1 INTRODUCTION

Sepsis is a common complication of severe infection, severe trauma, burns, shock, and surgery, which can lead to septic shock, namely multiple organ dysfunction syndrome.1 It is estimated to have an incidence of 31.5 million worldwide, including 19.4 million patients with severe sepsis, with an annual increase of 1.5% and a mortality rate of 25%–30% in recent years.2 Sepsis is confirmed to be accompanied with the presence of bacteria or a highly suspicious focus of infection.3 Inflammation plays an important role during sepsis, and excessive inflammation can result in organ damage, chronic inflammation, fibrosis, and scarring. Severe inflammation can be deleterious, resulting in multiple organ dysfunction, including acute lung injury (ALI). ALI is the most common organ injury in sepsis and causes severe lung inflammation.

Emerging evidence suggests that emodin exhibits many pharmacological effects, including anticancer, antivirus, anti-inflammatory, antibacterial, and hepatoprotective activities.4 In a study on lung cells, emodin could inhibit the proliferation of CH27 cells through the Bax and Fas death pathways.4 Furthermore, emodin was proved to attenuate sepsis-induced ALI in a cecal-ligation and puncture rat model.5 Emodin had a protective effect on intestinal epithelial tight junction barrier integrity in rats with sepsis induced by cecal ligation and puncture (CLP).6 However, the specific molecular mechanism has not been found yet.

Silent information regulator 1 (SIRT1) is closely related to the occurrence and development of the inflammatory response, in which nuclear factor (NF)-κB plays a key role. High mobility group box 1 (HMGB1) is critical in the pathogenesis of sepsis.7-9 HMGB1 is stored in the nucleus. Bacteria lipopolysaccharide (LPS) can cause HMGB1 acetylation, resulting in the localization of the protein to the cytosol.10 In addition, we validated that this mechanism might be related to NF-κB.11-13 NF-κB is a critical transcription factor for the maximal expression of numerous cytokines involved in the pathogenesis of ALI.14 One study showed that LPS-induced pulmonic injury was associated with the increased expression of inflammatory factors in an ARDS model. Metformin could target SIRT1 to increase its expression while suppressing the expression of p-NF-κB, thus alleviating ARDS.15 Another study indicated that SIRT1 antagonist nicotinamide (NAM) attenuated the inhibitory effects of AngsiRNA on LPS-induced NF-kB and p65 expression. This suggests that silencing angptl4 is protective against LPS-induced ALI via regulating the SIRT1/NF-kB signaling pathway.16 A study demonstrated that parenteral nutrition could increase the apoptosis level and activate the inflammatory HMGB1/RAGE/NF-kB signaling pathway.17 The current study aimed to investigate the specific mechanism of emodin by constructing in vivo and in vitro septic lung injury models via inhibition and reduction of the NF-kB and HMGB1 pathways.

2 METHODS 2.1 Animal study

Six-week-old adult male Sprague–Dawley rats (200 ± 20 g) were provided by the Experimental Animal Center of Changzhou No. 2 People's Hospital Affiliated to Nanjing Medical University (license No. SCXK (Su) 2016–0015). The rats were housed with a regular diet for 2 weeks.

2.2 CLP model

Rats were subjected to fasting for 12 h before operation. The rats were thoroughly anesthetized, and the abdomen was shaved and disinfected. A 1 cm long incision was cut into the abdomen of the anesthetized rats. The cecum was ligated below the ileocecal valve, and a 21-gauge needle was used for puncture, from mesenteric toward antimesenteric direction beneath ligation. After removing the needle, the bottom half of the cecum was gently squeezed until a droplet of intestinal contents spilled out of the puncture.18 The cecum was then relocated into the peritoneal cavity, and the abdominal incision was closed in two layers. The cecum was returned, and the abdominal cavity was closed. Dead animals were complemented in time during the observation period.2 All rats were returned to their cages and given food and water for 24 h. The lung samples were obtained 24 h after CLP.

2.3 Groups

The rats were divided into three groups randomly: SHAM group, CLP group, and CLP with emodin pretreated group (CLP + emodin) (n = 8 for each). The rats in CLP + emodin group received emodin (35 mg kg−1 day−1) in sodium carboxymethyl cellulose suspension by gavage for the following 5 days2 after CLP model formation every 24 h. The rats in the other two groups received the same amount of 0.5% sodium carboxymethyl cellulose solution by gavage, and they were prepared as follows 2 h after the last gavage.10 The animal experiments were approved by the Ethics Committees of Changzhou No. 2 People's Hospital Affiliated to Nanjing Medical University. All animals were treated humanely and with regard for alleviation of suffering.

2.4 Cell culture

Murine alveolar epithelial cell line (MLE-12) cells were purchased from American Type Culture Collection (ATCC, Manassas, VA). MLE-12 cells were seeded in Roswell Park Memorial Institute (RPMI)-1640 medium with 10% fetal bovine serum at 37°C with 5% CO2. The cells in logarithmic stage were transfected in the following experiment. The si-SIRT1 group was transfected with si-SIRT1 vector according to the instructions, while the non-transfected group was transfected with the empty vector. The cells were divided into three groups: control group, LPS group, and LPS with emodin pretreated group (LPS + emodin). Cells in LPS group were incubated with 1 mg/ml LPS (Solarbio, Beijing, China) for 24 h. Cells without stimulation of LPS were regarded as the control for the cellular sepsis model. The cells in the LPS + emodin group were treated with emodin for another 6 h.

2.5 Cytokine analysis

Tissue sample homogenates were taken and ground in liquid nitrogen. Concentrations of interleukin (IL)-1, IL-6, and tumor necrosis factor (TNF)-α in bronchoalveolar lavage fluid (BALF) were determined using commercially available ELISA kits for mouse cytokines (R&D Systems, USA) according to the instructions of the manufacturer.

2.6 Hematoxylin and eosin staining

The right lung inferior lobes obtained from rats were fixed with 10% formalin for 24 h and embedded in paraffin. Then, they were cut into 4 μm thick sections. After deparaffinization and hydration, the sections were stained with hematoxylin and eosin and observed under light microscopy (Nikon E400, Nikon Instrument Group, Japan). The lung injury score was calculated based on Smith's biopsy score.

2.7 Myeloperoxidase activity determination

Myeloperoxidase (MPO) activity of the lung tissue was detected by using the MPO kit (Jiancheng Bioengineering Institute, Nanjing, China) according to manufacturer's instructions. The absorbance was read using 450 nm wavelength in a spectrophotometer.

2.8 Flow cytometry

The apoptosis rate of MLE-12 cells was detected by flow cytometry using Annexin V-fluorescein isothiocyanate (V-FITC)/PI kit (Beyotime). Annexin V-FITC-positive and propidium iodide (PI)-negative cells were considered at early apoptotic phase, while Annexin V-FITC-positive and PI-positive ones were considered at late apoptotic phase. In this study, the apoptotic rate was shown as the percentage of cells at early and late apoptotic phases.

2.9 Western blot analysis

Lung tissue samples were collected at 24 h after CLP operation. Lung tissue samples and MLE-12 cells were lysed in automatic homogenizer in chilled RIPA Lysis buffer (Beyotime, Shanghai, China) on ice. The supernatant was retained to detect its protein concentration after high-speed centrifugation. Equivalent protein samples with buffer in each hole were loaded with 10% sodium dodecylsulfate–polyacrylamide gel electrophoresis (SDS-PAGE, Sigma, American) and transferred to a polyvinylidene fluoride (PVDF, Sigma, American) membrane when the objective protein was half dried. Then, the membrane was blocked in 5% defatted milk for more than 1 h and incubated with the specific primary antibody. The primary antibodies (rabbit monoclonal antibodies) were added as follows: anti-NF-κB (1:1000), anti-p-NF-κB (1:1000), anti-SIRT1 (1:1000), anti-p65, anti-p-p65, anti-HMGB, and anti-GAPDH (1:1000). Next, the membrane was incubated with the second antibody (1:5000) for 1 h. Anti-mouse or anti-rabbit antibodies against IgG conjugated with horseradish peroxidase were adopted as the secondary antibodies. The protein bands were visualized by using an enhanced chemiluminescence (ECL, Solarbio, Beijing, China) kit. Blots were repeated at least three times for every condition. After development, the band intensities were quantified using the Image-pro Plus 6.0 analysis software.

2.10 Statistical analysis

The data were analyzed using SPSS 20 software. Data are presented as the mean ± standard deviation (SD) with descriptive statistics. The n was equal to the number of animals in at least three independent experiments. Student's t-test or a one-way ANOVA (analysis of variance) was employed to the analyze data unless otherwise mentioned. The value of p <0.05 was considered to indicate a statistically significant difference.

3 RESULTS 3.1 Emodin alleviates CLP-induced injury in the sepsis model

HE staining was used to detect the lung tissue in different groups (Figure 1A). The levels of IL-1, IL-6, and TNF-α and MPO activity in lung tissues were detected by ELISA (Figure 1B–E). IL-1, IL-6, and TNF-α were increased in the CLP model, which were reduced after emodin treatment (Figure 1B). MPO activity was restored in the CLP + emodin group compared to the CLP model group (Figure 1F). The apoptosis rate was increased in the CLP model group, and it was partially decreased in the CLP + emodin group (Figure 1F). Generally, emodin may alleviate CLP-induced injury in the sepsis model.

Emodin weakens CLP-induced injury in the sepsis model. (A) HE staining was used to detect lung tissue in the SHAM, CLP, and CLP + emodin groups of mouse models. (B–D) The levels of (B) IL-1, (C) IL-6, and (D) TNF-α in the lung tissues of eight mice in each group (SHAM, CLP, and CLP + emodin) were detected by ELISA. (E) The MPO activity level of lung tissues in different groups of mouse models was detected by ELISA. (F) Flow cytometry was used to detect the apoptosis level of lung tissue in different groups of mouse models

3.2 Emodin upregulated SIRT1 protein and inhibited p-p65/p65 and HMGB1 protein levels

SIRT1 and p-p65/p65 and HMGB1 protein levels in different groups were analyzed. The results show that the SIRT1 protein level was decreased in the CLP model, which was restored after emodin treatment (Figure 2A). The protein levels of p-p65/p65 and HMGB1 were increased in the CLP model, which partly decreased after emodin treatment (Figure 2B,C). Therefore, emodin could upregulate the SIRT1 protein level and inhibit the p-p65/p65 and HMGB1 protein levels.

Emodin upregulated SIRT1 protein and inhibited p-p65/p65 and HMGB1 protein levels. (A) SIRT1, (B) p-P65/P65, and (C) HMGB1 protein levels in the lung tissues in different groups of mouse models were detected by Western blot. The results show that the SIRT1 protein level was decreased, while the p-P65/P65 and HMGB1 protein levels were increased. After emodin treatment, the SIRT1 protein level was partially increased, while the p-P65/P65 and HMGB1 protein levels were partially decreased

3.3 Emodin attenuates LPS-induced injury in MLE-12 cells

The expression of IL-1, IL-6, and TNF-α in MLE-12 cells of different groups was detected (Figure 3A–C). The expression levels of IL-1, IL-6, and TNF-α were elevated in LPS-induced model, which were decreased again with emodin treatment. Flow cytometry showed that the apoptosis rate of LPS-induced cells was increased, while the apoptosis rate was reduced in emodin-treated cells (Figure 3D). The protein level of SIRT1 was decreased, while those of p-p65/p65 and HMGB1 were increased in LPS-induced cells (Figure 3E–G). After emodin treatment, the protein level of SIRT1 was restored, while the protein levels of p-p65/p65 and HMGB1 were reduced. Thus, emodin attenuates LPS-induced injury in MLE-12 cells.

Emodin attenuates LPS-induced injury in MLE-12 cells. (A–C) The levels of (A) IL-1, (B) IL-6, and (C) TNF-α in different groups of MLE-12 cells were detected by ELISA. (D) Flow cytometry was used to detect the apoptosis level of lung cells in different groups of mouse models. (E–G) (E) SIRT1, (F) p-P65/P65, and (G) HMGB1 protein levels in different groups of MLE-12 cells were detected by Western blot. After LPS induction, the SIRT1 protein level was decreased, while the P-P65/P65 and HMGB1 protein levels were increased. After emodin treatment, the protein level of SIRT1 was partly increased, while the protein levels of P-P65/P65 and HMGB1 were partly decreased

3.4 Emodin alleviates LPS-induced inflammation and cell apoptosis via inhibition and reduction of NF-kB and HMGB1 protein levels mediated by SIRT1

The protein level of SIRT1 in MLE-12 cells was inhibited compared with the negative control tested according to Western blot (Figure 4A). According to the referred results, the expression levels of IL-1, IL-6, and TNF-α were elevated in LPS-induced cells model, while they were decreased after emodin treatment. When si-SIRT1 was transfected into the cells, IL-1, IL-6, and TNF-α were apparently elevated (Figure 4B–D). Similar results were obtained by flow cytometry and Western blot. The apoptosis rate was partially increased after transfection of si-SIRT1 (Figure 4E). The levels of p-P65/P65 and HMGB1 protein expression in different MLE-12 groups were detected by Western blot. The results indicate that the protein levels of P-P65/P65 and HMGB1 were partially reduced after emodin treatment in the LPS-induced cell model. The protein levels of P-P65/P65 and HMGB1 were partially increased after si-SIRT1 was transfected (Figure 4F,G).

Emodin alleviates sepsis-induced lung injury via inhibition and reduction of NF-kB and HMGB1 protein levels mediated by SIRT1. (A) The protein level of SIRT1 in MLE-12 cells (si-NC and si-SIRT1) was detected by Western blot. (B–D) ELISA was used to detect the levels of (B) IL-1, (C) IL-6, and (D) TNF-α in different groups of MLE-12 cells. IL-1, IL-6, and TNF-α were increased after LPS induction, while IL-1, IL-6, and TNF-α were partially decreased after emodin treatment. IL-1, IL-6, and TNF-α were partially elevated after transfection with si-SIRT1. (E) Flow cytometry was used to detect the apoptosis level of lung cells in different groups of mouse models. The apoptosis level was increased after LPS induction, partially decreased after emodin treatment, and partially increased after si-SIRT1 transfection. (F,G) The levels of (F) p-P65/P65 and (G) HMGB1 were measured in different MLE-12 groups. The levels of P-P65/P65 and HMGB1 were increased after LPS induction. After emodin treatment, the protein levels of P-P65/P65 and HMGB1 were partially reduced. The protein levels of P-P65/P65 and HMGB1 were partially increased after si-SIRT1 transfection

4 DISCUSSION

Sepsis, a life-threatening condition that arises when the body responds to infection, is currently the major cause of death in intensive care unit. ALI is a major cause of sepsis-induced acute respiratory failure. Emodin has been considered to play a protective role for acute lung edema in the CLP-induced sepsis model, which can improve the function of the liver, kidney, and small intestine and reduce the mortality rate. The therapeutic mechanisms include regulating the expression of pro-inflammatory cytokines and anti-inflammatory cytokines, inhibiting NF-kB scavenging of oxygen free radicals, and inhibiting apoptosis. This study indicated that emodin can alleviate sepsis-induced lung injury in vivo and in vitro. Meanwhile, inflammatory cytokine release and pulmonary apoptosis were remarkably reduced after emodin treatment in lung sepsis. Therefore, the results indicate that emodin could suppress inflammation, restore pulmonary epithelial barrier, and reduce mortality in CLP-induced ALI.

SIRT119 is a histone deacetylase dependent on nicotinamide adenine dinucleotide (NAD+),20 which regulates various cell functions including cell differentiation, survival, and metabolism and is involved in the inflammatory response.21 SIRT122 plays an important role in many metabolic processes, such as cellular energy metabolism, apoptosis, and cell survival.23 NF-κB transcription factor is a family of essential regulators of the immune response and cell proliferation and transformation.24 Therein, p65 is a member of the NF-κB family. Experimental studies have found that sepsis is often accompanied by upregulated activation of NF-kB, and inhibition of NF-kB activation can reduce the inflammatory response and improve the survival rate of experimental animals with sepsis. A study showed that C1q/tumor necrosis factor-related protein 3 (CTRP3) may exert its protective effects in severe acute pancreatitis mice via regulation of SIRT1-mediated NF-κB and p53 signaling pathways.25 Another study indicated that knockdown of SIRT1 attenuated the effect of miR-30d-5p depletion on proliferation via the NF-κB apoptosis signaling pathway in H9C2 cells with hypoxia treatment.26

Some studies have shown that HMGB1 plays a pivotal role in many cancers and diseases, such as gastric cancer, colon cancer, hepatocellular carcinoma, breast cancer, spinal cord injury, epilepsy, and lung cancer.27-31 In addition, it have demonstrated HMGB1 is a pivotal regulator in sepsis and ALI.32 Furthermore, HMGB1 suggests poor prognosis in patients with sepsis33 and downregulation of HMGB1 could reduce ALI severity in sepsis mice.34 Moreover, we validated that the mechanism might be related to HMGB1. We thought that the inhibition of HMGB1 mediated by SIRT1 has a protective effect on ALI. Previous studies have also suggested that HMGB1 is a pivotal mediator in sepsis and ALI.33, 35 Furthermore, HMGB1 indicates poor prognosis in patients with sepsis,34 and downregulation of HMGB1 could reduce ALI severity in sepsis mice.36 Based on the results of previous studies on HMGB1, we inferred that inhibition of HMGB1 has a protective effect on septic ALI. In this study, we found that emodin inhibited HMGB1 protein levels. However, it did not directly indicate the association between emodin and HMGB1 in sepsis because of the alteration of the microenvironment. Hence, to figure out the underlying mechanism, we further confirmed that HMGB1 was targeted and negatively regulated by SIRT1 in MLE-12 cells. HMGB1 could serve as a key therapeutic target for inflammatory diseases by promoting inflammatory cytokine secretion. Moreover, the study found that emodin inhibited HMGB1 protein levels mediated by SIRT1 and alleviated sepsis-induced lung injury. Therefore, inhibition of HMGB1 exhibits a protective function in sepsis-induced ALI.

This study had some limitations. Direct evidence was not ascertained, for example, by knocking down the HMGB1 gene. The role of HMGB1 will be further investigated, including silencing and overexpression of HMGB1 alone.

In conclusion, this study indicated that emodin could suppress inflammation, restore pulmonary epithelial barrier, and reduce mortality in CLP-induced ALI, suggesting the potential therapeutic application of emodin in sepsis. Moreover, this mechanism of sepsis-induced lung injury might be alleviated via inhibition and reduction of NF-kB and HMGB1 protein levels mediated by SIRT1.

CONFLICT OF INTEREST

All authors declare no conflict of interest.