Canonical insulin signaling is not significantly impaired in early stages of depression.

Patients with major depression (MD) are at high-risk for insulin resistance (IR), type-2 diabetes, metabolic syndrome, cardiovascular morbidity and mortality. However, our recent study published in this journal [Eur Arch Psychiatry Clin Neurosci. 2019 Jun;269(4):373-377], found no evidence of IR in acutely-ill drug-naive first-episode MD (FEMD) using the homeostatic model assessment of insulin resistance (HOMA-IR). We concluded, that MD may be related to impaired glucose/insulin homeostasis in the long-term but not in early disease stages. Now, we performed a complementary analysis of the canonical insulin signalling pathway containing the set of control and FEMD samples from the study mentioned above. The first node (pS312-IRS-1, pY-IRS-1) and downstream pathway which affects glucose and lipid homeostasis (phosphorylated proteins: pS473-AKT, pS9-GSK3β, pS2448-mTOR, pT389-p70S6K; total proteins AKT, GSK3β, mTOR, p70S6K) were analyzed by electrochemiluminescence (ECL) in neuronal extracellular vesicles (nEVs) enriched for L1 neural cell adhesion molecule (L1CAM) expression. No significant diagnosis-related differences were observed for the pS312-IRS-1 / pYIRS-1 ratio (P = 0.093), but the mean ratio was reduced by ~ 70% in FEMD versus controls. Moreover, omnibus analysis of downstream phosphorylated / total signaling protein ratios and respective post-hoc analyses revealed no significant changes in FEMD patients versus controls (P = 0.734). HAMD-21 scores were not correlated with pS312-IRS-1 / pY-IRS-1 or downstream phosphorylated/total signaling protein ratios. In summary, we did not find evidence for altered neuronal insulin signaling in early disease stages of MD. This is in contrast to schizophrenia, where we and other researchers have seen evidence of IR in first-episode patients.