Skip to main content
Download PDF

IL-33/ST2 signaling in monocyte-derived macrophages maintains blood-brain barrier integrity and restricts infarctions early after ischemic stroke

Journal of Neuroinflammation volume 21, Article number: 274 (2024) Cite this article

Abstract

Background

Brain microglia and infiltrating monocyte-derived macrophages are vital in preserving blood vessel integrity after stroke. Understanding mechanisms that induce immune cells to adopt vascular-protective phenotypes may hasten the development of stroke treatments. IL-33 is a potent chemokine released from damaged cells, such as CNS glia after stroke. The activation of IL-33/ST2 signaling has been shown to promote neuronal viability and white matter integrity after ischemic stroke. The impact of IL-33/ST2 on blood-brain barrier (BBB) integrity, however, remains unknown. The current study fills this gap and reveals a critical role of IL-33/ST2 signaling in macrophage-mediated BBB protection after stroke.

Methods

Transient middle cerebral artery occlusion (tMCAO) was performed to induce ischemic stroke in wildtype (WT) versus ST2 knockout (KO) male mice. IL-33 was applied intranasally to tMCAO mice with or without dietary PLX5622 to deplete microglia/macrophages. ST2 KO versus WT bone marrow or macrophage cell transplantations were used to test the involvement of ST2+ macrophages in BBB integrity. Macrophages were cocultured in transwells with brain endothelial cells (ECs) after oxygen-glucose deprivation (OGD) to test potential direct effects of IL33-treated macrophages on the BBB in vitro.

Results

The ST2 receptor was expressed in brain ECs, microglia, and infiltrating macrophages. Global KO of ST2 led to more IgG extravasation and loss of ZO-1 in cerebral microvessels 3 days post-tMCAO. Intranasal IL-33 administration reduced BBB leakage and infarct severity in microglia/macrophage competent mice, but not in microglia/macrophage depleted mice. Worse BBB injury was observed after tMCAO in chimeric WT mice reconstituted with ST2 KO bone marrow, and in WT mice whose monocytes were replaced by ST2 KO monocytes. Macrophages treated with IL-33 reduced in vitro barrier leakage and maintained tight junction integrity after OGD. In contrast, IL-33 exerted minimal direct effects on the endothelial barrier in the absence of macrophages. IL-33-treated macrophages demonstrated transcriptional upregulation of an array of protective factors, suggesting a shift towards favorable phenotypes.

Conclusion

Our results demonstrate that early-stage IL-33/ST2 signaling in infiltrating macrophages reduces the extent of acute BBB disruption after stroke. Intranasal IL-33 administration may represent a new strategy to reduce BBB leakage and infarct severity.

Introduction

Ischemic stroke rapidly initiates a series of molecular events that induce profound alterations in neural cells functioning at the interface of the blood and brain, such as vascular endothelial cells (ECs), mural cells, astrocytes, and microglia. Some of these acute cellular changes disrupt the integrity the blood-brain barrier (BBB) and contribute to the long-term progression of stroke injury [1]. Major BBB disruption appears several hours (4–6 h) after ischemic stroke and results in vasogenic edema, which is characterized by endothelial tight junction (TJ) breakdown and the detrimental influx of water and plasma proteins, such as IgGs, into the brain parenchyma [2, 3]. Early BBB damage and brain edema are known to predict poor long-term neurological outcomes after stroke [4,5,6]. Hence, a better understanding of the mechanisms that compromise the integrity of the BBB in the early stages of experimental stroke may reveal new therapeutic targets to improve clinical outcomes in the future.

Innate immune cells, including CNS-resident microglia/macrophages, monocyte-derived macrophages, and other infiltrating myeloid cells respond quickly to cerebral ischemia, through receptor-mediated recognition of danger-associated molecular patterns (DAMPs) [7]. The stimulation of microglia and mobilization of monocyte-derived macrophages is required to stabilize damaged blood vessels, clear debris, and restrict brain injury [8,9,10]. However, reactive microglia and macrophages also produce inflammatory cytokines and reactive oxygen species that initially induce the phosphorylation of TJ proteins, which raises BBB permeability [2, 11, 12], and result in recruitment of leukocytes that further amplify BBB breakdown and tissue damage [3]. Thus, identification of the DAMPs that trigger early signal transduction cascades in CNS microglia/macrophages, leading in turn to acute BBB disruption and inflammatory feed-forward cycles, will be critical for alleviating stroke injury.

As a member of the IL-1 cytokine family, interleukin 33 (IL-33) is a pleiotropic DAMP that initiates immune responses upon its release and activation after cellular damage [13, 14]. IL-33 is a chromatin-associated protein and is therefore localized in the nucleus under physiological conditions. Full-length IL-33 is released into the extracellular compartment during cellular necrosis, where it serves as a cytokine and chemoattractant to activate immune responses [15]. For example, CNS injury induces the release of IL-33 from mature oligodendrocytes and astrocytes [16,17,18]. The receptor for IL-33 consists of a heterodimer of ST2 with IL-1R accessory protein (IL-1RAP), which is expressed on the surface of myeloid cells and various other leukocytes [13, 14]. The IL-33/ST2 signaling pathway exerts immunomodulatory functions in diverse immune cells. For example, IL-33 acts directly on monocytes to promote beneficial, anti-inflammatory macrophage differentiation in the CNS [17]. IL-33 delivery is also beneficial in mouse heart transplant experiments, because it promotes monocyte differentiation into reparative and regulatory macrophages that prevent transplant rejection [19]. In the context of ischemic stroke, we have shown that IL-33/ST2 signaling diminishes acute neuronal loss [16] and promotes remyelination [20] after transient middle cerebral artery occlusion (tMCAO). The protection of oligodendrocyte lineage cells by IL-33/ST2 signaling depends on the functions of microglia/macrophages, because PLX5622-mediated microglia depletion prevented the protective effects of IL-33 [20]. In addition, IL-33 has been reported to exert beneficial effects after stroke by enhancing ST2-dependent regulatory T cell responses [21] and regulating Th1/Th2 balance [22, 23]. However, the specific impact of IL-33/ST2 on BBB integrity after stroke remains unknown.

The current study tests the hypothesis that IL-33/ST2 signaling limits the extent of BBB disruption during the subacute phase of experimental stroke in a macrophage-dependent manner. Our results show that the activation of ST2 on monocyte-derived macrophages serves as a natural brake on BBB injury progression after ischemia. To leverage the beneficial effects of this endogenous protective mechanism, recombinant IL-33 administration may represent a new strategy to reduce BBB leakage and infarct severity.

Methods

Animals

C57BL/6 mice, ST2 KO (ST2–/–) mice, CX3CR1-GFP mice, and CCR2-RFP mice (male, 3-months old) were obtained from Jackson Laboratories. All mice were housed on a 12-hour light/dark cycle with ad libitum access to food and water. Animal care and procedures were reviewed and approved by the University of Pittsburgh Institute Animal Care and Use Committee and were conducted in accordance with the ethical regulations.

Transient middle cerebral artery occlusion

Transient cerebral ischemia was induced in male mice (25–30 g) via intraluminal occlusion of the left middle cerebal artery (MCA) for 60 min, as previously described [16]. Briefly, mice were anesthetized with 1.5% isoflurane in 30% O2/70% N2O. The tMCAO model was performed by insertion of a silicon-coated nylon monofilament (Doccol Corporation, Sharon, MA, USA) into the external carotid artery. The microfilament was then advanced into the internal carotid artery to block the origin of the MCA. After 60–40 min of occlusion, the monofilament was withdrawn to restore blood flow. The occlusion was confirmed by measuring regional cerebral blood flow (rCBF) with a 2-D laser speckle imaging system (PeriCam PSI System; Perimed, Järfälla-Stockholm, Sweden). Only mice with a rCBF reduction of > 70% of pre-MCAO baseline levels were included for further experimentation. A total of 98 mice (51 WT C57BL/6, 32 ST2 KO, 3 CCR2-RFP, and 12 bone marrow chimera mice) underwent tMCAO surgery in this study. Five mice were excluded due to post-surgery death or insufficient reduction of rCBF. The sham-operated mice were subjected to the same surgical procedures, but without the MCA occlusion. During surgery, mouse body temperature (37 Â°C ± 0.5 Â°C) was maintained by a thermostatically controlled heating blanket. The tMCAO surgery was performed by investigators blinded to animal genotype and experimental groupings.

Measurement of infarct volume

Infarct volumes were assessed 3d after tMCAO. Six evenly spaced coronal brain slices (25 Î¼m thick) encompassing the MCA territory (~ 1.10 mm anterior to bregma to ~ 2.06 mm posterior to bregma) were stained with antibodies against the neuron-specific marker microtubule associated protein 2 (MAP2, Sigma-Aldrich, St. Louis, MO) for infarct volume assessment. Tissue volumes were determined using Image J software by an observer blinded to group assignments. The area of tissue loss was calculated as the area of the contralateral hemisphere minus the non-infarcted area of the ipsilateral hemisphere. Brain infarct volume was determined by multiplying the area of tissue loss by the distances between the analyzed brain sections.

Bone marrow isolation and differentiation of bone marrow-derived macrophages

Bone marrow (BM) was isolated from the femur and tibia of WT or ST2 KO mice at 8–10 weeks of age. Each bone was flushed with Dulbecco’s modified Eagle’s medium containing 2 mM EDTA to collect BM tissues. The BM tissues were passed through a 70 Î¼m filter to remove debris, followed by centrifugation for 3 min at 300 rpm to collect BM cells. The BM cells were then cultured in plastic dishes (100 mm diameter) in macrophage culture medium (RPMI-1640, 10% FBS, 20% L929 conditioned medium, 1% penicillin-streptomycin) to promote the growth of bone marrow-derived macrophages. After 7 days in culture, the medium was supplemented with 20-ng/mL macrophage colony stimulating factor (M-CSF) and cultured for 3 more days before further experiments.

Irradiation and BM transplantation

To construct BM chimeric mice, 6-week-old C57BL/6J WT or ST2 KO recipient mice were exposed to γ irradiation at 950 rad. BM cells obtained from 8 to 10-week-old C57BL/6J or ST2 KO donors (5 × 106 cells per recipient) were transferred intravenously to recipients 2 h after irradiation. After 6 weeks of reconstitution, the chimeric mice were subjected to tMCAO, as described above. The efficiency of chimerism in recipient mice was more than 90%, according to flow cytometric analyses of blood macrophages.

Macrophage depletion and adoptive transfer

Recipient WT mice were intravenously injected with clodronate liposomes (Liposoma, CP-010-010, 200 Âµl/mouse/day) for 2 days to deplete circulating monocytes/macrophages. Mice were subjected to tMCAO 24 h after the final dose. Bone marrow-derived macrophages (2 × 106 cells per recipient) from WT or ST2 KO mice were transferred intravenously immediately after reperfusion.

Microglia/macrophage depletion

For microglia/macrophage depletion, PLX5622 was supplied to mice (9–10 weeks old, 25–30 g body weight) in the diet (Research Diets) at 1200 ppm (1200 mg/kg of chow), starting 7 days prior to surgery and continuing until sacrifice.

Immunofluorescence staining

Immunohistochemistry was performed on 25-µm free-floating brain sections as we described previously [16]. Primary antibodies included the following: rabbit anti-ZO-1 (abcam, Waltham, MA, USA), goat anti-CD31 (R&D Systems, Western Springs, IL, USA), rat anti-ST2 (Mdbioproducts, St. Paul, MN, USA), goat anti-IL-33 (R&D system, Western Springs, IL, USA), and rabbit anti-Iba1 (Wako, Richmond, VA, USA). Images were captured with confocal microscopy (Olympus, Shinjuku-ku, Tokyo, Japan), loaded into Image J (NIH), and quantified by two investigators blinded to grouping. Three randomly selected microscopic fields in the peri-infarct area of each brain section were analyzed, and this analyses was repeated for three consecutive coronal sections of each mouse brain. Numbers from the nine images per mouse were then averaged.

Flow cytometry

Single-cell suspensions were prepared from blood and brain as we previously reported [16]. Isolated blood cells were resuspended at 1 × 106/mL and stained with fluorophore-labeled antibodies according to the manufacturer’s instructions: CD3-APC (Thermo Fisher Scientific, Waltham, MA, USA), Ly6G-BUV395 (BD Biosciences, San Jose, CA, USA), CD11b-APC-eF780 (Invitrogen, Carlsbad, CA, USA), F4/80-BV605 (BD Biosciences, San Jose, CA, USA), ST2-FITC (Mdbioproducts, St. Paul, USA). Brain cell suspensions were prepared with Neural Tissue Dissociation Kit (T) (Miltenyi Biotec; Auburn, CA) using the gentle MACS Octo Dissociator with heaters (Miltenyi Biotec, Auburn, CA). Single cell suspensions were separated from myelin and debris by Percoll gradient centrifugation (500 g, 30 min, 18 Â°C). The leukocytes in the interface were collected and washed with Hank’s balanced salt solution (Sigma-Aldrich, St. Louis, MO) containing 1% fetal bovine serum (Sigma-Aldrich, St. Louis, MO) and 2 mM EDTA (Sigma-Aldrich, St. Louis, MO) before staining. Cells were stained with the following antibodies: CD31-V450 (Thermo Fisher Scientific, Waltham, MA, USA), Ly6G-BUV395 (BD Biosciences, San Jose, CA, USA), CD11b-BUV737 (BD Biosciences, San Jose, CA, USA), CD45-FITC (BioLegend, San Diego, CA, USA), F4/80-BV605 (BD Biosciences, San Jose, CA, USA), ST2-PE-Cy7 (Invitrogen, Carlsbad, CA, USA). Flow cytometric analysis was performed using an LSRII flow cytometer (BD Biosciences, San Jose, CA, USA) and data were analyzed with FlowJo software.

Western blotting

Protein was isolated from the mouse brain in vivo and endothelial cultures in vitro. Western blots were performed using standard SDS-polyacrylamide gel electrophoresis. The primary antibodies used in this study include rabbit anti-ZO-1 (Cell signaling Technology, 78896, 1:500, Danvers, MA, USA) and mouse anti-β-actin (A2228, 1:20,000; Sigma-Aldrich, St. Louis, MO, USA). Polyvinylidene difluoride (PVDF) membranes were incubated in blocking buffer (LI-COR Biosciences, Lincoln, NE, USA) for 1 h at room temperature and then incubated with primary antibodies at 4 Â°C overnight. The membrane was incubated with secondary antibodies for 1 h at room temperature (1:10,000, LI-COR Biosciences, Lincoln, NE, USA), and scanned with the LI-COR Odyssey Infrared Imaging System 9201-550U (LI-COR Biotechnology, Lincoln, NE, USA). The results were normalized to β-actin expression.

Assessment of BBB permeability

The fluorescent tracer Alexa Fluor 555-conjugated cadaverine (0.95 kDa; Invitrogen, Carlsbad, CA, USA) was injected through the femoral vein at a dose of 200 Âµg per mouse, 60 min before sacrifice. Coronal brain Sect. (25 Î¼m thick) were prepared to visualize the leakage of tracers through the impaired BBB. To measure the extravasation of endogenous IgG, sections were blocked in an avidin-biotin solution (Vector Laboratories, Newark, CA, USA), followed by 5% (wt/vol) bovine serum albumin for 1 h. Sections were incubated with biotinylated anti-mouse IgG antibody (1:500; Vector Laboratories, Newark, CA, USA) at 4 Â°C overnight, and then incubated with Alexa Cy3 Streptavidin for 1 h (1:1,000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA). Whole-section images were acquired using an inverted fluorescence microscope (EVOS M7000, Thermo Fisher Scientific, Waltham, MA, USA). Six evenly spaced sections encompassing the MCA territory were quantified for cross-sectional area of fluorescence. These areas were summed and multiplied by the distance between Sect. (1 mm) to yield the volume of leakage in mm3. 3D plots were generated by the built-in Surface Plot function of ImageJ using the fluorescence luminance as height. Higher-resolution images were captured with an Olympus confocal microscope.

Oxygen–glucose deprivation

Primary mouse brain microvascular endothelial cells (BMECs) were purchased from Cell Biologics (C57-6023, Cell Biologics, Chicago, IL, USA). BMECs were grown in mouse endothelial cell medium (Cell Biologics, Chicago, IL, USA). For OGD, cultured BMECs were placed in an incubator chamber (Thermo Fisher Scientific, Waltham, MA, USA) containing 94% nitrogen, 5% CO2 and 1% O2 for 6 h. Control cultures were incubated for the same period at 37 Â°C in humidified 95% air and 5% CO2.

In vitro BBB model

The in vitro BBB model was established in cell culture inserts as described previously [24]. The 24-well transwell PET membranes (0.4 Î¼m pore, 6.5 mm diameter; Millicell) were coated with a gelatin-based coating solution (Cellbiologics, Chicago, IL, USA). BMECs were seeded onto the insert membrane at a density of 1.5 × 105 cells per membrane. Cultures were maintained in medium at 37 Â°C in humidified 95% (vol/vol) air and 5% (vol/vol) CO2 for 4d to reach confluence. BMECs were subjected to 6 h OGD or control non-OGD conditions. To measure the direct effect of IL-33 on BBB integrity, transwell BMECs were treated with IL-33 (50 ng/mL) or PBS after OGD. For BMEC-macrophage cocultures, bone marrow derived macrophages were cultured in the bottom chamber of the transwell system and were incubated in the presence of soluble IL-33 (50 ng/mL) or PBS in macrophage culture media for 24 h. The BMEC transwells were then inserted above the macrophages that cultured in the lower chambers.

To assess paracellular permeability, FITC–dextran (40 kDa; Sigma-Aldrich, St. Louis, MO, USA) was added into the luminal chamber at a concentration of 2.0 mg/mL in 500 µL media. Fluorescence intensity was measured with a fluorescence reader at 1, 2, 3, 4, and 6 h after OGD using 35 µL media from the lower (abluminal) chamber. Next, 35 µL fresh media were added to the lower chamber after each reading. The concentrations of tracers in samples were calculated from a standard curve fitted using known concentrations of tracers.

Real-time PCR and PCR array

Total RNA was isolated using the RNeasy Mini Kit, according to the manufacturer’s instructions (Qiagen, Germantown, MD, USA). Five micrograms of RNA were used to synthesize the first strand of cDNA using the First-Strand Synthesis System for RT-PCR (Bio-Rad, Hercules, CA, USA). Commercial quantitative PCR array plates (Qiagen PAMM-067Z, Germantown, MD, USA) were used to detect the expression of a series of immune response-related genes in IL-33- or PBS-treated macrophages. PCR was performed on the Opticon 2 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) using the SYBR green PCR Master Mix (Bio-Rad, Hercules, CA, USA). The cycle time values were normalized to GAPDH levels within the same sample. The expression levels of the mRNAs were then reported as fold changes versus sham control. Il10 primers: forward TTGAATTCCCTGGGTGAGAAG, reverse ATGGCCTTGTAGACACCTTG. IL13ra2 primers: forward TCTGCTCTTGGAAACCTGG, reverse CAAGTTGGACAGTTTGCATCC.

Statistical analyses

Sample sizes for animal studies were determined based on pilot studies or our published body of work. Results are presented as mean ± SD on scatterdot plots. GraphPad Prism was used for statistical analyses. The difference in means between two groups was assessed by the two-tailed Student’s t test (equal standard deviations) or Welch’s t test (unequal standard deviations). Differences in means among multiple groups were analyzed using a one-way or two-way ANOVA with treatment or time as the independent factor. When the ANOVA showed significant differences, pairwise comparisons between means were tested post hoc by the Bonferroni tests. In all analyses, p < 0.05 was considered statistically significant.

Results

ST2 knockout exacerbates BBB damage and neuronal tissue loss in the mouse brain after tMCAO

We first determined the effect of global ST2 knockout (KO) on ischemic infarct volume by immunolabeling mouse brain sections against the neuronal somatodendritic marker MAP2 at 3 days after tMCAO (Fig. 1A). Consistent with our previous study [16], ST2 KO mice had enlarged MAP2-negative brain infarcts compared to wild-type (WT) mice after both genotypes underwent tMCAO for 60 min (Fig. 1D).

Fig. 1
figure 1

Global ST2 deficiency enhances brain infarct volumes and BBB damage following experimental stroke. (A-C) MAP2 and IgG staining were used to determine the infarct volume and BBB damage, respectively, in WT and ST2 KO mice 3 days after 60 min–40 min tMCAO. Shown are representative coronal sections stained with anti-MAP2 (A, green) or anti-IgG (B, red) antibodies. (C) Top: Surface plot images generated from IgG immunostaining. Middle and Bottom: Representative images of IgG, and IgG/CD31 double staining in the peri-infarct areas. (D) Quantification of infarct volumes. (E) Quantification of IgG intensity in the ipsilesional hemisphere. (F) Representative double labeling of CD31 (red) and ZO-1 (green). Red inset boxes in the top left image illustrate the imaging areas for the quantification of CD31 and ZO-1 staining. Areas in the white insets were enlarged and reconstructed in 3D, as shown on the right column. (G) Quantification of the ratio of CD31+ vessels covered by ZO-1 staining (top) and diameters of CD31+ vessels (bottom) 3 days after tMCAO. (H) Representative immunoblots and quantification of ZO-1 protein levels in the ischemic hemisphere and contralateral hemispheres (Contra) collected 3 days after 60 min tMCAO. β-actin was used as the loading control. *P < 0.05, **P < 0.01, ***P < 0.001. Student’s two-tailed t test (G bottom), one-way ANOVA followed by post hoc Bonferroni test (D, E, G top), or Brown-Forsythe and Welch ANOVA test (H)

Full size image

BBB damage was then assessed in the same mice by labeling extravasated IgG (Fig. 1B). IgG was visible in the ipsilateral brain parenchyma outside of CD31 positive blood vessels (Fig. 1C), and IgG fluorescence intensity was higher in ST2 KO brains than WT brains after 60-minute tMCAO, indicating greater extravasation of blood proteins across the damaged BBB into the infarct territory in the absence of ST2 (Fig. 1E).

To confirm whether the increased BBB leakage in ST2 KO mice was a primary transformation or simply a secondary response to the enlarged infarct size, we subjected another group of ST2 KO animals to 40-min tMCAO. The ST2 KO mice with 40-min tMCAO displayed similar brain infarct volumes but significantly higher IgG extravasation compared to WT mice with 60-min occlusion (Fig. 1A-E). Thus, the greater breach in the BBB under ST2 KO conditions is unlikely to be a secondary consequence of greater infarction.

Next, we analyzed the expression of the tight-junction protein ZO-1 in CD31+ ECs, as another indicator of BBB permeability after tMCAO (Fig. 1F-G). The percentage of volumetric overlap between CD31+ blood vessels with ZO-1 signal was then assessed using 3D-reconstructed confocal images. Region-matched blood vessels from uninjured WT contralateral brains had an average ZO-1 vessel coverage greater than 95%, and this coverage dropped to 55.13% in WT vessels in the periinfarct areas after 60 min tMCAO. The ZO-1 coverage of CD31+ vessels was further reduced to 41.12% in ST2 KO brains after 40 min tMCAO (Fig. 1G). Of note, the average diameters of CD31+ vessels were greater in ST2 KO brains compared to WT brains after tMCAO (Fig. 1G). We also analysed ZO-1 protein expression levels using western blotting in the infarcted hemisphere after tMCAO, for further evidence of tight junction protein loss in ST2 KO brains after tMCAO (Fig. 1H, Figure S1A).

Taken together, these data suggest that ST2-linked signaling pathways play crucial roles in maintaining BBB integrity and restricting infarction after stroke.

ST2 is expressed in microglia and infiltrating macrophages in the ischemic brain after tMCAO

We explored the expression of ST2 on ECs and microglia/macrophages in the sham brain and ischemic brain at 3d after tMCAO (Fig. 2A-B). ST2 expression was detected in CD31+Ly6G− ECs, CD31−Ly6G−CD11b+CD45intermediate microglia, and CD31−Ly6G−CD11b+F4/80+CD45high macrophages in the sham-injured brains and was increased 3d after tMCAO. Robust ST2 immunostaining was apparent in the infarct core and peri-infarct regions 3 days after tMCAO, and this specific immunoreactivity was absent from ST2 KO brains (Fig. 2C). Iba1-expressing microglia/macrophages colocalized with ST2 in the peri-infarct area of WT mice (Fig. 2D). In CCR2-RFP reporter mice, ST2 was expressed by CCR2-RFP+ infiltrating macrophages and CD31+ ECs alongside the vessels (Fig. 2E). Double staining of CD68 and ST2 further confirmed the expression of ST2 on CD68+RFP+ macrophages in CCR2-RFP reporter mice (Figure S2). The endogenous ligand for ST2, IL-33, was expressed in ∼2,000 cells/mm2 in the infarct hemisphere of WT mouse brains 3 days after tMCAO. However, the density of IL-33+ cells increased to an average of ∼3,000 cells/mm2 in ST2 KO brains, along with an increase in overall fluorescence intensity (Fig. 2F). This suggests a compensatory increase in IL-33 ligand expression in the absence of functional IL-33 receptors.

Fig. 2
figure 2

ST2 expression in endothelial cells, microglia, and macrophages in the ischemic brain early after experimental stroke. (A-B) ST2 expression on ECs and microglia/macrophages was elevated after tMCAO. (A) Top: Representative gating strategy for CD31+Ly6G− EC, CD31−Ly6G−CD11b+CD45intermediate microglia, and CD31−Ly6G−CD11b+F4/80+CD45high macrophages in the brain 3d after tMCAO. Bottom: Histograms showing ST2 expression on ECs, macrophages, and microglia in sham and stroke brains. (B) Quantification of the percentage of ST2+ ECs, macrophages, or microglia in sham or ischemic brains. (C) ST2 staining (magenta) in the ischemic brains from WT and ST2 KO mice 3d after tMCAO. Cell nuclei are counterstained with DAPI (blue). (D) Co-immunostaining of ST2 (magenta) and Iba1 (green) in the peri-infarct area of WT mice 3d after tMCAO. Areas in the white insets are enlarged. (E) Co-immunostaining of ST2 (magenta) and CD31 (green) in the brain of CCR2-RFP reporter mice 3d after tMCAO. Areas in the white inset were enlarged and reconstructed in 3D. White arrows: ST2 expression in CD31+ ECs. Yellow arrows: ST2 expression in CCR2+ infiltrating macrophages. (F) Immunostaining of IL-33 and quantification of the number of IL-33+ cells and intensity of IL-33 staining in the peri-infarct area of WT and ST2 KO mice 3d following tMCAO. *P < 0.05. Two-tailed Student’s t test (B) or Welch’s t test (F)

Full size image

IL-33 exerts a protective effect on the BBB early after stroke, in a microglia/macrophage-dependent manner

We then determined whether IL-33 protects against BBB leakage after tMCAO in a manner dependent upon microglia/macrophages. WT mice were maintained on a diet containing the CSF1R-inhibitor PLX5622 to deplete microglia/macrophages or a normal diet for 7 days before 60 min tMCAO. Recombinant IL-33 (2 Âµg/mouse) or vehicle was administered intranasally to mice 2 h after tMCAO and repeated daily for 3d. IL-33 treatment resulted in smaller infarct volumes 3 days after tMCAO in mice that were fed control chow, but did not exert a protective effect in microglia/macrophage-depleted mice (Fig. 3A-B). IL-33 administration also lessened the extent of BBB leakage after tMCAO, as measured by mouse IgG immunostaining in the normal diet group. However, this effect was abolished in the PLX5622-treated group, suggesting that IL-33 acts on brain microglia/macrophages to promote BBB integrity and reduce the extent of ischemic injury (Fig. 3C-D). Representative high-resolution images of the peri-infarct area show effective Iba1+ microglia/macrophage depletion in PLX5622-treated groups (Fig. 3E). Consistent with the reduced leakage of IgG, IL-33-treated non-depleted mice (IL-33 + normal diet) had significantly higher ZO-1 coverage on CD31+ blood vessels compared to vehicle treated, non-depleted mice (PBS + normal diet) 3 days after tMCAO. However, IL-33 treatment did not improve ZO-1 blood vessel coverage in microglia/macrophage-depleted mice, and their ZO-1 coverage on vessels was comparable to PBS-treated stroke mice (Fig. 3F-G).

Fig. 3
figure 3

Intranasal IL-33 treatment protects against BBB damage in a microglia/macrophage-dependent manner. Mice were maintained on PLX5622 diet or control diet for 7 days and subjected to 60 min tMCAO. IL-33 or PBS was intranasally applied 2 h after ischemia and repeated daily for 3d. Brains were collected 3 days after stroke. (A) Representative coronal sections stained with antibodies against MAP2. (B) Quantification of infarct volumes on MAP2-stained sections. (C) Representative coronal sections immunolabeled for infiltrating IgGs. (D) Quantitative analysis of IgG intensity. (E) Representative images of CD31 (green) and Iba1 (white) immunostaining in the peri-infarct area. (F) Representative images of CD31 (red) and ZO-1 (green) double staining in the peri-infarct area. Cell nuclei are counterstained with DAPI (blue). (G) Quantification of the percentages of CD31+ vessels covered by ZO-1 staining. *P < 0.05, **P < 0.01. Two-way ANOVA followed by post hoc Bonferroni test

Full size image

Transplantation of ST2 KO bone marrow into WT recipients worsens infarct severity and exacerbates BBB damage after tMCAO

ST2 is expressed on both CNS-resident cells and peripheral immune cells that respond to brain injury. To distinguish whether the BBB-protective effect of IL-33/ST2 is dependent upon peripheral immune cells, we irradiated the WT animals to deplete hematopoietic stem cells and then reconstituted the bone marrow niche with either WT or ST2 KO hematopoietic stem cells (Fig. 4A). The recipients were given 6 weeks to recover from the procedure and to permit repopulation of peripheral immune cells from donor-derived leukocytes before tMCAO surgery. Based on prior evidence, donor bone marrow replaces peripheral immune cells of chimera mice, including the monocytes, with donor genotypes, whereas the microglia of recipient mice retain the host genotype [25]. WT bone marrow donors were used to reconstitute the bone marrow of recipient mice to control for off-target effects of irradiation that would otherwise be incorrectly attributed to ST2 knockout.

Fig. 4
figure 4

Transplantation of ST2 KO bone marrow into irradiated WT mice exacerbates BBB damage following experimental stroke. (A) Schematic illustrating bone marrow chimera mouse construction. Bone marrows from WT or ST2 KO donors were injected intravenously (5 × 106 cells/animal) into irradiated WT mice. After 6 weeks of reconstruction, the chimera mice (WT/WT or WT/ST2KO) were subjected to 60 min tMCAO. (B) Flow cytometry to confirm loss of ST2 expression in blood CD11b+F4/80+ macrophages and CD11b+F4/80- monocytes of WT/ST2 KO chimera mice. (C) Representative coronal sections stained with anti-MAP2 antibodies. (D) Quantification of infarct volume on MAP2 stained sections. (E) Representative coronal sections stained with anti-IgG antibodies. (F) Representative images of CD31 (green) and IgG (red) immunostaining in the peri-infarct area. (G) Quantification of infarct area in 6 levels of MAP2-stained coronal sections, spaced 1 mm apart, throughout the MCA territory. (H) Quantitative analysis of IgG intensity in 6 levels of IgG-stained coronal sections, spaced 1 mm apart, throughout the MCA territory. (I) The ratio of IgG intensity and infarct volume. (J) Representative images of CD31 (red) and ZO-1 (green) double staining in the peri-infarct area. Cell nuclei are counterstained with DAPI (blue). Areas in the white inset were enlarged and reconstructed in 3D. (K) Quantification of the percentages of CD31+ vessels covered by ZO-1 staining. *P < 0.05, ***P < 0.001. Two-tailed Student’s t test (D, I), one-way (K) or two-way (G, H) ANOVA followed by post hoc Bonferroni test

Full size image

Flow cytometric analysis confirmed a loss of ST2 expression on CD3−Ly6G−CD11b+ myeloid cells in the blood of ST2 KO bone marrow recipients (WT/ST2 KO) compared to WT bone marrow recipients (WT/WT) (Fig. 4B). As noted above, the efficiency of chimerism in recipient mice was above 90%. Notably, WT/ST2 KO mice displayed greater loss of MAP2 staining in the infarct hemisphere compared to WT/WT mice (Fig. 4C-D), corresponding to observations in global ST2 KO mice (Fig. 1A and D). There was also greater mouse IgG staining in the parenchyma of WT/ST2 KO brains, indicating worse BBB disruption (Fig. 4E-F). Importantly, the extent of plasma IgG extravasation was greater in WT/ST2 KO mice, even in brain sections with comparable infarct sizes between the groups, resulting in higher ratios of IgG intensity over infarct volume in WT/ST2 KO mice (Fig. 4G-I). These latter results suggest that the impact of peripheral ST2+ immune cells on BBB leakage does not depend on the overall severity of the stroke injury. Accordingly, recipients of ST2 KO bone marrow had less colocalization of the tight-junction protein ZO1 with CD31+ blood vessels after tMCAO (Fig. 4J-K). Taken together, a loss of ST2 expression in peripheral immune cells is sufficient to exacerbate BBB leakage and neuronal loss. Stated differently, ST2 receptors on monocyte-derived macrophages are instrumental in mediating endogenous protection against BBB damage, and this may have important consequences on infarct volumes.

Direct intravenous transfer of ST2 KO macrophages is sufficient to exacerbate BBB leakage and tissue loss after tMCAO

Recipient WT mice were intravenously injected with clodronate liposomes once a day for 48 h before tMCAO to deplete circulating monocytes. Next, 2 × 106 ST2 KO or WT bone marrow-derived macrophages were intravenously injected following tMCAO (Fig. 5A). The same procedure was first tested using bone marrow-derived macrophages from CX3CR1-GFP donors to confirm that intravenously injected macrophages infiltrated into the recipient brain by 3d post-tMCAO (Fig. 5B). Recipients of ST2 KO macrophages had significantly larger infarct volumes compared to WT macrophage recipients 3d after tMCAO (Fig. 5C), matching the results from global ST2 KO mice and ST2 KO bone marrow recipient mice (Figs. 1A and D and 4C-D). The transfer of ST2 KO macrophages resulted in significant loss of ZO-1 colocalization with CD31+ blood vessels compared to WT macrophage recipients in the ipsilateral but not contralateral brain (Fig. 5E). To test BBB dysfunction, we injected the fluorescent tracer Alexa Fluor 555-conjugated cadaverine (0.95 kDa) intravenously, 60 min before sacrifice at 3d post-tMCAO, to complement the mouse IgG staining. Recipients of ST2 KO macrophages displayed greater IgG staining as well as cadaverine tracer leakage in the ipsilesional parenchyma compared to WT macrophage recipient mice (Fig. 5F-I). We also performed macrophage replacement experiments in a cohort of ST2 KO mice (Figure S3A). Similar to what we observed in the WT recipients, WT macrophages reduced BBB damage (Figure S3B) and brain infact (Figure S3C) compared to ST2 KO macrophages, in ST2 KO recipients at 3d after tMCAO.

Fig. 5
figure 5

Depletion of ST2 on macrophages exacerbates BBB damage 3 days following experimental stroke. (A) Schematic diagram for macrophage depletion and adoptive transfer experiments. WT recipient mice received intravenous (i.v.) clodronate liposome injections to deplete monocyte/macrophage cells 48 h before tMCAO. Bone marrow-derived macrophages (MΦ) prepared from WT or ST2 KO mice were injected (i.v.) into recipient WT mice immediately after tMCAO. (B) Adoptive transfer of CX3CR1-GFP macrophages was followed by infiltration of donor GFP+IBA1+ cells (yellow arrows) in the peri-infarct area 3d post tMCAO. (C) Quantification of brain infarct volumes on MAP2-stained sections. (D) Representative images of CD31 and ZO-1 immunostaining in the peri-infarct area. Areas in the white insets were enlarged and reconstructed in 3D. Cell nuclei are counterstained with DAPI (blue). (E) Quantification of the percentages of CD31+ vessels covered by ZO-1 staining. (F) Fluorescent Cadaverine tracer was injected 60 min before sacrifice to quantify BBB leakage. Representative images showing leakage of exogenous tracer (red) or endogenous IgG (turquoise) outside of CD31+ vessels (green) in the peri-infarct area. (G) Representative high magnifications of tracer leakage (red), CD31 (white), and ZO-1 (green) in the peri-infarct area 3d following tMCAO. 3D-reconstructed images acquired from brains of ST2 KO macrophage-transplanted mice. (H-I) Quantification of tracer leakage (H) and IgG intensity (I). *P < 0.05, ***P < 0.001. Two-tailed Student’s t test (C, H, I), or one-way ANOVA followed by post hoc Bonferroni (E)

Full size image

IL-33/ST2 signaling in macrophages improves barrier function in brain microvascular endothelial cell (BMEM) cultures after oxygen-glucose deprivation (OGD)

We used an in vitro model of brain endothelial cell barrier ischemic damage by culturing brain microvascular endothelial cells (BMECs) in permeable cell culture inserts and subjecting them to OGD for 6 h. Next, the cell culture inserts were transferred to wells containing IL-33 or PBS pretreated macrophages from WT or ST2 KO donors and cocultured for another 6 h. FITC-labeled dextran was added to the upper chamber to test leakage from the luminal into abluminal chambers. FITC-dextran leakage through the BMEC layer was recorded 1, 2, 3, 4 and 6 h after OGD. Barrier leakage decreased when BMECs were cocultured with IL-33 treated macrophages. The permeability of the BMEC layer was indistinguishable from macrophage-free cultures when WT macrophages were treated with PBS, or when ST2 KO macrophages were treated with IL-33 or PBS (Fig. 6A).

Fig. 6
figure 6

IL-33/ST2 signaling in macrophages protects against BBB leakage in vitro in response to oxygen-glucose deprivation. (A) Primary mouse brain microvascular endothelial cells (BMECs) in cell culture inserts were exposed to 6 h of OGD. Reperfused BMECs were then cocultured with PBS-treated or IL-33-treated (25 Âµg/mL) WT or ST2 KO macrophages for 6 h. The diffusion of 40 kDa FITC-dextran from the luminal to abluminal chamber was measured over time. (B-C) Primary mouse BMECs in cell culture plates were exposed to 6 h of OGD. Reperfused BMECs were then treated for 4 h with conditioned media collected from PBS-treated or IL-33-treated WT or ST2 KO macrophages. (B) Representative western blot images and quantification of ZO-1. β-Actin was used as the loading control. (C) Representative images of BMECs immunolabeled for VE-cadherin (green) or ZO-1(magenta), and counterstained with DAPI (blue) for nuclear labelling. Images are representative of 3 independent experiments. (D) PCR arrays were used to measure immune regulatory changes in IL-33 or PBS-treated macrophages. The black line indicates fold-changes (2 ^ (-ΔCt)) relative to one. The purple lines indicate two-fold or negative two-fold changes in gene expression. Two arrays/group in two independent experiments. (E) RT-PCR was used to quantify the expression of IL10 and IL13ra2 in PBS or IL-33 treated macrophages. (F) BMECs in cell culture inserts were exposed to 6 h of OGD followed by reperfusion and treatment with PBS or IL-33 for 6 h. The diffusion of FITC-dextran from the luminal to abluminal chamber was measured over time. (G-H) BMECs in cell culture plates were exposed to 6 h of OGD followed by reperfusion and treatment with PBS or IL-33 for 4 h. (G) Representative western blot images and quantification of ZO-1. (H) Representative images of BMECs immunolabeled for VE-cadherin (green) or ZO-1(magenta), and counterstained with DAPI (blue) for nuclear labelling. Images are representative of 3 independent experiments. *P < 0.05. **P < 0.01. Two-tailed Student’s t test (E), one-way (B, G) or two-way (A, F) ANOVA followed by post hoc Bonferroni test

Full size image

In another set of in vitro experiments, macrophage cultures were treated with IL-33 or PBS for 24 h, and macrophage-conditioned medium was then transferred to BMEC cultures after 6 h of OGD. Degradation of tight junctions between ECs was evaluated through ZO-1 protein expression. Western blotting revealed that ZO-1 expression drops sharply in BMEC cultures (Fig. 6B; pink triangles, Figure S1B) after OGD compared to BMEC cultures not subjected to OGD, regardless of treatment with conditioned medium from WT (gray circles) or ST2 KO macrophages (black circles). BMECs that were exposed to PBS-treated WT macrophage-conditioned medium saw similar reductions in ZO-1 protein expression after OGD (Fig. 6B; red triangles). However, BMECs treated with conditioned medium from IL-33 treated macrophages preserved ZO-1 protein expression after 6 h of OGD (Fig. 6B; yellow diamonds), and this protective effect was lowered when ST2 KO macrophages were treated with either IL-33 (green circles) or PBS (blue circles) before transfer of conditioned medium.

We then immunostained BMEC cultures to label the adherens junction protein VE-cadherin and the tight junction protein ZO-1 after OGD. The results support the protective effects of conditioned medium from IL-33-treated WT macrophages (Fig. 6C). We used a high-throughput PCR array to explore the effects of IL-33 on the expression of a set of genes related to inflammation and immune responses in macrophages. IL-33 treatment upregulated the expression of genes that regulate immune responses and possess protective anti-inflammatory or anti-oxidative properties, such as Adrb2, Chia, Il13ra2, and Il10 (Fig. 6D). Conventional real time-PCR analyses confirmed the elevation of Il10 and Il13ra2 expression in macrophages after IL-33 treatment (Fig. 6E).

To confirm that the protective effects of IL-33 on endothelial cell junction integrity are mediated indirectly by macrophages, we directly exposed BMEC monocultures to IL-33 or PBS (without macrophages present) to measure changes in FITC-dextran diffusion at baseline and after OGD (Fig. 6F). PBS and IL-33-exposed BMECs were equally permeable to FITC-dextran during 6h of OGD (Fig. 6F; red vs. blue triangles). Accordingly, ZO-1 protein expression was reduced 6h after OGD in BMEC cultures after either IL-33 or PBS treatment, as measured by western blotting (Fig. 6G, Figure S1C). IL-33 treatment did not disrupt EC integrity under non-OGD conditions (Fig. 6F, gray circles vs. orange squares). Immunostaining for VE-cadherin and ZO-1 also showed a reduction in BMEC junction proteins after OGD, irrespective of IL-33 treatment (Fig. 6H).

Discussion

BBB disruption is a major detrimental consequence of cerebral ischemia and involves complex molecular interactions between ECs and other cells within the neurovascular unit, including microglia, astrocytes, mural cells, and infiltrating immune cells. However, the nature of the immune responses to stroke can be shifted depending on the particular DAMPs [1]. Microglia and macrophages are mobilized to the site of injury and perform vital functions such as phagocytosis of dead cells, regulation of cerebral blood flow, and initiation of vessel repair. On the other hand, microglia/macrophages also respond to DAMPs by producing reactive oxygen species, cytokines, and chemokines that directly raise BBB permeability. Thus, elucidating the mechanisms that induce protective properties of microglia/macrophages early after stroke may identify new therapeutic strategies to reduce acute ischemic brain injury. The present report indicates that IL-33 secretion in the ischemic brain serves as a DAMP signal for the ST2 receptor and induces a BBB-protective response in infiltrating macrophages, thereby preserving endothelial tight junctions and BBB integrity after stroke, and restricting infarct volumes.

The ST2 receptor is widely expressed in CNS resident cells and infiltrating immune cells. Previous studies from our [16, 18] and other research groups have reported the actions of this signaling axis on different types of cells, including microglia and T lymphocytes [21, 23]. Here, we observed that IL-33/ST2 signaling protects against BBB leakage after ischemic stroke in a microglia/macrophage-dependent manner. Microglia, perivascular macrophages, and infiltrating macrophages have long been implicated in regulating vascular integrity upon brain injuries [10, 26, 27], but our data newly reveal the importance of macrophages in IL-33/ST2-afforded BBB protection in stroke. Chimeric mice housing bone marrow from ST2 KO mice displayed worse BBB leakage than mice reconstituted with WT bone marrow. Bone marrow transplantation after total body irradiation repopulates not only peripheral immune cells but also leptomeningeal and perivascular macrophages with the transplanted genotype [28]. Therefore, we also replaced peripheral monocyte/macrophages in clodronate-treated WT mice with ST2 KO cells, to strengthen the interpretation that activation of the IL-33/ST2 signaling axis in infiltrating macrophages is an endogenous protective mechanism to mitigate BBB damage after stroke. Macrophages are a relatively accessible cell type, an important consideration when designing clinical approaches to target the CNS without breaching the BBB. Thus, the functional significance of the IL-33/ST2 signaling axis in CNS-infiltrating macrophages may allow clinicians to achieve early BBB preservation and brain protection in the future.

As the brain lesion volumes were larger in ST2 global KO mice or mice with ST2 KO in peripheral macrophages, we also tested whether BBB damage is a direct response to loss of the IL-33/ST2 signaling axis or a secondary consequence of lower infarct sizes. We added a group of ST2 KO mice subjected to 40 min of MCAO. ST2 KO mice with 40 min tMCAO exhibited worse BBB leakage, although they developed similar ischemic infarcts as WT mice subjected to 60 min of MCAO (Fig. 1). In addition, when comparing infarct areas and IgG leakage in WT/WT and WT/ST2 KO chimera mice, the extravasation of IgG was greater in the WT/ST2 KO brains than WT/WT brains, even when their infarct sizes were comparable. As a result, the ratio of IgG leakage to infarct sizes was significantly higher in WT/ST2 KO mice compared to WT/WT mice (Fig. 4). These data strongly suggest that loss of ST2 expression resulted in greater BBB leakage independent of infarct size. In vitro studies further confirmed the direct protective effects of IL-33-treated macrophages on BBB upon OGD.

Beyond infiltrating macrophages, the IL-33/ST2 axis may also work via additional cell types of the neurovascular unit, as we observed an upregulation of ST2 expression on brain ECs after ischemic stroke. However, direct application of IL-33 to EC cultures failed to protect against OGD-induced leakage. Some studies report that IL-33 treatment may increase endothelial permeability and angiogenesis in ECs isolated from human umbilical cord veins [29], which may necessitate the development of cell-specific IL-33 delivery strategies. In addition, the average diameter of microvessels in the infarct hemisphere was slightly larger in ST2 KO brains, an effect that may reflect greater contractile pericyte death, which occurs rapidly following cerebral ischemia [2]. Pericytes are major mediators of BBB integrity [30]. Although there is no report on the expression of ST2 in pericytes, it would be worthwhile to explore whether the IL-33/ST2 axis can promote pericyte survival by regulating cell-cell interations between pericytes and ST2+ cells. Astrocytes express ST2 [16] and also lie in close contact with blood vessels, swelling and retracting their endfeet following ischemia/reperfusion, leading to BBB leakage and cerebral edema [31]. Whether the ST2 expression on astrocytes influences BBB integrity also needs to be explored. In addition, the IL-33/ST2 axis has been reported to inhibit the expression of chemokines in macrophages and epithelial cells in lung, thereby reducing the infiltration of neutrophils and ameliorating neutrophil-mediated epithelial barrier injury [32]. Neutrophils are important players in BBB leakage after stroke [33, 34]. It is therefore possible that IL-33/ST2 signaling in macrophages also protects the BBB by preventing the recruitment or activation of infiltrating neutrophils. Although our in vitro experiments using macrophage-EC cocultures confirm protective effects of macrophage IL-33/ST2 signaling on EC barrier function, the possibility that IL-33 orchestrates ST2 signaling in multiple cellular targets to achieve even greater BBB protection deserves to be explored in future studies.

Our PCR array and RT-PCR analyses revealed that IL-33 treatment of macrophages induces expression of Il10. Similarly, our previous study reported that IL-33 enhanced IL-10 production from microglia, which was critical for neuronal protection [16]. IL-10 production by lymphocytes has been shown to reduce stroke severity via immunosuppression [35]. IL-10 production from activated astrocytes also protects the BBB after stroke [36]. Therefore, IL-10 production from IL-33-stimulated macrophages after stroke may contribute to BBB protection. Second, IL-33 treatment elevated the expression of Il13ra2, which encodes a subunit of IL-13 receptor. Il13ra2 drives macrophage differentiation toward a pro-angiogenic/pro-vascularization phenotype [37]. In addition, IL-33 treated macrophages display higher expression of Adrb2, which encodes the β-2-adrenergic receptor. Macrophages are receptive to catecholamines, such that β-2-adrenergic receptor activation downregulates macrophage inflammatory factors (e.g., TNFα and IL-6) to promote tissue remodeling and resolve inflammation [38]. IL-33 increased expression of macrophage Rorc, which encodes the retinoic-acid-related orphan receptor (RORC1/RORγ). RORC1 promotes the expansion of myeloid derived suppressor cells and anti-inflammatory macrophage polarization in tumors [39]. Hence, further research is warranted to test one or more of these mechanisms as mediators of the BBB-protective effects of IL-33/ST2 signaling in macrophages.

Finally, our data demonstrate that intranasal administration of IL-33 mitigates BBB damage and brain infarction early after stroke. Although the current study focuses on the protective effects of IL-33 at the acute/subacute stages of stroke, our previous results suggest that ST2 KO deteriorates long-term neurological functions after stroke [20]. Therefore, further studies are warranted to explore the role of IL-33 at delayed stages of stroke recovery, which will provide a comprehensive evaluation of the therapeutic potential of IL-33. As the intranasal route is a non-invasive CNS delivery route that decreases systemic exposure by bypassing the BBB and avoiding gastrointestinal and hepatic first-pass metabolism [40,41,42], the intranasal delivery of IL-33 will reduce the concern of systemic side effect compared to other routes of systemic delivery. In our study, intranasal IL-33 did not increase mortality early fter stroke, but it is possible that IL33-induced shifts to Th2-type immunosuppressive responses [43] will confound measurements at later timepoints.

Conclusion

The present study provides new evidence that the activation of IL-33/ST2 signaling in infiltrating macrophages decreases BBB leakage, preserves tight junctions between brain ECs, and reduces brain infarcts during ischemia/reperfusion injury. Further development of therapeutic strategies for targeted delivery of IL-33 or selective activation of ST2 in transformable macrophages is warranted to leverage the power of CNS-infiltrating immune cells for stroke treatment.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

BBB:

blood brain barrier

CNS:

central nervous system

DAMPs:

danger-associated molecular patterns

EC:

endothelial cell

IL-33:

interleukin 33

OGD:

oxygen-glucose deprivation

tMCAO:

transient middle cerebral artery occlusion

TJ:

tight junction

References

  1. Hu X, De Silva TM, Chen J, Faraci FM. Cerebral vascular Disease and Neurovascular Injury in ischemic stroke. Circ Res. 2017;120:449–71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Chen S, Shao L, Ma L. Cerebral edema formation after stroke: emphasis on blood-brain barrier and the lymphatic drainage system of the brain. Front Cell Neurosci. 2021;15:716825.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Jiang X, Andjelkovic AV, Zhu L, Yang T, Bennett MVL, Chen J, Keep RF, Shi Y. Blood-brain barrier dysfunction and recovery after ischemic stroke. Prog Neurobiol. 2018;163–164:144–71.

    Article  PubMed  Google Scholar 

  4. Xiao F. Bench to bedside: brain edema and cerebral resuscitation: the present and future. Acad Emerg Medicine: Official J Soc Acad Emerg Med. 2002;9:933–46.

    Article  Google Scholar 

  5. McKeown ME, Prasad A, Kobsa J, Top I, Snider SB, Kidwell C, Campbell BCV, Davis SM, Donnan GA, Lev M, et al. Midline Shift Greater than 3 mm independently predicts Outcome after ischemic stroke. Neurocrit Care. 2022;36:46–51.

    Article  PubMed  Google Scholar 

  6. Simard JM, Kent TA, Chen M, Tarasov KV, Gerzanich V. Brain oedema in focal ischaemia: molecular pathophysiology and theoretical implications. Lancet Neurol. 2007;6:258–68.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Iadecola C, Buckwalter MS, Anrather J. Immune responses to stroke: mechanisms, modulation, and therapeutic potential. J Clin Invest. 2020;130:2777–88.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Gliem M, Mausberg AK, Lee JI, Simiantonakis I, van Rooijen N, Hartung HP, Jander S. Macrophages prevent hemorrhagic infarct transformation in murine stroke models. Ann Neurol. 2012;71:743–52.

    Article  CAS  PubMed  Google Scholar 

  9. Szalay G, Martinecz B, Lenart N, Kornyei Z, Orsolits B, Judak L, Csaszar E, Fekete R, West BL, Katona G, et al. Microglia protect against brain injury and their selective elimination dysregulates neuronal network activity after stroke. Nat Commun. 2016;7:11499.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Mastorakos P, Mihelson N, Luby M, Burks SR, Johnson K, Hsia AW, Witko J, Frank JA, Latour L, McGavern DB. Temporally distinct myeloid cell responses mediate damage and repair after cerebrovascular injury. Nat Neurosci. 2021;24:245–58.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Dimitrijevic OB, Stamatovic SM, Keep RF, Andjelkovic AV. Effects of the chemokine CCL2 on blood-brain barrier permeability during ischemia-reperfusion injury. J Cereb Blood Flow Metab. 2006;26:797–810.

    Article  CAS  PubMed  Google Scholar 

  12. Shi Y, Zhang L, Pu H, Mao L, Hu X, Jiang X, Xu N, Stetler RA, Zhang F, Liu X, et al. Rapid endothelial cytoskeletal reorganization enables early blood-brain barrier disruption and long-term ischaemic reperfusion brain injury. Nat Commun. 2016;7:10523.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Drake LY, Kita H. IL-33: biological properties, functions, and roles in airway disease. Immunol Rev. 2017;278:173–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Dwyer GK, D’Cruz LM, Turnquist HR. Emerging functions of IL-33 in Homeostasis and Immunity. Annu Rev Immunol. 2022;40:15–43.

    Article  CAS  PubMed  Google Scholar 

  15. Liew FY, Pitman NI, McInnes IB. Disease-associated functions of IL-33: the new kid in the IL-1 family. Nat Rev Immunol. 2010;10:103–10.

    Article  CAS  PubMed  Google Scholar 

  16. Yang Y, Liu H, Zhang H, Ye Q, Wang J, Yang B, Mao L, Zhu W, Leak RK, Xiao B, et al. ST2/IL-33-Dependent microglial response limits Acute Ischemic Brain Injury. J Neurosci. 2017;37:4692–704.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Gadani SP, Walsh JT, Smirnov I, Zheng J, Kipnis J. The glia-derived alarmin IL-33 orchestrates the immune response and promotes recovery following CNS injury. Neuron. 2015;85:703–9.

    Article  CAS  PubMed  Google Scholar 

  18. Xie D, Miao W, Xu F, Yuan C, Li S, Wang C, Junagade A, Hu X. IL-33/ST2 Axis protects against Traumatic Brain Injury through Enhancing the Function of Regulatory T Cells. Front Immunol. 2022;13:860772.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Li T, Zhang Z, Bartolacci JG, Dwyer GK, Liu Q, Mathews LR, Velayutham M, Roessing AS, Lee YC, Dai H, et al. Graft IL-33 regulates infiltrating macrophages to protect against chronic rejection. J Clin Invest. 2020;130:5397–412.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Xie D, Liu H, Xu F, Su W, Ye Q, Yu F, Austin TJ, Chen J, Hu X. IL33 (interleukin 33)/ST2 (interleukin 1 receptor-like 1) Axis drives protective microglial responses and promotes White Matter Integrity after Stroke. Stroke. 2021;52:2150–61.

    Article  CAS  PubMed  Google Scholar 

  21. Liu X, Hu R, Pei L, Si P, Wang C, Tian X, Wang X, Liu H, Wang B, Xia Z, et al. Regulatory T cell is critical for interleukin-33-mediated neuroprotection against stroke. Exp Neurol. 2020;328:113233.

    Article  CAS  PubMed  Google Scholar 

  22. Xiao W, Guo S, Chen L, Luo Y. The role of Interleukin-33 in the modulation of splenic T-cell immune responses after experimental ischemic stroke. J Neuroimmunol. 2019;333:576970.

    Article  CAS  PubMed  Google Scholar 

  23. Korhonen P, Kanninen KM, Lehtonen S, Lemarchant S, Puttonen KA, Oksanen M, Dhungana H, Loppi S, Pollari E, Wojciechowski S, et al. Immunomodulation by interleukin-33 is protective in stroke through modulation of inflammation. Brain Behav Immun. 2015;49:322–36.

    Article  CAS  PubMed  Google Scholar 

  24. Mao L, Li P, Zhu W, Cai W, Liu Z, Wang Y, Luo W, Stetler RA, Leak RK, Yu W, et al. Regulatory T cells ameliorate tissue plasminogen activator-induced brain haemorrhage after stroke. Brain. 2017;140:1914–31.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Xu Z, Rao Y, Huang Y, Zhou T, Feng R, Xiong S, Yuan TF, Qin S, Lu Y, Zhou X, et al. Efficient strategies for Microglia replacement in the Central Nervous System. Cell Rep. 2020;32:108041.

    Article  CAS  PubMed  Google Scholar 

  26. Mondo E, Becker SC, Kautzman AG, Schifferer M, Baer CE, Chen J, Huang EJ, Simons M, Schafer DP. A Developmental Analysis of Juxtavascular Microglia Dynamics and interactions with the vasculature. J Neurosci. 2020;40:6503–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Santisteban MM, Ahn SJ, Lane D, Faraco G, Garcia-Bonilla L, Racchumi G, Poon C, Schaeffer S, Segarra SG, Korbelin J, et al. Endothelium-macrophage crosstalk mediates blood-brain barrier dysfunction in hypertension. Hypertension. 2020;76:795–807.

    Article  CAS  PubMed  Google Scholar 

  28. Santisteban MM, Schaeffer S, Anfray A, Faraco G, Brea D, Wang G, Sobanko MJ, Sciortino R, Racchumi G, Waisman A, et al. Meningeal interleukin-17-producing T cells mediate cognitive impairment in a mouse model of salt-sensitive hypertension. Nat Neurosci. 2024;27:63–77.

    Article  CAS  PubMed  Google Scholar 

  29. Choi YS, Choi HJ, Min JK, Pyun BJ, Maeng YS, Park H, Kim J, Kim YM, Kwon YG. Interleukin-33 induces angiogenesis and vascular permeability through ST2/TRAF6-mediated endothelial nitric oxide production. Blood. 2009;114:3117–26.

    Article  CAS  PubMed  Google Scholar 

  30. Armulik A, Genove G, Mae M, Nisancioglu MH, Wallgard E, Niaudet C, He L, Norlin J, Lindblom P, Strittmatter K, et al. Pericytes regulate the blood-brain barrier. Nature. 2010;468:557–61.

    Article  CAS  PubMed  Google Scholar 

  31. Patabendige A, Singh A, Jenkins S, Sen J, Chen R. Astrocyte activation in neurovascular damage and repair following ischaemic stroke. Int J Mol Sci 2021, 22.

  32. Ma Q, Qian Y, Jiang J, Wu J, Song M, Li X, Chen Z, Wang Z, Zhu R, Sun Z, et al. IL-33/ST2 axis deficiency exacerbates neutrophil-dominant allergic airway inflammation. Clin Transl Immunol. 2021;10:e1300.

    Article  CAS  Google Scholar 

  33. Jickling GC, Liu D, Ander BP, Stamova B, Zhan X, Sharp FR. Targeting neutrophils in ischemic stroke: translational insights from experimental studies. J Cereb Blood Flow Metab. 2015;35:888–901.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. He Q, Wang Y, Fang C, Feng Z, Yin M, Huang J, Ma Y, Mo Z. Advancing stroke therapy: a deep dive into early phase of ischemic stroke and recanalization. CNS Neurosci Ther. 2024;30:e14634.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Piepke M, Clausen BH, Ludewig P, Vienhues JH, Bedke T, Javidi E, Rissiek B, Jank L, Brockmann L, Sandrock I, et al. Interleukin-10 improves stroke outcome by controlling the detrimental Interleukin-17A response. J Neuroinflammation. 2021;18:265.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Suo Q, Deng L, Chen T, Wu S, Qi L, Liu Z, He T, Tian HL, Li W, Tang Y, et al. Optogenetic activation of astrocytes reduces blood-brain barrier disruption via IL-10 in stroke. Aging Dis. 2023;14:1870–86.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Xu N, Bo Q, Shao R, Liang J, Zhai Y, Yang S, Wang F, Sun X. Chitinase-3-Like-1 promotes M2 macrophage differentiation and induces Choroidal Neovascularization in Neovascular Age-Related Macular Degeneration. Invest Ophthalmol Vis Sci. 2019;60:4596–605.

    Article  CAS  PubMed  Google Scholar 

  38. Freire BM, de Melo FM, Basso AS. Adrenergic signaling regulation of macrophage function: do we understand it yet? Immunother Adv. 2022;2:ltac010.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Strauss L, Sangaletti S, Consonni FM, Szebeni G, Morlacchi S, Totaro MG, Porta C, Anselmo A, Tartari S, Doni A, et al. RORC1 regulates tumor-promoting emergency Granulo-Monocytopoiesis. Cancer Cell. 2015;28:253–69.

    Article  CAS  PubMed  Google Scholar 

  40. Erdo F, Bors LA, Farkas D, Bajza A, Gizurarson S. Evaluation of intranasal delivery route of drug administration for brain targeting. Brain Res Bull. 2018;143:155–70.

    Article  CAS  PubMed  Google Scholar 

  41. Kozlovskaya L, Abou-Kaoud M, Stepensky D. Quantitative analysis of drug delivery to the brain via nasal route. J Control Release. 2014;189:133–40.

    Article  CAS  PubMed  Google Scholar 

  42. Fitzgerald GS, Chuchta TG, McNay EC. Insulin-like growth factor-2 is a promising candidate for the treatment and prevention of Alzheimer’s disease. CNS Neurosci Ther. 2023;29:1449–69.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Zhang SR, Piepke M, Chu HX, Broughton BR, Shim R, Wong CH, Lee S, Evans MA, Vinh A, Sakkal S et al. IL-33 modulates inflammatory brain injury but exacerbates systemic immunosuppression following ischemic stroke. JCI Insight 2018, 3.

Download references

Acknowledgements

Not applicable.

Funding

This work was supported by a VA Merit Review grant (I01 BX003651) and NIH/NINDS grants (NS131169 and NS124673) to Xiaoming Hu.

Author information

Author notes

  1. Miao Wang and Connor Dufort contributed equally to this work.

Authors and Affiliations

  1. Geriatric Research, Education and Clinical Center, Veterans Affairs Pittsburgh Health Care System, Pittsburgh, PA, 15261, USA

    Miao Wang, Fei Xu, Zhentai Huang & Xiaoming Hu

  2. Department of Neurology, School of Medicine, University of Pittsburgh, 200 Lothrop Street, SBST 506, Pittsburgh, PA, 15213, USA

    Miao Wang, Connor Dufort, Zhihong Du, Fei Xu, Zhentai Huang, Ana Rios Sigler & Xiaoming Hu

  3. Department of Human Genetics, School of Public Health, University of Pittsburgh, Pittsburgh, PA, 15261, USA

    Ruyu Shi

  4. Division of Pharmaceutical Sciences, Duquesne University, Pittsburgh, PA, 15282, USA

    Rehana K. Leak

Authors
  1. Miao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Connor Dufort
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Zhihong Du
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ruyu Shi
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Fei Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Zhentai Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Ana Rios Sigler
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Rehana K. Leak
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Xiaoming Hu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

MW, CD, and XH wrote the main manuscript text. MW, CD, ZD, FX, ZH, RS and ARS performed experiments and data analyses. MW and XH prepared figures. RKL provided scientific feedback and revised the manuscript. XH designed and supervised the study. All authors reviewed the manuscript.

Corresponding author

Correspondence to Xiaoming Hu.

Ethics declarations

Ethical approval

All animal procedures were approved by the University of Pittsburgh Institutional Animal Care and Use Committee and performed following the Guide for the Care and Use of Laboratory Animals.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, M., Dufort, C., Du, Z. et al. IL-33/ST2 signaling in monocyte-derived macrophages maintains blood-brain barrier integrity and restricts infarctions early after ischemic stroke. J Neuroinflammation 21, 274 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-024-03264-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-024-03264-8

Keywords

Journal of Neuroinflammation

ISSN: 1742-2094