Connecting the Dots: Neurobiological Interplay Between Type 2 Diabetes and Alzheimer’s Disease

1. Introduction

Diabetes Mellitus (DM) is a metabolic disorder that accounts for 90% - 95% of all diabetes cases and is considered one of the most significant public health concerns (Burillo et al., 2021). The global prevalence of DM in 2019 was 9.3% (463 million people), and it is projected to rise to 10.2% (578 million people) by 2030 and 10.9% (700 million people) by 2045 (Takeishi et al., 2021). DM is characterized by disrupted insulin secretion and/or function, leading to persistent hyperglycemia and insulin resistance (IR) in peripheral tissues (Bakker et al., 2009; Saini, 2010).

In addition to its metabolic complications, DM has been increasingly associated with cognitive dysfunction, including impairments in executive functions, attention, memory, and processing speed (G. J. Biessels & Despa, 2018; Hameed et al., 2015; Sebastian et al., 2023). Recent studies have identified DM as a potential risk factor for cognitive decline, including mild cognitive impairment (MCI), vascular dementia, and particularly Alzheimer's disease (AD) (Hameed et al., 2015).

AD is the most prevalent form of dementia worldwide, affecting nearly 50 million people, and is marked by progressive memory loss and cognitive decline (Takeishi et al., 2021). At the molecular level, AD is characterized by extracellular amyloid-β (Aβ) plaques, intracellular neurofibrillary tangles (NFTs) composed of hyperphosphorylated tau protein, and activation of immune glial cells, mainly microglia (Calabrò et al., 2021; Lee et al., 2022).

Emerging evidence suggests that DM and AD share overlapping pathophysiological mechanisms, including insulin signaling dysfunction, Aβ accumulation, tau hyperphosphorylation, oxidative stress, vascular dysfunction, and neuroinflammation (Arvanitakis et al., 2004; Rasool et al., 2021). About 50-52% of patients with DM develop dementia as a late complication, with an increased risk of over 65% for AD (Arvanitakis et al., 2004; Barbagallo, 2014; Rasool et al., 2021). Understanding the intricate relationship between peripheral metabolic disorders and brain pathologies is critical to unraveling the molecular mechanisms that link DM -particularly DM2- and AD (Behl et al., 2022).

The relationship between metabolic dysfunction and cognitive impairment has gradually emerged over the past decades. In 1980, hypoglycemia was first observed to influence brain function (Adolfsson et al., 1980), and by 1994, researchers began to associate altered glucose metabolism with dementia (Razay & Wilcock, 1994). In 1996, further studies reported that DM2 patients and elevated blood glucose levels exhibited learning and memory deficits (Messier & Gagnon, 1996). Similarly, in AD patients, hyperglycemia was associated with neuronal damage due to Aβ accumulation (Messier & Gagnon, 1996). This led to the conceptualization of the term "brain diabetes" in 2005 (Li & Hölscher, 2007), to describe overlapping features between diabetes and AD, and to the identification of shared pathological features between DM2 and AD in subsequent years, such as the presence of fibrillar protein aggregates: amylin in pancreatic islets in DM2, and Aβ plaques and NFTs in AD patients (de la Monte & Wands, 2008; Götz et al., 2009).

Additional contributors to this intersection have been identified, including inflammation, oxidative stress, insulin receptor downregulation, and Aβ accumulation (Park, 2011; Saini, 2010). DM2 can impact brain function through various mechanisms, including glucose-induced toxicity, blood-brain barrier (BBB) disruption, increased production of reactive oxygen species (ROS), and IR in the brain (Behl et al., 2022; Garcia-Serrano & Duarte, 2020). Hyperglycemia and reduced brain glucose uptake also contribute to the formation of advanced glycation end products (AGEs), which are now considered a molecular bridge between DM2 and AD pathologies (Behl et al., 2022; Kong et al., 2020; Uribarri et al., 2020).

Patients with DM2 have a higher risk of neurodegeneration and AD development (Stanciu et al., 2020). In AD, insulin receptor density is reduced by 80%, and dysfunctional insulin signaling is believed to play a pivotal role in the development of AD (Hernández-Contreras et al., 2021). Therefore, DM2 patients not only have a higher predisposition to cognitive impairments but also an elevated risk of developing AD later in life (Pasquier et al., 2006; Verdelho et al., 2007). These cognitive impairments have also been replicated in animal models of diabetes, reinforcing the likelihood of a biological connection (Biju & Paulose, 1998; Winocur & Greenwood, 2005).

Additionally, the brains of AD patients exhibit features like those of insulin-deficient states, including reduced insulin and Insulin Growth Factor (IGF) signaling, as well as increased markers of oxidative and inflammatory stress (Craft, 2005, 2006; Rivera-Meza et al., 2017; Steen et al., 2005). Together, these findings support the notion that DM2 and AD are interrelated conditions, linked by common metabolic and neurobiological pathways.

This review aims to explore the neurobiological mechanisms linking DM2 to AD, identify additional risk factors contributing to this association, and analyze current advancements in shared biomarkers. By assessing these connections, we aim to provide a comprehensive understanding of the relationship between DM2 and AD from a neurobiological perspective and to suggest targeted therapeutic strategies (Figure 1).

2. Unraveling the Molecular Nexus Between Type 2 Diabetes Mellitus and Alzheimer's Disease: Insights Into Shared Pathways and Therapeutic Implications.

DM2 and AD represent two seemingly distinct disorders, but they increasingly intersect at multiple molecular levels (Petermann et al., 2018). Despite their clinical differences, accumulating evidence suggests shared features, particularly involving IR, chronic inflammation, oxidative stress, and the presence of common brain proteins such as Aβ (Costache et al., 2023; Formiga et al., 2015; Mestizo Gutiérrez et al., 2014; Zapata-Tragodara et al., 2020).

Insulin plays a fundamental role in brain function, particularly in cognition and memory. Beyond its well-known role in regulating glucose metabolism, insulin crosses the BBB and modulates various critical pathways involved in neuronal survival, including synaptic plasticity, neuroinflammation, and cell viability (Costache et al., 2023). In the context of AD, insulin can influence the dynamics of Aβ; for example, elevated levels of insulin in the brain may increase Aβ production or reduce its clearance, promoting its accumulation (Costache et al., 2023). This is because insulin shares metabolic pathways with Aβ, and dysfunction in insulin signaling can contribute to increased amyloid plaque deposition, a hallmark of the disease. Therefore, insulin levels can also serve as a surrogate marker for the efficiency of Aβ clearance, acting as an indirect indicator of Alzheimer’s-related pathological processes (Costache et al., 2023; Sánchez-Zúñiga et al., 2020). In this sense, brain IR or impaired signaling pathways may be key factors linking metabolic dysfunction characteristic of type 2 diabetes with the neurodegenerative changes observed in AD.

Soluble oligomers of the Aβ peptide (AβOs), widely distributed in the brain, have been implicated in pathologies like AD, inducing hyperphosphorylation of neuronal tau protein, oxidative stress, neurodegeneration, synaptic loss, and inhibition of synaptic plasticity (Chrem Mendez et al., 2019; Haass & Selkoe, 2007; Lamport et al., 2013; Watson et al., 2005). These features are also observed in insulin-involved AD pathogenesis in patients with DM2, suggesting that impaired insulin signaling may underlie a shared pathogenic mechanism (Sims-Robinson et al., 2010) (Figure 2).

Recent discoveries in neuroendocrinology have shown that both DM2 and AD share common features, including aberrant insulin signaling in the Phosphoinositide 3-Kinase (PI3K) and Mitogen-Activated Protein Kinase (MAPK) pathways in brain tissues, mitochondrial dysfunction, impaired glucose transport (via GLUT1 and GLUT3), and oxidative stress (Gutiérrez-Rodelo et al., 2017; von Bernhardi, 2004).

In fact, hyperglycemia in DM2 can affect GLUT transporter activity and compromise the integrity of the BBB, promoting the entry of neurotoxic substances into the brain and exacerbating inflammation (Sienes Bailo et al., 2022; Yaffe, 2004). These metabolic and vascular alterations contribute to cognitive dysfunctions, highlighting the importance of glycemic control in AD prevention (Bohórquez Moreno et al., 2020; Lamport et al., 2014).

Mitochondrial dysfunction also represents a common link between DM2 and AD. A study by Liu et al. (2021) demonstrated that vascular endothelial growth factor (VEGF) mitigated Aβ-induced mitochondrial depolarization and promoted mitochondrial biogenesis, thereby improving cognitive performance (X. Liu et al., 2021). Although VEGF did not significantly affect autophagy or mitophagy, its specific role in preserving mitochondrial function suggests a therapeutic bridge between these two conditions (Demetrius & Simon, 2012; Hansson Petersen et al., 2008; X. Liu et al., 2021; Naia et al., 2023) (Figure 2).

As mentioned, DM2 shares multiple pathological mechanisms with neurodegenerative diseases such as AD and vascular dementia, including alterations in cerebral blood flow, endothelial health, and an exacerbated inflammatory response. Moreover, recent research has demonstrated that dysfunction in brain insulin signaling—characteristic of cerebral insulin resistance or 'type 3 diabetes'—plays a central role in the pathogenesis of these conditions. Disrupted insulin pathways contribute to the accumulation of Aβ and hyperphosphorylation of tau, two key events in neurodegeneration. Therefore, the interplay between metabolic dysfunction and neurodegenerative processes not only supports the hypothesis that brain insulin resistance is a common factor but also highlights potential therapeutic avenues aimed at restoring insulin signaling to slow disease progression (Bedse et al., 2015).

Structural and functional integrity of the BBB is compromised in DM2 due to hyperglycemia, oxidative stress, and chronic inflammation (Bogush et al., 2017; Pooja Naik, 2014). This leads to the accumulation of glycolytic intermediaries, ROS, and activation of inflammatory pathways such as the AGEs-RAGE-NFκB axis (Chelombitko, 2018; Han & Kim, 2023; Jais & Brüning, 2017; Mittal et al., 2014; Tobon-Velasco et al., 2014).

While glycemic control remains the primary target in the management of diabetes, certain therapeutic strategies may inadvertently increase the risk of hypoglycemia, a factor linked to cognitive impairment and memory deficits. Ramirez-Rincón et al. have emphasized the importance of an individualized approach to diabetes treatment, considering patient-specific factors such as cardiovascular risk, comorbidities, and potential adverse events, including hypoglycemia (Ramírez Rincón et al., 2022). They highlight that overly aggressive glucose-lowering regimens, particularly in vulnerable populations, can elevate hypoglycemia risk and potentially accelerate cognitive decline. In contrast, some evidence suggests that maintaining optimal glycemic levels through careful management may help mitigate neurodegenerative processes associated with AD, if hypoglycemia is avoided (Cukierman-Yaffe et al., 2015; Fava, 2014; Vanhandsaeme & Benhalima, 2021). This underscores the need for a balanced approach that not only targets glycemic control but also minimizes hypoglycemia to preserve cognitive function. These observations underscore the intricate interplay between metabolic regulation and neurological outcomes.

IR in the brain has been associated with atrophy and AD phenotype, including tau pathology and Aβ accumulation (Ciudin, 2016). Insulin is vital not only for glucose uptake but also for maintaining axonal integrity through its regulation of cytoskeletal proteins and growth factors (detailed in the next section). The reduction in insulin production, commonly seen in aging and AD, interferes with these processes, thereby promoting synaptic dysfunction and neurodegeneration (Domínguez et al., 2014).

Genetic factors further modulate the link between DM2 and AD. The ε4 allele of apolipoprotein E (APOE), which is the strongest genetic risk factor associated with late-onset AD, is also linked to impaired lipid and glucose metabolism. It is important to note that late-onset AD typically appears after the age of 65, whereas early-onset AD occurs before this age and often involves different genetic and pathological mechanisms. The APOE ε4 allele particularly influences the development and progression of late-onset AD, highlighting its critical role in the neurobiological interplay between metabolic dysregulation and neurodegeneration in the aging population (Jabeen et al., 2022). Individuals carrying the ε4 allele exhibit increased risk of hyperglycemia, IR, and Aβ aggregation (Donoso S & Behrens P, 2005; Jabeen et al., 2022; Roda et al., 2019). By contrast, APOE ε2 confers protection for cognitive impairments and IR, while ε3 is considered neutral (Delikkaya et al., 2019; Martínez et al., 2022; Roda et al., 2019; Snyder, 2015).

The overlapping role of APOE in lipid transport, neuronal repair, and glucose metabolism suggests that it is a key integrator of vascular, metabolic, and neurodegenerative processes (Jabeen et al., 2022). Studies have found that APOE ε4 carriers show specific cognitive deficits—particularly in long-term memory—while individuals with diabetes are more prone to working memory impairments (Ravipati et al., 2022). Importantly, the co-occurrence of APOE ε4 and diabetes does not appear to synergistically elevate AD risk, suggesting distinct but converging mechanisms of damage (Ravipati et al., 2022) (Figure 3).

In summary, multiple overlapping mechanisms—vascular dysfunction, insulin signaling defects, neuroinflammation, Aβ deposition, and tau pathology—may contribute to the link between DM2 and AD. Understanding these intersections opens potential avenues for therapeutic intervention aimed at modulating common molecular targets.

3. Bridging the Gap: Exploring the Interplay Between Insulin Signaling, Diabetes, and Alzheimer's Disease.

Insulin is a hormone secreted by pancreatic β-cells permeable to the BBB and, therefore, able to enter the CNS via systemic circulation (Xourgia et al., 2019). In the brain, insulin acts as a potent neurotrophic factor, regulating metabolism, synaptic plasticity, and cognitive function through the activation of insulin receptors (Galea, 2021; Sima & Zhang, 2014; Sonar & Lal, 2018).

Insulin receptors are ubiquitously expressed in the brain, particularly in the hippocampus, cortex, hypothalamus, olfactory bulb, and pituitary, with a higher density in neurons compared to glia (J. Duarte, 2014; Kleinridders et al., 2014). Studies have shown that insulin receptors are densely expressed in the axons of hippocampal pyramidal neurons and are predominantly localized in brain regions essential for learning, memory, and cognitive functions. In addition to their structural presence, functional studies in healthy conditions provide strong evidence that insulin signaling is crucial for optimal cognitive performance (Freychet, 2000; McNay et al., 2010; Pomytkin et al., 2018). Specifically, when insulin receptors are blocked or inhibited within the hippocampus, there is a significant impairment in memory performance, indicating that endogenous insulin signaling is necessary for the maintenance of normal memory processes. This functional disruption emphasizes that insulin action within the hippocampus is not merely supportive but vital for proper cognitive function, with receptor blockade leading directly to deficits in spatial working memory and other hippocampal-dependent tasks (Freychet, 2000; McNay et al., 2010; Pomytkin et al., 2018).

In DM2 patients, a decrease in insulin receptor density and alterations in insulin signaling have been observed, particularly in the hippocampus, cortex, and choroid plexus (A. I. Duarte et al., 2012). Metabolic disturbances associated with diabetes, including hyperglycemia, hyperinsulinemia, and hypercholesterolemia, have been linked to brain atrophy and neuropathological features of AD (C.-C. Huang et al., 2014; Scheltens et al., 2021; Verdile et al., 2015). Furthermore, even in the absence of dementia, insulin resistance has been associated with cognitive impairment in aging (Burns et al., 2012; Gong et al., 2023).

Insulin regulates protein synthesis, post-translational modifications (PTMs), the assembly of cytoskeletal and adhesive proteins, and intracellular support molecules important for axonal integrity (Sima & Zhang, 2014; Yang et al., 2023). Dysregulation of these processes is associated with an increased risk of developing diabetes and can also lead to neuronal dysfunction, synaptic deficits, and increased vulnerability to neurodegeneration (J. M. N. Duarte, 2023; Yang et al., 2023).

Importantly, insulin regulates the metabolism of Aβ and tau, the basic components of amyloid plaques and NFTs, respectively (G. J. Biessels & Kappelle, 2005; Craft et al., 1998; Kim et al., 2013; Steen et al., 2005). Reduced insulin production, as observed in aging and AD, correlates with decreased insulin activity in regions such as the frontal cortex, hippocampus, and hypothalamus (De Felice et al., 2022).

Knopman, in 2014, studied glucose metabolism using positron emission tomography with fluorodeoxyglucose (PET-FDG) in cognitively normal APOE4 carriers, showing reduced uptake in various brain regions, including the temporal lobe (Knopman et al., 2014). Finally, the APOE3 allele is the most common isoform and tends to remain neutral regarding the disease, neither increasing nor decreasing the risk (Langella et al., 2023; Rebeck et al., 2002). Other studies confirm APOE ε4’s association with accelerated cognitive decline, especially when combined with different mutations, such as PSEN1 (Langella et al., 2023; Polsinelli et al., 2023) (Figure 4).

Interestingly, some studies suggest that APOE ε4 carriers with diabetes may experience distinct cognitive trajectories compared to non-carriers. While APOE ε4 primarily affects long-term memory, diabetes is more closely linked to deficits in working memory (Ravipati et al., 2022). This suggests that DM2 and APOE ε4 may influence cognition through partially independent but converging mechanisms.

The importance of considering APOE genotype and diabetes as potential risk factors for AD highlights the need for further research to elucidate how these factors contribute to the development and progression of the disease (Ravipati et al., 2022; Shinohara et al., 2021) (Figure 4). Insights gained from such studies could pave the way for targeted interventions and personalized treatment approaches in the field of AD research and management.

4. Therapeutic Opportunities and Pharmacological Interventions

Late-onset AD and DM2 share common features, including impaired insulin signaling, chronic inflammation, mitochondrial dysfunction, and glucose dysregulation, suggesting a link between the two conditions (Cholerton et al., 2013; Mancinetti et al., 2023). As mentioned, IR and impaired glucose metabolism may contribute to the development and progression of AD (Mancinetti et al., 2023). These shared mechanisms have raised interest in evaluating anti-diabetic drugs as potential therapies for AD and cognitive decline.

In experimental models, including rodent studies of ischemia and neurodegeneration, pioglitazone has been shown to exert vasculoprotective effects, such as reducing inflammatory responses in adipose tissue, decreasing urinary TGF-β1 excretion, and promoting endothelial progenitor cell mobilization, which enhances angiogenesis (Desouza & Shivaswamy, 2010). These effects are mediated through activation of PPAR gamma and beta/delta pathways, leading to reduced inflammation, improved endothelial function, and potential neuroprotective mechanisms. Clinically, in patients with DM2, pioglitazone's benefits include improved glycemic control, decreased inflammatory mediators like C-reactive protein and VEGF, and possible nephroprotective effects, as evidenced by reduced urinary TGF-β1 excretion. However, risks such as peripheral edema, weight gain, and hypoglycemia when combined with insulin or other hypoglycemic agents are notable considerations (Desouza & Shivaswamy, 2010). By inhibiting the reabsorption of glucose in the kidneys and promoting glucose excretion in the urine, SGLT2 inhibitors improve glycemic control while also modulating brain metabolism and neuroinflammatory pathways implicated in AD (Desouza & Shivaswamy, 2010; Dolan et al., 2010; Majid et al., 2025).

Recent population-based studies have compared the effects of SGLT2 inhibitors with other anti-diabetic drugs, such as dipeptidyl peptidase-4 (DPP4) inhibitors, indicating that SGLT2 inhibitors may be more effective in reducing the risk of dementia in DM2 patients (DeFronzo et al., 2013; Mancinetti et al., 2023). The neuroprotective effects of these drugs extend beyond their anti-diabetic properties, suggesting a direct impact on brain health and cognitive function (Cholerton et al., 2013).

Furthermore, SGLT2 inhibitors have been shown to prevent memory impairment in AD animal models, showing beneficial effects on neurogenesis, synaptic plasticity, and neurodegeneration (DeTure & Dickson, 2019). These findings suggest that targeting glucose metabolism through SGLT2 inhibition may offer novel therapeutic strategies for managing cognitive decline associated with AD (DeFronzo et al., 2013).

On the other hand, AD clinical presentations differ between early- and late-onset forms. Early-onset AD is associated with language, visuospatial, or executive function impairments, while late-onset AD patients exhibit the classic amnestic pattern of the disease (Palasí et al., 2015). Comorbidities such as diabetes, cardiovascular dysfunction, and obesity are more commonly observed in late-onset pathology, supporting the metabolic contribution to disease progression (Chen et al., 2017; Gerritsen et al., 2016).

Insulin plays a crucial role in brain health, influencing cerebral bioenergetics, synaptic viability, neurotransmitter turnover, and proteostasis (Kellar & Craft, 2020). Insulin signaling dysregulation in the brain can lead to neurodegeneration by impairing the clearance of Aβ and tau, promoting vascular damage, and increasing inflammation (Busiguina et al., 2000; Clark & Vissel, 2018; El Khoury et al., 2014).

Peripheral IR may result in reduced brain insulin concentrations, impairing insulin signaling within neurons (Könner et al., 2011). Notably, insulin directly interacts with Aβ, reducing its synaptotoxicity and facilitating clearance mechanisms (Craft et al., 2020). This interaction is critical because reduced insulin levels in the brain and cerebrospinal fluid (CSF) have been reported in AD patients (Craft et al., 2020; Kellar & Craft, 2020).

Emerging therapeutic strategies now aim to restore brain insulin function. Intranasal insulin administration has shown promise by enhancing cognitive performance in both preclinical and clinical studies (Craft et al., 2020). This approach increases brain insulin availability without affecting blood glucose, reducing the risk of hypoglycemia.

Other therapeutic targets include insulin-like growth factor-1 (IGF-1) and insulin receptor substrate-1 (IRS-1), which are frequently impaired in AD brains (Arvanitakis et al., 2020; Talbot et al., 2012). Postmortem studies of AD patients have revealed deficient insulin signaling in regions associated with cognition, such as the frontal cortex and hippocampus (Arvanitakis et al., 2020). These abnormalities are associated with both Aβ accumulation and tau hyperphosphorylation, underscoring the relevance of insulin-related pathways in AD pathogenesis (Craft et al., 2020).

In conclusion, therapies originally designed for DM2, such as SGLT2 inhibitors, may have potential in slowing or preventing cognitive decline associated with AD. These strategies target the intersection of metabolism, neuroinflammation, and neurodegeneration, offering a multi-modal approach to treating patients at the crossroads of diabetes and dementia. Further research is warranted to elucidate the specific mechanisms by which SGLT2 inhibitors exert their neuroprotective effects and to optimize their clinical use.

5. Biomarkers Related to Neurodegeneration in Diabetes Mellitus and Alzheimer's Disease

In recent years, the development of highly sensitive immunoassays has significantly advanced the early and less invasive detection of AD biomarkers in plasma and cerebrospinal fluid (Kivisäkk et al., 2024). Given the growing evidence that DM2 may contribute to the pathogenesis of AD, neurodegeneration has been considered a possible mechanism establishing a connection between DM2 and AD (Madhusudhanan et al., 2020).

Several converging mechanisms, such as vascular endothelial dysfunction, insulin dysregulation, dysglycemia, neuroinflammation, Aβ accumulation, and abnormal tau phosphorylation, have been identified as shared contributors in the pathogenesis of cognitive impairment associated with both DM2 subtypes and AD (Burillo et al., 2021; Sible et al., 2022; Trimm & Red-Horse, 2023; Tumminia et al., 2018; Willette et al., 2015) (Table 1).

AD is pathologically defined by extracellular Aβ plaques and intracellular NFTs composed of hyperphosphorylated tau (Zhang et al., 2021). In CSF, lower levels of Aβ42 and lower Aβ42:Aβ40 ratios have been related to AD (Andersson et al., 2023). This is due to the sequestration of Aβ42 into amyloid plaques, which induces elevated levels of total tau (t-tau) and tau phosphorylated at Thr 181 (p-tau181) in CSF (Lim et al., 2023). Additionally, increases in t-tau and p-tau181 in CSF tend to correlate with Aβ burden rather than with NFT load (Horie et al., 2023), suggesting that they may reflect early pathophysiological changes rather than late-stage neuronal loss (Barthélemy et al., 2023).

Recent advances in plasma biomarkers have demonstrated that phosphorylated tau species, particularly at Thr181 (p-tau181), Thr217 (p-tau217), and Thr231 (p-tau231), can predict AD pathology with high accuracy (Milà-Alomà et al., 2022). For example, p-tau231 in CSF is considered an early marker of Aβ pathology, and p-tau217 may offer better discrimination between AD patients and control groups, with better accuracy than proton emission tomography (Groot et al., 2022; Kimura et al., 2023; Milà-Alomà et al., 2022).

Tau hyperphosphorylation remains a central pathological mechanism in AD and may be exacerbated by chronic hyperglycemia, as seen in diabetes (R. Huang et al., 2020). Tau protein is essential for microtubule stability in neurons, but high glucose levels promote tau hyperphosphorylation in hippocampal neurons, leading to its detachment from microtubules, misfolding, and aggregation into NFTs, inducing cognitive dysfunction in diabetes (Barbier et al., 2019; Cheng et al., 2022; Ke et al., 2009; Tabeshmehr & Eftekharpour, 2023).

This pathological process disrupts axonal transport and cytoskeletal integrity, resulting in neuronal dysfunction, synaptic damage, and cell death (Cheng et al., 2022). Tau is also subject to proteolytic cleavage by endogenous enzymes, such as caspases and calpains, which enhances its aggregation propensity (Barthélemy et al., 2020; Mietelska-Porowska et al., 2014). These cleaved forms of tau further promote NFT formation and correlate with cognitive decline (Olesen & Quintanilla, 2023; Pérez et al., 2018).

Patients with DM2 often exhibit cognitive impairment, including deficits in verbal and visual memory, processing speed, and executive function (Y. Liu et al., 2024). Insulin has been implicated in the clearance of Aβ and the regulation of tau phosphorylation in the CNS (Mullins et al., 2017), with evidence supporting a protective role against Aβ synaptotoxicity by regulating Aβ elimination through the modulation of lipid metabolism, proteases, and insulin-degrading enzymes (Saraya et al., 2023).

C-peptide, traditionally considered a byproduct of insulin synthesis, has gained recognition as a bioactive molecule that may modulate insulin signaling, influencing diabetes (Wang et al., 2012). Emerging evidence suggests that treatment with a C-peptide receptor agonist, such as Glucagon-like peptide-1 receptor agonists (GLP1-RA), may offer therapeutic benefits by improving insulin signaling and reducing the incidence of dementia in type 2 diabetes (DM2) patients (García-Casares et al., 2023; Klausen et al., 2022).

On the other hand, a negative correlation has been observed between insulin signaling and tau phosphorylation (Gonçalves et al., 2019). Upon its binding, the insulin receptor activates different IRS, such as phosphorylated IRS-1, leading to PI3K activation, the phosphorylation of MAPK and protein kinase B (AKT) downstream, and inhibiting GSK-3β, a prominent tau kinase, white its inactivation by AKT signaling inhibits GSK-3β-mediated tau phosphorylation (Boucher et al., 2014; Gabbouj et al., 2019; Ye et al., 2017). When insulin signaling is impaired, GSK-3β remains active, resulting in tau hyperphosphorylation and NFT formation (Hobday & Parmar, 2021; Woodfield et al., 2022).

Hyperglycemia also exacerbates cognitive impairment through osmotic stress, oxidative damage, and inflammatory responses, particularly through the formation of advanced glycation end products (AGE), which subsequently leads to the production of ROS and the release of inflammatory cytokines such as IL-1β and IL-6 (C. Liu et al., 2015), causing systemic inflammation, microvascular dysfunction, and long-term neuronal injury (Ehtewish et al., 2022).

6. Conclusions and Perspectives

This comprehensive review highlights the intricate connections between insulin, neurodegeneration, and AD, shedding light on the multifaceted roles of insulin in maintaining cognitive function. Insulin receptors are ubiquitously expressed in brain regions critical for memory and learning, including hippocampus, cortex, hypothalamus, olfactory bulb, and pituitary (Mullins et al., 2017). Notably, the high density of insulin receptors in hippocampal neurons highlights insulin's crucial role in synaptic plasticity and memory consolidation (G.-J. Biessels et al., 1998; Malenka, 1994).

Both DM2 and AD exhibit reduced insulin levels and altered signaling, particularly in the hippocampus, cortex, and choroid plexus. These changes are associated with brain atrophy and cognitive decline, indicating a potential connection between metabolic disturbances and neurodegeneration.

Insulin plays a multifaceted role in several cellular processes relevant for neuronal survival, including cytoskeletal assembly, synaptic plasticity, energy metabolism, and the clearance of Aβ and phosphorylated tau (Mullins et al., 2017; Saraya et al., 2023). As insulin production declines with aging and disease progression, the loss of its neurotrophic and protective effects may directly contribute to AD pathology.

Genetic factors significantly contribute to the risk of AD. Variants of the APOE gene, particularly the ε4 allele, are strongly associated with increased AD risk and earlier onset. APOE4 influences lipid metabolism, neuroinflammation, and amyloid β (Aβ) aggregation. In contrast, the ε2 allele appears protective, while ε3 remains neutral (Raulin et al., 2022). Understanding these genetic influences not only clarifies the clinical variability seen in AD but also has the potential to guide the development of personalized interventions.

Biomarkers offer another promising avenue. Advances in immunoassays have enabled the measurement of p-tau isoforms in plasma and CSF, enhancing early and precise AD diagnosis. Key biomarkers, including insulin dysregulation, endothelial dysfunction, neuroinflammation, Aβ accumulation, and abnormal tau phosphorylation, serve as indicators of disease progression and therapeutic targets (Milà-Alomà et al., 2020).

Therapeutic interventions aimed at restoring insulin signaling and related molecules are under scrutiny. GLP1 receptor agonists and SGLT2 inhibitors show promise not only in controlling glycemia but also in reducing the incidence and progression of dementia in DM2 patients (Gonçalves et al., 2019). These agents may exert neuroprotective effects through multiple mechanisms, including reducing oxidative stress, enhancing neurogenesis, and improving mitochondrial function.

Hyperglycemia leads to neuronal damage through mechanisms such as osmotic imbalance, oxidative injury, and chronic inflammation. The accumulation of AGEs and release of cytokines like IL-1β and IL-6 further exacerbate microvascular damage and contribute to cognitive decline (Ehtewish et al., 2022; García-Casares et al., 2023; C. Liu et al., 2015; Luna et al., 2021). Insulin protective role in Aβ clearance and tau phosphorylation regulation highlights the potential of targeting insulin signaling pathways in both the prevention and treatment of AD.

Future research should focus on clarifying the molecular links between insulin resistance, genetic risk, and neurodegeneration, with the goal of identifying predictive biomarkers and tailoring individualized therapies. Moreover, non-invasive methods to improve brain insulin signaling, such as the use of intranasal insulin, should be further explored in clinical studies.

In conclusion, the overlap between DM2 and AD reflects a complex interplay between metabolism and neurodegeneration. A deeper understanding of insulin central role in these conditions opens new opportunities for early diagnosis, the development of biomarkers, and the design of targeted treatments aimed at delaying or preventing cognitive decline.