.