bioRxivImmunologyPreprint
Lipocalin-2-expressing red pulp macrophages link ischemic stroke to systemic immune reprogramming in mice and humans
Alexandra LucaciuiD, Patrick WurzeliD, Stine Rytz RasmusseniD, Eva Lueckhoff, Franziska Mayser, Jiffin BenjaminiD, Roxane‐Isabelle Kestner, Verena K. HaasiD, Lisa Sophie Huber, Deepthi Bevara, Nico KrausiD, Melanie Glueck, Karen R. GertziD, Raphael RaspeiD, Sethuraman SubramanianiD, Christoph WelschiD, Julia BeiniD, Peter Johannes WildiD, Helena RadbruchiD, Christian GrefkesiD, Adam StrzelczykiD, Waltraud PfeilschifteriD, Michael H. SiewekeiD, Josef M. PfeilschifteriD, Julien SubburayaluiD, Rajkumar VutukuriiD
Abstract
Background Acute ischemic stroke (AIS) triggers profound systemic immune alterations that contribute to post-stroke infections. The spleen is a key peripheral immune organ affected by AIS, but the molecular mediators linking cerebral ischemia to splenic immune dysfunction remain incompletely understood. Lipocalin-2 (LCN2) is an acute-phase protein implicated in neuroinflammation and host defense, yet its peripheral regulation, cellular sources and immune-modulatory actions after AIS remain unclear. Methods Male, 10-12-week-old C57BL/6J mice underwent transient middle cerebral artery occlusion (tMCAO) or sham surgery. Spleens were collected 24 hours, 3 days, or 7 days post-stroke for 3-MACE-Seq transcriptomics, multiplex cytokine profiling, RT-qPCR, immunoblotting, and immunofluorescence. Murine T cells and monocytes were magnetically isolated, stimulated with recombinant LCN2 protein or vehicle, and analyzed using (i) RT-qPCR of LPS-induced proinflammatory cytokines (Il1b, Il6, Tnf), (ii) LCN2-driven T cell chemotaxis, (iii) FACS-based phenotyping, (iv) FACS-based phagocytosis of pHrodo-labeled E. coli bioparticles, and (v) FACS-based quantification of reactive oxygen species (ROS) using DHR-123. Human plasma samples and post-mortem spleens from three patients with ischemic stroke and one non-stroke control were analyzed for LCN2 expression and cellular localization. Untouched human monocytes from healthy donors were exposed to recombinant LCN2 and assessed by flow cytometry for antigen-presentation, co-stimulatory markers, phagocytosis and ROS production. Results Transcriptomics revealed robust, biphasic splenic Lcn2 upregulation at 24 hours and 7 days after tMCAO. LCN2 localized predominantly to F4/80+ red pulp macrophages (RPMs), and accumulated at RPM-associated CD3+ contact sites. AIS induced a broad suppression of splenic innate and adaptive cytokine networks. Recombinant LCN2 induced T cell hyporesponsiveness, suppressed Il6 and Tnf induction, and impaired CCL5-directed chemotaxis. In monocytes, LCN2 promoted a shift toward Ly6Clow phenotypes, enhanced phagocytosis of E. coli, and uncoupled phagocytosis from ROS production. In human ischemic stroke spleens, LCN2 was detected in CD68+ RPMs and at RPM-associated immune-cell contact sites, consistent with the murine splenic phenotype. In line with the murine findings, human monocytes exposed to LCN2 displayed increased E. coli uptake but reduced ROS, alongside reduced HLA-DR, CD80, CD206, and CD11b expression. Conclusions AIS induces a conserved splenic LCN2-expressing RPM niche in mice and humans. Recombinant LCN2 is sufficient to reprogram T cells and monocytes toward hyporesponsive, tolerogenic and phagocytosis-ROS-uncoupled states, thereby linking cerebral ischemia to systemic immune dysfunction. These findings identify LCN2-expressing RPMs as a translationally relevant immunoregulatory niche after AIS and nominate receptor-specific LCN2 signaling as a candidate pathway for future studies aimed at mitigating post-stroke immunodepression and infection susceptibility without compromising antimicrobial host defense.
- organism: mouse
- organism: human
- assay: bulk RNA-seq
- assay: imaging
- research focus: differential expression
- research focus: cell-cell communication
- research focus: cell type annotation
- approach: experimental assay
- approach: statistical model
In-depth summary
Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (5 Oct 2026) · CC BY 4.0
Splenic Lcn2 is induced biphasically after mouse stroke and localizes to red pulp macrophages; recombinant LCN2 induces T cell and monocyte hyporesponsiveness and monocyte phagocytosis-ROS uncoupling, with human spleen support.
Post-stroke immunodepression raises infection risk, but the molecular link between cerebral ischemia and splenic immune dysfunction is unclear. Circulating LCN2 is a known stroke biomarker, yet it cannot resolve cellular source or tissue niche; this work places splenic LCN2 in a macrophage niche and tests whether it is sufficient to reprogram immune cells.
Key results
ShownAcute (24 h) stroke caused broad splenic transcriptional reprogramming in mice, with Lcn2 among the top up-regulated transcripts, confirmed at mRNA and protein level.
3'-MACE-Seq, tMCAO vs sham, n = 3 per group; PC1 47.81% and PC2 14.93% of variance; 347 up and 564 down genes (adjusted P < 0.001, log2FC > |2.0|). RT-qPCR confirmed Lcn2 up-regulation (P < 0.001); immunoblot showed higher LCN2 protein. · Transcriptomic profiling reveals robust splenic…; Figure legends
ShownSplenic Lcn2 is biphasic: high at 24 h, reduced at day 3, high again at day 7, within a four-signature temporal transcriptome pattern.
Day 3 spleens sat closer to sham on PC1 while 24 h and day 7 spleens separated from sham. Lcn2 mRNA and protein dipped at day 3 and rose by day 7 (RT-qPCR n = 4-6 per group, Mann-Whitney). Pseudotime analysis placed 24 h and day 7 at later states. · Temporal splenic transcriptomics reveal biphasic…; LCN2-expressing red pulp macrophages define a…; Figure legends
SuggestedIn mouse spleen LCN2 rose selectively in the red pulp and localized mainly to F4/80+ red pulp macrophages, which sat at close contact sites with CD3+ T cells; red pulp CD3+ T cells declined at days 3 and 7.
Immunofluorescence quantification, n = 3-6 mice per group with 3 ROIs per mouse (Kruskal-Wallis/Dunn or ANOVA/Tukey). A 40-plex cytokine array showed broad cytokine reduction (n = 2 per group, each pooled from four mice; no statistics). Contact sites were assessed qualitatively. · LCN2-expressing red pulp macrophages define a…; Figure legends
SuggestedRecombinant LCN2 made mouse splenic T cells and monocytes hyporesponsive, impaired CCL5-directed T cell migration, shifted monocytes toward Ly6C-low, and raised E. coli uptake while lowering ROS intensity.
T cells: reduced LPS-induced Il6 and borderline Tnf (n = 4, paired t test). Monocytes: reduced Il1b, borderline Tnf (P = 0.0513). Transwell with 5 µm pores (n = 6-7 technical replicates). Flow: n = 12 per group (6 technical replicates from two mice); ROS+ percentage unchanged, ROS MFI reduced. · LCN2 is sufficient to induce hyporesponsive and…; Figure legends
SuggestedHuman data parallel the mouse findings: plasma LCN2 was elevated in acute stroke, LCN2 sat in CD68+ red pulp macrophage-rich regions of stroke spleens, and LCN2 lowered antigen-presentation markers and ROS in healthy-donor monocytes.
Post-mortem spleens from three stroke patients plus one non-stroke reference (9 ROIs per patient, no formal stroke vs control statistics). Monocytes: HLA-DR, CD80, CD11b reduced, CD206 reduced on classical monocytes at higher doses, E. coli uptake increased, ROS MFI decreased in classical monocytes (n = 6 technical replicates). · Human ischemic stroke spleens show a conserved…; LCN2 induces tolerogenic and…; Image processing and quantitative analyses; Figure legends
Methods, data and reproducibility
Time-course 3'-MACE-Seq of mouse spleens after tMCAO, validated by RT-qPCR, immunoblot, cytokine array and immunofluorescence. Recombinant LCN2 was then tested in isolated mouse and human immune cells by RT-qPCR, transwell and flow cytometry.
- Spleen 3'-MACE-Seq, tMCAO (60 min occlusion) vs sham at 24 h, plus 3 d and 7 d time points · Mouse (male C57BL/6J, 10-12 weeks) · n = n = 3 per group at 24 h
- Mice used overall; analyzable animals after exclusions (24 h, 3 d, 7 d, sham) · Mouse (C57BL/6J) · n = 51 used; 9, 13, 12 and 8 analyzed
- Spleen RT-qPCR, immunoblot and immunofluorescence across time points · Mouse (C57BL/6J) · n = n = 4-6 per group (RT-qPCR); n = 3-6 mice (IF)
- 40-plex spleen cytokine array, sham vs tMCAO at three time points · Mouse (C57BL/6J) · n = n = 2 per group, each pooled from four mice
- Isolated splenic T cells and monocytes treated with recombinant LCN2 (cytokines, chemotaxis, phagocytosis, ROS) · Mouse (C57BL/6J) · n = n = 4 (RT-qPCR); n = 12 per group (flow, 6 technical replicates from two mice)
- Plasma LCN2 in acute stroke patients at onset, day 3, day 7, vs outpatient and healthy controls · Human
- Post-mortem spleens from ischemic stroke patients plus one non-stroke reference spleen · Human · n = 3 stroke + 1 non-stroke; 9 ROIs per patient
- Untouched monocytes from healthy donors (buffy coats) exposed to recombinant LCN2 · Human · n = n = 6 technical replicates per group, two independent donor experiments
- Methods & tools
- 3'-MACE-Seq (GenXPro, UMI-based), Cutadapt 4.6, FastQC 0.11.9, GSEA (Broad Institute), GSVA/ssGSEA module scoring, UMAP-based pseudotime, TaqMan RT-qPCR (delta-delta CT), QuPath, Python 3.11 (NumPy, scikit-image, OpenCV), napari v0.6.4, FlowJo, GraphPad Prism v9; R v4.5.0
- Reagents & environment
- Cutadapt 4.6, FastQC 0.11.9, Python 3.11, napari v0.6.4, GraphPad Prism v9, R v4.5.0.
- Reproducibility gaps
- Image-analysis scripts and processed quantification files available only on reasonable request. Sequencing differential expression results are in supplementary tables; no raw-data repository accession printed. UMI deduplication used in-house tools.
Limitations (6)
- Human spleen sample is small (three stroke patients, one non-stroke reference) and observational; post-mortem interval, severity, comorbidities, infection and treatment may confound LCN2 patterns, so causality cannot be inferred. · stated by the authors
- Human functional assays used healthy-donor monocytes with recombinant LCN2, which shows sufficiency but does not recapitulate the plasma and tissue milieu of stroke patients. · stated by the authors
- pHrodo E. coli uptake and DHR-123 ROS are surrogates; microbial killing was not measured, so effects on infection susceptibility remain untested (live-pathogen assays needed). · stated by the authors
- Receptor-specific roles of 24p3R and LRP2 are unresolved, and necessity of LCN2 signaling for post-stroke immune dysfunction was not tested. · stated by the authors
- Experiments used the tMCAO model in young, healthy male C57BL/6J mice, not capturing aged, comorbid or hemorrhagic stroke heterogeneity. · stated by the authors
- Ex vivo mouse and human assays rest largely on technical replicates from two mice or two donor experiments, and the 24 h sequencing used three mice per group, limiting biological inference. · noted in this summary
Read this if you study post-stroke immunodepression or splenic neuro-immune crosstalk and want evidence for LCN2-expressing macrophages as a candidate mediator in mice and humans.
- Citations
- 0
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- S100A8/A9 Inhibition Reduces Splenic Myelopoiesis and Improves Outcomes After Stroke
Hyun Ah Kim, Annas Al‐Sharea, … Man K.S. Lee·Frontiers in ImmunologyPeer-reviewed·Feb 17, 2026
- Endogenous protection from ischemic brain injury by preconditioned monocytes
Lidia García‐Bonilla, David Brea, … Josef Anrather·Journal of NeurosciencePeer-reviewed·Jun 26, 2018
- Alterations of the pulmonary immunity are independent of spontaneous pneumonia in an experimental model of ischemic stroke
B. Y. Farris, K. L. Monaghan, … E. C. K. Wan·bioRxiv·Apr 25, 2019
- An atypical role for the myeloid receptor Mincle in CNS injury
Thiruma Arumugam, Silvia Manzanero, … Christine Wells·Journal of Cerebral Blood Flow & MetabolismPeer-reviewed·Aug 19, 2016
- T cells modulate the microglial response to brain ischemia
C. Benakis, A. Simats, … A. Liesz·bioRxiv·Sep 23, 2021
- Microglia Piezo1 aggravates cerebral ischemia and reperfusion injury by chemotactic recruitment of T lymphocytes
Huinan Zhang, Hui-Min Chang, … Jing Huang·bioRxiv·Jan 13, 2026