Beyond the pancreas, type 2 diabetes also affects several other major organs. The disease is linked with liver disease, heart disease, and can damage the eyes (1). Scientists have made extensive efforts to characterize how the disease affects these and other tissues. However, one organ that has proven more challenging to study is skeletal muscle.
“For a long time, we've known that muscle from people with type 2 diabetes just does not respond to insulin in an adequate way,” said Anna Krook, a physiologist at Karolinska Institute and co-author of a new study published in Cell that investigated how insulin resistance — a condition where cells stop responding to the blood sugar-regulating hormone — arises at the molecular level (2). The authors explored proteins linked to insulin sensitivity and resistance, and identified locations on these proteins where phosphorylation occurred, a process that activates or deactivates proteins. The authors say that stimulating phosphorylation in certain protein hotspots may also sensitize the surrounding muscle tissue to insulin, which could open the way to developing drugs that reduce insulin resistance.
The new findings relied on the use of high-throughput proteomics, which has recently made it possible to study all the proteins in a tissue sample en masse. However, getting accurate samples from muscle is difficult. The muscle proteome is incredibly lopsided, explained Atul Deshmukh, study co-author and a metabolomics researcher at the University of Copenhagen. A handful of proteins are incredibly plentiful and clog up protein analysis techniques, such as mass spectrometry. “They dominate the signal on the mass spec. That masks the detection of low-abundance proteins,” said Deshmukh.
As a result of these challenges, researchers have waited decades to attempt a proteomics study of muscle. The new work used a high-throughput analysis approach that leveraged powerful mass spectrometers and innovative computational techniques to tease out low-abundance proteins in the muscle proteome.
The researchers recruited over 120 patients with type 2 diabetes and control volunteers for their analysis. Compared to omics studies that blast aside statistical roadblocks with seven-digit sample sizes, that doesn’t sound like a lot. However, the study’s participants had to undergo a grueling analysis program. This included thigh muscle biopsies and the use of a hyperinsulinemic-euglycemic clamp, a technique that allowed Krook and Deshmukh to investigate how their patients’ muscles responded to insulin and absorbed glucose at the cost of an IV lasting up to six hours (3).
For a long time, we've known that muscle from people with type 2 diabetes just does not respond to insulin in an adequate way.
– Anna Krook, Karolinska Institute
These analyses helped the team understand how their cohort’s muscles responded to insulin on a molecular level. They placed the resulting sensitivity ratings — called M-values — on a continuum. They found that the protein most strongly linked to insulin resistance was a subunit of AMP-activated protein kinase. This protein has “been studied forever” in the metabolomics field, said Krook. Their new work showed that in humans who were highly insulin-resistant, this protein was highly phosphorylated. Interestingly, this phosphorylation site is found uniquely in humans, meaning that preclinical research in rodents or even nonhuman primates would not have detected it. The analysis also revealed that the cohort’s insulin sensitivity varied widely. Some people who didn’t have type 2 diabetes had lower insulin sensitivity than patients with the condition, and the sample revealed a wide spectrum of insulin resistance among the entire study cohort.
Krook and Deshmukh said they are currently in the process of whittling down the most promising targets from their extensive list of altered proteins in collaboration with industry. This study will act as a “springboard,” for further research in linked pathways, said Krook. “If there’s phosphorylation, or a lack of phosphorylation on a protein, it tells us that there was an enzyme upstream that didn’t do its job properly.” The authors hope that boosting or reducing phosphorylation of key molecular sites may also sensitize the surrounding skeletal muscle tissue to insulin.
Paul Franks, an epidemiologist at Lund University and Queen Mary University of London, praised the study’s “outstanding” physiological analysis, but noted that translating the findings into clinical practice would be challenging. Krook and Deshmukh acknowledged that invasive muscle biopsy is unlikely to work as a mass diagnostic test to detect insulin resistance. They are exploring diagnostic options that could use blood tests in place of the muscle biopsies to investigate linked pathways.
Franks also highlighted that the study doesn’t capture all diabetic disease mechanisms. Some patients’ relatively high muscle insulin sensitivity showed that their symptoms were likely driven by impacts on other organs and tissues, he said. Significant minorities of type 2 diabetics are not overweight or obese and have much lower levels of insulin resistance. These individuals likely have failures in other cell types that regulate insulin (4).
Importantly, these patients are more commonly found in populations in the global south. Krook and Deshmukh stated that their cohort’s ethnicity was largely white European; however, no data on this could be reported in the study, as ethnicity recording is not permitted in Sweden.
Franks said that despite these limitations, in-depth molecular analysis represents the future of the field. “That's the direction of travel that we should be heading into to really understand the elements of heterogeneity in type 2 diabetes,” he said.
References
- Tomic D. et al. The burden and risks of emerging complications of diabetes mellitus. Nat Rev Endocrinol 18, 525–539 (2022).
- Kjærgaard J. et al. Personalized molecular signatures of insulin resistance and type 2 diabetes. Cell (2025).
- Tam C.S. et al. Defining insulin resistance from hyperinsulinemic-euglycemic clamps. Diabetes Care 35,1605–1610 (2012).
- George A.M. et al. Lean diabetes mellitus: An emerging entity in the era of obesity. World J Diabetes 6, 613–620 (2015).
