Diabetes has become a global epidemic, with the number of cases quadrupling over the past three decades, making it the seventh leading cause of death worldwide. Notably, 90 % of these patients suffer from Type 2 diabetes mellitus (T2DM) (Zheng et al., 2018). As T2DM progresses, its complications have garnered significant attention (Ali et al., 2022). Recent reports estimate that by 2035, the global diabetic population could reach 600 million, with 20–70 % of patients experiencing cognitive decline (Infante-Garcia and Garcia-Alloza, 2019). Diabetic encephalopathy (DE), a disabling complication of T2DM, is characterized by cognitive decline, neurodegeneration, and Alzheimer's disease (AD)-like neuropathology, including neurofibrillary tangles and synaptic dysfunction (Chen et al., 2023; Shi et al., 2023). Despite advancements in understanding DE pathogenesis, effective therapeutic strategies remain elusive (Cheng et al., 2022). Current treatments primarily focus on glycemic control, yet they fail to address the underlying neurodegenerative processes (M Khan et al., 2014). This unmet clinical need underscores the urgency to identify novel neuroprotective agents with dual metabolic and neuroregulatory properties.

Previous studies have demonstrated that patients with T2DM exhibit a significant reduction in the level of BDNF (Zhen et al., 2013). Animal experiments further revealed that the decrease in BDNF levels impairs cognitive and memory abilities in offspring rats, reduces dendritic spine density, and lowers the expression of Synaptophysin (SYN) and Postsynaptic density protein-95 (PSD-95) in the hippocampus (Chen et al., 2018). These findings indicate that reduced BDNF levels are closely associated with cognitive deficits and synaptic damage in DE, partly by exacerbating inflammatory responses through dysregulation of neurotransmitter release and impaired neuroplasticity following the activation of Nuclear factor kappa-B (NF-κB) (Gao et al., 2022). In the context of hyperglycemia, sustained high glucose levels trigger NF-κB translocation into the nucleus, promoting the production of pro-inflammatory cytokines, inducing neuronal apoptosis, and exacerbating brain injury, thereby accelerating DE progression (Shi et al., 2018).

Notably, the role of Glycogen synthase kinase-3β (GSK-3β) in DE pathogenesis extends beyond tau phosphorylation. Mechanistically, GSK-3β facilitates the ubiquitination and degradation of IκB-α via phosphorylation, thereby releasing NF-κB to activate inflammatory genes (e.g., TNF-α and IL-6) within the nucleus (Hoeflich et al., 2000). Additionally, GSK-3β directly phosphorylates tau protein at key residues (e.g., Ser396 and Thr231), leading to hyperphosphorylation, neurofibrillary tangle formation, and subsequent disruption of microtubule stability (Ma and Hottiger, 2016). Conversely, NF-κB reciprocally enhances GSK-3β activity through two distinct pathways: (1) inhibition of AKT, a negative regulator of GSK-3β, and (2) direct upregulation of GSK-3β expression (Liu et al., 2025; Zaki et al., 2022). This vicious cycle between NF-κB and GSK-3β exacerbates both tau pathology and neuroinflammation, highlighting their synergistic roles in DE progression.

The prior research offers a conceptual framework for this experiment. It is plausible to infer that diabetes induces a reduction in brain-derived neurotrophic factor (BDNF) levels within the brain. The deficiency of BDNF, coupled with a hyperglycemic milieu, synergistically activates the nuclear factor-kappa B (NF-κB) signaling pathway, thereby promoting the secretion of inflammatory cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α). Further investigations have demonstrated that glycogen synthase kinase-3 beta (GSK-3β), acting as an upstream regulator of NF-κB, enhances its degradation via phosphorylation of inhibitor of kappa B alpha (IκB-α), consequently augmenting the transcriptional activity of NF-κB (Beurel et al., 2015). Simultaneously, the activation of NF-κB releases the suppression of GSK-3β by inhibiting AKT phosphorylation at Ser473, establishing a positive feedback loop that exacerbates neuroinflammation in diabetic encephalopathy (DE) and contributes to tau protein pathology. Based on these findings, we propose that the BDNF/NF-κB/GSK-3β signaling axis represents a promising therapeutic target for mitigating diabetic encephalopathy.

D-cheria-inositol (DCI) is a naturally occurring polyol abundant in buckwheat and beans, with the molecular formula C6H12O6, as depicted in Fig. 1 (chemicalbook CB0718693) (Cheng et al., 2019a). Preclinical studies have demonstrated its promising hypoglycemic and anti-inflammatory effects. Our previous work and others have shown that DCI improves glucose metabolism in T2DM models and alleviates diabetic complications such as nephropathy and hepatic steatosis (Jha et al., 2024; Luthar et al., 2021; Thomas et al., 2016). Mechanistically, DCI enhances insulin sensitivity by modulating phosphatidylinositol signaling and reducing oxidative stress (Cheng et al., 2019b). However, the role of DCI in neuroprotection, particularly in the context of DE, remains poorly understood (Bulut et al., 2023). Emerging evidence suggests that DCI may exert neuroprotective effects through its anti-inflammatory and antioxidative properties. For instance, DCI attenuates neuroinflammation in polycystic ovary syndrome (PCOS) models and improves cognitive deficits in 5xFAD mice, a transgenic model of AD (Yang et al., 2022). However, direct evidence linking DCI to DE pathogenesis is lacking. Moreover, the molecular mechanisms underlying DCI's neuroprotective effects—specifically its interaction with key signaling pathways involved in synaptic plasticity (e.g., BDNF), neuroinflammation (NF-κB), and Tau hyperphosphorylation (GSK-3β)—remain undefined. These gaps in knowledge hinder the translation of DCI into clinical applications for DE.

Here, we hypothesize that DCI alleviates DE by restoring BDNF expression, suppressing NF-κB-mediated neuroinflammation, and inhibiting GSK-3β-driven Tau pathology. Using db/db mice—a spontaneous T2DM model with robust DE phenotypes—we evaluated DCI's effects on cognitive function, synaptic integrity, and molecular pathways. Our findings reveal that DCI rescues hippocampal neurodegeneration, reduces neuroinflammation, and normalizes Tau phosphorylation, mechanistically linked to BDNF/NF-κB/GSK-3β modulation. This study bridges a critical gap in understanding DCI's neuroprotective potential and provides a rationale for its therapeutic development in DE.