Trained immunity attenuated acute lung injury by activating alveolar macrophages via AKT2-PDK1 axis-mediated metabolic reprogramming

Abstract

Background

Acute lung injury (ALI) is a life-threatening clinical syndrome typically triggered by sepsis or severe trauma lacking effective treatment options. Alveolar macrophages (AMs), representing the most abundant immune cell population in pulmonary tissue, exhibited functional abnormalities that were closely associated with ALI pathogenesis. Notably, elevated pulmonary lactate secretion served not only as a characteristic pathological feature of ALI but also participated in disease progression through modulation of AMs activity. Trained immunity was found to activate innate immune cells including macrophages, regulating metabolic adaptations that alleviated ALI, though the precise mechanisms remained unclear.

Methods

We used β-glucan and LPS to establish both in vivo and in vitro models of trained immunity and ALI, enabling investigation of trained immunity effects on AMs immunoregulatory functions.

Results

The results demonstrated that trained immunity effectively attenuated ALI severity by up-regulating glycolytic activity in AMs, thereby potentiating their immune responsiveness, and primarily enabled alveolar macrophages to sustain immune responses in high-lactate environments through the AKT2-PDK1 axis, an effect that was abolished by relevant inhibitors.

Conclusions

We concluded that β-glucan induced trained immunity could enhance alveolar macrophage immune activity and improve lactate metabolic tolerance, offering a novel therapeutic approach for acute lung injury (ALI).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-06879-4.

Introduction

Acute lung injury (ALI) is a severe pulmonary inflammatory disorder triggered by endogenous and exogenous pathogenic factors [1]. First described in 1967, ALI is associated with high mortality rates and is characterized clinically by non-cardiogenic pulmonary edema and acute hypoxemia [2, 3]. The condition arises from an acute inflammatory response in the alveolar and pulmonary parenchyma, driven primarily by disruption of the alveolar-capillary barrier [4]. This damage initiates an inflammatory cascade involving macrophage and neutrophil infiltration (induced by infection or injury) and the release of pro-inflammatory mediators such as TNF-α, IL-1β, and IL-6. Notably, coronaviruses including SARS and COVID-1, have been shown to exacerbate ALI through severe lung infection [2, 5]. In progressive cases, ALI can escalate to acute respiratory distress syndrome (ARDS) [6].Currently, clinical management of ALI/ARDS lacks effective targeted therapies, and existing treatments often entail adverse effects, such as immune dysfunction and coagulation abnormalities [7]. Furthermore, studies confirm that lung injury can be exacerbated by immune responses to infectious agents like lipopolysaccharide (LPS). Given these challenges, there is an urgent need to explore novel therapeutic mechanisms to mitigate ALI-associated critical respiratory diseases in clinical practice.

In the context of ALI pathology, macrophages served as crucial immunomodulatory hubs, continuously monitoring and responding to microenvironmental cues to maintain physiological equilibrium. Alveolar macrophages (AMs), as the lung's primary immune sentinels, played dual roles in both maintaining pulmonary homeostasis and coordinating inflammatory responses [8]. During ALI, AMs demonstrated remarkable functional plasticity, capable of adopting either pro-inflammatory (M1) or anti-inflammatory (M2) phenotypes depending on local signals.It was reported that β-glucan induced systemic trained immunity conferred memory-like properties to innate immune cells, influencing disease pathogenesis and regression through the regulation of M1/M2 macrophage balance, sustained epigenetic modifications, and metabolic reprogramming [9]. Trained immunity referred to a phenomenon wherein the innate immune system returned to baseline after initial exposure to a stimulus (vaccine, pathogen, or metabolite) but exhibited an enhanced response upon secondary challenge. This heightened immune reactivity was primarily mediated by monocytes and macrophages [10]. Furthermore, studies demonstrated that alveolar macrophage (AMs) mediated trained immunity had a potential role in modulating lung malignancy progression during acute influenza virus infection [8]. Based on these findings, we hypothesized that localized β-glucan administration via lung perfusion might induce trained immunity, thereby converting immune tolerance into immune activation under pathological conditions characterized by acidosis and hypoxia.

A growing body of evidence highlights the critical relationship between trained immunity and cellular metabolism, with metabolites serving as key mediators in numerous biological processes. Lactate, the terminal product of glycolysis, has evolved from being considered merely a metabolic waste product to being recognized as a multifunctional signaling molecule that modulates diverse physiological and pathological processes [11]. The seminal discovery of the Warburg effect in 1921 first revealed the metabolic reprogramming associated with Lactate production. Subsequent research has established that intercellular lactic acid shuttling is facilitated by monocarboxylate transporters (MCTs), and that MCT mediated Lactate transport participates in immune cell recognition and subsequent immune response modulation. Mechanistically, Lactate exerts its effects through allosteric binding to various molecular targets, influencing processes ranging from lipid metabolism to macrophage-mediated adipose tissue inflammation [12]. In the context of acute lung injury (ALI), the characteristic metabolic shift toward glycolysis results in substantial lactic acid accumulation. This metabolic perturbation may exacerbate lung injury through multiple pathways, including mitochondrial dysfunction and altered macrophage polarization. Specifically, the Lactate-enriched microenvironment has been shown to significantly impact the dynamic equilibrium between pro-inflammatory M1 and anti-inflammatory M2 macrophage phenotypes [13, 14]. These findings position lactic acid metabolism as a promising therapeutic target and a novel research direction in ALI pathogenesis and treatment.

Pyruvate dehydrogenase kinase 1 (PDK1) is a critical serine/threonine kinase, composed of an N-terminal serine/threonine kinase domain responsible for regulating the phosphorylation of the E1α subunit of the pyruvate dehydrogenase complex (PDC), and a C-terminal lipoyl-binding domain that facilitates its interaction with the mitochondrial PDC complex. Notably, PDK1 lacks a PH domain [15].Previous studies have demonstrated that a key metabolic feature of trained immunity is the up-regulation of glycolysis. PDK1 (pyruvate dehydrogenase kinase 1) serves as a critical regulator of glucose metabolism by phosphorylating and inhibiting pyruvate dehydrogenase (PDH), thereby shifting cellular metabolism from oxidative phosphorylation to glycolysis [16]. This metabolic reprogramming is essential for enhancing the effector functions of immune cells. Additionally, emerging evidence suggests a potential link between PDK1 and trained immunity.In addition to, PDK1 demonstrated dual immunomodulatory effects: it regulated M1/M2 macrophage polarization while simultaneously modulating T-cell subset differentiation [17].Mechanistically, PDK1 contributed to the pathogenesis of inflammatory and immune-related diseases by orchestrating glycolysis, mitochondrial metabolism, immune cell activation, and inflammatory mediator release.The AKT signaling pathway, which is closely linked to trained immunity, was further implicated in this process. AKT2, a cytoplasmic member of the AKT protein family, exhibited phosphorylation patterns associated with PDK1-related pathological conditions. Notably, both PDK1 and AKT2 have been mechanistically tied to dysregulated immune responses, yet their interplay in acute lung injury remained poorly understood. These findings underscored the need for further investigation into the AKT2-PDK1 regulatory axis in the context of acute inflammatory lung injury.

Our study builds upon existing evidence to demonstrate that β-glucan induced trained immunity enhances macrophage responsiveness to secondary stimuli, significantly boosting host defense against infections and disease [9]. Importantly, we identified a novel protective role of trained immunity in acute lung injury (ALI), mediated through metabolic reprogramming. Specifically, trained immunity potentiates glycolytic flux while concomitantly increasing cellular tolerance to lactate accumulation—adaptations that correlate strongly with up-regulation of the glycolytic kinase PDK1 [18]. Mechanistically, we establish that PDK1 expression is regulated by AKT2, and that trained immunity activates this AKT2-PDK1 axis to mitigated lung injury while promoting a pro-inflammatory (M1-like) macrophage phenotype. Crucially, these protective effects are abolished by pharmacological inhibition of this pathway. Collectively, our findings reveal that trained immunity confers protection against early ALI through the AKT2-PDK1 signaling axis, highlighting its therapeutic potential in acute inflammatory lung diseases.

Results

Trained immunity improves LPS induced ALI and lactic acid accumulation

The pathogenesis of acute lung injury (ALI) involves the disruption of the alveolar-endothelial barrier, with alveolar macrophages playing a central role in modulating pulmonary inflammatory responses. Given the emerging evidence that trained immunity can reprogram innate immune cells to mount enhanced responses to subsequent challenges, we sought to investigate the potential protective role of trained immunity in ALI through its effects on alveolar macrophage function.To address this question, we established model of ALI using C57BL/6 mice, allowing us to systematically examine the interplay between trained immunity and alveolar macrophage polarization in the context of acute lung inflammation.6–8-week-old C57BL/6 mice were pretreated via airway perfusion with β-glucan (1 mg/kg) or PBS on days -7 and -4, followed by LPS challenge (5 mg/kg) administered through the same route on day 0 (Fig. 1a). Histopathological analysis revealed significantly exacerbated lung injury in LPS challenged mice compared to controls, characterized by alveolar space widening, structural disintegration, and extensive inflammatory cell infiltration in alveolar and interstitial compartments.Notably, trained immunity preconditioning substantially attenuated these pathological alterations, reducing lung injury scores by versus LPS only group (Fig. 1b). Immunofluorescence profiling demonstrated that trained immunity promoted M1-polarized macrophage accumulation in lung tissues (Fig. 1c) To assess lactate and inflammatory factor levels in the ALI model, bronchoalveolar lavage fluid (BALF) was collected. The results showed that lactate, TNF-α and IL-1β levels were significantly elevated following trained immunity (Fig. 1D, E), indicated a strong association between pulmonary inflammation and trained immunity. To further elucidate the role of alveolar macrophages in this process, QPCR analysis of cultured primary alveolar macrophages revealed enhanced expression of inflammatory factors concurrent with M1-type polarization (Fig. 1F). Collectively, our data demonstrated that trained immunity potentiates lactate-driven metabolic reprogramming and M1 macrophage polarization, paradoxically ameliorating LPS induced ALI through controlled proinflammatory activation.

Fig. 1.

Trained immunity improves LPS induced ALI and lactic acid accumulation. A An in vivo ALI mouse model was established using airway perfusion of 5 mg/kg LPS, with pretreatment of 1 mg/kg β-glucan or PBS (n = 3). B Representative H&E-stained lung sections from each group are shown (scale bar: 20 μM n = 3). C Macrophage activation was assessed by immunofluorescence staining of lung sections. D, E Lactate and inflammatory factor levels were measured in bronchoalveolar lavage fluid (BALF). F QPCR analysis of inflammatory factor mRNA levels in alveolar macrophages (n = 3).Statistical significance: TI + LPS vs. LPS vs. control groups (two-tailed t-test):p < 0.05, *p < 0.01, **p < 0.001, ***p < 0.0001

β-glucan mediated trained immunity enhances lactate tolerance and immune function in AMs

Our study builds upon established evidence that trained immunity regulates glycolytic pathways and metabolic reprogramming. Consistent with our in vivo findings of elevated lactate levels following trained immunity induction (Fig. 1D), we observed significantly increased lactate concentrations in alveolar macrophage (AMs) supernatants from trained mice compared to both LPS treated and control groups (Fig. 2A). To characterize the metabolic adaptations underlying this phenomenon, we exposed cells to graded concentrations of exogenous lactate. Trained AMs exhibited a concentration-dependent enhancement of lactate tolerance (Fig. 2B), concomitant with upregulated expression of inflammatory mediators TNF-α and IL-6 at both transcriptional and translational levels (Fig. 2C–E). Flow cytometric analysis revealed that trained immunity conferred cellular protection against lactate-induced stress, as evidenced by significantly reduced apoptosis rates compared to LPS treated controls (Fig. 2F). These findings collectively demonstrated that β-glucan induced trained immunity enhances immunocompetence through metabolic reprogramming that promoted lactate tolerance while sustaining inflammatory responses, thereby optimizing cellular immune function under metabolic stress conditions.

Fig. 2.

β-glucan mediated trained immunity enhances lactate tolerance and immune function in alveolar macrophages (AMs). A Lactate levels in cell supernatant p < 0.05, **p < 0.001 vs. control group; #p < 0.05 (LPS vs. TI + LPS groups). (Paired Student’s t-test) B Cell viability (CCK-8 assay) of LPS and TI + LPS groups treated with increasing lactate concentrations. C QPCR analysis of inflammatory factor expression in LPS + lactate (10 μM) and TI + LPS + lactate groups. D Protein expression measured by Western blot. E Apoptosis rate was significantly reduced in the TI + LPS + lactate group vs. LPS + lactate group (n = 3):p < 0.05, **p < 0.001.(Paired Student’s t-test for LPS + lactate vs. TI + LPS + lactate)

β-glucan induced trained immunity ameliorates acute lung injury via PDK1-dependent metabolic modulation

Building on evidence that trained immunity modulates immune responses through metabolic reprogramming, we investigated its role in acute lung injury (ALI).Transcriptomic analysis of GSE187664 and GSE57206 datasets revealed that trained immunity upregulates glycolysis-related gene expression.Analysis of the expression of glycolysois-related kinases at the mRNA level revealed that PDK1 is a key regulatory factor for metabolic reprogramming and immune cell activation. (Fig. 3B, supplemental Fig. 1 A-B).Immunohistochemistry demonstrated significantly elevated PDK1 expression in lung tissues from trained immunity mice compared to LPS treated controls (Fig. 3C). To validate this in vitro, we established a trained immunity model by macrophages with β-glucan (5 μg/mL, 24 h), resting them in low-serum medium (2%, 5 days), and stimulating with LPS (100 ng/mL, 6–8 h). QPCR confirmed that trained immunity upregulated PDK1 expression alongside proinflammatory cytokines (TNF-αIL-6; Fig. 3A, D). Given reports of mitochondrial AKT2-PDK1 interaction in tumors, we hypothesized a similar mechanism in trained immunity. Western blot analysis revealed that β-glucan preconditioning enhanced AKT2 activation and PDK1 expression upon LPS challenge, correlating with increased TNF-αand IL-6 production (Fig. 3E, F). Immunofluorescence further confirmed mitochondrial recruitment of AKT2 and its colocalization with PDK1 in trained cells (Fig. 3G, H). Collectively, these findings demonstrate that trained immunity alleviates ALI via the AKT2-PDK1 axis, linking metabolic adaptation to inflammatory regulation.

Fig. 3.

β-glucan Induced Trained Immunity Ameliorates Acute Lung Injury via PDK1-Dependent Metabolic Modulation. A An in vitro trained immunity ALI model was established. B Volcano map analysis revealed significant up-regulation of PDK1 in group A (TI) expression in the database. C PDK1 expression was assessed by immunohistochemical staining of mouse lung sections. (D) QPCR analysis showed differential mRNA expression of inflammatory factors between LPS and trained immunity (TI) groups (n = 3). E, F Western blotting detected activation levels of PDK1, its upstream regulators, and lung injury markers (n = 3). G, H Immunofluorescence demonstrated: Mitochondrial recruitment of p-AKT2 following trained immunityCo localization of p-AKT2 and PDK1Statistical significance:**p < 0.001, ***p < 0.0001 vs. control group ####p < 0.0001 (LPS vs. TI groups; paired Student’s t-test)

PDK1 inhibitors affect the improvement of ALI by trained immunity

To further elucidate PDK1 role in trained immunity, we employed the PDK1 inhibitor AZD7545 (10 μM, 24 h pretreatment) prior to β-glucan induction (Fig. 4A). AZD7545 significantly suppressed lactate secretion compared to the trained immunity group (Fig. 4B), and CCK-8 assays revealed impaired cellular proliferation under exogenous lactate stress, indicating that PDK1 inhibition disrupts metabolic adaptation critical for trained immunity (Fig. 4C). At the protein level, SiRNA and AZD7545 treatment reduced PDK1 expression and TNF-α production, as confirmed by Western blot and ELISA (Fig. 4D, E, supplemental Fig. 2A). Consistent with this, QPCR analysis showed diminished expression of M1-polarization markers (Fig. 4F, supplemental Fig. 2B),suggesting PDK1 drives both inflammatory and metabolic reprogramming.Intracellular phagocytosis and endocytosis were detected by macrophage phagocytosis test, and it was found that phagocytosis decreased significantly after inhibitor treatment (Fig. 4G). Given PDK1 mitochondrial localization,we assessed mitochondrial integrity using fluorescence probes.AZD7545 pretreatment abolished the protective effects of trained immunity against LPS induced mitochondrial damage, manifesting as fragmented networks versus controls (Fig. 4H).Collectively, these data demonstrated that PDK1 was indispensable for trained immunity establishment, as its inhibition blocks lactate-driven metabolic shifts, inflammatory responses, and mitochondrial homeostasis, ultimately compromising lung injury resolution.

Fig. 4.

PDK1 inhibitors affect the improvement of ALI by trained immunity. A Cells were pretreated with AZD7545 or CCT128930 for 24 h to establish inhibitor models. B Lactate levels in supernatant were significantly reduced in inhibitor-treated groups. C CCK-8 assay revealed decreased cell viability under high lactate conditions after inhibitor treatment. D, E Western blot and ELISA showed reduced protein expression of target genes. F qPCR confirmed decreased mRNA levels of PDK1 and inflammatory factors in inhibitor-treated groups. G Phagocytic capacity was impaired in inhibitor-treated groups compared to TI group. H Mitochondrial morphology alterations were observed by MitoTracker staining. Statistical significance:*p < 0.05, **p < 0.01, ****p < 0.0001 (vs control group; paired Student's t-test)$ p < 0.05, ##p < 0.01, ###p < 0.001, ####p < 0.0001 (TI + LPS + AZD7545 vs TI + LPS; paired Student's t-test)

Akt2 phosphorylation promotes the activation of PDK1 in ALI

On the observed correlation between AKT2 and PDK1, we next employed the AKT2-specific inhibitor CCT128930 to investigate whether the protective effects of trained immunity against acute lung injury are mediated through the AKT2-PDK1 axis. To confirm the metabolic interplay underlying this protection, we quantified lactate—a key metabolic byproduct of trained immunity—across experimental groups. Comparative analysis revealed that CCT128930 treatment significantly reduced lactate levels in cellular supernatants compared to both trained immunity (TI) and control groups, concurrently impairing glycolytic flux (Fig. 5A). These findings establish that AKT2 inhibition disrupts the metabolic reprogramming central to trained immunity's therapeutic efficacy. Exogenous lactate supplementation in CCK-8 assays mirrored the effects observed with PDK1 inhibition, showing comparable impairment of cellular viability (Fig. 5B). Western blot analysis further demonstrated that SiRNA and CCT128930 treatment significantly attenuated TNF-α expression, concurrently reversing AKT2-mediated PDK1 activation—a finding corroborated at the transcriptional level through mRNA quantification (Fig. 5C–E, supplemental Fig. 2C).ELISA quantification of cell supernatants revealed parallel reductions in TNF-α secretion (Fig. 5F), confirming that AKT2 inhibition suppresses both metabolic and inflammatory components of trained immunity.Critically, this dual suppression impaired lung injury repair mechanisms, as evidenced by compromised resolution of alveolar damage. These results collectively demonstrate that CCT128930 disrupts trained immunity's therapeutic efficacy by targeting the AKT2-PDK1 axis, thereby uncoupling metabolic adaptation from inflammatory regulation. Furthermore, trained immunity enhanced macrophage phagocytic capacity against E. coli, (Fig. 5G). Building on prior evidence that AKT2 activates PDK1 through mitochondrial recruitment, immunofluorescence analysis revealed that CCT128930 treatment significantly diminished P-AKT2-mitochondria colocalization (Fig. 5H). This subcellular redistribution indicated that AKT2 inhibition disrupts mitochondrial targeting of P-AKT2. Importantly, phospho-AKT2 (Ser474) levels correlated with disease severity in acute lung injury models, demonstrating the functional significance of AKT2 phosphorylation status in trained immunity-mediated tissue repair.

Fig. 5.

Akt2 phosphorylation promotes the activation of PDK1 in ALI. A Lactate secretion levels were measured following CCT128930 treatment and compared with the TI group. B CCK-8 assay demonstrated reduced cellular lactate tolerance after CCT128930 treatment. C–E Western blot and QPCR analyses revealed decreased PDK1 and inflammatory marker expression. F TNF-α ELISA showed significant reduction after CCT128930 treatment. G CCT128930 treatment impaired phagocytic capacity. H Immunofluorescence confirmed decreased p-AKT2 mitochondrial colocalization. Statistical significance:*p < 0.05, **p < 0.01, ****p < 0.0001 (TI + LPS + CCT128930 vs TI + LPS vs control; paired Student's t-test)##p < 0.01, ####p < 0.0001 (TI + LPS + CCT128930 vs TI + LPS; paired Student's t-test)

Inhibitors affect trained immunity and increase lung injury in mice

To validate the therapeutic potential of the AKT2-PDK1 axis in vivo, we established an acute lung injury (ALI) model with pharmacological inhibition (Fig. 6A). Histopathological analysis revealed that β-glucan induced trained immunity failed to attenuate lung injury in inhibitor pretreated mice, as evidenced by exacerbated alveolar damage increase in injury scores vs non-inhibitor group, and diminished PDK1 expression (Fig. 6B). Biochemical analysis of bronchoalveolar lavage fluid (BALF) showed concordant reductions in lactate and inflammatory cytokines (TNF-αIL-1β), confirming systemic impairment of trained immunity's metabolic-inflammatory coupling (Fig. 6C, D). QPCR profiling of alveolar macrophages demonstrated significant downregulation of TNF-α and IL-6 in inhibitor-treated groups (Fig. 6E). Immunofluorescence further revealed that CCT128930 and AZD7545 co-treatment suppressed M1 macrophage polarization, disrupting inflammatory regulation (Fig. 6F). These findings conclusively demonstrate that trained immunity alleviates LPS induced ALI through AKT2-PDK1-mediated coordination of metabolic reprogramming and immune activation, positioning this axis as a promising therapeutic target for inflammatory lung diseases.

Fig. 6.

Inhibitors affect trained immunity and increase lung injury in mice. A An in vivo mouse ALI model was established. B H&E and immunohistochemical staining assessed lung injury and PDK1 expression. C, D Lactate and TNF-α levels were measured in bronchoalveolar lavage fluid. E QPCR quantified inflammatory factor mRNA expression in alveolar macrophages. F Macrophage polarization in lung tissue was evaluated by immunofluorescence.***p < 0.001, ####p < 0.0001 ^^p < 0.01, ^^^p < 0.001, ^^^^p < 0.0001 Compared with the control group, the paired students were compared with TI + LPS + AZD7545 group, TI + LPS + CCT128930 group and TI + LPS group by T-test

Discussion

Acute lung injury (ALI) represents a life-threatening clinical syndrome characterized by diffuse alveolar epithelial damage and pulmonary vascular dysfunction, with mortality rates exceeding 35% in severe cases [19]. Notably, ALI shares metabolic similarities with cancer progression, particularly in lactate accumulation-mediated glycolytic dysregulation [20]. Concurrently, disruption of the body's microenvironment triggers widespread immune cell activation and subsequent release of proinflammatory cytokines and chemokines [21]. Emerging evidence indicates that vaccine-induced trained immunity in macrophages can provide enhanced protection against acute pulmonary bacterial infections.This innate immune memory effect exhibits long-term persistence and significantly reduces bacterial burden in infected tissues. Notably, respiratory mucosal exposure to LPS has been shown to establish trained immunity capable of preventing Streptococcus pneumoniae infection, though its efficacy against SARS-CoV-2 infection remains limited [22]. Current research on trained immunity in acute lung injury primarily focuses on viral infections, demonstrating that this phenomenon promotes neutrophil and macrophage population expansion while up regulating phagocytosis and cytotoxicity related genes to achieve protective effects [23].Recent studies have revealed the therapeutic promise of trained immunity (TI), particularly β-glucan induced systemic TI, which drives metabolic reprogramming and epigenetic remodeling in alveolar macrophages to mitigate disease progression. To investigate potential spatial specificity of this phenomenon, we conducted examining whether localized trained immunity elicited through β-glucan perfusion would mirror the therapeutic effects observed in systemic TI implementation. This process involves glycolytic flux enhancement leading to lactate accumulation, TCA cycle remodeling, and histone modifications (H3K4me3, H3K27ac), collectively priming cells for enhanced immune responsiveness. Building on these findings, current research focused on harnessing trained immunity to modulate lactate metabolism and improve cellular acid stress tolerance, representing a promising new approach for managing inflammatory diseases. It should be noted that our study primarily centered on the functional contributions of alveolar macrophages in acute lung injury (ALI), while the involvement of other macrophage populations—notably interstitial macrophages—has yet to be systematically investigated [24]. Additionally, although recent evidence indicates that viral-derived pathogens may mitigate lung injury through T-cell-dependent trained immunity pathways, our experimental design did not account for potential interactions with adaptive immunity mechanisms, representing a limitation of the current work.

It is understood that mitochondrial-localized PDK1, a key metabolic kinase, plays a pivotal role in both pulmonary function regulation and trained immunity induction, with its activation being mediated by the upstream regulator AKT2 [25, 26]. During trained immunity establishment, AKT2 was phosphorylated and subsequently translocated to mitochondria, where it orchestrates immunometabolic adaptations. While our findings established the critical role of the AKT2-PDK1 axis in trained immunity, two fundamental questions remain unresolved: the precise molecular triggers of AKT2 activation during trained immunity induction, and the mechanistic basis of PDK1-mediated acid resistance in immunometabolic reprogramming. These knowledge gaps will guide our future investigations into optimizing trained immunity for therapeutic applications.

This study investigates the interplay between lactate, trained immunity, and acute lung injury (ALI). As a key glycolytic metabolite, lactate serves as both a clinical biomarker and a functional regulator in disease pathogenesis [27]. Dysregulated lactate metabolism contributes to systemic metabolic imbalance, with elevated levels exacerbating inflammatory responses and potentiating sepsis-induced lung injury [28]. Notably, clinical studies have established clear associations between lactate accumulation and pulmonary fibrosis progression. Therefore, We hypothesize that trained immunity enhances the glycolytic activity of alveolar macrophages, enabling them to adapt to the lactation-rich microenvironment during acute lung injury (ALI).This metabolic reprogramming not only sustains macrophage viability under acidic stress but also potentiates neutrophil mediated pathogen clearance through lactate-dependent signaling [29, 30]. Building on these observations, future investigations will elucidate lactate pleiotropic effects across distinct macrophage subsets and their implications for ALI progression and resolution.

Macrophage metabolic reprogramming during cellular growth and differentiation involves dynamic shifts in core metabolic pathways, particularly glycolysis, oxidative phosphorylation (OXPHOS), and the tricarboxylic acid (TCA) cycle. As the terminal product of glycolysis and a pleiotropic signaling molecule, lactate plays a multifaceted role in regulating macrophage reprogramming through diverse molecular mechanisms [31]. In the pathophysiological microenvironment of acute lung injury (ALI), hypoxia cooperates with lactate accumulation to promote metabolic adaptation by inhibiting PHD2 activity, thereby stabilizing HIF-1α and enhancing expression of key glycolytic enzymes (HK2, PKM2, LDHA) [32]. Furthermore, lactate serves as a metabolic rheostat that coordinates the balance between pro-inflammatory and anti-inflammatory responses through concurrent activation of NF-κB and mTOR pathways while suppressing AMPK activity. The lactate-driven metabolic shift from OXPHOS to glycolysis represents a biochemical hallmark of M1 macrophage polarization, where enhanced glycolytic flux both provides biosynthetic precursors for pro-inflammatory mediators and suppresses M2-associated lipid synthesis pathways [33]. Emerging evidence reveals that intracellular lactate can be converted to lactoyl-CoA, serving as the substrate for novel post-translational modifications including histone lactoylation, thereby expanding its immunometabolic regulatory repertoire [34]. This sophisticated interplay between lactate metabolism and immune function underscores its central role in macrophage plasticity [35]. Despite growing recognition of lactylate role in immunometabolism, its impact on macrophage-trained immunity in ALI remains largely unexplored. These present a critical opportunity to develop lactate-targeted therapies that recalibrate immune responses while mitigating lung injury, potentially yielding transformative treatment strategies for ALI patients.

Conclusion

Altered lactate levels played a significant role in acute lung injury. Our study demonstrated that trained immunity led to increased AKT2 phosphorylation and PDK1 activation, which were associated with elevated lactate secretion and enhanced acid tolerance in alveolar macrophages. Notably, the typical immunosuppressive effect of lactate on alveolar macrophages was reversed to promote immune activation, resulting in improved acute lung injury outcomes. However, these beneficial effects were suppressed by AKT2 and PDK1 inhibitors. These findings suggest new therapeutic approaches for acute lung injury treatment.

Materials and methods

Reagents and antibodies

Β-glucan was purchased from Invivogen Biotechnology (Hong Kong, China). LPS and lactate were purchased from Sigma (St. Louis, MO, USA). Phospho-AKT2 (Ser474) (AF3264), PDK1 (DF4365), TNF-α (AF7014), IL-6 (DF6087) antibodies are purchased from Affinity Corporation (Changzhou, China). β-actin (81115-1-RR), CD86 (13395-1-AP), CD206(32647-1-AP) from Proteintech (Wuhan, China), CCT128930 and AZD7545 from MCE (New Jersey, USA). Purchase the mouse TNF-α enzyme-linked immunosorbent assay kit (SEKM-0034) Solarbio.

Cell culture

The mouse J774 cell line was obtained from the Shanghai Culture and Preservation Center of the Chinese Academy of Sciences. Alveolar macrophages were isolated from bronchoalveolar lavage fluid of C57BL/6 mice and cultured in an incubator at 37 °C with 5% CO₂ and 95% air. Cells were maintained in phenol red-free DMEM/1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin (100 U/ml penicillin and 100 µg/ml streptomycin). Cells were seeded at appropriate densities in different sized culture plates for subsequent experiments. Trained immunity and acute lung injury models were established using β-glucan and LPS, respectively.

Mouse

All animals used in this study were C57BL/6 mice aged 8 to 10 weeks. The mice were housed in an SPF-grade animal facility with approval from the Biological Academic Committee of Henan Normal University. All animal experiments followed international guidelines for animal use, with strict adherence to ethical standards to minimize animal suffering. During the study period, mice were provided with sterile rodent chow and drinking water, and maintained at a stable room temperature with a controlled light–dark cycle.

H&E staining

After sampling, fresh samples were fixed in 4% paraformaldehyde (PFA) and subsequently embedded in paraffin. The paraffin blocks mounted on tissue embedding cassettes were sectioned into 5 μm slices. Following baking at 60 °C, sections were dewaxed and rehydrated, then soaked in distilled water to prevent cracking. Hematoxylin staining was performed first, followed by washing under running water. Sections were then differentiated in HCl ethanol solution for 2 s and counterstained with eosin. Finally, stained sections were mounted with resin.

Immunohistochemistry

Lung sections from ALI model mice were dewaxed and rehydrated. Antigen retrieval was performed by microwaving in sodium citrate buffer for 10 min after membrane permeabilization with 3% Triton X-100. Sections were then treated sequentially with endogenous peroxidase blocking solution and immunostaining blocking solution, followed by overnight incubation with PDK1 primary antibody (1:100 dilution, Proteintech). The next day, staining was completed using the appropriate secondary antibody combination, and samples were examined under light microscopy.

Quantitative PCR

Total RNA was extracted from cells and tissues using Trizol reagent. Following the manufacturer’s protocol, 1 µg of RNA was reverse-transcribed into cDNA using reverse transcriptase (Vazyme Co., Ltd., China). For QPCR analysis, Hieff QPCR SYBR Green premix (Yeasen Biotechnology, Shanghai, China) was prepared and aliquoted into 384-well plates with three technical replicates per sample. Gene expression changes were analyzed using the 2^− ΔΔCt method. The oligonucleotide primers used for PCR amplification are listed in Table 1.

Table 1.

Primer sequences

Gene	Sense (5′-3′)	Anti-sense (5′-3′)
IL-6	TAGTCCTTCCTACCCCAATTTCC	TTGGTCCTTAGCCACTCCTTC
PDK1	GGACTTCGGGTCAGTGAATGC	TCCTGAGAAGATTGTCGGGGA
LDHA	TGTCTCCAGCAAAGACTACTGT	GACTGTACTTGACAATGTTGGGA
TNF-α	CAGCAAGGGACAGCAGAGG	AGTATGTGAGAGGAAGAGAACC
IL-1β	GCAACTGTTCCTGAACTCAACT	ATCTTTTGGGGTCCGTCAACT
HK2	CCCTGTGAAGATGTTGCCCACT	CCTTCGCTTGCCATTACGCACG
PKM2	CAGAGAAGGTCTTCCTGGCTCA	GCCACATCACTGCCTTCAGCAC
ALDOA	CACGAGACACTGTACCAGAAGG	TTGTCTCGCCATTGGTTCCTGC
BACT	GGCTGTATTCCCCTCCATCG	CCAGTTGGTAACAATGCCATGT

Western blotting

After cell treatment, proteins were extracted using RIPA lysis buffer. The protein lysates (20–50 µg per lane) were separated by SDS polyacrylamide gel electrophoresis. Following electrophoresis, proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore,Billerica, MA, USA). Membranes were blocked with skim milk powder at room temperature to prevent nonspecific binding. Primary antibodies were incubated overnight at 4 °C according to target protein molecular weights, with β-actin serving as the loading control. The next day, membranes were incubated with appropriate species-matched secondary antibodies in TBST buffer for 1.5 h, followed by detection using an imaging system.

Immunofluorescence

Cells were cultured on coverslips in 24-well plates and allowed to adhere in a CO₂ incubator before drug treatment. Following treatment, cells were washed three times with PBS and fixed with 4% paraformaldehyde at room temperature for 15 min. After fixation, cells were washed with PBS and permeabilized with 3% Triton X-100. Nonspecific binding was blocked using BSA solution. Primary antibodies (PDK1, phospho-AKT2, and Mitotracker) were applied and incubated overnight at 4 °C, followed by incubation with appropriate fluorescent secondary antibodies for 1.5 h in the dark. After three PBS washes, coverslips were mounted using DAPI containing mounting medium. Fluorescence images were acquired using a Nikon Eclipse Ti-U microscope equipped with a DS-Ri1 digital camera.

Cell viability assay

To assess changes in cell viability following lactate treatment, 100 µL of cell suspension was seeded into 96-well plates. CCK-8 reagent (10 µL per well) was added, and the plates were incubated for 1–4 h. Absorbance was then measured at 450 nM using a microplate reader.

ELISA

Following drug treatment, cell culture supernatants were collected, mixed, and centrifuged. TNF-α and IL-1β levels in the supernatant were then measured using a Solarbio ELISA kit according to the manufacturer's protocol.

Flow cytometry analysis

Following cell digestion and centrifugation, the harvested cells were washed with PBS and resuspended in 1 × binding buffer to normalize cell concentration. Cells were then gently mixed with Annexin V-FITC and propidium iodide (PI) in the dark at room temperature. After incubation, cells were washed with PBS and analyzed by flow cytometry to detect apoptosis.

Alveolar macrophage extraction

After euthanasia, mice were fixed and the trachea was exposed by dissection. An intravenous indwelling needle was inserted into the trachea toward the heart, and the lungs were lavaged with 800 µL PBS. The collected fluid was centrifuged at 500 × g for 10 min, and the supernatant was discarded. Following red blood cell lysis (3–5 min), cells were resuspended in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin (100 U/mL penicillin and 100 μg/mL streptomycin) and transferred to a CO₂ incubator.

Statistical analysis

All data were analyzed using GraphPad Prism 5 (GraphPad Software, USA) and assumed to follow a normal distribution. Differences between two groups were assessed using a two-tailed Student’s t-test, while comparisons of three or more groups were analyzed by one-way ANOVA. A p-value < 0.05 was considered statistically significant.

Author contributions

All authors participated in manuscript revising, data analysis and giving final approval of the version to be published. YW, XL and ZL conceived and designed the project. HM, XW, XG, WH, YZ and ZW participated in the animal experiment. ZS and HM wrote the manuscript. ZS and YW revised it. ZS, HM, XW, XG developed experimental protocols, directed the project, conceptualized and designed experiments, and interpreted results. This article was also supported to completion by ZL and XL.

Funding

This research was supported by grants from National Natural Science Foundation youth fund of China (32200714); Henan Normal University Research Launch Foundation (20220099) to Z.S; This research was also supported by Open Fund Project of Hebei Provincial Key Laboratory of Pulmonary Disease (2025001).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate

This study was carried out in accordance with the principles of the Basel Declaration and recommendations of the US NIH with Specific Pathogen Free conditions. The protocol was approved by the Model Animal Research Center of Nanjing University. All-time-available Standard rodent chow and water were also provided.

Competing interests

The authors declare no competing interests.