Synaptic Protein Lactylation: A Novel Mechanism Underlying Physical Exercise-mediated Stress Resilience

The health benefits of physical exercise are well established and have been observed in both human studies and rodent models [1], improving overall health and stress resilience. However, the underlying molecular mechanisms have not been comprehensively investigated. Previous studies have focused extensively on its neuromodulatory effects and have also identified multiple exercise-associated molecular substrates and blood-borne metabolites, including neurotrophic factors, monoamine neurotransmitters, neuroinflammatory cytokines, kynurenine, N-lactoyl-phenylalanine, and the ketone body β-hydroxybutyrate [2]. Notably, lactate, a common energy source derived from cellular glycolysis in response to intensive exercise, has recently been reported to exert antidepressant activity [3]. However, a detailed mechanistic explanation is lacking.

In addition to its canonical function as a cellular energy source, lactate is involved in several biological processes. A recent study has revealed that lactate functions as an antidepressant, mediating stress resilience by modulating the levels and activity of histone deacetylases in the hippocampus [3]. Histone lactylation in neural tissues primarily responds to neural excitation [4] and participates in brain injury and neurodegenerative diseases [5]. More importantly, exercise significantly increases the lactate levels in both blood and brain, suggesting that exercise-mediated protein lactylation patterns may confer stress resilience [6].

In a recent study published in Cell Metabolism, Yan et al. demonstrated that treadmill exercise prevents anxiety-like behaviors in chronic restraint stress (CRS) mice by elevating circulating and brain lactate levels. Proteomic analysis revealed an exercise-induced lactylation of synaptosome-associated protein 91 (SNAP91). Further combinations of genetic manipulation and anatomical, physiological, and behavioral assays indicated that SNAP91 lactylation maintains synaptic structure and function, contributing to stress resilience and exercise-induced anxiolytic effects (Fig. 1) [7]. This study uncovered a novel non-histone lactylation that modulates mental functions and elucidated the brain’s metabolic adaptation to exercise.

Fig. 1.

Schematic of how physical exercise-induced lactate contributes to stress resilience and anxiolysis. A Treadmill exercise-induced lactate prevents anxiety-like behaviors in a CRS model. B Exercise-induced lactate potentiates the lactylation of synaptic protein SNAP91 specifically in the mPFC. C Exercise-enhanced SNAP91 lactylation improves synaptic structure and function specific to the mPFC. CRS, chronic restraint stress. SNAP91, synaptosome-associated protein 91. mPFC, medial prefrontal cortex.

To investigate the underlying relationship between the lactate and exercise-induced anxiolytic effects, the authors chose a CRS model combined with 14 d-treadmill exercise. Surprisingly, after chronic exercise, the lactate levels were increased in the liver, muscle tissue, and medial prefrontal cortex (mPFC). Additionally, both the mRNA and protein levels of lactate dehydrogenase A/B (LDHA/B), the key enzymes for lactate biogenesis, were upregulated in mPFC tissue after exercise. They further showed that exercise-induced anxiolytic effects required lactate production by performing an intravenous injection of L-lactic acid instead of exercise. In contrast, administration of dichloroacetate, a potent inhibitor of lactate production, reversed the exercise-induced anxiolytic effects.

The authors then examined the lactylation pattern of neural tissue during exercise training. A total protein blotting with a pan-lactylation antibody revealed elevated lactylation levels during exercise. Utilizing a lactylation-based proteomics assay, they found prominent changes in the cytoskeleton and synapses within a cluster that exhibited reduced lactylation under CRS but increased abundance during exercise. Through bioinformatics analysis and immunoprecipitation, they further demonstrated that exercise training potentially increased lactylation of the synaptic protein SNAP91. Moreover, they found that exercise-induced SNAP91 lactylation improved the integrity of synaptic function and structure. Interestingly, such a lactylation pattern was found to be region-specific to the mPFC. When L-lactic acid was injected into CRS mice, only the prelimbic (PrL) region in the mPFC exhibited an increased Kla-SNAP91 level, whereas neither the hippocampus nor basolateral amygdala (BLA) showed a significant response to CRS or lactate injection. This finding is consistent with recent research revealing that the mPFC is sensitive to lactate metabolism in depressive disorders [8].

The authors proceeded to reveal the causal relationships between the synaptic protein lactylation and neuronal function. Using a human synaptophysin (hSyn)-driven adeno-associated virus (AAV) expressing a site-mutated SNAP91 (SNAP91-K885R), in which the lactylation potency at the K885 site was impaired, they first established a mouse model expressing a loss-of-function SNAP91 mutant (K885R) in the mPFC, leading to decreased SNAP91 lactylation. Disruption of synaptic vesicular structures and decreased presynaptic vesicle density were observed in the K885R mutant. Notably, these changes were associated with a marked reduction in mPFC-specific synaptic function, as shown by the ex vivo electrophysiological recording, in vivo two-photon calcium imaging, and behavioral assays. These data indicate that lactate predominantly mediates SNAP91 lactylation specifically in the mPFC, thus conferring stress resilience.

Using the established K885R mutant form, the authors further investigated whether synaptic protein lactylation is necessary for the exercise-induced anxiolytic effect. The quantification analysis showed decreased SNAP91 lactylation even in mice subjected to exercise training, while the total SNAP91 level did not change. By analyzing the synaptic ultrastructure, a reversal of exercise-induced vesicle formation at presynaptic sites, and reduced synaptic proteins or deformation of the postsynaptic density were found in exercised mice expressing the SNAP91(K885R) mutant. Furthermore, in vivo calcium imaging identified reduced neuronal calcium activity under impaired lactylation of synaptic proteins. In line with the behavioral tests, these data suggest that deficits in mPFC-specific synaptic protein lactylation abolish exercise-mediated anxiolytic effects.

Next, to eliminate potential confounding effects of overexpression of mutant SNAP91, the authors used Cas9-mediated knock-in of the SNAP91 (K885R) mutant to establish an mPFC-specific in vivo gene editing model. Notably, this manipulation reduced SNAP91 lactylation by approximately 50%–80% without altering the endogenous total SNAP91. Furthermore, transmission electron microscopy revealed a decreased density in synaptic vesicles, and behavioral assays verified the ablation of exercise-induced anxiolytic effects in this mutant model. Together, these findings highlight a novel function of non-histone protein lactylation in regulating stress resilience, likely through synaptic transmission and neural network modulation. More importantly, exercise training restores the homeostasis of synaptic protein lactylation, thus mitigating the CRS and anxiety-like behaviors.

To our knowledge, this is the first study to identify a specific lactylation site on the synaptic protein SNAP91 and the non-histone lactylation in the brain. However, several issues are worthy of further discussion. First, this study expands our understanding of lactate's dual functions, highlighting its post-translational modification of a synaptic protein. However, research on the non-metabolic functions of brain lactate in this study is still in its early stages, particularly regarding the specific mechanisms of its involvement in protein post-translational modification. Moreover, further studies are needed to delve deeper into the multiple functions of lactate in the brain and the underlying molecular mechanisms. Second, the authors have verified the association between chronic stress and lactate production, providing a new perspective for understanding the impact of stress on brain metabolism. However, the specific molecular mechanisms still require further exploration. Third, given the immunomodulatory effect of non-histone proteins, the authors underscored that the study of non-histone lactylation in the brain remains an underexplored field, and their work helps address this gap. However, the change in lactylation of histone proteins in the brain has not been assayed in this study, which determines whether exercise specifically induces lactylation of non-histone proteins. Exercise-induced epigenetic regulation, including DNA methylation, RNA methylation, and histone lactylation, has been elucidated as critical for long-term brain function and behavior. Although histone lactylation has been shown to play an important role in regulating gene expression involved in various biological processes, its role and mechanisms in stress resilience during exercise remain unclear. More importantly, non-histone lactylation and histone lactylation may act synergistically to mediate gene expression networks in the brain. Fourth, the authors have not investigated the gender-specific effects, overlooking the differences in lactate production during exercise and susceptibility to mental health disorders between females and males [9]. Thus, the female-specific patterns of non-histone protein lactylation are suggested to be further investigated. Fifth, considering that blood-borne lactate can be transported across different brain regions, the authors also investigated two other anxiety disorder-related brain regions besides the mPFC, the dorsal hippocampus and amygdala. However, previous studies have consistently demonstrated functional heterogeneity along the dorsal-ventral axis of the hippocampus [10]. Therefore, genetic manipulations of SNAP91 lactylation in the ventral hippocampus are suggested in future work, which may help unveil the functional specialization of the dorsal and ventral hippocampus. Last and most importantly, the authors demonstrated that lactylation at K885, a site located near the C-terminal of SNAP91, affects synaptic structure and function. However, the underlying molecular mechanisms remain poorly understood. For example, further investigation is warranted to determine whether lactylation at the K885 induces steric hindrance, conformational changes, or steric hindrance in SNAP91, thus modulating its function. Moreover, given that lactylation and acetylation both modify the lysine residues, with partial overlap in their target sites, indicating that these two modifications may engage in intricate interactions or competition, and affect SNAP91 degradation or its stability. Therefore, additional experiments are required to rule out potential crosstalk between SNAP91 lactylation and other post-translational modifications. Furthermore, the lactylation modification sites also colocalize with critical nodes in multiple signaling pathways, necessitating the exclusion of potential confounding effects arising from these crosstalks.

Overall, this study highlights the critical role of lactate as an intracellular signal in modulating cortical synaptic structure and function. Synaptic protein lactylation-induced changes in cortical neural networks confer enhanced stress resilience during persistent exercise. These findings reveal the non-metabolic neuromodulatory effect of lactate and a novel mechanism underlying the brain's metabolic adaptation under exercise training.

Conflict of interest

The authors declare no conflicts of interest.