# paper.pdf ## Metadata - PDFFormatVersion=1.4 - Language=en-US - IsLinearized=false - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260530091156Z - Creator=Adobe InDesign 18.3 (Windows) - Custom.CrossMarkDomains[1]=springer.com - Custom.CrossMarkDomains[2]=springerlink.com - Custom.CrossmarkDomainExclusive=true - Custom.CrossmarkMajorVersionDate=2010-04-23 - Custom.doi=10.1007/s12975-026-01451-8 - ModDate=D:20260602111340+02'00' - Producer=Adobe PDF Library 17.0; modified using iText® 5.5.13.4 ©2000-2024 iText Group NV (SPRINGER SBM; licensed version) - Trapped=False - crossmark:crossmarkdomains=springer.comspringerlink.com - pdfx:crossmarkdomains=springer.comspringerlink.com ## Contents ### Page 1 Introduction Ischemic stroke remains a leading cause of adult disabil-ity and mortality worldwide despite advances in acute careand secondary prevention strategies []. Following cere-bral ischemia, disruption of the blood–brain barrier (BBB)represents a pivotal pathophysiological event [] thatexacerbates secondary brain injury by increasing vascu-lar permeability, promoting inflammatory cell infiltration,and facilitating cerebral edema [,]. The BBB is a highlyspecialized interface between the circulating blood and thecentral nervous system, maintained by brain-endothelialtight junction proteins such as claudin-5, occludin, andzonula occludens-1 (ZO-1) [–]. BBB breakdown afterischemia occurs in a biphasic manner—an early disruptionwithin hours and a delayed phase spanning several days—allowing blood-borne molecules and inflammatory cellsto enter the brain parenchyma [,–]. Preserving BBBintegrity has, thus, emerged as a promising therapeutic Received: 21 January 2026 / Revised: 7 May 2026 / Accepted: 18 May 2026 © The Author(s) 2026 Communicated by Aditya Kumar. Hiroyasu Kidoya 1 Department of Integrative Vascular Biology, Faculty ofMedical Sciences, University of Fukui, 23-3 Matsuoka-Shimoaizuki, Eiheiji, Yoshida, Fukui 910-1193, Japan 2 Department of Neurosurgery, Division of Medicine, Facultyof Medical Sciences, University of Fukui, Fukui910-1193, Japan 3 Tenure-Track Program for Innovative Research, Universityof Fukui, Fukui 910-1193, Japan 4 Otothinolaryngology. Head & Neck Surgery, Faculty ofMedicine, University of Fukui, Fukui 910-1193, Japan # Abstract # Blood–brain barrier (BBB) disruption significantly exacerbates secondary injury following ischemic stroke. Althoughlysophosphatidic acid (LPA) is known to regulate vascular stability, its specific role in preserving BBB integrity duringcerebral ischemia remains unclear. Here, we investigated the neuroprotective effects of LPA using a mouse model of distalmiddle cerebral artery occlusion (dMCAO). Eight- to ten-week-old female C57BL/6 mice received 10 mg/kg LPA intra-peritoneally immediately after ischemia, and outcomes were assessed at 48 h after dMCAO via infarct volumetry, edemaquantification, BBB permeability assays (Evans blue and FITC-dextran), and transcriptomic analyses. We demonstrate thatLPA treatment significantly reduced infarct volume by approximately 60% and attenuated cerebral edema at 48 h post-ischemia; This protection was accompanied by preserved BBB integrity and maintained endothelial claudin-5 expressionin the ischemic territory. Single-cell analysis identified selective and sustained expression of the LPA4 receptor in ischemicvascular endothelial cells. Crucially, the protective effects of LPA on infarct size and BBB permeability were abolishedin LPA4 receptor-knockout mice. These findings indicate that LPA preserves BBB integrity and mitigates ischemic braininjury via an endothelial LPA4 receptor-dependent mechanism, identifying the LPA–LPA4 signaling axis as a promisingtherapeutic target for reducing secondary brain injury in ischemic stroke. Keywords Lysophosphatidic acid · Blood–brain barrier · Cerebral ischemia · LPA4 receptor · Tight junction · Vascularpermeability Lysophosphatidic Acid Reduces Ischemic Brain Injury by AttenuatingVascular Permeability Through LPA4 Receptor Signaling Shintaro Yamada1,2 · Kazuhiro Takara1,3 · Naoi Hosoe1 · Anna Shimizu1,4 · Yumiko Hayashi1 · Lamri Lynda1 ·Takayuki Sonoda1 · Kenichiro Kikuta2 · Hiroyasu Kidoya1 Translational Stroke Research (2026) 17:61 [ ![http://crossmark.crossref.org/dialog/?doi=10.1007/s12975-026-01451-8&domain=pdf&date_stamp=2026-5-30]() ](http://crossmark.crossref.org/dialog/?doi=10.1007/s12975-026-01451-8&domain=pdf&date_stamp=2026-5-30) ![Springer Logo]() ### Page 2 strategy to limit neuronal loss and improve functional out-comes after stroke. Lysophosphatidic acid (LPA) is a bioactive phospholipidmediator that signals through at least six G protein-coupledreceptors (LPA1–LPA6) to regulate diverse cellular func-tions, including proliferation, migration, and cytoskeletalrearrangements [–]. In the nervous system, LPA signal-ing influences neurogenesis, neuropathic pain, and neuro-inflammation [], and its effects in central nervous system(CNS) injury are highly context-dependent—varying withreceptor subtype, cell type, and temporal dynamics [,].Despite these established roles of LPA in CNS physiologyand injury, its specific effects on BBB integrity and strokeoutcomes, particularly in the context of cerebral ischemia,remain incompletely understood. LPA can modulate endothelial barrier function inperipheral tissues [–], and emerging evidence indi-cates that LPA receptor signaling contributes to vascularregulation and endothelial barrier stability. In particular,the LPA4 receptor regulates endothelial permeability andvascular integrity across multiple organ systems [–]. Because the BBB is formed by a specialized sub-set of endothelial cells that share core barrier-regulatorymechanisms with peripheral vasculature, these findingsraise the possibility that similar LPA4-dependent path-ways may operate in the cerebral endothelium. However,whether LPA4 regulates BBB permeability after strokehas not been well characterized. Furthermore, the spe-cific LPA receptor subtypes mediating these effects inthe cerebral vasculature remain unknown. Notably, priorstudies examining LPA signaling in cerebral microvas-cular endothelium and ischemia models have predomi-nantly reported increased permeability and detrimentaloutcomes, which are often attributed to LPA1 or theautotaxin–LPA axis [–]. These findings indicate thatLPA signaling in the brain vasculature is context-depen-dent and frequently disrupts the BBB, thereby makinga receptor-specific protective role mediated by LPA4 aparticularly novel and contrasting hypothesis. AlthoughLPA4 mRNA expression has been detected in brain tissue[,–], direct evidence of LPA4 expression specifi-cally in brain microvascular endothelial cells under nor-mal or ischemic conditions remains lacking. This gap inknowledge is a key obstacle to understanding the recep-tor-specific role of LPA signaling in BBB regulation andprovides a strong rationale for the present study. Based on these studies, we hypothesized that analo-gous LPA4-dependent mechanisms modulate BBB integ-rity after ischemic stroke. To test this hypothesis, weinvestigated the effects of exogenous LPA administrationon BBB integrity and stroke outcomes in a mouse modelof focal cerebral ischemia. Using a permanent distalmiddle cerebral artery occlusion (dMCAO) model, whichproduces consistent cortical infarcts with minimal mor-tality, we assessed infarct volume, behavioral recovery,BBB permeability, and vascular integrity markers fol-lowing LPA treatment. We further evaluated the role ofthe LPA4 receptor in mediating these effects using LPA4knockout mice. Through histological analysis, permeabil-ity assays, immunohistochemistry, and RNA-sequencing(RNA-seq), we aimed to determine whether LPA–LPA4signaling influences BBB stability and ischemic injury.Our findings reveal a previously unrecognized protec-tive function of LPA-LPA4 signaling in preserving BBBintegrity following cerebral ischemia, suggesting poten-tial therapeutic applications for modulating this pathwayin ischemic stroke. Methods Data Availability The data supporting the findings of this study are availablefrom the corresponding author upon reasonable request.RNA-seq datasets generated for this work will be depos-ited in the Gene Expression Omnibus (GEO) databaseupon acceptance of the manuscript. In addition, the pub-licly accessible dataset GSE261494 was utilized to assessthe expression of LPA receptors in the mouse cortex. Animals, Experimental Design, and ExclusionCriteria Adult female C57BL/6 mice (8–10 weeks old, 20–25 g,Japan SLC, Inc., Hamamatsu, Shizuoka, Japan) and LPA4receptor knockout mice on a C57BL/6 background wereused []. The experimental groups were as follows: (1)Sham + vehicle (wild-type [WT]), (2) dMCAO + vehicle(WT), (3) dMCAO + LPA (WT), (4) dMCAO + vehicle(LPA4 KO), and (5) dMCAO + LPA (LPA4 KO). All out-come measures were assessed in WT mice. In LPA4 KOmice, assessment was limited to Evans blue extravasa-tion and infarct volume measurement using 2,3,5-triphe-nyltetrazolium chloride (TTC) staining. A total of 138mice were initially included in this study, of which 14were excluded because they failed to develop infarctionon 2,3,5-triphenyltetrazolium chloride (TTC) staining(n = 3), exhibited abnormal behavior attributable to basalganglia involvement (n = 9), or died following Evansblue injection (n = 2); leaving 124 mice for final analy-ses. Animals were maintained under controlled housing ### Page 3 conditions (12 h light/dark cycle, 22 ± 2 °C, 55 ± 10%humidity) with ad libitum access to food and water. Allexperimental procedures were approved by the Institu-tional Animal Care and Use Committee of the Univer-sity of Fukui (approval number: R07013) and conductedin accordance with the ARRIVE guidelines (AnimalResearch: Reporting of In Vivo Experiments) for animalexperimentation. Distal Middle Cerebral Artery Occlusion (dMCAO)Model Focal cerebral ischemia was induced by permanent occlusionof the distal middle cerebral artery as previously described[]. Briefly, mice were anesthetized intraperitoneally with athree-drug mixture of medetomidine hydrochloride (0.3 mg/kg), midazolam (4 mg/kg), and butorphanol tartrate (5 mg/kg). Body temperature was maintained at 37 ± 0.5 °C using aheating pad throughout the surgery. After a midline skin inci-sion, a 2 × 2 mm craniotomy was created over the left mid-dle cerebral artery (MCA) territory. The dura was carefullyremoved, and the distal MCA was exposed and permanentlyoccluded using monopolar electrocoagulation (Low Tem-perature Cautery Kit, Clearwater, Florida, USA; #18019-00). Sham-operated animals underwent an identical surgicalprocedure without MCA occlusion. Animals were allowed torecover in a temperature-controlled chamber before return-ing to their home cages. Drug Administration LPA (18:1, 1-oleoyl-2-hydroxy-sn-glycero-3-phosphate,sodium salt, Avanti Polar Lipids, Alabaster, Alabama, USA;#857130P-25 mg) was dissolved in phosphate-bufferedsaline (PBS) to obtain a final concentration of 1 mg/mL.The dose of 10 mg/kg was selected based on prior stud-ies demonstrating the vascular-protective effects of LPAon mice [,] and based on preliminary dose-rangingexperiments testing 3, 5, and 10 mg/kg doses in our labo-ratory; among these concentrations, 10 mg/kg yielded themost consistent protective effect across outcome measures.The vehicle control consisted of PBS alone, administeredat the same volume to control for any solvent effects. LPAor vehicle was administered intraperitoneally at 10 mg/kg (200 μL per mouse) once immediately after dMCAO,twice within 24 h post-d-MCAO (between days 0 and 1),and then twice again between days 1 and 2. Specifically,the five injections were administered at approximately 0,12, 24, 36, and 48 h after dMCAO induction. All solutionswere freshly prepared before each experiment. Measurement of Infarct Volume Infarct volumes were assessed using TTC staining. Follow-ing induction of deep anesthesia, Animals were euthanizedby cervical dislocation at designated time points. Brains wererapidly removed and fully immersed in 2% TTC solution at37 °C for 30 min, followed by fixation in 4% paraformal-dehyde. After fixation, brains were sliced into 2-mm-thickcoronal sections, and images of the stained sections wereacquired using a digital camera (SX70HS; Canon, Tokyo,Japan) for quantitative analysis. Infarcted areas were mea-sured using ImageJ software (NIH, USA; version 1.54 g) byan investigator blinded to treatment groups. Infarct volumeswere calculated using the following formula: where h represents the slice width (2 mm), and S1, S2 are thecross-sectional areas of the infarct for consecutive slices []. Brain Water Content Brain edema was assessed by measuring brain water contentusing the wet-dry weight method. Following the induction ofdeep anesthesia, mice were euthanized 48 h after dMCAO,and the brains were quickly removed and dissected intoipsilateral and contralateral hemispheres. The wet weightof each hemisphere was immediately measured. The tissueswere then dried in a drying oven at 100 °C for 24 h to obtainthe dry weight. Brain water content was calculated based ona previous study using the following formula []: Behavioral Assessment Functional outcome was evaluated using the cylinder test toquantify forelimb use asymmetry. Mice were placed indi-vidually in a transparent cylinder (9 cm in diameter, 15 cmin height) and video recorded for 5 min. The number ofwall contacts made with the left forelimb (L), right forelimb(R), or both forelimbs simultaneously (both) was scored byan investigator blinded to the experimental groups. Limbpreference was calculated following a previously definedmethod using the following equation []: Positive values indicated left forelimb preference, and the neg-ative values indicated right forelimb preference. Assessments ### Page 4 were performed at baseline (before surgery) and on days 1, 3,and 7 after dMCAO. BBB Permeability Assessment BBB integrity was evaluated using two complementaryapproaches. For Evans blue extravasation (Miles assay),mice received an intravenous injection of Evans blue dye(4%, 4 mL/kg; FUJI FILM Wako, Osaka, Japan; #056–04061) 2 h before sacrifice. After euthanization by cervicaldislocation, brains were rapidly removed, and accumulationof dye was quantified spectrofluorometrically (excitation620 nm, emission 680 nm) and expressed as μg/g tissue[]. To evaluate macromolecular permeability, FITC-dextran (200 kDa, 10 mg/mL, 0.1 mL; Sigma-Aldrich, St.Louis, Missouri, USA; #FD2000s 100 mg) was adminis-tered intravenously 15 min before euthanasia. Subsequently,the animals were euthanized by cervical dislocation, andtheir brains were harvested immediately. After post-fixa-tion, brains were sectioned and imaged using fluorescencemicroscopy. FITC-dextran extravasation was quantified bymeasuring fluorescence intensity outside CD31-positivevessels using ImageJ software []. Immunofluorescence Staining Brain cryosections (40 μm) were blocked with 5% normalgoat serum in PBS containing 0.3% Triton X-100 for 1 hat room temperature, followed by overnight incubation at4 °C with primary antibodies against CD31 (1:200; Ratmonoclonal, BD Phamingen, San Diego, California, USA;#553370), claudin-5 (1:200; Rabbit polyclonal, ThermoFisher Scientific, Waltham, Massachusetts, USA; #34–1600), or MAP2 (1:500, Chicken polyclonal, abcam, Cam-bridge, England; #ab5393). After washing, sections wereincubated with appropriate fluorescent secondary antibod-ies for 2 h at room temperature. Images were acquiredusing a Leica Stellaris 5 (Leica Microsystems, Wetzlar,Germany) confocal microscope and analyzed with ImageJsoftware. For quantification of vessel density, the percent-age of CD31 or claudin-5 positive area relative to the totalarea of the region of interest (located between + 1.0 and− 1.0 mm relative to the bregma) was calculated using theAngio tool []. A minimum of three animals per groupwere included in quantitative analyses. For each animal,four non-overlapping fields of view were acquired fromthree consecutive brain sections. Flow Cytometry Brain microvascular endothelial cells (BMECs) wereisolated from ischemic brain tissue collected 48 h afterdMCAO. The isolation of BMECs, cell-surface antigenstaining, and flow cytometry were performed following pre-viously described methods. Briefly, the dissected brain tis-sue was enzymatically digested using collagenase/dispase(1 mg/mL) and DNase I (100 U/mL) at 37˚C for 30 min, fol-lowed by mechanical dissociation and density gradient cen-trifugation (Percoll, 37%/70%) to enrich for microvascularfragments. The resulting cell suspension was then stainedwith a viability dye and antibodies against CD31 and CD45.BMECs were identified and sorted as CD31 + CD45- livecells using a SONY SH800 Cell Sorter system (SONY,Tokyo, Japan) []. Live cells were identified using flowcytometry after staining with Zombie NIR Fixable Viabil-ity Kit (BioLegend, San Diego, California, USA; #423105)and labeled with antibodies against CD31 (Rat monoclonal,BioLegend; #102409) and CD45 (Rat monoclonal, BioLeg-end; #103108). Cells were sorted using a SONY SH800 cellsorter (SONY, Tokyo, Japan) and immediately processed forRNA extraction. RNA Extraction RNA was extracted from brain tissue and BMECs by lysingsamples in Buffer RLT, followed by thorough homogeniza-tion using a QIAshredder Spin Column (QIAGEN, Hilden,Germany), centrifuged at maximum speed (20,380 × g) for2 min. Total RNA was then purified using the RNeasy MicroKit and QIAcube system (QIAGEN) following the manu-facturer’s instructions. RNA concentration and RNA integ-rity number (RIN) were assessed using an Agilent 2100Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA).Only samples with a RIN > 7.0 were used for subsequentRNA -sequencing. RNA-seq and Bioinformatics Analysis RNA-seq data were analyzed using iDEP (ver2.5.2) () . Differentially expressedgenes (DEGs) were identified using DESeq with a maxi-mum false discovery rate-adjusted p < 0.05, and a 1.5-foldchange (FC) between groups [–]. The RNA-seq datahave been deposited in the GEO database (accession code:PRJDB39906). Gene Enrichment Analysis To characterize the functional pathways associated withtranscripts modulated by ischemic stroke and LPA treat-ment, two gene sets were defined: (i) genes upregulated afterdMCAO but downregulated following LPA administration,and (ii) genes downregulated after dMCAO but upregu-lated with LPA treatment. Gene set enrichment analysis was ### Page 5 performed using a comprehensive GMT-formatted collec-tion of curated gene sets derived from publicly availabledatabases, including Reactome, Kyoto Encyclopedia ofGenes and Genomes (KEGG, Homo sapiens), WikiPath-ways, Pathway Interaction Database, Integrative Networkof Organ System (INOH), BioCarta, and TarBase (miRNA-target gene sets) [–]. Re-analysis of Public Single-cell RNA-seq Data Single-cell RNA-seq datasets of the mouse brain were down-loaded from the GEO database (accession ID: GSE261494)and processed using Seurat (v5.0.0) in R (v4.3.2). Datasetswere integrated using the Harmony R package (v1.2.0) to cor-rect for batch effects while preserving biological variability.Uniform Manifold Approximation and Projection (UMAP)was applied for dimensionality reduction and visualization.Cell types were annotated based on canonical marker genesand guided by the classifications in the original publication.Major cell populations included endothelial cells (EC), neu-rons, oligodendrocytes (Oligo), microglia (MG), astrocytes(Astro), lymphocytes, epithelial cells, vascular leptomenin-geal cells, and venous endothelial cells []. Statistical Analysis All statistical analyses were pre-specified before unblind-ing. Data are presented as mean ± standard error of themean (SEM) with individual data points shown. Statisti-cal analyses were performed using GraphPad Prism soft-ware (version 9.5, GraphPad Software Inc., San Diego,CA). Normality was assessed using the Shapiro–Wilk test.Differences between the two groups were analyzed usingan unpaired two-tailed Student’s t-test. Multiple compari-sons were performed using one-way ANOVA followedby Tukey’s post-hoc test or two-way ANOVA followedby Bonferroni’s post-hoc test. Statistical significance wasdefined as p < 0.05. Results LPA Treatment Reduces Infarct Volume andImproves Behavioral Outcomes Following dMCAO To evaluate the neuroprotective potential of LPA in isch-emic stroke, we used a dMCAO model in mice. Follow-ing dMCAO induction, LPA was administered every 12 hfor a total of five doses. The overall experimental timeline,including the time points for various analyses, is detailed inFig.A. As illustrated in Fig.B, permanent occlusion ofthe distal MCA was achieved via craniotomy and electro-coagulation. This procedure consistently produced corticalinfarcts within the MCA territory, confirmed by TTC stain-ing. Brain sections demonstrated a clear distinction betweenthe ischemic core (white, TTC-negative) and surroundingviable tissue (red, TTC-positive) at 3 h (day 0), 24 h (day 1),and 48 h (day 2) post-occlusion (Fig.C–E). The dMCAOmodel also resulted in significant cerebral edema, evidencedby increased water content in the ipsilateral hemispherecompared with sham-operated controls (Fig.F). Next, weassessed the therapeutic effects of LPA administration. At48 h post-ischemia, TTC staining demonstrated that LPA-treated mice exhibited significantly reduced infarct volumescompared with PBS-treated control mice (Fig.G). Quanti-tative analysis indicated an approximately 60% reduction ininfarct size in the LPA group, supporting a robust neuropro-tective effect (Fig.H). To determine whether this anatomical protection trans-lated into functional improvement, we evaluated forelimbasymmetry using the cylinder test. Vehicle-treated dMCAOmice showed pronounced left forelimb preference (ipsi-lateral to the lesion) at all time points, reflecting impairedright forelimb function. In contrast, LPA-treated mice dis-played trends toward improved symmetry at days 1, 3, and7 post-ischemia (Fig.I, J). The cylinder test showed thatLPA treatment did not produce significant changes in fore-limb preference within the assessed period. Nevertheless,the observed trends toward improved forelimb symmetrysuggest possible functional benefits that may require lon-ger follow-up or complementary behavioral assays to detectthem more clearly. LPA Treatment Alters Gene Expression Profiles inIschemic Brain Tissue To investigate the molecular mechanisms underlying LPA-mediated neuroprotection, we performed bulk RNA-seqon ischemic cortical tissue collected from sham-operated,PBS-treated dMCAO, and LPA-treated dMCAO mice at48 h post-ischemia. Principal component analysis (PCA)revealed distinct transcriptional profiles across the threegroups (Fig.A), with Sham and ischemic samples sep-arated along PC1. In contrast, PBS- and LPA-treateddMCAO samples segregated along PC2, indicating thatLPA elicits marked gene expression changes within theinjured cortex. Differential expression analysis identified1,847 DEGs upregulated in PBS-treated dMCAO tis-sue relative to sham, with 40 of these genes overlappingwith those downregulated in the LPA-treated dMCAOgroup compared with the PBS control group (Fig.B).Conversely, among genes downregulated in PBS-treated ### Page 6 ### Page 7 dMCAO group versus sham group (1,473 genes), 84 over-lapped with those upregulated following LPA treatment.These reciprocal patterns highlight subsets of ischemia-responsive genes that LPA modulates. Volcano plot visual-ization further illustrated the magnitude and significance ofLPA-induced transcriptional changes (Fig.C), revealinga substantial number of genes exhibiting |log2(FC)| > 2 andadjusted p-values < 0.001. Gene ontology (GO) enrichmentanalysis of LPA-responsive DEGs demonstrated signifi-cant overrepresentation of biological processes involvedin vascular remodeling, inflammatory signaling, and celladhesion (Fig.D). Specifically, upregulated pathwaysincluded blood vessel morphogenesis, angiogenesis, cellmigration, and extracellular matrix organization, whereasdownregulated pathways were enriched for inflammatoryresponses, leukocyte trafficking, and cytokine-mediatedsignaling. Together, these findings indicate that LPA mod-ulates multiple transcriptional programs associated withBBB stabilization and post-ischemic tissue repair. LPA Treatment Preserves BBB Integrity AfterCerebral Ischemia Given the contribution of BBB disruption to secondaryinjury following stroke and the transcriptional changesobserved in vascular-related pathways, we next exam-ined whether LPA treatment alters BBB permeability afterdMCAO. At 48 h post-ischemia, the Miles assay demon-strated significantly reduced Evans Blue dye leakage in theipsilateral hemisphere of LPA-treated mice compared withthat in PBS-treated control mice (Fig.A, B), indicatingattenuation of BBB disruption. To more precisely assess BBB permeability, FITC-dex-tran extravasation was examined in brain sections followingdMCAO. Vehicle-treated dMCAO mice exhibited promi-nent FITC-dextran leakage throughout the ischemic terri-tory. Fluorescent signals were detected in the parenchymaoutside CD31-positive vessels (Fig.C; white arrowheads).LPA treatment markedly reduced FITC-dextran leakage,and quantitative analysis confirmed significantly reducedextravasation in the ischemic core compared with controltreatments (Fig.D). To further investigate the relationship between micro-vascular protection and neuronal preservation in theperi-infarct (penumbra) region, we performed dual immu-nofluorescence staining for FITC-dextran and MAP2at 48 h post-ischemia. In PBS-treated mice, extensiveFITC-dextran leakage coincided with areas of diminishedMAP2 immunoreactivity in the penumbra (Fig.E), con-sistent with compromised neurovascular unit integrity.In contrast, LPA-treated mice exhibited substantiallyreduced FITC-dextran extravasation together with betterpreserved MAP2 staining. Quantification of the area inwhich dextran signal overlapped with MAP2-positive tis-sue showed a significant reduction in LPA-treated mice(Fig.F), indicating that LPA preserves vascular bar-rier function in regions containing viable neuronal tis-sue. Collectively, these findings demonstrate that LPAtreatment reduces BBB hyperpermeability in both theischemic core and the penumbra, supporting a broad pro-tective effect on the neurovascular unit following cere-bral ischemia. LPA Treatment Maintains Tight Junction ProteinExpression in Peri-infarct Regions Because BBB integrity relies heavily on endothelial tightjunctions, we examined whether the protective effects ofLPA involve preservation of tight junction protein expres-sion in the peri-infarct region. Immunofluorescence stain-ing for the endothelial marker CD31 and the tight junctionprotein claudin-5 was performed at 48 h post-ischemia.Compared with those in sham controls, CD31-positivevessels in PBS-treated dMCAO mice exhibited markedlyreduced claudin-5 immunoreactivity, consistent with tightjunction disruption following ischemic injury. In contrast,LPA-treated mice displayed robust claudin-5 staining alongCD31-positive vessels, indicating preservation of tightjunction integrity, with no significant differences when com-pared with sham controls (Fig.A). Fig. 1 LPA treatment reduces infarct volume in a permanent dMCAOstroke model (A) Schematic diagram of the time course of LPA admin-istration and analysis in a model of permanent dMCAO using C57BL/6and LPA4KO mice. (B–E) Characterization of the dMCAO model. (B)Schematic of the dMCAO model showing craniotomy site (white box)and coagulation point (asterisk). (C) Representative images of isch-emic (+) and non-ischemic (-) hemispheres with white areas indicatingischemic core and red areas representing normal tissue. (D) Progres-sive development of the ischemic region at 3 h (day 0), day 1, and day2 post-dMCAO. (E) Quantification of stroke volume (mm³) over timefollowing dMCAO. (F) Brain water content (%) comparison betweensham-operated and dMCAO groups. (G) Representative images com-paring ischemic regions in PBS-treated versus LPA-treated brainsfrom multiple viewing angles. (H) Quantitative comparison of isch-emic core volume (mm³) between PBS and LPA-treated groups. (I)Cylinder test schematic for assessing forelimb motor function. L rep-resents left forelimb contacts, R represents right forelimb contacts,and “both” indicates simultaneous use of both forelimbs. (J) Cylindertest results showing limb preference percentages on days 1, 3, and 7post-stroke in sham, PBS-treated, and LPA-treated mice. Statisticalanalysis: Two-tailed unpaired t-test (F and H), ANOVA, and post-hocmultiple comparison tests (E and J). Data represent mean ± SEM withindividual data points shown. n = 4 per group for TTC analysis andbrain water content (E and F), n = 5 per group for TTC analysis (G),n = 5 per group for behavioral analysis (I). *p < 0.05 vs. day 0 (E), PBS(H), ***p < 0.01 vs. sham (F), PBS (J). Scale bar = 5 mm (C, D, and G) ### Page 8 Quantitative analysis showed no significant differ-ence in CD31-positive vessel density between the threegroups (Fig.B), suggesting that LPA does not alter over-all vascular density within the peri-infarct region. How-ever, claudin-5–positive vessel density was significantlyhigher in LPA-treated mice compared with PBS-treatedcontrols (Fig.C; P < 0.05), confirming that LPA attenu-ates the ischemia-induced loss of claudin-5 expression.Together, these findings indicate that LPA preserves BBBintegrity by maintaining tight junction protein expressionrather than by inducing angiogenesis or increasing vesseldensity. LPA Treatment Modulates Gene Expression in theInfarct Core To further investigate LPA-mediated molecular responseswithin the infarct core, we performed RNA-seq on tissue ### Page 9 isolated specifically from this region in LPA-treated andPBS-treated dMCAO mice. PCA analysis demonstratedclear separation between treatment groups (Fig.D),indicating distinct transcriptional profiles associatedwith LPA exposure. Consistent with this, volcano plotanalysis revealed numerous significantly regulated genes(Fig.E), further supporting robust LPA-dependent tran-scriptional modulation. Venn diagram analysis identified635 genes uniquely upregulated and 446 genes uniquelydownregulated in PBS-treated samples compared withsham tissue. Among the 635 upregulated genes in the PBSgroup, 12 overlapped with genes that were downregulatedin the LPA-treated dMCAO group compared with the PBScontrol group (Fig.F). GO enrichment analysis of DEGsrevealed significant enrichment of pathways associatedwith cell adhesion, extracellular matrix organization, andvascular development (Fig.G), consistent with the bulkRNA-seq findings from the whole cortex and supportingthe interpretation that LPA influences vascular functionthrough transcriptional mechanisms rather than structuralchanges. LPA4 Receptor Mediates the BBB-Protective Effectsof LPA To identify the LPA receptor subtype responsible for theBBB-protective effects of LPA, we evaluated LPA treat-ment in LPA4 receptor knockout (LPA4 KO) mice subjectedto dMCAO. The protective effect of LPA on BBB perme-ability was abolished in LPA4 KO mice, as evidenced bycomparable Evans blue extravasation between LPA-treatedand vehicle-treated LPA4 KO mice (Fig.A, B). In contrast,wild-type mice showed a reduction in Evans Blue leakagefollowing LPA administration. Evans Blue leakage in LPA-treated LPA4 KO mice remained comparable to vehicle-treated KO controls, indicating loss of LPA-mediated BBBprotection. Similarly, LPA treatment failed to reduce infarctvolume in LPA4 KO mice (Fig.C, D), whereas wild-typemice displayed the expected decrease in infarct size withLPA administration. These findings demonstrate that LPA4signaling is required for both BBB protection and infarctvolume reduction induced by LPA following cerebralischemia. LPA4 Receptor is Expressed in Vascular Endotheliumand Maintained After Ischemia To determine the cellular localization of LPA receptorexpression in the brain and identify the receptor sub-type likely mediating the protective effects of LPA, weperformed a comprehensive expression analysis of LPAreceptors across major brain cell populations and vascu-lar compartments. Single-cell analysis revealed distinctexpression patterns of different LPA receptor subtypes(Fig.E). UMAP visualization of cell clusters showedmajor brain cell types, including Oligo, Astro, neurons,MG, ECs, ependymal cells, and additional supportingpopulations (Fig.E, left panel). Analysis of receptor expression patterns revealedthat the LPA4 receptor was prominently and selectivelyexpressed in vascular EC (Fig.F). Feature plots con-firmed enrichment of LPA4 expression within the endo-thelial cell cluster, consistent with its putative role inregulating BBB integrity. The identity of each cluster wasverified using established cell-type marker genes, includ-ing Cldn5 for EC, Tmem119 for microglia, and Mbp foroligodendrocytes (Fig.F). Comparison between shamand ischemic conditions demonstrated that endothe-lial LPA4 expression was preserved following ischemicinjury (Fig.G). The maintenance of LPA4 expressionin post-ischemic vessels suggests that EC retain LPA4receptor availability after stroke, consistent with thepotential role of LPA4 in mediating BBB protection afterischemia. Discussion In this study, we show that LPA protects against BBBdisruption and mitigates ischemic brain injury followingfocal cerebral ischemia in mice. LPA administration sig-nificantly reduced infarct volume, attenuated brain edema,preserved BBB integrity, and showed trends towardimproved behavioral outcome after dMCAO. Notably,LPA did not reduce BBB permeability and infarct volumein LPA4 KO mice; this finding highlights LPA4 as a cru-cial mediator of the above-mentioned protective effectson vascular permeability and infarct size. Transcriptomicanalysis further revealed that LPA induces distinct shiftsin gene expression profiles favoring vascular stabilizationand anti-inflammatory responses. Additionally, single-cellanalysis confirmed LPA4 receptor expression in vascularendothelial cells, which was sustained after ischemia.These findings suggest LPA–LPA4 signaling axis as apotential therapeutic target for preserving cerebrovascularfunction after ischemic stroke. BBB disruption is a major determinant of secondaryinjury after ischemic stroke, driving vasogenic edema,inflammatory cell infiltration, and worsening neurologicaloutcomes []. Consistent with this, our findings showed thatLPA treatment significantly reduced Evans Blue and FITC-dextran extravasation, demonstrating effective preservation ### Page 10 ### Page 11 of BBB integrity. LPA reduced vascular permeability inboth the ischemic core and penumbra, indicating spatiallybroad stabilization of the neurovascular unit. This protec-tive effect correlated with preserved claudin-5 expressionin endothelial cells, suggesting that LPA helps maintaintight junction architecture that is otherwise compromisedduring ischemia. Additional studies are required to directlyassess whether LPA4 specifically mediates the preserva-tion of claudin-5 expression in LPA4 KO animals; however,LPA4 couples primarily to Gαs and Gαi proteins, modulat-ing cyclic AMP (cAMP) signaling [], which is knownto enhance endothelial barrier function through proteinkinase A (PKA)-dependent stabilization of tight junctions.LPA signaling can also activate Rho family GTPases, keyregulators of cytoskeletal organization and tight junctionassembly [,]. These mechanisms provide a plausiblemechanistic basis for LPA-mediated tight junction preser-vation, although direct evidence linking LPA4 signaling tomaintenance of claudin-5 expression requires further inves-tigation in LPA4-deficient animals. The essential role of LPA4 in these protective effects wasconfirmed in LPA4 KO mice, where LPA failed to reduceBBB leakage or infarct size. Furthermore, single-cell expres-sion analyses revealed prominent LPA4 receptor expressionin vascular endothelial cells, the principal structural com-ponent of the BBB, and demonstrated that this expressionis maintained after ischemic injury. Although our find-ings strongly support an endothelial mechanism, we can-not exclude contributions from other cell types expressinglower levels of LPA4, especially given the diverse and con-text-dependent effects of LPA receptors in CNS disorders[]. For example, LPA1 has been implicated in promot-ing BBB disruption in neuroinflammatory conditions [],whereas our results highlight an opposing, receptor-specificprotective role for LPA4. Several previous studies have reported deleterious effectsof LPA on ischemic brain injury and BBB disruption [–,,], predominantly implicating the autotaxin–LPA axisor LPA1/LPA5 signaling. The apparent discrepancy withour findings likely reflects the context dependence of LPAbiology, including differences in receptor subtype engage-ment, dose, and model systems. Notably, the protectiveeffects in the present study were strictly LPA4-dependent,as confirmed from their abolition in LPA4 KO mice. Theobserved preservation of MAP2-positive neurons in regionswith reduced FITC-dextran leakage in LPA-treated micesuggests that the effects of LPA extend beyond the endothe-lium to support the broader neurovascular unit. This find-ing aligns with the evidence that BBB integrity is essentialfor neuronal survival following ischemia, and supports theconcept that interventions that preserve BBB function canconfer indirect neuroprotection []. Our findings have several implications for therapeuticdevelopment. Current neuroprotective strategies for isch-emic stroke have had limited clinical success [], in partbecause many approaches fail to address BBB dysfunc-tion—a major contributor to poor outcomes and hem-orrhagic transformation following reperfusion therapy[,]. The observation that LPA provides neuropro-tection when administered immediately after ischemiaonset suggests that LPA4 stimulation could complementexisting therapies, including thrombolysis and thrombec-tomy, by reducing reperfusion-induced BBB breakdown.Given that different LPA receptors can mediate oppos-ing effects in the CNS, receptor selectivity is crucial fortherapeutic development. The requirement for LPA4 inmediating LPA-induced reduction in vascular permeabil-ity and infarct volume, as demonstrated by the results ofLPA4 KO experiments, highlights the potential utilityof developing selective LPA4 agonists, which may offerimproved safety compared with nonselective LPA recep-tor agonists [–]. Third, because ischemia paradoxi-cally exacerbates BBB breakdown during reperfusion[], LPA could be combined with existing reperfusiontherapies to provide additive protection by stabilizingthe BBB and mitigating reperfusion injury, potentiallyimproving outcomes in patients undergoing thrombolysisor thrombectomy. Despite these promising results, our study has some limi-tations. First, LPA was administered immediately following Fig. 3 LPA treatment preserves BBB integrity and reduces vascularpermeability following cerebral ischemia A) Miles assay of blood–brain barrier permeability showing Evans blue extravasation in PBScontrol versus LPA-treated (10 mg/kg) brains at 48 h post-dMCAO.B) Quantification of Evans blue content (μg/g brain tissue) in PBS andLPA-treated brains. C) Immunofluorescence images showing FITC-dextran (200 kDa, green) extravasation and CD31-positive blood ves-sels (magenta) in the ischemic territory. White arrowheads in highermagnification images of the ischemia territory indicate fluorescentsignal accumulating outside CD31-positive vessels. D) Quantificationof FITC-dextran extravasation area (intensity) in the ischemic core ofPBS and LPA-treated brains. E) Dual immunofluorescence staining forFITC-dextran (green) and MAP2 neuronal marker (magenta) in theperi-infarct (penumbra) region. Higher magnification images compar-ing FITC-dextran extravasation relative to MAP2-positive neuronalregions in PBS and LPA-treated brains. The area between dotted linesrepresents the penumbral region where neurons remain viable but areat risk. F) Quantification of the overlap area (mm²) between dextranextravasation and MAP2-positive regions in vehicle and LPA groups.Statistical analysis: Two-tailed unpaired t-test (B, D, and F). Data rep-resent mean ± SEM with individual data points shown. n = 5 per groupfor Miles assay (B), n = 12 per group for extravasation area of dextran(D), n = 8 per group for merge area of dextran and MAP2 positive (F).*p < 0.05 vs. PBS, **p < 0.01 vs. PBS. Scale bar = 5 mm (A), 500 μm(low-magnification of B and C), and 50 μm (high-magnification of Band C) ### Page 12 ### Page 13 dMCAO, and the therapeutic time window for LPA activa-tion remains undefined. Future studies to evaluate the effectsof delayed LPA administration are essential to determine theclinical applicability of the findings. Second, the present study employed a permanentdMCAO model, which does not recapitulate reperfusioninjury occurring in patients treated with thrombolysis orthrombectomy. Permanent dMCAO was selected becauseof its surgical reproducibility and low inter-animal vari-ability; however, reperfusion introduces additional patho-logical mechanisms, including reactive oxygen speciesgeneration and matrix metalloproteinase activation, thatmight differentially engage LPA receptor signaling. Futurestudies using transient MCAO or embolic models are war-ranted to assess LPA–LPA4 signaling in the setting ofreperfusion injury. Third, we focused primarily on endothelial mechanisms,and contributions from astrocytes, pericytes, or infiltratingimmune cells—each of which influences BBB function—were not examined in detail. Fourth, behavioral assessment was limited. Althoughthe cylinder test is sensitive for detecting forelimb asym-metry after cortical stroke, it does not capture the full spec-trum of neurological deficits. Future studies incorporatingadditional behavioral paradigms, such as the rotarod, adhe-sive removal test, and grid walk, would enable a more com-prehensive assessment of neuroprotection. Fifth, only adult female mice were used, and estrous cyclestage was not controlled. As hormonal status can influenceBBB integrity and stroke severity, the generalizability ofthe findings to male animals and different hormonal statesremains uncertain. Potential sex differences in the LPA–LPA4 axis warrant investigation in future studies. Sixth, all outcomes were assessed at 48 h post-ischemia,and the longer-term effects of LPA treatment on infarctmaturation, neurological recovery, and potential risks ofdelayed secondary injury were not examined. Seventh, the dose of LPA used (10 mg/kg) was selectedbased on preliminary comparisons of 3, 5, and 10 mg/kgdoses without performing a full dose-response analysis.As LPA acts as a non-selective agonist across multipleLPA receptor subtypes, the receptor selectivity of LPAat this dose in vivo cannot be guaranteed. Although theloss of protection in LPA4 KO mice indicates that LPA4signaling is required for the observed effects, we cannotexclude contributions from other receptors in wild-typeanimals. Eighth, BBB permeability measurements were not nor-malized to infarct volume, which limited our ability to con-clusively distinguish a direct BBB-stabilizing effect fromsecondary reduction in permeability resulting from smallerinfarcts. Finally, pharmacokinetic and safety profiling of LPA orLPA4-selective agonists were not evaluated in this study;such analyses will be required to assess the translational fea-sibility of targeting LPA–LPA4 signaling. In conclusion, this study identifies LPA–LPA4 signalingas a previously unrecognized mechanism for preservingBBB integrity and mitigating cerebral ischemia. LPA treat-ment reduces infarct volume and brain edema accompaniedby preserved claudin-5 expression and modulation of vas-cular stabilization and anti-inflammatory transcriptionalresponses. LPA4 KO experiments establish that LPA4 isrequired to reduce vascular permeability and infarct vol-ume, and single-cell analysis demonstrates selective LPA4expression in vascular endothelial cells together with itsrobust endothelial expression and sustained presence afterischemia, provides a compelling rationale for consider-ation of LPA4-targeted therapeutic strategies. Given thepersistently high global burden of stroke-related disabilityand mortality, and the paucity of effective neuroprotectivetherapies, the LPA–LPA4 pathway represents a promisingtarget for further translational investigation. Fig. 4 LPA treatment maintains tight junction protein expression inthe infarct cerebral vasculature and modulates gene expression in theinfarct core. A) High-magnification confocal microscopic images dis-playing claudin-5 expression (green) and CD31 endothelial marker(magenta). The merged images show their spatial relationship in sham,PBS, and LPA-treated brain sections. B) Quantification of CD31-pos-itive vessel density (% vessel area/total area) in the sham, PBS, andLPA groups. C) Quantification of claudin-5-positive vessel density (%vessel area/total area) in the sham, PBS, and LPA groups. D) Principalcomponent analysis (PCA) of single-cell RNA sequencing data fromvascular endothelial cells (VEC) isolated from ischemic cortex at 48 hpost-dMCAO, showing distinct gene expression patterns among sham,PBS-treated dMCAO, and LPA-treated dMCAO mice. E) Volcano plotshowing upregulated (red) and downregulated (green) DEGs uponLPA-treated VEC (left) and PBS-treated VEC (right). Labeled genesrepresent transcripts upregulated by PBS-treated VEC but downregu-lated by LPA-treated VEC, or downregulated by PBS-treated VEC butrestored/upregulated by LPA-treated VEC. F) Venn diagram analysisof differentially expressed genes (DEGs) in VEC, comparing PBS-treated VEC vs. sham VEC and LPA-treated VEC vs. PBS-treatedVEC (FDR < 0.05, |fold change| > 1.5). G) Heatmap showing theactivation of significantly enriched pathways, comparing PBS-treatedVEC with LPA-treated VEC. Revealing pathways related to cell adhe-sion, extracellular matrix organization, and vascular development,consistent with bulk RNA-seq findings. Statistical analysis: analysis ofvariance (ANOVA), and post-hoc multiple comparison tests (B and C).Data represent means ± SEM with individual data points shown. n = 12per group for vascular density (B and C), n = 3 per group for RNAsequencing and bioinformatics analysis. ****p < 0.0001 vs. sham andPBS groups, ns: not significant (B). Scale bar = 50 μm (A) ### Page 14 ### Page 15 Acknowledgements We thank Dr. Daisuke Motooka (RIMD, OsakaUniversity, Osaka) for performing RNA-seq analysis. Author Contributions S.Y., K.T., and H.K. conceived the study anddesigned the experiments. S.Y. performed the most experiments. K.T.,Y.H., and L.L. contributed to the analysis of vascular morphology, per-meability, and neuronal death. A.S., and T.S. assisted with the datacollection for the behavioral assessment of mice. N.H. performed RNAsequencing analysis. S.Y., K.T., and H.K. analyzed the data and wrotethe manuscript. K.K. and H.K. supervised the project and providedimportant feedback. All authors reviewed and approved the final ver-sion of the manuscript. Funding Open Access funding provided by University of Fukui. Thiswork was supported by JSPS KAKENHI (grant numbers: 23H04931,25K02516, 23K14590); JST FOREST Program (grant number: JPM-JFR2160); and University of Fukui Fund for promotion of Collabora-tive Research “Reff.” The funders played no role in the study design,data collection and analysis, decision to publish, or manuscript prepa-ration. Data Availability The data supporting the findings of this study areavailable from the corresponding author upon reasonable request.RNA-seq datasets generated for this work will be deposited in theGene Expression Omnibus (GEO) database upon acceptance of themanuscript. In addition, the publicly accessible dataset GSE261494was utilized to assess the expression of LPA receptors in the mousecortex. Declarations Compliance with Ethical Standards All experimental procedures wereapproved by the Institutional Animal Care and Use Committee of theUniversity of Fukui (approval number: R07013) and conducted in ac-cordance with the ARRIVE guidelines for animal experimentation. Competing Interests The authors declare no competing interests. Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format,as long as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons licence, and indicateif changes were made. The images or other third party material in thisarticle are included in the article’s Creative Commons licence, unlessindicated otherwise in a credit line to the material. If material is notincluded in the article’s Creative Commons licence and your intendeduse is not permitted by statutory regulation or exceeds the permitteduse, you will need to obtain permission directly from the copyrightholder. To view a copy of this licence, visit. References 1. Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, CallawayCW, Carson AP, et al. Heart disease and stroke statistics-2019update: A report from the American Heart Association. Circula-tion. 2019;139(10):e56–528.. 2. Zlokovic BV. Neurovascular pathways to neurodegenerationin Alzheimer’s disease and other disorders. Nat Rev Neurosci.2011;12(12):723–38.. 3. Jiang X, Andjelkovic AV, Zhu L, Yang T, Bennett MVL, Chen J,et al. Blood-brain barrier dysfunction and recovery after ischemicstroke. Prog Neurobiol. 2018;163–164:144–71.. 4. Candelario-Jalil E, Dijkhuizen RM, Magnus T. Neuroinflamma-tion, stroke, blood-brain barrier dysfunction, and imaging modal-ities. Stroke. 2022;53(5):1473–86.. 5. Daneman R, Prat A. The blood-brain barrier. Cold Spring HarbPerspect Biol. 2015;7(1):a020412.. 6. Sweeney MD, Zhao Z, Montagne A, Nelson AR, ZlokovicBV. Blood-brain barrier: From physiology to disease and back.Physiol Rev. 2019;99(1):21–78.. 7. Profaci CP, Munji RN, Pulido RS, Daneman R. The blood-brainbarrier in health and disease: Important unanswered questions. JExp Med. 2020.. 8. Shi Y, Zhang L, Pu H, Mao L, Hu X, Jiang X, et al. Rapid endothe-lial cytoskeletal reorganization enables early blood-brain barrierdisruption and long-term ischaemic reperfusion brain injury. NatCommun. 2016;7:10523.. 9. Kuroiwa T, Ting P, Martinez H, Klatzo I. The biphasic openingof the blood-brain barrier to proteins following temporary middlecerebral artery occlusion. Acta Neuropathol. 1985;68(2):122–9.. 10. Gelderblom M, Leypoldt F, Steinbach K, Behrens D, ChoeCU, Siler DA, et al. Temporal and spatial dynamics of cerebralimmune cell accumulation in stroke. Stroke. 2009;40(5):1849–57.. 11. Yung YC, Stoddard NC, Chun J. LPA receptor signaling: phar-macology, physiology, and pathophysiology. J Lipid Res.2014;55(7):1192–214.. 12. Kihara Y, Maceyka M, Spiegel S, Chun J. Lysophospholipidreceptor nomenclature review: IUPHAR Review 8. Br J Pharma-col. 2014;171(15):3575–94.. 13. Aikawa S, Hashimoto T, Kano K, Aoki J. Lysophosphatidic acidas a lipid mediator with multiple biological actions. J Biochem.2015;157(2):81–9.. 14. Yung YC, Stoddard NC, Mirendil H, Chun J. LysophosphatidicAcid signaling in the nervous system. Neuron. 2015;85(4):669–82.. 15. Geraldo LHM, Spohr T, Amaral RFD, Fonseca A, Garcia C,Mendes FA, et al. Role of lysophosphatidic acid and its receptorsin health and disease: novel therapeutic strategies. Signal Trans-duct Target Ther. 2021;6(1):45.. 16. Ren Y, Guo L, Tang X, Apparsundaram S, Kitson C, DeguzmanJ, et al. Comparing the differential effects of LPA on the barrierfunction of human pulmonary endothelial cells. Microvasc Res.2013;85:59–67.. 17. Minnear FL, Patil S, Bell D, Gainor JP, Morton CA. Plateletlipid(s) bound to albumin increases endothelial electrical resis-tance: mimicked by LPA. Am J Physiol Lung Cell Mol Physiol.2001;281(6):L1337-44.. 18. Pei J, Cai L, Wang F, Xu C, Pei S, Guo H, et al. LPA(2) contrib-utes to vascular endothelium homeostasis and cardiac remodel-ing after myocardial infarction. Circ Res. 2022;131(5):388–403.. 19. English D, Kovala AT, Welch Z, Harvey KA, Siddiqui RA,Brindley DN, et al. Induction of endothelial cell chemotaxis bysphingosine 1-phosphate and stabilization of endothelial mono-layer barrier function by lysophosphatidic acid, potential media-tors of hematopoietic angiogenesis. J Hematother Stem Cell Res.1999;8(6):627–34.. ### Page 16 20. Sumida H, Noguchi K, Kihara Y, Abe M, Yanagida K, Hamano F,et al. LPA4 regulates blood and lymphatic vessel formation dur-ing mouse embryogenesis. Blood. 2010;116(23):5060–70.. 21. Takara K, Eino D, Ando K, Yasuda D, Naito H, Tsukada Y, et al.Lysophosphatidic acid receptor 4 activation augments drug deliv-ery in tumors by tightening endothelial cell-cell contact. Cell Rep.2017;20(9):2072–86.. 22. Muramatsu F, Hosoe N, Suzuki T, Shimamura T, Hayashi Y,Takara K, et al. Vasoprotective effects of lysophosphatidic acidinhibit vascular injury caused by SARS-CoV-2 infection. Sci Rep.2025;15(1):24622.. 23. Eino D, Tsukada Y, Naito H, Kanemura Y, Iba T, Wakabayashi T,et al. LPA4-mediated vascular network formation increases theefficacy of Anti-PD-1 therapy against brain tumors. Cancer Res.2018;78(23):6607–20.. 24. Tiwari S, Basnet N, Choi JW. Lysophosphatidic acid receptor 1plays a pathogenic role in permanent brain ischemic stroke bymodulating neuroinflammatory responses. Biomol Ther (Seoul).2024;32(3):319–28.. 25. Bhattarai S, Subedi U, Manikandan S, Sharma S, Sharma P, MillerC, et al. Endothelial specific deletion of autotaxin improves strokeoutcomes. Cells. 2023;12(3):0511.. 26. Wang Y, Zhang J, Huang L, Mo Y, Wang C, Li Y, et al. The LPA-CDK5-tau pathway mediates neuronal injury in an in vitro modelof ischemia-reperfusion insult. BMC Neurol. 2022;22(1):166.. 27. Yanagida K, Kurikawa Y, Shimizu T, Ishii S. Current progress innon-Edg family LPA receptor research. Biochim Biophys Acta.2013;1831(1):33–41.. 28. Suckau O, Gross I, Schrotter S, Yang F, Luo J, Wree A, et al.LPA(1), LPA(2), LPA(4), and LPA(6) receptor expression duringmouse brain development. Dev Dyn. 2019;248(5):375–95.. 29. 29. Ishidou A, Ishiguro T, Nagata KI, Ito H. DevelopmentalChanges in the Expression of Lysophosphatidic Acid Recep-tor 4 in the Mouse Brain. Dev Neurosci. 2025:1–10. doi:10.1159/000550132. 30. Kuraoka M, Furuta T, Matsuwaki T, Omatsu T, Ishii Y, Kyuwa S,et al. Direct experimental occlusion of the distal middle cerebralartery induces high reproducibility of brain ischemia in mice. ExpAnim. 2009;58(1):19–29.. 31. Fang Y, Huang R, Su Z, Lin J, Li J. [Changes after dynamicobservation of ultrastructural cerebral infarction in renovascularhypertensive rats]. Chin Med J (Engl). 2000;113(6):520–4. 32. Kozler P, Maresova D, Pokorny J. Determination of brain watercontent by dry/wet weight measurement for the detection ofexperimental brain edema. Physiol Res. 2022;71(S2):S277–83.. 33. Zhu W, Gao Y, Wan J, Lan X, Han X, Zhu S, et al. Changes inmotor function, cognition, and emotion-related behavior afterright hemispheric intracerebral hemorrhage in various brainregions of mouse. Brain Behav Immun. 2018;69:568–81.. 34. Yoshida Y, Takagi T, Kuramoto Y, Tatebayashi K, ShirakawaM, Yamahara K, et al. Intravenous administration of humanamniotic mesenchymal stem cells in the subacute phase ofcerebral infarction in a mouse model ameliorates neurologi-cal disturbance by suppressing Blood Brain Barrier disrup-tion and apoptosis via immunomodulation. Cell Transplant.2021;30:9636897211024183.. 35. Ozen I, Roth M, Barbariga M, Gaceb A, Deierborg T, Genove G,et al. Loss of Regulator of G-Protein Signaling 5 leads to neuro-vascular protection in stroke. Stroke. 2018;49(9):2182–90.. 36. Zudaire E, Gambardella L, Kurcz C, Vermeren S. A computationaltool for quantitative analysis of vascular networks. PLoS One.2011;6(11):e27385.. 37. Spitzer D, Khel MI, Putz T, Zinke J, Jia X, Sommer K, et al. Aflow cytometry-based protocol for syngenic isolation of neuro-vascular unit cells from mouse and human tissues. Nat Protoc.2023;18(5):1510–42.. 38. Ge SX, Son EW, Yao R. iDEP: an integrated web application fordifferential expression and pathway analysis of RNA-Seq data.BMC Bioinformatics. 2018;19(1):534.. 39. Ge X. iDEP Web Application for RNA-Seq Data Analysis. In:Methods Mol Biol, vol. 2284. 2021. p. 417–43.. 40. Love MI, Huber W, Anders S. Moderated estimation of fold changeand dispersion for RNA-seq data with DESeq2. Genome Biol.2014;15(12):550.. 41. 41. Jassal B, Matthews L, Viteri G, Gong C, Lorente P, FabregatA, et al. The reactome pathway knowledgebase. Nucleic AcidsRes. 2020;48(D1):D498-D503. doi: 10.1093/nar/gkz1031. 42. Kanehisa M, Furumichi M, Sato Y, Ishiguro-Watanabe M, TanabeM. KEGG: integrating viruses and cellular organisms. Nucleic AcidsRes. 2021;49(D1):D545–51.. 43. Slenter DN, Kutmon M, Hanspers K, Riutta A, Windsor J, NunesN, et al. WikiPathways: a multifaceted pathway database bridg-ing metabolomics to other omics research. Nucleic Acids Res.2018;46(D1):D661–7.. 44. Schaefer CF, Anthony K, Krupa S, Buchoff J, Day M, Hannay T,et al. PID: the Pathway Interaction Database. Nucleic Acids Res.2009;37(Database issue):D674-9.. 45. Yamamoto S, Sakai N, Nakamura H, Fukagawa H, Fukuda K, Tak-agi T. INOH: ontology-based highly structured database of signaltransduction pathways. Database (Oxford). 2011;2011:bar052.. 46. 46. Vergoulis T, Vlachos IS, Alexiou P, Georgakilas G, Maragka-kis M, Reczko M, et al. TarBase 6.0: capturing the exponentialgrowth of miRNA targets with experimental support. NucleicAcids Res. 2012;40(Database issue):D222-9. doi: 10.1093/nar/gkr1161. 47. 47. Kim GS, Harmon E, Gutierrez MC, Kim S, Vance L, Bur-rous H, et al. Single-cell analysis identifies Ifi27l2a as a generegulator of microglial inflammation in the context of aging andstroke in mice. Nat Commun. 2025;16(1):1639. doi: 10.1038/s41467-025-56847-1. 48. 48. Komarova YA, Kruse K, Mehta D, Malik AB. Protein Interac-tions at Endothelial Junctions and Signaling Mechanisms Regu-lating Endothelial Permeability. Circ Res. 2017;120(1):179–206.doi: 10.1161/CIRCRESAHA.116.306534. 49. 49. Choi JW, Herr DR, Noguchi K, Yung YC, Lee CW, MutohT, et al. LPA receptors: subtypes and biological actions. AnnuRev Pharmacol Toxicol. 2010;50:157 − 86. doi: 10.1146/annurev.pharmtox.010909.105753. 50. 50. Dedoni S, Avdoshina V, Olianas MC, Onali P. Role of Lyso-phosphatidic Acid in Neurological Diseases: From Pathophysi-ology to Therapeutic Implications. Front Biosci (Landmark Ed).2025;30(5):28245. doi: 10.31083/FBL28245. 51. 51. Bhattarai S, Sharma S, Ara H, Subedi U, Sun G, Li C, etal. Disrupted Blood-Brain Barrier and Mitochondrial Impair-ment by Autotaxin-Lysophosphatidic Acid Axis in Postischemic ### Page 17 Stroke. J Am Heart Assoc. 2021;10(18):e021511. doi: 10.1161/JAHA.121.021511. 52. 52. Zhang X, Huang T, Lang L, Yu L. Effects of lysophospha-tidic acid receptor 5 on NLRC4 inflammasome in brain tissuesof transient cerebral ischemia/reperfusion rat. Hum Exp Toxicol.2022;41:9603271221078870. doi: 10.1177/09603271221078870. 53. 53. Campbell BCV, De Silva DA, Macleod MR, Coutts SB,Schwamm LH, Davis SM, et al. Ischaemic stroke. Nat Rev DisPrimers. 2019;5(1):70. doi: 10.1038/s41572-019-0118-8. 54. 54. Zhao Z, Nelson AR, Betsholtz C, Zlokovic BV. Estab-lishment and Dysfunction of the Blood-Brain Barrier. Cell.2015;163(5):1064-78. doi: 10.1016/j.cell.2015.10.067. 55. Obermeier B, Daneman R, Ransohoff RM. Development, main-tenance and disruption of the blood-brain barrier. Nat Med.2013;19(12):1584–96.. 56. Chamorro A, Dirnagl U, Urra X, Planas AM. Neuroprotection inacute stroke: targeting excitotoxicity, oxidative and nitrosativestress, and inflammation. Lancet Neurol. 2016;15(8):869–81.. Publisher’s Note Springer Nature remains neutral with regard to juris-dictional claims in published maps and institutional affiliations.