RPSA-OLFM4 axis governs neutrophil migration against bacterial infection and sepsis

Abstract

Neutrophil migration to bacterial infection sites is key for host defense. Host ribosomal protein SA (RPSA) has been recently reported to regulate the anti-infection immunity of immune cells; however, its role in neutrophil migration remains unclear. Here, using myeloid-specific Rpsa-deficient mice, we found that RPSA deletion inhibited neutrophil infiltration and markedly exacerbated Streptococcus suis serotype 2 infection. Adoptive cell transfer and neutrophil depletion assays identified RPSA as vital for the anti-infective function of neutrophils. Mechanistically, RPSA deficiency induced the overexpression of olfactomedin 4 (OLFM4), which in turn inhibited the activation of the RhoA/ROCK1/pMLC2 signaling pathway, reduced MYH9 expression, and caused aberrant MYH9 translocation from the uropod to the cytosol in migrating neutrophils. Ultimately, this disrupted cytoskeletal polarization and uropod extension, thereby abrogating migratory function. Clinically, septic patients’ neutrophils exhibited reduced RPSA and elevated OLFM4 expression, a phenotype that correlated with a marked impairment of migratory capacity. Therapeutic targeting of the RPSA-OLFM4 axis restored neutrophil migration and improved disease outcomes in both S. suis 2-infected and septic mice. Thus, our findings demonstrate that RPSA promotes neutrophil migration via downregulating OLFM4 to counter bacterial infection, and establish the RPSA-OLFM4 axis as a critical immune migratory checkpoint in host antibacterial immunity.

Whilst it is known that host ribosomal protein SA (RPSA) regulates the anti-infection immunity in immune cells, its role in neutrophil migration is yet to be explored. In this work, authors provide mechanistic insight into the impact of RPSA on promoting neutrophil migration via the regulation of olfactomedin 4 (OLFM4), in the context of sepsis.

Introduction

Neutrophils constitute the primary cellular defense against invading pathogens. Their ability to traverse the vascular endothelial barrier and efficiently extravasate into inflamed tissues is crucial for pathogen clearance^1,2. This trans-endothelial migration process, termed neutrophil extravasation, involves sequential stages including activation, rolling, adhesion, and crawling. Upon activation, neutrophils upregulate integrin expression and adhesion to endothelial cells, accompanied by morphological flattening and polarization³. Polarization—a defining characteristic of neutrophil motility—involves cytoskeletal reorganization to establish front-rear asymmetry, thereby enabling directional migration. This polarized state is characterized by a persistent leading-edge pseudopod and a distinct trailing uropod, facilitating cell crawling along the endothelium and in the tissues⁴. Notably, polarization is indispensable for efficient neutrophil migration, and its impairment significantly delays migratory responses⁵. Within polarized neutrophils, spatially segregated molecular regulators at the leading-edge pseudopod (e.g., Cdc42) or uropod (e.g., RhoA, ROCK, MLC2, MYH9) orchestrate actin and myosin polymerization dynamics to drive migration. Molecular regulator perturbations disrupt the expression and localization of filamentous actin (F-actin), ultimately inhibiting neutrophil migration^6–9.

Ribosomal protein SA (RPSA), a component of the small ribosomal subunit and a high-affinity receptor for laminin, is ubiquitously expressed in eukaryotes and participates in diverse biological processes¹⁰. In the context of bacterial infection, RPSA usually acts as a receptor for protein virulence factors from multiple meningitis-associated pathogens, including CbpA of Streptococcus pneumoniae, surface proteins PilQ and PorA of Neisseria meningitidis, and the OmpP2 protein of Haemophilus influenzae. Deletion of the critical regions of these proteins significantly reduces bacterial adhesion to RPSA and attenuates meningitic pathogenicity^11–14. Our previous work further identified that the virulence factor Enolase of Streptococcus suis serotype 2 (S. suis 2) binds to RPSA on brain microvascular endothelial cells, which exacerbates inflammation, induces apoptosis, and triggers meningitis¹⁵. These studies indicate that pathogens can exploit RPSA to facilitate their infection of host cells. In striking contrast, the role of RPSA in anti-infective responses of immune cells is largely unexplored.

RPSA mediates the adhesion of innate immune cells to the vascular wall by binding to laminin, a major adhesive ligand in the vascular basement membrane. Furthermore, RPSA expression on immune cells increases in response to inflammatory agonists, such as phorbol 12-myristate 13-acetate (PMA) or formyl-methionyl-leucyl-phenylalanine (fMLP)¹⁶. Our previous work demonstrated that RPSA expression is upregulated in multiple immune cell types during S. suis 2 infections in mice, and this expression exhibits dynamic changes that correlate with infection progression¹⁷. As an important zoonotic pathogen, S. suis can cause severe diseases, including sepsis, arthritis, and meningitis^18,19. Notably, a significant increase in RPSA⁺ neutrophils occurs in the brain parenchyma of S. suis 2-infected mice, and this subset has enhanced bactericidal capacity compared to RPSA^- neutrophils²⁰. Additionally, nuclear RPSA in macrophages functions as a sensor for viral nucleic acids, thereby promoting cytokine production and amplifying inflammatory immune responses²¹. These studies suggest that RPSA is involved in regulating the function of innate immune cells. However, the precise role and underlying mechanisms of RPSA in neutrophils during bacterial infections are unclear.

In this study, using myeloid-specific Rpsa-deficient mice, we demonstrated that RPSA deficiency significantly impairs neutrophil infiltration into infected tissues, exacerbating disease severity following S. suis 2 infection. Mechanistically, RPSA ablation promotes OLFM4 overexpression, which suppresses the RhoA/ROCK1/pMLC2 signaling pathway, reduces MYH9 expression, and results in MYH9 translocation from the uropod to the cytosol in migrating neutrophils. These alterations ultimately disrupt uropod protrusion and cytoskeletal polarization, thereby abrogating neutrophil migration. Notably, therapeutic targeting of the RPSA-OLFM4 axis effectively modulates neutrophil migration, thereby exerting a profound impact on host antibacterial defense. Our findings identify the underlying mechanism that enables RPSA to drive neutrophil migration, expanding the understanding of its immunological functions. Furthermore, we establish the RPSA-OLFM4 axis as a critical regulatory target for neutrophil migration, offering a promising therapeutic strategy to enhance host resistance to pathogens.

Results

RPSA enhances the innate immunity against S. suis 2 infection

To explore the RPSA’s function on host immune cell response, we first analyzed RPSA expression in murine peripheral blood mononuclear cells before and at 12 h after S. suis 2 infection. The infection significantly increased the proportion of RPSA⁺ cells within myeloid lineages (neutrophils, monocytes, dendritic cells), with no alteration in T or B cells (Fig. 1a, Supplementary Fig. 1). This myeloid-specific upregulation suggested RPSA may regulate innate immune cell function. To directly assess the function of RPSA in myeloid cell-mediated anti-infective immunity, we generated myeloid-specific conditional knockout mice Lyz2^IRES-iCreRpsa^fl/fl (Rpsa^−/−). Compared to Rpsa^fl/fl controls, Rpsa^−/− mice had significantly lower survival rates post-S. suis 2 challenge (Fig. 1b), along with higher clinical scores, and bacterial burdens in peripheral blood and multiple tissues (Fig. 1c, d). Pathological examination showed pronounced congestion and hemorrhages in the brains and lungs of Rpsa^−/− mice (Fig. 1e). Furthermore, histopathological analysis revealed exacerbated meningeal and pulmonary interstitial thickening, with more severe hemorrhage and lesions in Rpsa^−/− mice versus controls (Fig. 1f).

Fig. 1. RPSA deficiency exacerbates susceptibility of mice to S. suis 2 infection.

a Quantification of RPSA⁺ immune cells in peripheral blood before and at 12 h after infection using flow cytometry (n = 3). b Survival curves of Rpsa^fl/fl and Rpsa^−/− mice after infection (n = 8/group). c Clinical scores monitored for 120 h post infection (h p.i.) (for Rpsa^fl/fl group, n = 8 at 0–24 h, n = 7 at 48–120 h; for Rpsa^−/− group, n = 8 at 0–12 h, n = 7 at 24 h, n = 6 at 48–120 h). d Bacterial loads in peripheral blood and indicated tissues at 12 h p.i. (CFU/mL or CFU/g) (n = 5). e Representative gross pathology of brain and lung tissues at 72 h p.i. f Representative H&E-stained sections at 72 h p.i. Black dashed boxes indicate exacerbated pathology in Rpsa^−/− mice. Scale bar, 200 μm. Data were represented as mean ± SD and analyzed using two-way ANOVA with Bonferroni’s post hoc test (a, c), log-rank (Mantel-Cox) test (b), and two-tailed unpaired Student’s t-test (d). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

To determine whether the effects of RPSA deficiency were specific to S. suis 2 infection, we employed a distinct pulmonary infection model using the Gram-negative Klebsiella pneumoniae strain KP35 (temporarily designated, unpublished). In this independent model, Rpsa^−/− mice similarly recapitulated the detrimental phenotype observed in the S. suis 2 model, including significantly reduced survival, increased bacterial burdens, more severe clinical scores and weight loss, and exacerbated pulmonary pathology (Supplementary Fig. 2a–e). Taken together, our results demonstrate that RPSA mediates host defense against bacterial infections by regulating myeloid cell function, and this role is not specific to S. suis 2.

RPSA is crucial for neutrophil anti-infective function

Myeloid cells primarily encompass neutrophils, macrophages, monocytes, and dendritic cells. To identify which myeloid subset underlies RPSA-mediated enhancement of anti-infective capacity, we performed flow cytometric analysis of myeloid populations in tissues from S. suis 2-infected Rpsa^fl/fl and Rpsa^−/− mice. Analyses revealed that RPSA deficiency significantly reduced neutrophil infiltration in multiple tissues (brain, lungs, liver, spleen, and kidneys), while elevating their abundance in peripheral blood (Fig. 2a). Immunofluorescence staining of lung tissue further corroborated these findings (Fig. 2b). A similar phenomenon was observed in mice infected with Klebsiella pneumoniae strain KP35 (Supplementary Fig. 2f). Collectively, these results indicate that RPSA is likely involved in regulating neutrophil function.

Fig. 2. RPSA deficiency impairs neutrophil-mediated host defense against S. suis 2 infections.

a Quantification of myeloid subsets in tissues of Rpsa^fl/fl and Rpsa^−/− mice at 12 h p.i. via flow cytometry (n = 5). b Pulmonary neutrophil infiltration visualized by immunofluorescence. Anti-Ly6G (red), Nuclei: DAPI (blue). Scale bar, 100 μm. c Experimental scheme for neutrophil depletion assays (Created in BioRender. Wu, T. (2026) https://BioRender.com/aeifwu9). d, e Bacterial loads in Rpsa^fl/fl (d) and Rpsa^−/− (e) mice after administration of anti-Ly6G (500 μg, twice, i.p.) or isotype IgG control. NS, normal saline. Tissues harvested at 12 h p.i. (n = 5). f Neutrophil infiltration was assessed in the lung tissue of S. suis 2-infected Rpsa^fl/fl and Rpsa^−/− mice with/without neutrophil depletion by immunofluorescence. Anti-Ly6G (red), Nuclei: DAPI (blue). Scale bar, 100 μm. g Survival curves post-infection (n = 6/group). h Experimental scheme for adoptive transfer assays. 1 × 10⁸ neutrophils from Rpsa^fl/fl or Rpsa^−/− donor mice were intravenously (i.v.) transferred to Rpsa^−/− recipients before infection (Created in BioRender. Wu, T. (2026) https://BioRender.com/0kxzi5i). i Survival curves of recipient mice post-infection (n = 6/group). j Bacterial loads in recipient mice at 12 h p.i. (n = 3). k Infiltration of transferred neutrophils (dye-labeled) in recipient mice was quantified by flow cytometry (there was no detectable transferred neutrophil in either brain or kidney) (n = 3). Data were represented as mean ± SD and analyzed using two-way ANOVA with Bonferroni’s post hoc test (a), one-way ANOVA followed by Tukey’s post hoc test (d), two-tailed unpaired Student’s t-test (e, j, k), log-rank (Mantel-Cox) test (g, i). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

To further determine whether RPSA confers anti-infective protection through neutrophils, we established a neutrophil depletion model using anti-Ly6G antibody, which resulted in a ~ 94% depletion efficiency in peripheral blood as confirmed by flow cytometry (Fig. 2c, Supplementary Fig. 2g). Following S. suis 2 infection, neutrophil-depleted Rpsa^fl/fl mice exhibited significantly increased tissue bacterial burdens and decreased pulmonary neutrophil infiltration (Fig. 2d, f), confirming the essential role of neutrophils in host defense against S. suis 2. In Rpsa^−/− mice, neutrophil depletion did not significantly alter bacterial loads compared to isotype antibody treatment (Fig. 2e), and both groups displayed similarly impaired pulmonary neutrophil infiltration (Fig. 2f). Furthermore, the survival rates of neutrophil-depleted Rpsa^fl/fl and Rpsa^−/− mice were reduced to the levels comparable to those of Rpsa^−/− mice injected with isotype control antibody (Fig. 2g), indicating that neutrophil depletion causes detrimental effects similar to RPSA deficiency. To directly assess the impact of RPSA on neutrophil anti-infective function, we adoptively transferred Rpsa^fl/fl or Rpsa^−/− neutrophils, with a purity >90% (Supplementary Fig. 2h), into Rpsa^−/− recipient mice before infection (Fig. 2h). Recipients of Rpsa^fl/fl neutrophils exhibited significantly improved survival and reduced bacterial burdens across all tissues compared to those receiving Rpsa^−/− neutrophils (Fig. 2i, j), confirming that Rpsa^−/− neutrophils lack effective anti-infective capacity. Analysis of transferred neutrophils in the recipient mice revealed significantly greater tissue infiltration by Rpsa^fl/fl neutrophils compared to Rpsa^−/− neutrophils, with Rpsa^−/− neutrophils accumulating at higher levels in peripheral blood (Fig. 2k), consistent with previous observations.

Thus, these results demonstrate that RPSA deficiency largely abrogates neutrophils’ anti-infection function, severely impairing their tissue infiltration and causing aberrant retention in peripheral circulation.

RPSA deficiency impairs neutrophil migration

Neutrophils execute their anti-infection functions by migrating along chemotactic gradients to capillaries at infection sites, and extravasating across vascular walls to tissues^1,2. Compared to Rpsa^fl/fl controls, Rpsa^−/− mice exhibited significantly elevated neutrophil counts in peripheral circulation but decreased tissue infiltration. Given that RPSA deficiency had no significant effect on neutrophil production by the bone marrow (Supplementary Fig. 2i) or their maturation state (Supplementary Fig. 2j), as assessed using the canonical maturation markers CD101 and CXCR4^22–24, these data collectively suggest a cell-intrinsic migratory defect of Rpsa^−/− neutrophils from vasculature into tissues. To verify this proposal, we established an S. suis 2 dorsal skin wound model in mice. Histological analysis revealed abundant neutrophil infiltration in infected tissues of Rpsa^fl/fl mice, whereas Rpsa^−/− mice showed minimal tissue infiltration with neutrophils accumulating inside capillaries (Fig. 3a). To visualize migration defects, we injected lipopolysaccharide (LPS) into the scrotum of male mice to induce localized inflammation. Subsequent immunofluorescence staining of vascular endothelial cells and neutrophils in the cremaster muscle revealed robust trans-endothelial migration and tissue infiltration in Rpsa^fl/fl mice. In contrast, Rpsa^−/− neutrophils predominantly accumulated within capillaries, exhibiting markedly reduced extravasation into perivascular tissues (Fig. 3b). Thus, these results suggest that RPSA deficiency abrogates neutrophil migratory capacity, severely impairing its recruitment to infected tissues.

Fig. 3. Rpsa deletion impaired neutrophil migratory capacity across infection and inflammation models.

a H&E-stained section showing neutrophil intravascular accumulation (black arrows) in S. suis 2-infected skin wounds. Left scale bar, 50 μm; Right scale bar, 20 μm. b Neutrophil infiltration into the LPS-induced inflamed mouse cremaster muscle. Vessels (green), neutrophils (red). White arrows indicate stalled neutrophils. Scale bar, 200 μm. c, d Migration assay in vivo. Labeled Rpsa^fl/fl and Rpsa^−/− neutrophils were mixed and transferred into recipient WT mice. Neutrophil recruitment to the peritoneum was quantified after challenge with S. suis 2 (i.p.) (Created in BioRender. Wu, T. (2026) https://BioRender.com/yg5q4m6) (c). Data represent the ratio of transferred neutrophils that infiltrated into the peritoneal cavity (n = 3) (d). e, f Transwell migration assay ex vivo. Labeled Rpsa^fl/fl and Rpsa^−/− neutrophils were mixed and seeded in the inserts. After incubating with or without S. suis 2 for 3 h at 37 °C, the transmigrated cells in the lower chamber were quantified (Created in BioRender. Wu, T. (2026) https://BioRender.com/60nxhmy) (e). Data represent the ratio of migrated neutrophils (n = 3) (f). g, h Ex vivo flow chambers under physiological shear stress, comparison of the mean crawling speed of neutrophils adherent to a monolayer of bEnd.3 cells (n = 50 cells) (g), and comparison of the number of adherent neutrophils across different random fields of view (n = 7 fields), with cell counts normalized to the Rpsa^fl/fl group (h). i Transwell migration assay of DdHL-60 cells. RPSA shRNA-transfected DdHL-60 (CellTrace^TM CFSE-labeled) and vector-transfected control DdHL-60 cells (CellTrace^TM Violet-labeled) were mixed at a 1:1 ratio and seeded in inserts. Transmigrated cells in the lower chambers were quantified. Data represent the cell ratio of migrated DdHL-60 cells (n = 3). Data were represented as mean ± SD and analyzed using two-tailed unpaired Student’s t-test (d, f–i). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Neutrophil migration involves complex interactions across multiple tissues, cells, and soluble factors within the physiological or pathological microenvironment⁵. To explore the intrinsic role of RPSA in neutrophils, we adoptively transferred fluorescent dye-labeled neutrophils from Rpsa^fl/fl and Rpsa^−/− mice (mixed at a 1:1 ratio) into wild-type (WT) recipient mice via tail vein injection (Fig. 3c). Successful labeling of the transferred neutrophils was confirmed by flow cytometry (Supplementary Fig. 3a). To exclude the potential confounding effect of differential early organ sequestration on the input cell ratio, we collected peripheral blood 15 min post-transfer (prior to infection) and confirmed by flow cytometry that the circulating ratio of Rpsa^fl/fl to Rpsa^−/− neutrophils remained at ~1:1 (Supplementary Fig. 3b). Subsequently, recipient mice were challenged with S. suis 2 via intraperitoneal (i.p.) injection. Analysis of peritoneal lavage fluid revealed that nearly all extravasated donor neutrophils were Rpsa^fl/fl cells, whereas Rpsa^−/− neutrophils were significantly underrepresented (Fig. 3d). These results confirm that RPSA deficiency intrinsically impairs the migratory function of neutrophils. Consistent with in vivo findings, ex vivo Transwell assays showed significantly reduced migration of Rpsa^−/− neutrophils toward chemoattractant gradients, compared to Rpsa^fl/fl controls, regardless of S. suis 2 infection (Fig. 3e, f). In addition, RPSA deficiency impaired neutrophil adhesion to mouse brain-derived endothelial cells.3 (bEnd.3) and transmigration across bEnd.3 monolayer barriers (Supplementary Fig. 3c, d). To further dissect the defect under hemodynamic conditions, we employed a flow chamber assay that mimics physiological shear stress (1 dyne/cm²) to assess neutrophil adhesion and crawling on TNF-α-stimulated bEnd.3 monolayer (Supplementary Movie 1). Analysis of neutrophil dynamics revealed that the mean crawling velocities of Rpsa^fl/fl neutrophils were significantly higher than those of Rpsa^−/− neutrophils (Fig. 3g). Although the number of adherent Rpsa^−/− neutrophils was slightly lower, the difference was not statistically significant (Fig. 3h). β2-integrins are essential for the firm adhesion of neutrophils to endothelial cells and subsequent migration²⁵. We therefore assessed whether this migration defect could be attributed to altered β2-integrin expression or activation. We found that RPSA deficiency exerted no significant effect on the surface expression levels of CD11a or CD11b on neutrophils, under both resting and following CXCL1 (KC) stimulation (Supplementary Fig. 3e–g). Moreover, ligand-binding assays confirmed that RPSA deficiency had no significant effect on the activation-dependent binding of LFA-1 to ICAM-1 or Mac-1 to fibrinogen (Supplementary Fig. 3h). Taken together, these findings indicate that the migratory defect of Rpsa^−/− neutrophils is unlikely to result from altered integrin expression or activation, but rather stems from an intrinsic impairment in neutrophil motility.

Human HL-60 cells can differentiate into neutrophil-like cells upon DMSO induction (DdHL-60), serving as a model to study neutrophil chemotaxis^26,27. DdHL-60 cells showed increased CD11b expression and could migrate toward chemoattractant gradients in Transwell migration assays (Supplementary Fig. 4a, b), confirming their successful differentiation. Similarly, RPSA knockdown in DdHL-60 cells significantly suppressed their migration capacity (Fig. 3i). Crucially, Rpsa^−/− neutrophils exhibited enhanced survival compared to Rpsa^fl/fl neutrophils during co-culture with S. suis 2 (Supplementary Fig. 4c), excluding cell viability as a confounding factor on neutrophil migration.

Collectively, these data demonstrate that RPSA deficiency inhibits the extravasation of neutrophils into infected tissues by directly impairing their intrinsic migratory capacity.

RPSA deficiency impairs neutrophil uropod formation

Cell motility is crucial for neutrophil migration to the infectious tissues. Analysis of neutrophil migration trajectories under KC & fMLP chemoattractant gradients in an ex vivo agarose gel assay was used to determine whether RPSA deficiency impairs neutrophil motility. While Rpsa^−/− neutrophils retained directional migration along chemotactic gradients, they had significantly reduced crawling distances and velocities compared to Rpsa^fl/fl controls (Fig. 4a, Supplementary Movie 2), indicating compromised motility.

Fig. 4. RPSA deficiency disrupts uropod formation in chemoattractant-activated neutrophils.

a Time-lapse imaging of directional migration of Rpsa^fl/fl and Rpsa^−/− neutrophils toward gradients of KC (100 ng/mL) & fMLP (10 nM) in an agarose gel model (n = 19). Images were acquired over 1 h at 30 s intervals. Trajectories were generated using Chemotaxis and Migration Tool v2.0 and analyzed with ImageJ software. b RT-qPCR analysis of Rhoa, Rock1, and Myl2 expression in neutrophils stimulated with S. suis 2 (MOI = 10), LPS (100 ng/mL), KC (100 ng/mL), or fMLP (10 nM) for 2 h at 37 °C (n = 5). Data are normalized to Gapdh. c Western blotting analysis of RhoA, ROCK1, MLC2 and phosphorylated-MLC2 (p-MLC2) protein level in neutrophils. β-actin served as a loading control. d RT-qPCR analysis of RHOA, ROCK1, and MYL2 expression in RPSA-knockdown HL-60 cells. Cells were transfected with RPSA-targeting shRNA or control vector for 48 h before analysis (n = 3). Data are normalized to GAPDH. e Representative scanning electron microscopy images showing uropod morphology in KC & fMLP-stimulated Rpsa^fl/fl and Rpsa^−/− neutrophils and the quantification of uropod length (n = 10). Scale bars, 1 μm. Data were represented as mean ± SD and analyzed using two-tailed unpaired Student’s t-test (a, d, e), two-way ANOVA with Bonferroni’s post hoc test (b, c). A p-value ≤ 0.05 was considered statistically significant, ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Front-rear polarity is a defining feature, characterized by a persistent leading-edge pseudopod and a distinct trailing uropod, that regulates neutrophil motility⁴. Cdc42 and Rac2, localized at the leading edge, and RhoA, along with its downstream effectors ROCK and myosin light chains (MLC), residing in the uropod, collectively coordinate directional migration²⁸. Next, we determined the expression of these polarity-associated signaling molecules in neutrophils. After stimulation with S. suis 2, LPS, KC or fMLP, RPSA deficiency significantly suppressed the expression of Rhoa, Rock1, and Myl2, with no significant effect on Cdc42 and Rac2 expression (Fig. 4b, Supplementary Fig. 4d). Protein expression analysis further validated these findings, and revealed inhibited MLC2 phosphorylation upon RPSA loss (Fig. 4c). Notably, the transcriptional levels of key neutrophil-derived chemokines (Cxcl1, Cxcl2) and chemoattractant receptors (Cxcr2, Fpr1) were not suppressed by RPSA deficiency (Supplementary Fig. 5a), indicating that the observed motility defect is unlikely to stem from impaired chemokine signaling. Consistent with these findings in murine neutrophils, RPSA knockdown in HL-60 cells significantly downregulated transcription of genes in RHOA/ROCK1/MYL2 pathway (Fig. 4d). Importantly, compared to Rpsa^fl/fl, electron microscopy revealed significantly shortened uropod protrusions in Rpsa^−/− neutrophils (Fig. 4e), indicating impaired uropod formation. A similar phenomenon was observable in Supplementary Movie 1, Rpsa^fl/fl neutrophils exhibited pronounced ameboid shape changes with clear uropod-like extensions, whereas Rpsa^−/− cells largely failed to develop such protrusive structures despite retaining adhesion capacity. Collectively, these results demonstrate that RPSA deficiency inhibits the activation of the RhoA/ROCK1/pMLC2 signaling pathway and uropod protrusion, thereby impairing neutrophil motility.

OLFM4 upregulation induced by RPSA deletion suppresses neutrophil migration

To further dissect the molecular mechanism underlying how RPSA deficiency impairs neutrophil migration, we performed RNA sequencing analysis on neutrophils isolated from Rpsa^fl/fl and Rpsa^−/− mice. Bioinformatic analysis identified a significant increase in Olfm4 transcript abundance following RPSA ablation, regardless of infection status (Fig. 5a, b, Supplementary Fig. 5b). The upregulation of mRNA and protein levels, determined by RT-qPCR and western blotting analysis, respectively, was also found (Fig. 5c, d). In addition, elevated OLFM4 expression was detected in the serum of Rpsa^−/− mice and in conditioned medium from Rpsa^−/− neutrophil cultures (Supplementary Fig. 5c). RPSA knockdown in HL-60 cells also led to enhanced Olfm4 transcription and protein expression (Fig. 5e, f). Thus, these data indicate that RPSA deficiency results in OLFM4 overexpression.

Fig. 5. RPSA deficiency promotes OLFM4 overexpression.

a, b RNA sequencing analysis of neutrophils isolated from Rpsa^fl/fl and Rpsa^−/− mice. Volcano plot showing fold change and P-value for the comparison of gene expression in Rpsa^fl/fl versus Rpsa^−/− neutrophils (without S. suis 2 infection) (a). Heatmap showing the expression of the top 20 differentially expressed genes from the RNA sequencing data (unannotated genes were excluded) (b). c RT-qPCR analysis of Olfm4 expression in Rpsa^−/− neutrophils (n = 5). Data normalized to Gapdh. d Western blotting analysis of OLFM4 expression in Rpsa^fl/fl and Rpsa^−/− neutrophils (n = 3 biologically independent mice per group). β-actin served as a loading control. e, f OLFM4 expression after RPSA knockdown in HL-60 cells using RT-qPCR (e, n = 3, data are normalized to GAPDH) and western blotting (f, n = 3 biologically independent cell culture preparations per group, GAPDH served as a loading control). Cells were transfected with RPSA-targeting shRNA for 48 h before analysis. Data were represented as mean ± SD and analyzed using two-way ANOVA with Bonferroni’s post hoc test (c, d), one-way ANOVA followed by Tukey’s post hoc test (e, f). A p-value ≤ 0.05 was considered statistically significant, ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

To investigate whether OLFM4 participates in regulating neutrophil migration, OLFM4 was overexpressed in DdHL-60 cells, which resulted in a marked impairment of neutrophil migration in vitro (Fig. 6a). Consistent with this in vitro finding, in an in vivo adoptive transfer model using C-NKG mice, overexpression of OLFM4 in DdHL-60 cells significantly inhibited their extravasation into the peritoneal cavity in response to a gradient of injected chemoattractants (Fig. 6b, c). Consistent with previous studies^29,30, co-treatment with all-trans retinoic acid (ATRA) and decitabine (5-aza-2'-deoxycytidine, Aza) substantially potentiated OLFM4 expression, whereas BAY-11-7082 (B11) exerted a potent suppressive effect (Supplementary Fig. 5d). We then used these activators and inhibitors to assess the function of OLFM4 on neutrophil migration, and found that ex vivo exposure to ATRA+Aza significantly attenuated the migration of Rpsa^fl/fl neutrophils (Fig. 6d). Conversely, incubation with B11 effectively restored the impaired migratory function in Rpsa^−/− neutrophils (Fig. 6e). Moreover, adoptive transfer experiments in wild-type recipient mice revealed that the recruitment of ATRA+Aza-preconditioned Rpsa^fl/fl neutrophils to the infectious peritoneal focus significantly diminished compared to DMSO-preconditioned controls (Fig. 6f, g). Thus, these findings establish a critical functional role for OLFM4 in governing neutrophil migration.

Fig. 6. OLFM4 overexpression inhibits neutrophil migration.

a Transwell migration assay. The method was identical to that described in Fig. 3i. OLFM4-overexpressing DdHL-60 (CellTrace^TM CFSE-labeled), vector-transfected controls (CellTrace^TM Violet-labeled). Data represent the ratio of migrated DdHL-60 (n = 3). b, c Effect of OLFM4 on DdHL-60 migration in C-NKG mice. Labeled OLFM4-overexpressing and vector-transfected control DdHL-60 were mixed and transferred into C-NKG mice. Neutrophil recruitment to the peritoneum was quantified (b) (Created in BioRender. Wu, T. (2026) https://BioRender.com/e0q1qrr). Data represent the ratio of infiltrated DdHL-60 (n = 5) (c). d Transwell migration assay. ATRA+Aza or DMSO treated Rpsa^fl/fl neutrophils were labeled and mixed, and subjected to a Transwell migration assay. Data represent the ratio of migrated neutrophils (n = 5). e Parallel migration assay with Rpsa^−/− neutrophils using B11 versus DMSO (n = 5), otherwise identical to panel (d). f, g In vivo migration assay. ATRA+Aza or DMSO-treated Rpsa^fl/fl neutrophils were labeled and mixed and transferred into WT mice. Neutrophil recruitment to the peritoneum was quantified after infection (f) (Created in BioRender. Wu, T. (2026) https://BioRender.com/em2w2ur). Data represent the ratio of infiltrated transferred-neutrophils (n = 6) (g). h RT-qPCR analysis of RHOA, ROCK1, and MYL2 expression in HL-60 under three conditions (n = 3). Data normalized to GAPDH. i Following the same procedure as in panel 6b, compare the migration capacity between DdHL-60 with simultaneous knockdown of RPSA and OLFM4 (CellTrace^TM CFSE-labeled) and DdHL-60 with RPSA knockdown alone (CellTrace^TM Violet-labeled). Data represent the ratio of infiltrated DdHL-60 (n = 5). j RT-qPCR analysis of RHOA/ROCK1/MYL2 expression in OLFM4-overexpressing HL-60. Cells were transfected for 48 h before analysis (n = 5). Data normalized to GAPDH. k Western blotting analysis of RhoA/ROCK1/MLC2 and p-MLC2 in HL-60 from (j) (n = 3 biologically independent cell culture preparations per group). β-actin served as a loading control. Data were represented as mean ± SD and analyzed using two-tailed unpaired Student’s t-test (a, c–e, g, i–k), one-way ANOVA followed by Tukey’s post hoc test (h). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Next, we explored the link between OLFM4 and the RhoA/ROCK1/pMLC2 pathway. Knockdown of OLFM4 alone in HL-60 cells had no significant effect on the transcription of RHOA, ROCK1, or MYL2 (Supplementary Fig. 5e). However, concomitant knockdown of OLFM4 and RPSA significantly rescued suppression of RHOA/ROCK1/MYL2 induced by RPSA deficiency (Fig. 6h), and, importantly, reversed the in vivo migration defect, as evidenced by significantly promoted extravasation of the double-knockdown DdHL-60 cells in the C-NKG mouse model (Fig. 6i). Furthermore, OLFM4 overexpression markedly suppressed both the transcription and protein expression of RhoA, ROCK1, and MLC2 (Fig. 6j, k).

Altogether, these data demonstrate that RPSA deficiency leads to OLFM4 overexpression, which subsequently suppresses the activation of the RhoA/ROCK1/pMLC2 signaling pathway, thereby impairing neutrophil migration.

RPSA-OLFM4 signaling modulates MYH9 expression and spatial distribution

MYH9, a non-muscle myosin heavy chain, plays a critical role in regulating neutrophil migration. During migration, MYH9 co-localizes with F-actin in the uropod of migrating neutrophils⁷. MYH9 activity is determined by regulatory MLC, positioning it downstream of the ROCK1/MLC2 axis^31,32. OLFM4 and MYH9 physically interact in humans³³. Thus, we hypothesized that elevated OLFM4 expression following RPSA deficiency may alter MYH9 expression and spatial distribution, thereby impairing neutrophil migration. Consistent with this hypothesis, RPSA deletion significantly suppressed MYH9 transcription and protein expression in primary neutrophils (Fig. 7a, b). Concordantly, OLFM4 overexpression in HL-60 cells attenuated MYH9 expression (Fig. 7c), indicating MYH9 is a downstream target of the RPSA-OLFM4 axis. The addition of pharmacological inhibitors further confirmed that MYH9 is regulated by the RhoA/ROCK1/MLC2 pathway (Supplementary Fig. 5f). By integrating these observations with the well-established modulatory effect of RPSA-OLFM4 on the RhoA/ROCK1/MLC2 pathway, our data show that RPSA deficiency triggers OLFM4 upregulation, which then suppresses RhoA/ROCK1/MLC2 expression and activation, ultimately leading to MYH9 downregulation.

Fig. 7. The RPSA-OLFM4 axis regulates MYH9 expression and spatial distribution.

a RT-qPCR analysis of Myh9 expression in Rpsa^fl/fl and Rpsa^−/− neutrophils (n = 3). Data normalized to Gapdh. b Western blotting analysis of MYH9 protein expression in Rpsa^fl/fl and Rpsa^−/− neutrophils (n = 3 biologically independent mice per group). β-actin serves as a loading control. c Western blotting analysis of MYH9 protein expression in OLFM4-overexpressing HL-60 cells (n = 3 biologically independent cell culture preparations per group). β-actin serves as a loading control. d Representative immunofluorescence results displaying the distribution of MYH9 and F-actin in neutrophils activated with KC & fMLP. MYH9 (green), F-actin (orange), Nuclei: DAPI (blue). Scale bar, 10 μm. e Representative immunofluorescence results displaying the co-localization of MYH9 and OLFM4 in the cytoplasm (white arrows). MYH9 (green), OLFM4 (red), Nuclei: DAPI (blue). Scale bar, 10 μm. f Quantification of Pearson’s correlation coefficients (for Rpsa^fl/fl group, n = 8; for Rpsa^−/− group, n = 10). g Co-IP analysis of MYH9-OLFM4 interaction in murine neutrophil lysates (mouse α-MYH9 IP, rabbit α-MYH9 and rabbit α-OLFM4 IB). Data were represented as mean ± SD and analyzed using two-way ANOVA with Bonferroni’s post hoc test (a, b), two-tailed unpaired Student’s t-test (c, f). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Next, immunofluorescence assays were used to determine MYH9 localization. There was aberrant MYH9 redistribution in migrating neutrophils upon RPSA deficiency, characterized by a shift from the uropod to cytosolic accumulation (Fig. 7d). Concurrently, co-localization studies demonstrated enhanced spatial association between OLFM4 and MYH9, which coincided with cytosolic accumulation of MYH9 driven by RPSA deficiency (Fig. 7e), supported by significantly increased correlation coefficients (Fig. 7f). Co-immunoprecipitation (Co-IP) further confirmed the interaction between OLFM4 and MYH9 in murine neutrophils (Fig. 7g). Collectively, these results demonstrate that RPSA deficiency triggers aberrant MYH9 translocation from the uropod to the cytosol, co-localization of OLFM4 with MYH9, compromising uropod protrusion, and neutrophil migration.

RPSA and OLFM4 serve as regulatory targets for neutrophil antimicrobial immunity

To determine whether OLFM4 is a therapeutically targetable molecule for regulating neutrophil-mediated anti-infective responses, WT and Rpsa^−/− mice were administered with OLFM4 agonist (ATRA + Aza) or inhibitor (B11), respectively, followed by assessment of their susceptibility to S. suis 2 infection (Fig. 8a). In WT mice, OLFM4 agonist treatment significantly reduced survival rates (Fig. 8b), impaired neutrophil infiltration into tissues (Fig. 8c), increased bacterial burdens (Fig. 8d), and exacerbated pulmonary pathological damage (Fig. 8e). Conversely, administration of OLFM4 inhibitor in Rpsa^−/− mice significantly improved survival (Fig. 8b), enhanced neutrophil tissue infiltration (Fig. 8c), reduced bacterial loads (Fig. 8d), and attenuated the aggravated pulmonary pathology associated with RPSA deficiency (Fig. 8e).

Fig. 8. Modulation of RPSA and OLFM4 affects anti-infective immunity.

a Agonist/inhibitor intervention in WT and Rpsa^−/− mice: WT and Rpsa^−/− mice received daily injections of ATRA+Aza or B11 for 5 days, respectively. Then challenged with S. suis 2 (Created in BioRender. Wu, T. (2026) https://BioRender.com/1t9qhis). b Survival curves within 48 h p.i. (n = 8/group). c Neutrophil infiltration was quantified at 12 h p.i. (n = 3). d Bacterial loads at 12 h p.i. (n = 3). e Representative H&E-stained lung sections. Scale bars, 100 μm. f Transwell migration assay, conferred with anti-RPSA IgG treatment, wild-type neutrophils were preincubated with anti-RPSA IgG or isotype IgG, then labeled and mixed, and subjected to Transwell migration assays. Data represent transmigrated neutrophil ratios (n = 3). g, h Anti-RPSA IgG or isotype control IgG were administered at 0 h and 23.5 h before S. suis 2 challenge (g) (Created in BioRender. Wu, T. (2026) https://BioRender.com/blpu07g). Neutrophil recruitment to the peritoneum was quantified 2 h after challenge with S. suis 2. Data represent the ratio of recruited neutrophils (n = 3) (h). i–l Effects of anti-RPSA IgG treatment on mouse susceptibility to S. suis 2 infections (treatment described as in panel g). Survival curves within 48 h p.i. (n = 8/group) (i); Neutrophil infiltration was quantified at 12 h p.i. (n = 3) (j); Bacterial loads at 12 h p.i. (n = 3) (k); Representative H&E-stained lung sections (l). Scale bars, 100 μm. Data were represented as mean ± SD and analyzed using two-tailed unpaired Student’s t-test (c, d, f, h, j, k), log-rank (Mantel-Cox) test (b, i). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Taking into account the observed increase in surface RPSA⁺ neutrophils post-S. suis 2 infection and the established role of RPSA in regulating OLFM4 expression, we investigated whether targeting RPSA could modulate anti-infective immunity. Ex vivo, pre-incubation of primary neutrophils with anti-RPSA IgG significantly inhibited neutrophil migration in Transwell assays (Fig. 8f). Furthermore, in vivo administration of anti-RPSA IgG resulted in significantly reduced neutrophil infiltration into the peritoneal cavity following S. suis 2 infection (Fig. 8g, h). Similar to the effect of OLFM4 agonist (ATRA + Aza), administration of anti-RPSA IgG reduced mouse survival rates, markedly suppressed neutrophil infiltration into tissues, and promoted neutrophil accumulation in the peripheral blood (Fig. 8i, j). Consequently, this led to a significant increase in bacterial load (Fig. 8k) and exacerbated pulmonary pathology (Fig. 8l).

Altogether, these findings indicate that the RPSA-OLFM4 axis governs neutrophil infiltration into tissues, thereby tuning host anti-infection immunity. This positions RPSA and OLFM4 as promising therapeutic targets for augmenting neutrophil-mediated anti-infective responses.

The RPSA-OLFM4 axis regulates neutrophil migration in sepsis patients

Impaired neutrophil migration and subsequent immune dysregulation are pivotal factors in sepsis pathogenesis and progression^34–38. To investigate the role of the RPSA-OLFM4 axis in sepsis, we collected peripheral blood samples from intensive care unit (ICU) patients diagnosed with sepsis (n = 34) and healthy control donors (n = 17) from a physical examination center (Fig. 9a). The demographic and clinical characteristics of these subjects are summarized in Supplementary Table 1, and their comorbidities in Supplementary Table 2. Flow cytometry analysis revealed significantly reduced proportions of RPSA⁺ neutrophils in sepsis patients compared to healthy controls (median: 63.7% vs 81.0%, Supplementary Fig. 5g, Fig. 9b), whereas the proportions of OLFM4⁺ neutrophils were markedly increased (median: 14.3% vs 7.6%, Supplementary Fig. 5g, Fig. 9c). Furthermore, RPSA and OLFM4 expression in neutrophils was inversely correlated (Fig. 9d). Consistent with our results in Rpsa^−/− mice, ex vivo migration assays identified that the migratory capacity of neutrophils derived from sepsis patients was significantly impaired compared to those from healthy controls (Fig. 9e). Notably, treatment with an OLFM4 agonist suppressed the migration of neutrophil derived from healthy controls (Fig. 9f), while treatment with OLFM4 inhibitor B11 restored the migratory function of neutrophils from sepsis patients (Fig. 9g). Thus, these results suggest that RPSA-OLFM4 dysregulation underlies neutrophil migration defects in human sepsis.

Fig. 9. Neutrophil migration impairment in human sepsis is mediated by the RPSA-OLFM4 axis.

a Peripheral blood samples from sepsis patients and healthy control donors were collected for flow cytometry and ex vivo Transwell migration assays. (Created in BioRender. Wu, T. (2026) https://BioRender.com/cl8hh2f). b, c Representative plots showing the RPSA⁺ (b) and OLFM4⁺ neutrophils (c) in peripheral blood. The data represent positive cell ratios. d Correlation analysis of RPSA⁺ and OLFM4⁺ neutrophils. Scatter plots with linear regression: healthy cohort and sepsis cohort. e Transwell migration assay of human neutrophils: neutrophils from healthy (CellTrace^TM Violet-labeled) and sepsis (CellTrace^TM CFSE-labeled) donors were mixed at a 1:1 ratio (input), then subjected to a Transwell migration assay as described above. Transmigrated cells in the lower chambers were quantified (output). Data represent migrated neutrophil ratios (n = 6). f Effects of ATRA + Aza on the migration of healthy human neutrophils. Methodology was identical to that described in Fig. 6d, with Rpsa^fl/fl mouse neutrophils replaced with neutrophils from healthy controls. Data represent migrated neutrophil ratios (n = 6). g Rescue effects of B11 on the migration of sepsis patient neutrophils. Methodology was identical to that described in Fig. 6e, with Rpsa^−/− mouse neutrophils replaced with neutrophils from sepsis patients. Data represent migrated neutrophil ratios (n = 6). Data were represented as mean ± SD and analyzed using two-tailed unpaired Student’s t-test (b, c, e-g). Pearson’s correlation coefficient test (d), the coefficient of determination (R²) was used to assess the goodness of fit, with p < 0.05 indicating a statistically significant correlation. A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

OLFM4 inhibition confers therapeutic efficacy in murine sepsis models

To further validate OLFM4 as a therapeutic target for sepsis, we employed a cecal ligation and puncture (CLP) model to induce polymicrobial sepsis (Fig. 10a). OLFM4 inhibitor B11 administration following CLP significantly improved survival rates of mice (Fig. 10b), and ameliorated infection symptoms (Fig. 10c) and body weight loss (Fig. 10d). Treatment with B11 markedly attenuated sepsis-induced lung pathological damage (Fig. 10e). These results indicate that targeting OLFM4 ameliorates sepsis in vivo.

Fig. 10. OLFM4 inhibitor B11 rescues sepsis pathology in CLP mice.

a WT mice underwent CLP (ligation at 50% cecal length) to induce sepsis. Mice in the treatment group received B11 at 2 h and 12 h post-CLP; controls received the same amounts of saline (Created in BioRender. Wu, T. (2026) https://BioRender.com/a9uqwbz). b–d Survival curves after CLP (n = 8/group) (b), clinical scores (n = 3) (c), and body weight loss (n = 3) (d), were monitored for indicated times. e Representative H&E-stained lung sections at 24 h post-CLP. Scale bar, 100 μm. f Summary diagram illustrating the proposed mechanism by which the RPSA-OLFM4 axis regulates uropod formation and transvascular migration of host neutrophils following S. suis 2 infection (Created in BioRender. Wu, T. (2026) https://BioRender.com/kntcgkk). Data were represented as mean ± SD and analyzed using two-way ANOVA with Bonferroni’s post hoc test (c, d), log-rank (Mantel-Cox) test (b). A p-value ≤ 0.05 was considered statistically significant; ns indicates no significant difference between the compared groups. Source data are provided as a Source Data file.

Discussion

In this study, we unveil a previously uncharacterized RPSA-OLFM4-MYH9 axis in neutrophils that governs neutrophil migration during bacterial infections. We demonstrate that RPSA deficiency severely compromises host defense against S. suis 2 by impairing neutrophil infiltration into tissues. Mechanistically, RPSA ablation induces OLFM4 overexpression, which suppresses the RhoA/ROCK1/pMLC2/MYH9 signaling pathway and results in MYH9 translocation from the uropod to the cytosol in migrating neutrophils (Fig. 10f). These alterations ultimately disrupt uropod protrusion, cytoskeletal polarization and abrogate neutrophil migration. Therapeutic targeting of this axis (e.g., via OLFM4 inhibition) rescues neutrophil migration and improves outcomes in both S. suis 2 infection and polymicrobial sepsis models. This study not only elucidates the mechanism of RPSA in neutrophil migration, expanding the immunological functions of RPSA, but also provides therapeutic targets for treating bacterial infections and neutrophil-associated inflammation.

RPSA is a typical multifunctional protein that may have divergent roles across cell types and species¹⁰. Our previous study identified that S. suis 2 infection results in significantly elevated RPSA⁺ in neutrophils and murine brain tissue, leading to greater bactericidal capacity²⁰. Nevertheless, the immunological functions and mechanisms of RPSA in neutrophils remain poorly characterized. Analogous to pattern recognition receptors (PRRs), whose expression increases on immune cells during bacterial infection to facilitate antigen recognition, initiation of downstream inflammatory pathways, and subsequent pathogen clearance³⁹, we observed similar RPSA upregulation on myeloid-derived immune cells. Jiang et al. determined that nuclear RPSA functions as a sensor for viral nucleic acids in macrophages²¹. We further propose that pathogen stimulation may trigger dynamic RPSA re-localization between cytosol and membranes, enabling antigen sensing and participation in innate immune regulation. Further studies are required to determine if RPSA operates similarly to PRRs.

Here, we demonstrate that neutrophil RPSA regulates antibacterial defense by affecting their migration. Specifically, RPSA suppresses OLFM4 expression to maintain RhoA/ROCK1/pMLC2/MYH9-mediated uropod functionality, promoting uropod formation and driving cellular motility. RPSA ablation traps neutrophils in peripheral circulation, preventing their tissue infiltration. Thus, RPSA and OLFM4 represent critical targets for controlling infection and inflammation. Furthermore, using LPS-induced scrotal inflammation models, we show RPSA deficiency similarly impairs neutrophil recruitment to inflammatory sites. Thus, RPSA has an impact beyond S. suis 2 infection to include Gram-negative bacterial infections and even sterile inflammation. Critically, our work reveals that RPSA regulates neutrophil migration independent of its receptor function—distinct from its roles in structural cells. While the release of chemokines from the host significantly influences the intensity of neutrophil migration and extravasation during the early stages of infection, our study found that RPSA deficiency does not affect the ability of neutrophils to sense chemokine concentration gradients. Moreover, at the transcriptional level, the absence of RPSA did not alter the expression of neutrophil-derived chemokines or their corresponding receptors. However, whether RPSA deficiency alters host chemokine production, thereby inhibiting neutrophil recruitment, warrants further in-depth investigation.

The RhoA/ROCK1/pMLC2 pathway, which is predominantly localized to the uropod, is a critical regulator of cellular motility, modulating uropod contraction^40–43. Previous studies have shown that inhibition of this pathway results in uropod elongation and arrest of adhesion-mediated migration^44–46. In contrast, we observed significantly shortened uropod and suppressed migration in RPSA-deficient neutrophils. In fact, beyond leading-edge traction that propels cells forward, the uropod generates propulsion forces during migration. Neutrophil traction forces concentrate predominantly at the uropod, exceeding those at the leading edge, which appear primarily during directional “decision-making”⁴⁷. This rear-driven traction may facilitate movement through cytosolic pressurization, propelling cytoplasmic contents forward⁴⁸. The RhoA/ROCK1/pMLC2 pathway regulates neutrophil polarization, which is essential for chemotaxis. RPSA deficiency inhibits RhoA/ROCK1/pMLC2 signaling, compromises uropod formation, and ultimately abrogates neutrophil migration. Thus, our study provides compelling evidence for uropod protrusion as the driving force underlying neutrophil migration.

MYH9 localizes to leading-edge lamellipodia and uropod in migrating neutrophils, with uropod enrichment correlating with rapid motility. In mice, dysregulated MYH9 expression significantly reduces neutrophil migration speed, distance, and tissue extravasation post-infection⁷. Both RPSA deficiency and OLFM4 upregulation suppress MYH9 expression and trigger its translocation from the uropod to the cytosol, enhancing OLFM4-MYH9 co-localization. In human gastric precancerous lesion (PLGC) cell models, MYH9 is a direct binding partner of OLFM4, which engages MYH9 via its glycine-48 residue to promote PLGC cell proliferation and invasion³³. Consistent with the latter human study, we demonstrate OLFM4-MYH9 association in mouse neutrophils. However, the precise relationship among RPSA, OLFM4, and MYH9 remains elusive. Both RPSA and OLFM4 exhibit notably diffuse cytoplasmic localization rather than enrichment in migration-associated structures (uropod/leading edges). The connection between their subcellular distribution and uropod functional alterations demands further investigation.

This study establishes a previously unreported link between OLFM4 expression and impaired neutrophil migration in sepsis. While OLFM4 has a role in cancer cell migration^49–53, no evidence has identified links with neutrophil migration. Elevated serum OLFM4 levels and increased OLFM4⁺ neutrophil proportions correlate with a poor prognosis in septic patients^54–56. Notably, OLFM4 deficiency enhances resistance to infection-induced sepsis and inflammatory injury in mouse models, positioning OLFM4 as a promising target for augmenting antibacterial immunity^57,58. The precise mechanisms underlying OLFM4-mediated immunoregulation, however, were previously undefined. Here, we demonstrate for the first time that OLFM4 suppresses neutrophil migration—a key mechanism contributing to adverse infection outcomes. Analysis of peripheral blood neutrophils from septic patients in the ICU revealed significantly expanded OLFM4⁺ neutrophil populations concomitant with markedly impaired migratory capacity. Furthermore, we identified a significant reduction in RPSA⁺ neutrophils in these patients, implicating that dysregulation of the RPSA-OLFM4 pathway in sepsis induces migration defects. Therapeutic intervention with the OLFM4 inhibitor successfully restored migratory function in patient-derived neutrophils and improved survival rates in murine sepsis models, further validating the therapeutic potential of targeting this axis. While the absence of specific OLFM4-targeted compounds necessitated the use of established modulators (ATRA/Aza, BAY-11-7082) with pleiotropic effects, our core conclusion—that OLFM4 impairs neutrophil migration—is robustly supported by direct genetic evidence. OLFM4 overexpression recapitulated the migration defect, and its knockdown rescued the phenotype in RPSA-deficient cells, findings confirmed both in vitro and in vivo. The consistent results from pharmacological interventions provide complementary support. Future development of specific OLFM4-targeted compounds or genetic models will be crucial to precisely define the therapeutic potential of this axis in sepsis.

Interestingly, our study observed an apparent discrepancy: RPSA⁺ neutrophils were upregulated in a mouse model of S. suis 2 infection but downregulated in septic patients. A plausible explanation lies primarily in the distinct disease stages and immune contexts captured in the two settings. Our murine model examined a defined, early time point post-infection—an active immune activation stage in which, consistent with our prior observation of increased RPSA⁺ neutrophils in the brain tissue of S. suis 2-infected mice²⁰. This early phase reflects the host’s initial innate immune mobilization and provides the rationale for investigating the fundamental role of RPSA in neutrophil function during active infection. In contrast, clinical sepsis patients are typically enrolled at mid-to-late disease stages, often characterized by a state of immunological dysregulation, immunosuppression, or functional exhaustion^59–62. Confounding factors such as advanced age and underlying comorbidities further establish an immune baseline fundamentally distinct from that of healthy laboratory mice. In this clinical context, the downregulation of RPSA may reflect immune dysfunction or cellular exhaustion within the setting of sustained systemic inflammation. Despite these context-dependent regulatory trends, our finding remains consistent across both systems: RPSA protein levels positively correlate with neutrophil migratory capacity, and its absence or reduction directly impairs neutrophil recruitment. This indicates that while upstream regulatory signals may vary depending on timing, host status, and pathogenic context, the functional role of RPSA as a key intrinsic regulator of neutrophil migration is conserved.

To explore the potential of RPSA as a therapeutic target for regulating neutrophil migration, this study applied anti‑RPSA IgG both in vivo and in vitro, observing its significant inhibitory effect on neutrophil migration. This antibody strategy primarily targets RPSA that may be exposed or translocated to the cell surface under inflammatory conditions^20,63, rather than directly affecting its intracellular ribosome‑associated function, indicating a role for cell surface RPSA in the migration regulatory mechanism. However, the increase in RPSA⁺ neutrophils after infection clearly does not match the proportion of neutrophil infiltration suppressed by RPSA deficiency, suggesting that RPSA‑mediated regulation of neutrophil migration is not solely dependent on its cell surface form. Intracellular RPSA likely also plays an indispensable role, with both pools cooperating to ensure the integrity of this regulatory function. Although antibody treatment still inhibited migration in a laminin‑free Transwell migration system, implying that its effect may be independent of classical receptor‑ligand interactions, the complexity of RPSA’s multi‑domain structure and the potential steric hindrance introduced by the antibody preclude a strict delineation of the precise mechanism underlying its surface function. Thus, while this study confirms the essential regulatory role of RPSA in neutrophil migration, the relative functional weighting of its intracellular versus extracellular pools and the detailed molecular mechanisms involved require further elucidation through future investigations, such as rescue experiments using subcellular localization‑specific mutants.

Under physiological RPSA expression, OLFM4 knockdown alone in HL-60 cells fails to augment migration-associated pathways. To date, OLFM4 research predominantly focuses on pathological overexpression in malignant contexts, where targeted interference effectively reverses phenotypic outcomes. For example, in uterine cancer models, OLFM4 mediates suppression of migration and invasion, which is reversed by OLFM4 knockdown⁵¹. Whereas in head/neck squamous carcinomas, OLFM4 enhances cell motility, which is abrogated by its knockdown⁶⁴. We propose a threshold-dependent model for OLFM4-mediated regulation of uropod function and migration: in non-malignant HL-60 cells, RPSA constrains OLFM4 to low physiological levels (distinct from RPSA knockout-induced upregulation), permitting unimpaired basal migration. Consequently, OLFM4 knockdown under these conditions yields negligible pro-migratory effects. However, when OLFM4 exceeds pathological thresholds (via ectopic expression or RPSA ablation), subsequent OLFM4 knockdown restores migration by reducing expression toward subthreshold levels, thereby rescuing functionality of migration-associated pathways. Thus, OLFM4 manifests as a regulator of migration polarization exclusively under pathological overexpression conditions.

In summary, we redefine RPSA as a master immunoregulator that maintains effective neutrophil migration during pathogen challenge by suppressing OLFM4 expression. In the context of RPSA deficiency, aberrant OLFM4 disrupts the RhoA/ROCK1/pMLC2/MYH9 signaling pathway, inhibits uropod protrusion and migration, and traps neutrophils in peripheral circulation. The net result is ultimately compromised anti-pathogen defense. The conserved dysregulation of this pathway in human sepsis, coupled with its therapeutic tractability, underscores RPSA-OLFM4 as a pivotal actionable target for neutrophil-centric immunomodulation. While our study delineates the RPSA-OLFM4-RhoA/ROCK1/pMLC2/MYH9 regulatory axis, the precise molecular mechanism by which RPSA suppresses OLFM4 expression, as well as the exact biochemical interface through which OLFM4 inhibits the downstream RhoA/ROCK1/pMLC2 pathway, remains to be fully elucidated. Future investigations employing advanced techniques such as spatial proteomics and structure-function analysis will be essential to map these detailed interactions. Our work uncovers a crucial role for neutrophil anti-infective responses through OLFM4-dependent migration, while identifying promising therapeutic targets for enhancing anti-infective immunity and sepsis treatment.

Methods

Mice

C57BL/6 mice (8–12 weeks old) were supplied by Liaoning Changsheng Biotechnology Co., Ltd. C-NKG mice (8 weeks old, Catalog C001316, NOD.Cg background) were purchased from Cyagen (Jiangsu, China). Lyz2^IRES-iCre mice (Stock No.: C001358, C57BL/6 background) were purchased from Cyagen. Genotyping was performed by PCR using the following primers: For mutant allele: forward, 5′-CCCAGAAATGCCAGATTACG-3′; reverse, 5′-CTTGGGCTGCCAGAATTTCTC-3′. For wild-type allele: forward, 5′-TTACAGTCGGCCAGGCTGAC-3′; reverse, 5′-CTTGGGCTGCCAGAATTTCTC-3′.

Mice carrying a floxed allele of Rpsa (Rpsa^fl/fl, C57BL/6 background) were also obtained from Cyagen. To generate myeloid lineage-specific Rpsa knockout mice (Rpsa^−/−, C57BL/6 background), Rpsa^fl/fl mice were crossed with Lyz2^IRES-iCre mice, in which Cre recombinase expression is driven by the lysozyme 2 (Lyz2) promoter. Mice homozygous for the Rpsa mutation were viable, fertile, normal in size, and exhibited no overt physical or behavioral abnormalities. PCR genotyping of the Rpsa floxed allele was carried out with the following primers: Forward: 5′-GCAACGTCAGACACAAACAAAATAG-3′; Reverse: 5′-GACAAATTTATCCGAACACAGAGCA-3′. Cre recombinase was detected using the same primers as those for the Lyz2^IRES-iCre mice.

Mice treatment

To assess bacterial loads, clinical scores, and weight changes after infection, mice (female, 8–10 weeks old) were challenged as follows: for S. suis 2, mice received an intraperitoneal (i.p.) injection of 1 × 10⁸ CFU suspended in 200 μL of sterile 0.9% normal saline (NS); for Klebsiella pneumoniae strain KP35, infection was performed via intranasal administration of 1 × 10⁴ CFU in 30 μL of NS. Survival rates were determined separately using higher challenge doses: mice (female, 8–10 weeks old) infected with S. suis 2 were injected i.p. with 5 × 10⁸ CFU (approximately the median lethal dose) in 200 μL NS, whereas mice (female, 8–10 weeks old) infected with Klebsiella pneumoniae strain KP35 received 5 × 10⁴ CFU intranasally in 30 μL NS.

Sepsis was induced via cecal ligation and puncture (CLP). Briefly, mice (female, 8–10 weeks old) were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and underwent a midline laparotomy to expose the cecum. The distal 50% of the cecum was ligated with a 6–0 silk suture and punctured twice with a 21-gauge needle. A small amount of fecal content was gently extruded into the peritoneal cavity. The cecum was then returned to the abdomen, and the incision was closed in layers. Postoperative supportive care included subcutaneous administration of pre-warmed saline (1 mL per 20 g body weight) with free access to food and water.

To stimulate neutrophil recruitment into the peritoneum, mice (female, 8–10 weeks old) were injected i.p. with 1 × 10⁷ CFU of S. suis 2 in 200 μL NS. For subcutaneous neutrophil migration, a ~ 0.5 cm incision was made on the exposed back skin of anesthetized mice (female, 8–10 weeks old) using a sterile scalpel. A S. suis 2 bacterial suspension (1 × 10⁹ CFU/mL) was applied to the wound with a sterile cotton applicator.

For the scrotal inflammation model, anaesthetized mice (male, 8–10 weeks old) were injected intrascrotally with 300 ng LPS from Escherichia coli 055:B5 (Solarbio) in 100 μL NS; control animals received NS alone. To visualize capillaries, FITC-conjugated anti-mouse CD31 antibody (#FITC-65058, clone 390, 1:200, Proteintech, Wuhan, China) was co-injected. Ninety min after LPS challenge, PE-conjugated anti-mouse Ly-6G antibody (#127607, clone 1A8, 10–15 μg, BioLegend, San Diego, CA, USA) in 200 μL NS was administered (intravenously, i.v.) to label neutrophils. Thirty min later, cremaster muscles were surgically excised, fixed in ice-cold 4% paraformaldehyde (PFA) for 30 min, and mounted smoothly on slides using neutral resin. Imaging was performed on a confocal laser-scanning microscope (Nikon AXR Ti2-E, Japan) with a 100×/1.45 NA oil-immersion objective. Images were analyzed using NIS-Elements Viewer v5.21 (Nikon) and ImageJ v1.54 g (National Institutes of Health).

In adoptive transfer experiments, purified neutrophils from Rpsa^fl/fl and Rpsa^−/− mice (male, 10–12 weeks old) were stained with CellTrace™ Violet or CellTrace™ CFSE (Thermo Fisher Scientific, Waltham, MA, USA) for 30 min at 37 °C, respectively, and then transferred (1 × 10⁸ neutrophils per mouse in 200 μL saline, i.v.) 30 min before i.p. S. suis 2 challenge. For in vivo competitive migration assays, stained Rpsa^fl/fl and Rpsa^−/− neutrophils were mixed at a 1:1 ratio and injected (0.5–3 × 10⁷ neutrophils per mouse in 200 μL NS, i.v.). To assess the effect of surface RPSA expression on neutrophil migration, we blocked peripheral RPSA in mice (female, 8–10 weeks old) with anti-RPSA polyclonal antibody (500 μg, anti-RPSA IgG, purified by rabbit anti-RPSA serum) or an isotype control IgG. Antibodies were administered (i.v.) before S. suis 2 challenges (i.p.). To assess the effect of OLFM4 expression on neutrophil migration, mice (female, 8–10 weeks old) received daily injections of OLFM4 agonizts (ATRA 20 mg/kg + Aza 10 mg/kg) or OLFM4 inhibitor B11 (20 mg/kg) via i.p. for 5 days before S. suis 2 challenges, respectively.

Clinical scoring system: Six parameters (0–3 points each) assessed at indicated timepoints: mental status (0 = alert, 3 = lethargic), mobility (0 = normal, 3 = immobile), coat condition (0 = smooth, 3 = ruffled), ocular discharge (0 = none, 3 = severe), food intake (0 = normal, 3 = anorexic), fecal status (0 = formed, 3 = diarrhea). Total score range: 0–18.

Microbial strain

The virulent S. suis 2 strain JZLQ022 was isolated from the brain tissue of pigs with meningitis (Shandong Binzhou Animal Science & Veterinary Medicine Academy), and cultured in brain heart infusion (BHI) medium (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) with 5% newborn bovine serum (Clark Bioscience, Richmond, VA, USA) at 37 °C. The Klebsiella pneumoniae strain KP35, a clinical isolate obtained from the sputum of an infected patient at the First Hospital of Jilin University and maintained in our laboratory, was cultured in Luria-Bertani (LB) medium (Becton, Dickinson and Company) with 5% newborn bovine serum (Clark Bioscience) at 37 °C.

Cell culture and differentiation

The HL-60 (SCSP-5210) and bEnd.3 (SCSP-5267) cell lines were obtained from the National Collection of Authenticated Cell Cultures. bEnd.3 cells were cultured in high-glucose Dulbecco’s Modified Eagle Medium (DMEM; Gibco, New York, NY, USA), and HL-60 cells were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco). Both media were supplemented with 10% fetal bovine serum (Clark Bioscience, Richmond, VA, USA) and 1% penicillin–streptomycin (10,000 U/mL penicillin and 10,000 μg/mL streptomycin; Sigma-Aldrich, MO, USA). Human and murine primary neutrophils were isolated and cultured in RPMI 1640 medium containing 10% FBS and 1% penicillin–streptomycin. All cells were maintained at 37 °C in a humidified incubator with 5% CO₂. One h before experiments, the medium was replaced with serum-free and antibiotic-free base medium. To induce neutrophil-like differentiation, HL-60 cells were treated with 1.3% dimethyl sulfoxide (DMSO) for 5 days.

Flow cytometry

To analyze immune cell infiltration, tissues (brain, lung, liver, spleen, and kidney) were weighed, dissected into small fragments, and digested for 30 min at 37 °C under continuous agitation in RPMI 1640 medium containing 2 µg/mL collagenase type IV (Gibco) and 0.2 µg/mL DNase I (Thermo Fisher Scientific). The digestion was quenched with RPMI 1640 containing 2% bovine serum albumin (BSA; Solarbio, China). Tissue fragments were mechanically dissociated using a syringe plunger and filtered through a 70 μm nylon cell strainer. Brain cell suspensions were further purified over a 30% Percoll PLUS (Cytiva, USA) gradient to remove myelin debris. Erythrocytes were lysed using ACK lysis buffer (Gibco). Cell viability was assessed with the Zombie NIR™ Fixable Viability Kit (BioLegend) for 30 min at 4 °C. Peripheral blood samples were collected, subjected to erythrocyte lysis, and processed for staining. All single-cell suspensions were used for flow cytometric analysis.

For antibody staining, 40 µL of antibody cocktail was added to each sample and incubated for 30 min at 4 °C. Cells were then washed twice with staining buffer. The following antibodies were used: PE anti-human CD45 (#304007, clone HI30, 1:400), Brilliant Violet 421™ anti-mouse/human CD11b (#101236, clone M1/70, 1:400), APC/Cyanine7 anti-human CD66b (#305125, clone G10F5, 1:400), PerCP/Cyanine5.5 anti-mouse CD45 (#103132, clone 30-F11, 1:100), Brilliant Violet 605™ anti-mouse CD45 (#103155, clone 30-F11, 1:100), FITC anti-mouse/human CD11b (#101205, clone M1/70, 1:800), PE anti-mouse Ly-6G (#127607, clone 1A8, 1:300), Brilliant Violet 510™ anti-mouse Ly-6C (#128033, clone HK1.4, 1:100), PE/Cyanine7 anti-mouse CD11c (#117317, clone N418, 1:400), APC anti-mouse F4/80 (#123115, clone BM8, 1:100), Brilliant Violet 650™ anti-mouse MHC II (#107641, clone M5/114.15.2, 1:100), PerCP/Cyanine5.5 anti-mouse CD3 (#100217, clone 17A2, 1:100), PE anti-mouse CD4 (#100407, clone GK1.5, 1:400), PE/Cyanine7 anti-mouse CD8a (#100722, clone 53–6.7, 1:300), APC anti-mouse CD11a Antibody (#101119, clone M17/4, 1:200), PE anti-mouse Ly-6G/Ly-6C (Gr-1) (#108407, clone RB6-8C5, 1:200), Brilliant Violet 421™ anti-mouse CD184 (CXCR4) Antibody (#146511, clone L276F12, 1:200), Brilliant Violet 510™ anti-mouse/human CD11b Antibody (#101245, clone M1/70, 1:200), FITC anti-mouse Ly-6G Antibody (#127605, clone 1A8, 1:200) and APC anti-mouse CD19 (#115511, clone 6D5, 1:400) (all above from BioLegend), PE-Cyanine7 CD101 Monoclonal Antibody (#25-1011-82, clone Moushi101, 1:200, Invitrogen).

RPSA was detected using a rabbit polyclonal anti-RPSA antibody (#14533-1-AP, Proteintech, 1:200), followed by DyLight™ 649 donkey anti-rabbit IgG (#406406, BioLegend, 1:200) or FITC donkey anti-rabbit IgG (#406403, BioLegend, 1:200). For OLFM4 staining, cells were incubated with a monoclonal anti-OLFM4 antibody (#MA5-29457, clone 12, 1:200, Invitrogen) and then stained with PE/Cyanine7 anti-mouse IgG1 (#406613, BioLegend, 1:200).

A fixed volume of each stained cell suspension was used for flow cytometry. Absolute cell counts per gram of tissue or per mL of blood were calculated based on the initial sample volume. Data were acquired on an 11-color DxFLEX flow cytometer (Beckman Colter Biotechnology Co., Ltd, CA, USA) and analyzed using FlowJo v10.8.1 (BD Biosciences).

Immunofluorescence staining

The assessment of neutrophil infiltration was performed according to previously described methods^65,66. Tissue sections were fixed and subjected to antigen retrieval using antigen repair solution (Servicebio, China). After permeabilization and blocking of nonspecific sites with 5% BSA in 0.3% PBST for 1 h at room temperature, sections were incubated with anti-Ly6G rabbit polyclonal antibody (#GB11229, Servicebio, 1:200) for 12–16 h at 4 °C, followed by incubation with Cy3-conjugated goat anti-rabbit IgG (H + L) (#GB21303, Servicebio, 1:100) for 2 h at room temperature. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Sections were mounted using ProLong™ Gold Antifade Mountant (Invitrogen, Grand Island, NY, USA) and scanned using a Pannoramic MIDI II slide scanner (3DHISTECH, Hungary). Images were acquired with SlideViewer v2.4 software (3DHISTECH).

For primary neutrophil staining, cells were seeded onto 14 mm coverslips (Solarbio) at a density of 0.5–1 × 10⁵ cells per well and stimulated with 100 ng/mL mouse CXCL1 (KC, Peprotech, Rocky Hill, NJ, USA), which is a potent neutrophil chemoattractant, and 10 nM N-formylmethionyl-leucyl-phenylalanine (fMLP; Sigma-Aldrich) in serum-free medium for 30 min. Cells were fixed with 4% paraformaldehyde (Solarbio) for 30 min at room temperature, permeabilized with 0.5% Triton^TM X-100 (Sigma-Aldrich) for 30 min, and blocked with 10% goat serum for 30 min. Subsequently, cells were incubated overnight at 4 °C with OLFM4 polyclonal antibody (#PA5-115687, 1:200, Invitrogen) or MYH9 monoclonal antibody (#60233-1-Ig, clone 5D9D2, 1:200, Proteintech), followed by incubation with Alexa Fluor 488–conjugated anti-mouse IgG (#4408S, 1:200, Cell Signaling Technology) or Alexa Fluor 647–conjugated anti-rabbit IgG (#4414S, 1:200, Cell Signaling Technology) for 1 h at room temperature. Nuclei were stained with Hoechst 33342 (1:1000, Sigma-Aldrich) for 5 min, and F-actin was labeled with TRITC-phalloidin (100 nM, Yeasen Biotechnology) for 30 min. After washing with PBS, coverslips were imaged using a Nikon AXR confocal laser-scanning microscope (Ti2-E model) with a 100×/1.45 NA oil immersion objective. Image analysis was performed using NIS-Elements Viewer v5.21 (Nikon) and ImageJ v1.54 g (National Institutes of Health).

Neutrophil depletion

For neutrophil depletion, mice (female, 8–10 weeks old) were injected i.p. with anti-mouse Ly6G (BE0075-1, Bio X Cell, USA) or control IgG (BE0089, Bio X Cell, USA) antibodies (500 µg/mouse) 24 h and 0.5 h before the S. suis 2 injections^67–70. To avoid potential epitope masking from using the same clone (1A8) for depletion and detection, neutrophils were stained with a non-competing antibody PE anti-mouse Ly-6G/Ly-6C (Gr-1) (#108407, clone RB6-8C5, 1:200, BioLegend), together with PerCP/Cyanine5.5 anti-mouse CD45 (#103132, clone 30-F11, 1:100, BioLegend) and FITC anti-mouse/human CD11b (#101205, clone M1/70, 1:800, BioLegend), and identified as CD45⁺CD11b⁺Gr-1^hi cells by flow cytometry^70,71.

Mice primary neutrophil isolation

Primary murine bone marrow neutrophils were isolated using a Histopaque-based density gradient centrifugation method. Briefly, 10- to 12-week-old male mice were euthanized by cervical dislocation and sterilized by immersion in 75% ethanol for 3–5 min. Bone marrow cells were flushed from femurs and tibias using D-Hank’s Balanced Salt Solution (D-HBSS) containing 0.5% FBS and passed through a 70 μm nylon cell strainer. Erythrocytes were lysed using ACK lysis buffer.

A discontinuous density gradient was prepared by carefully overlaying 3 mL of Histopaque 1077 (density: 1.077 g/mL, #10771, Sigma-Aldrich) onto 3 mL of Histopaque 1119 (density: 1.119 g/mL, #11191, Sigma-Aldrich) in a 15 mL conical tube. The bone marrow cell suspension was then gently layered on top of the Histopaque 1077 layer and centrifuged at 872 × g for 30 min at room temperature with the brake turned off. Neutrophils were collected from the interface between the Histopaque 1077 and 1119 layers.

The harvested neutrophils were washed twice with RPMI 1640 medium supplemented with 10% FBS and 1% penicillin–streptomycin, followed by centrifugation at 400 × g for 7 min at room temperature. After the final wash, the cell pellet was resuspended in an appropriate buffer (e.g., NS or RPMI 1640). A small portion of this final cell suspension was taken to determine the neutrophil count and viability using trypan blue exclusion. Another aliquot was used to assess purity by flow cytometry staining for Ly6G and CD11b. The remaining neutrophils were then used for subsequent experiments.

Migration assay

In vivo and ex vivo competitive cell migration assays were based on methods described by Marco De Giovanni et al.⁷².

For the in vivo neutrophil extravasation assay, a co-transfer experiment was conducted. Primary neutrophils or DdHL-60 cells were stained as described above, mixed at a 1:1 ratio (input), and injected (i.v.). Thirty min later, S. suis 2 or chemokines was administered i.p. At specified time-points post-injection (i.p.), recruited neutrophils or DdHL-60 in the peritoneal cavity were collected by lavage with 2 mL of ice-cold sterile saline, followed by gentle peritoneal massage to dislodge adherent cells. The lavage fluid was aspirated and centrifuged at 500 × g for 10 min at 4 °C. The cell pellet was resuspended and analyzed by flow cytometry (output).

For the ex vivo migration assay, Transwell chambers (6.5 mm insert, 5 μm pore size, Corning, USA) were used. Primary neutrophils or DdHL-60 cells were stained and mixed at a 1:1 ratio as indicated in the text. Where applicable, cells were pretreated prior to staining with either the OLFM4 agonist (1 μM ATRA + 5 μM Aza for 6 h), OLFM4 inhibitor (100 μM B11 for 6 h), or anti-RPSA IgG (100 μg/mL for 1 h). Cells were resuspended in RPMI 1640 medium containing 0.5% FBS at a density of 2 × 10⁶ cells/mL. A total of 600 μL of chemoattractant solution (100 ng/mL KC and 10 nM fMLP or 100 ng/mL human IL-8 (PeproTech) and 10 nM fMLP in serum-free medium) was added to the lower chamber. Then, 100 μL of cell suspension (2 × 10⁵ cells) was seeded into the upper chamber. After incubation for 3 h at 37 °C, migrated cells in the lower chamber were collected and quantified by flow cytometry.

Soluble ICAM-1- and fibrinogen-binding assay

The soluble ICAM-1- and fibrinogen-binding assays were performed as previously described⁷³. Neutrophils were stimulated with CXCL1 (100 ng/ml; 10 min, 37 °C) or left unstimulated in the presence of ICAM-1/Fc (20 μg/ml, R&D Systems). Cells were then incubated with APC-conjugated anti-human IgG1 (#9042-11, clone H2, 1:100, Fc-specific, Southern Biotechnology) at 4 °C for 30 min. LFA-1-specific binding to ICAM-1/Fc was measured by flow cytometry (Beckman Colter Biotechnology). To investigate Mac-1 affinity to fibrinogen, isolated murine neutrophils were incubated with 150 μg/mL Murine Fibrinogen (MedChemExpress, Monmouth Junction, NJ, USA) and stimulated with CXCL-1 (100 ng/mL, 10 mins, 37 °C) or were left unstimulated. Subsequently, cells were incubated with a Fibrinogen Polyclonal Antibody (#PA5-141128, 1:100, Invitrogen) at 4 °C for 30 min, followed by staining with Alexa Fluor 488–conjugated anti-mouse IgG (1:200, Cell Signaling Technology) at 4 °C for 30 min. Fluorescence intensity was measured by flow cytometry (Beckman Colter Biotechnology) and analyzed using FlowJo v10.8.1 (BD Biosciences).

Plasmids and transfection

Overexpression plasmid OLFM4-3×FLAG-CMV-10 was synthesized by Sangon Biotech Co., Ltd (Shanghai, China). OLFM4 shRNA plasmid (CMV.G&PR.U6.shRNA) and RPSA shRNA plasmid (pLKO.1-TRC) were constructed by Sangon Biotech Co., Ltd. Plasmids were transfected into cells using Lipofectamine 2000 (Invitrogen, Grand Island, NY, USA), according to the manufacturer’s instructions.

Low-density agarose gel model

Ex vivo neutrophil migration tracking was performed according to the method described by Xu Wang et al.⁷⁴. A 1.2% agarose solution (Sigma-Aldrich) was heated to complete dissolution and mixed in a 1:3 ratio with prewarmed (68 °C) RPMI 1640 basal medium supplemented with 50% Hanks’ Balanced Salt Solution (containing Ca²⁺ and Mg²⁺; Thermo Fisher Scientific) and 20% FBS. Approximately 3 mL of the mixture was poured into a 35 mm culture dish (Corning, USA) and allowed to solidify at room temperature. After gelation, two wells (3.5 mm in diameter, 2.4 mm apart) were punched in the center of the gel. One well was filled with neutrophils (1 × 10⁵ cells). After 10 min of incubation at 37 °C to allow cell adhesion, nonadherent cells were removed and replaced with serum-free RPMI 1640 medium. The opposite well was filled with chemoattractant solution (100 ng/mL KC and 10 nM fMLP in serum-free medium). The dish was immediately transferred to an XD-202 Inverted Biological Microscope (Novel Optics, China), and time-lapse imaging was performed over 1 h with images captured at 30-s intervals. Cell migration tracks were analyzed using ImageJ v1.54 g (National Institutes of Health). Single-cell migration trajectories were generated with the Chemotaxis and Migration Tool v2.0 (ibidi, Germany).

Neutrophil adhesion and migration under physiological flow

Ex vivo flow chamber assays was performed according to the method described by J van Buul et al^75,76. bEnd.3 were cultured in a fibronectin-coated ibidi m-slide VI^0.4 (ibidi, Munich, Germany) the day before the experiment and stimulated overnight with TNF-a (10 ng/ml). Freshly isolated neutrophils were resuspended at 1 × 10⁶ cells/ml in RPMI 1640 medium and were incubated for 30 min at 37 °C. Cultured bEnd.3 in ibidi flow chambers were connected to an ibidi Pump System (ibidi) and exposed to RPMI 1640 medium shear flow for 10 min (1 dyne/cm²). Neutrophils were subsequently injected into the perfusion system, and real-time neutrophil-endothelial interactions were recorded for 30 min at intervals of 10 s and recorded in several random fields by an EVOS^® FL Auto Cell Imaging System (Thermo Fisher Scientific) using a 40 × NA 1.3 objective and Celleste 5.0 Image Analysis Software. All live imaging was performed at 37 °C in the presence of 5% CO2. Cell migration tracks were analyzed using ImageJ v1.54 g (National Institutes of Health). Ibidi Pump System was controlled by ibidi PumpControl software v1.6.2.

Morphological detection of neutrophils

Primary neutrophils were seeded onto 14 mm coverslips (Solarbio) at a density of 0.5–1 × 10⁵ cells per well and stimulated with 100 ng/mL KC and 10 nM fMLP in serum-free medium for 30 min. Cells were then washed twice with ice-cold PBS and fixed with 4% glutaraldehyde (Solarbio) for 24 h at 4 °C. After fixation, samples were rinsed twice with cold PBS and post-fixed with 0.5% osmium tetroxide for 1 h, followed by three washes with ice-cold PBS. The fixed cells were dehydrated through a graded ethanol series (70%, 80%, 90%, and 100% ethanol, each step for 2 min), subjected to critical point drying, and sputter-coated with gold. Morphological features of migrating neutrophils were examined using a Sigma 300 scanning electron microscope (ZEISS, Germany).

RNA extraction and RT-qPCR analysis

Total RNA was extracted from isolated primary neutrophils and HL-60 cells using the RNAsimple Total RNA Kit (TIANGEN, China). cDNA was synthesized with the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). Quantitative real-time PCR (RT-qPCR) was performed using SYBR Premix Ex Taq (Takara, Tokyo, Japan) on a QuantStudio 1 Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific). The comparative Ct (2^-ΔΔCt) method was used for relative quantification of gene expression, with Gapdh serving as the internal reference control. All primers were designed and synthesized by Sangon Biotech (Shanghai, China). The sequences are listed below:

Mouse primers

Rhoa: 5’-GTCTATGTGCCCACGGTGTTTG-3’ and 5’-CTGTGTCCCATAAAGCCAACTCTAC-3’; Rock1: 5’-CAAGGCGGTGATGGCTATTATGG-3’ and 5’-ACTGTAAGTCCCAACCAAAGAATCTG-3’; Myl2: 5’-GACCAGATGTTCGCAGCCTTTC-3’ and 5’-TCCTTCTCTTCTCCGTGGGTAATG-3’; Cdc42: 5’-AACAAACAGAAGCCTATTACTCCAGAG-3’ and 5’-GCAGAGCACTCCACATACTTGAC-3’; Rac2: 5’-AAGTGTGTGGTGGTGGGTGATG-3’ and 5’-GGGAAGGCGTTGGTGGTGTAG-3’;Myh9: 5’-TCAAGAACAAGCACGAGGCAATG-3’ and 5’-AGCAATCTGGTCACTGAGGTCTG-3’; Olfm4: 5’-ACAACTCTGACCTCTGCACATTCC-3’ and 5’-GCTGAGACACAGTCCCACCTTC-3’; Cxcl1: 5’-CATGGCTGGGATTCACCTCA-3’ and 5’-CCTCGCGACCATTCTTGAGT-3’; Cxcl2: 5’-CCACCAACCACCAGGCTACAG-3’ and 5’-TGGCAGGGTCTTCAGGCATTG-3’; Cxcr2: 5’-GTTCTGCTACGGGTTCACACTG-3’ and 5’-CAAGGACGACAGCGAAGATGAC-3’; Fpr1: 5’-GTATCTGCTGGCTACATCGTTCTG-3’ and 5’-CAAGTAAGAGATGGTGGTGACAGTG-3’; Gapdh: 5’-ACGGCAAATTCAACGGCACAG-3’ and 5’-ACACCAGTAGACTCCACGACATAC-3’.

Human primers

RHOA: 5’-AGGTGGATGGAAAGCAGGTAGAG-3’ and 5’-AGTATAACATCGGTATCTGGGTAGGAG-3’; ROCK1: 5’-AAGAAGCCAATGACTTACTTAGGACAG-3’ and 5’-ACTGACTAATTGACTTGCTCATCTCTG-3’; MYL2: 5’-GGGTGCTGAAGGCTGATTACG-3’ and 5’-GCGGCGAACATCTGGTCAAC-3’; CD11b: 5’-GGCCTCAGAGAATACCAGT-3’ and 5’-GGGTGAACACGGAATCGTT-3’; OLFM4: 5’-AACCATCTGCTTCTAACGCCTTC-3’ and 5’-GCCCTCTTTCCCTGTGTTTGTG-3’; GAPDH: 5’-AGTCCACTGGCGTCTTCACC-3’ and 5’-TGATCTTGAGGCTGTTGTCATACTTC-3’.

Western blotting

Total protein was extracted from cells using RIPA Lysis and Extraction Buffer (Thermo Fisher Scientific) supplemented with 1 mM protease inhibitor cocktail (Roche Diagnostics, IN, USA) and 1 mM Phosphatase Inhibitor Cocktail II (MedChemExpress, NJ, USA). Protein concentration was determined with a BCA Protein Assay Kit (Thermo Fisher Scientific). Equal amounts of protein were denatured in SDS sample buffer, separated by 12% SDS-PAGE, and transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore, MA, USA). The membrane was blocked by 5% skimmed milk and then incubated overnight at 4 °C with the following primary antibodies: ROCK1 Recombinant Rabbit Monoclonal Antibody (#ET1609-53, clone ST05-19, 1:5000, HUABIO, China), RhoA Recombinant Rabbit Monoclonal Antibody (#ET1611-10, clone SN0612, 1:2000, HUABIO), Myosin Light Chain 2 Polyclonal Antibody (#10906-1-AP, Proteintech, 1:5000), OLFM4 Polyclonal Antibody (#PA5-115687, Invitrogen, 1:2000), MYH9 Polyclonal Antibody (#11128-1-AP, Proteintech, 1:5000), rabbit polyclonal anti-RPSA antibody (#14533-1-AP, Proteintech, 1:4000), Phospho-Myosin Light Chain 2-S19 Rabbit mAb (#AP1433, clone ARC62046, 1:1000, Abclonal, China), GAPDH Rabbit Monoclonal Antibody (#A19056, clone ARC50888, 1:10000, ABclonal), β-Actin Rabbit Monoclonal Antibody (#AC026, clone ARC5115-01, 1:10000, ABclonal).

After washing, the membrane was incubated with HRP-conjugated Goat anti-Rabbit IgG (H + L) (#AS014, ABclonal, 1:10000) for 1 h at room temperature. Protein bands were visualized using an ECL chemiluminescent detection system (Tanon, China) and quantified with ImageJ v1.54 g (National Institutes of Health).

Bulk RNA-seq analysis

Primary neutrophils were isolated as described above. RNA was extracted from three biological replicates per group using the RNAsimple Total RNA Kit (TIANGEN, China) according to the manufacturer’s instructions. RNA quality was assessed using a Quawell Q6000 spectrophotometer (Quawell Technology Inc., USA). Sequencing libraries were constructed and sequenced on an Illumina NovaSeq™ 6000 platform (LC Bio Technology, China). Differential gene expression analysis was performed using the Cufflinks package and cuffmerge, with significance thresholds set at |fold change| > 2 and p < 0.05. Bioinformatic analysis was executed using the OmicStudio tools accessible at https://www.omicstudio.cn/tool. Visual representations such as volcano plots were generated utilizing the R programming language on the OmicStudio platform (https://www.omicstudio.cn/tool).

Co-Immunoprecipitation (Co-IP)

Immunoprecipitation assays were performed using the Pierce™ Classic Magnetic IP/Co-IP Kit (ACE Biotechnology, China) according to the manufacturer’s protocol. Briefly, primary neutrophils were washed twice with cold PBS and lysed in ice-cold IP Lysis/Wash Buffer for 5 min on ice with intermittent vortexing. Cell lysates were collected and incubated overnight at 4 °C with either MYH9 Monoclonal antibody (#60233-1-Ig, clone 5D9D2, 0.5–4 μg per 1–3 mg of total protein, Proteintech) or normal mouse IgG control (Proteintech, #B900620) to form immune complexes. Protein A/G magnetic beads were washed with IP Lysis/Wash Buffer and then added to the lysate–antibody mixtures, followed by incubation at room temperature for 1 h with gentle rotation. Beads were collected magnetically, and the unbound fraction was retained for further analysis. The beads were washed gently with IP Lysis/Wash Buffer and ultrapure water. Bound antigens were eluted with Elution Buffer, separated by SDS-PAGE, and detected by western blotting.

Human study subjects

Sepsis Patient Cohort: A total of 37 patients diagnosed with sepsis, according to the Sepsis-3 criteria⁷⁷, admitted to the Intensive Care Unit (ICU) of the First Hospital of Jilin University between May 2025 and July 2025, were prospectively enrolled. Peripheral blood samples were collected from all participants. Inclusion criteria comprised: age ≥ 18 years, confirmation of sepsis diagnosis, and ongoing systemic antimicrobial therapy. Exclusion criteria were as follows: age <18 years, pregnancy, known autoimmune diseases, active malignancy, immunosuppressive therapy or hematopoietic stem cell transplantation within the preceding 3 months, incomplete clinical records, a white blood cell (WBC) count <1000/mm³ (due to technical limitations in neutrophil isolation and flow cytometry under such conditions), inability to process samples for flow cytometry within 48 h (mostly within 24 h) of collection, and chronic ventilator-dependent respiratory failure.

Healthy Control Cohort: Peripheral blood samples from 17 healthy volunteers undergoing routine health examinations at the Physical Examination Center of the First Hospital of Jilin University during the same period were collected as controls. Healthy controls were free of acute or chronic infectious symptoms, had no history of major systemic or immune-related disorders, no recent history of vaccination, and no significant abnormalities on physical examination.

Human blood sample collection and processing

Collection: Peripheral venous blood from all subjects was collected into EDTA-anticoagulated vacuum tubes.

Patient sample characterization

Patient Sample Characterization: Samples were obtained from sepsis patients receiving antimicrobial therapy during hospitalization as part of clinical care, with no restriction on timing relative to treatment initiation.

Processing

Collected blood samples were processed immediately (≤ 2 h post-collection) at room temperature. Neutrophil RPSA and OLFM4 expression were quantified in RBC-lysed whole blood using a CD45⁺CD11b⁺CD66b⁺ gating strategy. Neutrophils were isolated by density gradient centrifugation using a human peripheral blood neutrophil separation solution reagent kit (Haoyang Huake Biotechnology, Tianjin, China). Isolated neutrophils were washed and used immediately for downstream migration assays.

Statistics & reproducibility

Data are expressed as mean ± SD. All data are from at least two to three independent experiments. For histological analyses (H&E staining) and immunofluorescence experiments, representative images are shown from at least three independent biological replicates per group, each performed with similar results. The number of humans, animals or samples in each experimental group is shown in the legend of each figure. Statistical analysis was performed using GraphPad Prism software v9.5.0 (GraphPad Software, La Jolla, CA). The normal distribution of data was assessed by the Shapiro-Wilk test. For comparisons between two groups, an unpaired two-tailed Student’s t-test (for parametric data) or Mann-Whitney U-test (for non-parametric data) was used. For comparisons among more than two groups, one-way ANOVA followed by Tukey’s post hoc test (for parametric data) or Kruskal-Wallis test followed by Dunn’s post hoc test (for non-parametric data) was applied. Two-way ANOVA followed by Bonferroni’s post hoc test was used to analyze experiments with multiple groups and two independent variables. The survival rates of mice were analyzed by the log-rank (Mantel-Cox) test. Pearson’s correlation coefficient test was performed to analyze the colocalization and correlation. A p-value ≤ 0.05 was considered statistically significant, ns indicates no significant difference between the compared groups. All statistics details, including the statistical test used, the exact value of n and statistical significance, are reported in the figure legend. For the reproducibility of the results, all the experiments were repeated independently with similar results.

Ethical statement

All human-related studies were conducted in accordance with the ethical guidelines of the Declaration of Helsinki and were approved by the Ethics Committee of the First Hospital of Jilin University (Approval Number: AF-IRB-032-07). Written informed consent was obtained from all participants or their legal guardians before sample collection. The sex of participants was determined based on medical records, and sex was not considered in the study design and analysis.

All animal experiments were performed in strict compliance with the Guidelines for the Ethical Review of Laboratory Animal Welfare (National Standard GB/T 35892-2018 of the People’s Republic of China). The protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Jilin University (Approval Numbers: SY202201009, SY202506040). Mice were housed under controlled conditions (20–22 °C, 50% humidity, 12 h/12 h light–dark cycle) with free access to food and water. Animals were maintained on a C57BL/6 or NOD.Cg background. Age- and sex-matched mice were used in each experiment. Sex was not considered a biological variable in the study design and analysis. All procedures adhered to relevant ethical regulations for animal testing and research.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

L.L. and N.L. conceived the project and supervised the study; T.W. performed all flow cytometry, immunofluorescence staining, neutrophil isolation, western blotting, RT-qPCR, migration assay, electron microscopy studies, cell culture, statistical analyses, and preparation of figures with assistance from X.Y., S.L. and S.W.; T.W. executed most in vitro and ex vivo experiments with assistance from X.Y., W.W., Y.S. and K.Z.; X.Y., W.W., Y.S., X.J., Y.T., K.Z. and Y.Z. assisted mice-related experiments; T.W. and H.C. performed all human clinical blood sample analyses under the supervision of H.J., N.L., F.L. and J.H. T.W. wrote an original manuscript. All authors provided intellectual input and edited the manuscript. L.L. supervised all aspects of the work.

Peer review

Peer review information

Nature Communications thanks Rance Berg and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The bulk RNA-seq raw data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) under accession code SRP658079. All data that support the findings of this study are included within the article, its Supplementary information and the Source Data file. The raw numbers for all graphs presented in the figures are available in the Source Data file. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-72365-0.