Formononetin Protects Against Cerebral Ischemia-Reperfusion-Induced Lung Injury: Insights from Animal and Cell Models

Abstract

Background

Cerebral ischemia-reperfusion (CIR) affects multiple organs, with lung injury being particularly prevalent. Formononetin, derived from legumes, has multiple pharmacological effects and demonstrated potential for the treatment of neurological diseases. However, its effects on CIR-induced lung injury are unknown.

Objective

To investigate the protective effect of formononetin in CIR-induced lung injury (in the Sprague-Dawley (SD) rats’ model).

Methods

Sprague-Dawley rats (n=27) underwent middle cerebral artery occlusion (MCAO) to establish a CIR model. Formononetin was then administered at specific time points. Longa neurological deficit score, histological staining, enzyme-linked immunosorbent assay (ELISA), and flow cytometry were subsequently used to assess lung injury. In addition, rat microglia were exposed to oxygen-glucose deprivation/re-oxygenation (OGD/R) to mimic CIR in vitro, and exosomes were then extracted from these cells for testing. A lipopolysaccharide (LPS)-induced lung epithelial cell injury model was subsequently used to investigate the protective effects of formononetin.

Results

ELISA revealed that formononetin exerted broad-spectrum anti-inflammatory effects, reducing TNF-α, IL-1β, IL-6, and IL-10 by 46.7%, 46.2%, 40.1%, and 54.6%, respectively, in bronchoalveolar lavage fluid after MCAO (P<0.05). In lung tissues, superoxide dismutase (SOD) expression was increased, while malondialdehyde (MDA) was decreased (22.8%). Additionally, formononetin reduced interferon-gamma positive (INF-γ⁺), IL-4⁺, IL-17A⁺, forkhead box P3 protein positive (Foxp3⁺), cluster of differentiation 4 positive (CD4⁺), and CD8⁺ by 72.3%, 76.7%, 66.1%, 67.7%, 78.4%, and 80.4%, respectively, all to sham levels (P < 0.05). The in vitro experiments also revealed a downward trend in the expression of inflammatory factors in lung epithelial cells, with increased SOD expression (26.2%) and decreased MDA levels (21.7%, P<0.05).

Conclusion

Formononetin protects against lung injury by inhibiting inflammatory cytokines, reducing oxidative stress, and regulating immune cells. Notably, exosomes derived from formononetin-treated microglia were shown to attenuate LPS-induced lung epithelial cell injury, highlighting a novel intercellular communication pathway. Together, these in vivo and in vitro studies offer new insights into the clinical application of formononetin in cerebrovascular diseases for secondary lung injury.

Introduction

Ischemic stroke is a devastating neurological disorder with high mortality and disability rates. By current estimates, the global annual incidence of ischemic stroke is 182 per 100,000 people,¹ although there has been an upward trend in recent years, and it is increasingly affecting younger people.^2,3 Ischemic stroke arises from transient or permanent cerebral ischemia, necessitating prompt restoration of blood flow after symptom onset to prevent further damage. In addition to the effects brought on by ischemia that result from the decrease and cessation of blood supply to tissues and organs; reperfusion, or the restoration of blood flow to ischemic tissues, worsens tissue damage and affects organs’ survival and function.⁴ Hence, reperfusion treatments such as thrombolytic therapy and endovascular thrombectomy are prevalent in clinical practice.⁵ In addition to posing a life-threatening risk, ischemic stroke can also cause peripheral organ damage, with lung injury being a prominent complication. Clinical data indicate that post-cerebral ischemia lung injury occurs in approximately 20–22% of cases,⁶ predominantly presenting as pulmonary edema and infection. This complication is particularly pronounced following cerebral ischemia-reperfusion (CIR). Because secondary lung injury prolongs treatment duration and diminishes patient survival rate and quality of life, its prevention and treatment must be seriously considered.⁷

In conjunction with CIR, secondary pulmonary complications constitute a “double blow to the body”, and the associated mortality rate is high.⁸ However, the specific mechanism underlying lung injury after CIR remains unclear, and there is presently no standard treatment scheme or process for lung injury after CIR in clinical settings. Recently, the concept of a cerebro-pulmonary axis has emerged, which describes interactions between the brain and lungs through neuroanatomical, endocrine, immune, metabolite microbial, and gaseous pathways.^9–11 According to current literature, the pathogenesis of CIR-induced lung injury is closely associated with the post-stroke inflammatory cascade.^7,12,13 Thus, the immune response to CIR is to safeguard the brain against further inflammatory stimuli, and the immunosuppression induced is subsequently responsible for stroke-induced immunodeficiency syndrome and infections.¹⁴

Formononetin, which is a bioactive isoflavone and phytoestrogen derived from various traditional Chinese herbs including Astragalus membranaceus, red clover, licorice, and Pueraria lobata, has shown promise in the treatment of neurological diseases such as ischemic stroke and traumatic brain injury.¹⁵ This natural phytoestrogen isoflavone possesses robust anti-inflammatory and antioxidant properties, leading to the inhibition of neuronal damage and ferroptosis.¹⁶ In previous research, formononetin was demonstrated to alleviate inflammatory responses via the JAK2/STAT3, NF-κB, PTEN/AKT/NF-κB, and PI3K/Akt signaling pathways.^17,18 In addition, inflammatory cytokines such as interleukins (ILs) and tumor necrosis factor alpha (TNF-α), and immune cells such as natural killer (NK) cells, are also affected by formononetin, further highlighting its potential for inducing anti-inflammatory effects.^19,20 Moreover, formononetin exhibits antioxidant properties, thereby reducing oxidative stress and its various interactions with the inflammatory response.²¹ In recent studies, the mechanism of lung injury after CIR has been associated with inflammatory reactions, oxidative stress, immune disorder, and other factors. Therefore, we have determined here to investigate whether formononetin has a protective effect on lung injury after CIR.

In previous research, CIR was demonstrated to cause lung tissue damage, increase the expression of inflammatory cytokines, and exacerbate oxidative stress responses.²² For reference, we note that lipopolysaccharides (LPS)-induced acute lung injury has also been shown to increase lung inflammation and oxidative stress levels in a similar manner.^23–25 While the pathways and mechanisms of these disorders may differ, the outcomes are relatively similar. To investigate the protective effects of formononetin on lung injury after CIR, an in vivo cerebral ischemia model was established in rats by middle cerebral artery occlusion (MCAO), and a formononetin treatment group was established. Several relevant indices were then measured after formononetin intervention to confirm the protective effect of formononetin on lung injury after CIR. For our in vitro studies, we used a microglial oxygen-glucose deprivation/re-oxygenation (OGD/R) model to simulate CIR, and LPS-induced cell damage in lung epithelial cells to mimic CIR-induced lung injury. By measuring inflammatory cytokine expression levels and antioxidant activity in lung epithelial cells after microglial exosome intervention, we provide evidence that formononetin mediates protection against CIR-induced lung injury via exosome cell-cell communication.

Methods

Animal Experimentation

Animal Grouping

Twenty-seven male SD rats (male, age, 8 weeks) of similar body weight (weight, ~200 g each, Hangzhou medical college experimental animal center, Hangzhou, China) were randomly selected and subjected to adaptive feeding conditions for one week. These conditions included a Specific Pathogen Free (SPF) environment under a 12-hour light/12-hour dark cycle with normal diet and water access. The 27 rats were then randomly divided into three groups: a sham-operated group (n=9); an MCAO group (n=9); and an MCAO with formononetin intervention (MCAO+FN) group (n=9). All animal experiments were first approved by our ethics committee (AL-JHYY202446), and all protocols strictly followed the relevant guidelines.

Animal Modeling and Drug Administration

An appropriate amount of formononetin (F141481-1g, Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China) was dissolved in approximately 40 mL of sterile saline to prepare a 24 mg/mL stock solution. After recording the weight of each rat, a volume of formononetin stock solution equivalent to 80 mg/kg was administered via intraperitoneal injection. An 80 mg/kg formononetin dose was chosen for the above experiments because inhibition of inflammatory response-related pathways was more significant at this dose when compared with lower doses.^26,27

The following methodologies for model establishment and experimental group allocation were implemented for this study. Sham-operated control group: the common carotid artery, external carotid artery, and internal carotid artery were separated. The common carotid artery and external carotid artery were ligated using surgical sutures, and the internal carotid artery was clamped with an artery clip. A V-shaped incision was performed at the distal end of the common carotid artery, but no suture was inserted. MCAO group: the same isolation procedure used with the sham group was followed, and after the V-shaped incision was performed at the distal end of the common carotid artery, a suture was inserted for 2 hours to induce MCAO. This was followed by reperfusion. MCAO+FN group: MCAO modeling was performed as described above. In addition, a formononetin solution (80 mg/kg) was administered via intraperitoneal injection 30 minutes before ischemia, immediately at reperfusion (0 hours), and then at 24 hours, 48 hours, and 72 hours post-reperfusion. All rats were then sacrificed 2 hours after the final formononetin dose. For consistency, the Sham and MCAO groups were administered an equivalent volume of saline (in comparison with the experimental group).

Longa Neurological Deficit Score

After modeling, the spontaneous movement of rats in each group was observed, and their neurological function and behavioral scores were assessed using the Longa neurological deficit score (behavioral assessments were performed by an investigator blinded to group allocation, though this was not explicitly stated in Methods). This scoring system ranges from 0 to 4 (with a higher score indicating more severe neurological deficit): 0, normal, no neurological deficit; 1, inability to fully extend the paralyzed front paw, mild neurological deficit; 2, circling towards the paralyzed side while walking, moderate neurological deficit; 3, falling towards the paralyzed side while walking, severe neurological deficit; 4, inability to walk spontaneously, loss of consciousness.²⁸

Staining to Observe Cerebral Infarction Area

Following euthanasia via cervical dislocation under isoflurane (at a concentration of 3%) anesthesia,²⁹ all animals had their brain tissues harvested and rapidly frozen at −20°C. The rat brains were subsequently sectioned continuously at a thickness of 2 mm. These sections were then placed in 2% triphenyl tetrazolium chloride (TTC) solution (PH1205, Phygene, Fuzhou, Fujian, China), stained in the dark at 37°C for 15 to 30 minutes, and subsequently fixed with 4% paraformaldehyde (LA0427, Labcoms, Fuzhou, Fujian, China). Under these conditions, normal brain tissue stained red, while infarcted tissue remained white.

Hematoxylin-Eosin (H&E) Staining

Rat lung tissue samples were paraffin-embedded, and 5-micrometer-thick sections were prepared from each embedded block. The sections were baked at 70°C for 40 minutes, dewaxed, and then stained with H&E. Following staining, the sections were mounted with neutral balsam and observed under a microscope.³⁰

ELISA to Detection of Inflammatory Factor Expression in Serum and Bronchoalveolar Lavage Fluid (BALF)

BALF samples were collected from rats in each group. ELISA kits (96T, Quanzhou, Fujian, China) were then utilized to detect the expression levels of TNF-α, IL-10, IL-6, and IL-1β in both serum and BALF. The optical density (OD) values of each well were measured at a wavelength of 450 nm, and the resulting data were collected for subsequent analyses.

Biochemical Analyses of Superoxide Dismutase (SOD) and Malondialdehyde (MDA) Content in Lung Tissue

Lung tissue samples from rats in each group were homogenized to obtain supernatants. The SOD (E-BC-K019-M-96T, Elabscience Biotechnology Co., Ltd., Wuhan, China) and MDA (96T, Quanzhou, Fujian, China) contents in these supernatants were then evaluated using biochemical kits (according to the manufacturer’s instructions). OD values were subsequently measured using a microplate reader, and the resulting data were collected for subsequent analyses.

Flow Cytometry to Detect Expression Levels of T-Helper (Th) 1, Th2, Th17, Regulatory T (Treg) Cells, and the CD4⁺/CD8⁺ Ratio in Lung Tissue

Antibody solutions were diluted 1:50 in TBST prior to use, and the 10× wash buffer was diluted 1:10 with ddH2O to prepare 1× working solution. Lung tissues were collected from rats and processed into single-cell suspensions. For each experimental group, unstained control tubes were prepared without antibodies against IFN-γ (PE-65153, Proteintech, Chicago, USA), IL-4 (APC-65074, Proteintech, Chicago, USA), IL-17A (66148-1-1g, vProteintech, Chicago, USA), Foxp3 (65089-1-Ig, Proteintech, Chicago, USA), CD4 (FITC-65104, Proteintech, Chicago, USA), or CD8 (PE-6509, Proteintech, Chicago, USA) for instrument voltage adjustment. Single-color positive controls were prepared by incubating cell suspensions (300–500 µL in medium with 10–20% FBS) with respective antibody working solutions (1:50 dilution) at 4 °C for 60 min, with gentle resuspension every 10–20 min. After incubation, cells were centrifuged (200×g, 5–10 min, room temperature), washed with 1–2 mL of 1× wash buffer, centrifuged again, and finally resuspended in 300 µL of 1× Apoptosis Wash Buffer. Double‑staining sample tubes followed the same procedure except that two corresponding antibody working solutions were added simultaneously during incubation. Following final resuspension, samples were analyzed immediately on the flow cytometer.

Cell Model Establishment

Microglial Cell Extraction

All microglial cells were sourced from SPF-grade SD rats within 24 hours of birth. Newborn SPF SD rats (within 24 h after birth) were sterilized with 75% ethanol. Following euthanasia via cervical dislocation under isoflurane (at a concentration of 3%) anesthesia, brain tissue was aseptically collected, placed in a PBS-containing dish, and dissected to remove the cerebellum, hippocampus, and cerebral medulla. The cerebral cortex was separated, and meninges and blood vessels were carefully stripped. The tissue was minced into approximately 1 mm³ pieces and digested with 0.2% collagenase type I for 60 min in an incubator, with gentle pipetting every 10 min. After digestion, the suspension was filtered through a cell strainer. The filtrate was collected and centrifuged at 1000 rpm for 4 min. The supernatant was discarded, and the pellet was washed once with PBS, followed by centrifugation (1000 rpm, 4 min). The supernatant was removed, and the pellet was resuspended in complete microglial medium (prepared by mixing 22.25 mL DMEM, 22.25 mL F12, 5 mL fetal bovine serum, and 0.5 mL penicillin‑streptomycin). The cells were then seeded into 25 T flasks and cultured at 37 °C under 5% CO₂.

OGD/R Model Induction and Formononetin Concentration Screening

In preparation for the in vitro experiments, the microglial cells were divided into three groups: a control group; an OGD/R group; and an OGD/R+FN (IF0190, Solarbio, Beijing, China) group. Using a 96-well plate, 200 μL aliquots of cell suspension (equivalent to~6×10³ cells) were added to each well, and the plates were then incubated at 37°C. After 24 hours, the control group wells were replenished with complete medium, while the OGD/R and OGD/R+FN wells were washed twice with phosphate-buffered (EH80256, ELGBIO Company, Guangzhou, Guangdong, China) saline, and then subjected to OGD by adding glucose-free Dulbecco’s modified eagle medium (AW-Med-017-500mL, Shanghai Anwei Biotechnology Co., Ltd., Shanghai, China) and serum-free medium, respectively.³¹ Next, the relevant wells of the 96-well plates were sealed with film, and the plates were transferred to an incubator for 6 hours of OGD. The medium was then exchanged for complete medium for 24 hours of reoxygenation to establish the OGD/R model. For the OGD/R+FN group, after 6 hours of OGD, the medium was replaced with complete medium containing 10, 50, and 100 μmol·L⁻¹ of formononetin and incubated for 24 hours of reoxygenation. To select the appropriate drug concentration, we evaluated cell morphology under a microscope and measured cell viability with the CCK-8 assay following the treatments.

Exosome Extraction

Cell supernatants were collected and centrifuged at 300×g (4°C) to pellet cell debris. The clarified suspensions were then subjected to ultrafiltration (3500×g, 30 min, DT5-2 centrifuge, Beijing Times North Centrifuge Co., Ltd.) for volume concentration. The resulting concentrate was sterile-filtered (0.22 μm) to yield the isolated exosomes.

Nanoparticle Tracking Analysis (NTA) Detection of Exosomes

Exosome isolates from each group were diluted to a suitable concentration using appropriate amounts of each stock. NTA of the exosomes was then performed to obtain particle size and exosome concentration. Prior to loading the exosome samples, instrument performance was tested using known standards. Only after satisfactory instrument performance was confirmed were the exosome samples loaded for subsequent analysis.

Electron Microscope Scanning of Exosomes

For analysis, 10 μL aliquots of the exosome samples were pipetted onto individual copper grids and allowed to precipitate for 1 minute. The remaining supernatant was then removed from each grid using filter paper. Next, 10 μL of uranyl acetate was added to each copper grid for another minute of precipitation. The supernatant was again removed with filter paper, and the sample was allowed to dry at room temperature for several minutes. A transmission electron microscope (HITACHI HT7700, Hitachi, Ltd., Tokyo, Japan) at 100 kV was used to obtain the imaging results.³²

Analysis of Exosome Surface Marker Proteins by Western Blot

For protein quantification, 100 μL aliquots of samples from each group were mixed with a 100 μL of exosome radio immunoprecipitation assay lysis buffer (20101ES60, Yeasen Biotechnology Co., Ltd., Shanghai, China) to facilitate denaturation, and the samples were then subjected to discontinuous SDS-PAGE on 12% or 8% resolving gels (with a 5% stacking gel). After sample loading (C0302B, HaiGene Biotech Co., Ltd., Haerbin, Heilongjiang, China), protein electrophoresis was conducted for 90 minutes. Next, the separated protein bands were electro-transferred to a polyvinylidene difluoride (ISEQ00010, Millipore, Darmstadt, Germany) membrane. The membrane was subsequently blocked for 2 hours and then incubated (at 4°C) overnight with primary antibodies against CD81 and TSG101 (27855-1-AP and 67381-1-Ig, Proteintech, Chicago, USA). After washing, secondary antibody incubation was performed at room temperature for 2 hours. Next, the membrane was washed once for 10 minutes. Finally, the membrane was developed using HRP substrate and then exposed to film.

LPS-Induced Lung Epithelial Cell Injury

Rat lung epithelial cells (AnWei-sci, Shanghai, China) were passaged and expanded for plating in 96-well plates. The cells were subsequently divided into four groups: a control group; an LPS+Control-exo group (control group exosomes added); an LPS+OGD/R-exo group (OGD/R group exosomes added); and an LPS+OGDR+FN-exo group (OGD/R+FN group exosomes added). Each well of the 96-well plate was seeded with 200 μL of a cell suspension containing 6×10³ cells, and the plates were then incubated for 24 hours. All treatment groups (except for the control group) were treated with 20 ng/mL LPS (L8880, Solarbio, Beijing, China) for 24 hours to mimic OGD/R-induced lung epithelial cell injury. The medium was subsequently replaced with medium containing 5 μg/mL of the respective exosomes, and the plates were incubated for an additional 48 hours.

Assay of Cell Viability

After LPS-induced injury of lung epithelial cells, 10 μL of cell counting kit-8 (CCK-8) solution (CA1210, Solarbio, Beijing, China) was added to each well. The cells were then incubated in an incubator for 4 hours, and the OD value at 450 nm was recorded.

ELISA Analyses of Inflammatory Factor Expression

ELISA assays for TNF-α, IL-10, IL-6, and IL-1β were performed using dedicated kits (96T, Quanzhou, Fujian, China) according to the manufacturer’s instructions. OD values for each group were measured using a microplate reader, and the data were collected for subsequent analysis.

SOD and MDA Assays

Cell supernatants from each group were assayed using dedicated kits (96T, Quanzhou, Fujian, China) according to the manufacturer’s instructions. OD values for each group were measured at 550 and 532 nm using a microplate reader, and the data were collected for subsequent analysis.

Statistical Analysis

All analyses were executed using GraphPad Prism (Version-10.0). Multiple group comparisons were assessed by one-way analysis of variance (ANOVA), with Tukey’s post hoc test applied for pairwise comparisons. All continuous variables are presented as mean ± standard error of the mean (SEM). A two-tailed P-value <0.05 was used to test for statistical significance.

Results

Rats in MCAO Formononetin Exerts Cerebroprotective Effects on Cerebral Ischemia-Reperfusion

To investigate the cerebroprotective effects of formononetin on CIR, we used MCAO modeling in rats. SD rats were randomly divided into three groups (with nine rats per group): a sham-operated group, an MCAO injure group, and an MCAO with formononetin intervention (MCAO+FN) group. After MCAO modeling and formononetin intervention, rats from the three experimental groups were assessed using the Longa neurological deficit score. In comparison with the sham-operated group, rats in the MCAO group exhibited obvious circling towards the paralyzed side while walking, indicating moderate neurological deficit, with higher neurological and behavioral scores. However, rats in the MCAO+FN group exhibited an attenuation in many of these symptoms, and an overall decrease in the Longa neurological deficit score (Figure 1A). In addition, TTC staining for brain tissue damage revealed that the area of infarction in the MCAO+FN group was reduced in comparison with the MCAO group (Figure 1B). So, rats in the MCAO+FN group had milder neurological deficits (compared with rats in the MCAO group), and the cerebral infarction area was smaller. Together, these results provide evidence that formononetin has certain cerebroprotective and therapeutic effects on CIR.

Figure 1.

(A) The TTC staining of brain tissue indicated that the damage in the MCAO+FN group was attenuated compared with the MCAO group. (B) The Longa score in the MCAO group were higher than MCAO+FN groups (** P <0.01, *** P<0.001).

Abbreviations: TTC, 2% triphenyl tetrazolium chloride; MCAO, middle cerebral artery occlusion; FN, formononetin.

Formononetin Improves Lung Injury in CIR by Reducing Inflammatory Cytokine Expression

After euthanasia and organ harvesting, lung tissue sections from rats in each group were stained with H&E and observed under a microscope to evaluate cellular morphology (Figure 2). While the lung tissue structure of rats in the sham-operated group was normal, with regular alveolar morphology, the lung tissue structure of rats in the MCAO group was significantly damaged-disruption of lung tissue structure, partial necrosis of lung epithelium, extensive disappearance of the alveolar lumen, large-scale inflammatory cell infiltration, hemorrhaging in lung blood vessels and the interstitial tissues. In comparison (with the MCAO group), these wholesale pathological changes were significantly alleviated in the lung tissues of rats in the MCAO+FN group, with partial recovery of the alveolar lumen, thickened alveolar walls, and reduced levels of inflammatory cell infiltration and hemorrhaging.

Figure 2.

(A) control: normal lung tissue, including alveoli (a). (B) MCAO: extensive areas of loss of normal alveolar architecture (arrow b) and mononuclear infiltration (arrow c). (C) MCAO + FN: alveolar structure nearly to normal (arrow d) with a focal area of loss of normal architecture (arrow e). (H&E×100, scale bar: 400μm).

Abbreviations: MCAO, middle cerebral artery occlusion; FN, formononetin.

Additionally, multiple ELISAs were conducted on BALF and serum from the rats to evaluate the expression levels of TNF-α, IL-10, IL-6, and IL-1β (Figure 3). In comparison with the sham-operated group, the expression levels of TNF-α, IL-10, IL-6, and IL-1β were increased in the BALF of rats in the MCAO group. However, TNF-α, IL-10, IL-6, and IL-1β expression levels were downregulated (to varying degrees) in the BALF of rats in the MCAO+FN group (when compared with the MCAO group). The ELISA results obtained using rat serum were comparable to those obtained with BALF, with reduced expression levels of TNF-α, IL-10, IL-6, and IL-1β in serum from rats in the MCAO+FN group (when compared with the MCAO group). The reduction in expression was more significant in BALF. In summary, formononetin reduced the levels of TNF-α, IL-10, IL-6, and IL-1β inflammatory cytokines in the BALF and serum of MCAO rats. Hence, formononetin improved the inflammatory state in model rats by reducing the expression levels of inflammatory cytokines in the blood and lungs, thus alleviating lung inflammation.

Figure 3.

(A) Cytokine levels in BLAF: TNF-α, IL-6, IL-1β, and IL-10 were lower in the MCAO + FN group than in the MCAO group. (B) Cytokine levels in serum: TNF-α, IL-6, and IL-1β were lower, and IL-10 showed a downward trend in the MCAO + FN group relative to the MCAO group (*P<0.05, **P<0.01, ***P<0.001, ns: not significance).

Abbreviations: MCAO, middle cerebral artery occlusion; FN, formononetin.

Formononetin Exhibits Anti-Oxidative Stress Activities

Next, the lung tissues of rats from each group were processed to facilitate analyses of their SOD and MDA contents. In comparison with the sham-operated group, rats in the MCAO group exhibited decreased SOD content (P<0.05) and increased MDA content (P<0.001). However, rats in the MCAO+FN group exhibited a significant reduction (compared with the MCAO group) in MDA levels (P<0.001). While a decrease in SOD activity was also observed, this trend did not reach statistical significance (P>0.05) (Figure 4). Hence, formononetin decreased MDA content and increased SOD content in the lung tissues of MCAO rats, thereby alleviating oxidative damage to lung tissue.

Figure 4.

(A) The MCAO + FN group showed a trend toward increased SOD expression compared to the MCAO group. (B) MDA expression was downregulated in the MCAO + FN group compared to the MCAO group. (***P<0.001, ns: not significance).

Abbreviations: MCAO, middle cerebral artery occlusion; FN, formononetin.

Formononetin Has Immunomodulatory Effects

Flow cytometry was utilized to analyze Th1, Th2, Th17, Treg, CD4⁺, and CD8⁺ levels in lung tissue, as well as the CD4⁺/CD8⁺ ratio. The above cell subsets were selected for analysis because Th1 and Th2 are in vivo markers for cellular immunity and humoral immunity (respectively), Th17 reports on the degree of inflammatory reaction, and the ratio of Treg to CD4⁺/CD8⁺ reports on whether the immune system is in normal operation.^33,34 In comparison with the Sham-operated group, the positive rates for Th1 (cytokine IFN-γ⁺) cells, Th2 (cytokine IL-4⁺) cells, Th17 (cytokine IL-17A⁺) cells, and Treg (transcriptional regulator Foxp3⁺) cells were significantly increased in the MCAO group (P<0.01). Moreover, CD4⁺ cell levels were significantly increased (P<0.01), and there were trending increases in CD8⁺ levels and the CD4⁺/CD8⁺ ratio. However, with the exception of the CD4⁺/CD8⁺ ratio, all of the above indicators were reduced (to varying degrees) in the MCAO+FN group (P<0.05) (Figure 5). Hence, formononetin regulated immune cell balance in lung tissues, thereby alleviating lung injury in CIR.

Figure 5.

Flow cytometry for lung tissue cells across groups. Immunological marker expression in the MCAO + FN group returned toward normal compared to the MCAO group. (*P<0.05, **P<0.01, ***P<0.001, ns: not significance).

Abbreviations: MCAO, middle cerebral artery occlusion; FN, formononetin.

Formononetin at a Concentration of 10 μmol·L⁻¹ Exerts Protective Effects on Microglia in vivo Model of OGD/R

In our in vivo model, OGD/R induced multiple changes in microglia cell structure (when compared with control cells). Thus, microglia cell processes were generally retracted, rounded cells were more prominent, floating cells were more numerous, and some microglial cells exhibited growth arrest. Conversely, formononetin at 10 μmol·L⁻¹ most effectively reversed these changes, enhanced viability, and promoted proliferation. CCK-8 results confirmed that 10 μmol·L⁻¹ yielded the highest cell viability among the three tested concentrations (Figure 6). In summary, while glucose and oxygen deprivation conditions inhibited microglia cell growth and promoted the retraction of microglial processes, cell damage could be reversed by exchanging the medium with complete medium containing formononetin and incubating for 24 hours. Additionally, formononetin treatment enhanced cell viability and promoted cell proliferation.

Figure 6.

(A–E) Morphology of microglial cells observed under electron microscopy after treatment with different concentrations of formononetin; (F) Cell viability of each group detected by CCK-8 assay. (*P<0.05, ***P<0.001, ns: not significance).

Abbreviations: OGD/R, oxygen-glucose deprivation/re-oxygenation; FN, formononetin.

The Extracted Substance Was Microglial Exosomes

As shown in Table 1, high concentrations of microglial exosomes were extracted from each experimental group (1.3×10⁸ particles/mL from the microglial control group vs 1.2×10⁸ particles/mL from the OGD/R group vs 1.3×10⁸ particles/mL from the OGD/R+FN group). The average particle size in each group ranged from 97.4 to 101.6 nm. The transmission electron microscope scanning results for these particles are presented in Figure 7A. Western blot analysis confirmed the expression of exosome markers CD81 and TSG101 in all groups (Figure 7B, the Western blot analysis in this study was performed as a qualitative assay to confirm the presence of target proteins; therefore, loading controls for normalization were not included). In summary, exosomes could be extracted from each experimental group, allowing us to proceed with subsequent experiments.

Table 1.

Analysis of Nanoparticle Tracking Analysis in Microglia Exosomes Scanned by Transmission Electron Microscope

Groups	Original Concentration/Sizes in nm	Median (Mean ± SD)
Control	1.3×108 Particles/mL	97.9±33.8
OGD/R	1.2×108 Particles/mL	99.5±32.9
OGD/R+FN	1.3×108 Particles/mL	97.4±35.5

Notes: No significant differences in particle concentration or size were observed between groups (P > 0.05).

Figure 7.

(A) The morphology of the extracted vesicles from each group, as observed under transmission electron microscopy, is consistent with that of exosomes. (B) All groups showed expression of the exosomal surface markers CD81 and TSG101.

Formononetin Protection Against OGD/R-Induced Lung Epithelial Cell Injury Is Facilitated via Exosome-Mediated Communication

LPS were used to generate a lung epithelial cell injury model, and the damaged cells were then treated with the isolated exosomes. First, a primary culture of rat lung epithelial cells was sub-divided to provide a control group, an LPS+Control-exo group (subsequently treated with control group exosomes), an LPS+OGD/R-exo group (subsequently treated with OGD/R group exosomes), and an LPS+OGDR+FN-exo group (subsequently treated with OGD/R+FN group exosomes). Next, the different experimental groups were treated with LPS to induce cell damage, and the damaged lung epithelial cells were then treated with exosomes (as indicated). Lung epithelial cell viability in the various groups was subsequently assessed using CCK-8. As expected, lung epithelial cell survival rate was highest in the control group. However, while the survival rate of lung epithelial cells was significantly reduced in the LPS+OGD/R-exo group (P<0.001), this reduction was attenuated in the LPS+OGD/R+FN-exo group (P<0.01) (Figure 8A). In summary, although the viability of lung epithelial cells was significantly reduced after LPS treatment, lung epithelial cell damage could be reversed using exosomes isolated from formononetin treated microglia cells.

Figure 8.

(A) CCK-8 assay: OGD/R-exo exacerbates LPS-induced lung epithelial cell damage; Compared to LPS+OGD/R-exo group, LPS+OGD/R-exo+FN group reverses the damage caused by LPS and OGD/R-exo to lung epithelial cells, increasing cell viability. (B) Formononetin demonstrates antioxidant stress capabilities in oxidative stress enzyme activity assays. (C) Inflammatory factor expression in lung epithelial cells across groups: Formononetin reduces IL-1β expression, TNF-α, IL-6, and IL-10 showed a downward trend in theLPS + OGD/R-exo + FN relative to the MCAO (*P<0.05, **P<0.01, ***P<0.001, ns: not significance).

Abbreviations: LPS, Lipopolysaccharide; OGD/R, oxygen-glucose deprivation/re-oxygenation; FN, formononetin.

Additionally, SOD content in the LPS+Control-exo group was significantly decreased when compared with the control group (P<0.001), and the MDA content was increased. Similarly, SOD content in the LPS+OGD/R-exo group was markedly reduced when compared with the control group (P<0.001), and the MDA content was significantly increased (P<0.001). However, the SOD content in the LPS+OGD/R+FN-exo group was significantly increased (P<0.001) when compared with the LPS+OGD/R-exo group, and the MDA content was decreased (P<0.05) (Figure 8B). Together, these results indicate that formononetin treatment can induce antioxidant effects via exosomes after OGD/R.

ELISA analyses of inflammatory cytokine levels indicated that the levels of TNF-α, IL-6, IL-1β, and IL-10 were increased to varying degrees in both the LPS+CON-exo and LPS+OGD/R-exo groups (when compared with the control group). However, the levels of these inflammatory cytokines were decreased (to varying degrees) in the LPS+FN-exo group when compared with the LPS+OGD/R-exo group (Figure 8C). In summary, formononetin-loaded exosomes reduced the levels of inflammatory cytokines (to varying degrees) after OGD/R. Notably, however, the anti-inflammatory cytokine IL-10 was also decreased, suggesting that formononetin may reverse a compensatory anti-inflammatory response rather than exerting a broadly suppressive effect on all cytokines. Nonetheless, this modulation was associated with an improvement in the inflammatory status of rat lung epithelial cells under LPS-induced injury combined with glucose-free hypoxic conditions.

Discussion

The aim of the present study was to investigate the protective effects and mechanisms of formononetin on CIR-induced lung injury. Our findings reveal that formononetin treatment reduced cerebral infarction area and inflammatory cytokine expression in MCAO rats. Interestingly, while IL-10 is conventionally defined as an anti-inflammatory factor, we observed a decline in in vivo IL-10 expression after formononetin administration. Because this decrease was accompanied by a parallel reduction in other inflammatory cytokines, we propose that a pre-treatment elevation in IL-10 expression is indicative of a significant inflammatory burden. Consequently, the observed formononetin-induced drop in IL-10 was the result of an alleviation of inflammation, possibly through a feedback mechanism that downregulates IL-10 production as the inflammatory drive subsides.³⁵ In addition, formononetin was observed to mitigate oxidative stress responses and to modulate immune cell function, thereby reducing the vicious cycle of oxidative stress, immune dysregulation, and inflammatory responses. Through these various mechanisms, formononetin was able to alleviate CIR-induced lung injury. Flow cytometry analysis revealed a significant increase in Treg (Foxp3⁺) cells within the lung tissues of MCAO model rats, consistent with activation and dysregulation of the immune response following cerebral ischemia. This elevation in Treg cells likely reflects a homeostatic feedback mechanism involved in the suppression of systemic immune imbalance and the dampening of inflammation.³⁶ Paradoxically, treatment with formononetin reduced the MCAO-induced accumulation of Treg cells. This counterintuitive finding suggests that formononetin may help relieve the primary immune dysregulation at its source, thereby reducing the compensatory need for heightened immunosuppression and ultimately helping to restore systemic immune homeostasis.

In our in vitro studies, we established a CIR-induced lung injury model using LPS to induce lung epithelial cell injury. After adding LPS to induce injury, the expression of inflammatory factors and oxidative stress levels in lung epithelial cells increased, which is highly consistent with the in vivo experimental results. This provided further confirmation of the feasibility of using an LPS-induced lung epithelial cell injury model to simulate CIR-induced lung injury. Furthermore, our in vitro cell-level experiments demonstrated that exosomes derived from microglial cells treated with formononetin can protect lung epithelial cells, although the underlying mechanism requires further investigation.

Currently, the mechanisms and pathways underlying CIR-induced lung injury remain unclear. According to previous studies, inflammatory cascades, oxidative stress, and immune regulation may all play a role in CIR-induced lung injury.^9,12,13 After CIR injury, the inflammatory cytokine expression levels and oxidative stress responses in the body increase significantly.³⁷ During the inflammatory response, inflammatory cytokines induce vascular endothelial cells to release adhesion factors and chemokines, thereby promoting the diffusion and maintenance of the inflammatory response.⁸ These cytokines activate the chemotaxis and adhesion of inflammatory cells, promoting their aggregation at the inflammatory site and stimulating the release of various inflammatory mediators, which further exacerbates the inflammatory response.^38–40 In addition, the body is known to experience excessive oxidative stress after CIR injury, which is characterized by the overproduction of ROS and the activation of oxidative stress-associated kinases. These events trigger inflammatory signaling pathways and the subsequent release of pro-inflammatory cytokines. Concurrently, post-stroke immune dysregulation occurs, wherein immune cells and inflammatory mediators synergistically amplify ROS generation, exacerbating oxidative stress. This reciprocal interaction establishes a self-reinforcing vicious cycle that intensifies CIR-induced pulmonary injury.⁴¹ The different possible mechanisms of cerebral ischemia-reperfusion induced lung injury are outlined in Figure 9 (from a consideration of the known literature).

Figure 9.

The possible mechanism of CIR-induced lung injury.

From the experimental results presented above, it can be concluded that formononetin exhibits anti-inflammatory, antioxidant, and immunomodulatory functions, and that it can help alleviate CIR injury and protect lung tissues. Moreover, formononetin was demonstrated to exert its protective and therapeutic effects on CIR-induced lung injury via exosome cell-cell communication. These interesting results notwithstanding, this study has several limitations. First, the optimal concentration of formononetin warrants further exploration, particularly regarding the efficacy and cytotoxicity of lower doses in vitro. Second, while our experimental results provide evidence that formononetin exerts a protective effect against cerebral ischemia-reperfusion-induced lung injury, its protective mechanisms remain unclear.

Additional research is also needed to validate the simulation of CIR-induced lung injury by lung epithelial cell injury induced with LPS, and to understand the mechanisms by which brain cell-derived exosomes might act on the lungs. Future experiments may consider using conditioned medium from OGD/R-exposed microglia directly on lung epithelial cells, or co-culture systems, to more closely model the brain-lung axis and utilize GW4869 along with exosome profiling to verify the mediating function of exosomes in lung injury following cerebral ischemia-reperfusion. Third, our review of the literature indicates that formononetin can mitigate brain injury when administered intravenously, while our research is primarily focused on the short-term protective effects of formononetin, there is a lack of research on whether formononetin can directly act on microglia by crossing the blood–brain barrier, and on its long-term efficacy and potential toxicity.

Conclusion

In this study, we provide evidence that formononetin exerts protective and preventive effects against CIR-induced lung injury. The underlying mechanisms involve an attenuation of inflammatory responses, antioxidant activities, and immunoregulatory mechanisms, providing a theoretical foundation for clinical strategies targeting CIR-induced lung injury.

Funding Statement

This study was supported by the Science and Technology Major Project of Jinhua City (No. 2024-3-007).

Abbreviations

CIR, Cerebral ischemia-reperfusion; MCAO, Middle cerebral artery occlusion; ELISA, Enzyme-linked immunosorbent assay; OGD/R, Oxygen-glucose deprivation/re-oxygenation; LPS, Lipopolysaccharide; TNF-α, Tumor necrosis factor alpha; IL-6, Interleukin 6; IL-1β, Interleukin 1β; IL-10, Interleukin 10; SOD, Superoxide dismutase; MDA, Malondialdehyde; INF-γ⁺, Interferon-gamma; Foxp3⁺, Forkhead box protein P3 positive; CD4⁺, Cluster of differentiation 4 positive; CD8⁺, Cluster of differentiation 8 positive; IL-4⁺, Interleukin 4 positive; IL-17A⁺, Interleukin 4 positive; SPF, Specific Pathogen Free; TTC, Triphenyl tetrazolium chloride; H&E, Hematoxylin-eosin; BALF, Bronchoalveolar lavage fluid; OD, Optical density; Th1, T-helper 1; Th2, T-helper 2; Th17, T-helper 17; Treg, Regulatory T; NTA, Nanoparticle tracking analysis; CCK-8, Cell counting kit-8; ROS, Reactive oxygen species.

Data Sharing Statement

This study follows the Taylor & Francis “Share Upon Reasonable Request” data policy. The original data presented in this study are available upon reasonable request, and further inquiries should be directed to the corresponding author, Saibin Wang.

Ethics Approval and Consent to Participate

The study was approved by the Jinhua Municipal Central Hospital’s Research Ethics Committee (AL-JHYY202446) and performed in accordance with Guide for the Care and Use of Laboratory Animals. All methods are reported in accordance with the ARRIVE guidelines 2.0 (https://arriveguidelines.org) for the reporting of animal experiments.

Consent for Publication

All the listed authors have carefully reviewed and approved this manuscript. We confirmed that the work described has not been published before.

Author Contributions

Conceptualization – Saibin Wang and Luying Chen. Data curation – Luying Chen and Yajie Zhou. Formal analysis –Saibin Wang, Luying Chen and Yajie Zhou. Funding acquisition – Saibin Wang. Investigation –Luying Chen and Yajie Zhou. Methodology – Huijun Chen, Renzhi Zhou and Lingyun Lou. Project administration – Saibin Wang, Dan Yan and Renzhi Zhou. Resources – Saibin Wang and Luying Chen. Software – Dan Yan, Renzhi Zhou and Luying Chen. Supervision – Saibin Wang. Validation – Saibin Wang, Luying Chen and Yajie Zhou. Visualization – Luying Chen and Yajie Zhou. Writing – original draft – Saibin Wang, Luying Chen and Yajie Zhou. Writing – review and editing – All authors. All authors took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no competing interests.