Lipopolysaccharide (LPS) is one of the common pathogens that causes mesentery hyperpermeability- and intestinal edema-related diseases. This study evaluated whether ginsenoside Rb1 (Rb1), an ingredient of a Chinese medicine Panax ginseng, has beneficial effects on mesentery microvascular hyperpermeability induced by LPS and the underlying mechanisms. Male Wistar rats were continuously infused with LPS (5 mg·kg−1·h−1) via the left jugular vein for 90 min. In some rats, Rb1 (5 mg·kg−1·h−1) was administrated through the left jugular vein 30 min after LPS infusion. The dynamics of fluorescein isothiocynate-labeled albumin leakage from mesentery venules was assessed by intravital microscopy. Intestinal tissue edema was evaluated by hematoxylin and eosin staining. The number of caveolae in endothelial cells of microvessels was examined by electron microscopy. Confocal microscopy and Western blotting were applied to detect caveolin-1 (Cav-1) expression and phosphorylation, junction-related proteins, and concerning signaling proteins in intestinal tissues and human umbilical vein endothelial cells. LPS infusion evoked an increased albumin leakage from mesentery venules that was significantly ameliorated by Rb1 posttreatment. Mortality and intestinal edema around microvessels were also reduced by Rb1. Rb1 decreased caveolae number in endothelial cells of microvessels. Cav-1 expression and phosphorylation, VE-Cadherin phosphorylation, ZO-1 degradation, nuclear factor-κB (NF-κB) activation, and Src kinase phosphorylation were inhibited by Rb1. Rb1 ameliorated microvascular hyperpermeability after the onset of endotoxemia and improved intestinal edema through inhibiting caveolae formation and junction disruption, which was correlated to suppression of NF-κB and Src activation.
- nuclear factor-κB
- Panax ginseng
lipopolysaccharide (LPS) is one of the common pathogens that causes mesentery hyperpermeability- and intestinal edema-related diseases [diarrhea, cirrhosis with ascites, inflammatory bowel disease, etc. (1, 2, 31)]. Shedding of LPS from gram-negative bacteria into the circulation results in endotoxemia that incites a systemic, overexuberant immune response contributing to eluding effective therapy with 30–40% mortality (33). Research into the pathogenesis of endotoxemia has traditionally been focused on leukocytes and inflammation. However, recent studies showed that, although widespread inflammation continued, reduction in vascular permeability correlated significantly with increased survival in animal models of endotoxin exposure (25), suggesting that vascular hyperpermeability and tissue edema are causes rather than simply epiphenomena in pathogenesis of endotoxemia (22). However, there have been few clinical trials of agents designed to prevent or treat leaky vasculature so far (9).
It has been widely accepted that transport of plasma proteins and solutes across the endothelium involves two different routes, one is transcellular via caveolae-mediated vesicular transport, and the other is paracellular through interendothelial junctions (20, 27). Studies demonstrated that caveolae-mediated transendothelial transport (transcytosis) of macromolecules, such as albumin, is an important mechanism responsible for hyperpermeability during many pathological processes, particularly at the early stage of inflammatory response. Transcytosis is known to correlate with its scaffold protein caveolin-1 (Cav-1) expression and phosphorylation through Src tyrosine kinase and nuclear factor-κB (NF-κB) activation (11, 20, 26, 30). On the other hand, the paracellular pathway governing endothelium permeability is mediated by interendothelial cell junctions, with Cadherin and tight junction proteins as major contributors (27). Previous study suggested LPS opens the paracellular pathway in pulmonary vascular endothelia through protein tyrosine phosphorylation (10). Therefore, regulation of caveolae formation and paracellular junction to prevent microvascular hyperpermeability has become a potential new strategy for septicemia therapy. However, few clinical drugs are available that direct at hyperpermeability in endotoxemia therapy, although preclinical experiments for some new agents have been undertaken to test their roles in endothelial barrier protection (7, 25).
Ginsenoside Rb1 (Rb1) is one of the major active ingredients of Panax ginseng (PG), an extensively used traditional medicine in Asian countries. As one of the most active ingredients of ginseng, Rb1 has been reported to have a variety of pharmacological effects, including anti-inflammation, anti-apoptosis, anti-oxidation, increasing endothelial cell nitric oxide production, and inhibiting angiogenesis, mainly through association to androgen or estrogen receptors (5, 14, 17, 23, 43). Our previous study has shown that Rb1 pretreatment inhibits LPS-induced leukocyte adhesion to microvascular wall and interstitial mast cell degranulation (39), which are implicated in vascular hyperpermeability (19, 38), suggesting a protective role of Rb1 against vascular protein leakage. Rb1 and its metabolite compound K have been reported to abrogate LPS-induced expression of proinflammatory cytokines in macrophage through inhibiting the interleukin-1 receptor-associated kinase-1 signaling pathway (18). Our recent study has demonstrated that pretreatment with Yiqifumai injection (a traditional Chinese medicine preparation containing PG as a major component) attenuates LPS-induced microcirculatory disturbances and hyperpermeability in mesenteric microvessels (44). However, the questions that remain are: 1) whether Rb1 has beneficial effects on albumin leakage at the early stage of endotoxemia and 2) what is the role of the transcellular and paracellular pathway in the beneficial actions of Rb1 on microvascular hyperpermeability. The present study was designed to answer these questions.
MATERIALS AND METHODS
Chromatographically pure Rb1 was purchased from Feng-Shan-Jian Medical (Kun Ming, Yunnan, China), and its structure is shown in Fig. 1A. LPS (Escherichia coli serotype O55:B5) and fluorescein isothiocynate (FITC)-conjugated bovine serum albumin (FITC-BSA) were purchased from Sigma Chemical (St. Louis, MO). Endothelial cell medium (ECM) was from ScienCell Research Laboratories (Carlsbad, CA). Antibodies against Cav-1, Src, phospho-Src, I-κBα, phospho-I-κBα, NF-κB p65, β-actin, GAPDH, and histone H3 were obtained from Cell Signaling Technology (Beverly, MA). Antibodies against phospho-VE-Cadherin and ZO-1 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Antibody against phospho-Cav-1 was obtained from BD Transduction Laboratories (San Jose, CA).
Male Wistar rats weighing 200–250 g were obtained from the Animal Center of Peking University Health Science Center. The certificate code of these animals was SCXK 2006–0008. The animals were given standard laboratory rodent chow (Animal Center of All Animals Peking University Health Science Center) and tap water. They were housed at 24 ± 1°C and relative humidity of 50 ± 1% with a 12:12-h light-dark cycle. The animals were fasted for 12 h before the experiment but allowed free access to water. All animals were handled according to the guidelines of the Peking University Animal Research Committee, and the surgical procedures and experimental protocol were approved by the Committee on the Ethics of Animal Experiments of the Health Science Center of Peking University (LA2011–38).
Human umbilical vein endothelial cells (HUVECs) were obtained from ScienCell Research Laboratories. The cells were plated at a density of 1 × 105 cells/cm2 on fibronectin-coated dishes and cultured in complete ECM supplemented with Supplement Mix (ScienCell Research Laboratories) containing fetal bovine serum, endothelial cell growth supplement, and penicillin/streptomysin solution. The cells were incubated (37°C) under a humidified atmosphere of 5% CO2 and 95% air and used at passages 5–6.
The rats were randomly divided into four groups, 26 animals in each. The number of animals for assessment of various parameters in each group is detailed in Table 1. After being anesthetized with urethane (2 g/kg body wt im), the left femoral vein and left jugular vein of the rat were cannulated. In the LPS group, LPS solution in saline was infused (5 mg·kg−1·h−1) for 90 min via the left femoral vein, as previously described (13, 44). The vehicle, instead of LPS solution, was administrated in Sham and Rb1 alone (Rb1) groups. In the Rb1 posttreatment (LPS + Rb1) group, Rb1 solution was infused continuously through the left jugular vein 30 min after LPS administration at the dose of 5 mg/kg according to our previous study (39). Rb1 solution and the same volume of saline were infused in Rb1 and Sham groups, respectively, without subsequent LPS administration. In a separate set of experiments, the rats were anesthetized with 2% penobarbital sodium (60 mg/kg body wt ip), and saline, LPS, and Rb1 were administered in corresponding groups in the same way as above. After recovery from anesthesia, the animals were allowed access to water and rodent chow, and survival rate was recorded over time until 4 days after LPS stimulation (32).
HUVECs were cultured to confluence and incubated with LPS (100 ng/ml) in complete ECM for 90 min (30). In the Rb1 posttreatment group, Rb1 was added 30 min after LPS stimulation to a concentration of 100 μM, which was equivalent to the Rb1 concentration applied in the aforementioned in vivo experiments, and showed no cell cytotoxicity in a pilot experiment.
FITC-BSA was used to evaluate the albumin leakage across the mesenteric venular wall by virtue of inverted intravital microscopy (DM-IRB; Leica, Mannheim, Germany) as described previously (12). In brief, after anesthesia, the rat abdomen was opened via a middle incision (2–3 cm long). The ileocecal portion of the mesentery, 10- to 15-cm region of caudal mesentery, was exteriorized and mounted on a transparent plastic stage. The mesentery was kept warm and moist by continuous superfusion with saline solution at 37°C. The animals were intravenously injected with 50 mg/kg of FITC-BSA. After 10 min basal observation, fluorescence intensity of FITC-BSA was recorded (excitation, 455 nm; emission, 530 nm) at a 10-min interval using a silicon-intensified target camera (C-2400-08; Hamamatsu, Shizuoka, Japan) and a DVD videocassette recorder (DVR-R25; Malata, Xiamen, China), and the fluorescent intensity in the venules (Iv) and in the perivenular interstitium (Ip) was measured with Image-Pro Plus 5.0 software (Bethesda, MD). Albumin leakage was estimated by dividing Ip by Iv, and the ratio of albumin leakage at a time point to that of the baseline was designated as the ratio of albumin leakage at that point.
Assessment of cytokines concentration in plasma.
The time courses of plasma TNF-α and IL-6 were measured at baseline and 30, 60, and 90 min after LPS challenge. For this purpose, whole blood was taken from rat abdominal aorta at each time point. The concentration of TNF-α was assessed by flow cytometry with a BD cytometric bead array kit (BD Biosciences, San Jose, CA) as previously described (45). Briefly, the blood was centrifuged (1,300 g, 4°C) for 10 min, and supernatants were collected as plasma. Bead (50 μl) was added to 50 μl plasma and incubated at room temperature in the dark for 1 h for bead capture. Phycoerthrin-labeled antibody (50 μl) was then added and incubated at room temperature for 2 h to form a “sandwich” complex. After incubation, the samples were washed thoroughly with 1 ml washing buffer. The mean fluorescence intensity of TNF-α was measured by flow cytometry (FACS Calibur; BD Biosciences), and the data were analyzed by the BD cytometric bead array analysis software. The concentration of IL-6 was measured by using a IL-6 ELISA kit according to the manufacturer's instruction (Abcam).
The animals were killed 90 min after LPS infusion, and the middle section of small intestine and the middle lobe of right lung were excised and rinsed in saline and fixed with 4% paraformaldehyde in 0.01 M PBS (pH 7.4). The tissues were cut into blocks, embedded in paraffin, and sectioned to 5-μm sections. The sections were stained by hematoxylin and eosin. The images were captured by a digital camera connected to a microscope (BX512DP70; Olympus, Tokyo, Japan).
The rat mesentery in each experimental condition was harvested after LPS stimulation for examination by electron microscopy. Briefly, after superficially dripping 3% glutaraldehyde in 0.1 M PBS for 20 min, the mesentery was cut into blocks smaller than 1 mm3 and further fixed by immersing in the same fixative for 1 h at room temperature and then overnight at 4°C. After being rinsed, the tissues were postfixed with 1% osmium tetroxide in 0.1 M PBS for 2 h at 4°C, dehydrated, and then embedded in Epon 812. Ultrathin sections were stained with uranyl acetate and lead citrate and observed with a transmission electron microscope (JEM 1230; JEOL, Tokyo, Japan).
Intestinal fresh frozen sections 8 μm thick were prepared using a cryostat (CM1800; Leica, Bensheim, Germany). After completely air drying, sections were treated with 0.01 M sodium citrate for antigen retrieval, washed by PBS, and permeabilized with 0.3% Triton X-100 for 30 min. Following blocking with goat serum at room temperature for 30 min, sections were incubated with anti-Cav-1 antibody diluted in PBS overnight at 4°C. After being rinsed with PBS, sections were incubated with Dylight 488-labeled secondary antibodies (KPL, Gaithersburg, MD) for 2 h at room temperature. Cav-1 expression, Cav-1 phosphorylation, cytoplasmic I-κBα degradation, and NF-κB p65 nuclear translocation in HUVECs were assessed by immunofluorescence staining. For this purpose, HUVECs were fixed with 4% paraformaldehyde for 20 min and permeabilized with 0.3% Triton X-100 for 30 min. After blocking with horse serum at room temperature for 30 min, cells were incubated overnight at 4°C with anti-Cav-1 (1:200), anti-phospho-Cav-1 (1:50), anti-I-κBα (1:100), and anti-NF-κB p65 (1:100) antibodies in 0.01 M PBS containing 3% Triton X-100. After three washes in PBS, cells were incubated for 1.5 h at 37°C with Dylight 488-labeled secondary antibodies. Both intestinal tissues and HUVECs were counterstained with Hoechst 33342 for nuclei. Immunostainings performed without primary antibodies served as negative controls. Images were acquired using a laser scanning confocal microscope (TCS SP5; Leica).
Western blotting analysis.
Intestinal tissues and HUVECs were removed and harvested, respectively, 90 min after LPS stimulation. Whole protein of tissues and cells was homogenized in lysis buffer containing the protease inhibitor. Cytoplasmic and nuclear protein of cells was extracted by NE-PER nuclear and cytoplasmic extraction reagents kit (Thermo Scientific) according to the manufacturer's instruction. An equivalent amount of protein was loaded on Tris-glycine SDS-PAGE for separation. The membranes were incubated overnight at 4°C with antibodies against Cav-1 (1:5,000), phospho-Cav-1 (1:1,000), ZO-1 (1:1,000), phospho-VE-Cadherin (1: 1,000), Src (1:5,000), phospho-Src (1:1,000), I-κBα (1:1,000), phospho-I-κBα (1:1,000), NF-κB p65 (1:1,000), β-actin (1:5,000), GAPDH (1:5,000), and histone H3 (1:1,000) in diluent buffer (3% BSA in Tris-buffered saline). Following washing, the membranes were incubated with the horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology) at room temperature for 1 h. Blots were developed using the enhanced ChemiLucent Detection System Kit (Millipore, Temecula, CA), and protein bands were visualized on X-ray film. Semiquantitation of the protein was performed using Image-Pro Plus 5.0 software. The β-actin or GADPH and histone H3 Western blotting was performed for each membrane as a loading control for cytoplasm and nucleus, respectively.
All parameters were expressed as means ± SD. For comparison of more than two conditions, a one-way ANOVA with Turkey's post hoc test or a repeated-measures ANOVA with Bonferroni post hoc test was used. For survival rate analysis, Kaplan-Meier analysis was used, followed by a Log-rank test. A probability <0.05 was considered to be statistically significant.
Rb1 posttreatment decreases rat mortality after LPS challenge.
The survival rate of rats in different groups was calculated over 4 days. As shown in Fig. 1B, LPS challenge for 90 min killed rats starting from 12 h, with a survival rate in the LPS group by day 4 reaching to 50%. Treatment with Rb1 for 30 min after LPS infusion increased the survival rate to 90% 4 days after stimulation, being significantly higher than that for the LPS group. Rb1 alone had no influence on animal survival rate.
Rb1 ameliorates albumin leakage from mesenteric microvessels in vivo.
To investigate the amelioration effect of Rb1 on microvascular hyperpermeability during endotoxemia, dynamics of FITC-BSA leakage out of mesenteric venules was assessed in vivo by inverted intravital microscopy. As shown in Fig. 2, the albumin leakage was undetectable in all groups before LPS infusion (Fig. 2A, a1–d1), and maintained undetectable over the entire observation period in Sham and Rb1 groups (Fig. 2A, a2, a3, b2, and b3). In contrast, the albumin leakage was observed apparently in both LPS and LPS + Rb1 groups 30 min after LPS infusion (Fig. 2A, c2 and d2), and further increased strikingly in the LPS group at 90 min (Fig. 2A, c3). Surprisingly, Rb1 posttreatment ameliorated albumin leakage even after the initiation of microvascular hyperpermeability (Fig. 2A, d3). The quantitative evaluation of the results was presented as a percentage of albumin leakage changed with time (Fig. 2B), which confirmed the results from image survey.
Rb1 inhibits the increase in TNF-α and IL-6 level in plasma after LPS.
In view of the importance of proinflammatory cytokine in the initiation and progression of inflammation, the time course of plasma levels of TNF-α and IL-6 was determined at 30, 60, and 90 min after LPS challenge. TNF-α and IL-6 in plasma of Sham and Rb1 groups were hardly detectable. In the LPS group, TNF-α increased significantly and linearly from 30 min after LPS infusion until the end of observation, whereas IL-6 upregulated only at 90 min after LPS infusion. Posttreatment with Rb1 attenuated IL-6 production (Fig. 3B) but had no effect on TNF-α increase at any time point assessed (Fig. 3A).
Rb1 mitigates histological changes in intestine and lung tissues.
Histology study performed on intestine revealed obvious changes in tissue 90 min after LPS challenge, including disintegration of intestinal villi and edema in mucosa (Fig. 4A, c2) and submucosa (Fig. 4A, c3) surrounding microvessels. LPS evoked lung edema around microvessels in pulmonary interstitium as well (Fig. 4B, c2). Interestingly, tissue edema in intestine and lungs was mitigated by Rb1 posttreatment evidently (Figs. 4A, d2, d3 and Fig. 4B, d2), indicating that amelioration effects of Rb1 on microvascular hyperpermeability during endotoxemia is systemic and independent on tissue.
Rb1 posttreatment decreases the number of caveolae in endothelial cells of mesentery microvessels.
Because transcellular trafficking for macromolecules such as albumin is carried out via vesicles, we assessed the effect of Rb1 on caveolae formation in endothelial cells of mesenteric microvessels using electron microscopy. Figure 5 presents the electron micrographs of rat mesenteric microvessels in each group. In the Sham (a1 and a2) and Rb1 (b1 and b2) groups, venules were lined by a layer of endothelial cells that exhibited a smooth inner face with occasionally occurring caveolae in the cytoplasm. At 90 min after LPS infusion, however, numerous caveolae emerged in the endothelial cells of mesenteric microvessel (Fig. 5, c1 and c2). In contrast, caveolae number decreased significantly in the Rb1 posttreatment group (Fig. 5, d1 and d2).
Rb1 decreases caveolae formation via suppressing Cav-1 expression and phosphorylation.
Cav-1 is known as a principal protein of caveolae that regulates the formation of vesicles in cytoplasm. With the use of confocal microscopy, the results of the present study showed that, in intestinal tissues, Cav-1 expression on both mucous and submucous layers enhanced impressively after LPS stimulation (Fig. 6A, c1–c3) compared with the Sham group (Fig. 6A, a1–a3), which was abrogated by Rb1 posttreatment apparently (Fig. 6A, d1–d3). Given the critical role of Cav-1 phosphorylation in vascular transportation, we conducted a Western blotting analysis for Cav-1 phosphorylation in different groups and revealed that LPS elicited a significant increase in the phosphorylation of Cav-1 in intestinal tissues compared with the Sham group, which was abrogated by posttreatment with Rb1 remarkably (Fig. 6B).
Rb1 inhibits Cav-1 expression and phosphorylation in HUVECs.
Consistent with in vivo results, LPS increased Cav-1 expression in HUVECs as well, which was inhibited by Rb1 administered 30 min after LPS stimulation (Figs. 7A, a–d, and Figs. 7B). Furthermore, both confocal and Western blotting revealed that phosphorylation of Cav-1 at tyrosine 14 also increased significantly in HUVECs after LPS exposure compared with the control group, and Rb1 posttreatment ameliorated Cav-1 phosphorylation as well (Figs. 7A, e–h and C).
Rb1 inhibits VE-Cadherin phosphorylation and ZO-1 degradation in HUVECs.
Because interendothelial cell junctions regulate endothelial permeability as well in addition to caveolae, we further detected changes in interendothelial junction proteins VE-Cadherin and ZO-1, which are the principal member of Cadherin and tight junction proteins, respectively. Western blotting analysis results showed that LPS evoked a significant VE-Cadherin phosphorylation and ZO-1 degradation in HUVECs, which was abrogated by Rb1 posttreatment (Fig. 8, A and B). Taken together, the aforementioned results suggest that Rb1 regulate microvascular hyperpermeability by both transcellular and paracellular pathways.
Rb1 decreases Cav-1 expression by inhibiting NF-κB activation.
Study has shown that NF-κB activation is involved in Cav-1 expression after LPS exposure. To further clarify the mechanisms underlying the amelioration effect of Rb1 on Cav-1 expression, we assessed the role of the NF-κB signaling pathway in Rb1-decreased Cav-1 expression. As expected, LPS led to a redistribution of NF-κB p65 compared with control groups, resulting in a dominant nuclear localization (Fig. 9A, c3). Noticeably, posttreatment with Rb1 inhibited LPS-induced NF-κB p65 translocation from cytoplasm to nuclei (Fig. 9A, d3). A quantitative Western blot analysis of NF-κB p65 in cytoplasmic and nuclear fractions of HUVECs was performed, showing a result consistent with the finding above (Fig. 9B). NF-κB activation is controlled by I-κBα. Thus we further evaluated the effect of Rb1 on I-κBα phosphorylation and revealed an increase in I-κBα phosphorylation after LPS that was attenuated by treatment with Rb1 (Fig. 10B). Confocal microscopy further confirmed the results described above, showing that I-κBα degraded in HUVEC cytoplasm after LPS stimulation, which was reversed by Rb1 obviously (Fig. 10A). Taken together, Rb1 inhibited Cav-1 expression and concurrently suppressed I-κBα-dependent NF-κB signaling activation.
Requirement of Src activation for suppression effects of Rb1 on Cav-1, I-κBα, and VE-Cadherin phosphorylation and ZO-1 degradation.
Because Src kinase activation is involved in regulation of both transcellular and paracellular pathways, we thus addressed whether Src activation is implicated in the suppression effects of Rb1 on Cav-1 and VE-Cadherin phosphorylation and ZO-1 degradation. In HUVECs, LPS significantly increased Src phosphorylation at tyrosine 416, whereas posttreatment with Rb1 prevented this response, a result similar to that by application of Src inhibitor 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine (PP2) (Fig. 11A). PP2 was also observed in the present study to inhibit LPS-provoked phosphorylation of Cav-1 (Fig. 11B), I-κBα (Fig. 11C), and VE-Cadherin (Fig. 11D) and ZO-1 degradation (Fig. 11E), highlighting a likely identical target for PP2 and Rb1.
The present study showed that posttreatment with ginsenoside Rb1 ameliorates LPS-induced albumin leakage from mesenteric venules, along with a reduction in plasma cytokine, intestinal edema, and mortality. Moreover, in response to administration of Rb1, a decrease was observed after LPS in the number of vesicles in endothelial cells of microvessels, Cav-1 expression and phosphorylation, I-κBα expression, NF-κB p65 nuclear translocation and VE-Cadherin phosphorylation, and ZO-1 degradation, implying both transcellular and paracellular routes being at work and an involvement of the NF-κB signaling pathway in Rb1 action. The resemblance of Rb1 and PP2 in blockage of Src tyrosine kinase suggests Src as the target for Rb1 to exert its effect.
Recent breakthroughs indicate that endotoxemia is characterized by diffuse microvascular leak and tissue edema both in animals and humans (3, 9), and experimental data demonstrate that preventing endothelial hyperpermeability per se could reduce animal mortality (25). However, few specific therapeutic drugs have been developed directed at microvascular hyperpermeability. In the present study, we provide compelling experimental evidence supporting the efficacy of Rb1 therapy for mesentery albumin leakage induced by LPS. Previous studies in our laboratory have demonstrated that LPS intravenous infusion at a concentration of 5 mg/kg effectively evoked hyperpermeability in microvessels (13, 44). Consistent with our previous studies, with the use of inverted intravital microscopy, the present study revealed that FITC-labeled albumin leaked markedly out of mesenteric microvessel 30 min after LPS infusion and increased linearly with time. Excitingly, the present study found that Rb1 posttreatment worked even given 30 min after LPS infusion, a time when LPS-induced microvessel hyperpermeability has initiated, which makes the finding clinically relevant.
It is well accepted that microvascular endothelium permeability is regulated by the paracellular pathway via intercellular junctions and the transcellular pathway mediated by caveolae. However, the results so far are inconsistent as to the role of the two pathways in the LPS-induced increase in endothelium permeability. Studies have shown that opening of the paracellular pathway is responsible for vascular leakage after LPS challenge (27). The disruption of endothelium integrity is reported to increase pulmonary micovessel permeability, leading to protein leakage and permeability edema in LPS-induced tissue injury in vivo (4). In agreement with these reports, we found that, in HUVECs, VE-Cadherin phosphorylation and ZO-1 degradation increased significantly after LPS stimulation. Interestingly, Rb1 posttreatment significantly reversed the aforementioned changes, implying the involvement of junction proteins in the regulation of microvascular permeability by Rb1.
On the other hand, reports have provided evidence that LPS mediates NF-κB-dependent Cav-1 expression that results in increased caveolae number and thereby contributes to the mechanism of increased transendothelial albumin permeability in human lung microvascular endothelial cells (41). In addition, recent studies also suggested Cerebralcare Granule, a compound Chinese medicine, reduced brain edema partly by inhibiting the expression of Cav-1 expression in cerebral endothelial cells (16), consistent with another study showing that Cav-1 was involved in BBB disruption in early stroke stages (24). In contrast, some other studies using Cav-1 null mice or endothelial cells demonstrated increased permeability to albumin resulting from opening of the paracellular route for albumin transport (29, 37), implying that Cav-1 is important in some cases while not indispensible in other cases in regulating microvascular hyperpermeability. The present study revealed that LPS induced a marked increase in Cav-1 expression and the number of caveolae in endothelial cells, indicating a role of caveolae-mediated albumin transport in microvascular hyperpermeability in the present condition. More importantly, Rb1 posttreatment significantly decreased the number of caveolae in endothelial cells in the presence of LPS, suggesting that Rb1 ameliorates albumin leakage at least in part via inhibition of caveolae formation.
Cav-1 is a structural protein of caveolae, and the level of Cav-1 expression correlates with the number of caveolae (6, 11). Study has shown that exogenous expression of Cav-1 in Cav-1 null cells induced the formation of caveolae (8). NF-κB activation is involved in Cav-1 mRNA transcription and protein overexpression after LPS exposure (41). In the present study, posttreatment with Rb1 inhibited LPS-induced NF-κB p65 nuclear translocation, I-κBα phosphorylation, and I-κBα degradation in HUVECs, suggesting that amelioration effects of Rb1 on caveolae-mediated albumin transcytosis were attributable to inhibition of NF-κB-dependent Cav-1 expression. In addition to Cav-1, NF-κB activation after LPS stimulation leads to expression of proinflammatory cytokines, such as TNF-α and IL-6, which are critical for initiating, amplifying organ injury. The persistent elevation of proinflammatory cytokines in the serum of patients with ALI has been reported to associate with increased mortality (26). In the present study, the time course of serum TNF-α and IL-6 level was detected. In line with microvascular hyperpermeability and tissue edema, the results suggested that both cytokines increased obviously, being significant from 30 and 90 min after LPS infusion, respectively. Rb1 had no inhibition effect on TNF-α production at any time point examined, which may account for the incomplete relief of microvascular hyperpermeability and tissue edema by Rb1. However, Rb1 was observed to inhibit the LPS-evoked production of IL-6 in serum, consistent with others showing that Rb1 and its metabolite compound K could abrogate LPS-induced expression of proinflammatory cytokines in macrophages (18), which may be attributable to its inhibition effect on NF-κB activation in endothelial cells.
Previous study demonstrated that tyrosine kinases are rapidly activated after LPS stimulation and related to vascular hyperpermeability (3, 35). Particularly, Src-mediated phosphorylation of Cav-1 at tyrosine 14 initiates plasmalemmal vesicle fission and transendothelial vesicular transport (21). Cav-1 phosphorylation has been reported to play a key role in albumin transcytosis in many inflammatory-related pathological processes, most of which are dependent on protein tyrosine kinase activation (15, 40, 42). On the other hand, it was reported that LPS reduces ZO-1 level and disrupts the tight junction of the epithelial monolayer by a Src kianse-dependent mechanism (36). In addition, study showed that LPS induced increment in tyrosine phosphorylation of VE-Cadherin and paracellular permeability in human lung microvascular endothelial cells (4), which was also abrogated by the Src kinase inhibitor PP2, implying the critical role of Src kinase activation in the regulation of paracellular permeability. In the present study, we found that, after LPS stimulation, Cav-1 phosphorylation as well as VE-Cadherin phosphorylation and ZO-1 degradation increased markedly in intestinal tissues or endothelial cells, which concurred with an increase in phosphorylation level of Src kinase in HUVECs. Impressively, Rb1 treatment significantly inhibited degradation of ZO-1 and phosphorylation of Cav-1, VE-Cadherin, and Src, an effect similar to that caused by the Src kinase inhibitor PP2, suggesting that Src kinase may participate in the regulation of Rb1 on both the transcellular and paracellular pathway. The mechanisms whereby Rb1 inhibits Src kinase, however, need further investigation.
A limitation of the present study is that the notion regarding the contribution of the transcellular and paracellular pathways to modulation of vascular permeability was based on the study of rat mesenteric and intestinal microvessels and HUVECs exposed to LPS; whether or not this notion can be applied to other vascular beds or other stimulation needs further investigations.
In summary, the present study showed that Rb1 posttreatment ameliorated mesenteric microvascular hyperpermeability induced by LPS, improving intestinal edema. This effect of Rb1 was correlated to inhibition of Cav-1 expression and phosphorylation, VE-Cadherin phosphorylation, and ZO-1 degradation via suppressing I-κBα degradation, NF-κB p65 nuclear translocation, and Src kinase activation. Rb1 was also found to inhibit the LPS-evoked production of IL-6 in serum. As an overall outcome, Rb1 improved the survival rate for the rats challenged by LPS. These results provide scientific evidence for Rb1 as a potential therapeutic strategy for LPS-induced mesentery hyperpermeability and intestinal edema-related diseases.
This study was supported financially by the Project H201-81072909 from the National Natural Science Foundation of China.
The authors declare no conflict of interest.
Author contributions: Y.Z., K.S., Y.-Y.L., Y.-P.Z., B.-H.H., X.C., L.Y., C.-S.P., Q.L., K.H., X.-W.M., L.T., and C.-S.W. performed experiments; Y.Z., K.S., Y.-Y.L., Y.-P.Z., B.-H.H., X.C., L.Y., C.-S.P., Q.L., K.H., X.-W.M., L.T., C.-S.W., and J.-Y.H. analyzed data; Y.Z., K.S., Y.-Y.L., Y.-P.Z., B.-H.H., X.C., L.Y., C.-S.P., Q.L., K.H., X.-W.M., L.T., C.-S.W., and J.-Y.H. interpreted results of experiments; Y.Z., K.S., and Y.-P.Z. prepared figures; Y.Z., K.S., and J.-Y.F. drafted manuscript; Y.Z., K.S., J.-Y.F., and J.-Y.H. edited and revised manuscript; J.-Y.H. conception and design of research; J.-Y.H. approved final version of manuscript.
- Copyright © 2014 the American Physiological Society