Integrating Immune-Cell Transcriptomic Data with Mendelian Randomisation Reveals Novel Causal Genes for Type 2 Diabetes and Its Complications
Type 2 diabetes is a long-term metabolic disease that can damage many organs over time. Its complications, such as diabetic neuropathy, kidney disease, eye disease, and vascular problems, are a major cause of disability and reduced quality of life worldwide. Although blood glucose control remains central to management, not all complications develop in the same way, and the biological pathways behind them are still being clarified. This study used genetic evidence, immune-cell gene expression data, and advanced statistical approaches to identify genes and immune mechanisms that may causally contribute to type 2 diabetes and its complications.
Why this study matters
The immune system plays an important role in type 2 diabetes, not only in the development of insulin resistance and beta-cell dysfunction, but also in the progression of chronic complications. However, many genes are active in more than one cell type, and signals from blood or tissue samples can sometimes reflect pleiotropy, meaning one gene influences multiple traits or pathways. The researchers aimed to separate true disease-related effects from signals driven by cell type-specific biology. By doing so, they hoped to improve the fine classification of diabetic complications and help prioritise more reliable drug targets.
How the researchers studied the problem
The team combined clinical, genetic, and immune-related data to sort diabetic complications into distinct clusters. This clustering approach helped identify whether certain complications shared similar biological features or formed their own pattern. They found that diabetic neuropathy stood out as a separate cluster, suggesting that nerve damage in diabetes may have a distinct immune-related signature compared with other complications.
Next, the researchers examined 18,611 immune expression quantitative trait loci, or immune eQTLs, covering 4,487 genes. An eQTL is a genetic variant associated with how strongly a gene is expressed. Using these data, they performed Mendelian randomisation, a method that uses inherited genetic variants as natural experiments to estimate whether changes in gene expression are likely to cause disease. They also used colocalisation analysis to check whether the same genetic region appeared to influence both gene expression and disease risk, strengthening the case for causality.
Key findings from the genetic analyses
The analyses identified 425 unique genes associated with type 2 diabetes and 123 unique genes associated with its complications. These genes were linked to immune-cell-related expression changes and were supported by external validation using single-cell RNA sequencing data. Single-cell RNA-seq allows scientists to see which genes are active in specific cell types, providing a higher-resolution view of how immune cells may contribute to disease.
One important result was that the proportion of genes showing pleiotropic effects increased substantially when cell-type-related effects were taken into account. Under a classic definition of pleiotropy, 40.0% of the genes showed pleiotropic behavior. When the authors included classic pleiotropy and/or cell-type-related pleiotropy, this proportion rose to 71.1%. In practical terms, this means that many apparent genetic signals are influenced by the cell context in which the gene acts, and not all are straightforward single-pathway effects.
Reducing cell-type-related pleiotropy
To address this issue, the researchers applied six multivariable Mendelian randomisation methods. Multivariable MR can estimate the independent effect of a gene while adjusting for related cell-type signals, making the results more robust. After this adjustment, the cell-type-related pleiotropy for the strongest findings was substantially reduced. This is an important step because it helps distinguish genes that are more likely to have direct biological relevance from those whose signals are confounded by immune-cell context.
What the results suggest about diabetic neuropathy
The finding that diabetic neuropathy forms a distinct cluster is clinically meaningful. Neuropathy is one of the most common and burdensome complications of diabetes, often causing pain, numbness, balance problems, and reduced function. The study suggests that neuropathy may be driven by immune-related mechanisms that differ from those involved in other diabetic complications. This could eventually support more tailored prevention or treatment strategies, rather than treating all complications as biologically identical.
Drug target prioritisation
Beyond identifying causal genes, the team integrated clinical trial evidence with genetic evidence to rank potential therapeutic targets. Using this approach, they prioritised ten immune-related drug targets for diabetic complications. These candidates may help guide future drug development or repurposing efforts. Importantly, prioritisation does not mean a treatment is ready for routine use; rather, it marks a target as biologically promising and worthy of further laboratory and clinical validation.
Clinical interpretation
This study reinforces the idea that type 2 diabetes is not only a disorder of glucose metabolism but also a disease with a strong immune component. Immune-cell transcriptomic data can reveal how genetic regulation differs across immune cells, and Mendelian randomisation can help infer whether those differences are likely to be causal. Together, these approaches provide a more refined picture of disease biology than traditional association studies alone.
For clinicians and researchers, the most important message is that diabetic complications may not arise from a single shared mechanism. Some complications, especially neuropathy, may have distinct immune signatures. Recognising these differences could improve risk stratification, enable more targeted prevention, and support the search for therapies that address the underlying biology rather than only the symptoms or glucose level.
Limitations and caution
As with any genetic study, the findings should be interpreted carefully. Mendelian randomisation is a powerful tool, but it relies on assumptions that may not hold perfectly in every setting. The results also depend on the ancestry and characteristics of the populations represented in the source datasets. In addition, gene prioritisation based on statistical evidence does not prove that a target will be safe or effective as a drug in humans. Experimental studies, functional validation, and clinical trials are still needed.
Another important limitation is that immune biology is highly dynamic. Gene expression can change with age, treatment, infection, obesity, and disease stage. Future work will need to examine how these factors influence the identified signals and whether they translate into actionable biomarkers or therapies.
Conclusion
Overall, this study uses immune-cell transcriptomic data, Mendelian randomisation, and colocalisation analysis to uncover novel causal genes for type 2 diabetes and its complications. By accounting for cell-type-related pleiotropy, the researchers improved the reliability of their findings and highlighted immune mechanisms as important contributors to diabetic complications. The prioritised immune-related drug targets provide a useful foundation for future translational research aimed at preventing or treating the long-term complications of diabetes.