Introduction

Neutrophils are the primary circulating innate immune cells and play crucial roles in the inflammatory response (1, 2). Following infection or tissue injury that increases vascular permeability, neutrophils rapidly extravasate into affected tissues, where they initiate inflammation by secreting lipid mediators, reactive oxygen species, chemokines, cytokines, and proteases. These events drive and amplify the early stages of inflammation, facilitating the removal of infectious or noxious agents and allowing subsequent mechanisms of tissue repair and the restoration of homeostasis to ensue (3). Neutrophils mediate phagocytosis, release reactive oxygen species (ROS), defensins, and lytic enzymes, and form neutrophil extracellular traps (NETs), all of which are critical for inflammation (4). While these mechanisms are essential for host defense against pathogens, they also contribute to sterile inflammation and, in some contexts, to carcinogenesis and tumor progression (5), unregulated neutrophil-derived mechanisms can also lead to collateral tissue damage during sterile inflammation (6, 7) and to atherothrombotic and autoimmune diseases (8).

Beyond their well-established roles in initiating inflammation, neutrophils are increasingly recognized for their contributions to the resolution phase of inflammation, a mechanistically programmed event necessary for the return to physiological homeostasis (9, 10). In this context, neutrophil elaboration of specialized pro-resolving lipid mediators (SPMs) and their responses have been characterized in numerous models of inflammation resolution and host defense (11–14). Known SPMs are largely derived from the oxidative metabolism of omega-3 polyunsaturated fatty acids, namely docosahexaenoic acid, eicosapentaenoic acid, and linolenic acid. These SPMs are thought to interact with GPCRs on cells in a stereospecific manner to regulate processes critical for the resolution of inflammation, such as phagocytosis and the production of anti-inflammatory factors. Whether other lipid mediators derived from the metabolism of membrane phospholipids, particularly those signaling through GPCRs, regulate neutrophil resolution programs remains unclear.

Recent studies have shown that neutrophils display marked heterogeneity, with distinct phenotypes defined by transcriptional, functional, and developmental programs (15–20). Recent work has further refined neutrophil classification frameworks, highlighting functional diversity across developmental and tissue-associated states (21). These subsets participate in a multitude of physiological processes, including inflammation resolution (22, 23), tissue remodeling (24, 25), angiogenesis (26, 27), and regulation of hematopoiesis (3, 28). Despite this functional diversity, the upstream signals that control neutrophil phenotypic plasticity and constrain inflammatory potential remain poorly understood.

Although neutrophil development is classically viewed as a bone marrow–restricted process, accumulating evidence indicates that the spleen serves as an important peripheral reservoir and regulatory niche for neutrophils (29, 30). Splenic neutrophils differ from bone marrow counterparts in maturation state, trafficking behavior, and functional programming, and can contribute to both host defense and the resolution of inflammation. Extramedullary myelopoiesis in the spleen can also occur under homeostatic and inflammatory conditions, providing a site for ongoing neutrophil differentiation and functional tuning. Moreover, neutrophils appear to undergo continued maturation and phenotypic adaptation after egress from the bone marrow, influenced by tissue-specific cues within secondary lymphoid organs such as the spleen (29, 31–33). These splenic microenvironments expose neutrophils to distinct cytokine, chemokine, and lipid mediator landscapes that shape their survival, trafficking, and inflammatory potential. However, specific lipid-sensing pathways that mediate these tissue-specific adaptations remain largely undefined, and differential expression of trafficking receptors and signaling molecules may reflect functional specialization rather than a simple developmental stage.

Sphingosine-1-phosphate (S1P) is a bioactive lipid mediator generated during membrane sphingolipid metabolism. It signals through five G protein-coupled receptors (GPCRs), now known as S1PR1-5 (34, 35), and plays crucial roles in lymphocyte trafficking, vascular development, and immune homeostasis (36, 37). S1P is best known for regulating lymphocyte egress from the thymus and secondary lymphoid organs into the circulation (38). S1PR1 functional antagonists that preferentially activate β-arrestin have been approved for the treatment of autoimmune diseases, underscoring the importance of adaptive immune cell S1PR1 signaling in autoimmune pathology (37, 39). More recently, S1P has been shown to influence T cell metabolism (40) and differentiation fate, including the balance between Th17 and regulatory T cells (41–43). However, the role of S1P receptors in myeloid lineage cells, particularly neutrophils, remains incompletely understood. Prior studies suggest that S1P can influence neutrophil functions, such as pain sensitization (hyperalgesia) (44) and fungus-induced vasculitis (45), while genetic disruption of S1P synthesis by knocking out sphingosine kinases does not markedly impair neutrophil migration in acute inflammatory models (46). Elevated systemic S1P levels in S1P lyase-deficient mice result in neutrophilia that is reversed by deletion of S1PR4, implicating receptor-specific S1P signaling in neutrophil biology (47). A rare missense mutation in S1PR4 has also been linked to human neutropenia (48). Although these findings support a role for S1P receptors in neutrophil regulation, they do not directly address the contribution of S1PR1 to neutrophil homeostasis or functional state. Notably, apolipoprotein M (ApoM)-bound S1P suppresses NETosis via activation of S1PR1 and S1PR4, suggesting receptor-specific anti-inflammatory functions (49). However, whether S1PR1 signaling is critical for neutrophil homeostasis, functions (inflammation, resolution), and heterogeneity remains unclear. Specifically, the direct contribution of S1PR1, a therapeutic target in many autoimmune diseases with five FDA-approved drugs in the clinic, to neutrophil trafficking, survival, phenotype, or function under inflammatory or infectious conditions remains unknown. Given the clinical use of S1PR modulators, understanding their effects on innate immunity is warranted.

In this study, we used myeloid-and neutrophil-restricted S1PR1 gain-and loss-of-function mouse models to define the role of S1PR1 signaling in neutrophil biology. We show that sustained S1PR1 signaling reprograms neutrophils into a long-lived, metabolically supported state characterized by reduced oxidative and inflammatory responses and altered surface marker profiles consistent with a distinct neutrophil state. Functionally, S1PR1^hi neutrophils retain intrinsic antibacterial activity under simplified ex vivo conditions, but exhibit impaired bacterial clearance in vivo, particularly in the lung, while also conferring enhanced protection against viral infection-induced lung injury. Together, our findings identify S1PR1 as a key regulator of neutrophil state that tunes the balance between host defense and inflammatory tissue damage in a context-dependent manner. These results may have therapeutic implications for infectious, autoimmune, and inflammatory diseases.

Results

Myeloid S1PR1 expands systemic neutrophil compartments under steady-state conditions

We generated two mouse models of S1PR1 loss-and gain-of-function: LysM-Cre; S1pr1^f/f(LysM-S1pr1 KO) and LysM-Cre; S1pr1^{flox-stop-flox}(LysM-S1pr1 TG). Both mouse strains were viable and appeared phenotypically normal. Complete blood count analysis revealed that LysM-S1pr1 TG mice had elevated numbers of circulating white blood cells, neutrophils, and monocytes, with no significant changes in other hematopoietic lineages (Fig. 1, A–D). In contrast, LysM-S1pr1 KO mice showed no significant changes in circulating neutrophil or monocyte numbers but displayed reduced lymphocyte counts and a modest decrease in total leukocyte counts (Fig. 1A–D). These data suggest that S1PR1 is not required for neutrophil and monocyte numbers in the circulation, but increased S1PR1 expression in the myeloid lineage is associated with expansion of the neutrophil and monocyte compartments in the blood. Reduced lymphocytes in LysM-S1pr1 KO mice could reflect recombination occurring in early hematopoietic progenitors that transiently express Lyz2, rather than being restricted exclusively to mature myeloid cells (50, 51). LysM-S1pr1 TG mice exhibited reduced red blood cell (RBC) count, hemoglobin levels, and hematocrit, along with increased mean corpuscular volume (MCV) and mean corpuscular hemoglobin (MCH) (Fig. S1A–E), indicating that the TG mice had mild macrocytic normochromic anemia. Wild-type neutrophils from bone marrow and the spleen expressed S1pr1 mRNA, and as expected, LysM-S1pr1 TG neutrophils showed increased S1pr1 mRNA expression (Fig. S1F). S1pr1 mRNA levels were higher in splenic vs. BM neutrophils in both LysM-S1pr1 TG and control mice, suggesting that S1PR1 expression is increased in peripheral neutrophils, which are more mature. We observed a modest increase in S1pr4 transcript levels in neutrophils from S1pr1 TG mice; however, the magnitude of this change was minor and more variable than the upregulation of S1pr1 (Fig. S1G).

Fig. 1.

Given the elevated blood neutrophil counts in LysM-S1pr1 TG mice, we next examined neutrophil morphology and distribution in hematopoietic organs. Circulating neutrophil morphology was unremarkable in both TG and KO mice (Fig. 1E). However, LysM-S1pr1 TG mice exhibited marked splenomegaly, with spleens ∼3× larger than those of littermate controls (Fig. 1F). Flow cytometry revealed a >9-fold increase in splenic neutrophils (CD45⁺CD11b⁺Ly6G⁺) (Fig. 1G). In contrast, LysM-S1pr1 KO mice showed no changes in spleen size or splenic neutrophil numbers (data not shown). To assess whether the splenomegaly phenotype is sensitive to pharmacologic S1PR inhibition, we treated LysM-S1pr1 TG mice with FTY720, a functional antagonist of S1PR1, for 2 weeks. Spleen size returned to normal littermate control levels after FTY720 treatment (Fig. 1H), and H&E staining showed restoration of normal splenic architecture (Fig. 1I), indicating that the splenomegaly is induced by neutrophil S1PR1^hi states. In addition, resident neutrophils in the lung and liver were elevated in LysM-S1pr1 TG mice (Fig. 1, J and K) without detectable signs of inflammation or damage to the lung or liver, such as high protein content in bronchoalveolar lavage fluid (BALF) or elevated alanine aminotransferase (ALT) enzyme activity in plasma (Fig. 1, L and M), respectively. Conversely, bone marrow neutrophils were significantly reduced in LysM-S1pr1 TG mice (Fig. 1N), suggesting increased homeostatic mobilization of neutrophils into the S1P-rich circulatory compartment.

To determine whether the consequence of S1PR1 overexpression is neutrophil-intrinsic, we used a neutrophil-selective Mrp8-Cre driver to overexpress S1PR1 (Mrp8-S1pr1 TG mice). Similar to LysM-Cre counterparts, Mrp8-S1pr1 TG mice displayed increased circulating neutrophil counts, enlarged spleens, elevated spleen/body weight ratio and splenic neutrophil frequency, and reduced bone marrow neutrophil percentage and count (Fig. S2), supporting a cell-intrinsic role for S1PR1 signaling. These findings suggest that S1PR1^hi neutrophils show increased egress from the bone marrow, circulate longer, and reside in peripheral organs without causing organ damage under homeostatic conditions. In contrast, lack of S1PR1 in neutrophils did not affect steady-state granulopoiesis, circulating neutrophil numbers, or tissue residence, suggesting that this pathway is not required for general neutrophil egress from the bone marrow, survival, and functions in vivo.

Sustained S1PR1 signaling alters neutrophil phenotypes

Flow cytometry showed that surface S1PR1 levels were low in both TG and control splenic neutrophils, whereas total S1PR1 signal was markedly higher in LysM-S1pr1 TG neutrophils (Fig. 2A). ImageStream cytometry analysis showed that S1PR1 was predominantly intracellular, a pattern consistent with ligand-induced S1PR1 endocytosis (Fig. 2B). Given that sustained GPCR activation is commonly associated with receptor endocytosis, these data suggest ligand-induced S1PR1 internalization and sustained receptor engagement in TG neutrophils.

Fig. 2.

Giemsa staining of splenic neutrophils revealed an increased frequency of neutrophils with band-shaped nuclei and reduced segmentation in LysM-S1pr1 TG mice (Fig. 2, C and D), consistent with phenotypes commonly associated with less mature neutrophil states (52–54). However, no apparent change was observed in the bone marrow, which contains newly formed neutrophils. Additionally, surface CXCR2 was downregulated and CXCR4 upregulated in LysM-S1pr1 TG neutrophils across bone marrow, spleen, and blood (Fig. 2, E and F), a pattern previously associated with altered neutrophil trafficking and post–bone marrow adaptation. To further characterize the phenotypic state of S1PR1^hi neutrophils, we assessed surface markers associated with neutrophil maturation and activation. CD62L expression was largely preserved across compartments, with only a modest reduction in bone marrow neutrophils (Fig. 2G), suggesting that S1PR1^hi neutrophils do not exhibit classical activation-associated CD62L shedding. In contrast, CD101 expression was markedly reduced, particularly in bone marrow and circulating neutrophils (Fig. 2H), indicating a shift toward a distinct neutrophil state. Together, these data indicate that S1PR1^hi neutrophils display altered phenotypic features consistent with modified maturation and trafficking states.

To determine whether S1PR1 overexpression influences myelopoiesis and/or granulopoiesis, we profiled myeloid subsets in the spleen and bone marrow by flow cytometry. Megakaryocyte/erythrocyte progenitors (MEPs) were significantly increased in the spleen (Fig. S3), consistent with compensatory erythroid responses. Common myeloid progenitors (CMPs) were mildly elevated, whereas granulocyte myeloid progenitors (GMPs) remained unchanged. Bone marrow progenitor frequencies were slightly elevated across all subsets. Together, these data indicate that increased S1PR1 signaling in neutrophils regulates its phenotype, characterized by CXCR4^hi, CXCR2^lo, with increased circulatory transit time and residency in several anatomical compartments, relatively late in its developmental trajectory in the bone marrow, i.e., during granulopoiesis.

S1PR1^hi neutrophils are long-lived and exhibit enhanced mitochondrial fitness

To understand the cellular basis for S1PR1^hi neutrophil accumulation in vivo in the circulation and organs, we examined cell survival and death parameters. During sterile peritonitis, neutrophil numbers at 12 h, but not 4 h, were significantly higher in TG mice, consistent with increased neutrophil survival at later time points in this model (Fig. 3A). Splenic neutrophils exhibited a reduced frequency of Annexin V+7-AAD+ cells, indicating reduced late apoptosis/necrosis. In contrast, early apoptotic neutrophils (Annexin V⁺/7-AAD⁻) were unchanged (Fig. 3B). LDH release assays confirmed reduced cell death in bone marrow neutrophils after 24 h of ex vivo culture (Fig. 3C). To evaluate in vivo survival, we co-injected fluorescently labeled TG and control neutrophils into the peritoneum of WT mice. TG neutrophils persisted at higher levels at 24 and 48 h post-injection (Fig. 3D), indicating enhanced survival of TG neutrophils in vivo. Together, these data support reduced neutrophil turnover and death of S1PR1^hi neutrophils in vivo, which likely contribute to their accumulation in circulation and peripheral organs.

Fig. 3.

Since S1PR1-mediated naïve T cell survival has been linked to metabolic reprogramming (55), which involves mitochondrial content and function (40), we examined whether mitochondrial alterations in LysM-S1pr1 TG neutrophils accompany increased persistence. Metabolic phenotyping by Seahorse analysis revealed an elevated oxygen consumption rate (OCR) in TG neutrophils under mitochondrial stress, suggesting increased oxidative phosphorylation (Fig. 3E). MitoTracker staining showed higher mitochondrial membrane potential with unchanged mass (Fig. 3F). Confocal imaging of Tom20-stained S1PR1^hi neutrophils revealed larger but fewer mitochondrial puncta, indicating reduced fragmentation and/or mitophagy (Fig. 3, G and H). Consistent with enhanced metabolic signaling, S1pr1-overexpressing BM neutrophils exhibited increased phosphorylation of mTOR downstream targets, including S6 and 4E-BP1, which was sensitive to rapamycin treatment (Fig. S4). Together, these results indicate that S1PR1 signaling enhances mitochondrial integrity and oxidative metabolism, thereby supporting metabolic fitness that promotes reduced cell death.

S1PR1^hi neutrophils display transcriptional programs associated with persistence, metabolism, and reduced inflammatory signaling

Recently, single-cell transcriptomic analysis has provided a more granular understanding of neutrophil differentiation pathways that yield heterogeneous, phenotypically distinct populations (16, 17, 56, 57). To understand the transcriptional programs of S1PR1^hi neutrophils, we performed single-cell RNA sequencing (scRNA-seq) on peritoneal cells collected 10 hours after thioglycollate elicitation. Uniform manifold approximation and projection (UMAP) analysis identified major immune cell populations, including neutrophils, monocytes/macrophages, resident macrophages, lymphocytes (T and B), dendritic cells, mast cells, and NK cells (Fig. 4A). A representative dot plot summarizes the expression of canonical marker genes used for cluster annotation, with dot size indicating the proportion of cells expressing each gene and color intensity reflecting the average normalized expression level (Fig. 4B). Within the neutrophil compartment, UMAP projection revealed three transcriptionally distinct subsets, G5a, G5b, and G5c, as previously described (16) (Fig. 4, C and D). As reported, G5a cells abundantly expressed genes related to neutrophil migration and inflammatory responses; G5b neutrophils expressed a set of interferon-stimulated genes (ISGs); and G5c neutrophils showed high expression of genes previously associated with aging and resolution-associated states (16). We found that LysM-S1pr1 TG mice displayed enrichment of G5c cells, consistent with enrichment of the G5c transcriptional state (Fig. 4E). These data provide a transcriptomic characterization of S1PR1^hi neutrophils.

Fig. 4.

To further validate the subcluster switch, we performed flow cytometry to distinguish splenic neutrophil populations G5a (CXCR4^lo IFIT1⁻), G5b (IFIT1⁺), and G5c (CXCR4^hi IFIT1⁻) (Fig. 4F). The results clearly documented an increase in G5c neutrophils in the spleen. These data suggested that the S1PR1^hi neutrophil subset resembles the G5c subset and exhibits a distinct gene expression program that correlates with increased neutrophil persistence observed in vivo.

The scRNAseq dataset was further analyzed using module scores with Hallmark gene sets (apoptosis) and GO/Reactome gene sets (autophagy and mitophagy). Reduced transcriptional programs for apoptosis and autophagy were observed in LysM-S1pr1 TG neutrophils compared with control neutrophils (Fig. 4, G–I). The mitochondrial biogenesis module score per neutrophil showed distributions near zero in both groups, with a slight upward shift in LysM-S1pr1 TG neutrophils (Fig. 4J). These results indicate that S1PR1^hi neutrophils exhibit transcriptional features consistent with reduced cell death signaling and suppressed mitochondrial turnover, aligning with the functional properties described above.

In addition, we noted that inflammation-and ROS-related module scores (inflammatory response, TNF_α signaling via NF_κB, NADPH oxidative stress, reactive oxygen species, neutrophil degranulation) were reduced in LysM-S1pr1 TG neutrophils (Fig. 4, K–O), suggesting a dampened inflammatory/oxidative state. Detailed analysis of transcript expression showed reduced Cxcr2 expression in TG neutrophils (Fig. S5A), consistent with flow cytometry data (Fig. 3E). In addition, expression of inflammation-associated genes (Il1b, Casp4, Ptgs2, Nfkbia, Fos, Junb, Jund, Nfkbiz) and cell death-associated genes (Map1lc3b, Atg3, Stk17b, Ubb, Rb1cc1, Sqstm1) was downregulated in LysM-S1pr1 TG neutrophils (Fig. S5, B and C). Together, these findings indicate that S1PR1^hi neutrophils adopt a transcriptional state associated with reduced inflammatory and oxidative signaling.

S1PR1^hi neutrophils retain phagocytic capacity but exhibit reduced ROS production

Phagocytosis and production of reactive oxygen species (ROS) are key effector functions of neutrophils that are essential for inflammatory, host defense, and resolution responses (58). Phagocytosis of pHrodo-E. coli particles was similar between LysM-S1pr1 TG and control thioglycollate-elicited peritoneal neutrophils (Fig. 5A), indicating that enhanced S1PR1 signaling does not impair neutrophil phagocytic capacity.

Fig. 5.

We next examined whether S1PR1 signaling affects neutrophil oxidative responses. Upon stimulation with the N-formyl peptide fMLP, activation of the NADPH oxidase complex was significantly reduced in LysM-S1pr1 TG neutrophils, whereas responses in WT and S1PR1-deficient neutrophils were preserved (Fig. 5B). Consistent with this, intracellular ROS levels, as measured by DCFDA staining, were significantly lower in both splenic and bone marrow neutrophils from TG mice (Fig. 5, C and D). Transcriptional profiling further supported these findings, revealing reduced expression of NADPH oxidase–related genes (e.g., Rac2, Ncf2, and Ncf4), increased expression of antioxidant genes (e.g., Prdx5 and Prdx6), and elevated Nampt, a key regulator of NAD+ biosynthesis (Fig. S6). Together, these data indicate that S1PR1 overexpression suppresses neutrophil oxidative responses. Given the central role of ROS in antimicrobial defense, we next assessed whether reduced oxidative activity affects bacterial killing. In ex vivo assays, extracellular bacterial killing was comparable between TG and control neutrophils. TG neutrophils exhibited a modest increase in intracellular bacterial burden at early time points, suggesting a potential delay in intracellular bacterial handling, although this difference was not statistically significant (Fig. S7). Overall, these findings indicate that reduced ROS production does not result in a major intrinsic defect in antibacterial activity under simplified ex vivo conditions.

To determine the functional consequences of reduced oxidative responses in vivo, we performed bacterial challenge experiments. After intraperitoneal (IP) bacterial challenge, LysM-S1pr1 TG mice exhibited a higher bacterial burden in the lungs than controls (Fig. 5E). Although bacterial loads were assessed across multiple compartments, no consistent differences were observed outside the lung (Fig. S8), indicating that the defect in bacterial clearance is most prominent in this tissue. To further evaluate local antibacterial responses in the airway, mice were subjected to intratracheal (IT) bacterial challenge. TG mice displayed a significantly higher bacterial load in bronchoalveolar lavage fluid (BALF) (Fig. 5F). This was accompanied by a lower proportion of neutrophils (CD11b⁺Ly6G⁺) among CD45⁺ cells in BALF (Fig. 5G), suggesting impaired neutrophil accumulation in the airway compartment. Together, these findings indicate that reduced ROS production in S1PR1^hi neutrophils is associated with impaired bacterial clearance in vivo, with a predominant effect in the lung.

S1pr1 TG mice exhibit improved survival and attenuated lung injury following influenza viral infection

Neutrophil-driven inflammatory responses are crucial for viral clearance, the induction of virus-specific adaptive immunity, and the subsequent resolution and regeneration that restore homeostasis. To assess the response of myeloid S1PR1^hi mice to influenza infection, mice were infected intranasally with an LD₅₀ dose of H1N1 virus and monitored for 21 days post-infection (dpi) (Fig. 6A). Approximately 40% of control mice died by 11 dpi, whereas only 5% of LysM-S1pr1 TG mice died (Fig. 6B). Weight loss in LysM-S1pr1 TG mice was significantly reduced compared with control mice (Fig. 6C). In addition, LysM-S1pr1 TG mice exhibited better circulatory O₂ saturation (SpO₂) at 9 and 11 dpi (Fig. 6D), suggesting improved preservation of pulmonary function after infection. In a sublethal model (0.16 LD50) (Fig. 6E), LysM-S1pr1 TG mice had fewer infiltrating cells in the BALF, including neutrophils and macrophages, at 3 dpi (Fig. 6, F and H). Lung viral load, as measured by nucleoprotein (NP), polymerase acidic protein (PA), and hemagglutinin (HA) gene expression, was lower in TG mice at 3 and 7 dpi (Fig. 6, I– K). BALF protein leakage and IL-6 levels were reduced at 7 dpi (Fig. 6, L and M), and IL-10 levels peaked earlier in TG mice (Fig. 6N), consistent with decreased inflammation and accelerated resolution. To determine whether these protective effects were neutrophil-intrinsic, we next evaluated influenza outcomes in Mrp8-S1pr1 transgenic mice. Mrp8-S1pr1 TG mice exhibited significantly improved survival, reduced weight loss, and better preservation of arterial oxygen saturation following H1N1 infection (Fig. 7, A-C). Mrp8-S1pr1 TG mice phenocopied key protective outcomes observed in LysM-S1pr1 TG mice, indicating that neutrophil-restricted S1PR1 overexpression is sufficient to confer protection in this model of viral pneumonia.

Fig. 6.

Fig. 7.

Discussion

Although S1PR1 has been extensively studied in lymphocytes and endothelial cells, its role in neutrophils remains poorly understood. Our study provides in vivo evidence that S1PR1 regulates neutrophil distribution, persistence, mitochondrial fitness, inflammatory phenotype, and host defense against bacteria and viruses. These effects reshape host responses in a context-dependent manner, impairing bacterial clearance while promoting protection during viral lung injury.

In this study, we used LysM-S1pr1 KO and TG mice to define how altered S1PR1 signaling affects neutrophil biology. Unlike the TG model, LysM-S1pr1 KO mice exhibited no discernible hematological or tissue phenotype, suggesting that S1PR1 is dispensable for neutrophil development and homeostatic functions under the steady-state conditions tested here. Although LysM-S1pr1 KO mice showed a significant reduction in circulating lymphocyte counts, neutrophil and monocyte numbers were not significantly altered (Fig. 1A–D). In contrast, LysM-S1pr1 TG mice displayed increased numbers of circulating and tissue-resident neutrophils, accompanied by a reduction in bone marrow neutrophils (Fig. 1). Given the established role of S1PR1 in sensing S1P gradients and regulating immune cell egress (59, 60), our findings suggest that increased S1PR1 signaling stimulates neutrophil egress from the bone marrow to peripheral tissues, including the spleen, lung, and liver. Importantly, despite increased neutrophil accumulation in peripheral tissues, S1PR1^hi neutrophils did not provoke overt tissue injury under homeostatic conditions.

In addition to the neutrophil phenotypes described above, LysM-S1pr1 TG mice displayed mild macrocytic anemia, splenomegaly, and increased splenic MEP populations (Fig. S1, A-E and S3B), consistent with compensatory erythropoietic responses. FTY720 treatment normalized splenic enlargement in TG mice, supporting S1PR1–dependent splenomegaly. Notably, the anemic phenotype was not recapitulated in Mrp8-S1pr1 TG mice (Fig. S2H). Nevertheless, Mrp8-S1pr1 TG mice exhibited improved survival, reduced weight loss, and preserved oxygen saturation during influenza infection (Fig. 7), indicating that the protective phenotypes conferred by S1PR1 overexpression can occur independently of altered erythropoiesis.

S1PR1 is known to mediate pro-survival signaling via Gi-coupled pathways (61). Consistent with this, S1PR1-overexpressing neutrophils exhibited improved survival both ex vivo and in vivo, along with increased mitochondrial membrane potential and oxidative phosphorylation, indicative of improved mitochondrial metabolic fitness. These findings suggest that enhanced mitochondrial function supports neutrophil metabolic fitness and persistence, thereby contributing to their accumulation in LysM-S1pr1 TG mice. Notably, S1PR1 signaling preserved mitochondrial integrity in neutrophils by reducing fragmentation and maintaining high membrane potential (Fig. 3). This observation contrasts with the well-characterized glycolytic bias of neutrophils (62–65), suggesting that S1PR1 signaling supports mitochondrial maintenance and delays mitochondrial turnover. Consistent with Gi-coupled signaling, S1PR1 may engage PI3K-mTOR pathways to support mitochondrial integrity and metabolic competence. Reduced NADPH oxidase activity and ROS output, potentially downstream of altered metabolic programming, may enhance neutrophil persistence while limiting collateral tissue damage during inflammation. Supporting this model, S1pr1-overexpressing neutrophils showed increased phosphorylation of mTOR downstream targets (Fig. S4), consistent with activation of metabolic signaling pathways linked to mitochondrial integrity and cell survival. Whether reduced ROS production arises directly from S1PR1-mediated signaling or secondarily from altered metabolic programming remains to be determined. While a modest alteration in S1pr4 expression was observed in S1pr1 transgenic neutrophils (Fig. S1G), this alteration was small and variable relative to S1pr1 overexpression and is unlikely to be a primary driver of the observed phenotypes.

Flow cytometry and morphological analysis revealed enrichment of CXCR4^hiCXCR2^lo neutrophils with band-shaped nuclei in the spleen and blood of LysM-S1pr1 TG mice (Fig. 2), consistent with altered peripheral neutrophil states and receptor remodeling associated with trafficking. Reduced CD101 expression further supports a shift toward a non-classical neutrophil state, potentially linked to altered maturation or activation trajectories. In parallel, single-cell RNA sequencing of peritoneal neutrophils revealed an expansion of the G5c subpopulation (Fig. 4), which has been associated with neutrophil aging and reverse migration (16). In this context, our data suggest that S1PR1 signaling reprograms neutrophils toward a distinct phenotypic state characterized by enhanced persistence and reduced inflammatory output. Although we observed reduced apoptotic markers and increased neutrophil persistence in select tissues, these findings do not directly establish prolonged neutrophil lifespan but rather support a shift toward a persistence-associated neutrophil state. In addition to intrinsic state reprogramming, altered neutrophil trafficking may also contribute to these phenotypes. Neutrophils can exit tissues via lymphatic vessels and re-enter secondary lymphoid organs, a process influenced by S1P receptor signaling (66). In this framework, S1PR1 overexpression may influence the balance between tissue retention, lymphatic egress, and recirculation, thereby shaping neutrophil distribution across peripheral compartments. Consistent with this model, LysM-S1pr1 TG mice exhibited increased neutrophil accumulation in lymphoid organs and enrichment of transcriptional features associated with aging and reverse migration. Moreover, during influenza infection, the combination of increased lung tissue neutrophils and reduced BALF neutrophils suggests altered compartmental trafficking rather than enhanced recruitment per se. Together, these findings support a model in which S1PR1 signaling reshapes neutrophil distribution and functional state through reprogramming of neutrophil phenotype, with additional contributions from altered trafficking dynamics.

Functionally, LysM-S1pr1 TG neutrophils exhibited reduced ROS production, accompanied by downregulation of NADPH oxidase-related genes and increased expression of antioxidant programs (Fig. 5 and Fig. S6). Although phagocytosis was preserved and antibacterial activity remained largely intact under simplified ex vivo conditions (Fig. S7), these findings suggest there is no major intrinsic defect in bacterial killing capacity. In contrast, in vivo bacterial challenge revealed impaired bacterial clearance, particularly in the lung (Fig. 5E and 5F), indicating that functional defects emerge under physiological conditions. Together, these findings indicate that S1PR1 signaling dampens oxidative responses without compromising intrinsic antibacterial capacity, but instead reshapes host defense in a context-dependent manner. Reduced oxidative output may limit collateral tissue damage during inflammation, consistent with protection in viral lung injury. At the same time, impaired bacterial clearance highlights a functional trade-off associated with this reprogrammed neutrophil state, in which reduced inflammatory damage may come at the expense of optimal antibacterial defense. Future studies will be required to elucidate how S1PR1-driven neutrophil reprogramming influences host defense across diverse infectious contexts. Given the clinical use of S1PR1 modulators in autoimmune diseases, our findings suggest a potential avenue for pharmacologically tuning neutrophil function to balance antimicrobial defense and inflammatory tissue damage.

During H1N1 infection, both LysM-S1pr1 TG and Mrp8-S1pr1 TG mice showed improved survival, reduced weight loss, and preserved oxygen saturation (Fig. 6 and Fig. 7). Despite increased neutrophil accumulation in lung tissue, BALF neutrophils were reduced (Fig. 6), suggesting altered compartmental trafficking and reduced transmigration across the alveolar barrier rather than increased recruitment per se. Reduced viral burden, together with decreased inflammatory injury markers (BALF protein and IL-6) and altered IL-10 dynamics, supports improved resolution of lung inflammation. Together, these findings indicate that S1PR1-dependent neutrophil reprogramming promotes tissue protection by limiting inflammatory damage while preserving host control of viral infection.

Our results support a model in which elevated S1PR1 signaling reprograms neutrophils into a long-lived, low-inflammatory state marked by reduced ROS production, enhanced metabolic fitness, and altered transcriptional programs. An apparent paradox is the expansion of the G5c neutrophil cluster in S1pr1 TG mice, previously associated with aging and apoptotic features (16). One possible explanation is that this transcriptional state reflects a late or post-maturation state rather than an irreversible commitment to cell death. In this context, S1PR1 signaling may decouple features of neutrophil aging, such as altered trafficking receptor expression and transcriptional reprogramming, from apoptotic execution, thereby promoting persistence without triggering cell death. This functional reprogramming may contribute to reduced lung damage and improved resolution during viral infection. The ability of S1PR1 to bias neutrophils toward a less-inflammatory, persistence-associated state may have therapeutic implications for inflammatory and infectious diseases. In settings of excessive neutrophil-driven inflammation, enhancing this axis may promote tissue protection and resolution. Conversely, in contexts where prolonged neutrophil persistence is detrimental, such as impaired pathogen clearance, overactivation of this pathway may be disadvantageous. Thus, the S1PR1–neutrophil axis represents a context-dependent regulatory pathway with both protective and potentially detrimental effects. Future work will be required to define how reduced ROS production influences pathogen control across different contexts. Importantly, multiple lines of evidence support a direct, cell-intrinsic role for S1PR1 in regulating neutrophil survival, metabolism, and inflammatory output.

One limitation of this study is the use of LysM-Cre, which targets multiple myeloid populations, including lymphocytes, monocytes, and macrophages. However, the core mechanistic and functional conclusions are supported by experiments in purified neutrophils, including analyses of survival, metabolic fitness, mitochondrial integrity, ROS production, and antibacterial activity, thereby reducing potential confounding effects from other cell types. Consistent with a neutrophil-intrinsic role, key phenotypes were recapitulated in neutrophil-restricted Mrp8-S1pr1 TG mice, which exhibited improved survival, reduced weight loss, and preserved oxygen saturation during influenza infection (Fig. 7). Together, these findings support a dominant role for neutrophil-intrinsic S1PR1 signaling in the observed phenotypes.

In conclusion, our study demonstrates that neutrophil-intrinsic S1PR1 reprograms neutrophils into a persistent, metabolically fit, and low-inflammatory state. This reprogramming promotes peripheral neutrophil accumulation and reduces oxidative and inflammatory responses, thereby reshaping neutrophil function in vivo. Functionally, S1PR1^hi neutrophils exhibit context-dependent effects on host defense, with impaired bacterial clearance but enhanced protection against viral lung injury. Together, these findings support a model in which S1PR1 serves as a key regulatory axis that tunes neutrophil persistence and inflammatory output in a context-dependent manner, balancing antimicrobial defense and inflammatory tissue damage. These findings position S1PR1 as a molecular switch that uncouples neutrophil persistence from inflammatory output. These insights highlight S1PR1 as a potential target for modulating neutrophil function in infectious and inflammatory diseases.

Materials and methods

Animals

All mice were on a C57BL/6J background and used after 12 weeks of age. Both the LysM-S1pr1 transgenic (TG) and knockout (KO) mouse lines were generated using the same myeloid-specific Cre driver line. This Cre line is commonly referred to as LysM-Cre (also known as Lyz2-Cre), in which Cre recombinase is knocked into and expressed from the endogenous lysozyme M (Lyz2) locus. LysM-S1pr1 TG mice (S1pr1^{flox/stop/flox}; LysM-Cre) were generated by crossing mice carrying a loxP-flanked transcriptional stop cassette upstream of the S1pr1 transgene (fsf allele) with the LysM/Lyz2-Cre driver. Cre-mediated excision of the stop cassette results in enforced S1pr1 expression in LysM-expressing myeloid cells, as previously described (67). LysM-S1pr1 KO mice (S1pr1^flox/flox; LysM-Cre) were generated by crossing S1pr1^flox/floxmice with the same LysM/Lyz2-Cre driver line, resulting in myeloid-restricted deletion of the endogenous S1pr1 allele. To assess neutrophil-intrinsic effects of S1PR1 signaling independently of broader myeloid populations, we generated Mrp8-S1pr1 TG mice using the Mrp8-Cre (S100a8-Cre) driver line. The Mrp8-Cre transgene drives Cre recombinase expression predominantly in mature neutrophils, with minimal recombination reported in monocytes or macrophages under steady-state conditions, as previously described (68–70). Mrp8-S1pr1 TG mice were generated by crossing Mrp8-Cre mice with S1pr1^{flox/stop/flox} mice, resulting in enforced S1pr1 expression preferentially in neutrophils. Unless otherwise indicated, Cre-negative littermates from the same breeding pairs were used as controls. All experimental procedures were approved by the Boston Children’s Hospital Institutional Animal Care and Use Committee.

FTY720 treatment

Mice (n ≥ 4) were treated intraperitoneally with 0.5 mg/kg (∼150 μL per animal) of FTY720 (Sigma-Aldrich, St. Louis, MO), a functional antagonist of S1P, or vehicle (2% DMSO) every other day for 2 weeks to assess the S1PR1 dependence of splenomegaly and neutrophil distribution.

Neutrophil isolation

Neutrophils were collected from bone marrow, spleen, or peritoneal fluid. To collect bone marrow cells, mice were euthanized with CO₂, and their femurs and tibias were dissected and maintained in RPMI-1640 medium (Thermo Fisher, Waltham, MA). Bone marrow was exposed by clipping the ends of the bones, then centrifuged at 600 × g for 5 minutes, and the cell suspension was collected. To collect splenic cells, the spleen was harvested, weighed, transferred into the gentleMACS C tube (Miltenyi Biotec, Waltham, MA) containing 5 mL of RPMI-1640 medium, and dissociated in the gentleMACS dissociator (Miltenyi Biotec). RBCs were lysed with ACK lysis buffer (Thermo Fisher), and cells were washed and passed through a 40-μm cell strainer. After washing, cells were resuspended in Hanks’ balanced salt solution (HBSS) containing 0.5% fatty acid-free bovine serum albumin (ff-BSA, Sigma-Aldrich) and 2 mM EDTA. For peritoneal cells, mice were injected intraperitoneally with 2 mL of a 3% (wt/vol) thioglycollate solution (Sigma-Aldrich). Peritoneal exudate cells were collected 4 hours after thioglycollate injection by collecting lavage with 10 mL of cold HBSS. Cells were hemolyzed with ACK lysis buffer, washed, and resuspended in HBSS containing 0.5% ff-BSA and 2 mM EDTA. Cells from bone marrow, spleen, and peritoneal lavage were then purified using the Neutrophil Isolation Kit (Miltenyi Biotec) or by density gradient centrifugation with Histopaque-1119/Ficoll-Paque (71). Histopaque-1119 and Ficoll-Paque Plus were purchased from Sigma-Aldrich and GE Healthcare Life Sciences (Marlborough, MA), respectively.

Giemsa stain

Cells in 200 μL of complete medium were loaded into an EZ Single Cytofunnel (Epredia, Kalamazoo, MI) and centrifuged at 800 rpm for 10 min to deposit onto a slide. Cells were fixed in methanol for 2 min, stained with Giemsa solution (Sigma-Aldrich) for 10 min, and rinsed with distilled water. After air-drying, cells were mounted with Neo-Mount (Sigma-Aldrich) and imaged by light microscopy. Neutrophil counts and segmented/band nuclei were quantified using ImageJ software (National Institutes of Health, Bethesda, MD).

Gene expression analysis by qRT-PCR

Total RNA was isolated from purified neutrophils using Trizol reagent (Thermo Fisher Scientific, Waltham, MA) and the Direct-zolTM RNA MicroPrep (Zymo Research, Irvine, CA). cDNA was synthesized by reverse transcription using the qScript XLT cDNA Supermix (Quantabio, Beverly, MA). RT-PCR was performed with the PerfecTA SYBR Green Fast Mix (Quantabio) on the Step One Plus PCR system (Applied Biosystems, Foster City, CA). Relative gene expression levels were calculated using the 2-ΔΔCt method. Primers: S1pr1, 5’-ATGGTGTCCACTAGCATCCC-3’ (forward) and 5’-CGATGTTCAACTTGCCTGTGTAG-3’ (reverse); Actb, 5’-AGCCATGTACGTAGCCATCC-3’ (forward) and 5’-CTCTCAGCTGTGGTGGTGAA-3’ (reverse).

Hematology profiling

Peripheral blood was collected from the retro-orbital cavity or the inferior vena cava and analyzed on a HEMAVET 950FS (Drew Scientific, Oxford, CT). Isolated BM and splenic neutrophils were attached to slides using a Shandon Cytospin 2 (Thermo Fisher). Blood smears or isolated neutrophils were stained with a Diff-Quick Kit (Thermo Fisher) and imaged on an Axioskop 2 mot Plus microscope (Carl Zeiss, Dublin, CA).

Flow cytometry

Cells (1 × 10⁶) from bone marrow, spleen, liver, lung, or peritoneal lavage were surface-stained with antibodies (listed in Table S1) and analyzed by flow cytometry on a FACS Calibur or Fortessa (Becton Dickinson, Franklin Lakes, NJ) or a Cytek Aurora/Northern Light (Bethesda, MD). After gating out dead cells, neutrophils were identified as CD45⁺/CD11b⁺/Ly6G⁺ cells. Progenitor populations were defined as previously reported (60). CLPs were gated as Lin⁻IL7Rα⁺ Flt3⁺ cKit⁺ Sca-1⁺. GMPs were gated as Lin⁻ IL7Rα⁻ cKit⁺ Sca-1⁻ CD34⁺ FcγRII/III^hi. MEPs were gated as Lin⁻ IL7Rα⁻ cKit⁺ CD34⁻ FcγRII/III^lo/-. CMPs were gated as Lin⁻ IL7Rα⁻ cKit⁺ CD34⁺ FcγRII/III^lo/-. For intracellular staining, surface-stained cells were fixed in 2% PFA for 10 min, washed, permeabilized in Intracellular Staining Permeabilization Wash Buffer (BioLegend, San Diego, CA), and then stained with antibodies (listed in Table S1). Fluorescence minus one (FMO) and single-stain controls were used to set gates.

Ex vivo neutrophil survival

Splenic neutrophils were incubated at 37°C in calcium-containing HBSS for 1h in a CO₂ incubator. After washing, cells were stained with Annexin-V and 7-AAD for 15 minutes on ice and analyzed by flow cytometry. Neutrophils were gated as CD11b⁺ Ly6G⁺ cells. Early apoptotic cells were Annexin V⁺7-AAD⁻; late apoptotic/necrotic cells were Annexin V⁺7-AAD⁺.

In vivo neutrophil survival

Isolated BM neutrophils from LysM-S1pr1 TG and control mice were stained with fluorescent dyes CFSE or CellTrace^TM Violet (1:1000 dilution) in HBSS for 20 min. After a brief wash with HBSS containing 0.5% ff-BSA, TG and control neutrophils were mixed 1:1 (∼1.4 ×10⁶ cells) and co-injected intraperitoneally into WT mice. Peritoneal cells were harvested at 0, 24, 48, and 72 hours and analyzed by flow cytometry.

Sterile inflammation

To induce sterile inflammation, mice were injected intraperitoneally with 2 mL of 3% (wt/vol) thioglycollate solution (72). Peritoneal lavage fluids were collected 4, 12, 18, 24, or 48 hours after injection. Lavages were centrifuged to separate cells and supernatant. Cells were washed and incubated with ACK lysis buffer to remove RBCs. After washing, cells were resuspended in HBSS, and cell numbers were counted using a hemocytometer. To determine neutrophil and macrophage numbers in the cell suspension, 1 × 10^⁶ cells were incubated with Fc-blocker, followed by antibodies, and analyzed by flow cytometry. Neutrophils were gated as CD11b⁺ Ly6G⁺, and macrophages were gated as CD11b⁺ Ly6G⁻ F4/80⁺.

ROS production

The detailed procedures (46) were followed. Briefly, peritoneal-derived neutrophils (5 × 10⁵ cells/well) were incubated in 100 μl of PBSG (PBS with glucose) containing 50 μM isoluminol and 5 U/mL horseradish peroxidase in a 96-well plate. Cell suspensions were then stimulated with 10 μM fMLP (N-formyl-Met-Leu-Phe, F3506, Sigma-Aldrich), and the plate was read on a SpectraMax M2 (Molecular Devices, San Jose, CA) over 5 minutes. The area under the curve was calculated using GraphPad Prism 7 (GraphPad Software, San Diego, CA). For DCFDA-based intracellular ROS measurement, neutrophils were stained with 10 μM DCFDA (Sigma-Aldrich) for 30 min at 37°C, washed, and analyzed by flow cytometry.

Phagocytosis assay

Peritoneal neutrophils (2 × 10⁵ cells) were incubated in HBSS with 100 μL of the pHrodo fluorescent bioparticle conjugate (1 mg/mL, Thermo Fisher) for 30-120 minutes at 37°C. Cells with pHrodo, placed on ice, served as the control groups. After incubation for the indicated time, 2 mL of ice-cold HBSS was added to stop phagocytosis. Cells were blocked with an Fc-blocker for 10 minutes, stained with PE-Ly6G and APC-F4/80 antibodies for 30 minutes, washed, and then stained with the nucleic acid dye 7-AAD for 15 minutes. Flow cytometry analysis was performed using a FACS Calibur. The neutrophil phagocytic index was quantified using the mean fluorescence intensity (MFI) of pHrodo in 7-AAD⁻Ly6G⁺ cells.

Neutrophil intracellular bacterial killing assay

Isolated bone marrow neutrophils were resuspended at 0.5 × 10⁶ cells/well in RPMI 1640 supplemented with 2% heat-inactivated FBS and incubated at 37 °C for 30 min before infection. E. coli (DH5α) was grown overnight in LB broth (Sigma-Aldrich), washed, and adjusted to an OD₆₀₀ of 0.5 (≈ 5 × 10⁸ CFU/mL). Bacteria were opsonized with 10% autologous mouse serum in PBS at 37 °C for 1 h, washed, and added to neutrophil cultures at a 1:10 (neutrophil:bacterium) ratio. After incubation for 0, 30, 60, and 120 min at 37 °C, cells were centrifuged at 500 × g for 5 min, and extracellular bacteria were removed by treatment with gentamicin (250 µg/mL, Sigma-Aldrich) for 20 min, followed by three PBS washes. The final wash supernatant was plated on LB agar to confirm elimination of extracellular bacteria. Neutrophil pellets were lysed in 1 mL of 0.1% Triton X-100 for 10 min at room temperature, vortexed briefly, and then serially diluted. Diluted samples were plated on LB agar for CFU enumeration after overnight incubation at 37 °C. Bacterial suspensions incubated without neutrophils served as growth controls.

Neutrophil extracellular bacterial killing assay

Purified neutrophils were resuspended at 0.5 × 10⁶ cells/well in RPMI 1640 supplemented with 2% heat-inactivated FBS and 10 µg/mL cytochalasin D (Sigma-Aldrich) to inhibit phagocytosis. Cells were incubated at 37 °C for 30 min before bacterial challenge. E. coli (DH5α) was cultured overnight in LB broth, washed, and adjusted to an optical density at 600 nm (OD₆₀₀) of 0.5, corresponding to approximately 5 × 10⁸ CFU/mL. Neutrophils were then incubated with S. aureus at a 1:2 (neutrophil:bacterium) ratio for 0, 30, 60, and 120 min at 37 °C. At each time point, culture supernatants were serially diluted in PBS and plated onto LB agar plates, which were incubated overnight at 37 °C for colony enumeration. Bacterial suspensions incubated in medium without neutrophils served as controls to assess bacterial growth in the absence of immune cells.

Seahorse metabolic quantification

Neutrophils (2 × 10⁵ cells/well) were resuspended in assay media and plated in microplates precoated with 22.4 μg/mL Cell-Tak (Corning) in PBS. Assay media was prepared in XF RPMI complete assay media supplemented with 10 mM glucose, 2 mM L-glutamine, and 1 mM sodium pyruvate. After incubating at 37°C without CO₂ for 45 min, transfer the cartridge and plate to the Seahorse XFe96 Analyzer (Agilent Technologies). Oxygen consumption rate (OCR) was measured at baseline and after sequential injections of ATP synthase inhibitor oligomycin (2.5 μM, Sigma-Aldrich), uncoupler FCCP (0.61 μM, Sigma-Aldrich), and complex I/II inhibitors rotenone/antimycin A (1 μM/0.1 μM).

Confocal microscopy

Neutrophils in 200 μL of 0.5% ff-BSA/HBSS were cytospun onto slides by centrifugation at 800 ×g for 5 min. Cells were fixed in 4% PFA for 10 min at room temperature, stained with anti-Tom20 (1:200, NBP2-67501, Novus Biologicals, Centennial, CO) overnight, and incubated with the secondary fluorescent antibody and DAPI for 1 h. Cells were then mounted with ProLong Gold. Airyscan images were acquired on a Zeiss LSM800 confocal microscope. Mitochondrial intensity, size, and count were quantified using ImageJ.

Intraperitoneal (IP) bacterial infection

For systemic bacterial challenge, mice were infected via intraperitoneal (IP) injection with Escherichia coli (E. coli K12). Briefly, bacteria were cultured to mid-log phase, washed, and resuspended in sterile phosphate-buffered saline (PBS). Mice were injected intraperitoneally with 1.66 × 10⁶ colony-forming units (CFU)/g in a total volume of 200 μL. At 14 hours post-infection, mice were euthanized, and samples were collected for bacterial burden analysis. Blood was collected via cardiac puncture, and organs, including the lung, liver, and brain, along with peritoneal lavage fluid, were harvested aseptically. Tissues were homogenized in sterile PBS, serially diluted, and plated on LB agar. After overnight incubation at 37°C, CFUs were quantified. Bacterial burden was expressed as CFU per gram of tissue or per milliliter of fluid.

Intratracheal (IT) bacterial infection

To assess local antibacterial responses in the lung, mice underwent intratracheal (IT) bacterial challenge. Mice were anesthetized with isoflurane, and E. coli K12 (3 × 10⁶ CFU in 50 μL PBS) was delivered intratracheally via a sterile catheter. At 24 hours post-infection, bronchoalveolar lavage fluid (BALF) was collected by flushing the airways with 3 mL of sterile PBS. BALF was centrifuged to separate cells and supernatant. Supernatants were used to quantify bacterial CFU and measure protein, while cell pellets were analyzed by flow cytometry. Lungs were harvested, homogenized in sterile PBS, and plated for CFU enumeration as described above. Bacterial burden was reported as CFU per lung or per milliliter of BALF. BALF cells were stained with fluorophore-conjugated antibodies against CD45, CD11b, and Ly6G to identify neutrophils (CD45⁺CD11b⁺Ly6G⁺). Data were acquired on a flow cytometer and analyzed using FlowJo software. Neutrophil frequency was expressed as a percentage of total CD45⁺ cells. Total protein concentration in BALF supernatants was measured using a bicinchoninic acid (BCA) protein assay according to the manufacturer’s instructions.

Influenza infections and viral load quantification

The Influenza A/Puerto Rico 8/1934 (PR8) H1N1 strain was obtained from Charles River (Cat# 10100374, Wilmington, MA), aliquoted, and stored in liquid nitrogen. Mice were anesthetized i.p. with Ketamine/Xylazine (Patterson Veterinary, Devens, MA) and infected intranasally with 60 egg infectious dose (EID₅₀)/gram or 0.16 lethal dose 50 (LD₅₀) (73, 74) for all experiments except survival experiments. Survival experiments were performed using an LD₅₀ dose, as determined by in vivo virus titration, at 360 EID₅₀/g (73, 74). Body weight was measured every day. Viral titers were determined by quantifying viral transcripts using reverse transcription (RT)-quantitative (q) PCR, as previously described (74–77). Primers: Polymerase acidic protein gene (PA), 5’-CGGTCCAAATTCCTGCTGA-3’ (forward) and 5’-CATTGGGTTCCTTCCATCCA-3’ (reverse); Nucleoprotein (NP), 5’-CAGCCTAATCAGACCAAATG-3’ (forward) and 5’-TACCTGCTTCTCAGTTCAAG-3’ (reverse); Hemagglutinin (HA), 5’-GAGGAGCTGAGGGAGCAAT-3’ (forward) and 5’-GCCGTTACTCCGTTTGTGTT-3’ (reverse).

Oxygen pulse meter

Oxygen saturation was measured every other day during the viral infection. Mice were shaved around the neck, where the O2 sensor was placed, and measurements were taken while the mice were awake using the MouseOx Plus (Starr Life Sciences Corp, Oakmont, PA).

Bronchoalveolar lavage fluid (BALF) collection

The BALF collection was performed as previously described (78). Briefly, mice were euthanized by intraperitoneal injection of a lethal dose of ketamine/xylazine. The neck was disinfected with 70% ethanol, and the trachea was exposed via a midline incision. A 26-gauge needle was used to puncture the trachea between two cartilage rings, and a sterile catheter was inserted (∼0.5 cm) and secured with cotton thread. A 1 mL syringe containing 1 mL of sterile HBSS with 2 mM

EDTA was connected to the catheter. The solution was gently instilled into the lungs and aspirated while the thorax was massaged. Approximately 700–900 µL of lavage fluid was recovered per instillation, and the procedure was repeated twice. The pooled fluid was centrifuged at 400 × g for 5 min at 4 °C. Supernatants were collected for protein analysis or stored at −80 °C. Cell pellets were resuspended in 200 µL of ACK lysis buffer for 2 min at room temperature, diluted with 1 mL of ice-cold HBSS, and centrifuged again at 400 × g for 5 min. The final cell suspension was prepared in HBSS with 0.5% ff-BSA and 2 mM EDTA for flow cytometry or cytological analysis.

Single-cell RNA sequencing (scRNA-seq) analysis

The scRNA-seq analysis was performed on peritoneal cells isolated from mice 10 h after thioglycolate injection. Approximately 1.6 × 10⁹ paired-end reads were obtained from 10× scRNA-seq using the Chromium Next GEM 3’ kit v3.1 and NovaSeq 6000 (79). Raw sequencing data were processed with CellRanger and analyzed in R (version 4.3.1) using the Seurat (version 4.0.5) package, yielding ∼33,000 cells with an average of 5,660 unique molecular identifiers (UMIs) and 1,672 genes per cell. Cell clusters were annotated using canonical markers, and neutrophil subsets (G5a/b/c) were identified as previously described (16). Differential expression was assessed using FindMarkers (log₂ fold change > 0.25; adjusted p < 0.05). Module scores were computed with AddModuleScore in Seurat. For each pathway, the mean expression of its member genes (MSigDB Hallmark sets for Apoptosis; GO/Reactome sets for Autophagy and Mitophagy) was centered using matched control genes with matched baseline expression, yielding a relative activity score per cell. Violin plots display the density of cells across module scores; center lines indicate modal values without assuming a parametric distribution. Pathway-level scores were standardized module scores across curated gene sets, with higher scores indicating stronger pathway activity.

Statistical analysis

All data, except scRNA-seq data, were analyzed in Prism (GraphPad, Boston, MA) and presented as mean ± SD. Statistical significance was assessed using one-or two-tailed unpaired Student’s t-tests or ANOVA, as appropriate. p values less than 0.05 were considered significant. Experiments were performed with at least 3 independent biological replicates. Survival curves in the infection model were compared using the Gehan–Breslow–Wilcoxon test (GraphPad Prism) to emphasize early differences, and similar results were obtained using the log-rank (Mantel– Cox) test. For scRNA-seq violin plots, statistical comparisons of modal values were performed using Wilcoxon rank-sum tests at the cell level, followed by Benjamini–Hochberg FDR correction. Bracket labels report FDR p-values. This cell-level analysis is exploratory.

Data availability

RNAseq data generated in this study have been deposited in the Gene Expression Omnibus (GEO) under accession number GSE316812

Additional information

Funding

National Institutes of Health grants R01AI173377 and R01HL167723 to T.H., R01 HL162642 to J.O.M., and R01AI142642, R01AI145274, R01AI141386, R01HL092020, and P01HL158688 to H.R.L.

Postdoctoral Research Abroad Program, Ministry of Science and Technology, Taiwan (Y.L)

NIH training grant T32HL066987 and the Cotran-Gimbrone Research Award (A.Y.H.)

American Heart Association grant 24POST1195961 (A.G.)

AbbVie-Harvard Medical School Alliance (J.O.M.)

The Pew Charitable Trusts Biomedical Scholars (J.O.M.)

The New York Stem Cell Foundation (J.O.M.)

The Cell Discovery Network, a collaborative funded by The Manton Foundation and The Warren Alpert Foundation at Boston Children’s Hospital (J.O.M.)

Cancer Research Institute CRI Irvington postdoctoral fellowship (S.W.K.)

Author contributions

Conceptualization: Y.L., T.S., T.H.

Methodology: Y.L., T.S., A.Y.H., A.C., A.K., M.V.L., A.G., I.F., V.A.B., S.G., R.C., J.O.M., S.W.K.

Investigation: Y.L., T.S., A.Y.H., A.C., A.K., M.V.L., A.G., I.F., V.A.B., S.G., R.C.

Visualization: Y.L., T.S.

Funding acquisition: T.H., H.R.L, J.O.M.

Project administration: T.H., Y.L., T.S.

Supervision: T.H.

Writing – original draft: Y.L., T.S., T.H.

Writing – review & editing: A.Y.H., A.C., A.K., M.V.L., A.G., I.F., V.A.B., S.G., R.C., J.O.M., S.W.K., H.R.L,

Funding

HHS | NIH | National Institute of Allergy and Infectious Diseases (NIAID) (R01AI173377)

• Timothy Hla

HHS | NIH | National Heart, Lung, and Blood Institute (NHLBI) (R01HL167723)

• Timothy Hla

HHS | NIH | National Heart, Lung, and Blood Institute (NHLBI) (HL162642)

• José Ordovas-Montanes

HHS | NIH | National Institute of Allergy and Infectious Diseases (NIAID) (R01AI142642)

• Hongbo Luo

HHS | NIH | National Institute of Allergy and Infectious Diseases (NIAID) (R01AI145274)

• Hongbo Luo

HHS | NIH | National Institute of Allergy and Infectious Diseases (NIAID) (R01AI141386)

• Hongbo Luo

HHS | NIH | National Heart, Lung, and Blood Institute (NHLBI) (P01HL158688)

• Hongbo Luo

Additional files

Supplementary Materials. Figs. S1 to S5. Table S1.