Numerous studies have reported an increased risk of cognitive impairment, known as “diabetes-related cognitive impairment”, in patients with diabetes mellitus (DM) (Umegaki, 2018; Liu et al., 2020). The incidence of cognitive impairment in diabetes is on the rise, with epidemiological evidence showing that patients with DM are three times more likely to develop dementia than the non-diabetic population. As a result, diabetes mellitus has become an independent risk factor for dementia, posing significant health risks and imposing substantial economic burden on patients' families. However, the exact pathogenesis of diabetes-related cognitive impairment remains unclear, and effective therapeutic measures are lacking.

Cytoskeletons have been implicated in various neurodegenerative diseases, including dementia, Parkinson's, Huntington's, ageing, and age-related diseases. Consequently, the regulation of the cytoskeleton has emerged as a new target for treating these conditions (Eira et al., 2016). While cytoskeletons have been extensively studied in diabetic retinopathy, diabetic nephropathy, and diabetic heart disease, their involvement in diabetic cognitive impairment has not been reported. Actin dynamics, which occur during cytoskeletal remodeling, are vital in maintaining neuronal morphology and long-term memory (Lamprecht, 2021). Multiple studies have established a link between the actin cytoskeleton and brain diseases characterized by dysfunctional dendritic spine morphology and memory impairment, such as Alzheimer's disease (Kommaddi et al., 2018; Pelucchi et al., 2020; Ben Zablah et al., 2020). This experiment aims to investigate the role of actin dynamics in cognitive impairment associated with diabetes.

Transient receptor potential channel 7 (TRPM7) is a calcium-permeable ion channel crucial for neuronal cell development (Ekaterina et al., 2021). Our previous research (Zhang et al., 2018) demonstrated increased TRPM7 expression in the hippocampus of mice with diabetic cognitive impairment, and silencing TRPM7 improved cognitive dysfunction. TRPM7 has been implicated in neuronal cell death through various pathways, including actin dynamics. The activation of the calcium/calmodulin-dependent phosphatase calcineurin (CaN) has been shown to have detrimental effects on neurons (Ankarcrona et al., 1996; Asai et al., 1999; See and Loeffler, 2001). In rat models of transient middle cerebral artery occlusion, CaN inhibitors FK506 and SDZ ASM 981 exhibited neuroprotective effects (Bochelen et al., 1999). Cofilin, an actin depolymerizing factor, can regulate the stability of F-actin. Inhibition of cofilin dephosphorylation partially ameliorates synaptic and cognitive deficits in animal models of Alzheimer's disease (Deng et al., 2016). TRPM7 has been found to mediate neuronal cell death through CaN, cofilin, and actin cascades in hypoxic-ischemic brain injury models (Ekaterina et al., 2021). We hypothesize that elevated TRPM7 expression in diabetic cognitive dysfunction may mediate increased actin depolymerization, leading to an imbalance in actin dynamics.

Troxerutin is a natural flavonoid with antioxidant and anti-inflammatory effects. Numerous studies have demonstrated its ability to improve cognitive dysfunction. Previous research by our group has confirmed that troxerutin can ameliorate cognitive dysfunction in type 1 diabetic rats; however, the specific mechanism remains unclear. Flavonoids have been reported to inhibit calcium overload in the literature (Feliciano et al., 2015). Bradykinin is a TRPM7 agonist, which can enhance the effect of TRPM7. If we observe that troxerutin can down-regulate the expression of TRPM7 when improving diabetic cognitive dysfunction, the effect of troxerutin is weakened and actin dynamics changes accordingly after the application of relieving peptide. We can speculate that troxerutin may decrease TRPM7 by inhibiting calcium overload, reduce actin depolymerization, increase actin stability, alleviate cytoskeleton damage, protect neurons and improve diabetic cognitive dysfunction.