1 Introduction

Cognitive impairment refers to a decline in abilities such as thinking, learning, and memory, which severely affects quality of life. It is estimated that by 2050, the number of people living with dementia worldwide will reach 152.8 million, nearly three times the 57.4 million cases estimated in 2019, posing a major global public health challenge (GBD 2019 Dementia Forecasting Collaborators, 2022). Notably, China has the largest population of dementia patients, imposing a substantial burden on public health systems (Jia et al., 2020). Mild cognitive impairment (MCI) is recognized as an early stage of dementia, with up to 15% of patients progressing to dementia within 2 years (Alzheimers and Dementia, 2023). Preventing or delaying cognitive decline and dementia can significantly improve quality of life in late adulthood and prolong functional independence (Petersen et al., 2018). The definition of cognitive impairment encompasses a broad spectrum ranging from MCI to severe dementia, including Alzheimer’s disease (AD) and Parkinson’s disease (PD) (Wang et al., 2024). Its etiology is multifactorial, involving genetic predisposition, environmental exposures, lifestyle factors, and comorbid health conditions (Zhu et al., 2020). Pathophysiologically, the underlying mechanisms are highly complex, with inflammation, oxidative stress, mitochondrial dysfunction, and blood–brain barrier disruption all implicated in disease progression (Kaur and Sharma, 2022).

Lactate, the end product of glycolysis, has long been a controversial subject in biology and exercise physiology. For more than 200 years, lactate was regarded merely as a “metabolic waste” in muscle, associated with fatigue and soreness (Ferguson et al., 2018). This view shifted in the mid-1980s when George Brooks proposed the “lactate shuttle” theory (Brooks, 1986). At its core, this theory posits that lactate functions as an energy intermediate, produced in tissues with high glycolytic activity and consumed by tissues with high oxidative capacity (Brooks et al., 2022; Brooks, 2018). Current evidence indicates that lactate serves as a metabolic bridge between glycolysis and mitochondrial respiration, acting as both a downstream product of glycolysis and a substrate for oxidative metabolism (Brooks, 2018). Importantly, according to the lactate shuttle hypothesis, this process occurs under fully aerobic conditions and can transcend cellular compartments, functioning across cells, tissues, and organs (Brooks, 2002, 2009). In the central nervous system (CNS), lactate has been recognized as both a crucial energy substrate and a signaling molecule (Magistretti and Allaman, 2018; Perry et al., 2016). It is derived from both glycolysis (Rogatzki et al., 2015) and gut microbiota (Bruning et al., 2019), and is markedly released during exercise (Goodwin et al., 2007). In the CNS, lactate functions as a rapid excitatory signal (Tang et al., 2014; Yang et al., 2014), exerting either neuroprotective (Berthet et al., 2009) or neurotoxic (Lama et al., 2014) effects depending on its concentration and duration of exposure. The astrocyte–neuron lactate shuttle (ANLS) hypothesis proposes that lactate is exported by astrocytes and subsequently taken up and oxidized by neurons, particularly in the context of glutamatergic signaling (Pellerin et al., 1998). Consistent with this concept (Barros, 2013; Bélanger et al., 2011), neurons express the molecular machinery required for glucose uptake and intracellular lactate utilization (Hashimoto et al., 2008). At the brain level, glucose–lactate interactions are critical for both physiological and pathological states. Lactate metabolism contributes to normal brain functions, including energy supply (Theparambil et al., 2024), maintenance of metabolism under hypoglycemia (Herzog et al., 2013), neurometabolic coupling and signal transduction (Li et al., 2025; Liu et al., 2017), and executive function (Hashimoto et al., 2018).

In the context of cognitive disorders, dysregulation of lactate metabolism has been linked to multiple conditions, including Alzheimer’s disease, Parkinson’s disease, traumatic brain injury, stroke, psychiatric disorders, and substance use disorders. This association underscores lactate metabolism as a fundamental biological basis for maintaining cognitive function and highlights its potential as an early diagnostic biomarker, prognostic indicator, and therapeutic target. With the advancement of technologies such as magnetic resonance spectroscopy and protein lactylation assays, dynamic monitoring of brain lactate metabolism and its molecular mechanisms has become increasingly feasible, establishing lactate as a growing focus in neuroscience and translational medicine. This review aims to provide a comprehensive overview of the relationship between lactate metabolism and cognition, spanning from fundamental mechanisms to clinical implications. We first introduce the physiological processes and regulatory mechanisms of lactate metabolism in the brain, followed by an exploration of its roles in energy supply, signaling pathways, synaptic plasticity, neurotransmission, and epigenetic regulation. On this basis, we integrate findings from both animal models and clinical studies across diverse conditions and physiological states, critically evaluate the current evidence, and propose future directions for research and clinical translation. Our goal is to provide insights that may advance mechanistic understanding and facilitate the development of novel strategies for the prevention and treatment of cognitive impairment.

Because lactate alterations are observed across many neurological and systemic conditions, lactate should not be framed as a universal “disease-specific” marker. Rather, lactate is best conceptualized as a context-sensitive readout of the brain’s metabolic state that becomes clinically informative only when interpreted within (i) disease stage (e.g., compensation vs. overload vs. exhaustion), (ii) brain-region specificity (network- and cell-type vulnerability), and (iii) sampling compartment and modality (brain tissue vs. cerebrospinal fluid (CSF) vs. plasma; Magnetic Resonance Spectroscopy (MRS) vs. metabolomics). In this framework, the same direction of lactate change may carry distinct biological meaning across disorders, while disease-relevant signatures emerge from patterned combinations–such as lactate together with monocarboxylate transporter (MCT)/lactate dehydrogenase (LDH) expression, redox/mitochondrial indices, inflammatory markers, and lactylation readouts–and from longitudinal or challenge-based dynamics (exercise, hypoxia, glycemic stress).

To increase transparency, we briefly describe our literature identification process. We searched PubMed, Web of Science, and Scopus for English-language studies published between 2015 and 2025 using keyword related to lactate metabolism, MCTs/ANLS, and cognition. We prioritized original cell/animal/clinical studies relevant to cognitive outcomes and excluded papers not addressing brain lactate biology or cognition.

2 Physiology and regulation of brain lactate metabolism2.1 Brain energy metabolism and the astrocyte–neuron lactate shuttle

Although the brain accounts for only ∼2% of total body weight, it consumes 20%–25% of the body’s energy to sustain its function. More than 10% of cardiac output is directed to cerebral blood flow, reflecting the brain’s high demand for glucose and oxygen (Magistretti and Allaman, 2015). Glucose is the principal energy substrate for mammalian cells. In the brain, glucose is almost completely oxidized to CO2 and H2O through sequential processes including glycolysis, the tricarboxylic acid (TCA) cycle, and oxidative phosphorylation. In astrocytes, glucose metabolism predominantly occurs through aerobic glycolysis, producing lactate, which is then taken up by neurons and converted to pyruvate for oxidation via the TCA cycle and the electron transport chain (Bélanger et al., 2011; Calì et al., 2019; Magistretti et al., 1999). Lactate thus serves as a key mediator of metabolic cooperation between astrocytes and neurons. The astrocyte–neuron lactate shuttle (ANLS) model posits that lactate produced by astrocytes is an essential energy substrate for neurons (Wu et al., 2023). During glucose deprivation, astrocyte-derived lactate acts as a neuroprotective metabolite, and exogenous lactate administration can restore neuronal activity (Sun et al., 2020). In addition to being an energy source, lactate functions as a signaling molecule or receptor agonist, modulating neuronal excitability, synaptic plasticity, and cognitive processes (Magistretti and Allaman, 2018). Evidence suggests that astrocytic lactate release and subsequent neuronal uptake are indispensable for learning, memory consolidation, and long-term potentiation (LTP) (Vezzoli et al., 2020). Suzuki et al. (2011) demonstrated that glycogenolysis and astrocytic lactate release are required for long-term memory formation. Thus, lactate contributes to learning and memory through redox- and energy-dependent mechanisms, highlighting its potential as a therapeutic target.

2.2 Transmembrane transport of lactate

The transport of lactate across cell membranes is mediated by monocarboxylate transporters (MCTs), a family of 14 transmembrane proteins (MCT1–14; SLC16A1–A14) that facilitate the movement of lactate, pyruvate, and β-hydroxybutyrate (Bergersen, 2015). In the CNS, three isoforms–MCT1, MCT2, and MCT4–show distinct regional and cellular distributions (Eid et al., 2018). Lactate transfer through the ANLS and MCTs exhibits cellular specificity: astrocytes express low levels of MCT1 and MCT4, while neurons express the high-affinity transporter MCT2, reflecting a specialized division of labor (Pierre and Pellerin, 2005). Neuronal MCT2, primarily located at postsynaptic membranes, facilitates the uptake of lactate, pyruvate, and ketone bodies as energy substrates (Pierre et al., 2002). Disruption of astrocytic lactate production or downregulation of astrocytic MCT1 and MCT4 expression in the hippocampus impairs long-term memory formation. Notably, this impairment can be reversed by exogenous L-lactate administration, underscoring the critical role of astrocyte–neuron metabolic cooperation in maintaining cerebral energy demands, redox homeostasis, and neurotransmitter receptor activity (Bonvento and Bolaños, 2021).

2.3 Metabolic pathways of lactate in the brain

Recent studies have established lactate as a key player in memory formation and neuroprotection (Proia et al., 2016). Lactate in the brain originates both from central glycolytic activity and peripheral sources transported across the blood–brain barrier. Within the CNS, lactate is primarily generated in astrocytes through glycolysis and glycogenolysis, and it plays a central role in astrocyte–neuron metabolic coupling (Dias et al., 2023). Astrocytes are crucial for brain metabolism: via glucose transporter 1 (GLUT1)-mediated glucose uptake, they metabolize glucose through glycolytic enzymes such as hexokinase (HK), phosphofructokinase (PFK-1) and its regulator PFKFB3, as well as pyruvate kinase M2 (PKM2), producing pyruvate (Takahashi, 2021). Pyruvate is then converted into lactate by lactate dehydrogenase A (LDHA), regenerating NAD+ to maintain energy balance (Denker and Dringen, 2024). Astrocytic glycogen reserves can be rapidly mobilized under norepinephrine stimulation, with glycogen phosphorylase (PYGB) releasing glucose-1-phosphate that enters glycolysis to generate abundant lactate (Coggan et al., 2018). Lactate is subsequently exported from astrocytes through MCT1 and MCT4, supplying energy substrates to nearby active neurons (Descalzi et al., 2019).

Neurons utilize high-affinity MCT2 transporters to import astrocyte-derived lactate. Once inside neurons, lactate is oxidized to pyruvate by lactate dehydrogenase B (LDHB), generating NADH for downstream oxidative metabolism (Descalzi et al., 2019; Lee et al., 2025). Pyruvate is then converted into acetyl-CoA by the pyruvate dehydrogenase (PDH) complex and enters the TCA cycle, driving oxidative phosphorylation and ATP production to support action potentials, neurotransmitter release, and synaptic plasticity (Tiwari et al., 2024). Increased synaptic activity leads to massive glutamate release, which is taken up by astrocytes via excitatory amino acid transporter 1 (EAAT1/GLAST) and excitatory amino acid transporter 2 (EAAT2/GLT-1). This process elevates Na+/K+-ATPase activity and energy consumption (Gegelashvili et al., 2007; Robinson and Jackson, 2016), thereby stimulating glycolysis and lactate production. Lactate exported via astrocytic MCT1/4 is rapidly taken up and oxidized by neurons through MCT2, not only fulfilling energy requirements but also activating signaling cascades that facilitate LTP and memory consolidation (Suzuki et al., 2011; Yang et al., 2014).

Multiple key enzymes, transporters, and receptors are involved in lactate regulation, including glycolytic enzymes HK1/2 (Nowak et al., 2018), PFK-1, PFKFB3 (Imbert-Fernandez et al., 2024; Xiao et al., 2023), and PKM2/PKM1 (Fukushi et al., 2022), lactate dehydrogenase isoforms LDHA (Tian et al., 2023) and LDHB (Park et al., 2022), PDH and its regulators PDK/PDP (Zhou et al., 2025), and glycogen metabolism enzymes GYS1 (Nitschke et al., 2025) and PYGB (Yang C. et al., 2024). Lactate receptor hydroxycarboxylic acid receptor 1 (HCAR1/GPR81), a Gi/o-coupled receptor, suppresses cAMP signaling, contributing to neurovascular coupling and protection against excitotoxicity (Colucci et al., 2023). More recently, lysine lactylation of proteins has emerged as a novel epigenetic mechanism linking lactate to inflammation regulation (Yu F. et al., 2025) and synaptic plasticity (Wu et al., 2024). Collectively, lactate is not only an intermediate metabolite in energy metabolism but also an active regulator in diverse pathophysiological processes within the brain (Figure 1).

Lactate metabolism in the brain and the astrocyte–neuron lactate shuttle. Schematic illustration of lactate metabolism in the brain and the astrocyte–neuron lactate shuttle (ANLS). Glucose enters astrocytes via GLUT1, where it undergoes glycolysis and glycogenolysis to generate lactate. Lactate is exported through MCT1 and MCT4 and subsequently taken up by neurons via MCT2, where it is converted into pyruvate and enters the TCA cycle to support oxidative phosphorylation. This metabolic coupling provides energy for synaptic activity and contributes to memory formation and cognitive function.

3 Mechanisms of lactate in cognitive function

Current evidence indicates that lactate exerts a “double-edged” effect on cognition, with outcomes that can be beneficial or detrimental depending on disease context. This heterogeneity is shaped by lactate concentration and temporal dynamics, regional and compartmental specificity, neuron–glia metabolic coupling, and the organism’s adaptive capacity to metabolic stress. In addition to serving as an energy substrate, lactate also acts as a signaling molecule and an epigenetic regulator, thereby influencing synaptic plasticity, neuroinflammation, mitochondrial function, and gene expression.

Mechanistic boundary conditions for lactate’s beneficial versus detrimental effects. Lactate-related cognitive effects can be reconciled by four interacting determinants: (i) concentration and dynamics (moderate/transient vs. sustained accumulation vs. depletion), (ii) duration of exposure (acute vs. chronic), (iii) cellular source and site of action (astrocyte-to-neuron shuttle vs. maladaptive glycolytic overdrive; tissue vs. CSF vs. plasma), and (iv) disease stage and metabolic capacity (preserved vs. impaired transport–utilization coupling). Because absolute lactate values vary substantially across assays, compartments, and protocols, we do not propose a universal numeric threshold. Instead, we interpret lactate based on directionality, duration, and coupling integrity, and summarize condition-specific patterns in Tables 1, 2. In general, lactate is most likely beneficial when it restores astrocyte–neuron metabolic coupling (ANLS) and supports neuronal oxidative capacity, but becomes detrimental when production exceeds utilization/clearance or when chronic exposure triggers maladaptive signaling and pathogenic protein modifications.

DiseaseStageKey brain
region(s)Lactate
changeNet interpretation
for cognitionKey accompanying
featuresADEarly/MCIHippocampus, frontal cortex↑ (often)Initially compensatory; may become maladaptive if sustainedLactate rises prior to Aβ deposition with enhanced astrocytic glycolysis; accumulation can drive A1 astrocyte activation via AKT–mTOR–HIF-1α; rapamycin mitigatesADEarly (subset reports)Astrocytes, region-dependent↓ (reported)Early glycolytic suppression may impair supportOxidative stress and glycolytic suppression in AD astrocytes may reduce lactate releaseADLate/dementiaCognition-related regions↓Metabolic exhaustion; impaired plasticity/memoryLactate decline with MCT2 downregulation; suppressed astrocytic glycolysis; LDHA↓; IDO1–KYN–AhR axis; ubiquitination-mediated LDHA degradation; lactate depletion reduces APP-K612 lactylation and increases AβADPathogenic lactylation armMolecular↑Harmful signalingExcess lactate can promote tau pathology via p300-mediated tau K331 lactylation; MAO-B/oxidative stressDACD (T1D)Very early (as early as week 3)Hippocampus, hypothalamus, striatum, cortex↑Potential early biomarker; may transiently support functionElevation precedes overt cognitive impairmentDACD (T1D)ProgressionHippocampus, cortex↑↑ (accumulation)More likely harmful when clearance/utilization failsLDHB activity↓ + MCT2↓ suggests overproduction + impaired clearance; FGF21 improves cognition by upregulating MCT2 and LDHB via PI3K/Akt/mTORDACD (stress condition)Recurrent hypoglycemiaBrain↓Biphasic pattern; stress-stage dependentLactate declines during recurrent hypoglycemia, contrasting typical diabetes elevationTBIEarly/acute (some reports)Brain; cortex, hippocampus (intervention)↓Lactate supplementation beneficial (metabolic support)Hypertonic sodium lactate reverses depletion and improves energy and function; lactate preconditioning via GPR81 improves plasticity/cognitionTBIAcute, variableBrain↑ (some reports)Possibly compensatory; context-dependentLLL (low-level light) + lactate/pyruvate improves mitochondrial function, protects hippocampus, restores cognitionPOCDPostoperative multiple time pointsHippocampus↓Energy supply deficit → cognitive declineSurgery-induced inflammation reduces hippocampal lactate; H2S restores Warburg effect and synaptic plasticityPOCD (aged)Certain postoperative stagesBrain/hippocampus↑Can be protective if moderate; harmful if accumulation-drivenLactate supplementation improves cognition via SIRT1 (blocked by EX-527); fructose pathway activation increases lactate synthesis and worsens cognition; inhibition lowers lactate and improvesExercise (acute)Post-exerciseSystemic, hippocampus↑Often beneficial; depends on intensity/clearance/ageLactate elevation associated with executive function changes; lactate activates SIRT1/PGC1α/BDNF and may reshape inflammatory phenotypesAgingYoung vs. agedHippocampusBidirectionalYoung: lactate supports LTP; aged: accumulation may be detrimentalYoung: lactate essential for LTP; aged: inhibiting lactate production can improve LTP; aerobic glycolysis increases with agingStroke/ischemiaAcuteAstrocyte PKM2 dependentSupply disruptionLactate supplementation beneficialAstrocytic PKM2 loss disrupts lactate energy supply; exogenous lactate reverses neuronal death/cognitive deficitsaSAHMetabolic crisisBrain tissue (CMD)↑ (often high)High lactate linked to poor outcomesHigh lactate correlates with cerebral microdialysis (CMD) total-tau; linked to hypoxia/axonal injury/poor cognitionSchizophreniaAcross stagesAnterior cingulate cortex↑Generally adverse associationElevated lactate negatively correlates with cognitive/functional scores; stage differences notedASD/developmental disordersDevelopmentalPlasma, systemic↑Higher lactate linked to poorer cognition/adaptive abilityPlasma lactate elevated; negative correlation with cognitive/adaptive abilities

Comparative lactate dynamics across diseases, stages, and brain regions relevant to cognition.

↑/↓ indicates the direction reported in the cited studies; interpretation depends on stage, region, and transport–utilization capacity. AD, Alzheimer’s disease; DACD, diabetes-associated cognitive dysfunction; POCD, postoperative cognitive dysfunction; TBI, traumatic brain injury; aSAH, aneurysmal subarachnoid hemorrhage.

FeatureBeneficial lactate signaling
(typically early/acute or
well-coupled)Harmful lactate signaling
(typically
overload/mismatch)Exhaustion phenotype
(late/chronic failure)Lactate patternModerate/transient ↑Sustained ↑/accumulation↓/depletionANLS/transport–utilization couplingCoupled shuttle; neuronal uptake/oxidation preserved (e.g., adequate MCT2/LDHB)Mismatch: production exceeds utilization (e.g., MCT2↓, LDHB↓)Supply failure: LDHA↓/MCT2↓, suppressed astrocytic glycolysisDominant roleMetabolic fuel supporting LTP/plasticity and cognitionMetabolic stress marker + maladaptive signaling (inflammation/oxidative stress/protein lactylation)Energy insufficiency → synaptic failure, LTP impairment, cognitive declineGlial/inflammation contextCan promote reparative shifts (exercise-related microglial phenotype changes)A1 astrocyte activation and feed-forward glycolysis amplification (e.g., AKT–mTOR–HIF-1α in AD)Chronic metabolic collapse with impaired glial supportLactylation-related examplesProtective arm: APP-K612 lactylation promotes lysosomal degradation and reduces Aβ; lactate can enhancePathogenic arm: p300-mediated tau K331 lactylation increases phosphorylation/aggregation; MAO-B/oxidative stressReduced lactate may also reduce beneficial lactylation programsRepresentative mappingsEarly AD/MCI; acute TBI with lactate depletion rescued by supplementation; exercise-induced lactateSubarachnoid hemorrhage (aSAH) metabolic crisis (high lactate + tau/poor outcome); POCD fructose-pathway lactate accumulation; AD overload statesLate AD dementia (lactate↓ + MCT2↓ + LDHA↓); POCD hippocampal lactate↓Practical implicationSupport lactate availability/uptake when depletion exists; preserve couplingReduce drivers of accumulation; restore utilization capacityRestore synthesis/transport (LDHA/MCT2 axis) and protect synapses

Beneficial versus harmful lactate signaling across disease stages: a practical decision matrix.

↑, increase in lactate; ↓, decrease in lactate; →, leads to/results in (causal direction).

Consistent with this framework, the recurrent cognitive domains and their mappings across conditions are summarized in Figure 2, while condition-specific lactate trajectories are synthesized in Tables 1, 2.

Lactate-associated cognitive disorders.

3.1 Alzheimer’s disease (AD)

Alzheimer’s disease (AD) is the most common neurodegenerative disorder, characterized by β-amyloid (Aβ) deposition (Magalingam et al., 2018), tau hyperphosphorylation (Muralidar et al., 2020), synaptic dysfunction (Knopman et al., 2021), and neuroinflammation (Kloske and Wilcock, 2020). Lactate is no longer viewed solely as a metabolic byproduct but as a regulator of brain energy metabolism, signaling, and epigenetic modulation. In AD, these functions are closely tied to astrocyte–neuron metabolic coupling and may contribute to energy deficits and cognitive decline. Evidence suggests a stage-dependent relationship between lactate and cognitive impairment in AD. In the early stage and mild cognitive impairment (MCI), lactate levels are often elevated. Preclinical evidence from cellular and animal models indicates that in multiple AD mouse models, lactate concentrations–especially in the hippocampus and frontal cortex–rise prior to Aβ deposition, accompanied by enhanced astrocytic glycolysis (Harris et al., 2016; Santos et al., 2022; Yang X. et al., 2024; Zhu et al., 2025). In microglia exposed to AD plasma, glycolytic enzymes such as glyceraldehyde-3-phosphate dehydrogenase and pyruvate kinase are upregulated, increasing lactate production and apoptosis, thereby impairing cellular energy metabolism (Jayasena et al., 2015). Mechanistically, enhanced astrocytic glycolysis and lactate export may initially act as a compensatory response to energy stress. However, sustained lactate accumulation may promote A1 astrocytic activation via AKT–mTOR–HIF-1α signaling, further amplifying glycolysis and lactate production. This feed-forward loop has been linked to impaired LTP and increased Aβ aggregation and can be attenuated by rapamycin (Harris et al., 2016; Yang X. et al., 2024). Nevertheless, some studies report early lactate reductions, suggesting that glycolytic suppression and oxidative stress in AD astrocytes may lead to decreased lactate release and cognitive impairment (Oksanen et al., 2017; Tarczyluk et al., 2015). In the late stage and dementia phase, lactate decline and downregulation of MCT2 indicate impaired energy metabolism, correlating with advanced pathology and memory deficits (Lu et al., 2015). In late-stage AD brains, reduced lactate is associated with suppressed astrocytic glycolysis, LDHA downregulation, and activation of the IDO1–KYN–AhR axis, all of which exacerbate neuronal energy deficits and LTP impairment (Minhas et al., 2024; Sun et al., 2020). Mechanisms such as VGLL4 downregulation and increased ubiquitination-mediated degradation of LDHA may further reduce lactate synthesis (Tian et al., 2023). Lactate transport dysfunction is another pathogenic mechanism: overexpression of MCT4 in astrocytes increases lactate export but disrupts neuronal energy homeostasis, impairing function and survival (Hong et al., 2020). At the molecular level, under certain conditions, excess lactate can promote tau pathology through p300-mediated tau lactylation at K331, enhancing tau phosphorylation and aggregation (Zhang X. et al., 2025), as well as upregulating MAO-B expression and oxidative stress (Lee et al., 2018). Clinically, observational studies further suggest that CSF lactate is significantly elevated in the MCI stage (Zebhauser et al., 2022), and large-scale studies indicate that CSF lactate levels fluctuate across the AD continuum–elevated in MCI but often declining as dementia progresses–reflecting a shift from metabolic overactivation to metabolic exhaustion; notably, lactate inversely correlates with tau pathology, suggesting that its dynamics may mirror neuronal metabolic impairment in AD (Bonomi et al., 2021; Liguori et al., 2015). Interventional and translational studies provide preliminary causal support. For example, rapamycin mitigates lactate-linked A1 astrocytic activation and AKT–mTOR–HIF-1α signaling (Harris et al., 2016; Yang X. et al., 2024), separately, lactate depletion can reduce APP-K612 lactylation, whereas exogenous L-lactate enhances APP lactylation, consistent with a potential role for lactylation in modulating Aβ generation (Zhang X. et al., 2025); In parallel, lactate serves as both an energy substrate and a neuroprotective factor; under metabolic stress, moderate supplementation can improve synaptic plasticity and cognition (Wang et al., 2025). Exercise or exogenous lactate elevates brain lactate, enhances histone H3 lactylation (H3Kla), and induces a microglial phenotype shift from pro-inflammatory to reparative, thereby reducing neuroinflammation and improving cognition (Han et al., 2023). Conversely, inhibition of MCT4 reduces neuronal apoptosis and inflammation, ultimately preserving LTP and memory (Hong et al., 2020). A stage-resolved synthesis of AD-related lactate findings (including early/MCI versus late dementia patterns) is summarized in Table 1 and interpreted using the decision matrix in Table 2.

In summary, lactate levels in AD exhibit stage-dependent changes: elevated in early stages but reduced in late stages due to impaired synthesis and transport. Lactate participates in multiple pathological processes–including energy metabolism, inflammatory states, synaptic plasticity, tau pathology, and APP processing–with astrocytic metabolic reprogramming as a central driver. Lactate may act both as a pathological mediator and as a therapeutic target, reflecting its double-edged properties. Dynamic lactate monitoring thus holds promise as an early diagnostic and stratification biomarker in AD, particularly when combined with stage definitions and complementary AD-relevant markers (e.g., Aβ/tau) and region-specific readouts.

For rapid cross-condition comparison of lactate directionality, key brain regions, and cognitive interpretation across the disorders covered, please refer to Table 1.

3.2 Diabetes-associated cognitive dysfunction (DACD)

Diabetes-associated cognitive dysfunction (DACD) has increasingly been recognized as a critical complication of diabetes, exerting profound effects on patients’ quality of life. Metabolic disturbances in diabetes may disrupt regional brain energy balance and synaptic plasticity, thereby contributing to the onset and progression of cognitive impairment. Across T1D models, multiple studies report elevated brain lactate, particularly in cognition-related regions (e.g., hippocampus, hypothalamus, striatum, and cortex) (Ando et al., 2022; Dong et al., 2019; Zhang et al., 2020; Zhao et al., 2018, 2022; Zheng et al., 2017a). These changes are commonly linked to disrupted neuron–glia metabolic coupling and neurotransmitter imbalance, which together may contribute to cognitive deficits. Remarkably, lactate elevation occurs as early as the third week, before overt cognitive impairment develops, suggesting its potential role as an early metabolic biomarker (Ando et al., 2022). As the disease progresses, lactate accumulation intensifies, accompanied by reduced LDHB activity and decreased MCT2 expression. This indicates a dual pathology of excessive lactate production and impaired clearance. Notably, fibroblast growth factor 21 (FGF21) ameliorates learning and memory deficits in DACD mice by enhancing neuronal lactate uptake (via MCT2 upregulation) and utilization (via LDHB upregulation) (Zhao et al., 2022). Mechanistically, this effect involves PI3K/Akt/mTOR-dependent translation of MCT2, which promotes pyruvate generation and ATP/NADH production and helps restore hippocampal energy metabolism and synaptic plasticity. Other studies, however, suggest that early in diabetes, astrocytic metabolism is initially upregulated, leading to increased lactate that temporarily supports neuronal function; yet with disease progression, astrocytic support diminishes, reflecting metabolic failure in the diabetic brain (Wang et al., 2015). Interestingly, during recurrent hypoglycemia, brain lactate levels decline rather than rise, indicating biphasic changes in lactate metabolism across different pathological stages and stress conditions (Wu et al., 2025). In type 2 diabetes (T2D), lactate changes appear more heterogeneous across models; nevertheless, many studies report elevated lactate in brain tissue, particularly in the hippocampus (Hackett et al., 2019; Shima et al., 2017; Zheng et al., 2016, 2017b,c). Stress-related increases in amygdalar lactate further disrupt energy homeostasis, providing a metabolic basis for cognitive dysfunction under diabetic stress (Xu et al., 2019). Lactate accumulation is accompanied by enhanced glycolysis, increased activity of lactate-alanine shuttling, and disrupted neuron–astrocyte metabolic communication. As a result, lactate becomes inefficiently utilized for energy production and neurotransmitter synthesis, leading to energy network reprogramming and cognitive decline (Zheng et al., 2017b). At the epigenetic level, histone lactylation (e.g., H4K12la) is upregulated under diabetic conditions, activating the FOXO1/PGC-1α pathway, which exacerbates mitochondrial oxidative stress and neuronal apoptosis, providing molecular evidence of lactate’s pathogenic role (Yang et al., 2025). Importantly, even at the prediabetic stage (e.g., 6 months of high-fat diet), brain metabolic changes are evident, suggesting that lactate dysregulation may precede clinical cognitive decline (Choi et al., 2019). Clinically, observational studies indicate that DACD manifests as impairments in attention, memory, executive function, visuospatial skills, and language abilities (Yu X. et al., 2025). Interventional and translational studies provide preliminary causal support showing that lactate modulation may have therapeutic potential: intracerebroventricular lactate injection improves cognition in diabetic mice (Kobayashi et al., 2019; Wu et al., 2025), and lactate supplementation alleviates recurrent hypoglycemia-induced brain dysfunction by restoring ANLS, reducing oxidative stress, and supporting mitochondrial function and synaptic plasticity (Wu et al., 2025). Exercise exerts similar benefits by increasing lactate production, enhancing lactate-dependent mitophagy, and activating the lactate–SIRT1–FOXO3–PINK1/Parkin axis, thereby improving T2D-related cognitive dysfunction (Khosravi et al., 2024). Exercise may also promote lactate transport by upregulating MCT2 (Shima et al., 2017), enhances BDNF expression (Jesmin et al., 2022), restores hippocampal lactate metabolism, and attenuates diabetes-associated cognitive decline (