Ginsenoside Rb1 Improves Brain, Lung, and Intestinal Barrier Damage in Middle Cerebral Artery Occlusion/Reperfusion (MCAO/R) Mice Through the PPARγ Signaling Pathway

Background Ischemic stroke causes brain inammation and multi-organ injury, in which the peroxisome proliferator-activated receptor-gamma (PPARγ) signaling pathway plays a key role. Studies have indicated that ginsenoside Rb1 (GRb1) can protect blood–brain barrier integrity after stroke. Here, we aimed to determine whether GRb1 can ameliorate brain/lung/intestinal barrier damage in mice with acute cerebral ischemia through the PPARγ signaling pathway.


Introduction
Stroke accounts for almost 5% of all disability-adjusted life years (DALYs) [1] and 10% of all deaths worldwide [2], with the bulk of this burden (over 75% of deaths from stroke and 81% of DALYs) falling on low-and middle-income countries [3]. It has become increasingly clear that stroke causes multi-organ systemic diseases and triggers systemic responses [4]. Pulmonary infection and stress ulceration are the most common complications for severe stroke patients [5].
Several studies have focused on stroke-induced lung infection. The lung is particularly vulnerable after stroke, including ischemic and hemorrhagic strokes [6,7]. Elmer [8] et al. reported that 15.6% of stroke patients develop acute lung injury(ALI) within 36 h of hospital admission, while 7.8% develop pneumonia or bronchitis during hospitalization. Current clinical studies frequently report that CNS injury patients develop gut microbiota dysbiosis, gut dysmotility, gastrointestinal bleeding, and even gut-origin sepsis, effects that are regularly associated with worse outcomes, including deteriorating functional de cits and increased mortality rates [9]. After an ischemic and hemorrhagic stroke, approximately 50% of patients experience gastrointestinal complications, including dysphagia, gastrointestinal hemorrhaging, constipation, and bowel incontinence [10,11]. Furthermore, gut permeability has been shown to increase after stroke in mice, and correlates with stroke severity [12]. However, there is a lack of systematic and indepth research on the role and mechanism underlying the effects of acute ischemic stroke on the pulmonary and intestinal barriers.
Peroxisome proliferator-activated receptor-gamma (PPARγ), a stress-induced transcription factor, plays essential roles in the regulation of cellular differentiation, development, and metabolism [13]. Numerous studies have shown that PPARγ plays a key role in the barrier functions of epithelial and endothelial cells. PPARγ maintains the integrity of the blood-brain barrier (BBB) via inhibiting hypoxia-induced nuclear factor kappa B (NF-κB) nuclear translocation [14]. It has been shown that magnolol can mitigate lipopolysaccharide (LPS)-induced ALI, and this is achieved, at least partly, through PPARγ activation [15].
There is also convincing evidence that the microbiota in uences intestinal PPARγ in maintaining intestinal mucosal homeostasis [16].
Panax ginseng (Renshen, Chinese ginseng) is a traditional Chinese herbal medicine that is extensively used worldwide as a medicinal and nutritional supplement for a variety of disorders [17,18]. More than 40 different ginsenosides have been isolated from ginseng [19], including GRb1, which was identi ed as one of its main active ingredients [20]. GRb1 has been reported to play a neuroprotective role in the acute phase of cerebral ischemia by protecting BBB integrity, increasing regional cerebral blood ow, and enhancing the stability of the neuronal ultrastructure [21]. Moreover, GRb1 can signi cantly attenuate LPSinduced lung microcirculatory injury [22], and can protect the intestinal mucosal barrier against damage induced by peritoneal air exposure [23]. Recent studies have demonstrated that GRb1 promotes adipogenesis through the regulation of PPARγ [24] and is a potential ligand of PPARγ, as evidenced via a newly developed method called polarized total internal re ection uorescence microscopy (polTIRFM) [25]. Despite the large number of reports on the wide-ranging effects of GRb1, whether this ginsenoside can alleviate multi-organ injury induced by ischemic stroke via PPARγ remains unknown. Therefore, in the present study, we investigated how GRb1 affects the brain/lung/intestinal barriers using a middle cerebral artery occlusion/reperfusion (MCAO/R) mouse model, and explored the putative underlying mechanisms.  23, revised in 1996). The animals were housed in a controlled environment (12:12 h light/dark cycle, humidity of 60 ± 5%, and a temperature of 22 ± 3 °C) and were allowed access to food and water ad libitum. Before surgery, all the mice were allowed to acclimatize to the laboratory conditions for at least three days. All the measurements of outcome were performed by investigators who were blinded to the experimental conditions. 2.3. Preparation of the cerebral ischemia/reperfusion (I/R) model, and induction of acute colitis, and the mouse model of ALI MCAO/R was performed by intraluminal nylon lament intrusion in accordance with Longa's method [26].
Brie y, mice were anesthetized by intraperitoneal injection of 3% pentobarbital sodium (0.1 mL/10 g body weight); then, the neck vessels were exposed by a midline incision and the branches of the right external carotid artery were carefully isolated and cauterized. A 6 − 0 nylon mono lament suture, blunted at the tip and coated with 1% poly-L-lysine, was advanced 9-10 mm into the internal carotid to occlude the origin of the middle cerebral artery. During surgery and ischemia, the temperature of the animal was maintained at 37.0 ± 0.5 °C with a heating pad (ALCBIO, Shanghai, China). After 60 min of occlusion, reperfusion was achieved by the withdrawal of the suture. The same treatment was carried out on mice in the Sham group, except that the suture was not advanced into the internal carotid artery [27]. A laser Doppler owmeter (FLPI2, Moor Instruments Ltd, Axminster, UK) was used to con rm that middle cerebral artery blood ow had decreased to < 30% of the basic cerebral blood ow immediately after the occlusion.
Animals with a blood ow of less than 30% that of preischemic levels were used for further study.
Acute colitis was induced by DSS (3%, w/v) administered in double-distilled drinking water for seven consecutive days [28].

Groups
Ninety-six ICR mice were randomly divided into the following six groups (n = 16 per group) according to the random number table generated by SPSS software 22.0 (SPSS Inc., Chicago, IL, USA): i) Sham group, where the mice received a sham operation and an equal volume of normal saline; ii) Model group, where the mice received MCAO/R and an equal volume of normal saline; iii) GRb1 group, where the mice received MCAO/R and were treated with 20 mg/kg GRb1 (dissolved in normal saline) administered by intraperitoneal injection after 1 h of ischemia; iv) Eda group, where the mice received MCAO/R and were treated with 3 mg/kg Edaravone (dissolved in normal saline) administered by intraperitoneal injection after 1 h of ischemia; v) DSS group, where the mice received 3% DSS for seven days; vi) LPS group, where the mice received 5 mg/kg LPS administered intratracheally. Subsequently, the mice were administrated with GRb1with or without GW9662. ICR mice (n = 36) were randomly divided into the following six groups: Sham group, Model group, GRb1 group (Model + GRb1), GW9662 plus Sham group (Sham + GW), GW9662 plus Model group (Model + GW), and GRb1 plus GW9662 plus Model group (Model + GRb1 + GW). One hour before surgery, mice from the Sham + GW, Model + GW, and Model + GRb1 + GW groups were intraperitoneally injected with GW9662 (4 mg/kg body weight).

2,3,5-Triphenyltetrazole ammonium chloride (TTC) staining
Brains were quickly removed at 24 h after reperfusion. The 2, 3, 5-triphenyl tetrazolium chloride (TTC) staining was performed to evaluate tissue viability and measure the infarct size. The infarct area was measured in NIH Image J software (Version 1.42; National Institutes of Health, Bethesda, Md). The infarct areas on each slice were summed and multiplied by slice thickness to give the infarct volume. Infarct volume was expressed as a percentage of infarction per ipsilateral hemisphere [30].
2.6. Determination of Evans blue (EB) leakage in the brain, lung, and small intestine As previously described [31], 2 h before the mice were euthanized (n = 6 per group), 2% EB (Sigma-Aldrich, St. Louis, MO, USA) in normal saline was injected into mice through the tail vein. After 2 h, the mice were anesthetized and perfused with 0.9% normal saline. The brains, lungs, and small intestines were taken from the mice and photographed. The brain, lungs, and intestines were dried with lter paper, weighed, and frozen in liquid nitrogen. Tissue homogenates were prepared (100 mg tissue/mL formamide), incubated at 60 ℃ for 18 h, and centrifuged at 5,000 × g for 30 min. The absorbance of the tissue supernatants was measured using a microplate spectrophotometer at 620 nm. The EB content was calculated according to an EB standard curve.

Changes in mesenteric blood ow
After anesthesia with 3% pentobarbital sodium, an incision of about 1-2 cm was made in the abdominal cavity of mice. Blood ow in the mesentery was measured using a laser Doppler owmeter [32].

Neutrophil count in bronchoalveolar lavage uid (BALF)
The lung lumen was sampled by ushing with 500 µL of PBS using a syringe. The BALF was centrifuged at 1500 rpm for 5 min at 4 °C and the supernatant was collected for further analysis [33].

Histomorphological analysis
Mice (n = 3, for each group) were sacri ced after MCAO/R, and transcardially perfused with physiological saline from the heart to the systemic circulation until the liver appeared white, followed by 4% paraformaldehyde solution of 4℃. The brains were removed and immersed in the 4% formaldehyde solution overnight at 4℃. Afterwards, the brains were processed through graded alcohols and xylene. By using a brain matrix, the forebrain was cut coronally into two equally spaced slices and embedded in para n blocks after detaching the hindbrain. The brain sections located at 1.5 mm behind the bregma in the coronal plane were cut into 5 mm sections with vibratome (Leica, Solms, Germany) and mounted on the glass slides. The sections were depara nized in xylene and rehydrated in 100-70% gradient ethanol. Finally, they were stained with hematoxylin-eosin (H&E staining). The slices were washed with double distilled water, dehydrated in ethanol, cleaned with xylene, and examined with light microscopy.
The pictures were taken with a digital camera. The sections corresponding to the Plate 35 contained the brain atlas were selected for analysis. The images were captured using the computer assisted image analyzer system consisting of a microscope (Olympus BX-51, Tokyo, Japan) magni cation at × 100.

Enzyme-linked immunosorbent assay (ELISA)
The serum expression of TNF-α and IFN-γ were detected using the corresponding ELISA kits (Yifeixue Bio Tech, Nanjing, China) according to the manufacturer's instructions. Absorption at 450 nm was determined using a microplate reader (Bio-Rad, Hercules, CA, USA), and TNF-α and IFN-γ concentrations were determined according to the standard curve generated at the same time.

Statistical analysis
All the results are expressed as means ± standard deviation. The Student's t-test was used for comparisons between two groups. One-way ANOVA was used for comparisons among three or more groups. All the tests were performed using GraphPad Prism v7 (GraphPad Software, La Jolla, CA, USA).
Signi cance was established at P < 0.05.

GRb1 ameliorated brain injury in MCAO/R mice
As illustrated in Figure 1A, TTC staining showed that the area of ischemic infarction of the right brain was increased in the Model group compared with the Sham group, while infarction volumes in the GRb1 group were signi cantly reduced compared with those in the Model group. Edaravone, the positive control, also signi cantly reduced infarct volume at the dose of 3 mg/kg. Representative samples of H&E-stained brain sections are shown in Figure 1B. Mice in the Model group also displayed neuronal loss and shrunken cells in the right cerebral ischemic area, as well as a loss of nucleoli and wider perivascular space. In contrast, brain damage was signi cantly mitigated in the GRb1 and Eda groups. EB staining showed that there was a signi cant increase in EB leakage in the Model group when compared with that in the Sham group (P < 0.05). GRb1 and Eda treatment both led to a signi cant reduction in EB leakage (P < 0.05, P < 0.01) (Fig. 1C, D).
3.2. GRb1 increased the expression of tight junction proteins in the brain after MCAO/R To identify tight junction (TJ) remodeling, the expression of ZO-1 and occludin were analyzed using immuno uorescence staining. Compared with the Model group, the average optical density for ZO-1 and occludin was signi cantly increased in the GRb1 group ( Fig. 2A, B). The protein expression levels of ZO-1 and occludin were also determined by western blotting. The expression levels of both proteins were decreased in the Model group compared with the Sham group; whereas they were signi cantly increased in the GRb1 group compared with those in the Model group. (Fig. 2C, D).

GRb1 ameliorated pulmonary injury in MCAO/R mice
Quantitative spectrometric analysis showed extensive EB leakage in the lung after cerebral ischemia/reperfusion. EB leakage was signi cantly reduced in the GRb1 group when compared with the Model group (P < 0.05) (Fig. 3A). Pulmonary barrier injury was accompanied by an in ux of neutrophils into the interstitium and bronchoalveolar space [35]. Neutrophil counts in alveolar lavage uid were then determined, and were found to be signi cantly higher in the Model group than in the Sham group (P < 0.05). The neutrophil content was markedly decreased in the GRb1 group, reaching levels similar to those of the Sham group (Fig. 3B). The results of H&E staining showed that lung sections excised from mice in the Sham group were normal. In contrast, the alveolar space from animals in the Model group was wider, the alveolar cavity had shrunk, and normal alveolar tissue was lost. Interestingly, treatment with GRb1 greatly suppressed lung injury and also reduced histological damage (Fig. 3C).

GRb1 increased the expression of junction proteins in the lung after MCAO/R
To identify the effects of GRb1 on the pulmonary barrier, the expression of VE-cadherin and claudin-5 in the lung was analyzed by immuno uorescence staining. The results showed that the expression levels of both proteins were signi cantly decreased in the Model group compared with those in the Sham group, whereas in the GRb1 and the Eda groups they were increased (Fig. 4A, B). The expression of VE-cadherin and claudin-5 in the lung was also determined by western blotting. The results showed that, the expression levels of these proteins in the GRb1 group were remarkably increased compared with the Model group (P < 0.05) (Fig. 4C, D).

GRb1 ameliorated intestinal injury in MCAO/R mice
To determine the effect of GRb1 on the intestinal barrier after cerebral I/R, an EB leakage assay was performed. The results showed that EB leakage was signi cantly increased (P < 0.05) in the Model group, whereas in the GRb1 group it was signi cantly decreased (P < 0.05) (Fig. 5A, B). The results of H&E staining showed that, in mice of the Model group, the intestinal villi were shortened and thickened, the tips were broken, and neutrophils were in ltrated in the intestinal mucosa and submucosa. However, mice treated with GRb1 and Eda showed a marked improvement in intestinal injury compared with the Model group. The most severe intestinal injury was seen in the DSS group (Fig. 5C). Laser Doppler owmetry is a practical and sensitive technique to evaluate the viability of the intestine [36]. Here, we used a Doppler blood ow analyzer to measure the mesenteric blood ow. Compared with the Sham group, the blood ow in the Model group was signi cantly decreased (P < 0.05); however, GRb1 treatment could extensively increase the blood ow compared with those in the Model group. (P < 0.05) (Fig. 5D).

GRb1 increased the expression of tight junction proteins in the intestine after MCAO/R
Considering that the tight junction proteins ZO-1 and occludin are critical for the maintenance of an intact intestinal permeability barrier, the effects of GRb1 on these proteins in the intestine after cerebral I/R were also examined. As expected, immuno uorescence staining results showed that, the expression levels of ZO-1 (Fig. 6A, left) and occludin (Fig. 6B, right) proteins were signi cantly decreased in the Model group compared with those in the Sham group, whereas in the GRb1 and the Eda groups they were increased compared with those in the Model group. The expression of ZO-1 and occludin in the intestinal epithelium was also determined by western blotting. The results demonstrated that the levels of both proteins were signi cantly increased in the GRb1 group compared with those in the Model group (P < 0.05) (Fig. 6C, D). These data suggested that the observed improvement in intestinal permeability was facilitated by the GRb1-mediated maintenance of tight junction protein levels.

GRb1 improved brain/lung/intestinal barrier damage through the PPARγ/NF-κB pathway
The integrity of the BBB is known to be associated with PPARγ/NF-κB signaling pathway [37]. In addition, GRb1 can reportedly increase intracellular PPARγ activity [25]. To further validate the link between GRb1 and activation of PPARγ/NF-κB pathway, the levels of PPARγ and phospho-NF-κB p65 in the brain, lung, and intestinal tissues of mice were measured by western blotting. As shown in Figure 7B & C, after cerebral I/R, the level of PPARγ was signi cantly decreased in the Model group compared with the Sham group, whereas PPARγ levels were signi cantly increased in the GRb1 group compared with those in the Model group (P < 0.05). In the brain, an increase in PPARγ expression was observed in the Model group compared with the Sham group, however, GRb1 administration also signi cantly increased the levels of PPARγ compared with those in the Model group (Fig. 7A). Stroke-induced NF-κB activation was markedly inhibited by GRb1 in the brain, lung and intestine compared with those in the Model group. Finally, the serum levels of in ammatory mediators, including TNF-α and IFN-γ, were measured to evaluate the severity of in ammation. The levels of TNF-α and IFN-γ were signi cantly lower in the GRb1 group than in the Model group (P < 0.05) (Fig. 7D).
3.8. The PPARγ inhibitor, GW9662, abolished the protective effects of GRb1 on the different barriers in MCAO/R mice To investigate the essential role of PPARγ in the three barriers, we administered GW9662, a PPARγ inhibitor. Compared to the Model+GRb1 group, the Model+GRb1+GW group showed that GW9662 treatment reduced the integrity of the various barriers, as evidenced by the reduced expression of junction proteins in the brain, lung, and intestine (Fig. 8A). In addition, GW9662 administration also reduced the levels of PPARγ, whereas those of phospho-NF-κB p65 were increased compared with the Model+GRb1 group (Fig. 8B).

Discussion
In this study, we presented evidence that the permeability of the brain, lung, and intestinal barriers increased after stroke, and that GRb1 could effectively mitigate the stroke-induced damage to the three barriers. We also demonstrated that GRb1 exerted these effects through the PPARγ/NF-κB signaling pathway, at least partly.
An increasing number of studies have shown that ischemic stroke causes a series of complications, including pneumonia and damage to the intestinal mucosa [38]. For instance, focal ischemic stroke has been associated with brain-lung crosstalk, leading to increased pulmonary damage and permeability of the pulmonary capillary membrane [39]. Additionally, intestinal mucosal barrier disturbance can be identi ed even in the early stages of cerebral ischemic stress. Necrosis occurs at the tip of the villi and the intestinal mucosa is damaged in the early stages of ischemic stroke, leading to gastrointestinal kinetic changes [40].
In our study, we found that both the pulmonary and intestinal barriers were simultaneously damaged in an experimental model of focal ischemic stroke. Several studies have demonstrated that GRb1 can be useful as a therapy for cardiovascular diseases [21], and exerts bene cial effects on the intestinal mucosal barrier [23] and lung in ammation [41]. However, its protective effects on brain/lung/intestinal barrier damage after ischemic stroke remain unclear.
In the present study, EB leakage was used to evaluate the integrity of the barriers of the three organs (brain, lung, and intestine). EB has a very high a nity for serum albumin, which cannot cross the blood barrier [42]. However, when a blood vessel is compromised, albumin, with bound EB, leaks out. Our ndings demonstrated that treatment with GRb1 reduced EB leakage and ameliorated histopathological damage in the brain, lung, and intestine after cerebral I/R (Figs 1C, 3A, and 5A).
In the brain, a physical paracellular barrier is created by the presence of TJs between endothelial cells. TJs are the luminal-most cellular junctions and are composed of claudins, occludins, and zona occludens [43].
TJs are more often studied in terms of their role in the BBB, where they have been found to induce high transendothelial electrical resistance and reduce paracellular diffusion to a greater extent than adhesion junctions [44]. In our study, the expression of ZO-1 and occludin in brain endothelial cells was signi cantly higher in the GRb1 group than in the Model group (Fig. 2).
In the lung, junction protein levels were reduced in mice with cerebral I/R. Adherens junctions play a major role in the regulation of pulmonary barrier permeability [45], while the transmembrane protein VE-cadherin is a central player in adherens junction formation and the regulation of pulmonary barrier integrity. Moreover, lung barrier injury has been associated with decreased mRNA and protein levels of claudin-5 [46]. In our study, the expression of VE-cadherin and claudin-5 in the lung was signi cantly increased in the GRb1 group compared with that in the Model group (Fig. 4A, B).
Patients with severe brain injury have been reported to experience reduced gastrointestinal motility. In our study, GRb1 administration led to a signi cant increase in mesenteric blood ow (Fig. 5D), indicating its potential e cacy for treating brain injury-induced gastrointestinal motility. Laboratory and clinical evidence have both indicated that the intestinal TJ-mediated barrier plays a critical role in the pathogenesis of intestinal and systemic diseases [47]. TJs encircle the apical ends of the lateral membranes of epithelial cells and determine the selectivity of their paracellular permeability to solutes [48,49]. In the current study, we showed that the levels of ZO-1 and occludin were signi cantly decreased after cerebral I/R, but increased after treatment with GRb1 or Eda. This indicated that GRb1 can help maintain TJ-mediated barrier integrity following stroke-induced barrier disruption (Fig. 6A, B).
Combined, these ndings implied that GRb1 could effectively mitigate brain, pulmonary and intestinal barrier dysfunction in mice with cerebral I/R.
There is clear evidence that human stroke causes multi-organ systemic disease through multiple mechanisms. After a stroke, the stimulation of the vagus nerve through nicotinic acetylcholine receptor α7 (nAChRα7) induces microglial activation, resulting in reduced alveolar macrophage phagocytic capability and leading to lung injury [50]. Meanwhile, systemic in ammation consequent to stroke leads to an increase in the release of in ammatory mediators such as interleukin (IL)-6 and TNF-α, resulting in lung in ammation and alveolar-capillary injury [51]. There exists bidirectional communication and interaction between the gut and the brain [52]. This link is thought to be based on the brain-gutmicrobiota axis, and involves the autonomic nervous system (ANS), the hypothalamic-pituitary-adrenal (HPA) axis, and the gastrointestinal tract [53]. Moreover, in ammatory molecules and cellular debris, the so-called DAMPs, are released from damaged tissue after a stroke. The release of DAMPs and cytokines, as well as the activation of the vagus nerve, induces gut dysmotility, gut dysbiosis, and increased gut permeability, resulting in the translocation of intestinal bacteria [54].
PPARγ is an important member of the nuclear receptor superfamily and predominantly expressed in adipocytes. However, PPARγ is also expressed in vascular cells, and data generated have shown the great importance of this receptor in vascular barrier integrity [55]. Studies have suggested that PPARγ activation can inhibit in ammation, thereby maintaining barrier integrity [56]. PPARγ may reduce proin ammatory phenotypes via suppression of the AP-1 and NF-κB pathways [57]. PPARγ agonists can also signi cantly reduce ischemia-triggered increases in cerebrovascular/BBB permeability in mice by suppression of proapoptotic microRNA-15a (miR-15a) [58]. Systematic administration of the PPARγ agonist pioglitazone was shown to ameliorate DSS-induced colitis by helping to repair the intestinal mucosal barrier. Moreover, pioglitazone could also inhibit ERK and NF-κB phosphorylation, as well as upregulate the expression of the tight junction proteins ZO-1 and claudin-5 in a PPARγ signalingdependent manner [59].
Increasing evidences have shown that PPARγ plays a upstream signal of NF-κB in the anti-in ammatory process [60]. The direct impact of PPARγ on NF-κB may be associated with its enzymatic properties. PPARγ is an E3 ubiquitin ligase, which cooperates with E2 UBCH3. PPARγ mediates the ubiquitination of Lys48 NF-κB p65, which leads to the proteolytic degradation of this NF-κB subunit [61]. Recent studies have reported that GRb1 binds to the PPARγ ligand-binding domain as determined by surface plasmon resonance, which suggests that GRb1 may be able to activate PPARγ [62], while polTIRFM analysis also indicated that GRb1 could bind to PPARγ [25]. GRb1 has also been shown to exert anticalci cation effects through the PPARγ/Wnt/β-catenin axis [63] and can promote the browning of 3T3-L1 adipocytes through the activation of PPARγ [64]. Based on these observations, we assessed the protein levels of PPARγ and phospho-NF-κB p65 in MCAO/R mice treated with GRb1. We found that GRb1 treatment upregulated PPARγ levels and inhibited those of phospho-NF-κB p65, with a concomitant reduction in the concentrations of proin ammatory factors, thereby helping to maintain the integrity of the brain, lung, and intestinal barriers. Moreover, we also found that these effects were suppressed by cotreatment with the PPARγ antagonist GW9662.
The vascular barrier maintains vascular and tissue homeostasis and modulates many physiological processes. The integrity of vascular barriers can be disrupted by a variety of soluble permeability factors, and changes in barrier function can result in multi-organ damage during disease progression [65].
Dysregulated vascular hyperpermeability can participate in the progression of many pathological states, such as rheumatoid arthritis, in ammatory bowel disease, asthma, acute lung injury, and cardiovascular diseases [66]. PPARγ expression is highest in brown and white adipose tissues, the large intestine, and immune cells, but can also be found in various other tissues, including the muscle, pancreas, liver, small intestine, and kidney [67]. Activators of PPARγ exert a broad spectrum of biological functions, such as regulating fatty acid metabolism, reducing in ammation, in uencing immune cell balance, inhibiting apoptosis and oxidative stress, and improving endothelial function. Such pleiotropic activity makes them interesting therapeutic targets for the treatment of various conditions [68]. Therefore, the protective effects elicited by GRb1 on barrier damage in multiple organs via PPARγ suggest that it may have potential as a therapeutic option for the treatment of vascular barrier-related diseases.

Conclusion
In summary, we demonstrated for the rst time that treatment with GRb1 can activate PPARγ, inhibit phospho-NF-κB p65 expression, and ameliorate cerebral, pulmonary, and intestinal barrier impairments in mice with MCAO/R. Our ndings demonstrated that GRb1 exerts bene cial effects on multi-organ damage after ischemic stroke and may serve as a potential clinically translatable therapy during recovery from stroke.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing Interest
The authors declare that they have no con icts of interest.  Quantitative analysis of EB extravasation by spectrophotometry. The data (n = 6) are expressed as means ± SD. ## P < 0.01 vs. the Sham group; * P < 0.05, ** P < 0.01 vs. the Model group.  The results of H&E staining of lung tissue shows congestion in thick alveolar septa and alveolar collapse (black arrowheads). Scale bars = 100 µm. The data (n = 6) are expressed as means ± SD. # P < 0.05, ## P < 0.01 vs. the Sham group; * P < 0.05, ** P < 0.01 vs. the Model group. The effects of Ginsenoside Rb1 (GRb1) on the expression of tight junction proteins in the intestine after cerebral ischemia/reperfusion (I/R). (A, B) ZO-1 (right) and occludin (left) are depicted in green and DAPIstained nuclei in blue. Scale bars = 20 µm. Representative western blots and quantitative analysis of the ratios of ZO-1 (C) and occludin (D). The data (n = 3) are expressed as means ± SD. # P < 0.05, ## P < 0.01 vs. the Sham group; * P < 0.05, ** P < 0.01 vs. the Model group.