Neurons, in comparison to glial cells, maintain a higher energy level due to their continuous need to generate action potentials, synaptic potentials, and neurotransmitters. Similar to glial cells, neurons are capable of fully oxidizing glucose and lactate. However, neuronal energy metabolism displays distinct characteristics, especially when responding to increased energy demand (see Table 1 for an overview).

A study employing high-resolution immunocytochemistry at both light and electron microscopic levels in rodents has demonstrated that neurons efficiently uptake lactate via monocarboxylate transporter 2 (MCT2), a high-affinity transporter localized at postsynaptic densities to facilitate lactate import from astrocyte-released sources (Bergersen, 2007) and subsequently convert it to pyruvate via lactate dehydrogenase isotype 1. When both glucose and lactate are available, neurons preferentially utilize lactate for energy production, with lactate contributing up to 75% to neuronal oxidative metabolism under physiological concentration conditions (Bouzier-Sore et al., 2006). Although neurons possess a full complement of glycolytic enzymes, their expression of 6-phosphofructo-2-kinase (PFK2) is extremely low, especially the Pfkfb3 isoform (Almeida et al., 2004). This enzyme catalyzes the production of fructose-2,6-bisphosphate, an activator of phosphofructokinase-1 (PFK), which, in turn, reduces the efficiency of glycolysis in neurons.

While Pfkfb3 expression is comparable between neurons and astrocytes, there is a significant difference in the degradation process within neurons. Specifically, Pfkfb3 undergoes rapid degradation through the ubiquitin-proteasome pathway in neurons, resulting in lower levels of Pfkfb3 compared to astrocytes (Riera et al., 2003). This distinction becomes particularly critical under conditions of mitochondrial dysfunction. In such scenarios, neurons are prone to spontaneous apoptosis due to insufficient ATP production, whereas astrocytes demonstrate greater resilience. Astrocytes can upregulate rapid glycolysis mediated by PFKFB, thereby enhancing energy production and mitigating mitochondrial dysfunction.

Beyond ATP production, neurons also depend on NADPH for antioxidant functions. As a key product of the pentose phosphate pathway, NADPH strengthens cellular antioxidant defenses by driving the regeneration of reduced glutathione (Bolaños et al., 2010). Additionally, lactate can be converted to pyruvate intra-mitochondrially via mitochondrial lactate dehydrogenase (mLDH), rather than by the cytosolic isoform (cLDH), a reaction that initiates entry into the tricarboxylic acid (TCA) cycle and oxidative phosphorylation.

However, overexpression of Pfkfb3 activates glycolysis in neurons through the APC/C-Cdh1 pathway, which inhibits the pentose phosphate pathway (PPP), reduces the oxidation of glucose-6-phosphate (G6P), decreases the production of reduced glutathione, and finally leads to oxidative stress and apoptosis (Herrero-Mendez et al., 2009). Meanwhile, the enhancement of glycolysis inhibits the pentose phosphate pathway (PPP), resulting in a decrease in the oxidation rate of G6P and a reduction in the production of reduced glutathione (Tietze, 1969). Notably, this inhibitory effect has no impact on the apoptosis of astrocytes (García-Nogales et al., 2003). These observations suggest that neurons have a limited capacity to sustain high rates of glycolysis, warranting further investigation into whether neurons can enhance glycolytic rates under specific conditions.

Moreover, studies have shown that neurons can meet their energy demands by increasing the efficiency of glucose uptake and metabolism in response to stimulation. For example, using ¹³C-labeled 2-fluoro-2-deoxy-D-glucose (FDG) combined with nuclear magnetic resonance, a study found that glucose phosphorylation in nerve terminals significantly increases during acute neuronal activity (such as seizures), confirming the activity-dependent glucose uptake and glycolytic capacity of neurons. Studies observed through two-photon imaging that neurons in awake mice preferentially uptake the glucose analog 2-deoxyglucose-IR (2DG-IR) during sensory stimulation, with higher hexokinase expression than astrocytes, indicating that neurons have an intrinsic ability to metabolize glucose through enhanced glycolysis (Tang, 2018; Díaz-García and Yellen, 2019).

As our understanding of neuronal energy metabolism deepens, current research increasingly focuses on the dynamic regulation of metabolic pathways within neurons under diverse physiological and pathological conditions. Advances in single-cell transcriptomics, metabolomics, and live-cell imaging—such as the genetically encoded nicotinamide adenine dinucleotide (NADH) sensor Peredox—enable researchers to dissect the metabolic profiles of individual neurons and glia with unprecedented precision. For example, single-cell sequencing has revealed that neurons upregulate glucose transporter GLUT3 and hexokinase 2 gene expression under oxidative stress, while live-cell imaging directly visualizes mitochondrial reorganization and glycolytic enzyme relocalization during surges in energy demand. These techniques are unraveling how neurons finely tune the balance between glycolysis and oxidative phosphorylation to cope with challenges like oxidative stress or mitochondrial dysfunction (Barros, 2013).

A key area of interest is the interplay between neuronal energy metabolism and neurodegenerative diseases. Recent studies have revealed that neuronal metabolic reprogramming, characterized by PDK4-mediated inhibition of pyruvate dehydrogenase and altered expression of PKM2/PKM1 isoforms, contributes to early-stage Alzheimer’s and Parkinson’s diseases. This reprogramming leads to a metabolic shift where a greater proportion of glucose is diverted through glycolysis to produce lactate, instead of being fully oxidized in mitochondria. This results in cellular energy deficits and increased oxidative stress (Magistretti and Allaman, 2015). Understanding these processes may uncover novel therapeutic targets aimed at restoring metabolic balance in affected neurons, potentially slowing or even reversing disease progression (Bolaños et al., 2010).

Another emerging trend is the investigation of metabolic coupling between neurons and astrocytes (Ashrafi and Ryan, 2017). In vivo studies using two-photon fluorescence lifetime imaging and genetically encoded sensors (e.g., the laconic lactate biosensor) have confirmed that astrocytes generate lactate via glycolysis during neuronal activation, which is transported to neurons through monocarboxylate transporter MCT2. This process not only supplies an oxidative energy substrate but also enhances synaptic plasticity by regulating the cytosolic NADH/NAD + redox balance in postsynaptic neurons (Möller et al., 2021). Metabolic flux analysis further reveals that this lactate shuttle relies on AMP-activated protein kinase (AMPK)-mediated glycolytic activation in astrocytes and dynamic regulation of mitochondrial pyruvate carriers in neurons, providing both energy and signaling support for synaptic structural remodeling essential for memory formation (Yellen, 2018).

Furthermore, the potential of metabolic interventions as therapeutic strategies is becoming a prominent research focus (Riera and Dillin, 2015). For instance, modulating glucose uptake and utilization in neurons via the AMPK signaling pathway, or enhancing mitochondrial metabolic flux through PGC-1α-dependent mitochondrial biogenesis, could offer neuroprotective effects under metabolic stress or neurodegenerative conditions (Lee and Silva, 2009). This approach may foster the creation of innovative therapies that specifically target neuronal insulin resistance (e.g., GLUT4 dysfunction) and enhance MCT2-mediated ketone uptake, thereby addressing the metabolic vulnerabilities of neurons and offering a promising pathway for intervening in neurodegenerative diseases (Antal et al., 2025).

Looking ahead, integrating multi-omics approaches such as 4-dimensional fast data-independent acquisition (4D-FastDIA) quantitative proteomics and targeted metabolomics [e.g., dissecting how Danggui-Shaoyao-San regulates the glycogen synthase kinase 3 beta (GSK3β)/PGC1α signaling pathway to improve brain glucose uptake and mitochondrial complex function in AD models (Wu et al., 2024)] with computational modeling is unraveling systematic regulatory networks in neuronal energy metabolism, pinpointing critical nodes like pyruvate dehydrogenase complex phosphorylation and dynamic MCT2 transporter expression (Brown and Ransom, 2007). As the field advances toward personalized medicine, single-cell transcriptomics has revealed interindividual variability in neuronal glycolysis-oxidative phosphorylation balance (e.g., differential GLUT1/4 expression affecting substrate selection in APP/PS1 mice), offering new strategies for targeting AMPK-mediated astrocyte glycogenolysis to optimize neuronal energy supply (Wu et al., 2024).

In conclusion, while the fundamental aspects of neuronal energy metabolism are well-established, ongoing exploration of its complexities continues to reveal new layers of regulation and interaction. Future research will likely focus on the specific regulatory mechanisms of these metabolic processes in different neuronal populations, such as the lactate shuttle pathway and monocarboxylate transporter (MCT)-mediated interactions, how they are involved in the pathological processes of neurodegenerative diseases through key signaling pathways, and how to target specific metabolic enzymes and transporters for therapeutic intervention (Bélanger et al., 2011). This forward-looking perspective underscores that energy metabolism functions not merely as a support system for neuronal activity but as a pivotal regulator in brain health and disease through mechanisms such as mitochondrial ATP synthesis, calcium buffering, and oxidative stress response (Nicholls and Budd, 2000). Specifically, dysregulated neuronal energy metabolism can trigger apoptotic cascades via mitochondrial membrane potential collapse or abnormal pyruvate dehydrogenase complex phosphorylation, while targeting monocarboxylate transporter MCT2-mediated lactate shuttling or mitochondrial calcium uniporter (MCU) activity has emerged as a promising therapeutic strategy for neurodegenerative disorders.

Astrocytes, while requiring less energy than neurons, account for 5–15% of the brain’s total energy consumption (Attwell and Laughlin, 2001). In contrast to neurons with higher energy demands, astrocytes exhibit a greater capacity for glucose uptake, which exceeds their immediate energy requirements, as their glycolytic rate is several-fold higher than that of neurons (Barros et al., 2009). This excess glucose is utilized to synthesize glycogen, making.

astrocytes the brain’s sole glycogen reservoirs. Glycogen plays a crucial role during brain development, as neurons rely on astrocytes for energy during this critical stage. Astrocytes support neuronal energy metabolism through three primary pathways:

These pathways highlight the critical role of astrocytes in supporting neuronal energy metabolism and overall brain function.

Astrocytes exhibit a greater glycolytic capacity relative to oxidative phosphorylation, supported by elevated Pfkfb3-driven fructose-2,6-bisphosphate production that accelerates glycolysis, while reduced pyruvate dehydrogenase expression restricts oxidative phosphorylation (Köhler et al., 2020). The elevated expression of Pfkfb3 in astrocytes facilitates the continuous production of fructose-2,6-bisphosphate, an allosteric activator of phosphofructokinase (PFK), which accelerates glycolysis (Bélanger et al., 2011). In contrast, astrocytes exhibit lower expression of pyruvate dehydrogenase (Laughton et al., 2007), which limits oxidative phosphorylation (Itoh et al., 2003) and reduces pyruvate utilization efficiency in the TCA cycle (Rahman et al., 2020). The concurrent production of NADH during pyruvate’s conversion to lactate also fosters a potent antioxidant environment, enabling astrocytes to sustain high glycolytic rates (Bélanger et al., 2011).

In addition to glycolysis, astrocytes store energy by synthesizing glycogen via Glc-6-P. Glycogen is indeed present in astrocytes within the brain (Briski et al., 2021), and under conditions of glucose deficiency or intense neural activity, astrocytic glycogen can be rapidly broken down via glycolysis to produce lactate (Brown and Ransom, 2007). This process, akin to glycolysis, is less prominent in neurons (Vilchez et al., 2007).

As research into astrocytic energy metabolism progresses, several emerging trends are reshaping our understanding of the crucial roles astrocytes play in brain function. One significant area of interest is the interaction between astrocytes and neurons, particularly how astrocytes respond to neuronal signals to regulate their metabolic output. Recent studies suggest that astrocytes can dynamically adjust their metabolic pathways according to the demands of surrounding neurons. In response to neuronal activity, they modulate glycogen breakdown, glucose uptake, and lactate production through calcium-dependent signaling pathways, such as the release of glutamate which is triggered by elevated intracellular calcium levels (Chen et al., 2023). This crosstalk underscores the adaptability of astrocytes, positioning them as active participants in brain energy metabolism rather than merely passive supporters (Verkhratsky and Nedergaard, 2018).

Another trend is the exploration of astrocytic metabolism in neurodegenerative diseases (Bélanger et al., 2011). Increasing evidence suggests that astrocytes may undergo metabolic reprogramming in conditions such as Alzheimer’s and Parkinson’s diseases. In these contexts, astrocytes may shift from a neuroprotective role to one that exacerbates neuronal dysfunction, potentially due to downregulated expression of lactate dehydrogenase A (LDHA), a key glycolytic enzyme, which impairs lactate shuttling by inhibiting large conductance Ca²⁺ activated potassium channel (BK channel)-mediated neuronal hyperpolarization (Yao et al., 2023). Understanding these molecular changes could reveal novel therapeutic targets aimed at restoring normal astrocytic function and protecting neurons from degeneration (Lalo et al., 2021).

Furthermore, advances in imaging and metabolic tracing techniques enable researchers to observe real-time astrocyte-neuron metabolic interactions, such as astrocytes taking up glutamate via transporters and converting it to glutamine for neurons, providing new insights into how these cells coordinate to maintain brain homeostasis (Schousboe et al., 1997). These techniques are expected to clarify the specific mechanisms by which astrocytes sense and meet neuronal energy demands (such as through the monocarboxylate transporter-mediated astrocyte-neuron lactate shuttle), potentially revealing new therapeutic targets in brain metabolism (Yang et al., 2024).

Looking ahead, the potential of targeting astrocytic metabolism as a therapeutic strategy is garnering increased attention (Sun et al., 2022). By modulating key metabolic pathways in astrocytes, it may be possible to enhance their neuroprotective functions or prevent their transition to a harmful state in neurodegenerative diseases (Bak et al., 2018). For instance, enhancing astrocytic glycogen storage and promoting lactate production through the activation of glycogen phosphorylase and monocarboxylate transporters (MCTs) could ensure a stable energy supply for neurons during high-frequency firing or ischemic stress (Simard and Nedergaard, 2004).

In conclusion, while astrocytes have long been recognized for their supportive roles in the brain, current research is increasingly highlighting their active participation in brain energy metabolism and their potential involvement in neurological disorders. As we continue to unravel the complexities of astrocytic metabolism, these cells may emerge as crucial targets for interventions aimed at maintaining brain health and combating neurodegenerative diseases. This forward-looking perspective underscores the importance of continued research into the dynamic and multifaceted roles of astrocytes in the brain.

Astrocytes not only sustain their own physiological activities but also provide crucial energy support to neurons, regulating synapse formation and information transmission. The energy metabolism of astrocytes is closely intertwined with that of neurons, facilitating a functional synergy between the two cell types. For example, neuronal activation is consistently associated with a 3–5 fold increase in astrocytic glucose uptake, as demonstrated by [³H]2DG autoradiography (Pellerin and Magistretti, 1994). During synaptic activity, astrocytes rapidly clear synaptic glutamate via Na⁺-dependent transporters (GLT-1/GLAST), activating the Na⁺/K⁺-ATPase which establishes a Na⁺ gradient driving GLUT1/3-mediated glucose uptake. This process culminates in lactate production via LDHB (EC50 = 80 μM), as confirmed by pharmacological inhibition of glutamate transporters (THA/L-CCG III) and Na⁺ substitution experiments.

Beyond binding to postsynaptic membrane glutamate receptors to generate postsynaptic currents or initiate cell signaling pathways, glutamate released into the synaptic space by the neuronal presynaptic membrane can also be taken up by astrocytes through glutamate transporters. This glutamate uptake in astrocytes requires the energy generated by the entry of Na⁺ ions into the astrocytes through the cis-electrochemical gradient. These Na⁺ ions are then actively transported out of the cell via the Na⁺/K⁺-ATPase enzyme. This increased ATP consumption in astrocytes leads to greater glucose uptake, mediated by an increase in glucose transporter production, which in turn boosts glycolysis and lactate production via LDHA.

The lactate produced by glycolysis is then transported out of astrocytes through monocarboxylate transporters (MCT1 and MCT4) and taken up by neurons via MCT2 (Figure 1). In neurons, lactate is converted into pyruvate by lactate dehydrogenase 1, which then enters the TCA cycle and undergoes oxidative phosphorylation, resulting in the production of significant amounts of ATP to meet the energy demands of normal physiological activities.

Lactate also plays a role in enhancing neuronal excitability in the prefrontal cortex by modulating BK channels, specifically reducing the level of fast afterhyperpolarization (fAHP) during action potentials. The lactate signaling pathway primarily targets ATP-sensitive potassium (K-ATP) channels to regulate neuronal firing, with these channels being a key target for lactate signaling (Karagiannis et al., 2021). However, the precise molecular mechanisms remain unclear, with emerging evidence suggesting lactate may indirectly modulate BK channel function via metabolic-epigenetic crosstalk, such as activating Sirtuin 1 (SIRT1) deacetylase through NAD+/NADH redox modulation or altering channel phosphorylation via AMPK signaling, thereby influencing potassium current properties (Shipston and Tian, 2016; Yu et al., 2018; Li B. et al., 2020). Indeed, lactate acts as an epigenetic regulator by affecting histones, proteins that are critical in the packaging and organization of DNA, thereby influencing gene transcription and expression.

While lactate’s BK channel regulation remains unclear, its epigenetic role is revealed: lactate modifies histone lysine with 28 lactylation sites (e.g., H3K18la) in human/mouse cells. In M1 macrophages, elevated lactate induces Arg1 via lactylation, differing from acetylation dynamics. This offers a new path for neuro-glial metabolic crosstalk; future studies could focus on lactylation-mediated microglial cytokine regulation (Zhang et al., 2019).

It is crucial to acknowledge that the ANLS hypothesis, while supported by a substantial body of evidence as detailed above, remains a subject of active debate within the field. Critics point to methodological challenges and alternative interpretations of in vivo data. A prominent alternative viewpoint argues that the increased glucose uptake observed during neural activation is primarily metabolized via “aerobic glycolysis” (non-oxidative glucose consumption) (Fox et al., 1988) within the activated tissue itself, producing lactate that is largely released rather than shuttled to neurons. This perspective is often supported by neuroimaging studies (e.g., BOLD-fMRI) that report a mismatch between increases in glucose utilization (CMRglc) and oxygen consumption (CMRO₂) during brain activation. Proponents of this view suggest that neuronal energy demands during activation are met directly by glucose or other mechanisms, minimizing the quantitative importance of an intercellular lactate shuttle (Dienel, 2012, 2017).

A critical point of contention lies in the interpretation of the CMRO₂ measurements underlying this debate. The “non-oxidative” conclusion heavily depends on methods like BOLD-fMRI, which infer oxygen consumption from changes in vascular deoxyhemoglobin. A growing critique posits that such methods may not fully capture instantaneous oxygen consumption from pre-existing stores within the brain tissue itself (the parenchyma) (Schurr, 2025). This perspective is strengthened by recent findings demonstrating that lipid-rich structures, most notably the myelin sheaths that envelop axons, can act as significant oxygen reservoirs. Computational and biophysical models suggest myelin has a high oxygen capacitance and can temporarily supply oxygen to adjacent axons during increased metabolic demand (Morelli and Scholkmann, 2023; Vervust et al., 2025). Therefore, the observed hemodynamic uncoupling may reflect, in part, the initial utilization of these local oxygen pools rather than a true lack of oxidative metabolism (Schurr, 2025). The resolution of this debate likely hinges on technical advancements that allow direct, compartment-specific measurement of metabolic fluxes and oxygen dynamics in the tissue parenchyma.

Advances in astrocyte-neuron metabolic interaction research are uncovering mechanisms by which these cells maintain brain homeostasis. A pivotal approach employs cutting-edge imaging techniques like two-photon microscopy and fluorescence lifetime imaging, enabling real-time visualization of metabolic exchanges—such as lactate shuttle dynamics during synaptic activity. These methods reveal that astrocyte Ca²⁺ elevations, triggered by mechanical or electrical stimuli, drive glutamate release to modulate neuronal activity via ionotropic (iGluR) or metabotropic (mGluR) receptors, as seen in hippocampal slices where astrocyte Ca²⁺ waves inhibit or enhance presynaptic neurotransmission, highlighting context-dependent metabolic coupling in physiological and stress conditions (Chen et al., 2023). These technologies reveal the dynamic astrocyte-neuron lactate shuttle: during learning/memory, astrocytic glycogen-derived lactate transported via MCTs promotes Arc/cAMP response element-binding protein (CREB)-dependent synaptic plasticity; under stress, lactate supports neuronal KATP channels via mitochondrial oxidation, highlighting adaptive metabolic coupling (Bélanger et al., 2011).

Another area of increasing interest is the signaling role of astrocytic lactate beyond its role as an energy substrate: emerging evidence demonstrates that lactate induces histone lysine lactylation (e.g., H3K18la), activating transcription factors like Klf4 to regulate genes associated with synaptic plasticity (e.g., Arc, CREB), thereby contributing to epigenetic remodeling during long-term memory formation—a mechanism that underscores metabolic signaling as a target for cognitive enhancement (Yao et al., 2023). This concept opens up exciting possibilities for new therapeutic strategies targeting metabolic pathways to enhance cognitive functions, especially in aging and neurodegenerative diseases (Lalo et al., 2021).

Furthermore, the link between astrocytic metabolism and neurodegeneration is gaining traction: during brain aging or degenerative processes, the efficiency of astrocyte-neuron metabolic coupling—critical for glutamate homeostasis via transporters like GLT-1 and GLAST—declines, leading to reduced glutamate clearance, extracellular accumulation, and NMDA receptor-mediated excitotoxicity, which drives neuronal dysfunction. In Alzheimer’s disease, for example, downregulation of GLT-1 correlates with synaptic damage and exacerbates disease progression (Schousboe et al., 1997). In Alzheimer’s disease, astrocytic dysfunction impairs glutamate clearance via downregulated GLT-1/GLAST and reduces lactate production through decreased lactate dehydrogenase A (LDHA) activity, leading to neuronal energy deficits and exacerbated NMDA receptor excitotoxicity. Targeting astrocyte-mediated glutamate transport (e.g., enhancing GLT-1 function) or lactate shuttling (e.g., activating MCT1/2 transporters) represents a promising therapeutic strategy to preserve neuronal bioenergetics and slow disease progression (Yang et al., 2024).

Looking ahead, research is also beginning to explore how systemic metabolic states, such as those influenced by diet and exercise, affect astrocyte-neuron interactions (Sun et al., 2022). There is growing interest in how metabolic interventions, like ketogenic diets or intermittent fasting, might enhance brain energy metabolism by modulating astrocytic functions (Bak et al., 2018). These approaches could offer new ways to support cognitive health and combat neurodegenerative disorders (Simard and Nedergaard, 2004).

In conclusion, while significant progress has been made in understanding the metabolic interplay between astrocytes and neurons, many questions remain unanswered. Future research will likely focus on elucidating the molecular mechanisms underlying astrocyte-mediated modulation of neuronal activity, the role of lactate as a signaling molecule, and how these processes are altered in neurodegenerative diseases. The continued integration of advanced imaging, molecular biology, and metabolic studies will be crucial in uncovering the full extent of astrocyte-neuron metabolic cooperation and its implications for brain health and disease (Verkhratsky and Nedergaard, 2018). This forward-looking perspective highlights the potential of targeting astrocytic metabolism as a novel therapeutic strategy in the fight against neurodegenerative diseases and cognitive decline (Simard and Nedergaard, 2004).

Glutathione (GSH) is a critical antioxidant in the brain, playing a vital role in protecting neurons from oxidative damage by scavenging reactive oxygen species (ROS). Astrocytes efficiently synthesize GSH using cysteine (Cys) and release it extracellularly (Figure 2). Once outside the cell, GSH is cleaved into Cys-Gly by the enzyme γ-glutamyl transpeptidase (γ-GT), which is located on the astrocytic cell membrane. Neurons then take up Cys-Gly and convert it into Cys and Gly via the enzyme aminopeptidase-N. This process enables neurons to use Cys for synthesizing their own GSH, which is essential for defending against oxidative stress.

Numerous studies have highlighted the significance of oxidative stress and reduced GSH levels as major risk factors for neurodegenerative lesions in the central nervous system (CNS) across various neurological diseases (Dringen, 2000). GSH levels in the brain tend to decline in age-related neurodegenerative diseases, such as amyotrophic lateral sclerosis (ALS), Parkinson’s disease (PD), and Alzheimer’s disease (AD) (Aoyama and Nakaki, 2013). Postmortem analyses of brains from PD patients have shown a marked decrease in GSH levels, particularly in the substantia nigra of the midbrain (Rae and Williams, 2017). This reduction in intracellular GSH concentrations, combined with diminished ATP production, has been linked to increased vulnerability of dopaminergic neurons in PD, as observed in both in vitro and in vivo studies (Sian et al., 1994). In the MPTP mouse model of PD, GSH depletion resulted in elevated oxidative stress in the midbrain and dysfunction of the Excitatory amino acid carrier 1(EAAC1) transporter (Zeevalk et al., 1998). These findings suggest that a decline in neuronal GSH levels is a critical event preceding the onset of PD.

In AD, the abnormal aggregation of amyloid-beta (Aβ) protein contributes to increased oxidative stress, which is believed to play a central role in disease onset. Aβ has been shown to impair EAAC1 transporter function, thereby inhibiting cysteine uptake and further exacerbating oxidative stress (Aoyama et al., 2008). These observations emphasize the importance of maintaining optimal GSH levels and minimizing oxidative stress to preserve neuronal health and prevent the progression of neurodegenerative diseases. However, further research is needed to fully elucidate the underlying mechanisms and to develop targeted therapeutic interventions.

In patients with AD or mild cognitive impairment, proton magnetic resonance spectroscopy (1H-MRS) has revealed significantly lower hippocampal GSH levels compared to healthy elderly controls (Hodgson et al., 2013), suggesting that GSH depletion may play a role in AD pathogenesis. Similarly, oxidative stress is a hallmark of ALS (Mandal et al., 2015). In clinical studies where brain glutathione (GSH) levels were quantified using standard J-edited spin-echo differential magnetic resonance spectroscopy (MRS), results indicate that demonstrating decreased GSH levels in the brains of ALS patients compared to age-matched healthy volunteers, particularly in the motor cortex (Aoyama et al., 2000; Weiduschat et al., 2014). These findings further underscore the involvement of GSH in the development and progression of ALS.

Table 2 summarizes the key metabolic alterations observed in neurons, astrocytes, and microglia across major neurodegenerative diseases.

GSH’s pivotal role in protecting neurons from oxidative damage has sparked significant interest in its potential as a therapeutic target for neurodegenerative diseases. As understanding of oxidative stress in neurodegeneration deepens, current research is increasingly focusing on the mechanisms that drive GSH depletion and the broader implications of oxidative imbalance in the brain.

A key emerging trend is the exploration of GSH-boosting strategies as potential therapeutic interventions. For example, research is investigating the use of GSH precursors, such as N-acetylcysteine (NAC), to enhance intracellular GSH levels and counteract oxidative stress in neurodegenerative conditions (Arakawa and Ito, 2007). Additionally, novel blood–brain barrier-penetrant compounds (e.g., lipoic acid derivatives, GSH precursor transporter agonists) are being developed to boost CNS GSH levels by activating glutamate cysteine ligase in neurons or enhancing precursor uptake via OATP at the blood–brain barrier, thereby restoring neuronal antioxidant capacity and protecting against tau oxidation in AD and dopaminergic neuron loss in PD (Sun and Chen, 1998).

In addition to the GSH system, the brain employs other endogenous antioxidant mechanisms. For instance, lactate plays a dual role in aerobic metabolism: it serves not only as an important energy substrate but also contributes to antioxidant defense. The concomitant production of reduced NADH during the mitochondrial conversion of lactate to pyruvate can directly scavenge ROS. Schurr and Gozal demonstrated that during neuronal activation, NADH—produced via mLDH—acts as an endogenous ROS scavenger, neutralizing oxidative stress induced by glutamate excitotoxicity and thereby conferring neuroprotection (Schurr and Gozal, 2011). This lactate-NADH antioxidant mechanism operates in concert with the GSH system, collectively maintaining neuronal redox balance under conditions of activation or metabolic stress. Its role may be particularly significant during early metabolic disturbances in neurodegenerative diseases.

Another area of growing interest is the role of astrocyte-neuron interactions in maintaining GSH homeostasis: astrocytes synthesize GSH by importing cystine via the cystine/glutamate antiporter (Sxc⁻) and glutamine synthetase, releasing GSH directly through pannexin-1 channels or gap junctions (Cx43), while regulating extracellular glutamate levels via excitatory amino acid transporter 1/2(EAAT1/2) transporters to sustain neuronal GSH precursor supply, thereby mediating neuroprotection under oxidative stress (Verkhratsky and Nedergaard, 2018). Future research is likely to focus on enhancing astrocytic support for neuronal GSH levels, particularly in aging and neurodegeneration: strategies may include activating Nrf2-mediated antioxidant responses to upregulate cystine/glutamate antiporter (Sxc⁻) and glutamine synthetase in astrocytes, or targeting AMPK/SIRT1 pathways to facilitate GSH precursor transfer to neurons. Elucidating such signaling mechanisms could uncover novel therapeutic targets by leveraging astrocyte-neuron metabolic crosstalk (Dringen, 2000).

Moreover, the potential of personalized medicine in managing oxidative stress-related neurodegeneration is gaining momentum: genetic variants in the Nrf2 signaling pathway or SIRT1-mediated deacetylation affecting astrocytic GSH synthesis (e.g., regulating glutamate cysteine ligase catalytic subunit expression), combined with biomarkers like individual GSH levels and Nrf2 activity, could enable tailored interventions. For instance, activating AMPK/SIRT1 pathways to enhance neuronal GSH precursor supply or targeting Nrf2-dependent antioxidant gene expression, particularly addressing SIRT1 hypofunction in Alzheimer’s disease, may optimize neuroprotective outcomes by addressing patient-specific metabolic vulnerabilities (Simon et al., 2020). The field is also exploring the link between metabolic health and GSH levels: emerging evidence indicates that systemic metabolic states, including dietary patterns (e.g., fiber-rich diets enhancing short-chain fatty acids) and gut microbiota, modulate brain GSH levels via the gut-brain axis. Mechanisms involve Nrf2-dependent upregulation of glutamate cysteine ligase in astrocytes, thereby balancing redox status and oxidative stress (Sharon et al., 2016). This opens up possibilities for lifestyle interventions to modulate GSH levels and support brain health: ketogenic diets, by generating ketone bodies like β-hydroxybutyrate, activate astrocytic Nrf2 signaling to upregulate glutamate cysteine ligase and enhance GSH synthesis; intermittent fasting may improve neuronal GSH precursor transport via AMPK/SIRT1 pathways. Investigating how such interventions affect brain GSH metabolism could reveal new strategies for neuroprotection (McDougall and Stewart, 2005). The neuroprotective potential of the ketogenic diet is closely linked to its induced metabolic remodeling: by generating ketone bodies like β-hydroxybutyrate, it activates astrocytic Nrf2 signaling to upregulate glutamate cysteine ligase, directly enhancing glutathione (GSH) synthesis and brain antioxidant capacity. Additionally, the diet modulates gut microbiota (e.g., enriching Akkermansia muciniphila), indirectly improving neuroinflammatory microenvironments via the gut-brain axis, offering new targets for interventions in neurodegenerative diseases through metabolic-microbial crosstalk (Dilmore et al., 2023). The neuroprotective effects of the ketogenic diet are closely linked to its induction of metabolic remodeling and protein modification: by increasing β-hydroxybutyrate levels, it specifically promotes β-hydroxybutyrylation of key tricarboxylic acid (TCA) cycle enzymes (such as citrate synthase [CS] and succinate-CoA ligase subunit alpha [SUCLG1]), enhancing enzymatic activity and ATP production to improve neuronal energy metabolism. Additionally, the diet reduces amyloid- β plaque deposition and microglial overactivation in Alzheimer’ s model mice, providing a novel mechanism for intervening in neurodegenerative diseases through regulation of metabolic enzyme modifications (Han et al., 2025). The neuroprotective effects of the ketogenic diet further extend to synaptic plasticity regulation: by increasing β-hydroxybutyrate (BHB) levels, it directly activates the ERK/CREB signaling pathway in the hippocampus, promoting brain-derived neurotrophic factor (BDNF) synthesis (especially in females) to restore impaired long-term potentiation (LTP) in Alzheimer’s model mice. Additionally, the diet reduces the expression of microglial activation markers (e.g., Iba1, CD11b), mitigating neuroinflammation, with its protective effects independent of amyloid- β plaque reduction, thus offering new targets for intervening in early cognitive dysfunction (Di Lucente et al., 2024).

While GSH has long been recognized as a key brain antioxidant, its role in neurodegenerative diseases is now being unraveled with mechanistic clarity: future research will likely focus on activating astrocytic Nrf2 signaling to enhance glutamate cysteine ligase (GCL)-mediated GSH synthesis, exploring the role of MCT1/2 transporters in neuronal GSH precursor supply, and dissecting how oxidative stress modulates microglial activation via pathways like NF-κB-dependent proinflammatory cytokine secretion (Sun et al., 2022). The continued integration of molecular biology, genetics, and systems biology approaches is pivotal to unraveling the full therapeutic potential of glutathione (GSH). This forward-looking perspective highlights that maintaining brain redox balance—by targeting astrocyte-neuron metabolic coupling and microglial oxidative stress pathways—represents a central strategy in combating neurodegenerative diseases (Aoyama, 2021).

Microglia, which constitute 10–15% of brain cells, play a critical role in the innate immune response and neuroplasticity (Figure 3). The density of microglia varies across brain regions, with the highest densities found in the cortex, striatum, corpus callosum, hippocampus, subventricular zone, and rostral migratory stream. Intermediate densities are observed in the olfactory bulb, thalamus, hypothalamus, midbrain, and brainstem, while the cerebellum has the lowest microglial density (Campos-Torres et al., 2005; Stratoulias et al., 2019; Tan et al., 2020). Microglia exhibit distinct gene expression profiles and functional heterogeneity, maintaining a ramified morphology at rest via CSF1R signaling. Upon activation, CX3CR1-mediated chemotaxis induces a shift to an amoeboid shape, accompanied by upregulation of TREM2-dependent phagocytic receptors and P2X7R activation, triggering dynamic secretion of pro-inflammatory cytokines (e.g., IL-1β, TNF-α) or anti-inflammatory factors (e.g., IL-10, TGF-β). This phenotypic plasticity manifests as biphasic regulation in neurodegeneration: early protective phagocytosis of Aβ plaques versus late-stage pro-inflammatory exacerbation, modulated by microenvironmental signals like local ATP concentration and neuron-secreted IL-34 (Eyo et al., 2013; Colonna and Butovsky, 2017; Tan et al., 2020).

Glucose is the primary energy source for microglial survival and function, with hypoxic glucose deprivation leading to microglial cell death (Kacimi et al., 2011; Kalsbeek et al., 2016). Microglia express multiple glucose transporters, including GLUT1, 3, 4, 5, 6, 8, 9, 10, 12, and 13, as well as high levels of hexokinase 2 (HK2), which collectively ensure sufficient glucose influx for their metabolic needs (Baik et al., 2019; Wang et al., 2019). Among these, GLUT1 is the most highly expressed in microglia and plays a key role in glucose uptake, especially under inflammatory conditions (Baik et al., 2019). In addition to glucose, microglia utilize glutamine as an energy substrate, facilitated by the high expression of glutamine transporters SLC1A5 and SLC38A1 (Liu et al., 2018).

The energy metabolism of microglia varies depending on their activation state. Quiescent microglia rely on oxidative phosphorylation of glucose to meet their energy demands. However, when microglia are stimulated, they shift from oxidative phosphorylation to glycolysis (Cheng et al., 2014). Although glycolysis produces less ATP compared to mitochondrial respiration, it generates lactate more rapidly, which supports energy-intensive processes such as proliferation, migration, and secretion (Werneburg et al., 2017; Elmore et al., 2018). AMP-activated protein kinase (AMPK), a sensor for AMP and ADP, inhibits mechanistic target of rapamycin (mTOR) phosphorylation, thereby suppressing the expression of hypoxia-inducible factor 1-alpha (HIF-1α), the master transcriptional regulator of glycolysis (Peek et al., 2017). Inhibition of the mTOR pathway significantly reduces the inflammatory response in microglia, highlighting how cellular metabolic pathways, particularly the mTOR-HIF-1 pathway, can influence the functional phenotype of microglia (Cheng et al., 2014).

Microglia, often referred to as the brain’s macrophages, play a vital role in regulating brain function. They perform various essential tasks, including clearing dead neurons, eliminating non-functional synapses, and producing ligands that support neuronal survival, all of which contribute to the overall regulation of brain function (Saw et al., 2020). Under normal physiological conditions, motile microglia continuously monitor their microenvironment. Through interactions with neurons, microglia regulate crucial processes such as neuronal activity, synapse formation and survival, and the remodeling of presynaptic circuit (Badimon et al., 2020; Saw et al., 2020; Sun et al., 2020).

Neuronal activity plays an active role in modulating microglial function, particularly in synapse pruning, which ultimately influences synaptic plasticity (Kim et al., 2020). Synaptic plasticity is the central nervous system’s ability to modify synaptic and neural connections in response to synaptic activity, sensory input, and motor experiences. Microglia are key players in this regulation, affecting both long-term potentiation (LTP) and long-term depression (LTD), which are fundamental mechanisms underlying synaptic plasticity (Raghuraman et al., 2019; Kim et al., 2020).

Moreover, in the adult brain, microglia promote dendritic spine formation through the secretion of brain-derived neurotrophic factor (BDNF). Remarkably, studies have shown that microglial regeneration in humans is feasible. Replantation of microglia in the brains of individuals with Alzheimer’s disease (AD) has been found to restore synaptic density and neurosynaptic plasticity, thereby alleviating AD-related learning and memory impairments (Wang W. et al., 2023). These findings underscore the intricate relationship between microglia and neurons, highlighting the critical role of microglial metabolism in maintaining optimal brain function and offering promising therapeutic potential for neurodegenerative diseases like AD.

In addition to regulating neuronal activity, microglia are adept at sensing and responding to neural activation. Similar to inhibitory neurons, microglia can detect excessive neuronal activity and inhibit it, thereby protecting the brain from disease (Akiyoshi et al., 2018). When neurons are activated, they release ATP molecules that microglia can sense. Microglia then break down ATP into adenosine, which binds to adenosine receptors on the surface of active neurons. This binding effectively inhibits neuronal activity, preventing overactivation. This negative feedback mechanism is crucial for regulating neuronal activity, as excessive hyperactivity can lead to neuronal death, synaptic loss, and reduced plasticity (Badimon et al., 2020).

Microglia also play a significant role in responding to brain injury, infection, or neuroinflammation. Upon activation, microglia undergo morphological changes, becoming amoeboid in shape and migrating to sites of inflammation. There, they secrete proteins such as cytokines, chemokines, and reactive oxygen species, which can impact synaptic plasticity and contribute to learning and memory dysfunction. These dysfunctions are associated with various conditions, including aging, AD, traumatic brain injury, HIV-related neurocognitive impairment, and psychiatric disorders such as autism, depression, and post-traumatic stress disorder (Elmore et al., 2018; Krukowski et al., 2018).

Aberrant activation of microglia and the subsequent neuroinflammation have been identified as key mechanisms underlying cognitive deficits associated with normal aging and various diseases. This suggests that microglia serve not only as immune and neural support cells in the brain but also play a crucial role in regulating synaptic structure, memory strength, and precision. Therefore, targeting microglia holds potential as a therapeutic strategy for treating cognitive disorders associated with aging, AD, and other related conditions (Lee et al., 2018; Zuckerman et al., 2018; Li et al., 2021).

As our understanding of microglial function and metabolism deepens, several emerging trends are shaping future research in this field. One significant area of focus is the development of therapeutic strategies aimed at modulating microglial activation states. Given that microglia can exhibit both pro-inflammatory and anti-inflammatory phenotypes depending on the context, future research is likely to explore ways to fine-tune microglial responses to promote neuroprotection and reduce neuroinflammation. For instance, interventions that selectively inhibit the pro-inflammatory mTOR-HIF-1 pathway while enhancing anti-inflammatory signaling could be a promising avenue for treating neurodegenerative diseases such as AD and Parkinson’s disease (Lawson et al., 1990; Mittelbronn et al., 2001; Lalo et al., 2021).

Another promising trend is the exploration of microglial metabolism as a therapeutic target. Since microglial function is closely tied to their metabolic state, strategies that modulate glucose uptake, glycolysis, or oxidative phosphorylation could profoundly impact microglial activity and, by extension, brain health. For example, enhancing oxidative phosphorylation in microglia might promote a more quiescent, neuroprotective state, while targeted inhibition of glycolysis could reduce harmful inflammation (Barros et al., 2009; Yellen, 2018; Wu et al., 2024). Understanding the metabolic signatures associated with different microglial activation states could lead to the development of novel therapeutics that specifically target these metabolic pathways (Bélanger et al., 2011; Sun et al., 2022).

The interplay between microglial function and systemic metabolic states is gaining increasing attention. For example, dietary polyphenols (such as anthocyanins and ellagitannins in berries) can activate the AMPK/mTOR signaling pathway to inhibit NF-κB-mediated secretion of pro-inflammatory cytokines like IL-1β and TNF-α in microglia, thereby improving the neuroinflammatory microenvironment associated with metabolic disorders such as diabetes. Exercise or metabolic interventions may enhance CSF1R-dependent microglial ramification, promoting the release of brain-derived neurotrophic factor (BDNF) to maintain synaptic plasticity and mitigate cognitive decline linked to aging or neurodegenerative diseases (McDougall and Stewart, 2005).

Furthermore, microglial regeneration as a therapeutic approach holds significant promise, with adult brain-resident nestin-positive progenitor cells capable of rapidly proliferating and differentiating into ramified microglia under CSF1R signaling. These cells secrete neurotrophic factors like brain-derived neurotrophic factor (BDNF) to promote synaptic remodeling, addressing synaptic dysfunction in neurodegenerative diseases. Advances in stem cell biology and regenerative medicine enable the generation of functional microglia from embryonic or induced pluripotent stem cells, specifically restoring dysregulated microglial networks in conditions such as Alzheimer’s disease and traumatic brain injury. This process reinstates CX3CR1-mediated synaptic pruning and complement-dependent synapse elimination, offering a novel strategy to rebuild neural circuits and cognitive function (Elmore et al., 2014).

While microglia are well-established as central to brain immunity, recent findings reveal their mechanistic roles in synaptic plasticity via C1q/C3 complement-mediated pruning and TREM2-DAP12-dependent phagocytosis, directly regulating neuronal survival and circuit homeostasis. Future research will focus on decoding metabolic regulation—such as CSF1R-mediated resting-state energy metabolism and AMPK/mTOR-controlled activated-state metabolic remodeling—and targeting BDNF neurotrophic or TNF-α pro-inflammatory signaling to develop precision therapies for neurodegenerative diseases, leveraging microglial metabolic-immunological crosstalk (Salter and Stevens, 2017).

As we advance our understanding of brain energy metabolism, the emerging research trends and unsolved scientific questions reveal several promising avenues for future exploration. The intricate balance between energy supply and demand in neurons, astrocytes, and microglia, and their dynamic interactions under both normal and pathological conditions, underscores the complexity of maintaining brain health. A critical immediate step is to design experiments that can directly resolve the longstanding controversy regarding fuel utilization during brain activation. This requires moving beyond inferences from vascular signals and towards techniques capable of distinguishing metabolic fluxes within specific cellular compartments. Ideal experiments would simultaneously quantify, with high spatiotemporal resolution, key parameters such as: (1) lactate and glucose dynamics in astrocytes and neurons using genetically encoded biosensors; (2) oxygen concentration and consumption within the tissue parenchyma itself, not solely in the vasculature; and (3) the contribution of local oxygen reservoirs, such as those in myelin sheaths (Morelli and Scholkmann, 2023; Vervust et al., 2025), to meeting transient metabolic demands. By directly testing whether activated neurons primarily oxidize astrocyte-derived lactate versus other substrates, and by accurately accounting for all sources of oxygen (vascular and extravascular), such approaches could provide definitive evidence to reconcile the conflicting models (Tietze, 1969; Bolaños et al., 2010). Moving forward, the integration of cutting-edge technologies such as multi-omics approaches, advanced imaging techniques, and computational modeling will be pivotal in unraveling these complexities.

While single-cell technologies have revealed metabolic diversity in neurons and glia, most studies remain static. The dynamic metabolic reprogramming (e.g., rapid glycogen mobilization in astrocytes during ischemia) and its role in disease progression are poorly understood. For instance, astrocytic LDHA downregulation in Alzheimer’s disease may not only impair energy supply but also directly suppress synaptic plasticity genes via epigenetic modifications (e.g., H3K18 lactylation)(Yao et al., 2023). Future studies should develop time-resolved metabolic tracing technologies to uncover how metabolic heterogeneity drives degeneration in specific neuronal subtypes.

Although ketogenic diets show promise in animal models, clinical outcomes are inconsistent. Recent work found that β-hydroxybutyrate (BHB) enhances mitochondrial function but may exacerbate neuroinflammation by inhibiting histone deacetylases (Sun et al., 2022). Such dual effects highlight the need to stratify patients by disease stage and metabolic status. Moreover, brain-region specificity of metabolic interventions (e.g., hypothalamic sensitivity to ketones vs. cortical reliance on lactate) remains unexplored, limiting broad application. Integrated platforms (e.g., brain organoids coupled with clinical cohorts) are needed to define therapeutic windows and contraindications.

Microglial metabolic plasticity (e.g., M1 phenotype relying on glycolysis vs. M2 phenotype preferring oxidative phosphorylation) is well-documented, but its regulation of synaptic pruning remains unclear. For example, in Parkinson’s disease, does TREM2-mediated α-synuclein phagocytosis by microglia depend on lactate levels? Live imaging reveals lactate-induced rapid microglial morphological changes, yet the signaling pathways (e.g., via monocarboxylate transporters or GPCRs) are unresolved (Salter and Stevens, 2017; Dilmore et al., 2023). Targeting these metabolic-immune interfaces may yield dual-action therapeutics with anti-inflammatory and neuroprotective properties.

The gut-brain axis links microbial short-chain fatty acids (SCFAs) to astrocytic glutathione metabolism, but less than 1% of SCFAs reach the brain due to blood–brain barrier heterogeneity (Sharon et al., 2016). Personalized approaches must address two conflicts: (1) discrepancies between systemic biomarkers (e.g., plasma BHB) and intracerebral metabolic states; (2) non-linear associations between genotypes and metabolic phenotypes. Promising strategies include nanoparticle-mediated targeted delivery and CRISPR activation of endogenous metabolic enzymes, combined with AI-driven prediction of individual metabolic responses.

As the field moves towards personalized medicine, understanding individual variability in brain metabolism will be crucial for developing targeted treatments. There are sex-specific differences in brain energy metabolism. Genetic and environmental factors can influence metabolic responses and disease susceptibility. By identifying patient-specific metabolic profiles, researchers can design more effective, personalized therapeutic approaches that address the unique metabolic challenges faced by individuals with neurodegenerative diseases.

Progress in neuronal energy metabolism research has been substantial, yet several critical constraints impede further translation. It is debated whether and to what extent neurons depend on astrocytic lactate, given evidence of their autonomous glycolytic capacity in specific contexts (Tang, 2018; Díaz-García and Yellen, 2019). A major constraint is the predominant use of reduced preparations (e.g., in vitro models, acute slices), which fail to capture the integrated milieu of the intact brain. Additionally, metabolic heterogeneity among neuronal subtypes is widely recognized but difficult to resolve with current spatially-limited techniques. Ultimately, extrapolating findings from animal models of neurodegeneration to humans is complicated by differences in disease chronology and etiology.

The study of astrocyte metabolism is complicated by dynamic functional states and translational gaps. Their substantial regional heterogeneity challenges the concept of a uniform astrocytic phenotype, with direct implications for metabolic support. In neurodegenerative contexts, a critical dichotomy emerges: a loss of support function contrasts with reactive, potentially inflammatory metabolic reprogramming. The molecular triggers and progression of this shift are poorly resolved. These therapeutic ambitions are further tempered by the practical difficulties of achieving cell-type-specific modulation without disrupting homeostasis or compromising delivery across the blood–brain barrier.

While models of astrocyte-neuron metabolic coupling, such as the lactate and glutathione (GSH) shuttles, are well-supported, their universality and pathophysiological relevance remain subjects of debate. For example, the quantitative contribution of the lactate shuttle likely varies by brain region and neural activity, and the interplay between its roles as an energy substrate and a signaling molecule (e.g., in epigenetic regulation) is poorly defined. Technological limitations pose a major barrier: real-time, quantitative in vivo tracking of metabolite flux is hindered by trade-offs between spatiotemporal resolution, specificity, and invasiveness in current methodologies. Regarding the GSH shuttle, despite its central role in antioxidant defense, clinical trials aiming to boost brain GSH levels through precursor supplementation have shown inconsistent outcomes. This is attributable to the blood–brain barrier, variable cellular uptake, and the irreversibility of oxidative damage in late disease stages. Collectively, these gaps highlight the need for next-generation tools capable of precise metabolic mapping and for validation in more human-relevant models, such as brain organoids.

The primary obstacle in microglial metabolism research is its profound plasticity and context-dependency. The classic M1/M2 paradigm is insufficient to describe their phenotypic continuum, making corresponding metabolic states highly dynamic and difficult to target predictably. Compounding this is the poorly understood metabolic crosstalk within the microglia–neuron-astrocyte triad, where signaling mechanisms and functional outcomes are unclear. This functional ambiguity extends to disease, where causal relationships between metabolic reprogramming and pathology are hard to dissect. Translational efforts face an additional barrier: while CSF1R inhibitors allow experimental microglial manipulation, recreating their physiological complexity—including precise spatial distribution and functional heterogeneity—presents a significant challenge for therapeutic development.