If insulin can’t get into tissues, it can’t act

Developing resistance to insulin is a key step on the road to diabetes, making insulin resistance a target for treating metabolic dysfunction. Insulin orchestrates blood glucose levels by signaling fat and muscle cells to absorb the sugar and liver cells to stop pumping it out, so most research on insulin signaling focuses on these tissues. But a new study from researchers at the Max Planck Institute for Heart and Lung Research shows that insulin resistance in the endothelial cells that line blood vessels could also drive diabetes—and targeting signaling in those cells by another hormone, adrenomedullin, can improve insulin sensitivity in diabetic mice (Science 2025, DOI: 10.1126/science.adr4731).

If insulin can’t get out of blood vessels, it can’t act in its target tissues. “Endothelial cells are full of insulin receptors,” says Stefan Offermanns, who led the study. When the endothelial cells sense elevated insulin, they dilate blood vessels, increasing tissue perfusion, and may also ferry the hormone through cells themselves.

Offermanns and his team made their discovery while studying an unrelated signaling pathway. They observed to their surprise that removing a protein in that pathway changed endothelial cells’ response to insulin. They then found that when adrenomedullin activates this pathway, it reverses insulin’s effects in endothelial cells by dephosphorylating the insulin receptor.

Having worked out the tug-of-war between insulin and adrenomedullin, the researchers tested its effect on mice with endothelial cells that could not sense adrenomedullin. In animals that had already developed insulin resistance akin to human type 2 diabetes, removing adrenomedullin signaling improved systemic insulin sensitivity.

The results suggest that adrenomedullin signaling could become a new drug target for diabetes. Blocking adrenomedullin itself long term is impractical, Offermanns says, because of the hormone’s other roles. But the team suspects that blocking its interaction with a plasma protein, complement factor H, which potentiates the interaction between adrenomedullin and its receptor, might be more promising.

According to Annette Beck-Sickinger, who studies peptide hormones, including adrenomedullin, at the University of Leipzig, the study “opens up a new field.” Whether the insulin-adrenomedullin interplay has the same effects in humans remains to be seen, she says, but “the data clearly show that many hormonal interactions are not as simple as thought.”