Diabetes study reveals previously overlooked genes tied to disease, pointing to new therapies

Dozens of unexpected genes are strongly linked to type 2 diabetes, new research from The Jackson Laboratory (JAX) shows.

The findings, based on a new genomic atlas of pancreatic cells from non-diabetic, prediabetic, and diabetic people, suggest the disease depends on expression of genes key to cell death and vitamin A metabolism.

Published in The EMBO Journal, the work provides the most detailed view yet of molecular changes that could be driving a disease affecting more than 500 million people worldwide. It may also help inform therapeutic strategies to preserve pancreatic cell function.

“This is a huge problem. Almost everyone knows someone affected by diabetes,” said JAX Professor Michael L. Stitzel, a human geneticist who directed the research. “We really have to think about the fact that there are many ways the disease can arise, which means there should also be many ways to target it. As a gene hunter, identifying the genes that contribute to disease is exciting because it gives us actionable information for precision medicine.”

By analyzing hundreds of thousands of cells and pinpointing genes that behave differently in diseased versus healthy states, the team identified links between genetics and how pancreatic cells respond to stress as diabetes progresses.

The study focused on human pancreatic islets. These cell clusters contain different types of cells that produce insulin, an important hormone the pancreas releases into the bloodstream to convert food into energy and manage glucose levels. Diabetes develops when the body can’t produce insulin (type 1) or when insulin production and response stop working properly (type 2).

A new view of diabetic gene expression

The researchers analyzed nearly 250,000 islet cells from 48 human donors, characterizing each cell and identifying 14 distinct cell types. The team set out to learn how changes in cells’ gene expression or activity—which works like turning the volume down on a stereo rather than shutting it off—affects the glucose-regulating abilities of the islets.

The analysis shows people with T2D lose about 25% of their beta cells, the islet cells primarily responsible for insulin production. A subgroup of these cells enters a senescent (aged and less functional) state, the study confirmed.

“We originally thought we would see alpha or delta cells change and impinge on beta cells,” Stitzel said. “Surprisingly, this all looks like it’s pretty beta cell specific.”

The team combined and compared multiple datasets from observational studies and genome-wide association studies and discovered 511 genes that were more or less active in beta cells from people with T2D. Genetic, protein, and metabolic analyses in mouse models narrowed that list to 58 genes that may contribute to beta cell defects linked to T2D. This integrated approach helped the team distinguish signals that are merely associated with disease from those more likely to play a causal role.

Further experiments showed many of these genes could trigger increased beta cell death, suggesting they may play a direct role in the loss of glucose-regulating cells seen in people with the disease.

Overlooked therapeutic targets?

The team also discovered the genes GRAMD2B and PDZK1 appear to help maintain beta cell mass. A gene involved in insulin release, GRAMD2B levels were consistently lower in T2D. Deleting the same gene in mice led to glucose management problems, making it a top suspect for further studies in T2D. While deleting PDZK1 in mice did not similarly disrupt glucose management, the team showed that reducing PDZK1 levels increased cell death in human islets.

“No one had previously shown these genes were relevant to type 2 diabetes,” said Khushdeep Bandesh, a staff scientist in the Stitzel lab who co-led the study.

Additionally, beta cells in T2D lowered the activity of a set of genes involved in vitamin A metabolism. These genes help convert dietary vitamin A into a molecule called retinoic acid, which helps beta cells survive stress. When the genes are less active, beta cells become more fragile and may die sooner. This points to previously underappreciated biology.

“This vitamin A pathway had been loosely linked to diabetes before, but this is the first time we’ve been able to clearly show at the gene level that it is systematically changing in type 2 diabetes,” Bandesh said. “What we don’t know, and what only careful clinical studies can answer, is whether interventions such as vitamin A supplementation would be helpful.”

The team has made their dataset available to the scientific community, reflecting their commitment to accelerating discovery by enabling scientists worldwide to build on foundational genomic resources. The next step is to expand it. Pancreatic islets are made up of several interacting cell types, and understanding how they communicate with one another is also crucial.

“This isn’t a closed‑ended study,” Bandesh said. “It’s designed as an open resource, and multiple researchers in several countries are already taking our data and building on it. That tells us this atlas is already helping move the field forward.”

Reference:

Bandesh, K., Motakis, E., Nargund, S. et al. Deep single-cell decoding of human pancreatic islets reveals T2D β-cell gene expression defects. EMBO J (2026). https://doi.org/10.1038/s44318-026-00744-w