IFP35, a novel DAMP, aggravates neuroinflammation following acute ischemic stroke via TLR4/NF-κB/NLRP3 signaling

Abstract

Background

Acute ischemic stroke is a disastrous disease characterized by damaging blood flow in the brain, leading to acute brain injury. Acute brain ischemia elicits severe inflammation, thus in turn, aggravates neural injury. Interferon-Induced Protein 35 (IFP35), is a 35 kDa protein, a novel type of DAMP that trigger inflammatory responses, exacerbating acute and chronic inflammatory disease. This study aimed to investigate the potential neuroinflammation role of IFP35 in acute ischemic stroke in a mouse model of MCAO.

Methods

C57BL/6 male mice were subjected to middle cerebral artery occlusion (MCAO) to establish an animal model of acute ischemic stroke. Leveraging serum from stroke patients, serum and brain tissue after MCAO mice, IFP35 was released. Immunofluorescence assay was used to investigated the cell sources of IFP35 expression after MCAO. The impact of IFP35 on neuroinflammation and neural injury was assessed by siRNA-mediated cerebral IFP35 knockdown. Behavioral tests, and brain tissues were harvested for histological analysis and biochemical assays. TUNEL assays were used to evaluate neuronal damage. TTC staining was performed to assess infarction volumes. Additionally, using western blotting and immunofluorescence assays, we further assessed the contribution of TLR4/NF-κB/NLRP3 signaling in MCAO mice and BV2 cells.

Results

IFP35 was accumulated in peripheral blood of cerebral ischaemia patients, ischemia mice serum, as well as peri-infarct regions in focal cerebral ischemia mice. Although endothelial cells, microglia, and astrocytes are capable of expressing IFP35, cerebral neural cells seem to express and release more IFP35 compare to other cell types. Knockdown of IFP35 alleviated the production of neuroinflammatory cytokines, decreased neuronal death, and minimized infarct volumes, ultimately leading to improved neurological outcomes. Importantly, IFP35 triggered the activation of NF-κΒ and NLRP3 signaling, exacerbating neuroinflammation and brain injury by binding its receptor TLR4.

Conclusions

This study revealed IFP35 as a novel DAMP released during cerebral ischemia that promotes neuroinflammation and injury, expanding the current understanding of inflammatory networks following stroke.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12974-025-03492-6.

Introduction

Acute cerebral ischemia is a devastating neurologic injury disease due to its high disability and mortality rate and limited treatment options [1, 2]. Emerging evidences revealed that inflammation and immune cells invasion are key factors involved in the ongoing of brain injury within minutes to a few hours after acute ischemic stroke [3–5]. Immediately after ischemia, brain-resident microglia are the first activated immune cells and followed by infiltrating leucocytes such as neutrophils, monocytes, natural killer cells, T cells and B cells [1, 6]. The release of alarmins or danger signals by dying neuronal cells such as damage-associated molecular patterns (DAMPs) guide these immune cells to the injured brain region and elicit localized inflammation that, in turn, exacerbate brain infarction and microvascular dysfunction [7, 8].

As the integral part of the innate immune system, inflammation is initiated by various chemical mediators, including some derived directly from pathogens called pathogen-associated molecular patterns (PAMPs) [9]. Alternatively, the inefficient clearance of dying cells and cellular debris leads to the release of intracellular inflammatory substances into the extracellular environment, so-called damage associated molecular patterns (DAMPs) [10, 11]. Once DAMPs are released, they engage pattern recognition receptors (PRRs) such as Toll-like receptors (TLRs) and receptor for advanced glycation end-products (RAGE) in activating nuclear factor κB (NF-κB), mitogen-activated protein kinases (MAPK), as well as other pathways to trigger production of inflammatory cytokines (e.g., TNF-α, IL-1β and IL-6) even in the absence of microorganisms, which further accelerate localized inflammatory responses and tissue injury [12–14]. This process is called sterile inflammation, which is often found in ischemia-reperfusion injury (IRI), trauma, or chemically induced injury [10, 11]. Some classic DAMPs have been identified, such as heat shock proteins (HSPs) [15], high-mobility group box protein 1 (HMGB1) [16], and S100A [17] et al.

Interferon-Induced Protein 35 (IFP35), is a 35 kDa protein firstly identified as an interferon-inducible leucine zipper protein that translocate from the cytoplasm to the nucleus upon α/β-interferon stimulation for anti-viral immune responses [18, 19]. Additionally, the function of IFP35 extends to the modulation of autoimmune diseases such as rheumatoid arthritis (RA) [20] and systemic lupus erythematosus (SLE) [21], as well as cancer like colorectal [22] and renal cancer [23]. Besides, in cases of cellular damage, existing reports indicate that intracellularly released IFP35 might function as a novel type of DAMP that trigger inflammatory responses, exacerbating conditions like lipopolysaccharide (LPS)-induced acute lung injury [24] and chronic neurological diseases like multiple sclerosis (MS) [25]. Meanwhile, released IFP35 has also been recognized as a danger signal that can induce and worsen cytokine storm syndrome (CSS) so that it is considered a potential therapeutic target for CSS induced by SARS-CoV-2 or influenza viruses [26]. Thus, the evidence indicates that IFP35 assumes a pivotal role as a regulator in both the biology and pathophysiology of inflammation. Therefore, we propose that, acting as a novel DAMP, IFP35 could potentially promote inflammation and worsen acute cerebral ischemia.

Recently, the inflammasomes have been identified as a key element in the innate immune response following ischemic stroke [27]. Inflammasomes, a type of complexes formed by cytosolic proteins, orchestrate the inflammatory response to both pathogen infections and host damage [28]. Typically, an inflammasome consists of a cytosolic PRR, an adaptor molecule named ASC, and the effector caspase-1 [29]. Various PRRs have the potential to assemble inflammasomes, among which are NLRP1, NLRP3, AIM2, NLRC4, and IFI16 [30]. Of these, NLRP3 stands out as the most extensively studied and characterized [31]. Upon activation by a variety of sterile danger signals in tissues, NLRP3 inflammasome initiates its self-oligomerization, then assembles the complexes in conjunction with ASC and pro-caspase-1, leading to the self-processing of caspase-1, converting it into its active form cleaved-caspase-1. Active caspase-1 then facilitates the cleavage of pro-IL-1β and pro-IL-18 to augment inflammatory responses [32, 33].

In the present study, we found that serum IFP35 increased in individuals with acute ischemic stroke and a significant increase in IFP35 levels in serum and peri-infarct areas in MCAO mice. In addition, we also observed that although endothelial cells (at the early phase of stroke), microglia, and astrocytes are capable of expressing some IFP35, cerebral neural cells seem to express and release more IFP35 compare to other cell types. Furthermore, administering siRNA against IFP35 prior to MCAO markedly reduced pro-inflammatory cytokine levels. Silencing IFP35 also decreased the frequency of neuronal death in the peri-infarct cortex and resulted in smaller infarct sizes after ischemia-reperfusion in MCAO mice. Ultimately, the MCAO mice treated with IFP35 siRNA demonstrated better neurological outcomes. In term of mechanism, IFP35 triggered the activation of the NLRP3 inflammasome, exacerbating neuroinflammation by TLR4/NF-κΒ signaling. IFP35-stimulated microglial cells (BV2 cells) may serve as a potential source of pro-inflammatory cytokines secreted in vitro.

Materials and methods

Serum samples and ethics statement

Peripheral blood was collected from individuals with acute ischemic stroke (n = 31; 16 females) attending the Department of Emergency Medicine of the Affiliated Hospital of Xuzhou Medical University between 2024.01 and 2024.07. Acute ischemic stroke patients met the reported diagnostic criteria by William in 2019 [34]. Serum samples from healthy volunteers (n = 21; 10 females) were included as controls. The blood samples were processed within 2 h of collection and the sera were stored at − 80 °C until use. All blood samples from patients and healthy volunteers were obtained within 4–6 h of hospital admission, prior to or without any drug treatment. All patients were informed and signed written informed consent. This study was reviewed and approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (Approved #: XYFY2024-KL297-01).

Cell culture

Murine microglial BV2 cell lines were obtained from Chinese Academy of Sciences (Shanghai, China). BV2 cells were cultured in DMEM medium supplemented with 10% FBS. All cells were incubated at 37 °C in a 5% CO₂ humidified chamber. In some experiments, cells were then placed into a hypoxic incubator (Forma 3110, Thermo Fisher Scientific) with 1% O₂, 5% CO₂, and 94% N₂ for suitable time at 37 °C to mimic OGD injury with or without recombinant IFP35 protein (5 µg/ml, P2-58698PEP, Novus). Cultures were then restored with glucose at DMEM and recovered at normoxic conditions (37 °C, 5% CO₂) for 12 h (OGD restoration).

ELISA

The peripheral blood was extracted from patients and mice and placed at 4˚C. Brain tissues were homogenized in RIPA buffer and the protein concentration was determined by BCA kit (Beyotime, Shanghai). Human and mouse IFP35 ELISA kit (Bioswamp, Wuhan) was used according to the instructions to detect the concentration of IFP35 in the peripheral blood plasma and the homogenate of brain tissues. In some experiments. TNF-α, IL-1β related cytokines were measured by ELISA kits (BioLegend, San Diego, CA) according to the manufacturers’ instructions.

Middle cerebral artery occlusion (MCAO) procedure

8–12 weeks aged, male C57BL/6J mice were purchased from experimental animal center of Xuzhou Medical University to develop a model of acute ischemic stroke. The mice were housed in groups of four or five per cage in a temperature and humidity-controlled animal facility with a 12 h light– dark cycle. Food and water were available ad libitum. All procedures using laboratory animals were approved by and conducted consistently with the guidelines of the Animal Care and Use Committees (202203A549) of Xuzhou Medical University. For experiments, the mice underwent focal cerebral ischemia with silicone rubber-coated 6–0 nylon monofilaments (602256PK5Re; Doccol, Sharon, MA, USA) as previously described [35]. Briefly, mice were anesthetized with 1% pentobarbital sodium (100 mg/kg. i.p.). A carotid incision was made at the left common carotid artery (CCA), the internal carotid artery (ICA) and external carotid artery (ECA) were separated. Once ligation of the ECA and CCA was achieved, the silicone tipped filament was inserted and passed through the ICA. Advancement of the filament was abated when resistance is encountered at 8–9 mm and the filament was secured. After 60 min of occlusion, the filament was withdrawn, allowing reperfusion. During the procedure, the body temperature was maintained at 37 °C. Of the 283 mice used in this study, 40 mice mice died or insufficient after reperfusion and excluded from the experiment. After preliminary statistical analysis, the mortality of MCAO group was 11.3% (32 in 283). The success rate of the MCAO model was 85.9% (243/283).

Immunofluorescence assay

Immunofluorescence staining experiments, primary antibodies were performed on 10% formalin fixed brain tissue sections. The sections were sealed with BSA and incubated overnight at 4 °C with the respective primary antibody, including anti-IFP35 (1:100, sc-393513, Santa Cruz), anti-Iba1 antibody (1:100, 17198, CST), anti-GFAP antibody (1:100, 80788, CST), anti-NeuN antibody (1:100, 24307s, CST). Then, the sections were incubated with a second antibody at room temperature for 1 h. After the second antibody was removed, the nuclei were counterstained with DAPI (100 µl/well). Fluorescence images were captured by DM3000B (Leica, Germany).

Western blot assay

Ischemic cerebral hemispheres and cultured cells were extracted total proteins. protein concentration was determined by BCA kit (Beyotime, China). Cells or tissues were lysed in cold lysis buffer containing 20 mM Tris–HCl pH7.4, 150 mM NaCl, 1 mM EDTA, and 0.5% NP-40. Debris was depleted by centrifugation at 6000 g for 5 min. Equal amount of protein was loaded to 8-12% SDS-PAGE gel for each sample, separated by electrophoresis and transferred to PVDF membrane. After blocking with 5% non-fatted milk at RT for 1 h, the membrane was probed the corresponding primary antibody at 4℃ overnight, then incubated with the corresponding secondary antibody at RT for 1 h after washing with TBST. The PVDF membrane was washed with TBST and observed with Chemiluminescence Gel Imaging (Protein Simple, USA). Image J software was used to calculate the gray value. Antibodies (Abs) were purchased from Abcam, CST or Santa Cruz. Primary Abs (1:000 or 1:2000 dilution) used were: anti-IFP35 Abs (sc-393513), anti-NF-κB p65 Abs (ab16502), anti-Phospho-NF-κB p65 (3033), anti-TLR4 Abs (sc-293072), TLR4 polyclonal Abs (19811-1-AP), anti-mouse IgG (A0168), anti-rabbit IgG (A1949), anti-NLRP3 Abs (ab263899), anti-ASC Abs (67824), anti-Caspase 1 Abs (sc-398715), anti-IL-1β (B122) Abs (sc-12742), anti-IL-6 Abs (12912), anti-iNOS Abs (ab283655), anti-TNF-α Abs (11948). The secondary Abs (1:10000 dilution) used were HRP-anti-Rabbit Abs, HRP-anti-Mouse Abs, and HRP-anti-Goat Abs.

Administration of IFP35 SiRNA

With the anterior bregma point as the center, the cranial projection site of the lateral ventricle was found according to the position 0.5 mm back and 1.0 mm side open, and the skull was carefully drilled at the mark point with a micro cranial drill (to avoid damage to the brain parenchyma of the mouse due to rough manipulation). Then take a micro syringe, align the tip of the needle with the opening, and search down 2.0–2.5 mm, that is, the lateral ventricle. Pull back the micro syringe. If the liquid level rises or small bubbles appear, the positioning is accurate. A single-channel micro-injection pump was used to inject the siRNA mixture (4 µl (20µM) IFP35 siRNA, and 1 µl Lipofectamine 2000) into the lateral ventricle at a constant injection rate of 0.5 µl/min for 8 min. After injection, the needle was stopped for 5 min, and then the needle was slowly removed, the surgical incision was sutured, and iodophor disinfection was performed. IFP35 siRNA was injected continuously for three days. IFP35 siRNA sequence:5’-AUGAUGAGCUGGUGUCCAUTT-3’.

TAK-242 treatment

Mice were treated with TAK-242 at an optimal dose as previously described [36]. Briefly, TAK-242 was dissolved in DMSO and then diluted in sterile endotoxin-free water. The final concentration of DMSO was 1%. The dissolved TAK-242 or DMSO (1%) was injected intraperitoneally (3 mg/kg body weight) 1 h after middle cerebral artery occlusion or sham surgical operation.

TTC staining

For TTC staining, brain tissues were cut into five sections in the coronal plane (2 mm), and then dipped in 2% TTC (Solarbio, China) for 30 min at 37 ℃. The infarcted volume was measured and presented as a percentage of the non-ischemic hemisphere to correct for edema. The healthy area pixels of contralateral (VC) and ipsilateral hemisphere (VL) were calculated by using an image analysis software. The relative infarct percentage (%) was obtained using the following formula: % = 100 × (VC– VL)/VC.

Immunoprecipitation

Take 1000 µg of protein sample and place it in a 1.5 ml EP tube. Add Immunoprecipitation buffer (IP buffer) to bring the total system volume to 600 µl. Following the dilution ratio specified in the antibody instructions, add the primary antibody. Place the tube on a rotating mixer at 4 °C, gently shake overnight to ensure thorough binding between the antibody and target protein. On the next day, add 30 µl Protein A/G PLUS-Agarose to the sample, and rotate on a mixer at 4 °C for 2 h. After 2 h, remove the sample and centrifuge (4 °C, 5000 g×3 min). Discard the supernatant. Wash the beads in the sample with IP buffer. Wash each time with 600 µl IP buffer, then centrifuge to remove the wash solution. Repeat this washing process three times (perform all operations gently to prevent bead loss). After washing, carefully remove the supernatant, avoiding sediment inhalation. Add 16 µl IP buffer and 4 µl loading buffer (5×), then heat at 100 °C for 5 min. Centrifuge again and take the supernatant for protein content detection.

Neurobehavioral assessment

Twelve mice in each group were randomly selected for neurological function testing. Mice were placed in the test chamber for 1 h prior to testing to allow acclimatization to the environment. Neurological deficits were tested using the Zea Longa (0–4) score [37] method for scoring neurological function in animals, in which higher scores indicate more severe neurological deficits.

Behavioral tests

Cylinder test

Mice were placed in a transparent cylinder (10 cm in diameter and 15 cm in height), and record the use of forelimbs in the exploratory behavior within 5 min. Place a mirror at the appropriate position to ensure that the forelimb activity can be recorded even when the mouse was turned back to the camera lens. The evaluator used a video recorder with slow-motion and clear freeze-frame functions to record and score. When analyzing the behavior in the cylinder, record the number of times that the right paw, the left paw, and both paws touch the cylinder wall simultaneously. Data were analyzed using asymmetry, which was calculated as follows: (left + 0.5 × both)/(right + left + both) × 100%.

Corner test

The Corner test was performed as previously described [38]. Mice were placed between two wooden boards measuring 30 cm×20 cm×1 cm, forming a 30°. mice approached the corner, its whiskers would first touch the boards, causing the mouse to move forward and upward with hind legs standing. Each animal underwent 10 tests, and the score was calculated using the following formula: # of turns to the healthy side / Total # of turns ×100%.

Beam balance tests

normal = 0, maximum = 6; balances with steady posture: 0 point; slides off beam: 1 point; hugs beam and 1 limb falls down from beam: 2 points; hugs beam and 2 limbs fall down from beam, or spins on beam (> 60s): 3 points; attempts to balance on beam but falls off (> 40s): 4 points; attempts to balance on beam but falls off (> 20s): 5 points; falls off on attempt to balance or hang on to beam.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8 Software (La Jolla, CA). Data were presented as means ± SEM and statistically analyzed with two-tailed independent Student’s t-test (two groups). For multiple groups, data were analyzed with one-way ANOVA or two-way ANOVA. The level of statistical significance was set at P < 0.05.

Results

Increased circulating and cerebral level of released IFP35 in patients who had an acute ischemic stroke and mice subjected to focal ischemia

The danger signal theory introduced that tissue injury generated DAMPs govern the initiation of the inflammatory response [11]. Thus, whether IFP35, as a novel DAMP, elevated in acute ischemic stroke patients and mice is a crucial factor in understanding the initiation of inflammatory responses. To address it, we first assessed the level of human serum IFP35 from patients who had an acute ischemic stroke for 4–6 h without medical intervention by ELISA; the human serum from healthy volunteers who had non-neurological diseases were applied as controls. To our surprise, a significantly increasing level of human serum IFP35 in the ischemic stroke group versus those of control was observed in Fig. 1A. Subsequently, using a mouse model induced by 1 h MCAO and reperfusion, the dynamics of mice IFP35 at different time points after MCAO are illustrated in Fig. 1B&C. There was a statistically increment of mice IFP35 in brain homogenates in the peri-infarct areas at 6 h persisting until 72 h after MCAO (Fig. 1B). The serum IFP35 results were consistent with those observed in brain tissue. Specifically, IFP35 levels significantly increased at 3 h after ischemia-reperfusion and remained elevated for up to one week (Fig. 1C). Under oxygen-glucose deprivation (OGD) conditions, the level of IFP35 in the supernatant of mice primary neurons was also significantly increased (Fig. S1A). Taken together, these data indicated that circulating and brain level of released IFP35 escalated rapidly in both acute ischemic stroke patients and MCAO mice, sustaining for an extended duration.

Fig. 1.

IFP35 is enhanced in acute ischemic stroke patients and MCAO mice. (A) Detection of IFP35 in patient sera. Serum samples from patients with stroke (n = 31) and from healthy human subjects (n = 21) were individually assayed for the presence of IFP35 by ELISA, (B&C) Detection of serum IFP35 and brain homogenates IFP35 in the peri-infarct areas in MCAO mice (n = 3). Samples were individually assayed for the presence of IFP35 by ELISA. The results are expressed as IFP35 concentrations (ng/ml for patients; pg/ml for mice). One-way ANOVA (Dunnett’s multiple comparison test) was used. Data are the means ± SEM, *P < 0.05 versus Sham; ****P < 0.0001 versus Healthy control

Differential expression and release of IFP35 by brain cell types after MCAO

Given that IFP35 can be released by activated macrophages [24], and DAMPs are also frequently released by damaged cerebral neurons during acute cerebral ischemia [39, 40], our next step is to investigate whether the increased release of IFP35 during acute cerebral ischemia originates from neurons, microglia, or other brain cell types. Initially, we analyzed the brain sections in the cortical areas after MCAO by immunofluorescence assay. There was the dynamics of IFP35 at different time points after MCAO in Fig. 2A&B. A notable surge in IFP35 levels within the cortical region of the ischemic at 3 h, reaching its peak at 24 h, and persisting for at least 1 week. These findings aligned with our ELISA results. Further immune-stained with antibodies (Abs) against NeuN to detect neuron cells showed in Fig. 2A&C, the majority of IFP35⁺ cells co-stained with NeuN⁺ cells across various time points and spatial locations after MCAO. Fig.S1A also indicated that the expression IFP35 from cerebral neurons could be released upon ischemic-like conditions. Interestingly, at the early phase of stroke (especially at 3 and 6 h), our results seemed that IFP35 also exhibited vasculature-like staining patterns (Fig. 2A). To investigate whether IFP35 is expressed in vascular cells, we further performed co-immunofluorescence staining of IFP35 with CD31, an endothelial cell marker. The results showed co-localization of IFP35 with CD31-positive endothelial cells at 3–6 h, suggesting that IFP35 is also expressed in vascular structures at the early phase of stroke. However, the intensity of IFP35 expression in endothelial cells was relatively low compared to that in neurons. (Fig. S2). Besides, to determine whether IFP35 protein was also upregulated in other brain cell types, such as microglia (Iba1⁺) and astrocytes (GFAP⁺), following MCAO, brain tissues were additionally fixed and immunostained with antibodies against NeuN, Iba1, and GFAP to identify neurons, microglia, and astrocytes. In Fig. 2D, it’s evident that IFP35 predominantly expressed in neuronal cells, displaying markedly lower levels in the other two cell types. Finally, our in vitro experiments using BV2 microglial cells under OGD conditions showed that the cell culture supernatant of IFP35 level was indeed upregulated in activated microglia under OGD conditions. However, the level of IFP35 induction in microglia was significantly lower compared to that observed in neurons under OGD conditions (Fig. S1B). The outcomes indicated that, following MCAO, although endothelial cells (at the early phase of stroke), microglia, and astrocytes are capable of expressing IFP35, cerebral neural cells seem to express and release more IFP35 compare to other cell types.

Fig. 2.

Differential expression and release of IFP35 by brain cell types after MCAO. The mice were subjected to ischemia for 60 min, followed by various time points of reperfusion. (A) Immunofluorescence for detection of IFP35 in different time points of reperfusion by specific Abs. (B&C) The IFP35 positive cells density was expressed as the number of cells per 1 mm length of cortical cells, counted under a fluorescence microscope. (D) Immunofluorescence for detection of IFP35 co-stained with 3 cell type markers in 24 h of reperfusion by various specific Abs against NeuN, IBa1, and GFAP. One-way ANOVA (Dunnett’s multiple comparison test) was used. Data are the means ± SEM (n = 3), ****P < 0.0001 versus Sham

The repression of neuroinflammation by downregulation of IFP35 in MCAO mice

To investigate whether IFP35 proteins are involved in neuroinflammation during acute ischemic stroke, we knocked down IFP35 by siRNA before MCAO to study its effect on pro-inflammatory-related cytokinse release. 20µM IFP35 siRNA mixture (Fig. 3A) was administrated into mice lateral ventricle by a single-channel micro-injection pump at a speed of 0.5 µl/min for 3 consecutive days before MCAO. Subsequently, the levels of pro-inflammatory related-cytokines were measured in the peri-infarct areas tissue and peripheral blood to evaluate neuroinflammation following MCAO. The findings indicated a significant elevation in IL-1β and TNF-α levels within both the peri-infarct cortex and peripheral blood in the MCAO group compared to those of the Sham group (Fig. 3B&C). Intriguingly, a sharp reduction in both peri-infarct cortex and peripheral blood IL-1β and TNF-α levels was observed in the siRNA IFP35 group in contrast to the MCAO group (Fig. 3B&C). However, there was no statistical difference between the MCAO group and the Vehicle group (P > 0.05) (Fig. 3B&C). Additionally, we further verified the levels of pro-inflammatory factors iNOS, IL-6, and TNF-α by immunoblotting. The results revealed significantly higher levels of these pro-inflammatory factors in the MCAO group compared to the Sham group, demonstrating statistical differences (P < 0.05). However, subsequent to down-regulate IFP35 expression, the levels of iNOS, IL-6, and TNF-α, showed a significant decrease compared to the MCAO group, exhibiting statistical significance (P < 0.05) (Fig. 3D-G). Thus, these results suggested that the reduction in IFP35 repressed brain inflammation and production of pro-inflammatory cytokines after ischemia.

Fig. 3.

Downregulation of IFP35 by siRNA represses brain neuroinflammation in MCAO mice. The mice were injected i.c.v. with IFP35 siRNA into the right lateral ventricle for 3 consecutive days before ischemia-reperfusion, then, blood samples and peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h of reperfusion. (A) Immunoblotting for detection of IFP35 silence efficiency by specific IFP35 siRNA. (B & C) IL-1β and TNF-α ELISA Kit for measurement of the levels of IL-1β and TNF-α in both the peri-infarct cortex and peripheral blood in MCAO mice. (D) Immunoblotting for detection of other pro-inflammatory related cytokines such as iNOS, IL-6, and TNF-α. (E-G) Bands were scanned, and the intensities are represented as the fold changes versus sham or IFP35 siRNA treatment. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM (n = 3), ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

The alleviation of neuronal cell death by downregulation of IFP35 in MCAO mice

In response to cellular damage or death, released danger signal, including DAMPs, activating brain-resident immune cells such as microglia, amplifies inflammatory responses, ultimately leading to neuronal cell death [7]. To answer whether down-regulation of IFP35 could mitigate cortical neuronal damage before MCAO. TUNEL staining conducted on the peri-infarct cortex 24 h post ischemia-reperfusion revealed a notably increased the frequency of neuronal death in the MCAO group and the Vehicle group compared to the Sham group (Fig. 4A). However, the administration of IFP35 siRNA notably decreased the rate of neuronal death (Fig. 4B), indicating that decreasing IFP35 levels led to an improvement in neuronal death after MCAO, potentially achieved through the suppression of brain inflammation.

Fig. 4.

Downregulation of IFP35 by siRNA alleviates neuronal cell death in MCAO mice. The mice were injected i.c.v. with IFP35 siRNA into the right lateral ventricle for 3 consecutive days before ischemia-reperfusion, then peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h reperfusion. (A) TUNEL assay was performed to detect the apoptotic cells. (B) The apoptosis index is calculated by dividing the number of TUNEL-positive cells by the total number of cells per-field of view. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM (n = 5), ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

Targeted downregulation of IFP35 reduces acute ischemic brain injury

After downregulation of IFP35 by siRNA, we also tested the neurological function of MCAO mice. The MCAO and the Vehicle mice had more severe neurological deficits than the Sham mice, and the administration of IFP35 siRNA mice were less impaired in the Neurological Deficient Score (Fig. 5A). Additionally, neurological outcomes were significantly improved in the IFP35 siRNA group of mice than that in the MCAO and the Vehicle groups of mice, confirming by corner-turning test that evaluates mice comprehensive sensory motor function (Fig. 5B). Performance in beam balance test was statistically indistinguishable between the MCAO and the IFP35 siRNA groups of mice (Fig. 5C). Mice treated with IFP35 siRNA exhibited a significant decrease in the frequency of left forelimb usage in contrast to mice from the MCAO and Vehicle groups by cylinder test (Fig. 5D).

Fig. 5.

Downregulation of IFP35 by siRNA mitigates acute ischemic brain injury in MCAO mice. The mice were injected i.c.v. with IFP35 siRNA into the right lateral ventricle for 3 consecutive days before ischemia-reperfusion then, peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h reperfusion. (A) Neurological function as shown by neurological deficit score; The score ranged from 0 (without observable neurological deficit) to 4 (no spontaneous motor activity and loss of consciousness) (n = 12). (B-D) Cylinder test、Beam balance test and Corner test was performed to evaluate neurobehavioral of mice at day1, day3, day7 (n = 12). (E) TTC assay was conducted to assess the cerebral infarct volume. (F) Cerebral infarct volume rate was calculated. Infarct volume (%) = infarct volume/contralateral hemisphere volume × 100% (n = 6). Data are the means ± SEM. One-way ANOVA (Tukey’s multiple comparison test) and two-way ANOVA (Bonferroni’s multiple comparison test) was used. ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

Due to infarct volume is a major determinant of brain inflammation and stroke outcome [41], and variations in immune responses to immune modulation might occur in mice with differing infarct volumes. Indeed, analysis of 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 5E&F) aligned with the behavioral results. After reperfusion, both the MCAO and the Vehicle-treated mice exhibited significantly larger infarcts compared to the Sham mice. Conversely, mice treated with IFP35 siRNA showed smaller infarcts than that of the MCAO mice. The deteriorative effect of IFP35 on brain injury might, to some extent, endure due to its influence on inflammatory responses.

Released IFP35 mediated neuroinflammation by activating nuclear factor kappa B through Toll-Like receptor 4 during acute ischemic stroke

Considering that TLR4 signaling is involved in augmenting inflammatory responses mediated by IFP35 during septic shock [24]and acute liver injury [42], we hypothesize that IFP35 also triggers neuroinflammation via TLR4 pathway in acute ischemic stroke. To verify this hypothesis, a known TLR4 inhibitor, TAK-242 was intraperitoneally administrated to evaluate the interaction between IFP35 and TLR4 during a focal cerebral ischemia model. The co-immunoprecipitation results showed that the cerebral ischemia-reperfusion increased the interaction between IFP35 and TLR4 compared to the Sham group, and treatment with the TLR4-specific inhibitor TAK-242 weakened the interaction between IFP35 and TLR4 (P < 0.05) (Fig. 6A&B), indicating that IFP35 might interact with TLR4 to mediated neuroinflammation during acute ischemic stroke.

Fig. 6.

Released IFP35 mediated neuroinflammation through TLR4 during acute ischemic stroke. The mice were injected intraperitoneally with TAK-242 (3 mg/kg) or DMSO (1%) 1 h before ischemia-reperfusion, then, peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h of reperfusion. (A) IFP35 was pulled down using an anti-IFP35 antibody, followed by immunoblotting to detect the level of TLR4 expression. (B) TLR4 was pulled down using an anti-TLR4 antibody, followed by immunoblotting to detect the level of IFP35 expression. Bands were scanned, and the intensities are represented as the fold changes versus sham or TAK-242 treatment. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM (n = 3), ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

Given that the capability of TLR4 signaling pathway to activate downstream mediators, such as the transcription factor nuclear factor (NF-κB), which in turn increases the production of pro-inflammatory agents like cytokines and chemokines [43], we next examined the effects of IFP35 on NF-κB activation. We found that the subunits of NF-κB p65, was decreased in the cytoplasm and increased in the nuclear in MCAO mice (Fig. 7A-C). However, after administration of IFP35 siRNA into mice lateral ventricle for 3 consecutive days before MCAO, the levels of NF-κB p65 nuclear protein (Fig. 7A-C) and phosphorylation were reduced compared to the MCAO and Vehicle groups (Fig. 7D&E). The immunofluorescent staining also revealed the translocation of p65 from the cytoplasm to the nucleus (Fig. 7F). Overall, these findings suggest that the released IFP35 protein induces NF-κB-dependent neuroinflammation via the TLR4 signaling pathway during acute ischemic stroke.

Fig. 7.

Released IFP35 mediates neuroinflammation by activating NF-κB signaling in MCAO mice. The mice were injected i.c.v. with IFP35 siRNA into the right lateral ventricle for 3 consecutive days before ischemia-reperfusion then, peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h of reperfusion. (A) Immunoblotting for detection of NF-κB (p65) in the cytoplasm and NF-κB (p65) protein in nucleus. β-actin in the cytoplasm and Histone 3 in the nucleus were used as control. (B & C) Bands were scanned, and the intensities are represented as the fold changes versus sham or IFP35 siRNA treatment. (D) Immunoblotting for detection of phosphorylation of p65 in the whole cell lysates. (E) Bands were scanned, and the intensities are represented as the fold changes versus sham or IFP35 siRNA treatment. (F) Immunofluorescent staining for detection of the localization of NF-κB p65. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM (n = 3), ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

Released IFP35 induces activation of NLRP3 signaling to aggravates neuroinflammation during acute ischemic stroke

IFP35 serve as a novel pro-inflammatory DAMPs trigger innate immune response and exacerbate inflammation to promote multiple sclerosis [25]. Meanwhile, NLRP3 inflammasome can be activated by various stimuli, including PAMPs and DAMPs, which activate NF-κB to induce the transcription of NLRP3 and key pro-inflammatory cytokines such as pro-IL-1β [44]. Thus, to explore the downstream mechanisms of TLR4/NF-κB activation signaling pathway, NLRP3 activation could be considered a potential initiating event in the inflammatory responses during acute ischemic stroke. Consistent with our hypothesis, the activation of NLRP3 signaling was also observed in a MCAO mouse model. Actually, we found significantly increased levels of NLRP3, ASC, caspase-1 protein, pro-IL-1β as well as IL-1β, which are downstream key inflammatory factor of inflammasome in MCAO mice at 24 h after ischemia-reperfusion (Fig. 8). However, pre-treated with IFP35 siRNA before MCAO in mice, the levels of NLRP3, ASC, caspase-1 protein, pro-IL-1β and IL-1β were statistically decreased (Fig. 8). Concurrently, we measured the levels of pro-caspase-1 protein and found no significant differences between Sham and MCAO mice. Moreover, there were no notable changes between MCAO mice pretreated with IFP35 siRNA and those not pretreated (Fig. 8). Taken together, these above results indicated that as a novel DAMP, released IFP35 could trigger innate immune response and exacerbate inflammation responses via TLR-4/NF-κB/NLRP3 signaling during acute ischemic stroke.

Fig. 8.

Released IFP35 aggravates neuroinflammation by activating NLRP3 signaling in MCAO mice. The mice were injected i.c.v. with IFP35 siRNA into the right lateral ventricle for 3 consecutive days before ischemia-reperfusion then, peri-infarct areas tissues were collected from sham mice or from mice subjected to ischemia for 60 min, followed by 24 h of reperfusion. (A) Immunoblotting for detection of NLRP3, ASC, pro-caspase-1, caspase-1, pro-IL-1β as well as IL-1β respectively (B-H) Bands were scanned, and the intensities are represented as the fold changes versus sham or IFP35 siRNA treatment. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM (n = 3), ^#P < 0.05 versus MCAO; ^*P < 0.05 versus sham

Exogenous IFP35 regulates TLR-4/NF-κB/NLRP3 signaling pathway to exacerbates neuroinflammation in BV2 microglial cells

Microglia, the predominant immune cells within the brain parenchyma, have been increasingly implicated in neuroinflammation during acute ischemic stroke [45]. IFP35 serving as a novel proinflammatory DAMPs, could activate macrophage to promote inflammatory responses. Therefore, we tested if extracellular IFP35 protein regulates TLR-4/NF-κB/NLRP3 inflammasome signaling pathway to exacerbate BV2 microglial neuroinflammation in OGD condition. Consistent with animal experimental results, extracellular IFP35 protein (5 µg/ml) were pretreated with BV2 microglial cells for 3 h in OGD condition, followed by a return to normal conditions for another 12 h. 1 h later, although we found that the subunits of NF-κB p65, was slight decreased in the cytoplasm and increased in the nuclear without extracellular IFP35 protein treatment, the levels of NF-κB p65 nuclear protein were greatly increased upon extracellular IFP35 protein treatment (Fig. 9A-C). 6 h later, we also found that significantly increased levels of NLRP3, ASC, caspase-1 protein upon extracellular IFP35 protein treatment, however, extracellular IFP35 protein treatment was not able to induce significant changes of the level of pro-caspase-1 protein (Fig. 9D-H). 24 h later, IL-1β was significantly increased upon extracellular IFP35 protein treatment (Fig. 9J). Similarly, the level of pro-IL-1β had no notable changes upon extracellular IFP35 protein treatment (Fig. 9I). These results suggested that microglial cells might serve as the potential target cell type through which released IFP35 protein modulates the TLR-4/NF-κB/NLRP3 inflammatory signaling pathway to exacerbate neuroinflammation.

Fig. 9.

Exogenous IFP35 exacerbates neuroinflammation by regulating TLR-4/NF-κB/NLRP3 signaling in BV2 microglial cells. The BV2 cells with or without rIFP35 protein (5 µg/ml) were exposed to OGD conditions for 3 h, followed by a return to normal conditions for another 12 h. (A) 1 h later, immunoblotting for detection of NF-κB (p65) in the cytoplasm and NF-κB (p65) protein in nucleus. β-actin in the cytoplasm and Histone 3 in the nucleus were used as control. (B & C) Bands were scanned, and the intensities are represented as the fold changes versus sham or rIFP35 treatment. (D) 6 h later, Immunoblotting for NLRP3, ASC, pro-caspase-1, caspase-1, pro-IL-1β as well as IL-1β were detected respectively. (E-J) Bands were scanned, and the intensities are represented as the fold changes versus sham or rIFP35 siRNA treatment. One-way ANOVA (Tukey’s multiple comparison test) was used. Data are the means ± SEM from three independent experiments (n = 3). ^*P < 0.05 versus control

Discussion

Considering ischemic stroke, a leading cause of death and disability, particularly in China induces neurofunctional impairment imposes a substantial economic burden on society and families [46], there is an urgent need to explore the underlying mechanisms to elucidate the causes behind stroke occurrence. The innate immune responses have consistently been linked to the pathophysiology of stroke, and their activation is correlated with adverse outcomes [1]. In this study, we identified the contribution of a virus-related inflammatory protein, IFP35, as a novel DAMP in regulating neuroinflammatory events during acute ischemic stroke.

DAMPs are typically described as endogenous signaling molecules that are released into the extracellular space in response to non-infectious stimuli, where they can trigger and perpetuate an immune response [47]. IFP35 serve as a novel pro-inflammatory DAMP trigger innate immune response and exacerbate inflammation to promote acute and chronic inflammatory diseases. For example, in acetaminophen-induced acute liver injury, the danger signal IFP35, released from necrotic hepatocytes, amplifies live injury by upregulating inflammatory factors and chemokines [42]. Multiple sclerosis is a chronic inflammatory disease, released IFP35 as a biomolecular marker in multiple sclerosis [48], is able to promote the development of multiple sclerosiss [25]. Therefore, we hypothesize that IFP35 may function as a universal pro-inflammatory DAMP triggered by injury tissues, potentially playing a role in the pathophysiology of acute ischemic stroke. Firstly, we evaluated the levels of IFP35 in acute ischemic stroke patients and MCAO mice model. The results showed that serum IFP35 increased in individuals with an acute ischemic stroke for 4–6 h and a statistically increment of mice IFP35 in the peri-infarct areas at 6 h and mice serum IFP35 at 3 h persisting until 1 week after MCAO (Fig. 1). Moreover, we performed OGD stimulation on primary neurons and found that the level of IFP35 in the supernatant was significantly elevated (Fig. S1A). These results indicate that there is an excessive released IFP35 in acute ischemic stroke patients and mice model.

In our study, we focused on the ipsilesional hemisphere because it is the primary site of ischemic injury and neuroinflammation in the MCAO model, and where IFP35 is most likely to be involved in pathological processes. Notably, in similar MCAO studies, the contralesional hemisphere is sometimes used as an internal control. However, increasing evidence suggests that the contralesional hemisphere may also be affected by ischemic injury, including secondary inflammation [49] and altered gene expression [50], which could confound interpretation of results. Therefore, we chose to focus solely on the ipsilesional hemisphere in our analysis. A longitudinal time-course analysis could offer a more comprehensive understanding of IFP35 dynamics over time. However, we selected the 4–6 h post-stroke time window based on clinical and ethical considerations, as well as biological relevance. This period corresponds to the therapeutic window in stroke patients, during which interventions are typically initiated—making later measurements potentially confounded by treatment effects. Additionally, previous reports [3–5] indicated that early inflammatory responses, including the activation of innate immune pathways, typically occur within this window following ischemic injury. Thirdly, from an ethical standpoint, performing a longitudinal time-course analysis solely for research purposes could interfere with timely clinical treatment, which limits its feasibility in patient settings.

As we know, the MCAO model is a classical and widely used experimental approach for studying ischemic stroke. Multiple methods are available to confirm the successful establishment of the MCAO model, including real-time cerebral blood flow monitoring using laser Doppler flowmetry (LDF), neurological deficit scoring, TTC staining, and histological or behavioral assessments. Although LDF is indeed a classic and well-established technique for verifying the success of MCAO by providing real-time monitoring of cerebral blood flow, in our study, we did not employ LDF due to equipment limitations. However, we used a set of well-validated and widely accepted alternative indicators to assess the success of MCAO, including contralateral forelimb flexion, spontaneous circling, and reduced activity (Supplementary recorded video 1–3). In addition to these behavioral indicators, we also performed TTC staining at 24 h post-occlusion to confirm infarct presence. Animals that did not show consistent infarct formation in the MCA territory were excluded from further analysis to ensure the reliability of our model.

Previous studies have shown that DAMP is either passively released from necrotic cells or actively secreted from activated immune cells [51]. In damaged tissue, IFP35 are released to serum by active macrophages [24] or damaged cells [42]. Therefore, we next investigated the primary source of increased IFP35 in MACO mouse model. Interestingly, we found that differential expression and release of IFP35 by brain cell types after MCAO, such as endothelial cells (at the early phase of stroke), microglia, and astrocytes, cerebral neural cells seem to express and release more IFP35 compare to other cell types (Fig. 2, Fig. S1 and Fig.S2). Interestingly, at the early phase of stroke, vascular endothelial cells seems express IFP35(Fig.S2). Vascular endothelial cells (ECs) are not merely passive barriers but active participants in innate immunity, responding to DAMPs released during tissue injury. These DAMPs can bind to PRRs and the RAGE on ECs, triggering inflammatory signaling pathways. This activation leads to the secretion of pro-inflammatory cytokines and chemokines, increased vascular permeability, and disruption of endothelial integrity, contributing to various pathological conditions including atherosclerosis, sepsis, and ischemia-reperfusion injury [52] For instance, HMGB1 released during ischemic events can interact with RAGE on ECs, activating the p38 MAPK pathway and leading to increased endothelial permeability and inflammation [53]. Similarly, in sepsis, DAMPs can induce oxidative stress in ECs, activating NF-κB and AP-1 transcription factors, which upregulate the expression of adhesion molecules and cytokines, exacerbating vascular inflammation [54]. As a novel DAMP, IFP35 expressed in vascular endothelial cells may contribute to neuroinflammatory processes through similar mechanisms. Additionally, the presence of IFP35 in vascular endothelial cells may help explain why elevated levels of IFP35 were detected earlier in the peripheral blood (at 3 h, Fig. 1C) than in the brain tissue (at 6–12 h, Figs. 1B and 2) in mice. This suggests that during acute stroke, the periphery may be mobilized earlier than the brain to release DAMPs like IFP35, which could then enter the brain through the vasculature and contribute to the neuroinflammatory response. These potential roles of vascular IFP35 has been briefly discussed in the discussion section of the updated version.

Notably, in acute ischemic stroke, the formation of emboli in the arteries that supply blood to the brain [55] to a specific brain region may result in the death of brain cells, such as neurons [56, 57]. The death of neurons triggered by ischemic stroke is able to promote the activation of brain resident immune cells such as microglia, astrocytes, and oligodendrocytes. This activation occurs due to the recognition of various DAMPs leaked from dead neurons, which act as ligands for different Toll-like receptors (TLRs) expressed on these resident immune cells [7]. Although our data showed that IFP35 exacerbated neuroinflammation, which indirectly contributed to neuronal damage, our in vitro experiments to also assess the direct cytotoxic effects of IFP35 on neurons. To assess whether IFP35 directly induces neuronal death, SH-SY5Y cells and primary neurons were exposed to OGD conditions for 48 h in the presence or absence of recombinant IFP35 protein (5 µg/ml). Cell viability was measured using the CCK-8 assay. The results showed no statistically significant difference between the IFP35-treated and untreated groups (Fig. S3), suggesting that IFP35 does not directly promote neuronal death under ischemic-like conditions.

Although the initial injury causes of acute ischemic stroke and intracerebral hemorrhage differ, the commonality lies in the damage to neurons and the release of DAMPs [58], which activate a unified the innate and adaptive immune response that eliciting inflammation within the brain, and possibly, extends to peripheral areas [7]. A range of DAMPs, including adenosine, heat shock proteins, high mobility group box 1, and interleukin-33, play a role in the pathogenesis of both acute ischemic stroke and intracerebral hemorrhage [7]. For instance, acute ischemic stroke leads to the release of DAMPs like HMGB1, which triggers macrophage activation, causing infiltration and activation of pro-inflammatory cells, along with upregulation of pro-inflammatory factors such as TNF-α, IL-6, and IL-1β [59]. IFP35, identified as a novel DAMP, was next examined for its role in the secretion of inflammatory mediators. Pre-administration of siRNA targeting IFP35 into the lateral ventricles of mice before MCAO significantly attenuated the levels of pro-inflammatory-related cytokines in both the peri-infarct tissue and peripheral blood (Fig. 3), suggesting the involvement of IFP35 in exacerbating neuroinflammation during acute ischemic stroke.

As we know, viral vectors such as lentivirus or AAV, CRISPR-Cas system or siRNA method are current popular with editing gene expression. Viral vector or CRISPR-Cas system are widely used for stable gene manipulation in the brain. However, siRNA-mediated knockdown with appropriate concentration remains a well-established and effective approach, particularly for short-term gene silencing in acute models like MCAO [60–62]. This method allows for timely intervention without the need for viral packaging or long expression periods, which is especially advantageous in acute stroke studies. Since our siRNA was not labeled with a fluorescent tag (e.g., FAM), we were unable to directly visualize transfection efficiency via fluorescence-based methods. However, the knockdown efficiency of IFP35 was consistently demonstrated by Western blot analysis in multiple independent experiments, including those shown in Figs. 3A and 8A. These results confirm the effectiveness of siRNA-mediated silencing of IFP35 in our study.

In the acute phase of brain injury, the focal inflammation exacerbates brain damage by amplifying excitotoxicity, direct cell lysis, oxidative stress, and thromboinflammation [63]. Neuronal cell death can significantly impact brain function, resulting in a diverse array of cognitive, motor, and sensory impairments. Thus, we evaluated the impact of IFP35 on neuronal cell death and brain function in our setting. Silencing IFP35 prior to MCAO also reduced the frequency of neuronal death on the peri-infarct cortex 24 h post ischemia-reperfusion (Fig. 4) and exhibited significantly smaller infarcts by TTC staining in MCAO mice (Fig. 5E&F). For TTC staining, we used the widely accepted 2 mm slice thickness [5, 6, 64]. following 60-minute MCAO, and consistent with previous studies, the resulting infarct sizes were comparable with reported studies [65–67], considering reperfusion injury and differences in collateral circulation. Additionally, neurological function outcomes were significantly improved in the IFP35 siRNA pre-treated MCAO mice 1, 3, 7 days’ post ischemia-reperfusion respectively, as demonstrated in corner-turning, beam balance and cylinder test (Fig. 5A-D).

The Toll-like receptor (TLR) is a type of PRR responsible for recognizing microbial components, and it plays a crucial role in initiating the immune response. TLR4, expressed in neurons, microglia, and astrocytes in the brain, senses lipopolysaccharide (LPS) and initiates the release of inflammatory cytokines by activating the NF-kB signaling pathway. Previous studies have demonstrated that HMGB-1 promoted cerebral ischemia/reperfusion injury via TLR4 signaling pathway and TLR4-deficient mice had lower infarct volumes and better outcomes in neurological and behavioral tests [68]. Recent studies showed that IFP35 family members are recently identified as DAMPs which exacerbate inflammatory responses, by binding to Toll-like receptor 4 (TLR4) and triggering NF-kB pathway [24, 25, 42]. These observations prompted us to hypothesize that IFP35 may modulate the immune response through TLR4/NF-kB-mediated inflammatory signaling following acute ischemic stroke. Therefore, we administered the TLR4 inhibitor TAK-242 intraperitoneally and used immunoprecipitation to validate whether IFP35 interacts with TLR4 in a focal cerebral ischemia model. The results (Fig. 6) indicated that treatment with the TLR4-specific inhibitor TAK-242 weakened the interaction between IFP35 and TLR4 (P < 0.05), suggesting that IFP35 might interact with TLR4 to mediated neuroinflammation in MCAO model mice.

TLR4 in response to diverse stimuli, initiates transcription through the NF-κB-dependent pathway to induce a variety of inflammatory genes expression [43]. NF-κB, a pivotal nuclear transcription factor, plays a crucial role as a transcriptional regulator in cells, particularly during inflammatory responses. Initially localized in the cytoplasm, NF-κB (p65) translocates into the nucleus upon agents sitimulation. This translocation regulates the release of diverse inflammatory chemokines and cytokines, thus serving as a critical process in inflammation development [69]. Although reporter assays or NF-κB reporter mice would provide direct functional evidence of NF-κB activation, these tools were not available to us during the course of this study. Instead, we assessed NF-κB activation by examining its nuclear translocation using immunofluorescence and/or nuclear protein extraction followed by Western blotting. Nuclear translocation of NF-κB is a well-established marker of its activation and has been widely used in previous studies as an indicator of NF-κB signaling activity [24, 25]. We believe this approach provides reliable evidence supporting NF-κB activation in our model. In this study, the results in Fig. 7 indicate that IFP35 siRNA pretreatment could significantly prevent NF-κB (p65) in the cytoplasm from translocating into the nucleus.

Recently, the inflammasome protein complex has emerged as a pivotal component in the innate immune response to ischemic stroke [27]. In the central nervous system (CNS), four main types of inflammasomes are recognized: NOD-like receptor pyrin domain-containing protein 1 (NLRP1), NLRP3, NLR family CARD domain-containing protein 4 (NLRC4), and absent in melanoma-2 (AIM2), with NLRP3 being the most thoroughly investigated in CNS disorders. Notably, NLRP3 inflammasome has been extensively studied in CNS diseases [30, 31]. NLRP3 inflammasome activation involves two main steps [44, 70]. First, it requires priming through the activation of the TLR4/NF-kB signaling pathway, which promotes the transcription of NLRP3 components. Subsequently, the NLRP3 protein forms a complex with apoptosis-associated speck-like protein containing a CARD (ASC), which then binds to the cysteine protease caspase-1 to assemble the inflammasome. This activation of caspase-1 leads to the cleavage of pro-IL-1β and pro-IL-18 into their mature forms (IL-1β and IL-18), initiating inflammatory responses. These findings prompted us to explore the potential involvement of the NLRP3 inflammasome in IFP35-mediated inflammation during acute ischemic stroke. Figure 8 showed that elevated levels of the NLRP3 inflammasome components including NLRP3, ASC, and caspase-1, along with the inflammatory cytokine IL-1β, were detected in MCAO mice 24 h following ischemia-reperfusion. Conversely, pre-administering IFP35 siRNA prior to MCAO in mice led to significantly lowered the levels of these inflammasome associated proteins, indicating that apart from the neuroinflammation driven by TLR-4/NF-κB signaling, the released IFP35 exacerbates inflammatory responses through the TLR-4/NF-κB/NLRP3 inflammasome during acute ischemic stroke.

Microglia, as the brain resident macrophages, are the primary mediators of the brain’s innate immune responses to injury in ischemic stroke [45]. Accumulating evidence indicates that M1 polarized microglia contribute to neuronal dysfunction, and cell death by releasing pro-inflammatory mediators such as cytokines, reactive oxygen species and MMPs [71, 72]. Despite the coexistence of both microglial phenotypes during ischemic stroke, the vulnerable brain environment typically favors M1 microglial polarization, resulting in the expansion of damage and neurological deficits [73]. Although emerging evidences suggest that activated microglia in vivo shows the transient M2 phenotype followed by a quick shift to the M1 phenotype in ischemic stroke [73, 74], the findings also indicate that both phenotypes of microglia are activated and polarized at early stage of cerebral ischemia [75]. Meanwhile, in vitro studies revealed that ischemic neurons or some agents prime the polarization of microglia toward M1 [73, 76]. Therefore, BV2 cells (microglia cell line) were employed in vitro to explore the role of IFP35 in microglia polarization. In line with the previous data, our results showed that pro-inflammatory cytokines such as TNF-α, IL-6 and IL-1β were statistically increased in exogenous IFP35 treatment group, however, pre-treatment of IFP35 siRNA before MCAO, these M1-like cytokines were greatly decreased. Mechanistically, IFP35 induced activation of NLRP3 inflammasome to aggravates neuroinflammation via binding its receptor TLR4, subsequently NF-κΒ activation in OGD-conditioned BV2 cells.

In conclusion, IFP35, an antiviral protein, was recognized as a novel DAMP implicated in modulating neuroinflammatory events during acute ischemic stroke. From the clinical findings to laboratory research, we found that an increased serum IFP35 in acute ischemic stroke patients and an elevated sera and peri-infarct areas IFP35 in MCAO mice. Pre-treatment of siRNA against IFP35 before alleviated neuroinflammatory cytokines secreted, reduced neuronal death, smaller infarction volumes, and thus, resulted in better neurological outcomes. In regards to mechanism IFP35 triggered the activation of NF-κΒ and NLRP3 inflammasome, exacerbating neuroinflammation by binding its receptor TLR4 in vitro and in vivo. However, the precise mechanism of IFP35 in regulating types of resident immune cells and signal pathway needs to be further addressed.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Mengmeng Zhang, Xiaowei Zhang: Data curation, writing-original and methodology; Bingnan Guo: Formal analysis, investigation and writing-original; Dong Han: Methodology and funding; Lanxin Lv: Methodology and funding; Xiaoqing Yan: Methodology and funding; Chenglei Su: Software; Dafei Chai: Methodology and funding; Ningjun Zhao: Conceptualization and funding; Xianliang Yan: Supervision and funding; Shuqun Hu: Writing-reviewing, editing and funding.

Funding

This research was supported by Science and Technology Development Fund from Affiliated Hospital of Xuzhou Medical University (XYFM2021006, Dong Han, XYFY2021011, Lanxin Lv, 2021ZA37, Xiaoqing Yan), Medical Science and Technology Innovation Project of Xuzhou Health Commission (XWKYHT20220144, Lanxin Lv), Natural Science Fund for Colleges and Universities in Jiangsu Province (22KJB320026, Xiaoqing Yan), Xuzhou Key Research and Development Plan (Social Development) Project—General Medical and Health Project (KC22232, Ningjun Zhao), Natural Science Foundation of Jiangsu Province (BK20231162, Xianliang Yan), Xuzhou National Clinical Key Specialty Cultivation Project (2018ZK004, Xianliang Yan), Key Project of Jiangsu Provincial Health Commission (K2023020, Shuqun Hu), Major Project of Biotherapy Research Special Project of Xuzhou Health Research Institute (XJZ2023002, Shuqun Hu) and National Natural Science Foundation of China (82072814 Dafei Chai).

Data availability

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

Declarations

Ethics approval and consent to participate

All procedures using laboratory animals were approved by and conducted consistently with the guidelines of the Animal Care and Use Committees (Approved #: 202203A549) of Xuzhou Medical University. All patients were informed and signed written informed consent and the study was reviewed and approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (Approved #: XYFY2024-KL297-01).

Competing interests

The authors declare no competing interests.