Metabolic control of neuronal redox homeostasis: implications of pentose phosphate pathway loss in the nervous system

Neurons have exceptionally high energetic demands and so rely on metabolic support from glial cells, which provide substrates such as lactate to fuel neuronal oxidative metabolism (Bonvento & Bolaños, 2021). Despite this support, neurons continue to take up glucose, suggesting roles beyond ATP production. One consequence of the highly oxidative metabolism of neurons is the generation of reactive oxygen species (ROS), which must be tightly regulated to prevent cellular damage. Maintaining redox balance is therefore critical for neuronal function and survival. One proposed role of neuronal glucose metabolism is to divert glucose into the pentose phosphate pathway (PPP), which generates reducing equivalents in the form of NADPH (Herrero-Mendez et al., 2009). NADPH is essential for antioxidant defence, particularly by regenerating reduced glutathione, a key molecule that neutralizes ROS. Given the relatively limited antioxidant capacity of neurons, PPP-derived NADPH has been hypothesized to play a central role in protecting neurons from oxidative stress (Bonvento & Bolaños, 2021). However, a direct assessment of the functional importance of neuronal PPP activity in vivo has remained challenging, and the extent to which it contributes to maintaining NADPH levels and regulating ROS in the living nervous system remains unclear. In this issue of The Journal of Physiology, Müller et al. (2026) investigate whether neuronal PPP activity is required to maintain redox homeostasis in vivo. Using neuron-specific genetic manipulations in Drosophila and genetically encoded biosensors for NADPH and hydrogen peroxide (H2O2), they examine how disruption of PPP-dependent NADPH production affects oxidative stress and neuronal integrity in the living nervous system. Müller et al. (2026) employed a Drosophila model to investigate the role of neuronal PPP activity in maintaining redox homeostasis in vivo. Neuron-specific knockdown of key PPP enzymes involved in NADPH production, including glucose-6-phosphate dehydrogenase (G6Pdh) and 6-phosphogluconate dehydrogenase (Pgd), was achieved using RNA interference (RNAi), with knockdown efficiency validated at the transcript level. To avoid developmental confounds, gene expression was temporally controlled and induced in adult animals. The authors assessed the consequences of PPP disruption at both structural and functional levels, quantifying neurodegeneration in brain tissue and evaluating behavioural changes using activity-monitoring assays. To probe the contribution of oxidative stress, antioxidant supplementation was used to determine whether redox imbalance underlies the observed phenotypes. To directly examine the metabolic consequences of PPP disruption, Müller et al. (2026) utilized genetically encoded fluorescent biosensors to monitor intracellular redox states in living neurons. NADPH levels were assessed using the iNap1 sensor, whereas H2O2 levels were measured using the HyPerRed sensor. Live imaging was performed in both larval neuronal tissues and adult brain structures, enabling real-time assessment of metabolic changes in vivo. This combination of neuron-specific genetic manipulation with dynamic metabolic imaging allowed Müller et al. (2026) to link PPP activity to NADPH availability and oxidative stress at the cellular level, providing a framework to understand how metabolic pathways influence neuronal integrity. Müller et al. (2026) first established that their genetic manipulation reduced expression of the targeted PPP enzymes, with quantitative real-time PCR showing ∼35–78% reductions in quantitative real-time PCR and Pgd transcript levels across the RNAi lines. They then assessed whether neuronal PPP loss in adult flies caused progressive neurodegeneration. Using semi-thin brain sections and confocal analysis of N-cadherin-labelled optic lobes, neuron-specific knockdown of G6Pdh or Pgd increased the number of degenerative holes in the adult brain, with the phenotype more pronounced by 30 days of age (G6PdhdsRNA50667: P < 0.0001; PgddsRNA65078: P < 0.0001). Antioxidant supplementation with vitamins C and E significantly ameliorated this degeneration (G6PdhdsRNA50667: P = 0.0421; PgddsRNA65078: P = 0.0037), suggesting that oxidative stress contributes to the structural phenotype rather than nonspecific neuronal injury from PPP loss. Müller et al. (2026) next considered whether these structural changes were accompanied by functional impairment. Activity monitoring showed little difference at 15 days, but, by 30 days, PPP knockdown animals displayed reduced peak activity (G6PdhdsRNA50667: P < 0.0001; PgddsRNA65078: P = 0.0002) and increased activity during rest phases (G6PdhdsRNA50667: P = 0.0119; PgddsRNA65078: P < 0.0001), consistent with a progressive loss of normal circadian rhythmicity. The most mechanistically important findings came from live metabolic imaging. To test whether PPP loss directly alters neuronal redox state, Müller et al. (2026) generated flies expressing the NADPH sensor iNap1 and the H2O2 sensor HyPerRed in neurons. In larval eye imaginal disc neurons, PPP knockdown reduced baseline NADPH levels, reaching statistical significance for Pgd knockdown and showing a similar, but non-significant, trend for G6Pdh knockdown (PgddsRNA100269: P = 0.0009). A pH control experiment showed that these changes were not explained by altered intracellular pH. Complementing this, HyPerRed imaging revealed elevated basal H2O2 levels after neuronal PPP disruption, with a significant increase in G6Pdh knockdown tissue (G6PdhdsRNA101507: P = 0.0292) and a confirmed increase for Pgd using an independent RNAi line (PgddsRNA65078: P = 0.0138). Together, these experiments provide direct in vivo support for the proposed metabolic sequence in which impaired PPP activity lowers neuronal reducing capacity and increases oxidative stress. These findings were extended to adult neurons. In antennal lobe neurons imaged 5 days after adult-onset knockdown, Pgd deficiency significantly reduced NADPH levels (PgddsRNA65078: P = 0.0032), whereas G6Pdh knockdown again showed only a tendency toward reduction. At this early adult time point, no increase in H2O2 was detected, probably reflecting insufficient time for ROS accumulation. However, older animals showed increased oxidative lipid damage, measured through 4-HNE staining, following G6Pdh knockdown (G6PdhdsRNA50667: P = 0.0053), suggesting that the consequences of PPP disruption may emerge in stages, with redox imbalance preceding overt oxidative damage. This study provides in vivo evidence linking neuronal PPP activity to redox homeostasis in the nervous system. Müller et al. (2026) demonstrate a functional relationship between reduced PPP activity, decreased NADPH availability and increased oxidative stress in living neurons, extending previous observations largely confined to cell culture systems (Kuehne et al., 2015). These findings establish the PPP as a key contributor to maintaining redox balance in intact neural tissue. These results also refine our understanding of neuronal metabolism. Beyond biosynthetic roles, glucose metabolism through the PPP appears critical for sustaining antioxidant capacity, supporting the idea that neurons use glucose in part to buffer oxidative stress. This highlights the importance of metabolic specialization within neuron–glia interactions and suggests that redox regulation is a central function of neuronal glucose utilization. These findings are also relevant to ageing and neurodegenerative disease, where oxidative stress is a common feature. The observation that PPP disruption induces progressive neurodegeneration, partially rescued by antioxidant supplementation, suggests that impaired NADPH homeostasis may contribute to neuronal vulnerability. However, the relatively mild phenotype observed indicates that compensatory NADPH-producing pathways may mitigate PPP loss under baseline conditions. Several limitations also remain. Differences between G6Pdh and Pgd knockdowns suggest pathway-specific regulation or compensation, and redox imaging in adult neurons was restricted to early time points. Future studies should investigate long-term redox dynamics, interactions with alternative NADPH sources and whether similar mechanisms operate in mammalian systems. Müller et al. (2026) provide direct in vivo evidence that neuronal PPP activity is essential for maintaining NADPH levels and redox balance in the nervous system. By combining genetic manipulation with live metabolic imaging, the study establishes a functional link between impaired PPP activity, oxidative stress and progressive neuronal dysfunction. These findings advance our understanding of neuronal metabolism by highlighting a critical role for glucose utilization in redox homeostasis and underscore the importance of metabolic pathways in shaping neuronal health and vulnerability. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. No competing interests declared. All authors have approved the final version of the manuscript submitted for publication. All authors listed meet the criteria for authorship, and no qualified contributors have been omitted. No funding received.