GPR35 protects against reperfusion injury in ischemic stroke by binding with the CR2 domain of Raf1.

GPR35 protects against reperfusion injury in ischemic stroke by binding with the CR2 domain of Raf1 - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice J Neuroinflammation . 2026 Feb 10;23:123. doi: 10.1186/s12974-026-03726-1 Search in PMC Search in PubMed View in NLM Catalog Add to search GPR35 protects against reperfusion injury in ischemic stroke by binding with the CR2 domain of Raf1 Xiaojun Li Xiaojun Li 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 2 Department of Pain Medicine, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Xiaojun Li 1, 2, # , Qi Wang Qi Wang 3 Department of Neurology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Qi Wang 3, # , Qingyu Cheng Qingyu Cheng 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Qingyu Cheng 1 , Jingyong Zhou Jingyong Zhou 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Jingyong Zhou 1 , Tingting Qu Tingting Qu 4 Department of Neurology, the First Affiliated Hospital of Anhui Medical University, Hefei, China 5 Anhui Public Health Clinical Center, Hefei, China Find articles by Tingting Qu 4, 5 , Jianling Xu Jianling Xu 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Jianling Xu 1 , Yuming Du Yuming Du 3 Department of Neurology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Yuming Du 3 , Bowen Miao Bowen Miao 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Bowen Miao 1 , Yongquan Chen Yongquan Chen 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 2 Department of Pain Medicine, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Yongquan Chen 1, 2 , Weidong Yao Weidong Yao 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Weidong Yao 1, ✉ , Bin Wang Bin Wang 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 2 Department of Pain Medicine, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China Find articles by Bin Wang 1, 2, ✉ Author information Article notes Copyright and License information 1 Department of Anesthesiology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 2 Department of Pain Medicine, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 3 Department of Neurology, the First Affiliated Hospital of Wannan Medical College, Yijishan Hospital, Wuhu, China 4 Department of Neurology, the First Affiliated Hospital of Anhui Medical University, Hefei, China 5 Anhui Public Health Clinical Center, Hefei, China ✉ Corresponding author. # Contributed equally. Received 2025 Aug 17; Accepted 2026 Feb 3; Collection date 2026. © The Author(s) 2026 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/ . PMC Copyright notice PMCID: PMC13085350  PMID: 41664072 Abstract Ischemic stroke remains a leading cause of global mortality and disability. While timely vascular recanalization is the most direct and clinically validated intervention for cerebral ischemia, reperfusion often induces secondary brain injury, with neuroinflammation playing a central role. Single-cell sequencing data from an ischemic stroke mouse model identify G protein-coupled receptor 35 (GPR35) as a potential regulator of post-ischemic inflammatory responses. GPR35 expression was markedly increased in both in vivo cerebral ischemia–reperfusion models and in vitro oxygen–glucose deprivation systems, predominantly localizing to microglia. Functional studies revealed that genetic knockdown of GPR35 in murine brain tissue significantly exacerbated cerebral infarction volume, neurological deficits, and neuroinflammation in animal models, whereas GPR35 overexpression produced therapeutic effects. Consistently, pharmacological activation of GPR35 using zaprinast attenuated ischemia–reperfusion injury and reduced proinflammatory cytokine production. Mechanistically, zaprinast-mediated GPR35 activation suppressed proinflammatory cytokine production via modulation of the Raf1/ERK1/2/MAPK signaling cascade. Notably, Raf1 knockdown attenuated the pathological exacerbation induced by GPR35 deficiency in peri-infarct regions. Co-immunoprecipitation analyses revealed a direct interaction between GPR35 and Raf1, with the CR2 domain, a critical region for Raf1 autoinhibition, identified as the primary binding interface. Collectively, these findings demonstrate that zaprinast confers cerebroprotective effects in cerebral ischemia–reperfusion injury by activating GPR35, ultimately attenuating infarct progression and neuroinflammation. This mechanistic insight positions GPR35 as a promising therapeutic target for mitigating reperfusion injury in ischemic stroke. Supplementary Information The online version contains supplementary material available at 10.1186/s12974-026-03726-1. Keywords: GPR35, Reperfusion injury, Raf1, Microglia, Zaprinast Introduction Ischemic stroke (IS) [ 1 ] is a leading global cause of mortality [ 2 ] and disability [ 3 ], constituting a major public health challenge [ 4 ]. Current management strategies primarily involve thrombolytic therapy and mechanical thrombectomy [ 5 , 6 ], although these are constrained by a narrow therapeutic window (being applicable in < 3% of patients), with attendant risks of hemorrhagic complications [ 7 ] and ischemia–reperfusion injury, which is defined as paradoxical tissue damage following the restoration of blood flow [ 8 ]. The ischemic cascade initiated by oxygen–glucose deprivation involves excitotoxicity, oxidative/nitrosative stress [ 9 ], and inflammatory responses [ 10 ]. Microglial-mediated neuroinflammation has been identified as a critical pathophysiological component, with activated microglia driving immune responses post-ischemia [ 11 ]. Emerging evidence positions neuroinflammation and immunomodulation [ 12 , 13 ] as pivotal therapeutic targets for managing cerebral ischemia–reperfusion injury [ 14 ]. G protein-coupled receptors (GPCRs), the largest membrane receptor family characterized by seven transmembrane domains, mediate extracellular signal transduction through diverse ligands [ 15 ]. GPR35, an orphan GPCR [ 16 ], is expressed in immune, gastrointestinal, and neural tissues [ 17 ]. GPR35 activation modulates calcium transients, cAMP levels, and exerts anti-inflammatory effects [ 18 ]. Schneditz et al. [ 19 ] demonstrated that GPR35-mediated binding to the α1 subunit of Na⁺/K⁺-ATPase potentiates pump activity, leading to alterations in membrane electrochemistry, calcium flux dynamics, and metabolic reprogramming. However, the expression patterns and functional roles of GPR35 in ischemia–reperfusion injury remain unclear. Utilizing quantitative proteomic analysis, we identified Raf1-a ubiquitously expressed serine/threonine kinase that serves as a central node in Ras-dependent MAPK signaling, regulating cell proliferation and survival, and has increasingly been recognized as a modulator of inflammatory and stress responses in the central nervous system. Nevertheless, the extent to which Raf1 interacts with GPCR signaling, and specifically, how it functionally interacts with microglial GPR35 in ischemia–reperfusion injury remains unclear. To date, to our knowledge, no truly specific agonists or antagonists for GPR35 have been identified. The reported small-molecule agonists (e.g., pamoic acid, zaprinast, kynurenic acid [KYNA], and 5-hydroxyindoleacetic acid [5-HIAA]) and antagonists (e.g., ML194 and ML145) exhibit substantial off-target effects. Consequently, pharmacological studies utilizing these compounds to modulate GPR35 cannot exclude confounding non-specific actions. For example, Sharmin et al. [ 20 ] employed pamoic acid and ML194 to investigate GPR35 in permanent ischemic stroke models. While permanent ischemia represents the majority of clinical stroke cases, the non-renewable nature of neurons underscores the significant clinical benefit of reperfusion therapy in eligible patients. Critically, the pathophysiological mechanisms underlying transient ischemia–reperfusion injury differ substantially from those in permanent ischemia. In this study, we comprehensively investigated GPR35 function in a transient middle cerebral artery occlusion (MCAO) model using a multi-tiered approach: (1) spatiotemporal profiling of GPR35 expression in post-ischemic brain tissue; (2) phenotypic and functional analyses in GPR35 transgenic mice subjected to MCAO; (3) proteomic identification and mechanistic validation of downstream effectors; (4) characterization of GPR35-interacting proteins via co-immunoprecipitation and domain mapping; and (5) pharmacological screening of GPR35 agonists. Our findings indicate that GPR35 is selectively upregulated in the ischemic penumbra, and its activation mitigates neuroinflammation and infarct expansion through CR2 domain-mediated interaction with Raf1, resulting in suppression of ERK1/2–MAPK signaling. Notably, zaprinast was identified as a highly potent GPR35 agonist. These findings establish GPR35 as a novel neuroprotective target in cerebral ischemia–reperfusion injury and provide mechanistic insights into GPCR-mediated modulation of post-ischemic neuroinflammation. Materials and methods Ethics, participants, and animals Patients with acute ischemic stroke (AIS) were recruited from the First Affiliated Hospital of Wannan Medical College, with healthy controls selected through standardized medical screenings. Inclusion and exclusion criteria were referenced from previous studies [ 21 ]. Blood samples were collected one day after admission and patients were scored using The National Institute of Health Stroke Scale (NIHSS). The relevant content is detailed in the Inclusion and Exclusion Criteria and Human Samples sections, as well as in Supplemental Table S1 of the Supplementary Information. Animal experiments We adhered to the ARRIVE guidelines [ 22 ] and the study was approved by the Animal Welfare and Ethics Committee of Wannan Medical College (Approval No. WNMC-AWE-2024210). Mice were acclimated for one week and were randomly assigned to groups using a random number table. A control group was included in each segment of the experiment. To avoid the protective effect of estrogen in ischemic stroke, only male mice were used in the study. Adult male C57BL/6J mice (8–10 weeks old, weighing 25–30 g) were purchased from Jiangsu Qinglongshan Biotechnology Co. Ltd. (Nanjing, China). The GPR35 conventional knockout mice (Gpr35 KO ) and the microglia-specific conditional knockout mice (Gpr35 CKO ) were obtained from Shanghai Model Organisms Center, Inc. (Nanjing, China) ( Supplemental Fig. S1 , S2). The dosing regimen of the GPR35 agonists in mice was based on previous studies [ 23 , 24 ]. Briefly, zaprinast (5 mg/kg), L-KYNA (300 mg/kg), and 5-HIAA (2 mg/kg) were administered via intraperitoneal injection 10 min before establishing the mouse MCAO model. All mice were acclimatized for one week under controlled conditions (25 °C, 70% humidity, 12 h light/dark cycle) with ad libitum access to food and water. We implemented measures to minimize murine model utilization and alleviate experimental distress. All experimental procedures were rigorously conducted by operators blinded to group allocation, thereby mitigating potential bias in data collection and analysis. Information on animal numbers are presented in the figure legends. Different individuals conducted animal experiments, collected human blood samples, performed NIHSS scoring on the patients with stroke, and statistically analyzed the data. Data points identified through outlier detection analysis, along with data obtained from mice in the modeling cohort in which the model was not successfully established, were excluded from subsequent analyses. Lentiviral microinjection Mice were anesthetized with isoflurane (induction: 5%; maintenance: 1%) and stereotactically injected with lentiviruses (Lv-sh-Gpr35 KD , Lv-GPR35 OE , Lv-sh-Raf1 KD , or non-targeting controls (NC); 5 µL, 10⁹ viral genomes/µL, GenePharma) into the right lateral ventricle (coordinates: 0.8 mm lateral to bregma, 2.5 mm depth). After 14 days, MCAO was established in successfully transduced mice. Middle cerebral artery occlusion (MCAO) Cerebral ischemia–reperfusion injury was induced as previously described [ 25 ]. Mice underwent 60 min of MCAO under 2% isoflurane anesthesia. Sham controls underwent identical procedures without embolization. Neurological deficits were evaluated using the Zea-Longa score 24 h post-reperfusion. See the Middle Cerebral Artery Occlusion (MCAO) section and Supplemental Table S2 in the Supplementary Information for further details. Laser speckle contrast imaging (LSCI) Cerebral blood flow was monitored using a 2D Laser Blood Flow Imager (OMEGAZONE OZ-2, Omegawave, Inc., Tokyo, Japan). Bilateral flux ratios (ischemic/contralateral hemispheres) were calculated from regions of interest. Modified neurological severity score (mNSS) Blinded evaluators assessed neurological function at 6 h, 12 h, 1 d, 3 d, and 7 d post-MCAO using standardized criteria (Supplementary Table S3). 2,3,5-triphenyl-tetrazolium chloride solution (TTC) staining Brains were sectioned coronally (1-mm slices), stained with 1% TTC (37 °C, 15 min), and analyzed using ImageJ (Bio-Rad Laboratories Inc., Richmond, CA, USA). Unstained white regions represent infarct areas. Cell culture HT22, BV2, and HEK293T cells (Guangzhou Cellcook Biotech) were maintained in DMEM/10% FBS (Hyclone) with penicillin-streptomycin (Beyotime) at 37 °C/5% CO₂. Oxygen–glucose deprivation (OGD) model Cells underwent 6 h OGD in glucose-free RPMI-1640 (Thermo Fisher) within a hypoxia chamber (95% N₂/5% CO₂), followed by 24 h reperfusion. Western blot Mouse brain protein lysates were resolved using 10% SDS-PAGE and electrophoretically transferred onto Bio Trace™ NT nitrocellulose membranes (Pall, #66485). Membranes were blocked at room temperature for 2 h using bovine serum albumin and subsequently incubated overnight at 4 °C with the indicated primary antibodies: anti-GPR35 (1:1000, Novus, NBP2-24640), anti-Raf1 (1:1000, CST, 9422 S), anti-phospho-ERK/anti-ERK (1:1000, Affinity, AF1015/AF0155), anti-phospho-JNK/anti-JNK (1:1000, Affinity, AF6138/AF3318), anti-phospho-p38 MAPK/anti-p38 MAPK (1:1000, Affinity, AF4001/AF6456), anti-phospho-AKT/anti-AKT (1:1000, HUABIO, HA721870/ET1607), anti-FLAG/anti-HA (1:5000, HUABIO, M1403-2/HA721750), and anti-β-actin/anti-GAPDH (1:100000/1:10000, ABclonal, AC026/AC001). Following three washes with TBST, membranes were incubated with appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies for 2 h at room temperature. Protein bands were visualized using a chemiluminescent HRP substrate (Merck-Millipore, WBKLS0050) and imaged with a ChemiDoc XRS system (Bio-Rad). Band intensity quantification was performed using ImageJ software (NIH; Bethesda, MD, USA). For normalization, band intensities were calculated relative to β-actin or GAPDH, which served as the loading control. Quantitative real-time PCR (RT-qPCR) Total RNA was extracted from tissues using the TRIzol reagent (15596026, Invitrogen) and quantified by spectrophotometry. RNA was reverse transcribed using Hifair ® III SuperMix (11184ES08, 11141ES60, Yeasen) and analyzed via SYBR Green qPCR (Bio-Rad CFX Connect), The primer sequences used are detailed in Supplemental Table S4. Immunofluorescence Immunofluorescence analysis of mouse brain tissue sections and staining were performed as previously mentioned [ 26 ]. Sections were incubated with NeuN (1:100, 94403, CST), Iba1 (1:100, MA5-27726, Invitrogen), GFAP (1:200, GB12096-100, Servicebio), CD68 (1:200, ab283654, Abcam), or Raf1 (1:100, sc7267, Santa Cruz), followed by DyLight-conjugated secondary antibodies (1:200, E032210-01, E032420-01, EarthOx) and DAPI counterstaining. Co-immunoprecipitation (Co-IP) Total protein extracts isolated from the right cerebral hemispheres of mice were incubated with specific antibodies for 4 h at 4 °C. Following incubation, samples were combined with protein A/G agarose (Santa Cruz) beads and rotated overnight at 4 °C. The immunoprecipitated complexes were subsequently washed, separated by SDS-PAGE, and transferred for immunoblotting using target-specific antibodies. The Raf-1 gene and its deletion mutants (Δ1, Δ2, Δ3, and Δ4) were subcloned into the pCMV-Flag vector, whereas the GPR35 gene was subcloned into the pCMV-HA vector. The co-IP primary antibodies are as follows: Anti-Flag (1: 5000; HA601167 , HUABIO), anti-HA antibodies (IP: 1 µg/sample, Western blot: 1: 5000; HA721750, HUABIO) and IgG (HA722127, HUABIO). Molecular docking The three-dimensional structures of the GPR35 and Raf1 proteins were predicted using AlphaFold3 ( https://alphafoldserver.com/ ), followed by an assessment of their binding affinity and potential interaction sites. The resulting models were subsequently visualized using PyMOL (v1.8). Molecular dynamics simulations The selected receptor protein–protein complex was subjected to molecular dynamics simulations using the Gromacs 2020 software package. The AMBER99SB-ILDN force field was applied to the protein, whereas the TIP3P model was employed for explicit water solvation. A minimum distance of 1.0 nm was maintained between the protein atoms and the edges of the water box. System charge was neutralized with sodium or chloride ions based on the docking results. The molecular dynamics workflow comprised four stages: energy minimization, heating, equilibration, and production MD simulation. Detailed procedures and corresponding results are provided in the Results of Molecular Dynamics Simulations section of the Supplementary Material. Plasmid and lentivirus For plasmid construction, the full-length Raf-1 gene and its deletion mutants (Δ1, Δ2, Δ3, and Δ4) were subcloned into the pCMV-Flag vector. The GPR35 gene was subcloned into the pCMV-HA vector. The Raf-1 deletion mutants were defined as follows: Δ1(amino acids 1–254), Δ2 (amino acids 1–330), Δ3 (amino acids 254–648), and Δ4 (amino acids 330–648). HEK293T cells were transiently transfected using Lipofectamine™ 3000 (L3000001, Thermo Fisher Scientific, Waltham, MA, USA) when they reached 80–90% confluence. All plasmids used in this study were sourced from Miaoling Biology (Wuhan, China). Lentiviral particles for GPR35 knockdown/overexpression and Raf1 knockdown were procured from GenePharma. A schematic of the corresponding lentiviral vector constructs is provided in Supplementary Material S5. All vectors include a green fluorescent protein (GFP) expression cassette, enabling the monitoring of transduction efficiency and subsequent functional assays. The targeting sequences for Lv-shRNA-GPR35 KD and Lv-shRNA-Raf1 KD were AGGTCTCCCTGAACCTCAATA and TCGAGTTATTACTGGGAAATA. The mRNA expression levels of GPR35 and Raf1 were quantified via RT-qPCR. Single-cell sequencing analysis Data downloaded from the GEO database (accession: GSE22765 ) were processed using Scanpy (v1.11.4) within a Python environment. Low-quality cells were initially filtered out. Library size normalization was then performed for each cell, followed by a log transformation of the normalized expression matrix. Highly variable genes were selected, and the total UMI count per cell and the mitochondrial transcript percentage were regressed out, with gene expression values subsequently scaled. Batch effects were further corrected using Harmony via Scanpy to reduce technical variability. Uniform Manifold Approximation and Projection was subsequently applied for low-dimensional visualization. Cell clustering was performed using the graph-based Leiden algorithm. Following the aforementioned pipeline, unsupervised clustering of all brain cells identified a total of 24 distinct cell clusters. To identify marker genes for each cluster, differential expression analysis was performed with thresholds of log2FC > 1 and an adjusted p-value (pvals_adj) < 0.05. Statistical analysis Data are expressed as mean ± standard error of the mean (SEM). Statistical analyses were conducted using GraphPad Prism software (version 9.4.0; GraphPad Software, LLC). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. P < 0.05 was considered statistically significant. Results Upregulation of GPR35 expression in cerebral ischemia–reperfusion injury To determine the relevance of GPR35 in ischemic brain injury, we initially performed single-nucleus RNA sequencing (snRNA-seq) analysis to characterize GPR35 expression in the ischemic penumbra of MCAO mice. Clustering analysis identified nine seven major cell types using representative marker genes (Fig. 1 A). GPR35 was significantly upregulated in microglia in the ischemic penumbra of MCAO mice compared to that of sham controls ( P < 0.0001; Fig. 1 B). The clinical relevance of GPR35 was subsequently assessed by quantifying circulating GPR35 transcript levels in peripheral blood samples from healthy donors and patientswith AIS undergoing endovascular therapy [ 27 , 28 ]. Patients with AIS exhibited elevated circulating GPR35 mRNA levels, compared with non-stroke controls (Fig. 1 C). Linear regression analysis revealed that, at 24 h post-admission, higher GPR35 expression levels were significantly associated with higher NIHSS scores, indicating more severe neurological deficits (Fig. 1 D). Infarct volume was assessed using diffusion-weighted imaging sequences on a 3.0-T magnetic resonance imaging system within 24 h following admission (demographic data are summarized in Supplementary Table S1). Fig. 1. Open in a new tab GPR35 expression is elevated in neurons and microglia in ischemic stroke. A , B Single-cell sequencing data revealed that GPR35 was significantly increased in microglia in brain tissue of stroke model mice. C GPR35 is elevated in the blood of patients with stroke. D Correlation between GPR35 expression and NIHSS scores using Pearson’s correlation coefficient ( n =22). E Schematic diagram of MCAO modeling. F Laser speckle blood flow imaging and ( G ) Statistical data before and after modeling ( n = 6). H mNSS of model mice ( n = 10). I TTC-stained brain sections and ( J ) quantitative data ( n = 4). K Western blot representative images of ischemic penumbra brain tissue of MCAO model mice and ( L ) Quantitative data ( n = 3). M Schematic diagram of OGD modeling. N Western blot representative images of BV2 cells after establishment of the OGD model and ( O ) Quantitative data ( n = 6). P , Q GPR35 is expressed in neurons and microglia. Co-localization of GPR35 with NeuN ( P ), and Iba1 Q . Scale bar: 20 μm. Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student; s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and Data are presented as mean ± SEM; P < 0.05 indicating significance Next, we established a reproducible MCAO model (Fig. 1 E–J) for temporal analysis of GPR35 protein dynamics. Western blot quantification demonstrated progressive GPR35 upregulation in penumbral tissue post-reperfusion, peaking at 24 h (Fig. 1 K–L), coincident with maximal neurological impairment (Fig. 1 H). Based on these observations, subsequent experiments were conducted at the 24-h reperfusion time point. This temporal pattern was recapitulated in OGD models, with BV2 microglia and HT22 neurons exhibiting early GPR35 induction (2 h), maximal expression at 6 h, and gradual decline by 12 h (Fig. 1 M–O; Supplementary Fig. S3). Immunofluorescence analysis further confirmed GPR35 co-localization with neuronal (NeuN) and microglial (Iba1) markers (Fig. 1 P, Q), but not with the astrocytic marker GFAP (Supplementary Fig. S4). Collectively, these findings indicate that GPR35 expression is upregulated in both peripheral blood from patients with AIS and ischemic penumbral brain tissue, suggesting a potential involvement of GPR35 in the pathophysiological processes of ischemic stroke. Knockdown of GPR35 expression aggravates cerebral ischemia–reperfusion injury in MCAO mice To investigate the therapeutic potential of GPR35 in cerebral ischemia–reperfusion injury, we generated lentiviral vectors carrying shRNA targeting GPR35 (Lv-shRNA-GPR35 KD ) for gene silencing. Optimal constructs were screened and validated (Supplementary Fig. S5,6) before stereotaxic injection into the lateral ventricles (Fig. 2 A). Two weeks after injection, widespread GFP expression confirmed successful viral transduction, accompanied by significant downregulation of GPR35 in brain tissue (Supplementary Fig. S5). Behavioral assessments revealed exacerbated neurological deficits in GPR35-knockdown mice, with modified neurological severity scores (mNSS) significantly elevated compared to both wild-type and scramble control (Lv-shRNA-NC1) groups subjected to MCAO (Fig. 2 B). Quantitative analysis of cerebral infarction via TTC staining demonstrated increased lesion volumes (white areas) in GPR35-silenced mice (Fig. 2 C–D). Molecular characterization of penumbral tissue revealed elevated mRNA levels of proinflammatory cytokines (IL-1β, TNF-α, and IL-6) in the Lv-shRNA-GPR35 KD group (Fig. 2 H). Fig. 2. Open in a new tab Knockdown of GPR35 expression aggravates cerebral ischemia–reperfusion injury in MCAO mice. A Experimental procedure and timeline. B mNSS score in the brains of model mice ( n = 10). C TTC staining and infarct volume numerical statistics D , n = 4. E Immunofluorescence staining of brain tissue sections from the peri-infarct zone showing CD68 levels and quantitative analysis F , n = 5; G , n = 4. Scale bar: 20 μm. H PCR results of inflammatory factors in the peri-infarct brain tissue of mice in different groups ( n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicates significance CD68, a lysosomal transmembrane protein [ 29 ], reflects enhanced phagocytic activity and is closely associated with neuroinflammation [ 30 – 32 ]. During the acute phase post-stroke, CD68 localizes to peri-infarct regions and progresses to the infarct core by day 7 [ 33 ]. Therefore, CD68 serves as an indicator of the overall inflammatory level in the ischemic penumbra during the acute phase of stroke. We performed quantitative analysis of both fluorescence intensity and area [ 30 ] of CD68 immunofluorescence staining in peri-infarct regions to assess the inflammatory status in these areas. These results demonstrate that both the fluorescence intensity and the stained area of CD68 were increased in the peri-infarct tissues of the GPR35 knockdown group (Fig. 2 E–G), which is consistent with the elevated levels of inflammatory factors observed at the transcriptional level (Fig. 2 H). Collectively, these findings establish that GPR35 deficiency exacerbates ischemic brain injury through amplified neuroinflammatory responses. Knockdown of GPR35 expression aggravates cerebral ischemia–reperfusion injury via Raf1 To elucidate the role of GPR35 in ischemia–reperfusion injury, we performed tandem mass tag-based quantitative proteomic profiling of penumbral tissue from GPR35-knockdown (Lv-sh-Gpr35 KD + MCAO) and control (Lv-sh-NC1 + MCAO) mice. This analysis identified 238 differentially expressed proteins (DEPs), with 215 upregulated and 23 downregulated in GPR35-deficient mice (fold change ≥ 1.5; P < 0.05, Benjamini–Hochberg corrected FDR < 0.05; Fig. 3 A, B, C). Among the signaling pathways with the most significant enrichment of differential proteins, the MAPK pathway was the most prominent (Fig. 3 D). The Venn diagram (Fig. 4 E) of the intersection of signaling pathways associated with ischemic stroke indicated that Raf1 ( Q99N57 ; https://www.uniprot.org/uniprotkb/Q99N57/entry ) is one of the most critical proteins. Raf1 is a 75 kDa protein kinase ubiquitously expressed in mammalian tissues [ 34 ]. The regulation of ERK phosphorylation by Raf1 has been repeatedly demonstrated [ 35 , 36 ]. Next, we verified the difference in Raf1 expression in the ischemic penumbra brain tissue between Lv-sh-Gpr35 KD + MCAO mice and Lv-sh-NC1 + MCAO mice by western blot (Fig. 3 F). As the MAPK pathway is primarily composed of ERK, p38-MAPK, and JNK pathways, the expression of Raf1 in the ischemic penumbra brain tissue of Lv-sh-Gpr35 KD + MCAO mice was significantly higher than that in Lv-sh-NC1 + MCAO mice. The results indicate that GPR35 acted primarily via the Raf1-ERK1/2 pathway (Fig. 3 G–H, Supplemental Fig. S7). Fig. 3. Open in a new tab Knockdown of GPR35 expression aggravates cerebral ischemia–reperfusion injury through Raf1. A Mice in the empty virus group and the GPR35 knockdown virus group were injected with MCAO 2 weeks later, and the proteome of brain tissue in the ischemic penumbra was sequenced and analyzed, respectively. B Volcano map. C Heat map. D Pathway enrichment map. E Venn diagram of highly differential proteins in stroke-related signaling pathways. F Western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( G ) GPR35 ( n = 3), ( H ) Raf1( n = 6), p-ERK ( n = 6), ERK ( n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way ANOVA, or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicates significance Fig. 4. Open in a new tab GPR35 overexpression alleviates cerebral ischemia–reperfusion injury in MCAO mice. A Timeline of modeling after GPR35 overexpression. B mNSS score in the brains of model mice. C TTC staining ( n = 11), ( D ) infarct volume and numerical statistics ( n = 4). E immunofluorescence staining of brain tissue sections in the peri-infarct zone revealed CD68 levels and quantitative analysis ( F , n = 5, G , n = 4). Scale bar: 20 μm. H Quantitative PCR analysis showed decreased proinflammatory mediator expression (IL-1, n = 6; IL-6, n = 6) in GPR35-overexpressing penumbral tissue. I western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( J , n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way ANOVA, or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicating significance GPR35 overexpression alleviates cerebral ischemia–reperfusion injury in MCAO mice To complement the loss-of-function studies, we generated Lv-GPR35 OE lentiviral vectors for cerebral GPR35 overexpression via stereotaxic ventricular delivery (Fig. 4 A). The construction of lentiviral plasmids and validation of transfection efficiency are shown in Supplemental Figures S5 and S6. Behavioral assessment demonstrated ameliorated neurological deficits in GPR35-overexpressing mice, exhibiting significantly lower mNSS scores compared to wild-type and scramble controls (Fig. 4 B). TTC-based volumetric analysis revealed a reduction in infarct size in the Lv-GPR35 OE group relative to that of controls (Fig. 4 C, D). The improved outcomes were associated with reduced infiltration of CD68⁺ cells (Fig. 4 E-G). In line with this, quantitative PCR analysis of the penumbral tissue further confirmed a concomitant decrease in the expression of proinflammatory mediators, specifically IL-1 and IL-6, following GPR35 overexpression (Fig. 4 H). Mechanistic investigations demonstrated that GPR35 overexpression suppressed Raf1-ERK1/2 signaling activation, with phospho-ERK1/2 levels reduced compared to that of controls (Fig. 4 I, J). These findings indicate that GPR35 overexpression alleviates cerebral ischemia–reperfusion injury by inhibiting Raf1-ERK1/2 signaling in the ischemic penumbra, thereby suppressing neuroinflammation. Raf1 is essential for GPR35-mediated protection against cerebral ischemia–reperfusion injury in MCAO mice To elucidate the functional interaction between GPR35 and Raf1, we employed a dual-knockdown strategy using combinatorial lentiviral vectors (Lv-shRNA-Gpr35 KD + Lv-shRNA-Raf1 KD ) administered via stereotaxic ventricular injection (Fig. 5 A). Following lentiviral validation (Supplementary Fig. S5, 6), Raf1 suppression was confirmed by immunoblotting 14 days post-injection. Neurological assessment revealed partial rescue of GPR35 deficiency-induced deficits, with mNSS scores in dual-knockdown mice approximating wild-type levels (Fig. 5 B). TTC-based volumetric analysis demonstrated infarct reduction in GPR35/Raf1 co-silenced mice compared to GPR35 single-knockdown controls (Fig. 5 C, D). Fig. 5. Open in a new tab Raf1 is essential for GPR35-mediated protection against cerebral ischemia–reperfusion injury in MCAO mice. A Timeline of modeling after GPR35 and Raf1 knockdown. B mNSS score in the brains of model mice ( n = 11). C TTC staining and infarct volume numerical statistics ( D , n = 4). E Immunofluorescence staining of brain tissue sections from the peri-infarct zone showing CD68 levels and quantitative analysis ( F , n = 5; G , n = 4). Scale bar: 20 μm. H Quantitative PCR analysis showed decreased proinflammatory mediator expression (IL-1, n = 5; IL-6, n = 6) in dual-knockdown mice. I Western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( J ) Raf1, p-ERK ( n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicating significance Molecular profiling revealed attenuated neuroinflammation in dual-knockdown mice, evidenced by: (1) reduction in CD68 + cells infiltration (Fig. 5 E–G). (2) decrease in proinflammatory cytokine mRNA levels (Fig. 5 H); Western blot analysis confirmed pathway-specific modulation, with Raf1 knockdown normalizing ERK1/2 hyperphosphorylation in GPR35-deficient mice (Fig. 5 I, J). These results mechanistically establish Raf1 as the critical downstream mediator of the neuroprotective effects of GPR35. GPR35 targets the CR2 region of Raf1 to inhibit Raf1 activation Protein–protein interaction networks serve as essential mechanistic determinants for GPCR functionality, given their role in modulating receptor conformation and downstream effector coupling [ 37 ]. Therefore, we conducted systematic investigations of the protein–protein interactions between GPR35 and Raf1. We first performed AlphaFold3 structural predictions, which revealed potential binding interfaces between GPR35 and Raf1, including salt bridges (H175–D300, D521–H296, K591–D204) and hydrogen bonds (C184–R123, N198–S292, Y565–Q207, Q253–S305) (Fig. 6 A, Supplemental Table S5). Molecular dynamics simulations over a 100 ns trajectory provided further validation of the complex’s stability, demonstrating that the identified interactions—including salt bridges, hydrogen bonds, and hydrophobic contacts—effectively enhance the thermodynamic stability of the RAF1–GPR35 complex. The persistence of these favorable interactions throughout the simulation indicates their critical role in maintaining a stable protein–protein association. (Fig. 6 B–F). Immunofluorescence colocalization analysis demonstrated spatial proximity between GPR35 and Raf1 in murine brain parenchyma, HT22 neurons, and BV2 microglia (Fig. 6 G–I). Co-immunoprecipitation (Co-IP) assays in HEK293T cells confirmed physical interaction between the two proteins (Fig. 6 J). Fig. 6. Open in a new tab GPR35 targets the CR2 region of Raf1 to inhibit Raf1 activation. A Schematic representation of the 3D structures of the predicted binding sites of GPR35 and Raf1. B– F Molecular dynamics simulations of GPR35 and Raf1 stability. G GPR35 and Raf1 colocalize in mouse brain tissue. H GPR35 and Raf1 colocalize in HT22 cells. I GPR35 and Raf1 colocalize in BV2 cells. Scale bar: 20 μm. J GPR35 and Raf1co-IP in BV2 cells. K , L different Raf1 domains interact with GPR35 To identify the specific Raf1 domain responsible for binding GPR35, we performed domain mapping using truncation mutagenesis, which localized the primary interaction interface to Raf1’s CR2 domain (Fig. 6 K, L). Notably, the CR2 region is a critical domain for Raf1 autoinhibition [ 38 ]. These findings mechanistically establish that GPR35 directly engages Raf1 through CR2 domain interaction, thereby suppressing Raf1-mediated signaling cascades. Global KO of GPR35 aggravates cerebral ischemia/reperfusion injury in MCAO mice To establish genetic evidence for GPR35’s neuroprotective role, we developed constitutive GPR35 knockout (GPR35 KO ) mice (Fig. 7 A, Supplementary Fig. S1). Following MCAO induction, GPR35 KO mice exacerbated neurological deficits, as reflected by higher mNSS scores compared to wild-type controls (Fig. 7 B). Quantitative analysis of TTC-stained revealed larger infarct volumes in Gpr35 KO mice versus wild-type mice (Fig. 7C, D). Increased infiltration of CD68⁺ cells in the penumbral tissue of GPR35 KO mice (Fig. 7 E-G) corresponded to a significant upregulation of proinflammatory cytokines at the molecular level (Fig. 7 H). Western blot analysis confirming upregulation of Raf1 protein and increase in phospho-ERK1/2 levels (Fig. 7 I, J). Collectively, these findings indicate that GPR35 deficiency amplifies neuroinflammation and exacerbates ischemic brain injury, in part through activation of the Raf1-ERK1/2 signaling pathway. Fig. 7. Open in a new tab Global knockout of GPR35 exacerbates cerebral ischemia/reperfusion injury in MCAO mice. A Generation and experimental timeline of knockout mice. B mNSS score in the brains of model mice ( n = 10). C TTC staining and infarct volume numerical statistics ( D , n = 4). E Immunofluorescence staining of brain tissue sections from the peri-infarct zone showing CD68 levels and quantitative analysis ( F , n = 5; G , n = 4). Scale bar: 20 μm. H PCR results of inflammatory factors in the peri-infarct brain tissue of mice in different groups ( n = 6). I Western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( J ) Raf1, p-ERK, ERK ( n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicates significance Microglial GPR35 deficiency exacerbates cerebral ischemia–reperfusion injury in MCAO mice Immune and inflammatory responses play pivotal roles in ischemic stroke pathogenesis [ 39 ]. As essential innate immune cells and resident sentinels of the brain, microglia are the first non-neuronal cells activated in response to acute cerebral injury [ 40 ]. Immunofluorescence analysis revealed that GPR35 was expressed in both neurons and microglia. However, in vitro OGD models using HT22 neurons and BV2 microglial cells revealed the most pronounced induction in microglia, consistent with single-cell RNA sequencing data from publicly available datasets demonstrating microglia-specific upregulation. Therefore, we subsequently investigated the functional role of microglial GPR35 in this injury context, generated microglia-specific GPR35 conditional knockout mice (Gpr35 CKO ) by crossing Gpr35 flox/flox mice with Cx3cr1-cre mice, yielding Cx3cr1-cre +/− /Gpr35 flox/flox progeny (Fig. 8 A, Supplemental Fig. S2). Fig. 8. Open in a new tab Conditional knockout of GPR35 exacerbates cerebral ischemia–reperfusion injury in MCAO mice. A Generation and experimental timeline of conditional knockout mice. B mNSS score in the brains of model mice ( n = 10). C TTC staining and infarct volume numerical statistics ( D , n = 4). E Immunofluorescence staining of brain tissue sections from the peri-infarct zone showing CD68 levels and quantitative analysis ( F , G , n = 4). Scale bar: 20 μm. H PCR results of inflammatory factors in the peri-infarct brain tissue of mice in different groups ( n = 6). ( I ) Western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( J ) Raf1, p-ERK, ERK ( n = 3). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t -test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicating significance As anticipated, Gpr35 CKO mice exhibited exacerbated neurological deficits, as evidenced by higher mNSS Scores compared with control mice (Fig. 8 B). Quantitative analysis of TTC-stained revealed significantly larger infarct volumes in Gpr35 CKO mice versus controls (Fig. 8 C–D). Increased infiltration of CD68⁺ cells in the ischemic penumbra of Gpr35 CKO mice (Fig. 8 E-G) corresponded to elevated levels of pro-inflammatory cytokines as demonstrated by molecular analyses (Fig. 8 H). Western blot analysis confirmed upregulation of Raf1 protein and increased levels of phosphorylated ERK1/2 (p-ERK1/2) in these mice (Fig. 8 I, J). Collectively, these findings indicate that microglial GPR35 deficiency amplifies neuroinflammation and exacerbates ischemic brain injury, partly through activation of the Raf1-ERK1/2 signaling pathway. GPR35 agonists attenuate cerebral ischemia–reperfusion injury in MCAO mice GPR35 is activated by multiple small acidic molecules, but its specific agonist remains unclear [ 41 ]. We selected several commonly studied candidate agonists, including the recently identified endogenous agonist 5-HIAA [ 45 ]. Results demonstrated that both L-KYNA and zaprinast conferred protective effects in the MCAO mouse model, with zaprinast showing the most pronounced efficacy (as assessed by neurological function scores, infarct volume, and inflammatory cytokine levels) (Supplemental Fig. S8). However, 5-HIAA did not provide significant improvement across these parameters. Therefore, we selected zaprinast (an approved clinical PDE5 inhibitor) as the candidate agonist for subsequent studies, highlighting its translational potential. We investigated whether supplementation with zaprinast could ameliorate cerebral ischemia–reperfusion injury. wild-type (WT) mice and GPR35 CKO mice were administered zaprinast (5 mg/kg, i.p [ 23 , 24 ]). 10 min before ischemia induction, whereas WT mice receiving phosphate-buffered saline (PBS) served as controls. Following MCAO, neurological function, infarct volume, major downstream signaling pathways, and inflammatory indices were systematically evaluated (Fig. 9 A). Compared with PBS-treated WT mice, zaprinast treatment significantly reduced mNSS scores (Fig. 9 B) and infarct volumes (Fig. 9 C–D) in WT mice. Quantitative immunofluorescence analysis further revealed a substantial reduction in CD68 + cells in zaprinast-treated WT mice, consistent with suppressed microglial activation and phagocytic activity (Fig. 9 E–G). In parallel, quantitative PCR analysis demonstrated that zaprinast treatment significantly decreased the mRNA expression of pro-inflammatory cytokines, including IL-1β, IL-6, and TNF-α, in WT mice (Fig. 9 H). Western blot analysis revealed reduced Raf1 protein levels and decreased expression of key effector molecules in the ERK1/2/MAPK signaling pathway following zaprinast treatment (Fig. 9 I–J). Importantly, these behavioral, histological, inflammatory, and molecular effects were not observed in zaprinast-treated GPR35 CKO mice, indicating a requirement for microglial GPR35. Collectively, these findings demonstrate that zaprinast alleviates cerebral ischemia–reperfusion injury by GPR35-dependently inhibiting the Raf1-ERK1/2/MAPK signaling cascade and downstream neuroinflammatory responses. Fig. 9. Open in a new tab Pharmacological agonist of GPR35 attenuates ischemia–reperfusion injury in MCAO mice. A Diagram of zaprinast structure and timelines for three different groups. B mNSS scores in the brains of model mice ( n = 10). C TTC staining and infarct volume numerical statistics ( D , n = 4). E Immunofluorescence staining of brain tissue sections from the peri-infarct zone showing CD68 levels and quantitative analysis ( F , n = 5; G , n = 4). Scale bar: 20 μm. H PCR results of inflammatory factors in the peri-infarct brain tissue of mice in different groups ( n = 6). ( I ) Western blot representative diagram of protein expression in ischemic penumbra brain tissue of different groups of mice, and numerical statistics ( J ) Raf1, p-ERK, ERK ( n = 6). Normality of data distribution was assessed using the Shapiro–Wilk test. Normally distributed data were subjected to Student’s t-test, one-way analysis of variance (ANOVA), or two-way ANOVA as appropriate. Numerical analysis and data are presented as mean ± SEM; P < 0.05 indicates significance Discussion Our study demonstrates that microglial GPR35 plays an important role in modulating neuroinflammatory responses and tissue injury during cerebral ischemia–reperfusion. Following ischemia–reperfusion, GPR35 expression is substantially upregulated in microglia, and its activation is associated with suppression of the Raf1-ERK1/2-MAPK signaling cascade, reducing pro-inflammatory cytokine expression, and limiting infarct progression. Importantly, both microglial GPR35 and its downstream effector Raf1 are essential to mitigate post-ischemic inflammatory responses and attenuating neurological deterioration after stroke (Fig. 10 ). Collectively, these findings identify a GPR35-Raf1-mediated signaling axis as a previously unrecognized microglia-intrinsic immune regulatory mechanism contributing to the pathogenesis of cerebral ischemia–reperfusion injury. Fig. 10. Open in a new tab GPR35 protects against ischemic stroke by binding with the CR2 domain of Raf1 GPR35 is an incompletely characterized orphan receptor with demonstrated involvement in anti-inflammatory responses [ 42 , 43 ], analgesic pathways [ 23 , 44 ]. Genome-wide association studies have identified GPR35 as a susceptibility locus for coronary artery calcification and atherosclerotic cardiovascular diseases [ 45 ]. Emerging evidence supports its association with hypoxic–ischemic pathologies [ 46 ], where its expression is significantly upregulated. In examining the role of GPR35 in adipose tissue energy homeostasis, Agudelo et al. found that increased GPR35 expression correlated with the expression of anti-inflammatory genes [ 43 ]. Given that GPR35 is not distributed in adipocytes, the anti-inflammatory effect is derived from immune cells in adipose tissue. Further research has shown that GPR35 is widely involved in immune regulation of peripheral tissues. However, the expression of GPR35 in the immune cells of the central nervous system microglia and its mechanism have not been studied. Pharmacological inhibition studies using selective G protein antagonists demonstrate that GPR35 signaling engages Gi/o [ 47 ], Gα13, and β-arrestin2 [ 48 ] pathways to modulate downstream cAMP signaling and ERK1/2 phosphorylation cascades [ 49 , 50 ]. Immunity and inflammation are the key factors in stroke [ 13 , 14 ]. As a key cell population in the brain involved in immunity and inflammation, the research on its regulatory mechanism is of great significance, with obvious value in clinical application. The involvement of GPR35 in non-central immune regulation has been repeatedly demonstrated [ 24 , 43 , 51 ], but the mechanism of the regulation of microglia in the central system has not been studied. Our commitment to exploring the role of inflammatory factors in cardiovascular and cerebrovascular contexts [ 26 , 52 ], has promoted our strong interest. In this study, we first observed that GPR35 was significantly increased in microglia in the ischemic penumbra of the mouse model of ischemic stroke. Further analyses revealed that GPR35 was increased in the blood of patients with ischemic stroke, in ischemic penumbra brain tissue from ischemic stroke model mice, and in HT22 and BV2 cells after OGD. Subsequently, we demonstrated that GPR35 depletion exacerbates cerebral ischemia–reperfusion injury, whereas its overexpression attenuates such damage, indicating its essential protective role in ischemic stroke. Proteomic profiling revealed that Raf1 as a pivotal mediator of GPR35-conferred protection against cerebral ischemia–reperfusion injury, wherein it orchestrates inflammatory responses via the ERK/MAPK signaling axis. Notably, the synchronous elevation of GPR35 and inflammatory cytokines, as well as the positive association between circulating GPR35 levels and stroke severity in the acute clinical setting, suggests acute stress responses, paralleling the rapid activation of microglia as early-phase injury responders [ 53 , 54 ], which likely reflects disease stage–dependent regulation rather than a direct causal pathogenic role. Collectively, these observations indicate that increased GPR35 expression during ischemia–reperfusion injury represents a context-dependent, reactive upregulation linked to ischemia-induced inflammatory responses. Protein interactomes constitute fundamental mechanistic determinants of GPCR functionality by modulating receptor reconfiguration and effector protein engagement [ 37 , 55 ]. Recent studies revealed distinct GPR35 interaction paradigms: Schneditz et al. [ 19 ] identified plasma membrane-localized complex formation with the Na⁺/K⁺-ATPase α-subunit, augmenting ion transport activity, whereas Wyant et al. [ 18 ] demonstrated mitochondrial inner membrane binding to ATP synthase inhibitory factor 1, effectively suppressing its inhibitory function. To investigate the relationship between GPR35 and Raf1, we predicted the binding sites of GPR35 and Raf1 using Alphafold3 and simulated the binding state using a computer. According to the domain of Raf1 on the uniprot website ( Q99N57 ; www.uniprot.org/uniprotkb/Q99N57/entry ), different plasmids [ 56 ] were constructed to verify that the binding region of GPR35 and Raf1 was CR2. The CR2 region is closely associated with autoinhibition of Raf1 [ 38 ]. GPR35 has been recognized as a receptor activated by multiple small acidic molecules, yet its truly specific agonist has remained a subject of debate [ 44 ]. In selecting candidate agonists, we adhered to the following principles: (1) Prioritizing experimental animal species with minimal genetic divergence from human GPR35, considering clinical translational potential. (2) Choosing agonists with demonstrated efficacy in both experimental models and humans, given the species-specific differences in agonist activity. (3) Favoring endogenous agonists due to biosafety considerations. Given the high homology between humans and mice was highest [ 45 , 46 ], we further evaluated the candidate GPR35 agonists in mouse models of cerebral ischemia–reperfusion. Among all candidate agonists, KYNA and zaprinast exhibited significant activity in both humans and mice, while the recently identified endogenous agonist 5-HIAA showed no neuroprotective effect in our cerebral ischemia–reperfusion model, potentially due to its known neutrophil-recruiting properties. Although KYNA possesses established neuroprotective effects, its poor blood-brain barrier permeability limits its direct application. Following previous studies [ 23 ], we administered its BBB-permeable precursor L-KYNA peripherally, achieving central elevation of KYNA levels. Notably, zaprinast demonstrated the most pronounced therapeutic efficacy, consistent with prior reports that it enhances cGMP signaling, promotes vasodilation, and improves cerebral blood flow during reperfusion [ 57 ]. Crucially, earlier studies did not identify GPR35 as a primary target of zaprinast, leaving its cell-specific mechanisms unexamined. Our data now demonstrate that zaprinast’s protective effects are mediated predominantly through microglial GPR35 activation, reducing post-stroke inflammation and injury. These results suggest a dual mechanism of action—microglia-dependent modulation of neuroinflammation and vasodilatory support—providing a more comprehensive understanding of zaprinast-mediated neuroprotection. The pathophysiological distinctions between permanent MCAO and transient MCAO with reperfusion models reflect clinically divergent patient populations [ 58 ]. Although permanent occlusion represents a larger patient cohort, the potential for meaningful neurological recovery is significantly enhanced in patients receiving timely reperfusion therapy. Furthermore, the reperfusion phase facilitates improved delivery of potential neuroprotective agents to the vulnerable ischemic penumbra, rendering the transient MCAO model particularly relevant for therapeutic efficacy assessment. Critically, the underlying mechanisms diverge substantially between these models, primarily in the severity and duration of ischemia and the dynamic composition of the ensuing inflammatory infiltrate. The transient MCAO model offers distinct advantages for investigating the mechanisms of ischemia–reperfusion injury [ 59 ] and inflammatory cell infiltration dynamics [ 60 ]. Given these distinct temporal and cellular profiles, investigating GPR35 function specifically within the context of transient MCAO reperfusion injury is highly warranted. Crucially, our findings reveal a distinct mechanism of GPR35-mediated protection compared to the research by Sharmin et al. by: (1) focusing on GPR35 in cerebral ischemia–reperfusion injury within clinically relevant cohorts; (2) employing transient stroke models distinct from permanent ischemia paradigms; (3) identifying divergent pathological mechanisms including disease progression kinetics and immune infiltration dynamics [ 54 ]; (4) characterizing differential downstream signaling pathways; and (5) implementing distinct analytical approaches—whereas their flow cytometry-based cell sorting coupled with high-sensitivity ELISA(Enzyme-Linked Immunosorbent Assay) kits may partly account for the observed phosphorylation dynamics (e.g., p-p38 MAPK changes). Our study has some limitations. First, our current findings only demonstrate that GPR35 plays a neuroprotective role through its binding to Raf1 to downregulate the MAPK signaling pathway during ischemic stroke; however, Raf1 has complex and important functions and we have only explored one mechanism, with the precise process of MAPK downregulation after binding remaining to be elucidated. Second, although our study showed that GPR35 protects neurons in vivo, it remains to be verified whether it exerts neuroprotective effects in vitro. Third, while our data indicate that zaprinast ameliorates post-stroke inflammation and injury predominantly through microglial GPR35 activation, previous studies have also reported that zaprinast increases cerebral blood flow and reduces infarct volume in rodent models of stroke. These findings suggest that zaprinast-mediated neuroprotection likely involves multiple cell-type–specific mechanisms in addition to vascular effects, thereby further defining its cell-type specificity and contribution to cerebrovascular regulation, which is essential for mechanistic clarification and future targeted therapeutic strategies. Finally, only male mice were used in this study. However, sex differences exist in cerebrovascular, metabolic, ischemic neuronal damage, and stroke outcomes. The lack of data from both sexes is a limitation of this study and should be addressed in future studies. Conclusions This study establishes GPR35 as a microglia-enriched protector against cerebral ischemia–reperfusion injury. Mechanistically, GPR35 binds Raf1’s CR2 domain to inhibit ERK1/2/MAPK signaling, reducing neuroinflammation and infarct progression. Zaprinast also exerts protective effects by activating GPR35 to alleviate ischemia–reperfusion injury. These findings propose GPR35 as a novel target for stroke therapy, warranting additional mechanistic studies and further investigation into its translational applicability. Supplementary Information Supplementary Material 1. (37.1MB, docx) Acknowledgements This work was financially supported by the Anhui Provincial Health Commission (AHWJ2022b021), Anhui Provincial Natural Science Foundation (2308085MH259), Natural Science Research Project of Anhui Educational Committee (2022AH040169, 2022AH051219, 2023AH051776, 2023AH010074, 2025AHGXZK20029), Clinical Medical Research Project of Anhui Province (202304295107020002, 202304295107020004), Science and Technology Project of Wuhu City (2022jc63), Wannan Medical College Combination Fund (XQHR202424), Anhui province key laboratory university-level open subject (LAB202203), Provincial-Level Quality Engineering Project for Higher Education Institutions in Anhui Province (2024xfkc009), the Open Research Fund of Anhui Province Key Laboratory of Non-coding RNA Basic and Clinical Transformation (NcRNA202512). Authors’ contributions Bin Wang conceived the project; Bin Wang, Qi Wang, Xiaojun Li, Yongquan Chen and Weidong Yao designed the experiments; Xiaojun Li and Jianling Xu completed the animal experiment; Yuming Du participated in the design and drawing of the illustrations; Tingting Qu and Jingyong Zhou completed the cell experiment and data analysis; Qingyu Chen and Bowen Miao participated in the collection and analysis of clinical patient data; Bin Wang, Qi Wang, and Jianling Xu provided experimental drugs and quality control; All the authors reviewed and approved the final manuscript. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate All animal experiments were approved by the Animal Welfare and Ethics Committee of Wannan Medical College (Approval No: WNMC-AWE-2024210). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Xiaojun Li and Qi Wang contribute equally to this work. Contributor Information Weidong Yao, Email: [email protected]. Bin Wang, Email: [email protected]. References 1. Mendelson SJ, et al. Diagnosis and management of transient ischemic attack and acute ischemic stroke: a review. JAMA. 2021;325:1088–98. [ DOI ] [ PubMed ] [ Google Scholar ] 2. Tsao CW, et al. Heart disease and stroke statistics-2022 update: a report from the American heart association. Circulation. 2022;145:e153–639. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Andrew NE, et al. Hospital presentations in long-term survivors of stroke. Stroke. 2020;51:3673–80. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Rudd KE, et al. Global, regional, and National sepsis incidence and mortality, 1990–2017: analysis for the global burden of disease study. Lancet. 2020;395:200–11. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Nguyen TN, et al. Endovascular management of acute stroke. Lancet. 2024;404:1265–78. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Chen H, et al. Endovascular stroke thrombectomy for patients with large ischemic core: a review. JAMA Neurol. 2024;81:1085–93. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Powers WJ, et al. 2018 guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2018;49:e344-418. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Eltzschig HK, et al. Ischemia and reperfusion—from mechanism to translation. Nat Med. 2011;17:1391–401. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Qin C, et al. Signaling pathways involved in ischemic stroke: molecular mechanisms and therapeutic interventions. Signal Transduct Target Ther. 2022;7:215. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Schulz C, et al. Inflammaging aggravates stroke pathology. Nat Immunol. 2023;24:887–8. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Paolicelli RC, et al. Microglia States and nomenclature: a field at its crossroads. Neuron. 2022;110:3458–83. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Kelly PJ, et al. Anti-inflammatory approaches to ischaemic stroke prevention. J Neurol Neurosurg Psychiatry. 2018;89:211–8. [ DOI ] [ PubMed ] [ Google Scholar ] 13. Iadecola C, et al. The immunology of stroke: from mechanisms to translation. Nat Med. 2011;17:796–808. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Fisher M, et al. Future directions of acute ischaemic stroke therapy. Lancet Neurol. 2015;14:758–67. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Alavi MS, et al. Orphan g protein-coupled receptors: the role in CNS disorders. Biomed Pharmacother. 2018;98:222–32. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Foord SM, et al. International union of pharmacology. XLVI. G protein-coupled receptor list. Pharmacol Rev. 2005;57:279–88. [ DOI ] [ PubMed ] [ Google Scholar ] 17. Wang J, et al. Kynurenic acid as a ligand for orphan G protein-coupled receptor GPR35. J Biol Chem. 2006;281:22021–8. [ DOI ] [ PubMed ] [ Google Scholar ] 18. Wyant GA, et al. Mitochondrial remodeling and ischemic protection by G protein-coupled receptor 35 agonists. Science. 2022;377:621–9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Schneditz G, et al. GPR35 promotes glycolysis, proliferation, and oncogenic signaling by engaging with the sodium potassium pump. Sci Signal. 2019;12:aau9048. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Sharmin O, et al. Activation of GPR35 protects against cerebral ischemia by recruiting monocyte-derived macrophages. Sci Rep. 2020;10:9400. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Zhao H, et al. Inhibition of histone deacetylase 3 by MiR-494 alleviates neuronal loss and improves neurological recovery in experimental stroke. J Cereb Blood Flow Metab. 2019;39:2392–405. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Percie du Sert N, et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol. 2020;18:e3000410. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Cosi C, et al. G-protein coupled receptor 35 (GPR35) activation and inflammatory pain: studies on the antinociceptive effects of kynurenic acid and Zaprinast. Neuropharmacology. 2011;60:1227–31. [ DOI ] [ PubMed ] [ Google Scholar ] 24. De Giovanni M, et al. GPR35 promotes neutrophil recruitment in response to serotonin metabolite 5-HIAA. Cell. 2022;185:815–e830819. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Lemmerman LR, et al. Nanotransfection-based vasculogenic cell reprogramming drives functional recovery in a mouse model of ischemic stroke. Sci Adv. 2021;7:eabd4735. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Wang Q, et al. Chemokine receptor 7 mediates miRNA-182 to regulate cerebral ischemia/reperfusion injury in rats. CNS Neurosci Ther. 2023;29:712–26. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Baird AE. Blood biologic markers of stroke: improved management, reduced cost? Curr Atheroscler Rep. 2006;8:267–75. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Raman K, et al. Peripheral blood MCEMP1 gene expression as a biomarker for stroke prognosis. Stroke. 2016;47:652–8. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Chistiakov DA, et al. CD68/macrosialin: not just a histochemical marker. Lab Invest. 2017;97:4–13. [ DOI ] [ PubMed ] [ Google Scholar ] 30. Huang Y, et al. Regulation of cell distancing in peri-plaque glial Nets by Plexin-B1 affects glial activation and amyloid compaction in alzheimer’s disease. Nat Neuros. 2024;27:1489–504. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Gao M, et al. Loss of MFE-2 impairs microglial lipid homeostasis and drives neuroinflammation in alzheimer’s pathogenesis. Nat Aging. 2025;5:2279–96. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Zhu D, et al. Microglial Nr4a1 deficiency and neuronal C3 deposition mediate TMJ inflammation-induced hippocampal excessive synaptic pruning and depression-like behaviors in mice. Brain Behav Immun. 2025;131:106128. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Perego C, et al. Temporal pattern of expression and colocalization of microglia/macrophage phenotype markers following brain ischemic injury in mice. J Neuroinflammation. 2011;8:174. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Storm SM, et al. Expression of Raf family proto-oncogenes in normal mouse tissues. Oncogene. 1990;5:345–51. [ PubMed ] [ Google Scholar ] 35. Hodis E, et al. A landscape of driver mutations in melanoma. Cell. 2012;150:251–63. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Chan LH, et al. PRMT6 regulates RAS/RAF binding and MEK/ERK-mediated cancer stemness activities in hepatocellular carcinoma through CRAF methylation. Cell Rep. 2018;25:690–e701698. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Hilger D, et al. Structure and dynamics of GPCR signaling complexes. Nat Struct Mol Biol. 2018;25:4–12. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Riaud M, et al. The role of CRAF in cancer progression: from molecular mechanisms to precision therapies. Nat Rev Cancer. 2024;24:105–22. [ DOI ] [ PubMed ] [ Google Scholar ] 39. Castellani G, et al. Transforming the understanding of brain immunity. Science 2023, 380:eabo7649. [ DOI ] [ PubMed ] 40. Beuker C, et al. Immune cell infiltration into the brain after ischemic stroke in humans compared to mice and rats: a systematic review and meta-analysis. Transl Stroke Res. 2021;12:976–90. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Im D-S. Recent advances in GPR35 pharmacology; 5-HIAA serotonin metabolite becomes a ligand. Arch Pharm Res. 2023;46:550–63. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Farooq SM, et al. Disruption of GPR35 exacerbates dextran sulfate sodium-induced colitis in mice. Dig Dis Sci. 2018;63:2910–22. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Agudelo LZ, et al. Kynurenic acid and Gpr35 regulate adipose tissue energy homeostasis and inflammation. Cell Metab. 2018;27:378–e392375. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Gupta RA, et al. GPR35 agonists inhibit TRPA1-mediated colonic nociception through suppression of substance P release. Pain. 2025;166:596–613. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Sun YV, et al. Application of machine learning algorithms to predict coronary artery calcification with a sibship-based design. Genet Epidemiol. 2008;32:350–60. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Ronkainen V-P, et al. Hypoxia-inducible factor 1-induced G protein-coupled receptor 35 expression is an early marker of progressive cardiac remodelling. Cardiovasc Res. 2014;101:69–77. [ DOI ] [ PubMed ] [ Google Scholar ] 47. Ohshiro H, et al. GPR35 is a functional receptor in rat dorsal root ganglion neurons. Biochem Biophys Res Commun. 2008;365:344–8. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Zhao P, et al. Targeting of the orphan receptor GPR35 by Pamoic acid: a potent activator of extracellular signal-regulated kinase and β-arrestin2 with antinociceptive activity. Mol Pharmacol. 2010;78:560–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Jenkins L, et al. Identification of novel species-selective agonists of the G-protein-coupled receptor GPR35 that promote recruitment of β-arrestin-2 and activate Gα13. Biochem J. 2010;432:451–9. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Jenkins L, et al. Agonist activation of the G protein-coupled receptor GPR35 involves transmembrane domain III and is transduced via Gα13 and β‐arrestin‐2. Br J Pharmacol. 2011;162:733–48. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Kaya B, et al. Lysophosphatidic acid-mediated GPR35 signaling in CX3CR1 + macrophages regulates intestinal homeostasis. Cell Rep. 2020;32:107979. [ DOI ] [ PubMed ] [ Google Scholar ] 52. Wang B, et al. V1-Cal hydrogelation enhances its effects on ventricular remodeling reduction and cardiac function improvement post myocardial infarction. Chem Eng J. 2022;433:134450. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Shi FD, et al. Neuroinflammation across neurological diseases. Science. 2025;388:eadx0043. [ DOI ] [ PubMed ] [ Google Scholar ] 54. Neumann J, et al. Microglia cells protect neurons by direct engulfment of invading neutrophil granulocytes: a new mechanism of CNS immune privilege. J Neurosci. 2008;28:5965–75. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Sanchez-Soto M, et al. A structural basis for how ligand binding site changes can allosterically regulate GPCR signaling and engender functional selectivity. Sci Signal. 2020, 13:eaaw5885. 10.1126/scisignal.aaw5885 [ DOI ] [ PMC free article ] [ PubMed ] 56. Park H-B, et al. USP7 regulates the ERK1/2 signaling pathway through deubiquitinating Raf-1 in lung adenocarcinoma. Cell Death Dis. 2022;13:698. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Gao F, et al. Phosphodiesterase 5 inhibitor, zaprinast, selectively increases cerebral blood flow in the ischemic penumbra in the rat brain. Neurol Res. 2013;27:638–43. [ DOI ] [ PubMed ] [ Google Scholar ] 58. McBride DW, et al. Precision stroke animal models: the permanent MCAO model should be the primary model, not transient MCAO. Transl Stroke Res. 2017;8:397–404. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Nour M, et al. Ischemia-reperfusion injury in stroke. Interv Neurol. 2012;1:185–99. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Pan J, et al. Reperfusion injury following cerebral ischemia: pathophysiology, MR imaging, and potential therapies. Neuroradiology. 2006;49:93–102. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1. (37.1MB, docx) Data Availability Statement No datasets were generated or analysed during the current study. Articles from Journal of Neuroinflammation are provided here courtesy of BMC ACTIONS View on publisher site PDF (17.8 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top