Type 2 diabetes (T2D) is a rapidly growing metabolic condition worldwide characterized by increased blood glucose levels during fasting or after food consumption. T2D is also associated with peripheral insulin resistance, which affects the skeletal muscle, liver, and adipose tissue. A recent study documented that over 500 million people live with T2D worldwide. Genetic and environmental factors influence the heterogeneous pathogenesis of T2D. Subgroup stratification and deep phenotyping enabled the identification of distinct T2D clusters associated with various clinical outcomes. This finding highlights the need to consider continuous variation in metabolic function when diagnosing and treating patients, as conventional diagnostic categories (such as T2D or normal glucose tolerance) may not fully capture the underlying biology.
Previous studies have shown that skeletal muscle is the primary tissue associated with insulin-stimulated glucose uptake and the major site of insulin resistance in T2D. Improper insulin-stimulated glucose uptake could be due to a post-receptor defect, such as insufficient recruitment of the glucose transporter 4 (GLUT4) to the plasma membrane after phosphorylation. It reduces the abundance of signaling molecules or glucose transporters in normal conditions. A comprehensive system-wide evaluation is required to develop personalized treatments to identify individual insulin signaling variations contributing to T2D heterogeneity. Although mass spectrometry-based proteomics has been significantly exploited in cancer research, few proteomics-related studies in relevant tissues related to insulin resistance have utilized this strategy.
Identifying the differences in phenotypic traits, proteome and phosphoproteome signatures, and varied responses to environmental stimuli could enable the development of personalized medicine for diabetes. A latest study used proteomics technology and deep in vivo phenotyping to map diabetogenic traits based on the skeletal muscle protein landscape of normal and diabetic individuals. Both men and women with normal glucose tolerance (NGT) or T2D were recruited. All participants were paired based on age, sex, body mass index (BMI), and smoking status. Biopsy samples were obtained from the vastus lateralis muscle of the eligible participants before and during the hyperinsulinemic-euglycemic clamp. The discovery cohort comprised 77 participants and was used to determine the molecular landscape of insulin resistance and T2D.
Of these, 34 participants were diabetic and 43 individuals had NGT. Experimental findings indicated the importance of skeletal muscle, particularly phospho-signaling, in whole-body insulin sensitivity. A variation in the proteomic landscape within the diagnosis groups was observed. Stratified proteome-phenotype associations revealed mitochondrial protein content strongly correlated with whole-body insulin sensitivity. However, mitochondrial abundance was not a distinct feature of T2D diagnosis, suggesting it reflects insulin sensitivity, not disease status. Additionally, the study newly implicated protein degradation and turnover pathways, including the proteasome and ubiquitin-mediated proteolysis, as well as Wnt and adrenergic signaling, as being negatively correlated with insulin sensitivity. This suggests altered protein turnover may contribute to insulin resistance.
In contrast, a higher abundance of glycolytic enzymes was negatively correlated with insulin sensitivity. The study also emphasized that the ratio of lactate dehydrogenase isoforms (LDHA/LDHB) and the overall stoichiometric relationships between glycolytic and oxidative phosphorylation proteins provided added insight into metabolic variation beyond individual protein abundance. A total of 118 phosphosites were found to be linked with insulin resistance in the fasted state, compared with 66 phosphosites exclusively in the insulin-stimulated state. Unexpectedly, the study found that fasting-state phosphoproteome signatures were even more predictive of insulin sensitivity than those in the insulin-stimulated state. The enrichment analysis indicated that the activation of JNK and p38 stress kinases was linked to insulin resistance.
Therefore, the JNK-p38 pathway could be a predominant driver of aberrant human skeletal muscle signaling in insulin resistance, an information that is actually known since the 2000 from earlier studies in vitro and in rats. Insulin Receptor Substrate 1 (IRS-1), indeed, was shown to be phosphorylated by either p38 or JNK1, resulting in impairment in its docking with insulin receptor but mostly with PI-3K kinase, which mediated GLUT4 receruitment to plasma membrane. Cellular assays also determined the role of protein kinase MAPKAPK2 downstream to p38 as an upstream regulator of AMPKγ3 and its phosphorylation on residue S65, crucial in regulating skeletal muscle insulin sensitivity. The AMPKγ3 S65 site was uniquely found in humans and strongly correlated with insulin resistance, suggesting it could serve as a human-specific marker or therapeutic target.
The current investigation demonstrated the complex nature of dysregulated signaling pathways in insulin resistance. Importantly, the researchers found that although there was impairment in certain signaling pathways, other components, such as c-Akt (downstream to IRS1-PI3K-PDK1) and some of its downstream substrates, remained functional even in severely insulin-resistant individuals, showing that insulin resistance does not uniformly affect all signaling nodes. The research observed distinct sex-specific differences in the proteome and phosphoproteome. While males showed higher expression of glucose metabolism-related proteins, females showed higher expression of lipid metabolism-related proteins. However, differences in kinase activity, such as CAMK2, casin kinase 2° (CK-2alpha) GSK-3alpha and mTOR signaling, also emerged, highlighting the relevance of sex as a biological variable.
This has mostly to be connected to estrogen signaling which is known to has wide connections with growth factors and insulin receptors intracellular platforms. Beside, IRS-1 expression was previously found to be regulated by estradiol in the MCF-7 human breast cancer cell line. Estradiol increased insulin receptor substrate-1 mRNA and protein levels at concentrations consistent with a mechanism involving the estrogen receptor. On the contrary, IRS-2, -3 and -4 were not sensitive to estradiol. In other cell lines, ER-alpha activation rapidly induces insulin-like growth factor receptor (IGF-1R) phosphorylation and downstream IRS-1 signaling. This is due to direct interaction of ER-alpha with tyrosine kinases like c-Src and coregulators (MNAR and Hsp90), which enhance growth factor trans-phosphorylation once they dissociate from ER-alpha.
However, in this investrigations molecular signatures of insulin resistance remained broadly similar between men and women. Despite these differences, insulin resistance-related signaling signatures were largely conserved across sexes. In conclusion, although the JNK-p38 pathway has been implicated in inflammatory responses and the development of insulin resistance and type 2 diabetes, this study identifies the JNK-p38 pathway as a predominant driver of aberrant human skeletal muscle signaling in insulin resistance. In addition, several previously undescribed insulin-activated kinases were revealed, including DCLK1, cyclin-dependent kinase (CDK)-2, -6, -16 and NEK11, indicating that they could become future target to smooth insulin resistance, since JNK or p38 targeting has already been proven difficult to manage in inflammatory situations.
- edited by Dr. Gianfrancesco Cormaci, PhD, specialist in Clinical Biochemistry.
Scientific references
Jeppe K, Stocks B et al. Cell 2025 May 27; in press.
Nikolic I et al. Nat Rev Endocrinol. 2020; 16(12):697.
Lawan A, Min K et al. Diabetes. 2018; 67(4):624-635.
Kawamoto E et al. Physiol Rep. 2016; 4(15):e12876.
Masharani UB et al. PLoS One. 2011; 6(5):e19878.
DeFronzo RA et al. Diab Care. 2009; 32:S157-S163.
Molloy CA et al. J Biol Chem. 2000; 275(17):12565.
Kahlert S et al. J Biol Chem. 2000; 275(24):18447.
