A brain-lung circuit drives LPS-induced lung injury via sympathetic neutrophil axis Authors: Shichun Ren, Shuo Yang, Baotong Liu, Qian Zhang, Zhenzhen Zhao, Yan Meng, Qi Shao, Xiaoming Deng, Jinjun Bian, Li Cao, Fengfeng Mo, Jiafeng Wang Published: 2026 ABSTRACT Acute respiratory distress syndrome (ARDS) is a fatal complication of sepsis. However, the neural mechanisms underlying the exacerbation of pulmonary inflammation during systemic infection remain largely undefined. We employed an intraperitoneal lipopolysaccharide (LPS)-induced systemic inflammatory lung injury model and found that systemic inflammation strongly activates corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus (PVN), whereas bilateral subdiaphragmatic vagotomy abolished this response. Chemogenetic inhibition of CRHPVN neurons alleviated lung injury and the formation of neutrophil extracellular traps (NETs), whereas chemogenetic activation of CRHPVN neurons exacerbated pathological damage through sympathetic efferent signaling. Additionally, in patients with septic ARDS, β2-adrenergic receptor (ADRB2) expression was upregulated on neutrophils, and ADRB2 signaling delayed neutrophil apoptosis through the PI3K/Akt pathway, whereas ADRB2 blockade promoted neutrophil apoptosis and attenuated lung injury. In summary, we delineate a dysregulated neuroimmune axis in LPS-induced lung injury: systemic inflammatory signals activate the vagal afferent-PVN-sympathetic circuit, which in turn drives a neutrophil-mediated ADRB2-PI3K/Akt signaling pathway that amplifies lung injury. Our study reveals a mechanistic basis for neuroregulatory intervention and indicates that targeting this Brain–lung pathway may offer new therapeutic strategies for ARDS.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12974-026-03914-z. Background Acute respiratory distress syndrome is a devastating complication of systemic infection, characterized by diffuse alveolar damage, pulmonary edema, and an overwhelming neutrophilic inflammatory response [1]. As a primary driver of mortality in critically ill patients with sepsis and pneumonia, ARDS represents a major global health challenge [2–4]. Despite advances in supportive care including mechanical ventilation and extracorporeal membrane oxygenation [5, 6], no specific pharmacological treatment is available that directly targets the inflammatory cascade responsible for lung injury. This therapeutic impasse underscores the urgent need to elucidate the fundamental mechanisms that orchestrate pulmonary inflammation during systemic sepsis. Inflammation is now recognized to be under sophisticated neural control. The central nervous system actively senses and modulates immunity through dedicated “neuroimmune axes”, which allow for rapid, organ-specific regulation of inflammatory responses [7–9]. The lungs are densely innervated by both the vagus and sympathetic nerves [10, 11]. The vagus nerve is a prime example, with its efferent arm constituting the “cholinergic anti-inflammatory pathway” that suppresses cytokine production and is generally protective against lung injury [12]. Furthermore, while the sympathetic nervous system (SNS) is a key mediator of the septic stress response, its specific impact on pulmonary neutrophil survival and function—and whether it acts as a protective or maladaptive force in sepsis—remains controversial [13, 14]. The hypothalamic paraventricular nucleus (PVN) is a key hub that integrates neuroendocrine and autonomic responses to systemic stressors, coordinating the hypothalamic-pituitary-adrenal (HPA) axis and sympathetic outflow [15–17]. Previous studies have demonstrated that the PVN receives afferent signals from circulating cytokines and the vagus nerve, and modulates immune organ function by regulating glucocorticoid release (via the HPA axis) and sympathetic nerve activity in the context of sepsis and systemic inflammatory response syndrome (SIRS) [18, 19]. However, its specific contribution to the pathogenesis of ARDS has not been systematically explored. Recent studies have highlighted distinct brain regions, such as the central amygdala, in modulating lung immunity [9], suggesting that specific central nodes may be differentially recruited depending on the nature of the inflammatory insult. A recent study demonstrated that CRH-expressing PVN neurons regulate acute lung injury via a sympathetic nerve–neutrophil axis, identifying a protective role for this circuit in intranasal LPS-induced localized airway injury [13, 14]. However, the functional role of this circuit in systemic endotoxemia — a condition that more closely models clinical sepsis with its characteristic sympathetic storm and multi-organ involvement — has not been examined. Moreover, neutrophils are pivotal to innate defense against infection, but their overactivation during sepsis leads to excessive tissue damage [20, 21], particularly in the lungs. As the main effector cells in ARDS, neutrophils contribute to lung injury through robust infiltration, the release of neutrophil extracellular traps (NETs), and delayed apoptosis. While these processes are known to be influenced by local inflammatory mediators and adrenergic signaling, it remains unclear whether specific brain nuclei directly commandeer these neutrophil-driven pathological processes via a defined neural pathway, and the precise regulatory mechanisms involved have yet to be elucidated. In this study, we hypothesize that the CRHPVN neurons–sympathetic nerve pathway may be pathologically recruited during systemic inflammation, critically amplify neutrophil-mediated lung injury in systemic inflammation. Here, we demonstrate that systemic inflammatory signals are sensed by abdominal vagal afferents, triggering the robust activation of CRHPVN neurons. Contrary to canonical anti-inflammatory reflexes, activation of this circuit exacerbates lung injury via a maladaptive sympathetic-neutrophil axis. We show that sympathetic outflow acts on upregulated β2-adrenergic receptors (ADRB2) on neutrophils to activate the PI3K/Akt pathway, thereby delaying apoptosis and preventing the resolution of inflammation. Thus, we identify a neuroimmune mechanism—mediated by the CRHPVN–sympathetic–ADRB2 axis—that drives pathological lung injury, revealing novel therapeutic avenues for septic ARDS. Methods Mice Animal experiments were approved by the Committee on Ethics of Medical Research, Naval Medical University, PLA (NO.2021AAA010802), and all experiments were conducted following the relevant rules. All experiments were conducted on male mice older than 8 weeks of age housed in a 12-h light/12-h dark cycle at 22 °C and 30–60% humidity. Food and water were provided ad libitum. We used the following genotypes of mice: C57BL/6J (wild type; GemPharmatech Laboratory Animal Company, Nanjing, China), FOS2A-iCreER (TRAP2) mice (The Jackson Laboratory, stock number 030323), CRH-Cre (The Jackson Laboratory, stock number 012704). For all Cre mouse line experiments, only heterozygous animals were used. Sample sizes were estimated based on pilot experiments and similar studies. Age-balanced litter-mate mice were randomly assigned to experimental groups. Adult male mice aged 3–4 months were used for experiments. Male mice were used to minimize confounding sex-related biological variability. Investigators were blinded to group allocation during histological scoring, and cell counting. Patients and healthy volunteers ARDS patients: Patients were recruited between May 2024 and September 2025 at Shanghai Changhai Hospital. Informed consent was obtained from the families of all the subjects, and the protocol was approved by the Committee on Ethics of Medical Research, Naval Medical University, PLA (NO.2021AAA010802). Patients meeting the Berlin criteria for ARDS admitted to the Central Intensive Care Unit were consecutively enrolled. Inclusion criteria included a diagnosis of sepsis as the primary etiology. Peripheral blood was collected upon ICU admission when patients were not receiving vasopressors, β-agonists, or β-blockers. Fresh neutrophil samples were selected from donors with sufficient blood volume, preserved cell viability, and complete medication records. The exact sample size for each assay is indicated in the corresponding figure legend. Healthy volunteers: Age-matched healthy donors with no history of acute inflammation, autoimmune disease, or recent medication use were recruited as controls. LPS-induced acute lung injury model To model LPS-induced lung injury, mice received a single i.p. injection of LPS (E. coli 0111:B4, Sigma-Aldrich, L2630). LPS was resuspended in saline (5 mg ml− 1) and frozen into single-use aliquots. Further dilutions in saline to the appropriate experimental dose and a standardized volume (about 200 µl per animal) were further generated before each experiment. For acute injury assessment, a dose of 10 mg/kg was used to induce severe systemic inflammation and lung injury [22]. For TRAP2 labeling experiments, a sub-lethal dose of 0.5 mg/kg was used to ensure survival during the labeling window while sufficiently activating the PVN [23]. Control mice received an equivalent volume of sterile saline. For in vivo ADRB2 blockade, ICI 118,551 (5 mg/kg, i.p.) was administered 1 h before LPS [24]. Immunofluorescence Mice were transcardially perfused with saline followed by 4% paraformaldehyde. The brains were postfixed for 16–24 h at 4 °C and dehydrated in 30% sucrose solution. Brains were sectioned using a frozen microtome (Leica CM1950S) to a thickness of 50 μm. Brain slides were blocked in 5% goat serum, 0.3% Triton X-100, and 5% bovine serum albumin (BSA) in 1x PBS for 2 h at room temperature. Slides were stained with the following primary antibodies: Rabbit anti-FOS antibody (1:1,000, abcam ab214672), Rabbit anti-TH (1:1000, Sigma-Aldrich ab152), Rabbit anti-myeloperoxidase (MPO) (1:200, abcam ab208670), Rabbit anti-Ly6G (1:500, Servicebio GB11229), Rabbit anti-F4/80 (1:500, Cell Signaling Technology 70076 S), Rabbit anti-CD3 (1:200, abcam ab16669), Rabbit anti-CitH3 (1:200, abcam ab219407) for 24 h at 4 °C. After washing in 1x PBS three times (15 min each), the slides were incubated in Goat anti-Rabbit 488 (1:500, Beyotime A0423), Donkey anti-Mouse 555 (1:500, Beyotime A0460), Goat anti-Rabbit 647 (1:500, Beyotime A0468), Goat anti-Mouse 488 (1:500, Beyotime A0428), Donkey anti-Rabbit 555 (1:500, Beyotime A0453), Goat Anti-Rat Cy3 (1:500, Beyotime A0507) or Goat anti-Mouse 647 (1:500, Beyotime A0473) for 90 min at room temperature. The slides were then washed three times (15 min each), followed by mounting with anti-fluorescence quenching mounting medium containing DAPI (Biosharp). Confocal images were obtained with an OLYMPUS IX83-FV3000 system or Zeiss LSM 980 AiryScan2 system. H&E staining and lung injury score Lung tissues were harvested and immediately fixed in 4% paraformaldehyde. Following dehydration via a graded ethanol series, tissues were embedded in paraffin wax and sectioned into 5 μm-thick slices. Sections underwent deparaffinization in xylene and rehydration, followed by hematoxylin and eosin (H&E) staining according to standard protocols. Digital images were acquired using an Olympus VS120 slide scanner. To minimize bias, histopathological assessment was performed by two independent experienced pathologists blinded to the experimental group allocation. Five randomly selected, non-overlapping fields were analyzed per sample. A quantitative scoring system, ranging from 0 to 5, was established to grade the extent and characteristics of pulmonary injury. The specific criteria were defined as follows: 0, normal alveolar architecture; 1, mild tissue injury; 2, presence of inflammatory cells without significant parenchymal destruction; 3, marked inflammatory cell infiltration accompanied by interstitial congestion and septal thickening; 4, severe compromise of lung integrity with alveolar space compression; and 5, diffuse and severe destruction of the entire lung tissue within the visual field. Brain surgery and stereotaxic viral injection Mice were anesthetized using isoflurane anesthesia (induction 3%, maintenance 1.5–2%) and placed in a stereotactic frame (RWD). A craniotomy hole was drilled, and the virus was injected via a glass micropipette connected to a Nanoliter Inject (RWD) and its controller (RWD) at a speed of 20 nL/min. For the PVN region, the following coordinates relative to the bregma were used: AP, − 0.65 mm to − 0.7 mm; ML, ± 0.2 mm; DV, − 4.7 mm. For chemogenetic experiments, 300 nL of AAV2/9-hSyn-DIO-hM3D(Gq)–mCherry (activating), AAV2/9-hSyn-DIO-hM4D(Gi)–mCherry (inhibitory) or AAV2/9-hSyn-DIO–mCherry (control) was injected bilaterally. Mice were allowed to recover and express virus for 3 weeks. Animals were injected with 1 mg/kg of CNO or saline (control) 3 weeks post-injection. For fiber photometry recording, 300 nL of AAV-hSyn-Dio-jGCaMP7b was injected into PVN, and an optic fiber was implanted approximately 200 μm above the viral injection site. The fiber was secured to the skull using dental cement. We performed post hoc verification of the accuracy of viral infection and optic fiber implantation. For retrograde tracing to map the polysynaptic neural circuitry connecting the lung to the central nervous system, 3 µL of PRV-CAG-EGFP (BrainVTA) was injected into three distinct sites (upper, middle, and lower regions; 1 µL per site) of the left lung lobe in adult CRH-Cre mice (CRHPVN neurons hade been pre-labeled with red fluorescence). The needle was left in place for approximately 5 min after injection to prevent viral extravasation. Animals were monitored daily for health status. After 5 days of viral infection and transsynaptic spread, the mice were euthanized and processed for histology. Brain and spinal cord tissues were harvested for subsequent immunofluorescence analysis. Subdiaphragmatic vagotomy Mice were anesthetized with isoflurane anesthesia (3% induction, 1.5-2% maintenance) and placed in a supine position on a surgical board. A midline laparotomy (approximately 1.5–2 cm) was performed along the linea alba, caudal to the xiphoid process. At the esophagogastric junction, where the esophagus passes through the diaphragm, the anterior and posterior vagal trunks were identified as fine, white, thread-like structures adherent to the esophageal wall. A segment of approximately 2–4 mm of each identified vagal trunk was resected and removed. In sham-operated mice, all procedures were performed identically, but the nerves were not transected. Following resection, the surgical site was inspected for hemostasis, and the liver lobes were repositioned anatomically. The peritoneum, muscle layers, and skin were closed sequentially with sutures. Finally, mice were placed on a warming pad for recovery from anesthesia and subsequently returned to its home cage with ad libitum access to food and water. Sympathetic denervation To deplete systemic sympathetic nerves, mice received intraperitoneal injections of 6-hydroxydopamine (6-OHDA; 100 mg/kg, Sigma-Aldrich 162957) dissolved in sterile saline containing 0.1% ascorbic acid (vehicle control: saline + 0.1% ascorbic acid). Injections were administered once daily for three consecutive days prior to LPS challenge. Efficacy was confirmed by Western blot for tyrosine hydroxylase (TH) in lung tissues. Fiber photometry recording AAV-DIO-jGCaMP7b virus was unilaterally injected into the PVN of CRH-Cre mice. An optic fiber with a 200-µm outer diameter was implanted dorsal to the viral injection site and secured to the skull using dental cement. The accuracy of viral infection and optic fiber placement was verified post hoc. Calcium dynamics of CRHPVN neurons were recorded using a multichannel fiber photometry system (Thinker Tech, Nanjing Bioscience Inc.) Excitation light at 488 nm (calcium-dependent signal) and 410 nm (isosbestic control) was delivered through the fiber. Fluorescence signals were acquired at a sampling frequency of 100 Hz, with laser power at the fiber tip adjusted to 20–40 µW to minimize photobleaching. To dissect the neural circuitry driving PVN activation during systemic inflammation, mice were assigned to four experimental cohorts: (1) LPS Model: mice received an i.p. injection of LPS; (2) Sympathectomy + LPS: mice received chemical sympathectomy (6-OHDA) once daily for three consecutive days prior to recording; (3) Vagotomy + LPS: mice underwent bilateral subdiaphragmatic vagotomy 7 days prior to recording; and (4) Saline Control: mice received an i.p. injection of saline. On the recording day, mice were habituated in a light-tight chamber to minimize ambient light interference. The recording protocol consisted of a 1–2 h stabilization period to establish a steady baseline, followed by i.p. injection of LPS or saline. Recording continued uninterrupted until the fluorescence signal returned to pre-injection baseline levels, followed by an additional 20-minute acquisition period. Raw data were exported in .mat format and analyzed using MATLAB software. The relative change in fluorescence (ΔF/F) was calculated as: ΔF/F = (F-F0) / (F0-Foffset), where F represents the real-time fluorescence signal, F0 is defined as the mean signal during the 5-minute baseline period preceding injection, and Foffset represents the mean background fluorescence (machine offset). The acute neuronal response was quantified by comparing the mean ΔF/F intensity during the 50-minute post-injection window. Bulk RNA-seq and analysis To characterize the transcriptomic profile of the PVN during LPS-induced lung injury, mice received intraperitoneal injections of LPS or saline (control). Total RNA was extracted from bilateral PVN tissue micropunches using TRIzol Reagent (Takara Biotechnology, Shiga, Japan) according to the manufacturer’s protocol. RNA sequencing libraries were constructed and sequenced on an Illumina NovaSeq 6000 platform by OE Biotech Co., Ltd. (Shanghai, China). Raw sequencing reads underwent quality control filtering and were aligned to the mouse reference genome. Gene expression was quantified, and differential expression analysis was performed using the DESeq2 package in R. Significantly differentially expressed genes (DEGs) were identified based on a False Discovery Rate (FDR) adjusted P-value < 0.05 and an absolute log2 fold change ( |log2 FC|) > 1. Functional enrichment analysis for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways was conducted on the DEGs using clusterProfiler package. Quantitative real-Time PCR (RT-qPCR) Total RNA was extracted from tissue samples with TRIzol Reagent (Takara Biotechnology, Shiga, Japan), and then was reverse transcribed with the PrimeScript™ RT Reagent Kit (Takara) according to the manufacturer’s instructions. Quantitative real-time PCR was performed using the SYBR® Premix Ex Taq™ II kit (Takara). Gene expression levels were normalized to the internal control Gapdh. The specific primer sequences are listed in Supplementary Table S2. Enzyme-linked immunosorbent assay Serum concentrations of IL-6 and TNF-α were quantified using commercial ELISA kits (Elabscience, Wuhan, China) in strict accordance with the manufacturer’s instructions. Briefly, standards and samples were added to the plates and incubated, followed by the addition of detection reagents. The optical density (OD) was measured using a microplate reader. Cytokine concentrations were calculated based on the linear regression of the standard curves. Isolation and culture of human neutrophils Peripheral blood samples were collected from ARDS patients and healthy volunteers into heparinized tubes. Neutrophils were isolated via density gradient centrifugation using Ficoll-Hypaque (GE Healthcare) and 3% dextran sedimentation, followed by hypotonic lysis to remove residual erythrocytes. Cell viability was confirmed to be > 95% by trypan blue exclusion. Neutrophils were maintained in complete RPMI 1640 medium containing 10% FBS and 1% penicillin/streptomycin. For in vitro experiments, neutrophils were seeded in 24-well plates at a density of 1 × 10⁶ cells/mL. The following treatments were administered: the selective β2-adrenergic receptor antagonist ICI 118,551 (50 µM; Sigma-Aldrich I127), Norepinephrine (10 µM; MedChemExpress HY-13715), the Akt activator SC79 (10 µM; MedChemExpress HY-18749), and the Akt inhibitor AKTi (5 µM; MedChemExpress HY-10355). Control wells received an equivalent volume of vehicle (DMSO or PBS). Cells were incubated at 37 °C in a humidified 5% CO2 atmosphere for 24 h for apoptosis and NETosis before being harvested for analysis. Flow cytometry For mice experiments, samples of peripheral blood (100 µL) and bronchoalveolar lavage fluid (BALF) were collected. Red blood cells in peripheral blood were eliminated using RBC Lysis Buffer (BioLegend). BALF samples were centrifuged at 1500 rpm for 5 min at 4 °C to pellet cellular components. Single-cell suspensions from both blood and BALF were stained with FITC-conjugated anti-Ly6G (BioLegend) and PE-conjugated anti-CXCR2 antibodies (BioLegend) for 15 min at 4 °C. Human neutrophils were isolated from whole blood via density gradient centrifugation (Ficoll-Hypaque; GE Healthcare) and 3% dextran sedimentation. For phenotypic analysis, cells were stained with FITC-conjugated anti-human CD11b (BioLegend) and PE-conjugated anti-ADRB2 antibodies (santacruz 81577) for 15 min at 4 °C. Apoptosis was assessed using an Annexin V-FITC and Propidium Iodide (PI) Apoptosis Detection Kit (BD Biosciences) according to the manufacturer’s instructions. Apoptotic cells were defined as Annexin V+/PI−. Data were acquired using a FACSCanto II flow cytometer (BD Biosciences, San Jose, CA, USA). Fluorescence Minus One (FMO) controls were utilized to establish gating boundaries. All data were analyzed using FlowJo software (V10.8.1; Tree Star, Ashland, OR, USA). Western blot Lung tissues were homogenized and lysed in RIPA lysis buffer (Epizyme Biotech, Shanghai, China). Lysates were centrifuged at 12,000 rpm for 10 min at 4℃. Supernatants were collected and total protein concentrations were quantified by a BCA assay Kit (Epizyme Biotech). Equal amounts of protein were separated by 10% SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes (Merck, Darmstadt, Germany). Membranes were blocked with 5% non-fat dry milk for 1 h at room temperature and incubated overnight at 4 °C with the following primary antibodies: anti-TH (1:2,000, Sigma, AB152), Rabbit anti-AkT( 1:1000 Cell Signaling Technology 9272), Rabbit anti-pAkT( 1:1000 Cell Signaling Technology 4060), and Rabbit anti-GAPDH (1:2,000, Engibody, AT0010). Subsequently, membranes were incubated with Goat anti-Rabbit IgG-HRP antibody (1:2,000, Engibody, AT8301) secondary antibody for 1 h at room temperature. Finally, the protein bands were visualized using the chemiluminescence kit (Beyotime, P0018AS, China) and analyzed by ImageJ software. Statistical analysis All quantitative data are expressed as mean ± SEM. The sample size ‘n’ indicates the number of individual animals per group. Parametric data were analyzed using a two-tailed Student’s t-test for single comparisons or a one-way ANOVA with Tukey’s post-hoc adjustments for multiple group comparisons. Differences in lung injury scores between two groups were analyzed using the Mann-Whitney U test. All statistical analyses were performed using GraphPad PRISM 9 software or MATLAB. Results LPS-induced acute lung injury robustly activates CRH neurons in the PVN To systematically identify the central neuronal nuclei responsive to acute lung injury, we employed a well-established mouse model of ALI induced by LPS [22]. Following intraperitoneal administration of a lethal dose of LPS (10 mg/kg) and confirmation of lung injury (Figure S1A, B), we performed whole-brain screening for neuronal activation using c-Fos immunofluorescence. Brain tissues were collected 24 h after LPS administration for analysis (Fig. 1A). Our results revealed a significant increase in c-Fos+ neurons across multiple brain regions upon LPS challenge, including the bed nucleus of the stria terminalis (BNST), lateral preoptic area (LPO), mediodorsal thalamic nucleus (MD), and CeA (Figure S1C). Among these regions, the PVN of the hypothalamus exhibited the pronounced activation (Fig. 1B, C), implicating the PVN as a key node in ALI. The PVN serves as a critical hub that integrates neural, endocrine, and immune responses. It modulates systemic inflammatory tone by engaging the HPA axis to release glucocorticoids and by regulating sympathetic nervous system (SNS) outflow via descending pathways [7, 25]. Fig. 1CRH neurons in the PVN are highly activated in LPS-induced acute lung injury. A Schematic of the LPS-induced lung injury model. Mice received intraperitoneal (i.p.) LPS (10 mg/kg, E. coli 0111:B4) to induce systemic inflammation. Brains were collected at 24 hours for c-Fos mapping. B, C Representative immunofluorescence images and quantification of c-Fos expression in the Paraventricular Nucleus (PVN). LPS challenge significantly increased the number of c-Fos+ neurons compared to saline controls (n=3 mice/group). Scale bar, 300 µm. D Schematic of the TRAP2 (Fos-iCreER) chemogenetic strategy. To label sepsis-responsive neurons without inducing lethality during the labelling window, TRAP2 mice received a low dose of i.p. LPS (0.5 mg/kg) concurrent with 4-hydroxytamoxifen (4-OHT). After a 3-week recovery, neurons were reactivated with Clozapine N-oxide (CNO, 1 mg/kg). E–H Flow cytometric analysis of neutrophils (Ly6G+ CXCR2+) in peripheral blood and bronchoalveolar lavage fluid (BALF) 24 hours after CNO reactivation. (E, F) Reactivation of sepsis-labeled PVN neurons (hM3Dq-CNO) significantly increased circulating peripheral blood neutrophils counts and reduced surface CXCR2 expression compared to controls (n≥5 mice/group). (G, H) Reactivation drove neutrophil recruitment into the BALF. (n≥5 mice/group). I, J Representative H&E staining and histological scoring of lung tissues. CNO treatment in hM3Dq-expressing mice resulted in significant alveolar wall thickening and leukocyte infiltration. Lung injury scores were analyzed using the non-parametric Mann-Whitney U test (n=5 mice/group). Scale bar, 100 µm. K, L Immunohistochemical staining for myeloperoxidase (MPO). Quantification shows a significant increase in MPO+ neutrophils in the lung parenchyma of the hM3Dq-CNO group (n=6 mice/group). Scale bar, 100 µm. Local enlargings are shown below, Scale bar, 50 µm. M Heatmap of differentially expressed genes (DEGs) from RNA-seq of PVN tissue harvested 24h post-LPS (10 mg/kg). Crh was among the top upregulated transcripts (n=4 replicates). N qPCR validation confirming upregulation of Crh mRNA in the PVN following LPS challenge (n=6 mice/group). O, P Double immunofluorescence staining showing colocalization of c-Fos (green) and CRH (red) in the PVN. Approximately 81% of LPS-activated c-Fos+ neurons were CRH+, identifying them as the dominant responsive population (n=4 mice). Scale bar, 100 µm. *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; ns, not significant. Statistical comparisons were performed using one-way ANOVA with Tukey’s post-hoc test CRH neurons in the PVN are highly activated in LPS-induced acute lung injury. A Schematic of the LPS-induced lung injury model. Mice received intraperitoneal (i.p.) LPS (10 mg/kg, E. coli 0111:B4) to induce systemic inflammation. Brains were collected at 24 hours for c-Fos mapping. B, C Representative immunofluorescence images and quantification of c-Fos expression in the Paraventricular Nucleus (PVN). LPS challenge significantly increased the number of c-Fos+ neurons compared to saline controls (n=3 mice/group). Scale bar, 300 µm. D Schematic of the TRAP2 (Fos-iCreER) chemogenetic strategy. To label sepsis-responsive neurons without inducing lethality during the labelling window, TRAP2 mice received a low dose of i.p. LPS (0.5 mg/kg) concurrent with 4-hydroxytamoxifen (4-OHT). After a 3-week recovery, neurons were reactivated with Clozapine N-oxide (CNO, 1 mg/kg). E–H Flow cytometric analysis of neutrophils (Ly6G+ CXCR2+) in peripheral blood and bronchoalveolar lavage fluid (BALF) 24 hours after CNO reactivation. (E, F) Reactivation of sepsis-labeled PVN neurons (hM3Dq-CNO) significantly increased circulating peripheral blood neutrophils counts and reduced surface CXCR2 expression compared to controls (n≥5 mice/group). (G, H) Reactivation drove neutrophil recruitment into the BALF. (n≥5 mice/group). I, J Representative H&E staining and histological scoring of lung tissues. CNO treatment in hM3Dq-expressing mice resulted in significant alveolar wall thickening and leukocyte infiltration. Lung injury scores were analyzed using the non-parametric Mann-Whitney U test (n=5 mice/group). Scale bar, 100 µm. K, L Immunohistochemical staining for myeloperoxidase (MPO). Quantification shows a significant increase in MPO+ neutrophils in the lung parenchyma of the hM3Dq-CNO group (n=6 mice/group). Scale bar, 100 µm. Local enlargings are shown below, Scale bar, 50 µm. M Heatmap of differentially expressed genes (DEGs) from RNA-seq of PVN tissue harvested 24h post-LPS (10 mg/kg). Crh was among the top upregulated transcripts (n=4 replicates). N qPCR validation confirming upregulation of Crh mRNA in the PVN following LPS challenge (n=6 mice/group). O, P Double immunofluorescence staining showing colocalization of c-Fos (green) and CRH (red) in the PVN. Approximately 81% of LPS-activated c-Fos+ neurons were CRH+, identifying them as the dominant responsive population (n=4 mice). Scale bar, 100 µm. *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; ns, not significant. Statistical comparisons were performed using one-way ANOVA with Tukey’s post-hoc test To define the function of LPS-activated neurons within the PVN, we stereotaxically injected AAV-DIO-hM3Dq-mCherry into the PVN of TRAP2 (FOS2A-iCreER) mice (Fig. 1D). This approach enabled the tamoxifen-inducible CreER system to selectively target and label the population of neurons activated by LPS—a strategy known as Targeted Recombination in Active Populations (TRAP). Ten days after viral injection, mice received a low dose of LPS (0.5 mg/kg) together with 4-hydroxytamoxifen (4-OHT, 50 mg/kg) to induce DREADD expression specifically in Fos-activated neurons. Following a 3-week recovery period to allow resolution of acute inflammation, we administered clozapine N-oxide (CNO) intraperitoneally to chemogenetically activate these “TRAPed” neurons. Mice were euthanized 24 h later, peripheral blood, and bronchoalveolar lavage fluid (BALF) were collected for analysis (Fig. 1D). Compared to control groups (mCherry + Saline, hM3Dq + Saline, mCherry + CNO), chemogenetic reactivation of the LPS-activated PVN neuronal ensemble significantly increased neutrophil counts in both peripheral blood and BALF (Fig. 1E and G). Moreover, we observed a decrease in the proportion of CXCR2+ neutrophils in the blood alongside a marked increase in the BALF, indicating robust recruitment of neutrophils from the circulation into the lung (Fig. 1F and H). Consistent with these findings, Hematoxylin-eosin (H&E) staining of lung sections revealed enhanced alveolar structural damage and neutrophil infiltration (Fig. 1I-L). These findings demonstrate that activation of the LPS-responsive PVN neuronal ensemble is sufficient to drive pulmonary neutrophil recruitment and tissue injury, suggesting the existence of a Brain–lung neuroimmune pathway. To molecularly characterize the LPS-activated neurons in the PVN, we performed RNA sequencing (RNA-seq) on PVN tissue. Differential expression analysis (FDR < 0.05) revealed that systemic LPS challenge induced a broad transcriptional reprogramming in the PVN. The enrichment analyses of the differentially expressed genes highlighted pathways related to immune-neuro-endocrine integration (Figure S2). Notably, Crh was identified as one of the most significantly upregulated transcripts (Fig. 1M). This was validated by quantitative real-time PCR (qPCR) (Fig. 1N). At the protein level, immunofluorescence staining further demonstrated that approximately 81% of c-Fos+ neurons were co-labeled with CRH (Fig. 1O, P), indicating that CRH-expressing neurons in the PVN constitute the predominant neuronal population robustly activated by systemic inflammation to drive pulmonary pathology. Systemic inflammation activates CRHPVN neurons through the vagus nerve Building on these findings, we next monitored CRHPVN neuronal activity in vivo to capture its dynamic responses to LPS challenge via fiber photometry. We injected AAV-DIO-jGCaMP7b virus and implanted optic fiber above the PVN of CRH-Cre mice (Fig. 2A). The accuracy of optic fiber placement was validated post-hoc (Fig. 2B). Upon LPS challenge, CRHPVN neurons exhibited a rapid and sustained activation profile, with significant calcium transients beginning at ~ 20 min, peaking at 2 h, and lasting over 12 h (Fig. 2C, D). In contrast, saline injection elicited no response, consistent with our c-Fos immunofluorescence data. In parallel, we assessed the effects of vagotomy and chemical sympathectomy by 6-hydroxydopamine (6-OHDA), a neurotoxin that selectively targets catecholaminergic nerve terminals, on the activation of CRHPVN neurons [26]. Vagotomy nearly abolished LPS-induced calcium transients in CRHPVN neurons, whereas 6-OHDA produced a more modest attenuation. We next performed bilateral subdiaphragmatic vagotomy to assess the role of vagal signaling in CRHPVN neuronal activation. Vagotomy completely abolished LPS-induced c-Fos expression in CRHPVN neurons (Fig. 2E, F). These data establish a “viscera-to-brain” axis wherein abdominal vagal afferents sense systemic inflammation and relay this signal to activate the PVN. Fig. 2Systemic inflammation activates CRHPVN neurons through vagus nerve. A, B Schematic and histological verification of fiber photometry recording in the PVN of CRH-Cre mice injected with AAV-hSyn-DIO-jGCaMP7b. Scale bar, 200 µm. C, D Real-time Ca2+ dynamics of CRHPVN neurons in freely moving mice. i.p. LPS challenge triggered a rapid and sustained increase in calcium signaling compared to saline, mice with sympathetic nerve ablation, the calcium signaling was attenuated, whereas in those with subdiaphragmatic vagotomy, calcium signals were nearly undetectable. (n≥3 mice/group). E, F Impact of afferent blockade. Bilateral subdiaphragmatic vagotomy (VGX) completely abolished LPS-induced c-Fos expression in CRH neurons, confirming that the PVN responds to abdominal/systemic signals. (n=3 mice/group). Scale bar, 100 µm Systemic inflammation activates CRHPVN neurons through vagus nerve. A, B Schematic and histological verification of fiber photometry recording in the PVN of CRH-Cre mice injected with AAV-hSyn-DIO-jGCaMP7b. Scale bar, 200 µm. C, D Real-time Ca2+ dynamics of CRHPVN neurons in freely moving mice. i.p. LPS challenge triggered a rapid and sustained increase in calcium signaling compared to saline, mice with sympathetic nerve ablation, the calcium signaling was attenuated, whereas in those with subdiaphragmatic vagotomy, calcium signals were nearly undetectable. (n≥3 mice/group). E, F Impact of afferent blockade. Bilateral subdiaphragmatic vagotomy (VGX) completely abolished LPS-induced c-Fos expression in CRH neurons, confirming that the PVN responds to abdominal/systemic signals. (n=3 mice/group). Scale bar, 100 µm CRHPVN neurons drive pulmonary neutrophil infiltration through the sympathetic nervous system We next sought to identify the descending pathway through which CRHPVN neurons regulate pulmonary neutrophilic inflammation. Previous studies have established the vagus nerve as a critical component of the “cholinergic anti-inflammatory pathway”, exerting a protective role in restraining excessive pulmonary inflammation induced by LPS challenge [27, 28], whereas our findings show that activation of CRHPVN neurons aggravates lung injury. Thus, we speculated that pro-inflammatory effects of CRHPVN neurons are not mediated by vagal efferents. We therefore hypothesized that the sympathetic nervous system may serve as the maladaptive effector arm. To further explore the polysynaptic autonomic pathway between the PVN and the lung, we injected retrograde transsynaptic tracer PRV-CAG-EGFP into the lung parenchyma of CRH-Cre mice (Fig. 3A). Five days later, EGFP-labeled neurons were detected in multiple lung-associated regions, including the rostral ventrolateral medulla (RVLM), locus coeruleus (LC), periaqueductal gray (PAG), central amygdala (CeA), and PVN (Figure S3). Notably, labeling was observed in the RVLM, a major sympathetic premotor region, and in the spinal IML (intermediolateral column), where sympathetic preganglionic neurons reside (Fig. 3C). These tracing data support the potential involvement of the RVLM and spinal IML as candidate nodes in a descending polysynaptic autonomic pathway linking the PVN to pulmonary sympathetic output, although they do not establish their functional necessity. We performed a quantitative analysis of retrogradely labeled neurons within the PVN. The results revealed that CRH neurons (mCherry+) comprised 72% of the neurons retrogradely traced from the lung (GFP+) (Fig. 3B and F), suggesting that CRH neurons constituted the predominant lung-associated PRV-labeled neuronal population in the PVN. Fig. 3CRHPVN neurons drive pulmonary neutrophil infiltration through the sympathetic nervous system. A Retrograde polysynaptic tracing using PRV-CAG-EGFP injected into the lung. B Representative images of PRV-labeled neurons in the PVN, colocalizing with CRH-mCherry. Scale bar, 100 µm. A local magnification is shown on the right, Scale bar, 50 µm. C Representative images of PRV+ neurons in the rostral ventrolateral medulla (RVLM) and spinal IML/lateral horn, delineating the descending sympathetic outflow. Scale bar, 500 µm (low mag), 200 µm (high mag). Scale bar, 200 µm (low mag), 100 µm (high mag). D, G Immunofluorescence of lung sections showing tyrosine hydroxylase (TH, sympathetic fibers) and MPO (neutrophils). LPS increased the density of both TH+ fibers and MPO+ cells (n=3 mice/group). Scale bar, 100 µm. A local magnification is shown on the right, Scale bar, 50 µm. E, H immunofluorescent staining assessing the spatial relationship between sympathetic fibers and immune cells. TH+ fibers (green) showed preferential spatial proximity to Ly6G+ neutrophils (yellow) compared to F4/80+ macrophages (magenta) or CD3+ T cells (red) (n=3 mice/group). F Quantification reveals that ~72% of lung-projecting PVN neurons are CRH+ (n=3 mice). *Data are presented as mean ± SEM. ***P < 0.001, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA with Tukey’s test CRHPVN neurons drive pulmonary neutrophil infiltration through the sympathetic nervous system. A Retrograde polysynaptic tracing using PRV-CAG-EGFP injected into the lung. B Representative images of PRV-labeled neurons in the PVN, colocalizing with CRH-mCherry. Scale bar, 100 µm. A local magnification is shown on the right, Scale bar, 50 µm. C Representative images of PRV+ neurons in the rostral ventrolateral medulla (RVLM) and spinal IML/lateral horn, delineating the descending sympathetic outflow. Scale bar, 500 µm (low mag), 200 µm (high mag). Scale bar, 200 µm (low mag), 100 µm (high mag). D, G Immunofluorescence of lung sections showing tyrosine hydroxylase (TH, sympathetic fibers) and MPO (neutrophils). LPS increased the density of both TH+ fibers and MPO+ cells (n=3 mice/group). Scale bar, 100 µm. A local magnification is shown on the right, Scale bar, 50 µm. E, H immunofluorescent staining assessing the spatial relationship between sympathetic fibers and immune cells. TH+ fibers (green) showed preferential spatial proximity to Ly6G+ neutrophils (yellow) compared to F4/80+ macrophages (magenta) or CD3+ T cells (red) (n=3 mice/group). F Quantification reveals that ~72% of lung-projecting PVN neurons are CRH+ (n=3 mice). *Data are presented as mean ± SEM. ***P < 0.001, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA with Tukey’s test We then asked whether sympathetic innervation spatially correlates with neutrophil recruitment in LPS-induced ALI. Immunofluorescence staining revealed that LPS challenge markedly increased tyrosine hydroxylase (TH), a marker of sympathetic nerves, and myeloperoxidase (MPO), a neutrophil marker, with strong spatial correlation between TH+ fibers and MPO+cells (Fig. 3D and G). To determine the cell-type specificity of this interaction, we performed high-resolution multiplex imaging of TH alongside markers for neutrophils (Ly6G), macrophages (F4/80), and T cells (CD3). We observed a striking spatial proximity between TH+ sympathetic nerves and Ly6G+ neutrophils, but rarely with macrophages or T cells (Fig. 3E and H). This distinct spatial arrangement suggests that pulmonary sympathetic fibers are spatially associated with neutrophils, providing a structural basis for the precise adrenergic modulation of neutrophilic inflammation. To further characterize the catecholaminergic milieu associated with LPS-induced lung injury, we measured norepinephrine (NE) and epinephrine (EPI) levels in plasma, BALF, and lung homogenates. LPS challenge increased NE levels in all three compartments, including plasma, BALF, and lung homogenates (Figure S4H–J). In contrast, EPI was significantly increased in plasma but remained unchanged in BALF and lung homogenates (Figure S4K–M). This compartmental pattern suggests that LPS-induced systemic inflammation is associated with a NE-dominant adrenergic response in the pulmonary compartment, supporting a contribution of pulmonary sympathetic/adrenergic signaling rather than circulating adrenal EPI alone. Inhibition of CRHPVN neurons alleviates LPS-induced acute lung injury To establish the optimal window for assessing the impact of neuronal modulation, we first characterized the temporal profile of inflammation and PVN activation following LPS challenge. In wild-type (WT) mice, we induced an LPS-induced ALI model, with saline injection serving as the control. Mice were sacrificed at 6, 12, and 24 h post-modeling for sample collection and analysis (Figure S4A). Our findings revealed a significant increase of IL-6, TNF-α, and lung injury scores at 24 h post-LPS compared to controls (Figure S4B-G). This inflammatory peak coincided with robust c-Fos expression in the PVN, leading us to select the 24-hour time point for all subsequent endpoint analyses. We then asked whether the timing of CRHPVN neuron inhibition influences the severity of ALI. We bilaterally injected AAV-DIO-hM4Di-mCherry into the PVN of CRH-Cre mice to enable chemogenetic suppression. After a 3-week period for viral expression, we administered CNO at 0, 6, and 12 h post-LPS induction, with subsequent doses given every 6 h to maintain continuous inhibition of CRH neurons until sampling (Fig. 4A). Prophylactic inhibition (CNO administration immediately at 0 h post-LPS induction) was most effective in reducing IL-6, TNF-α, and the lung injury score (Fig. 4B-E). This suggests that early intervention targeting CRHPVN neurons confers maximal protection. Consequently, we chose to administer CNO immediately upon LPS modeling for subsequent experiments. Fig. 4Inhibition of CRHPVN neurons alleviates acute lung injury and reduce NET release. A Schematic of the prophylactic inhibition protocol. CRH-Cre mice expressing hM4Di (inhibitory DREADD) received CNO (1 mg/kg) immediately (0h), 6h, or 12h after i.p. LPS (10 mg/kg). B–E Determination of the optimal therapeutic window. CRH-Cre mice expressing hM4Di (inhibitory DREADD) received CNO (1 mg/kg) immediately (0h), 6h, or 12h after i.p. LPS (10 mg/kg). Immediate inhibition (0h) provided the most significant reduction in lung injury scores, serum IL-6, and serum TNF-α. Lung injury scores were analyzed using the Mann-Whitney U test (n=6 mice/group). Scale bar, 100 µm. F Experimental design for the optimized inhibition study (CNO administered at 0h). G, H ELISA quantification of serum IL-6 and TNF-α. Inhibition of CRHPVN neurons significantly blunted the systemic cytokine storm induced by LPS (n=8 mice/group). I, K Representative H&E staining and histological scoring. hM4Di-mediated inhibition preserved alveolar architecture and reduced interstitial edema compared to mCherry-LPS controls. (n=5 mice/group). Scale bar, 100 µm. J, L Evaluation of Neutrophil Extracellular Traps (NETs). Lung sections were stained for Citrullinated Histone H3 (CitH3, green) and MPO (red). Inhibition significantly reduced the percentage of NET-forming neutrophils (n=5 mice/group). Scale bar, 100 µm (low mag), 50 µm (high mag). M, N Western blot analysis of tyrosine hydroxylase (TH) in lung tissues homogenates. LPS-induced upregulation of TH protein was significantly attenuated by PVN inhibition, linking central inhibition to reduced peripheral sympathetic tone (n=3 biological replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA (cytokines/WB) Inhibition of CRHPVN neurons alleviates acute lung injury and reduce NET release. A Schematic of the prophylactic inhibition protocol. CRH-Cre mice expressing hM4Di (inhibitory DREADD) received CNO (1 mg/kg) immediately (0h), 6h, or 12h after i.p. LPS (10 mg/kg). B–E Determination of the optimal therapeutic window. CRH-Cre mice expressing hM4Di (inhibitory DREADD) received CNO (1 mg/kg) immediately (0h), 6h, or 12h after i.p. LPS (10 mg/kg). Immediate inhibition (0h) provided the most significant reduction in lung injury scores, serum IL-6, and serum TNF-α. Lung injury scores were analyzed using the Mann-Whitney U test (n=6 mice/group). Scale bar, 100 µm. F Experimental design for the optimized inhibition study (CNO administered at 0h). G, H ELISA quantification of serum IL-6 and TNF-α. Inhibition of CRHPVN neurons significantly blunted the systemic cytokine storm induced by LPS (n=8 mice/group). I, K Representative H&E staining and histological scoring. hM4Di-mediated inhibition preserved alveolar architecture and reduced interstitial edema compared to mCherry-LPS controls. (n=5 mice/group). Scale bar, 100 µm. J, L Evaluation of Neutrophil Extracellular Traps (NETs). Lung sections were stained for Citrullinated Histone H3 (CitH3, green) and MPO (red). Inhibition significantly reduced the percentage of NET-forming neutrophils (n=5 mice/group). Scale bar, 100 µm (low mag), 50 µm (high mag). M, N Western blot analysis of tyrosine hydroxylase (TH) in lung tissues homogenates. LPS-induced upregulation of TH protein was significantly attenuated by PVN inhibition, linking central inhibition to reduced peripheral sympathetic tone (n=3 biological replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA (cytokines/WB) We next bilaterally delivered AAV-DIO-hM4Di-mCherry into the PVN of CRH-Cre mice to test the effect of silencing CRHPVN neurons on ALI progression. Three weeks later, mice were challenged with LPS and administered CNO to engage Gi-DREADD–mediated inhibition (Fig. 4F). Compared with mCherry controls, hM4Di-mediated inhibition of CRHPVN neurons markedly reduced circulating proinflammatory cytokines (Fig. 4G, H). Consistent with these systemic effects, lung histopathology revealed attenuated alveolar injury and neutrophilic infiltration, and the burden of NETs was significantly decreased in hM4Di mice (Fig. 4I-L). Notably, this protection was accompanied by a marked reduction in pulmonary TH immunoreactivity in hM4Di mice (Fig. 4M, N), indicative of diminished sympathetic tone. Together, these findings demonstrate that inhibition of CRHPVN neurons alleviates LPS-induced acute lung injury, and suggest that the protective effect is mediated, at least in part, through downregulation of sympathetic outflow to the lung. Activation of CRHPVN neurons aggravates LPS-induced acute lung injury To determine whether the timing of CRHPVN neuron activation modulates ALI severity—paralleling our inhibition approach—we expressed the excitatory DREADD hM3Dq in these neurons and administered CNO at 0, 6, or 12 h post-LPS, with supplemental doses every 6 h (Fig. 5A). Activation initiated concurrently with LPS challenge (0 h) produced the most pronounced elevations in lung injury scores and proinflammatory cytokines, including IL-6 and TNF-α (Fig. 5B-E), prompting the adoption of this early-activation paradigm in subsequent experiments. Fig. 5Activation of CRHPVN neurons aggravates acute lung injury and increase NET release. A Schematic of the activation protocol using hM3Dq (excitatory DREADD). B–E Time-course of activation. Concurrent activation (0h) with LPS induced the most severe phenotype, elevating lung injury scores and cytokines beyond LPS alone. (n≥5 mice/group). Scale bar, 100 µm. F Experimental design for specific activation. G, H ELISA quantification of serum pro-inflammatory cytokines. Activation of CRHPVN neurons significantly amplified IL-6 and TNF-α levels, suggesting a central driver of the "cytokine storm" (n=6 mice/group). I, K Histological assessment. Activation exacerbated alveolar hemorrhage and neutrophil infiltration. Lung injury scores were analyzed using the Mann-Whitney U test. (n=5 mice/group). Scale bar, 100 µm. J, L Quantification of NETosis. Activation increased the abundance of CitH3+MPO+ complexes in the lung, indicating enhanced neutrophil pathogenicity (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. M, N Western blot analysis. Chemogenetic activation further increased lung TH protein expression relative to LPS controls, confirming enhanced sympathetic innervation/tone (n=3 biological replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA (cytokines/WB) Activation of CRHPVN neurons aggravates acute lung injury and increase NET release. A Schematic of the activation protocol using hM3Dq (excitatory DREADD). B–E Time-course of activation. Concurrent activation (0h) with LPS induced the most severe phenotype, elevating lung injury scores and cytokines beyond LPS alone. (n≥5 mice/group). Scale bar, 100 µm. F Experimental design for specific activation. G, H ELISA quantification of serum pro-inflammatory cytokines. Activation of CRHPVN neurons significantly amplified IL-6 and TNF-α levels, suggesting a central driver of the "cytokine storm" (n=6 mice/group). I, K Histological assessment. Activation exacerbated alveolar hemorrhage and neutrophil infiltration. Lung injury scores were analyzed using the Mann-Whitney U test. (n=5 mice/group). Scale bar, 100 µm. J, L Quantification of NETosis. Activation increased the abundance of CitH3+MPO+ complexes in the lung, indicating enhanced neutrophil pathogenicity (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. M, N Western blot analysis. Chemogenetic activation further increased lung TH protein expression relative to LPS controls, confirming enhanced sympathetic innervation/tone (n=3 biological replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ****P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA (cytokines/WB) We next specifically activated CRHPVN neurons by delivering AAV-DIO-hM3Dq-mCherry into the PVN of CRH-Cre mice (Fig. 5F). Chemogenetic stimulation of CRHPVN neurons significantly elevated plasma levels of IL-6 and TNF-α (Fig. 5G, H), and was accompanied by exacerbated lung injury, enhanced neutrophil aggregation, and increased NET release relative to mCherry controls (Fig. 5I-L). Notably, this aggravated phenotype correlated with upregulated TH expression in lung tissues (Fig. 5M, N). Collectively, these data suggest that activation of CRHPVN neurons was associated with increased pulmonary sympathetic tone, thereby worsening the lung injury phenotype. Sympathetic ablation ameliorates LPS-induced acute lung injury To determine whether this sympathetically mediated pathway is necessary for LPS-induced ALI, we performed chemical sympathectomy using 6-OHDA [26]. Mice were pretreated with 6-OHDA (100 mg/kg, i.p.) for three days prior to LPS challenge (Fig. 6A). Compared with controls, 6-OHDA pre-treatment markedly reduced plasma IL-6 and TNF-α (Fig. 6B, C) and ameliorated histopathological lung injury (Fig. 6D and F). Concordantly, pulmonary neutrophil accumulation and NETs burden were substantially decreased (Fig. 6E and G), indicating that ablation of sympathetic nerves blunts neutrophil-driven acute lung injury. Effective denervation was verified by a pronounced decrease in TH immunoreactivity (Fig. 6H, I). These data indicate that sympathetic signaling is a key driver of neutrophilic inflammation and tissue damage in LPS-induced ALI. Fig. 6Sympathetic ablation reverses the exacerbation of lung injury induced by CRHPVN neuron activation. A Schematic of the chemical sympathectomy model. Mice were pre-treated with 6-OHDA (100 mg/kg, i.p.) for 3 days to deplete systemic sympathetic terminals prior to LPS challenge. B, C Impact of sympathectomy on systemic inflammation. 6-OHDA treatment significantly reduced plasma IL-6 and TNF-α levels in LPS-challenged mice (n=6 mice/group). D, F Histological assessment. Sympathetic ablation ameliorated LPS-induced lung injury, reducing alveolar wall thickening and cellular infiltration. Lung injury scores were analyzed using the Mann-Whitney U test (n=5 mice/group). Scale bar, 100 µm. E, G Quantification of NETs (MPO-DNA complexes) in lung. 6-OHDA significantly reduced NET burden (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. H, I Western blot analysis confirmed a significant decrease in TH protein content in mouse lung tissues following 6-OHDA treatment. (n=3 replicates). J Experiment Schematic. hM3Dq-expressing mice received 6-OHDA pre-treatment to determine if the exacerbating effects of CRHPVN activation depend on the sympathetic nervous system.K, L Cytokine analysis. 6-OHDA blocked the hM3Dq-driven increase in IL-6 and TNF-α (n=8 mice/group). M, O Histological assessment. Sympathectomy prevented the severe lung damage induced by concurrent LPS and CRHPVN activation (n=5 mice/group). Scale bar, 100 µm. N, P Quantification of NETs (MPO-DNA complexes) in lung. The increase in CitH3+ neutrophils driven by CRHPVN activation was abolished by sympathetic ablation (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. Q, R Western blot validation the TH depletion. (n=4 replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA with Tukey’s test Sympathetic ablation reverses the exacerbation of lung injury induced by CRHPVN neuron activation. A Schematic of the chemical sympathectomy model. Mice were pre-treated with 6-OHDA (100 mg/kg, i.p.) for 3 days to deplete systemic sympathetic terminals prior to LPS challenge. B, C Impact of sympathectomy on systemic inflammation. 6-OHDA treatment significantly reduced plasma IL-6 and TNF-α levels in LPS-challenged mice (n=6 mice/group). D, F Histological assessment. Sympathetic ablation ameliorated LPS-induced lung injury, reducing alveolar wall thickening and cellular infiltration. Lung injury scores were analyzed using the Mann-Whitney U test (n=5 mice/group). Scale bar, 100 µm. E, G Quantification of NETs (MPO-DNA complexes) in lung. 6-OHDA significantly reduced NET burden (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. H, I Western blot analysis confirmed a significant decrease in TH protein content in mouse lung tissues following 6-OHDA treatment. (n=3 replicates). J Experiment Schematic. hM3Dq-expressing mice received 6-OHDA pre-treatment to determine if the exacerbating effects of CRHPVN activation depend on the sympathetic nervous system.K, L Cytokine analysis. 6-OHDA blocked the hM3Dq-driven increase in IL-6 and TNF-α (n=8 mice/group). M, O Histological assessment. Sympathectomy prevented the severe lung damage induced by concurrent LPS and CRHPVN activation (n=5 mice/group). Scale bar, 100 µm. N, P Quantification of NETs (MPO-DNA complexes) in lung. The increase in CitH3+ neutrophils driven by CRHPVN activation was abolished by sympathetic ablation (n=5 mice/group). Scale bar, 100 µm. Local enlargings are shown below. Scale bar, 50 µm. Q, R Western blot validation the TH depletion. (n=4 replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; ns, not significant. Statistical comparisons via one-way ANOVA with Tukey’s test Sympathetic ablation reverses the exacerbation of lung injury induced by CRHPVN neuron activation We next investigated whether the detrimental effects of CRHPVN neuron overactivation were mediated by the sympathetic nervous system. To this end, we combined chemogenetic activation of CRHPVN neurons via AAV-DIO-hM3Dq-mCherry in CRH-Cre mice with systemic 6-OHDA–mediated sympathetic ablation (Fig. 6J). Sympathetic ablation largely attenuated the aggravated lung injury phenotype induced by neuronal activation: circulating IL-6 and TNF-α were significantly reduced (Fig. 6K, L), histopathological injury was attenuated, and both neutrophil accumulation and NETs burden were diminished (Fig. 6M-P). Pulmonary TH levels were also decreased (Fig. 6Q, R), confirming effective suppression of sympathetic drive. Together, our results demonstrate that the lung injury driven by CRHPVN neuron activation is strongly influenced by sympathetic output, and that targeting this downstream axis can therapeutically mitigate neuroimmune exacerbation of ALI. Sympathetic nervous system delays neutrophil apoptosis via ADRB2 in ARDS Building on our evidence that CRHPVN neuron activation exacerbates neutrophil-driven lung injury predominantly via the sympathetic nervous system, and that sympathetic terminals are closely apposed to neutrophils in the lung, we sought to investigate the molecular mechanism mediating this sympathetic-neutrophil crosstalk. We focused on β-adrenergic receptors, the predominant adrenoceptor on neutrophils and its known regulation of neutrophil migration, apoptosis, and inflammatory cytokine release [29, 30]. To explore this issue, we collected peripheral blood samples from patients with septic ARDS and healthy volunteers (Fig. 7A) (Table S1). We then applied qPCR to quantify the transcript levels of ADRB1 and ADRB2 in neutrophils. Compared with healthy controls, both ADRB1 and ADRB2 mRNA were upregulated in neutrophils from septic ARDS patients, with a more pronounced increase in ADRB2 (Fig. 7B). Flow cytometry further revealed a higher frequency of ADRB2+CD11b+ neutrophils (Fig. 7C, D). These findings suggest that neutrophils from septic ARDS patients may exhibit increased ADRB2-associated responsiveness to catecholaminergic signaling. Fig. 7Sympathetic nervous system regulates neutrophil apoptosis via ADRB2-PI3K/Akt pathway in ARDS. A Schematic of human neutrophil isolation from healthy volunteers (Ctrl) and ARDS patients. Neutrophils were isolated via density gradient centrifugation using Ficoll-Hypaque (GE Healthcare) and 3% dextran sedimentation, followed by hypotonic lysis to remove residual erythrocytes. Cell viability was confirmed to be >95% by trypan blue exclusion. Neutrophils were maintained in complete RPMI 1640 medium containing 10% FBS and 1% penicillin/streptomycin. For in vitro experiments, neutrophils were seeded in 24-well plates at a density of 1 × 10⁶ cells/mL. B qPCR analysis showing specific upregulation of ADRB2 mRNA, but not ADRB1, in ARDS neutrophils compared to healthy controls (n=3 donors/group). C, D Flow cytometric analysis. Representative image and quantification show increased surface expression of ADRB2 on CD11b+ neutrophils in ARDS patients. Fluorescence Minus One (FMO) controls were used to determine gating (n≥3 donors/group). E, F Western blot analysis of Akt phosphorylation. ARDS neutrophils exhibit elevated basal p-Akt, which is significantly reduced by ex vivo treatment with the selective β2-antagonist ICI 118,551 (n=3 independent experiments). G, H Apoptosis assay (Annexin V/PI). Neutrophils were cultured for 24 hours. ARDS neutrophils showed delayed apoptosis (low Annexin V+). Treatment with ICI 118,551 significantly increased the percentage of apoptotic cells (n=3 donors/group). I, J Mechanistic validation in Healthy Neutrophils. The PI3K/Akt activator SC79 (Aa) mimicked the survival effect of catecholamines, and ICI 118,551 neutralized the effect (n=6 replicates). K, L Mechanistic rescue in ARDS Neutrophils. The ability of ICI 118,551 to promote apoptosis was reversed by co-treatment with SC79, confirming the ADRB2-PI3K/Akt axis (n=6 replicates). M–P Validation with Norepinephrine (NE) and Akt Inhibitor (AKTi). (M, N) In healthy cells, NE delayed apoptosis, an effect blocked by AKTi (n=6 replicates). O, P In ARDS cells, AKTi alone significantly promoted apoptosis, overriding the pro-survival effect of endogenous or exogenous adrenergic signaling (n=6 replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Statistical comparisons via one-way ANOVA with Tukey’s test Sympathetic nervous system regulates neutrophil apoptosis via ADRB2-PI3K/Akt pathway in ARDS. A Schematic of human neutrophil isolation from healthy volunteers (Ctrl) and ARDS patients. Neutrophils were isolated via density gradient centrifugation using Ficoll-Hypaque (GE Healthcare) and 3% dextran sedimentation, followed by hypotonic lysis to remove residual erythrocytes. Cell viability was confirmed to be >95% by trypan blue exclusion. Neutrophils were maintained in complete RPMI 1640 medium containing 10% FBS and 1% penicillin/streptomycin. For in vitro experiments, neutrophils were seeded in 24-well plates at a density of 1 × 10⁶ cells/mL. B qPCR analysis showing specific upregulation of ADRB2 mRNA, but not ADRB1, in ARDS neutrophils compared to healthy controls (n=3 donors/group). C, D Flow cytometric analysis. Representative image and quantification show increased surface expression of ADRB2 on CD11b+ neutrophils in ARDS patients. Fluorescence Minus One (FMO) controls were used to determine gating (n≥3 donors/group). E, F Western blot analysis of Akt phosphorylation. ARDS neutrophils exhibit elevated basal p-Akt, which is significantly reduced by ex vivo treatment with the selective β2-antagonist ICI 118,551 (n=3 independent experiments). G, H Apoptosis assay (Annexin V/PI). Neutrophils were cultured for 24 hours. ARDS neutrophils showed delayed apoptosis (low Annexin V+). Treatment with ICI 118,551 significantly increased the percentage of apoptotic cells (n=3 donors/group). I, J Mechanistic validation in Healthy Neutrophils. The PI3K/Akt activator SC79 (Aa) mimicked the survival effect of catecholamines, and ICI 118,551 neutralized the effect (n=6 replicates). K, L Mechanistic rescue in ARDS Neutrophils. The ability of ICI 118,551 to promote apoptosis was reversed by co-treatment with SC79, confirming the ADRB2-PI3K/Akt axis (n=6 replicates). M–P Validation with Norepinephrine (NE) and Akt Inhibitor (AKTi). (M, N) In healthy cells, NE delayed apoptosis, an effect blocked by AKTi (n=6 replicates). O, P In ARDS cells, AKTi alone significantly promoted apoptosis, overriding the pro-survival effect of endogenous or exogenous adrenergic signaling (n=6 replicates). *Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Statistical comparisons via one-way ANOVA with Tukey’s test ADRB2 delays neutrophil apoptosis and exacerbates pulmonary inflammation via the PI3K/Akt signaling pathway in ARDS Next, we evaluated whether the functional properties of these ADRB2-high neutrophils—such as chemotaxis, apoptosis, and NETosis—were altered. Given the critical role of delayed neutrophil apoptosis in sustaining inflammation during ARDS, we examined the impact of upregulated ADRB2 expression on neutrophil survival. Neutrophils isolated from the peripheral blood of healthy volunteers and septic ARDS patients were treated in vitro with the highly selective ADRB2 antagonist ICI 118,551, and subsequently subjected to apoptosis quantification by flow cytometry. ADRB2 blockade significantly promoted apoptosis in neutrophils from septic ARDS patients (Fig. 7G, H), indicating that ADRB2 signaling contributes to delaying neutrophil apoptosis in ARDS. Chemotaxis and NETosis were evaluated in parallel (Figure S5A-C). Our results indicate that selective blockade of ADRB2 suppressed neutrophil chemotaxis and NETs. Additionally, ICI 118,551 attenuated the inflammatory response and alleviated the lung injury in mice (Figure S5D-G). As an important downstream pathway of β2-adrenergic receptor signaling, PI3K/Akt is also a central pro-survival pathway in neutrophils [31]. Therefore, we investigated whether the PI3K/Akt pathway was involved in the regulation of neutrophil function by β2-adrenergic receptor. We examined Akt phosphorylation after ADRB2 inhibition. ICI 118,551 treatment significantly reduced phospho-Akt (p-Akt) in ARDS neutrophils (Fig. 7E, F). We next used pharmacological interventions to test the functional involvement of Akt signaling. The Akt activator SC79 decreased apoptosis and partially rescued the pro-apoptotic effect of ADRB2 blockade (Fig. 7I-L), whereas catecholamine stimulation with norepinephrine reduced apoptosis and this effect was abolished by the Akt inhibitor AKTi VII (Fig. 7M-P). These data support the involvement of an ADRB2–PI3K/Akt pathway in regulating neutrophil survival in septic ARDS, which may contribute to sustained neutrophilic inflammation. This cell-intrinsic mechanism works in concert with the sympathetic neuroimmune pathway defined above to drive the progression of ALI. Discussion The nervous system is increasingly recognized as a master regulator of immunity, but the neural circuits that shape pulmonary inflammation during systemic sepsis remain poorly defined [32]. Here, by integrating functional manipulation of specific brain nuclei with immunological phenotyping and human neutrophil analyses, we show that CRHPVN neurons–sympathetic nerve pathway drives acute lung injury during ARDS. We demonstrate that LPS-induced systemic inflammatory signals are transmitted to the brain via the vagus nerve, leading to the robust activation of CRH-expressing neurons in the PVN region. These CRHPVN neurons amplify lung injury through sympathetic outflow to the pulmonary microenvironment. There, adrenergic signaling acts on ADRB2 on neutrophils to delay apoptosis and enhance NETosis via the PI3K/Akt pathway, ultimately exacerbating tissue damage (Fig. 8). Fig. 8Schematic of Brain–lung circuit in regulating LPS-induced lung injury via a sympathetic-neutrophil axis. (Left) CRHPVN Inhibition: Under basal conditions or with PVN inhibition, sympathetic outflow is low. Neutrophils undergo normal apoptosis and clearance, resolving inflammation. (Right) CRHPVN Activation (ARDS): Systemic inflammatory signals are relayed to the brain via the vagus afferents, triggering robust activation of CRH-expressing neurons in the PVN of the hypothalamus. These CRHPVN neurons engage the RVLM → IML → sympathetic chain pathway to drive pulmonary sympathetic outflow. Norepinephrine released from sympathetic terminals acts on ADRB2 on neutrophils, activating PI3K/Akt signaling to delay neutrophil apoptosis, increase NET formation, and exacerbate lung injury. Abbreviations: ARDS, acute respiratory distress syndrome; ADRB2, β2-adrenergic receptor; CRH, corticotropin-releasing hormone; ICI, ICI 118,551; PVN, paraventricular nucleus; RVLM, rostral ventrolateral medulla; IML, intermediolateral column; NE, norepinephrine; NETs, neutrophil extracellular traps; 6-OHDA, 6-hydroxydopamine Schematic of Brain–lung circuit in regulating LPS-induced lung injury via a sympathetic-neutrophil axis. (Left) CRHPVN Inhibition: Under basal conditions or with PVN inhibition, sympathetic outflow is low. Neutrophils undergo normal apoptosis and clearance, resolving inflammation. (Right) CRHPVN Activation (ARDS): Systemic inflammatory signals are relayed to the brain via the vagus afferents, triggering robust activation of CRH-expressing neurons in the PVN of the hypothalamus. These CRHPVN neurons engage the RVLM → IML → sympathetic chain pathway to drive pulmonary sympathetic outflow. Norepinephrine released from sympathetic terminals acts on ADRB2 on neutrophils, activating PI3K/Akt signaling to delay neutrophil apoptosis, increase NET formation, and exacerbate lung injury. Abbreviations: ARDS, acute respiratory distress syndrome; ADRB2, β2-adrenergic receptor; CRH, corticotropin-releasing hormone; ICI, ICI 118,551; PVN, paraventricular nucleus; RVLM, rostral ventrolateral medulla; IML, intermediolateral column; NE, norepinephrine; NETs, neutrophil extracellular traps; 6-OHDA, 6-hydroxydopamine The PVN is a key neuroendocrine and autonomic hub that coordinates systemic stress responses through both the HPA axis and sympathetic outflow [33]. Here, we establish the PVN and its CRH-expressing neurons as a central neuroimmune hub in LPS-induced lung injury. Using bilateral subdiaphragmatic vagotomy coupled with in vivo calcium imaging, we delineate a dedicated ascending pathway “systemic inflammation → vagal afferents → PVN activation”. This indicates that the PVN is not a passive responder to circulating cytokines but is engaged through a defined neural route that senses inflammation in the visceral compartment. Functionally, chemogenetic excitation of CRHPVN neurons exacerbated hallmark ALI phenotypes, including neutrophil infiltration, NETosis, and cytokine surges, whereas their inhibition significantly attenuated LPS-induced lung injury, indicating that CRHPVN neurons as both necessary and sufficient drivers of pathology. Notably, the pro-inflammatory effects of CRH neuron activation were accompanied by increased pulmonary TH expression and were reversed by chemical sympathectomy (6-OHDA), supporting sympathetic output as an important effector arm of CRHPVN neuron-driven neutrophilic inflammation. However, CRHPVN neurons can also engage the HPA axis. Because corticosterone levels were not measured and HPA-axis dependence was not directly tested, a modulatory contribution of glucocorticoid signaling cannot be excluded and should be addressed in future studies. We identify a CRHPVN–sympathetic–neutrophil axis that anatomically aligns with the circuit recently reported by Li et al. [14]; however, our findings reveal a divergent functional outcome in systemic inflammatory lung injury—exacerbation rather than protection. This conclusion is also consistent with another recent independent study by Li W et al. [13], which showed that in a severe pneumonia model, the CeA–RVLM–lung sympathetic pathway promotes pulmonary inflammatory responses and lung injury through ADRB2+ interstitial macrophages. Together, these studies suggest that brain-to-lung sympathetic signaling is not uniformly protective or injurious; rather, its net effect depends on the inflammatory trigger, route of stimulation, disease phase, and immune cell target. Our PRV tracing data identified the RVLM and spinal IML as candidate nodes in a polysynaptic autonomic pathway linking the brain to pulmonary sympathetic output. The RVLM may serve as a convergence point for brain regions implicated in pulmonary sympathetic control, including the PVN in our study and the CeA reported by Li W et al. Direct RVLM-specific recording, inhibition, or projection-defined manipulation will be required to establish its causal role in this pathway. The vagus nerve also showed a functionally complex role in this model. On one hand, subdiaphragmatic vagotomy abolished LPS-induced CRHPVN neuronal activation, supporting an essential role for vagal afferents in transmitting systemic inflammatory signals to the brain. On the other hand, vagotomy aggravated lung injury and inflammatory responses (Figure S6), consistent with disruption of the protective cholinergic anti-inflammatory pathway. This apparent paradox highlights the mixed sensory and efferent nature of the vagus nerve. Rather than being categorized as uniformly anti-inflammatory, vagal signaling should be considered in terms of its distinct afferent and efferent arms. In systemic endotoxemia, abdominal vagal afferents may initiate a pathogenic central sympathetic response, whereas vagal efferents may simultaneously restrain inflammation. The overall inflammatory outcome likely reflects the balance between these opposing functions. At the immune effector level, our data identify ADRB2–PI3K/Akt signaling as a mechanism by which sympathetic output may prolong neutrophil-mediated inflammation. ADRB2 has conventionally been ascribed an anti-inflammatory role, linked to its capacity to suppress macrophage activation [34]. However, accumulating evidence indicates that ADRB2 signaling can also amplify inflammation by enhancing LPS-induced proinflammatory gene expression in monocytes and macrophages [35, 36]. In the present study, our data show that in neutrophils from septic ARDS patients, ADRB2 expression was increased, and pharmacological ADRB2 blockade promoted apoptosis. Conversely, norepinephrine reduced neutrophil apoptosis, and this effect was dependent on Akt activity. These findings support a model in which enhanced adrenergic signaling in systemic inflammatory lung injury engages ADRB2–PI3K/Akt signaling to delay neutrophil apoptosis, thereby sustaining neutrophil persistence and pulmonary inflammation. The human ARDS cohort, however, had substantial clinical heterogeneity and was not comorbidity-matched with healthy controls. Larger comorbidity-matched cohorts and validation in mouse primary neutrophils will be needed to strengthen the translational relevance of this pathway. NET-associated neutrophil activation may represent another inflammatory feature downstream of this neuroimmune axis. In this study, increased Cit-H3 signals were interpreted as NET-associated neutrophil activation, rather than definitive biochemical evidence of extracellular NET release. This interpretation is consistent with previous work from our group linking NET release to LPS- or endotoxemia-induced lung injury, including studies showing that suppression of NET-associated markers such as MPO-histone colocalization and MPO-DNA complexes is accompanied by reduced pulmonary inflammation and tissue injury [37], that PD-L1 maintains NET release in endotoxin-induced lung injury [38], and that GSDMD-mediated NETosis promotes ARDS development [39]. Nevertheless, direct extracellular NET quantification and NET-targeting interventions were not performed in the present study. Therefore, although NET-associated changes may contribute to inflammatory amplification downstream of the CRH^PVN–sympathetic–ADRB2 axis, whether NET formation is required for this lung injury phenotype remains to be determined. Multiple regulated cell death pathways may also participate in this neuroimmune inflammatory network. In this study, we focused on neutrophil apoptosis and NETosis because neutrophils emerged as the principal immune effector downstream of the CRHPVN–sympathetic pathway. However, pyroptosis, necroptosis, ferroptosis, and PANoptosis have all been implicated in ALI/ARDS in a cell type–dependent manner [40, 41]. Endothelial pyroptosis, particularly through the caspase-11–GSDMD pathway, may increase vascular permeability and facilitate neutrophil recruitment [42], whereas macrophage or microglial inflammasome activation may amplify cytokine production and further modulate neural and immune responses [43, 44]. Conversely, prolonged neutrophil survival and NET release may provide DAMPs that reinforce inflammasome activation, endothelial injury, and additional cell death programs. Therefore, the CRHPVN–sympathetic–neutrophil axis should be viewed as part of a broader inflammatory amplification network rather than as an isolated mechanism. From a translational perspective, our work suggests several potential therapeutic avenues. Targeting the CRHPVN-sympathetic-neutrophil axis may help modulate deleterious neuroimmune crosstalk in acute lung injury. Vagus nerve stimulation and sympathetic inhibition have shown protective effects in experimental lung injury models. Li et al. [45] reported that vagus nerve stimulation reduced IL-1β and TNF-α levels and promoted macrophage polarization from an M1-like toward an M2-like phenotype. Other neuromodulatory strategies, including α7 nicotinic acetylcholine receptor activation, stellate ganglion block, and adrenergic pathway modulation, have also shown protective effects in experimental or clinical inflammatory settings [46–49]. Together, these findings support a broader therapeutic framework that integrates neuromodulation with immune-specific targeting. However, several issues require further study before this concept can be generalized. The animal experiments in this study primarily used an intraperitoneal LPS-induced systemic inflammatory lung injury model, and future studies using polymicrobial sepsis, infectious pneumonia, or local airway inflammation models will be needed to determine the broader applicability of this pathway. Only male mice were used, and potential sex-dependent differences should also be assessed. In addition, although lung injury was evaluated using histological, cellular, and inflammatory readouts, direct measurements of pulmonary edema, vascular permeability, MPO enzymatic activity, and IL-1β/IL-18 levels in serum or BALF would further strengthen phenotypic characterization and clarify inflammasome-related inflammation. Conclusions In summary, our study identifies a maladaptive brain–lung neuroimmune pathway that exacerbates LPS-induced systemic inflammatory lung injury. We demonstrate that systemic inflammatory signals are conveyed to the brain via the abdominal vagal afferents and strongly activate CRH-expressing neurons in the PVN of the hypothalamus. These CRH neurons then target the neutrophils in the lung through the sympathetic nervous system, driving the neutrophil-mediated ADRB2-PI3K/Akt signaling pathway, resulting in delayed cell apoptosis and NETosis, ultimately exacerbating lung injury. Together, our findings provide a mechanistic framework for understanding how the central nervous system senses systemic inflammatory signals and amplifies pathological immune responses in the lung. This suggests that we can target specific nodes within this neuroimmune axis—from CRHPVN neurons to the neutrophil PI3K/Akt pathway—to provide new therapeutic intervention approaches for ARDS and potential other neuro-inflammatory disorders. Supplementary Information Supplementary Material 1: Figure S1 Brainwide mapping of LPS-induced c-Fos expression. Figure S2 Enrichment analysis of the differentially expressed genes. Figure S3 PRV retrograde tracing of lung-associated neural pathways. Figure S4 Defining the optimal time points for assessing inflammation, cytokine dynamics, and neuronal modulation. Figure S5 ADRB2 antagonism suppresses neutrophil chemotaxis and NETosis to mitigate lung injury. Figure S6 Bilateral subdiaphragmatic vagotomy aggravates LPS-induced lung injury. Table S1 The demographic data of the septic ARDS patients included. Table S2 Primer pairs for RT-qPCR. Supplementary Material 1: Figure S1 Brainwide mapping of LPS-induced c-Fos expression. Figure S2 Enrichment analysis of the differentially expressed genes. Figure S3 PRV retrograde tracing of lung-associated neural pathways. Figure S4 Defining the optimal time points for assessing inflammation, cytokine dynamics, and neuronal modulation. Figure S5 ADRB2 antagonism suppresses neutrophil chemotaxis and NETosis to mitigate lung injury. Figure S6 Bilateral subdiaphragmatic vagotomy aggravates LPS-induced lung injury. Table S1 The demographic data of the septic ARDS patients included. Table S2 Primer pairs for RT-qPCR.