GIV/Girdin Modulation of Microglial Activation in Ischemic Stroke: Impact of FTO-Mediated m6A Modification

Abstract

Ischemic stroke (IS) is one of the most common causes of death in the world. The lack of effective pharmacological treatments for IS was primarily due to a lack of understanding of its pathogenesis. Gα-Interacting vesicle-associated protein (GIV/Girdin) is a multi-modular signal transducer and guanine nucleotide exchange factor that controls important signaling downstream of multiple receptors. The purpose of this study was to investigate the role of GIV in IS. In the present study, we found that GIV is highly expressed in the central nervous system (CNS). GIV protein level was decreased, while GIV transcript level was increased in the middle cerebral artery occlusion reperfusion (MCAO/R) mice model. Additionally, GIV was insensitive lipopolysaccharide (LPS) exposure. Interestingly, we found that GIV overexpression dramatically restrained microglial activation, inflammatory response, and M1 polarization in BV-2 microglia induced by oxygen-glucose deprivation and reoxygenation (OGD/R). On the contrary, GIV knockdown had the opposite impact. Mechanistically, we found that GIV activated the Wnt/β-catenin signaling pathway by interacting with DVL2 (disheveled segment polarity protein 2). Notably, m⁶A demethylase fat mass and obesity-associated protein (FTO) decreased the N6-methyladenosine (m6A) modification-mediated increase of GIV expression and attenuated the inflammatory response in BV-2 stimulated by OGD/R. Taken together, our results demonstrate that GIV inhibited the inflammatory response via activating the Wnt/β-catenin signaling pathway which expression regulated in an FTO-mediated m⁶A modification in IS. These results broaden our understanding of the role of the FTO-GIV axis in IS development.

Graphical Abstract

Introduction

Ischemic stroke (IS) is a major cause of mortality and disability worldwide, ranking third in terms of leading causes of death [1, 2]. Reperfusion therapy, specifically thrombolysis, is the current standard treatment for IS. However, it often results in cerebral ischemia/reperfusion injury (CIRI) [3, 4]. CIRI involves various pathological processes, with the inflammatory response mediated by microglia playing a significant role [5, 6]. Therefore, the suppression of the neuroinflammatory cascade is considered a potential therapeutic approach for treating ischemic stroke.

Microglia, a type resident immune cell of the central nervous (CNS) [7], have been regarded as the first line of defense against ischemic stroke [8, 9]. In response to pathophysiological stimulation, activated microglia migrate to the ischemic penumbra in response to pro-inflammatory mediators derived from damaged cells [10]. Microglia can be considered a “double-edged sword” as they can switch between a “pro-inflammatory phenotype” (M1-type) and an “anti-inflammatory phenotype” (M2-type) depending on their morphology and function [11]. It has been proven that following ischemic stroke, microglia initially polarize to the M2 phenotype, releasing IL-4, IL-10, and Arg-1, which are involved in synaptic remodeling, angiogenesis, and neurogenesis [12]. However, during the late stage of ischemic stroke, microglia polarize to the M1 phenotype and secrete pro-inflammatory mediators such as TNF-α, IL-6, and IL-1β, which contribute to brain tissue damage and worsen neuroinflammatory response [13]. Therefore, suppressing M1 microglia or increasing M2 microglia is considered a potentially effective therapeutic approach for preventing ischemic stroke. However, the molecular profile of microglia activation and polarization in ischemic stroke remains unclear.

Coiled-coil domain containing 88A (Ccdc88a), also known as GIV/Girdin, is an actin-binding protein that plays a crucial role in cytoskeletal remodeling and cell migration [14]. It is activated by the serine/complex protein kinase AKT [15]. GIV is involved in various cellular processes, including autophagy [16], vascular endothelial cell growth, and regulation of DNA replication [17]. The GIV family protein has an essential regulatory role in the Wnt/β-catenin signaling pathway [18]. Interestingly, activation of this pathway has been found to have a neuroprotective effect in ischemic stroke [19]. Furthermore, overexpression of GIV suppresses macrophage inflammatory response in colitis [20]. However, the effect of GIV on microglia polarization and inflammatory responses during ischemic stroke remains unclear.

N6-methyladenosine (m⁶A) is a common reversible epigenetic post-transcriptional modification found in RNAs [21]. It plays a crucial role in regulating the splicing, stability, maturation, and translational efficiency of RNAs [22, 23]. The control of this modification is mainly carried out by methyltransferases METTL3/METTL14 complex, demethylase FTO/ALKBH5 [24, 25], and other factors. Recent studies have shown that m⁶A is highly enriched in both embryonic and adult brains, suggesting its significance in brain development and the development of neurological diseases such as stroke and Alzheimer’s disease [26]. Additionally, research has indicated that FTO-mediated modification of m⁶A is involved in neuronal apoptosis [27] and oxidative stress during ischemic stroke [28]. However, the specific role of FTO in regulating neuroinflammatory responses and microglia polarization induced by ischemic stroke has not yet been fully understood.

The results in this study demonstrated that the overexpression of GIV had an inhibitory effect on neuroinflammatory responses and the polarization of microglia towards the M1 phenotype. This inhibition was achieved by activating the Wnt/β-catenin signaling pathway. Additionally, our findings revealed that deficiency of FTO led to the methylation and stabilization of GIV mRNA, resulting in a reduction of GIV protein expression. These findings provide significant evidence for the involvement of the FTO-GIV axis in the regulation of microglial polarization and neuroinflammation during ischemic stroke.

Materials and Methods

Reagents and Antibodies

Lipopolysaccharide (0111: B4, L4391) was purchased from Sigma-Aldrich. The following antibodies were used in this study: FTO (D2V1I) Rabbit mAb (#45980; CST), anti-GIV antibody (ab179481; Abcam), N6-methyladenosine(m6A) (D9D9W) rabbit mAb (#56593; CST), β-actin mouse mAb (#3700; CST), GAPDH monoclonal antibody (60004-1-Ig; Proteintech; Wuhan, China), iNOS rabbit pAb (A14031; ABclonal; Wuhan, China), arginase 1 (ARG1) rabbit pAb (A1847; ABclonal), CD16 polyclonal antibody (16559-1-AP; Proteintech), CD206 polyclonal antibody (18704-1-AP; Proteintech), Iba1 polyclonal antibody (10904-1-AP; Proteintech), GSK-3β rabbit mAb (#12456; CST; USA), β-catenin rabbit mAb (#8480; CST; USA), AXIN2 polyclonal antibody (20540-1-AP; Proteintech), cyclin D1 rabbit mAb (#55506; CST; USA), DVL2 polyclonal antibody (12037-1-AP; Proteintech).

Establishment of the Cerebral Ischemia/Reperfusion Model

The middle cerebral artery occlusion reperfusion (MCAO/R) mice model was established to mimic cerebral ischemia-reperfusion according to a previous study [29]. Male C57BL/6 mice weighing approximately 25 g and aged between 8 and 10 weeks were used in this experiment. All experiments were conducted in accordance with the Animal Ethics Committee of Guizhou Medical University. Mice were anesthetized with 1–1.5% isoflurane and placed in a supine position. The right common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA) were then sequentially exposed and carefully separated. Subsequently, the CCA and ECA were ligated, followed by clamping of the internal carotid artery with an arterial clamp. A silicone nylon monofilament was inserted from the CCA into the middle cerebral artery and temporarily secured. After 1 h of ischemia, the monofilament was removed, and the blood vessels were ligated at the incision site. The neck wound was closed with surgical sutures. Experiments were conducted after 24 h of reperfusion. In the sham operation group, the same surgical procedures as the MCAO/R model group were performed, with the exception of the monofilament insertion.

Histopathological Staining

The brains were extracted from different groups of mice and fixed in 4% paraformaldehyde solution for 24 h. The brain tissue was dehydrated in graded concentrations of ethanol and xylene, followed by embedding in paraffin. Subsequently, paraffin sections (approximately 4 μm) were stained with H&E or Nissl staining sealed with resin. The histology of the cortex was examined using an optical microscope. For TTC staining, the whole brain was isolated after MCAO/R operation and placed in saline. Approximately 5 slices were cut from each brain and placed in a TTC staining solution with a concentration of 2%. The slices were then incubated for 30 min at 37 °C in an incubator. The infarcted and un-infarcted areas of the slices were analyzed using ImageJ.

Cell Lines and Culture Conditions

BV-2, HMC3, and THP-1 cells were obtained from ATCC (Shanghai, China) and cultured in Gibco™ BASIC DMEM/1640 medium containing 5% fetal bovine serum (FBS) under 5% CO₂ conditions. For the OGD/R treatment, the cells were cultured in free-glucose DMEM medium and placed in a chamber with N₂/CO₂/O₂ (95% N₂, 5% CO₂, and 1% O₂₎ for 4 h followed by 24 h of reoxygenation and normal glucose DMEM medium [30]. BV-2 cells were treated with LPS at a concentration of 1 μg/ml for 24 h to induce LPS challenge [31].

Immunofluorescence Staining

THP-1 cells were seeded in 12-well plates and treated with 400 nM phorbol myristate acetate (PMA; MCE, USA) for 24 h [32]. The cells were then washed with 1 × PBS and fixed with 4% paraformaldehyde. After permeabilization with 0.5% Triton X-100, the cells were treated with 5% BSA. Subsequently, the cells were incubated overnight at 4 °C with primary antibodies DVL2 (1:200; Proteintech; Wuhan, China) and GIV (1:250; Abcam; USA). Sections were then incubated with fluorescein secondary antibodies (1:250; Proteintech) before being measured using a microscope (Olympus, Japan)

RNA Extraction and Quantitative RT-PCR

Total RNA was isolated from murine brain tissues and BV-2 cells using TRIzol reagent (Absin, Shanghai, China). Reverse transcription was performed using the PrimeScript RT Reagent Kit (Vazyme, Nanjing, China). qPCR was conducted on an ABI Prism 7900HT (Applied Biosystems, Foster City, CA, USA) using ChamQ SYBR qPCR Master Mix (Vazyme, Nanjing, China). The mRNA expressions of target genes were normalized to β-actin and calculated using the 2−ΔΔCt method. The primers used are listed in Table 1.

Table 1.

The qRT-PCR primers

Gene	Forward 5’-3’	Reverse 5’-3’
CCDC88A	AGGACAGTGGACAGGTAG	AGGACAGTGGACAGGTAG
ACTB	TGCGTGACATTAAGGAGAA	AAGGAAGGCTGGAAGAGT
Ccdc88a	GACAAGAGCAGCCAAGAC	CTGTTGTGGTCCTTCCTTC
m6A (Ccdc88a)	AGAGGAGCACAAGACAGA	TTCGTTGGTGTCTCTTGC
Actb	GTGCTATGTTGCTCTAGACTTCG	ATGCCACAGGATTCCATACC
Nlrp3	AGACCTCCAAGACCACTAC	ACATAGCAGCGAAGAACTC
Cox2	ATCCTTGCTGTTCCAATCC	TTCACCATAGAATCCAGTCC
TNF-a	GTGGAACTGGCAGAAGAG	GAGAAGAGGCTGAGACATAG
IL-6	TCCATCCAGTTGCCTTCT	TAAGCCTCCGACTTGTGA
IL-1β	CTTCAGGCAGGCAGTATC	CAGCAGGTTATCATCATCATC
iNOS	CAGGAGATGTTGAACTATGTC	TTGGTGTTGAAGGCGTAG
Gsk3β	AGGAACACCAACAAGGGAGC	TCCTGGGGTGAAATGTCCTG
Ctnnb1	CAGCGACTAAGCAGGAAG	GACGAAGAGCACAGATGG
Fzd1	TCCTACCTCAACTACCACTT	ACGGACCAGATGCCTATC
Fzd2	CGCCTGCTACTTCTATGAG	CTGTTGGTGAGACGAGTG
Ccnd1	CAGAAGTGCGAAGAGGAG	GGATAGAGTTGTCAGTGTAGA
Axin 2	GTTCACCACCACTACATCC	CATCCACTGCCAGACATC

RNA Immunoprecipitation (RIP)

RIP assays were conducted following the instructions provided by BersinBioTM RNA Immunoprecipitation Kit (BersinBio; Guangzhou, China) as described in ref. [33]. A total of 5 × 10⁶ cell samples were collected, DNA was eliminated, and 20 μL of protein A/G beads were prepared for each RIP sample group. Antibody FTO was added to the IP samples, which were then incubated overnight at 4 °C in a vertical mixer. RNA was subsequently eluted and extracted for RT-qPCR analysis.

m6A Dot Blot

As described in reference [34], total RNA was isolated from BV-2 cells using TRIzol (Absin; Shanghai, China) following the manufacturer’s instructions. The quality of the RNA was analyzed using a NanoDrop (Thermofisher; USA). Each sample of RNA (100, 250, and 500 ng) was examined by M6A antibody (#56593; 1:1000; CST; USA), and the solution was blotted onto the NC membrane.

Flow Cytometry

BV-2 cells, transfected with indicated plasmids and subjected to oxygen-glucose deprivation/reoxygenation (OGD/R), were harvested and stained with anti-CD16 and anti-CD206 antibodies (Proteintech; Wuhan, China). The cells were then analyzed by flow cytometry.

Western Blot

Mouse brain tissues and cultured cells were lysed in RIPA buffer (Beyotime, China) containing a protease inhibitor. After centrifugation, the protein concentrations in the supernatants were determined using a BCA protein assay kit (Biosharp, China). The protein was separated by SDS-PAGE on a 12% gel and subsequently transferred to PVDF membranes. The membranes were blocked with 5% non-fat milk and incubated overnight at 4 °C with the related primary antibodies. After washing, the membranes were incubated with the secondary antibody for 1 h at room temperature. Finally, the blots were captured using ECL and analyzed using Image J (V1.8.0).

Bioinformatics Analysis and Molecular Docking

The GSE58294 dataset was obtained from GEO Data Sets. Tissue distribution expression data for GIV were extracted from the Human Protein Atlas database (https://www.proteinatlas.org/). Enrichment analysis, specifically for the analysis of DVL2 and GIV, was performed using GSEA 4.2.3. The structural data was sourced from the PDB database. Protein-protein molecular docking analysis was conducted using the ClusPro2.0 protein-protein docking online software (https://cluspro.org/login.php). The docking results were visualized using Pymol 2.1.

Co-immunoprecipitation

Co-IP assays were performed using Protein A/G-Agarose (Absin; Shanghai, China) following the manufacturer’s instructions. After thorough washing, the electrophoresis loading buffer was added to the complexes and incubated at 95 °C for 10 min. Immunoprecipitated proteins were separated by SDS-PAGE and analyzed by immunoblotting using the indicated antibodies.

MeRIP-qPCR

As previously described in ref. [34], in accordance with the guidelines provided by the commercial kit (BersinBio™ Methylated RNA Immunoprecipitation Kit, Guangzhou, China), total RNA was extracted from BV-2 cells and fragmented into 100-nt-long fragments. The MeRIP-qPCR assay was conducted using a positive antibody (m6A antibody) and a negative antibody (IgG antibody) in separate groups. The samples were then incubated in a vertical mixer. Subsequently, RNA was extracted for RT-qPCR analysis.

RNA Decay Assay

Cells were treated with 2.5 μg/mL of actinomycin D (MCE, USA) for 0, 2, 4, 6, and 8 h after FTO knockdown. Total RNA was then extracted for real-time quantitative RT-PCR.

Statistical Analysis

The statistical analyses were conducted using GraphPad prism9.0 software. The data are presented as mean ± SE. Significant differences were assessed using Student’s t-test and one-way analysis of variance (ANOVA), with a significance level of P < 0.05.

Results

GIV Was Highly Expressed in the CNS

The distribution and expression of GIV in various tissues of mice were observed using THE HUMAN PROTEIN ATLAs (https://www.proteinatlas.org/). GIV mRNA expression was mainly found in the cerebral cortex and white matter, as shown in Fig. 1A. Additionally, moderate levels of GIV expression were observed in the spleen and lymphatic tissues. In terms of protein expression, GIV was prominently localized within the cortex, hippocampus, and thalamus regions of the CNS (Fig. 1B). Further examination of GIV mRNA and protein expression was conducted in the heart, liver, spleen, lung, kidney, cerebellum, and cerebrum of mice. The results were detected by qRT-PCR and Western blotting assay, as depicted in Fig. 1C and D respectively. The mRNA of GIV was primarily expressed in the cerebrum, cerebellum, kidney, lung, and heart. Meanwhile, the protein was predominantly expressed in the cerebrum, cerebellum, and spleen. These findings highlight that GIV is predominantly expressed in the CNS and moderately expressed in the spleen and lymph.

Fig. 1.

Expression and distribution of GIV in mouse tissue. A Relative expression of the GIV gene in various tissues generated using Human Protein Atlas (www.proteinatlas.org). B Relative expression of GIV protein in various tissues generated using Human Protein Atlas (www.proteinatlas.org). C mRNA expression of GIV in different organs of mice examined by qRT-PCR. D Protein expression of GIV in different organs of mice detected by Western blotting

GIV Protein was Reduced But GIV mRNA Was Increased in Experiment Stroke Model

To investigate the involvement of GIV in ischemic stroke-induced neuroinflammation, we established an ischemic stroke model on C57BL/6 mice using the MCAO/R mice model. The MCAO/R group showed a significant increase in the infract volume compared to the Sham group, as depicted in Fig. 2A and B. Additionally, the Longa score was significantly higher in the MCAO/R group (Fig. 2C). Histological analysis using H&E staining (Fig. 2D) and Nissl staining (Fig. 2E and F) revealed pathological changes in the brain of the MCAO/R group, including nuclear fragmentation of neuronal cells, cytoplasmic loss, and loss of Nissl bodies. The results above mentioned indicated that MCAO/R-operation induced cerebral ischemia-reperfusion injury in mice.

Fig. 2.

GIV expression was downregulated in the mice model of ischemic stroke and cell model of OGD/R. A, B TTC staining of brains showed that MCAO/R induced an increase in the infarct volume compared with the sham group. C Neurological function was evaluated using the Longa score. D H&E staining of the cerebral cortex region revealed histopathological changes induced by MCAO/R, including loose cytoplasm, edema, and nuclear division. E, F Nissl staining of cerebral cortex region revealed that MCAO/R decreased the percentage of Nissl bodies and promoted neuronal loss compared with the sham group. G GIV protein levels were reduced in the brain periinfarct area from MCAO/R group mice. H Immunofluorescence of GIV in Sham or MCAO/R group mice. I GIV protein levels were reduced in OGD/R group BV-2 cells. J GIV protein levels were reduced in OGD/R HMC3 cells. K LPS exposure did not affect the protein expression of GIV. L GIV mRNA was increased in the brain periinfarct area from MCAO/R group mice and OGD/R group HMC3(M) cells (n ≥ 3, **P < 0.01)

Previous study found that GIV was preferentially expressed in monocyte (such as macrophages and dendritic cells), and modulate macrophage polarization [20]. To explore possible functions of GIV involved in ischemic stroke, we detected the dynamic changes of microglial GIV protein level by Western blotting assay and immunofluorescence staining in mice brain slices. We found that the GIV protein level was decreased after ischemia-reperfusion injury (Fig. 2G). Notably, GIV localized in Iba-1-positive cells in physiological condition, but when exposed to ischemia-reperfusion, GIV protein level was dramatically reduced (Fig. 2H). Similar results were observed in BV-2 cells and HMC3 cells induced by OGD/R, as shown in Fig. 2H and I. However, exposure of BV-2 cells to LPS for 6 h, 12 h, and 24 h did not significantly alter the protein expression of GIV (Fig. 2J). Notably, mRNA levels of GIV were increased in the experimental stroke model of mice and HMC3 cells, as depicted in Fig. 2K and L. There are two main subtypes of activated microglia, called M1 microglia (pro-inflammatory type) and M2 microglia (anti-inflammatory type) followed by ischemic stroke. Accordingly, our results suggested that GIV was involved in modulation microglia polarization after ischemic stroke.

Knockdown of GIV Increased Production of Inflammatory Cytokines and Microglia M1-Type Polarization

To evaluate the impact of GIV on microglia-mediated inflammation, BV-2 cells were transfected with specific LV-shRNAs targeting GIV. Pro-inflammatory cytokine production was then measured after OGD/R challenge. Western blot results confirmed the successful knockdown of GIV, as shown in Fig. 3A, B. Knockdown of GIV resulted in increased mRNA levels of TNF-a, IL-6, IL-1β, iNOS, Cox-2, and NLRP3 in BV-2 microglia with or without OGD/R treatment, as depicted in Fig. 3C–H. Additionally, Western blot assay was performed to assess the expression of specific biomarkers for M1 and M2 phenotypes. The protein level of iNOS and CD16 increased in BV-2 cells after OGD/R challenge (Fig. 3I–J), while the protein level of CD206 and Arg-1 were decreased (Fig. 3K). Furthermore, the expression of Iba1 was increased in BV-2 cells after GIV knockdown (Fig. 3L). These findings indicate that GIV loss promoted microglia M1-type polarization and inflammatory response in ischemic stroke.

Fig. 3.

GIV deficiency aggravates inflammatory response in BV-2 cells induced by OGD/R. A, B Representative images of GIV shRNA knockdown efficiency detected by Western blot. C–G The relative mRNA expression levels of TNF-α, IL-6, IL-1β, iNOS, Cox-2, and NLRP3 after GIV knockdown. H Representative images of Western blot. I–J The relative protein expression levels of iNOS and CD16, CD206, and Arg-1 after GIV knockdown. K The Iba1 protein expression level after GIV knockdown. (n ≥ 3, *P < 0.05, **P < 0.01)

Upregulation of GIV Suppressed Inflammatory Cytokines Release and Promoted Microglia M2-Type Polarization

To investigate the inhibitory effects of GIV on neuroinflammatory responses induced by OGD/R, we conducted adeno-associated virus (AAV) transfection to overexpress GIV [16] in BV-2 cells. The results, as depicted in Fig. 4A–H, demonstrated that GIV overexpression decreased the mRNA expression levels of TNF-a, IL-6, IL-1β, iNOS, Cox-2, and NLRP3. Moreover, GIV overexpression resulted in a decrease in the protein expression of CD16 and iNOS, classical markers of M1 phenotype (Fig. 4I–J), increase in the protein expression of CD206 and Arg-1, markers of M2 phenotype (Fig. 4I and K), and a decrease in the expression of Iba1protein level, a marker for microglia activation (Fig. 4L). Overall, our results suggested that upregulation GIV effectively suppresses the expression of pro-inflammatory factors induced by OGD/R and promotes the transformation of microglia from the M1 phenotype to the M2 phenotype.

Fig. 4.

GIV overexpression inhibits the inflammatory response in BV-2 cells induced by OGD/R. A, B Representative images of GIV overexpression were detected using Western blot. The relative mRNA expression levels of TNF-α (C), IL-6 (D), IL-1β (E), iNOS (F), Cox-2 (G), and NLRP3 (H) were measured after GIV overexpression. The representative images of Western blot are shown in I. The relative protein expression levels of iNOS and CD16 after GIV overexpression are shown in J. The relative protein expression levels of CD206 and Arg-1 after GIV overexpression are shown in K. The Iba1 protein expression level after GIV overexpression is shown in L (n ≥ 3, *P < 0.05, **P < 0.01)

GIV Regulated Microglia M1 and M2 Polarization

Flow cytometry was employed to evaluate the levels of CD16-positive cells and CD206-positive cells in BV-2 microglial cells exposed to OGD/R after knockdown and overexpression of GIV. The findings, depicted in Fig. 5A–D, demonstrated that overexpression of GIV notably enhanced the presence of CD206-positive cells while decreasing the number of CD16-positive cells. These outcomes suggest that GIV has the potential to enhance the ratio of M2/M1 microglial transformation, thereby mitigating the neuroinflammatory response induced by OGD/R.

Fig. 5.

GIV regulates BV-2 microglia M1/M2 polarization. A The representative images of CD16-positive cells by flow cytometry after GIV knockdown and overexpression. B The CD16-positive cells count after GIV knockdown and overexpression. C The representative images of CD206-positive cells by flow cytometry after GIV knockdown and overexpression. D The CD206-positive cells count after GIV knockdown and overexpression (n = 3, **P < 0.01)

GIV Activated the Wnt/β-Catenin Signaling Pathway

Based on the bioinformatics analysis of the GEO Data Sets, the transcript level of GIV was increased in peripheral blood samples from stroke patients compared with healthy controls (Fig. 6A). Additionally, the results of GSEA showed that the high expression of GIV significantly activated the Wnt/β-catenin signaling pathway (Fig. 6B). Furthermore, the transcript level of GIV was observed in the OGD/R model of BV-2 cells (Fig. 6C). When GIV was overexpressed in BV-2, it led to an upregulation of the transcript levels of Gsk3β, Ctnnb1, Ccnd1, Fzd1, and Fzd2, which are classical markers of the Wnt/β-catenin signaling pathway. On the other hand, knockdown of GIV significantly inhibited the expression of Gsk3β, Ctnnb1, Ccnd1, Fzd1, and Fzd2 (Fig. 6D–I). Therefore, these results suggest that the upregulation of GIV activates the Wnt/β-catenin signaling cascade.

Fig. 6.

GIV regulates the Wnt/β-catenin signaling pathway. A comparison of GIV transcript levels in the control and Stroke groups from GEO DataSets (GSE58294) revealed significant differences. B Gene set enrichment analysis indicated an upregulation of the Wnt/β-catenin signaling pathway in the Stroke group. C The mRNA expression of GIV was compared between the control and OGD/R groups in BV-2 microglia. The relative mRNA expression levels of Gsk3b (D), Ctnnb1 (E), Ccnd1 (F), Fzd1 (G), Fzd2 (H), and Axin2 (I) were measured after GIV knockdown or overexpression (n ≥ 3, **P < 0.01)

GIV Interacted with DVL2 and Increased Its Stability to Regulate the Wnt/β-Catenin Signaling Pathway

The mechanism by which GIV regulates the Wnt/β-catenin signaling pathway in ischemic stroke remains unknown. To investigate this, a protein-protein interaction network analysis (PPI) was conducted to identify the interaction partner of GIV. The analysis revealed that GIV can interact with DVL2, a key regulatory molecule in the Wnt signaling pathway (Fig. 7A). This interaction was further confirmed through molecular docking, which showed a minimum binding energy of − 1624.8 Kcal/mol (Fig. 7B). Moreover, the overexpression of GIV significantly increased DVL2 expression and activated the Wnt/β-catenin signaling pathway (Fig. 7C). Conversely, the knockdown of GIV inhibited GIV expression and the activation of the Wnt/β-catenin signaling pathway (Fig. 7D). Further evidence for the interaction between GIV and DVL2 was provided through immunofluorescence staining (Fig. 7E) and co-immunoprecipitation (Fig. 7F). Dose GIV interacted with DVL2 enhance the protein stability of DVL2? Further the results of cycloheximide (CHX) assay demonstrated that the overexpression of GIV increased the half-life of DVL2 (Fig. 7G). In summary, our results indicate that GIV can directly bind to DVL2, thereby increasing its stability and activating the Wnt/β-catenin signaling pathway.

Fig. 7.

GIV interacts with DVL2 and activates Wnt/β-catenin signaling pathway. A Protein-protein interaction network analysis (PPI). B Structure-based protein interaction interface analysis between GIV and DVL2. C Representative images of Western blot after GIV overexpression. D Representative images of western blot after GIV knockdown. E Representative images of immunofluorescent staining. F DVL2, GIV, and IgG antibody, THP-1 lysates were immunoprecipitated using DVL2 antibody and then analyzed by Western blot using the indicated antibodies. G Half-life of DVL2 in BV-2 cells with Vector or overexpression of GIV, treated with cycloheximide (CHX) at the indicated times and analyzed by Western blot (n = 3, **P < 0.01, scale bar = 20 μm)

FTO Modulated Neuroinflammation and GIV Expression

Previous studies have reported that FTO has neuroprotective effects in ischemic stroke [35]. However, it has not been reported whether FTO regulates the neuroinflammatory response induced by ischemic stroke. Our results demonstrate that the expression of GIV and FTO was decreased in BV-2 cells induced by OGD/R (Fig. 8A). Moreover, overexpression of FTO suppressed the expression of Cox-2 and IL-1β (Fig. 8B), while knockdown of FTO increased their expression (Fig. 8C). Additionally, through online m6Atarget prediction analysis, we identified GIV as a potential target gene of FTO in the RNA sequencing results of mouse B16-OVA cells, showing m6A modification and increased expression (Fig. 8D). Knockdown of FTO resulted in reduced expression of GIV (Fig. 8F), whereas FTO overexpression led to increased expression of GIV (Fig. 8E). These results suggest that FTO may regulate the expression of GIV through m⁶A modification and suppress neuroinflammation induced by OGD/R.

Fig. 8.

FTO suppressed inflammatory response and modulates GIV expression. A BV-2 cells challenged with OGD/R, FTO, and GIV protein expression was reduced. B FTO suppressed COX-2 and IL-1β expression in BV-2 cell challenged with OGD/R. C FTO knockdown increased COX-2 and IL-1β expression. D GIV is a target gene of FTO predicated by RM2Target software. E FTO overexpression promotes GIV expression of BV-2 cells challenged with OGD/R. F FTO Knockdown suppresses GIV protein expression in BV-2 (n = 3, **P < 0.01)

FTO Regulated GIV mRNA Stability by m6A Modification

Our previous results demonstrated that FTO can regulate the expression of GIV, but its specific mechanism is unclear. Using online SRAMP software, we identified multiple high-confidence m⁶A modification sites in the mRNA CDS region of GIV (Fig. 9A). Among these sites, the motif GAACU consistently scored the highest confidence (Fig. 9B). Furthermore, we detected the m⁶A modification level of total RNA in BV-2 cells after overexpressing FTO, and observed a reduction in m6A modification level (Fig. 9C). MeRIP-PCR results indicated that overexpression of FTO significantly decreased the level of m⁶A-modified GIV mRNAs (Fig. 9D). Additionally, RIP-PCR results demonstrated the binding of FTO to GIV mRNAs (Fig. 9E). Finally, knockdown of FTO was found to increase the RNA half-life of GIV (Fig. 9F). Collectively, these findings suggest that FTO regulates the expression of GIV in experimental models of ischemic stroke by increasing the m⁶A modification level, thereby enhancing the stability of GIV mRNAs. Consequently, this leads to a decrease in GIV protein translation efficiency.

Fig. 9.

FTO knockdown increased GIV mRNA stability by m6A modification. A The GIV m⁶A site was predicted using the SRAMP software. B The sequence of GIV m⁶A modification was determined. C Overexpression of FTO resulted in a reduction in m⁶A modification, as confirmed by dot blot assay. D After FTO overexpression, the m⁶A modification of GIV mRNA expression was found to be reduced, as determined by MeRIP-PCR. E The RIP-PCR assay detected the enrichment of GIV mRNA precipitated by the anti-FTO antibody. F RNA stability analysis revealed a decrease in GIV mRNA expression in FTO knockdown BV-2 cells treated with Actinomycin D (Act. D) (n = 3, **P < 0.01)

Discussion

Neuroinflammation, which refers to a sequence of inflammatory responses occurring in nervous tissue, is increasingly recognized as the key regulator of the pathological progression of ischemic stroke [36, 37]. Inflammation plays a role in all stages of cerebral ischemia/reperfusion injury [38, 39]. Interestingly, GIV mRNA levels were also elevated in samples from ischemic stroke patients. Moreover, GIV did not show significant changes upon exposure to LPS. Overexpression of GIV leads to a decrease in the expression of pro-inflammatory factors and promotes M2 polarization of microglia by activating the Wnt/β-catenin signaling pathway. Conversely, knockdown of GIV inhibits the Wnt/β-catenin signaling pathway and increases the expression of pro-inflammatory factors, thereby promoting M1 polarization of microglia. Additionally, we discovered that GIV is a target gene of FTO, and FTO regulates the mRNA stability of GIV through m6A modification, thereby affecting the translation efficiency of GIV. Intriguingly, overexpression of FTO reduces the release of pro-inflammatory factors and inhibits the inflammatory response induced by ischemia-reperfusion.

Microglia dysregulation is known to contribute to progressive tissue damage after ischemic stroke [40, 41]. However, the mechanisms controlling pathogenic microglial gene expression remain unclear [42, 43]. In this study, we demonstrate that GIV distributes in microglia and is reduced in an experimental model of ischemic stroke in vitro and in vivo. However, GIV is insensitive to LPS exposure and extremely sensitive to ischemia and hypoxia. This suggests that GIV is involved in regulating the inflammatory response induced by ischemic stroke. Previous studies have reported that GIV binds to TLR4, inhibiting the inflammatory response of macrophages by suppressing the NF-κB signaling pathway in colitis [20]. However, the role of GIV in microglia-mediated neuroinflammation remains unclear during ischemic stroke. In this study, we demonstrate that overexpression of GIV reduces inflammatory cytokine production in microglia after OGD/R exposure. Interestingly, previous studies have also shown that GIV phosphorylation is crucial for synaptic plasticity and memory through the regulation of neuron migration [44]. Nevertheless, our present study shows that GIV can suppress M1 microglia polarization and neuroinflammation induced by OGD/R. Additionally, we discovered that GIV interacts with DVL2 and increased its protein stability to activate the Wnt/β-catenin signaling pathway. DVL2 is an adaptor protein of Wnt/β-catenin signaling, with its upregulation leading to subsequent pathway activation [45]. Emerging studies have confirmed that GIV family proteins can regulate the Wnt/AKT signaling pathway [14, 18]. Activation of the Wnt/AKT signaling pathway exerts a neuroprotective effect in ischemic stroke [46, 47]. Collectively, our present study and previous studies also demonstrate that GIV plays vital role in CNS disease. Notably, our study found that upregulation of GIV alleviates neuroinflammatory response via promotion microglia M1 to M2 phenotype transition by activating the Wnt/β-catenin signaling pathway during ischemic stroke.

N6-methyladenosine demethylase FTO is responsible for catalyzing and regulating the m6A modification of target genes, thereby affecting their expression and function [48, 49]. Numerous studies have reported a decrease in FTO levels in both ischemic stroke patients and experimental models, indicating its involvement in regulating the function of molecular targets in ischemic stroke models [27, 35]. Overexpression of FTO has been found to inhibit brain injury induced by ischemic stroke and promote cerebral vascular repair in stroke model animals [48]. These findings suggest a neuroprotective role for FTO. Additionally, FTO-mediated m⁶A modification modulates microglial phenotype via GPC4/TLR4/NF-κB signaling axis to ameliorate autoimmune uveitis [50]. In ischemic stroke, FTO reduces cerebral ischemia/reperfusion-induced neuroinflammation in an m⁶A-dependent manner by reducing cGAS mRNA stability [51]. These studies suggested that FTO-mediated m⁶A modifications play an important role in regulation of neuroinflammation. Consistent with previous studies, we demonstrated a significant reduction in FTO expression in OGD/R-induced microglia models. Moreover, overexpression of FTO suppressed the release of inflammatory cytokines induced by OGD/R. Further investigation revealed that FTO acts as a demethylase, regulating the target gene GIV. Overexpression of FTO upregulated GIV expression, while knockdown of FTO significantly increased m6A modification, resulting in an extended mRNA half-life of GIV. This, in turn, led to the obstruction of GIV translation and a subsequent decrease in protein expression, thereby exacerbating the neuroinflammation induced by ischemia-reperfusion. Above all, we provide the first evidence that FTO-m⁶A-GIV axis in regulation of microglia polarization and pathogenesis.

However, it is important to acknowledge some limitations in our study. Firstly, although our findings have confirmed that GIV levels are significantly decreased in both mouse and cellular models of ischemic stroke. However, the mRNA level of GIV was increased in peripheral blood samples from patients diagnosed with stroke. It is acknowledged that experimental models developed in young, healthy rodents do not adequately replicate the pathological conditions encountered in clinical cases of ischemic stroke. Patients suffering from ischemic stroke frequently display a diverse array of heterogeneity of pathophysiological factors, including position, severity, and duration. Consequently, future research should further investigate the influence of variables such as age, sex, genetic background, and comorbidities on the pathological changes associated with ischemic stroke. Secondly, regarding to the mechanisms through which GIV regulates microglial phenotype transformation and inflammatory responses, we did not perform a comprehensive analysis in animal models, such as mice, which may impose limitations on the conclusions drawn from our study. In subsequent studies, we intend to acquire additional clinical samples from stroke patients and perform detailed analyses in animal models, such as mice. Nevertheless, despite these limitations, our research contributes valuable theoretical insights and experimental evidence concerning the role of the FTO-m⁶A-GIV axis in modulating neuroinflammatory pathological changes induced by ischemic stroke.

Taken together, our study revealed a novel role for GIV in controlling microglia M1/M2-type polarization and inflammatory response through the activation of the Wnt/β-catenin pathway in ischemic stroke. Furthermore, we found that the expression of GIV is regulated by FTO-mediated m6A modification. This study provides valuable insights into the neuroprotective effects of FTO and suggests that the FTO-GIV axis could be a promising therapeutic target for ischemic stroke.

Authors’ Contributions

Peng Xie and Wenfeng Yu designed the research; Peng Xie and Mingyan Xia, Tingting Long, Dongfen Guo and Wenpeng Cao performed most experiments; Peng Xie analyzed the data and drafted the manuscript. Ping Sun and Wenfeng Yu revised the manuscript. All authors read and approved the final manuscript.

Funding

This study was funded from the National Natural Science Foundation of China (No. 82060232 and No.82160225), Basic Science Technology Project of Guizhou Province, China [grant no. ZK [2021]412], Department of Education of Guizhou Province (Guizhou Teaching and Technology [2023]015), and Key Projects of Science and Technology Fund of Guizhou Provincial Department of Science and Technology [No: Qiankeheji [2020] 1Z060].

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics Approval

The use and care of animals as well as all the experimental protocols were approved by the Ethic Committee of Guizhou Medical University (approval number:2000792) and were strictly in accordance with animal care and use guidelines of the National Institutes of Health.

Consent to Participate

Not applicable.

Consent for Publication

Consent for publication was obtained from the participants.

Competing Interests

The authors declare no competing interests.