PD-1 mediates microglia polarization via the MAPK signaling pathway to protect blood-brain barrier function during cerebral ischemia/reperfusion

Background: Cerebral ischemia is characterized by its rapid onset and high rates of recurrence, morbidity, and mortality, with blood-brain barrier (BBB) permeability playing a vital role in brain injury. Therefore, it is important to understand the molecular mechanism which regulates the BBB during cerebral ischemia. Materials and methods: An in vitro model of oxygen-glucose deprivation (OGD) and an in vivo model of cerebral ischemia/reperfusion (I/R) were constructed. PD-1 overexpression vectors and vectors containing si-RNA were transfected and injected into in vitro and in vivo models. Western blotting, real-time quantitative PCR (qPCR), immunofluorescence (IF) analysis, and immunohistochemical staining were employed to evaluate the expression levels of programmed cell death-1 (PD-1), microglia M1 and M2 biomarkers, and tight junction proteins. Flow cytometry and ELISA were used to measure the levels of pro-inflammatory cytokines. The BBB permeability of brain tissues was evaluated by Evans blue dye (EBD) extravasation and transendothelial electrical resistance (TEER). Brain water content was measured to assess the extent of inflammatory exudation. The infarct volume and neurological severity score (NSS) were used to assess the severity of brain injury. Brain cell apoptosis was assessed by the TUNEL assay and hematoxylin-eosin (H & E) staining. Results: PD-1 helped to convert the microglia M1 phenotype to the M2 phenotype and to reduce BBB permeability both in vitro and in vivo . Overexpression of PD-1 promoted a shift of the M1 phenotype to the M2 phenotype and reduced BBB permeability via the ERK and p38 MAPK signaling pathways. PD-1 reduced inflammatory exudation, BBB permeability, cell apoptosis, and brain injury in vivo . Conclusion: Our present study verified that PD-1 exerts an anti-inflammatory effect by converting the microglia M1 phenotype to the M2 phenotype, reducing BBB permeability, and thereby relieves brain injury caused by cerebral ischemia. PD-1 is potential therapeutic target for brain injury caused by cerebral ischemia.


Introduction
Cerebral ischemia disease is a common cause of stroke and has high rates of morbidity and mortality worldwide (Jurcau and Simion, 2021).The disease is caused by blockage of a blood vessel by a thrombus or embolus, resulting in insufficient blood flow to the brain and a resultant decrease in the supply of oxygen and other nutrients to brain tissue (Dodd et al., 2021).The blood-brain barrier (BBB) is a complex cellular system that exists between brain tissue and circulating blood.It controls the transport of substances from circulating blood into central nervous tissue, and thereby ensures the environmental stability of central nervous tissue (Zhang et al., 2020).The inflammatory coupling reaction that occurs during cerebral ischemia damages the BBB damage.This damage allows for external ingredients to enter capillaries, and increases the amount of water in contained in intercellular spaces to form cerebral edema; which in turn, increases the amount of brain tissue injury (Jurcau and Simion, 2021).After ischemia-reperfusion of cerebral ischemia, many factors contribute to damage of the BBB, and changes in BBB function contribute to pathological changes such as edema and inflammation following cerebral ischemia.Therefore, detailed studies of factors that influence the BBB can help to optimize the clinical treatment of cerebral ischemia.
Programmed cell death-1 (PD-1) is a cell surface receptor widely expressed in immune cells and other cell types, including microglia (Zhao et al., 2021).It is widely used in cancers as one immune checkpoint, and is also implicated in central nervous system (CNS) disorders, including ischemic stroke, Alzheimer's disease, cognitive function disorders, and spinal cord injury.PD-1 signaling can suppress the CNS immune response of microglia, which might be a novel strategy for treating brain diseases.As reported, microglia activation and polarization influence cerebral ischemic injury by inducing pro-inflammatory or anti-inflammatory effects (Dong et al., 2021;Lyu et al., 2021).Microglia are vital components of the neurovascular unit, which when activated, promote the release of pro-inflammatory cytokines such as interleukin-6 (IL-6), IL-1β, and tumor necrosis factor-α (TNF-α), which are detrimental to BBB integrity during the early phase of ischemia (da Fonseca et al., 2014;Richter et al., 2017;Chen et al., 2019).Therefore, in depth studies of PD-1 are of great importance for treatment of cerebral ischemia.
Like macrophages, microglia also can be activated into M1/M2 polarization, which plays an important role in cerebral ischemic injury.Previous study has reported that the activation of PD-1/programmed death ligand 1 (PD-L1) signaling induced macrophage M2 polarization in the tumor microenvironment of melanoma (Liu et al., 2021).Trichinella spiralis adult worms -induced M2 polarization was inhibited in PD-1 gene knockout mice (Wang et al., 2020), which suggested that PD-1 mediates M2 macrophage polarization.A recent study indicated that PD-1/PD-L1 axis is involved in the interaction between microglial polarization and glioma (Wang et al., 2024).However, the role of PD-1-mediated M2 microglial polarization in cerebral ischemia remains unclear.The mitogen-activated protein kinase (MAPK) pathway is a vital signal transfer system that mediates extracellular signals from various growth factors and cytokines to induce internal cell reactions needed for the cell progression process.It also activates the microglia, and thus mediates the occurrence of CNS disorders and neuropathic pain (Wu et al., 2023).The RhoA/p38MAPK signaling pathway participates in regulating the transformation of microglia from the M1 to the M2 phenotype and the polarization process, and thus helps to relieve neuropathic pain (Wu et al., 2023;Li et al., 2023a).Moreover, the MAPK signaling pathway is also associated with PD-1 in regulating various diseases, and especially cancers (Li et al., 2023b;Pokhrel et al., 2021).However, whether PD-1 is involved in BBB function by regulating microglia polarization via the MAPK signaling pathway remains unknown.
In this study, we investigated the role of PD-1 in regulating microglia M1 and M2 phenotypes, as well as BBB integrity and permeability under conditions of oxygen-glucose deprivation (OGD) in vitro and cerebral ischemia/reperfusion (I/R) in vivo.Moreover, we also assessed whether the p38 MAPK signaling pathway participates in the molecular mechanism by which PD-1 affects brain injury due to cerebral ischemia.

OGD model construction
HMC3 human microglial cells were purchased from the American Type Culture Collection (Manassas, VA, USA) and cultured in medium containing 10 % FBS and 1 % antibiotics in a 5 % CO 2 atmosphere at 37℃.Next, the cells were treated with OGD to create an in vitro I/R injury model.Control cells were cultured in DMEM under normoxic conditions (5 % CO 2 and 21 % O 2 ) for 24 h.Cells in the OGD model group were cultured in serum-free DMEM in an atmosphere of 95 % N 2 and 5 % CO 2 for 5 h, and subsequently incubated under normoxic conditions for 6 h.

Animals and cerebral I/R model
Sprague-Dawley male rats (n = 30; age = 8 weeks, weight range, 200-250 g) were obtained from the Guangdong Medical Experimental Animal Center (Guanghzou, China) and used for the present study.The rats were housed in separate cages in room with a 12 h:12 h light/dark cycle at room temperature.The study protocol was approved by the Guangdong Provincial People's Hospital [No. GDREC2017100A].

Real-time Quantitative PCR (qPCR)
The total RNA in cells and transfected cells was extracted using TRIzol reagent (Invitrogen, USA).The primers used for PD-1, TNF-α, IL-10, and β-actin were designed and synthesized by Sangon (Shanghai, China).An HiScript II One Step qRT-PCR SYBR Green Kit (#Q221-01) was used for amplification, and the products were analyzed on one ABI 7900 Real-time PCR system (Applied Biosystems, Foster City, CA, USA), Relative levels of gene expression were calculated using the 2 -ΔΔCt method.

Enzyme linked immunosorbent assay (ELISA)
Cells were collected and lysed with RIPA buffer and then centrifuged at 12,000 rpm for 15 min at 4 • C. The levels of TNF-α and IL-10 were measured with commercial kits (Shanghai Boyun Biotech Co., Ltd.) according to the manufacturer's instructions.

Flow cytometry
Approximately 2 × 10 6 cells were suspended and stained with PEconjugated antibodies against iNOS, CD86, Arg1, and CD206 at 4℃ for 30 min in the dark.The cells were then washed with PBS and detected with a FACSCalibur™ flow cytometer (BD Biosciences, San Jose, CA, USA).The data were processed using FlowJo software (Tree Star, San Carlos, CA, USA).

Evans blue dye (EBD) extravasation
EBD extravasation was used to evaluate BBB permeability in both cells and brain tissues.An Evans blue-bovine serum albumin (EB-BSA) working solution was prepared with 2 % EB solution and 4 % BSA, and then diluted with PBS to the desired concentration.Transwell chambers were washed with PBS and inserted into a 24-well plate.Next, 100 μL of EB-BSA and 600 μL of 4 % BSA were added to the top and bottom chambers, respectively, and the plates were cultured at 37℃ in a 5 % CO 2 atmosphere or 1 h.After culture, the liquid in the bottom chamber was removed and centrifuged at 1000 rpm for 5 min.The EBD absorbance at 620 nm was measured by EBD fluorescence.

Transendothelial electrical resistance (TEER)
TEER is considered to be a marker for evaluating brain permeability.Microglia cells were digested and counted.Next, approximately 1 × 10 5 cells were seeded into a Transwell chamber that was inserted into a 24well plate that was maintained at 37℃ in 5 % CO 2 atmosphere.A Mil-licell® ERS-2 Voltohmmeter (Merck Millipore, Burlington, MA, USA) was used to measure the cellular monolayers once every two days.The baseline reading was recorded, and the TEER was calculated.

Brain water content measurement
After treatment, the rats were sacrificed, and their brain water content was measured to assess the extent of inflammatory exudation.The wet weight of each brain was determined immediately after the brain tissues were removed.The dry weight of each brain was recorded following four days of incubation at 60℃; after which, the rat brain water content was calculated.

Infarct volume assessment
Brain infarct volume was calculated after reperfusion in vivo.Slices of rat brain were stained with 2 % 2,3,5-triphenyltetrazolium chloride (TTC) at 37 • C for 30 min, and then fixed in 4 % paraformaldehyde for 12 h.Images of TTC-stained sections were observed under a digital scanner.Normal areas of brain tissue were stained red, and brain infarct lesions appeared as white areas.

Neurological severity score (NSS)
NSS scores were determined by two blinded observers who used scoring criteria described in previous studies (Boyko et al., 2013;Zlotnik et al., 2012).Briefly, the scores were assigned based motor functions and behavior.The scores ranged from 0 to 25; representing an intact neurological condition to the greatest neurological dysfunction.

Hematoxylin-eosin (H&E) and immunohistochemical (IHC) staining
Sections of rat brain tissue were stained with H&E to evaluate the tissue injury.
For IHC staining, the sections were incubated with primary antibodies against PD-1, iNOS, TNF-α, IL-10, CD86, and CD206 at 4℃ for 12 h, and subsequently incubated with an HRP-labeled goat anti-rabbit IgG secondary antibody at room temperature for 40 min.The numbers of positive cells were then calculated.

TUNEL staining
An In-Situ Cell Death Detection Kit (Roche) was used to perform TUNEL assays.The red fluorescent-tagged enzyme solution in the assay kit was added to cells, which were then analyzed under a fluorescence microscope (DM2500; Leica Microsystems, Germany).Five high-power lens visual fields were randomly selected and evaluated in each section.TUNEL-positive cells with red fluorescent were observed using Image-Pro Plus version 6.0 software (Media Cybernetics, USA).After each experiment, images of DAPI-and TUNEL-stained cells were recorded and merged.

Statistical analysis
GraphPad Prism 10 (San Diego, CA, USA) software was employed to draw images and analyze data using the unpaired student's t-test.Statistical results are presented as a mean value ± standard deviation (SD).A p-value < 0.05 was considered to be statistically significant.

PD-1 helped to convert the microglia M1 phenotype to the M2 phenotype and reduce BBB permeability in vitro
We evaluated the role of PD-1 in regulating microglia M1/M2 polarization and BBB permeability under conditions of PD-1 overexpression and OGD.First, the levels of PD-1 expression were measured by western blotting, qPCR, and IF.The results showed that OGD significantly decreased PD-1 expression (P < 0.001, Fig. 1A-1 C).When cells were transfected with the PD-1 overexpression vector, the levels of PD-1 expression were significantly increased, indicating a successful transfection.Next, biomarkers for type M1 (iNOS, CD86, and TNF-α) and type M2 (Arg1, CD206, and IL-10) microglia were measured by western blot, qPCR, ELISA, and flow cytometry assays.Those results showed that OGD significantly increased the levels of iNOS and Arg1 expression when compared with those levels in control cells (P < 0.001, Fig. 1A).When compared with the OGD+O-NC group, iNOS expression was significantly decreased and Arg1 expression was significantly increased (P < 0.001).QPCR results indicated that TNF-α gene expression was significantly decreased and IL-10 gene expression was significantly increased when PD-1 was overexpressed (P < 0.001; Fig. 1B).Flow cytometry assays showed that iNOS (P < 0.05) and CD86 (P < 0.01) expression were significantly decreased, and Arg1 (P < 0.001) and CD206 (P < 0.001) expression were significantly increased in the OGD+O-PD1 group when compared to the OGD+O-NC group (Fig. 1D).The concentrations of TNF-α were significantly decreased (P < 0.001), while those of IL-10 were significantly increased (P < 0.001) under conditions of PD-1 overexpression (Fig. 1E), which was consistent with the qPCR results.Therefore, we concluded that the microglia M1 phenotype was converted to the M2 phenotype under conditions of PD-1 overexpression in the OGD model, thus showing its anti-inflammation function.
We next assessed the influence of PD-1 expression on BBB function.First, we evaluated BBB permeability in the OGD models and transfections.When compared with the Control group, the EB-BSA concentrations in OGD group were significantly increased, indicating that the BBB was severely damaged in the OGD model, but was recovered by overexpression of PD-1 (P < 0.001; Fig. 1F).As a physical parameter, TEER is employed to characterize BBB barrier tightness.Our data showed that OGD significantly decreased the TEER values when compared with the control (P < 0.001, Fig. 1G).In addition, the TEER values were higher in the OGD+O-PD-1 group than in the OGD+O-NC group (P < 0.05).Expression of the tight junction proteins claudin-5 and ZO1 was significantly decreased in the OGD model when compared with the Control (P < 0.001) and significantly increased in OGD+O-PD-1 group when compared with the OGD+O-NC group (P < 0.001, Fig. 1H).These results demonstrated that overexpression of PD-1 reduced BBB permeability, which might help to protect BBB function.

Overexpression of PD-1 promoted a shift from the M1 phenotype to the M2 phenotype and reduced BBB permeability via the ERK and p38 MAPK signaling pathways
To evaluate the role of PD-1 in regulating the MAPK signaling pathway, we assessed the expression levels of PD-1, iNOS, and Arg1 in the OGD+O-NC, OGD+O-PD1, and OGD+O-PD1+p38 MAPK-IN-1 groups.Results showed that a p38 MAPK inhibitor did not significantly dysregulate PD-1 (P > 0.05, Fig. 2A).Next, the effects of the p38 MAPK inhibitor on microglial M1 and M2 polarization were assessed by ELISA, flow cytometry, and qPCR.The levels of iNOS, TNF-α, and CD86 were significantly increased (P < 0.001), and those of Arg1, CD206, and IL-10 were significantly decreased in the OGD+O-PD1+p38 MAPK-IN-1 group when compared with the OGD+O-PD1 group (P < 0.001, Fig. 2A-2D), indicating an increased number of M1 microglia and a decreased number of M2 microglia.These results demonstrated that the p38 MAPK inhibitor blocked the promoting effect of PD-1 on regulating microglia M1/M2 polarization.Western blot analyses showed that the levels of p-ERK1/2 and p-p38 were significantly increased in the OGD+O-PD1 group when compared with the OGD+O-NC group, and were significantly decreased in the OGD+O-PD1+p38 MAPK-IN-1 group when compared with the OGD+O-PD1 group (P < 0.001, Fig. 2E).The results revealed that both the ERK1/2 and p38 MAPK signaling pathways were involved in PD-1 regulation.
Next, we assessed the effects of the p38 MAPK inhibitor on BBB permeability.The EB-BSA concentrations were significantly decreased in the OGD+O-PD1+p38 MAPK-IN-1 group when compared with the OGD+O-PD1 group (P < 0.001, Fig. 2F).When compared with the OGD+O-PD-1 group, the TEER values were significantly decreased after treatment with the p38 MAPK inhibitor (P < 0.001, Fig. 2G).The expression levels of claudin-5 and ZO1 were significantly decreased in the OGD+O-PD1+p38 MAPK-IN-1 group when compared with the OGD+O-PD1 group (P < 0.001, Fig. 2H).These results demonstrated that overexpression of PD-1 promoted a shift of the M1 phenotype to the M2 phenotype and reduced BBB permeability by affecting the ERK and p38 MAPK signaling pathways.

PD-1 inhibited BBB permeability, cell apoptosis, and brain injury in vivo
Brain water content was measured to assess the extent of inflammatory exudation.Our calculations showed that brain water content was significantly increased in the I/R model rats when compared with the Sham group (P < 0.05, Fig. 3A), indicating that the rat cerebral I/R model had been successfully established.The brain water content of rats treated with sh-PD-1 (AAV-KD group) was significantly increased when compared with rats in the AAV-EV kd group (P < 0.05).In addition, the brain tissues in the I/R group were stained darker than those in the Sham group (Fig. 3B).We observed that brains in the AAV-KD group had the darkest staining, while brains in the AAV-OE group showed almost normal staining, indicating that PD-1 reduced BBB permeability.TTC staining showed that brain infarct size was different among groups (Fig. 3C).The infarct volume percent was higher in the AAV-KD group than in the AAV-EVkd group.Rats with PD-1 overexpression showed a small infarct size, which almost recovered to the level of Sham rats.These results suggested that PD-1 decreased infarct size.We also evaluated brain injury by using the NSS.We found that the NSSs were significantly higher in the I/R group than in the Sham group, The scores were also higher in the AAV-KD group than in the AAV-EV kd group, and lower in the AAV-OE group when compared to the AAV-EVOE group (P < 0.001, Fig. 3D).Furthermore, the cells in the Sham and AAV-OE groups showed a bright side on neatly arranged neurons, dense cytoplasm, and an intact nucleus, while the cells were severely damaged in the AAV-KD group (Fig. 3E).When compared with control cells, the numbers of TUNEL-positive cells were increased in I/R groups, indicating enhanced levels of cell apoptosis in I/R rat models (Fig. 3F).We also found that cell apoptosis was promoted by sh-PD-1 and inhibited by PD-1 overexpression.These results indicated that PD-1 helps to inhibit inflammatory exudation and reduce BBB permeability, cell apoptosis, and brain injury.

PD-1 helped to convert the microglia M1 phenotype to the M2 phenotype and reduce BBB permeability in vivo
Next, we further evaluated the effect of PD-1 on the M1/M2 ratio and BBB permeability in vivo.In the in vivo models, the levels of PD-1, claudin-5, and ZO1, as well as the levels of microglial M1 biomarkers (iNOS, TNF-α, and CD86) and M2 biomarkers (Arg1, CD206, and IL-10) were assessed by western blotting, qPCR, and IF.Western blotting showed that PD-1 levels were significantly decreased in the AAV-KD group when compared with AAV-EV kd group, and were increased in the AAV-OE group when compared with the AAV-EVOE group, demonstrating the success of knockdown and overexpression treatments in vivo (P < 0.001, Fig. 4A).Western blot results further showed that the expression levels of TNF-α (an M1 biomarker) were significantly increased by sh-PD-1 and reduced by PD-1 overexpression (P < 0.001), while the levels of IL-10 (an M2 biomarker) were significantly reduced by sh-PD-1 and increased by PD-1 overexpression (P < 0.001).Furthermore, claudin-5 and ZO1 expression were significantly increased by PD-1 overexpression and reduced by sh-PD-1 (P < 0.001, Figs.4A  and 4B), showing the inhibitory effect of PD-1 on BBB permeability.In addition, IF assays revealed that the levels of claudin-5 protein were consistent with those of claudin-5 mRNA (Fig. 4C).
The M1 biomarkers (iNOS, TNF-α, and CD86) and M2 biomarkers (CD206, and IL-10) were also assessed by IHC.IHC staining showed that the numbers of iNOS-, TNF-α-, and CD86-positive cells were increased by sh-PD-1 treatment and decreased by PD-1 overexpression (Fig. 5A-5 C), while the biomarkers for M2 microglia exhibited the opposite results (Fig. 5D and E).When taken together, the above results demonstrated that PD-1 reduced the microglia M1/M2 ratio and BBB permeability in vivo.

Discussion
Cerebral ischemia is characterized by its rapid onset and high rates of recurrence, morbidity, and mortality (Shen et al., 2021).BBB leakage plays a vital role in cerebral ischemia-reperfusion injury.Therefore, it is important to understand the molecular mechanism which regulates the BBB during cerebral ischemia.Our present study revealed that the M1 polarization of microglial cells decreased PD-1 secretion during an ischemia-reperfusion injury, which might be one of the causes of BBB damage.Furthermore, PD-1 might help to protect BBB function.
Recent studies have focused on microglia, because they have been found to be the driver of many neurological and neurodegenerative diseases.Microglia under undergo morphological and phenotypic transformations depending on their microenvironment (Hensley et al.,  ).Microglia-mediated neuroinflammation is a potential target for the treatment of neurodegenerative diseases, and works by converting the M1 and M2 phenotypes (Biswas, 2023;Kwon and Koh, 2020).However, such transformation is considered to be a double-edged sword, because M1 microglia release inflammatory mediators and M2 microglia release anti-inflammatory mediators that exert both harmful and helpful effects in cerebral ischemia (Guo et al., 2022).Therefore, molecules that can shift microglia M1 and M2 phenotypes might be essential for the development of certain diseases.In our present study, microglia in rat and cell models appeared to switch from an M1 phenotype observed at the beginning of pathology to an M2 phenotype after treatment with PD-1 overexpression, This conversion was accompanied by increased levels of CD86, NOS, and TNF-α (a pro-inflammatory cytokine).
PD-1 is a transmembrane glycoprotein, that when combined with its ligand (PD-L1), plays essential roles in the pathogenesis of autoimmune diseases (Manenti et al., 2022).PD-1/PD-L1 can suppress and attenuate the immune response, and exert a negative feedback-regulating effect on inflammation (Martinic and von Herrath, 2008).The PD-1/PD-L1 pathway can negatively regulate inflammation after a stroke via intercellular interaction, and thereby provide protective effects (Lin et al., 2012;Shi et al., 2015).Overexpression of PD-1/PD-L1 can decrease the inflammatory response and reduce secondary brain injury after intracerebral hemorrhage (ICH) (Wu et al., 2017).A recent study revealed a similar effect of PD-1 in ICH, in which the authors demonstrated that increased PD-1 expression around the lesion promoted a shift of microglia from the M1 phenotype to the M2 phenotype (Liu et al., 2023).Moreover, Yao et al (Yao et al., 2014).showed that PD-1 deficiency induces macrophage/microglia M1 polarization after a spinal cord injury.Our present study showed results that were similar to those of other studies on the conversion of microglia M1 and M2 phenotypes.However, most of the current studies have relied animal models, and numerous other studies will be required before microglia phenotype conversion can be used as a therapeutic strategy in the clinic.
Inflammatory cytokines transmit their inflammatory signals to the central nervous system, where they activate microglia and astrocytes to produce reactive oxygen species, nitrogen, inflammatory cytokines, and a series of neurotoxic molecules that adversely affect BBB integrity (Huang et al., 2020).In our present study, we found that PD-1 activates anti-inflammatory cytokines in microglia, protects the BBB integrity, and therefore reduces brain injury.As the main interface between the central nervous system and circulatory system, changes in the BBB may either protect the central nervous system or induce diseases (Xiao et al., 2020).Our present study suggests PD-1 as a therapeutic target for cerebral ischemia patients.
The mechanisms which regulate PD-1 and p38 MAPK signaling have been extensively studied in cancers.For example, a blockade of p38 MAPK signaling or PD-1 was found to promote the proliferation of senescent human CD8+ T-cells by distinct pathways (Henson et al., 2015).A PD-1/PD-L1 binding inhibitor was shown to increase cell apoptosis via the p38 MAPK signaling pathway in the pituitary gland and reproductive system (Jiang et al., 2022).However, no study has investigated the role of PD-1 and p38 MAPK signaling in cerebral ischemia.Our present study revealed for the first time that overexpression of PD-1 shifted the microglia M1 phenotype to the M2 phenotype and reduced BBB permeability via the ERK and p38 MAPK signaling pathways.

Conclusion
Our present study verified that PD-1 exerts an anti-inflammation function by converting the microglia M1 phenotype to the M2 phenotype, reducing BBB permeability, and thereby relieving brain injury caused by cerebral ischemia.PD-1 regulated microglia polarization maybe through MAPK pathway.We also identified a potential therapeutic target for cerebral ischemia.

Fig. 1 .
Fig. 1.PD-1 helped to convert the microglia M1 phenotype to the M2 phenotype and reduce BBB permeability in vitro.(A) Western blot analyses were performed to evaluate the levels of PD-1, iNOS, and Arg1 expression.(B) QPCR analysis was performed to evaluate the levels of PD-1, TNF-α, and IL-10 expression.(C) An IF analysis was conducted to assess the PD-1 expression.(D) Flow cytometry assays were performed to measure the levels of iNOS, CD86, Arg1, and CD206.(E) ELISA assays were performed to evaluate the levels of TNF-α and IL-10.(F) BBB permeability for the OGD model and transfections were assessed by EBD extravasation.(G) The TEER value was calculated.(H) Western blot analyses were performed to evaluate the levels of claudin-5 and ZO1 expression.*P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2 .
Fig. 2. Overexpression of PD-1 promoted a shift of the M1 phenotype to the M2 phenotype and reduced BBB permeability via the ERK and p38 MAPK signaling pathways.(A) Western blot analyses were performed to evaluate the levels of PD-1, iNOS, and Arg1.(B) Flow cytometry assays were conducted to measure the levels of iNOS, CD86, Arg1, and CD206 expression.(C) ELISA assays were performed to evaluate the levels of TNF-α and IL-10.(D) QPCR analyses were performed to evaluate the levels of TNF-α and IL-10 expression.(E) Western blot analyses were performed to evaluate the levels of ERK1/2, p-ERK1/2, p38, and p-p38 expression.(F) BBB permeability was assessed by EBD extravasation.(G) The TEER value was calculated.(H) Western blot analyses were performed to evaluate the levels of claudin-5 and ZO1 expression.ns, not significant.**P < 0.01, ***P < 0.001.ns: no significant difference. 2003

Fig. 3 .
Fig. 3. PD-1 inhibited inflammatory exudation and reduced BBB permeability, cell apoptosis, and brain injury in vivo.(A) The brain wet-to-dry weight ratio was measured.(B) BBB permeability was detected by Evan's blue staining.(C) Representative photographs of the whole brain in each group after TTC staining.(D) NSS scores were calculated based on tissue damage.(E) H&E staining for evaluation of cell morphological characteristics.(F) TUNEL staining was performed to assess cell apoptosis.*P < 0.05, ***P < 0.001.ns: no significant difference.