RETRACTED ARTICLE: Effects of Panax notoginseng saponins on severe acute pancreatitis through the regulation of mTOR/Akt and caspase-3 signaling pathway by upregulating miR-181b expression in rats

To obtain PNS extracts, 500 g crude material of Panax notoginseng (Sanqi) was subjected to extraction and purification according to the following procedures. Crushed powder was added into 4 L of 70% (v/v) ethanol, and extracted twice for 2 h with refluxing. Extracts were evaporated by a rotary evaporator. PNS was purified on D101 resins and washed with 0.1% (v/v) ammonia, water, and 70% (v/v) ethanol. The fraction washed with ethanol was collected and evaporated. The residue was dried in a freezing dryer. All solution fractions were respectively diluted to 5 μg/mL (herbal weight per solution volume) with 1640 medium for analysis.

The Waters 2960 HPLC system (Milford, MA) was used to anlyze PNS extracts. The separation was performed on an Alltech Ultrasphere C18 column (5 μm, I.D.: 250 × 3.2 mm) (Deerfield, IL) and a C18 guard column (5 μm, I.D.: 7.5 × 3.2 mm). Firstly, 20 μL of diluted sample solution was injected into the column for elution (1.0 mL/min, room temperature). Acetonitrile and water were respectively used as Solvent A and Solvent B. The gradient elution was performed in the following steps: starting with 17.5% Solvent A and 82.5% Solvent B, 21% A for 20 min, 26% A for 22 min, 36% A for 13 min, 50% A for 9 min, 95% A for 5 min, and 17.5% A for 11 min. in the analysis, the detection wavelength was 203 nm. According to the HPLC analysis results, 100 mL of PNS extracts contained 3.84 mg Panax notoginseng Rl, 17.25 mg ginsenoside Rgl, 18.16 mg ginsenoside Rbl, and 13.8 mg ginsenoside Rd. (Fig. 1).

AR42J cell lines from the American Type Culture Collection (ATCC) were cultured at 37 °C in DMEM composed of 5% FBS, streptomycin (100 units/mL), penicillin (100 units/mL), and amphotericin B (50 μg/L) in a humidified atmosphere containing 5% CO₂. Then the cells were sub-cultured into 6-well plates and maintained until subconfluence.

AR42J cells were cultured to the confluence of 40% in 6-well plates. MiR-181b mimics, MiR-181b mimic-negative control (NC), miR-181b inhibitor, or miR-181b inhibitor-NC (Invitrogen, Carlsbad, CA, USA) were mixed with Lipofectamine 2000 (Invitrogen), and then added into the culture medium. After transfection for 24 h, total RNA and proteins were extracted for qRT-PCR and western blotting.

AdCMV-miR-181b was constructed as previously described [32, 33]. The constitutively active miR-181b expression construct was delivered to rats by intravenous administration of 1 × 10⁹ PFU of AdCMV-miR-181b at 10 d, at the time of taurocholate-induced pancreatitis. Control rats received empty adenoviral vector at the same time.

Male Sprague Dawley rats were housed under pathogen-free conditions at a relative humidity of 30~ 70% in Kunming Medical University. At the beginning of the experiments, 8~ 9 weeks old rats were maintained on standard laboratory chow and water ad libitum. All the animal experiments were approved by the Ethics Committee of the Institute of Yunan University of Traditional Chinese Medicine (TCM).

The rats were anaesthetized under aseptic conditions by intraperitoneal injection of 1% pentobarbital sodium (35 mg/kg body weight) (Wuhan Dinghui Chemical Co., Ltd., Wuhan, China). According to the previous method [34], SAP models were prepared. After median epigastric incision in the abdomen, hepatic hilus, the bile-pancreatic duct, common hepatic duct, and the duodenal papilla inside the duodenum duct wall were identified. A segmental epidural catheter was firstly inserted into the duodenum cavity and then into the bile-pancreatic duct along papilla. Two microvascular clamps were used to nip the two ends of the bile-pancreatic duct. Then, with a microinfusion pump, 5% sodium taurocholate was injected into the bile-pancreatic duct (0.2 mL/min, 1.5 mL/kg body weight). In 5 min, the epidural catheter and microvascular clamp were removed. The abdomen was closed until active bleeding was not observed in the abdominal cavity.

Sixty rats were divided into the following 3 groups in a random manner: the SAP model group (model group), the Sham operation group (SO group), and the PNS group (treatment group) and each group contained 20 rats. Before the surgery, the rats of the treatment group were injected with PNS extracts (50 mg/kg) by intravenous injection administration every 8 h. The rats of the SO and SAP model groups were injected with the same volume of saline. The rats in each group were anesthetized with 1% pentobarbital sodium (35 mg/kg body weight, Wuhan Dinghui Chemical Co., Ltd., Wuhan, China) and euthanized by cervical dislocation after 24 h and then the right internal carotid artery was identified. Then 5 mL of blood was extracted and centrifuged. The supernatant was dispensed into two sterile tubes, sealed, and stored in a freezer at − 20 °C for analysis. Ascites and serum were collected and amylase levels and ascitic capacity were determined. Pancreata were quickly obtained and fixed in 10% formalin for observation. Portions of the pancreas were freshly treated to determine pancreatic water contents.

Total RNA was extracted with TRIzol reagent (Invitrogen). The RNA concentration was measured with a Nanodrop Spectrophotometer (ND-100, Thermo, Waltham, MA, USA). cDNA was reversely transcribed from total RNA with a HiScript 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China) according to the manufacturer’s protocol. cDNA was generated from miRNAs with a stem-loop RT-qPCR method. Quantitative real-time PCR was performed in triplicate in an ABI StepOnePlus Real-Time PCR system (Applied Biosystems, CA, USA). β-actin and U6 were adopted as endogenous controls respectively in mRNA and miRNA expression profiles. The expression levels were normalized to endogenous controls and the fold change in relative gene expression was calculated as 2^−∆∆Ct. Primer sequences are listed in Table 1. After 10-min pre-incubation at 95 °C, PCR was performed according to the following program: 35 cycles of denaturation at 95 °C for 15 s, annealing at 60 °C for 5 s, and elongation at 72 °C for 12 s. The experiments were performed in triplicate.

Retrieved pancreatic tissues of rats were powdered in liquid nitrogen and then lysed with a nuclear and cytoplasmic protein extract kit (Beyotime, Beijing, China). The whole proteins of pancreatic tissues were reconstituted in ice-cold RIPA buffer containing a cocktail of protease inhibitors (1:100; Sigma-Aldrich) and phenylmethanesulfonylfluoride (PMSF, 1 mM). Homogenates of pancreatic tissues were centrifuged for 15 min (12,000×g, 4 °C). Acinar cell supernatants were centrifuged for 100 min (4000×g, 4 °C) in an ultrafiltration tube (Vivaspin 20, 3000 MWCO PES, Sartorius, Germany). The concentrations of cytoplasmic, nuclear, and whole proteins were determined according to the BCA method (Pierce, Rockford, LA, USA). An aliquot of protein (80 μg) or the same proportion of concentrated supernatant was firstly separated by sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS-PAGE Bio-Rad, Hercules, CA, USA) and then transferred to a nitrocellulose/PVDF membrane. The membrane was blocked with 5% (w/v) dry non-fat milk in Tris-buffered saline/0.05% Tween-20 (TBST) for 1 h at room temperature in a covered container. Blots were incubated with rabbit polyclonal anti-p-mTOR (1:1000), rabbit polyclonal anti-mTOR (1:1000), rabbit polyclonal anti-Beclin1 (1:200), rabbit polyclonal anti-LC3-II rabbit polyclonal (1:1000), rabbit polyclonal anti-LC3-I rabbit polyclonal (1:1000), anti-Akt antibody (1:1000), anti-p-Akt antibody (1:1000), and rat monoclonal anti-β-actin (1:1000) diluted in 5% bovine serum albumin (BSA) overnight at 4 °C. Lamin-A and β-actin were respectively used as the markers for nuclear and cytoplasmic fractions. Then the membranes were washed with TBST and incubated for 1 h at room temperature with a secondary goat anti-rabbit IgG-horseradish peroxidase (HRP) antibody (1:2000) or goat anti-rat IgG-HRP antibody (1:2000) (Santa Cruz, CA, USA) diluted in 5% (w/v) dry non-fat milk in TBST. The same protein loading of the samples was confirmed by β-actin. All western blots were determined with a densitometer.

After antigen retrieval with Retrievagen A (Zymed Laboratories, Inc., San Francisco, CA, USA), the pancreatic tissue sections were subjected to 20-min immunostaining at 100 °C and then endogenous peroxidases were quenched with 3% H₂O₂ (Tianjin Jinqiang Chemical Co., Ltd.). After the sections were blocked with 2% bovine serum albumin (BSA) in phosphate-buffered saline (PBS, staining was performed at room temperature with primary anti-anti-caspase-3 and Bcl-2 (BD Pharmingen, San Jose, CA, USA) for 1 h. After washing, the sections were treated with the secondary antibody (R & D Systems, Inc.). The tissues were developed with Vectastain ABC and 3,3′-diaminobenzidine (Vector Laboratories, Inc., Burlingame, CA, USA). After staining, in each slide, five high-power fields (Magnification, 400×) were randomly selected and then the average proportion of positive expression in each field was calculated with the true color multi-functional cell image analysis management system (Rockville, MD, USA). Caspase-3 and Bcl-2-positive expressions in pancreatic tissue were measured and expressed in the form of positive units.

After trimming pancreatic tissues to the size of 1 × 1 × 1 mm³, the trimmed tissues were subjected to 24-h fixation with 2.5% pentanediol and 2-hpost-fixation with 2% osmium tetroxide. The samples were respectively dehydrated according to the gradient of 50, 70, 90, and 100% dehydrated acetone and the dehydration was performed for 10 min in each concentration of dehydrated acetone for three times. Then dehydrated samples were treated at 37 °C for 24 h in the mixture of epoxy resin and pure acetone (1:1), followed by 24-h embedding at 60 °C in the mixture of Epon812 resin, dodecenylsuccinic anhydride, dimethylaminomethyl phenol, and methyl nadic anhydride. After trimming the blocks with a semi-thin glasscutter and staining with toluidine blue, the samples were observed under an optical microscopy in order to select the areas with pancreatic acinar structures. After obtaining ultrathin sections (500 Å thick) with an ultrathin microtome, the sections were stained with uranyl acetate and lead nitrate and then observed under transmission electron microscopy (Nissan HT7700; Nissan Corporation, Tokyo, Japan) to explore the changes of the pancreatic acinar autophagosomes.

Apoptosis could be analyzed in the TUNEL assay (Roche Molecular Biochemicals, Mannheim, Germany). TUNEL-positive nuclei were counted in 10 random high-power fields (640× objective). The apoptotic index was calculated as a ratio of the apoptotic cell number to the total cell number in each field and expressed as a percentage. The quantification of apoptotic cells was carried out by three different observers in a blinded fashion.

Serum levels of endotoxin were determined with quantitative chromogenic endpoint Limulus Amebocyte Lysate (LAL) QCL-1000 kit (Lonza, Walkersville, MD). Firstly, 300 μL of plasma was diluted with 10 mM MgCl₂ (Lonza) according to the ratio of 1:3 and then inactivated for 30 min at 70 °C. After the samples were further diluted to the ratio range of 1:30~ 1:40, 50 μL of the diluted sample was added into a 96-well pyrogen-free culture plate. Other procedures were performed according to the manufacturer’s instructions. Endotoxin levels (EU/mL) in the samples were obtained from a standard curve plotted with pure endotoxin standards. All the assays were performed in duplicate.

Serum levels of D-lactate and DAO were measured with commercial kits (Genmed, China) via spectrophotometric measurements.

Serum activities of lipase and amylase were determined with commercial kits in a Roche/Hitachi modular analytics system (Roche, Mannheim, Germany).

The changes in pancreatic weight were evaluated as the indicator of pancreatic interstitial edema. The whole pancreas was obtained and weighed. The weight of each pancreatic sample was used to estimate the water content in pancreas according to the previous method [35] and expressed as a ratio of pancreas/body weight.

The pancreas was obtained from each rat, fixed overnight at 4 °C in 10% formalin and then embedded in paraffin. Full-length (4 μm) sections were obtained and stained with hematoxylin and eosin for histologic evaluation. According to the previous method [35], edema, hemorrhage and necrosis of the pancreas were graded into 4 levels (0 to 3).

Immediately obtaining parts of distal ileum, the fixation was performed in 40 g/L phosphate buffered formaldehyde. Paraffin-embedded tissue sections (5 μm thick) were stained with hematoxylin and eosin. Mucosal damage was assessed according to the standard scale proposed by Chiu et al. [36]. Grading was classified as follows: 5 = disintegration of the lamina propria; 4 = denuded villi; 3 = massive epithelial lifting; 2 = extension of the space with epithelial lifting; 1 = development of subepithelial space at the tip of the villus; 0 = normal mucosa.

The other 45 rats were divided into 3 groups: the sham operation group, SAP group and SAP plus PNS group (n = 15 per group) and the survival rate was calculated. The treatments were identical to those of previous experiments [3]. Observation started from PNS treatment and ended in 72 h after PNS treatment.

In SPSS 11.0, statistical analysis was performed and the experimental data were expressed as mean ± SD. Statistical differences between two groups were evaluated by t-test. The analysis of linear correlation was used to evaluate the correlation between two variances and q-test of analysis of variance (ANOVA) was used to perform multiple comparisons. The survival data were analyzed with log-rank or χ2. P < 0.05 was considered to be significantly different.

LC3 and Beclin-1 are the markers of autophagy [35]. To explore the effects of miR-181b on Beclin1 and LC3-II expressions, we transfected AR42J cells with MiR-181b mimics, MiR-181b mimic-negative control (NC), miR-181b inhibitor, or miR-181b inhibitor-NC. After 24-h transfection, the expression levels of Beclin1 and LC3-II were measured by Western Blotting. The increase in miR-181b expression significantly decreased Beclin1 and LC3-II expressions in the miR-181b mimics group compared to the control group and all treatment groups (P < 0.05) (Fig. 2). These results suggested that miR-181b might inhibit autophagy and regulate Beclin1 and LC3-II expressions.

To analyze the effects of overexpression of miR-181b on pancreatic damage in rats with taurocholate-induced SAP, we injected 1 × 10⁹ PFU of AdCMV-miR-181b into rats by tail intravenous administration, followed by intraperitoneal taurocholate challenge in 10 days. Histopathological examination of the pancreas of taurocholate-treated miR-181b-expressing rats revealed the markedly extenuated pancreatic damage compared with control animals (Fig. 3).

To analyze the effects of overexpression of miR-181b on Beclin1 and LC3-II expression in rats with taurocholate-induced acute pancreatitis, we performed western blotting on pancreatic tissue samples. As shown in Fig. 4, taurocholate-induced SAP tissues in SAP/miR-181b group and SAP group showed the significantly increased Beclin1 and LC3-II expressions compared to the SO group and SO/miR-181b group (P < 0.05). However, overexpression of miR-181b in rats with taurocholate-induced AP (the SAP/miR-181b group) resulted in the decreased Beclin1 and LC3-II expressions compared to the rats induced by taurocholate (SAP group) (P < 0.05).

We evaluated the effects of PNS treatment on SAP rats by examining the expressions of miR-181b, mTOR, Akt, Beclin1 and LC3-II in taurocholate-induced pancreas tissue by RT-PCR and Western Blotting. In taurocholate-induced rats, the mRNA and proteins of miR-181 and mTOR were significantly decreased (P < 0.05) and Akt, Beclin1 and LC3-II expressions were significantly upregulated compared to the control group (P < 0.05). After PNS treatment, the expressions of miR-181 and mTOR were markedly enhanced (P < 0.05), whereas Akt, Beclin1, and LC3-II expressions were significantly lower compared to the SAP group (P < 0.05). The changes suggested that PNS could suppress the expressions of Akt, Beclin1 and LC3-II, and upregulate miR-181b and mTOR expressions (Figs. 5–6).

The formation of autophagosomes is a pivotal process in autophagy. To explore the effect of PNS on autophagy in SAP, the ultrastructures of pancreatic cells from the three treatment groups were observed under the electron microscopy. Phagophores, autophagosomes and autolysosomes were increased in the taurocholate-treated group and decreased in the PNS treatment group (Fig. 7). These observations suggested that PNS inhibited pancreatic autophagy in taurocholate-induced rat pancreas tissues.

The effects of PNS treatment on apoptosis in the SAP model was explored by evaluating Bcl-2 and caspase-3 expressions in pancreas tissues via immunohistochemistry. The expressions of Caspase-3 and Bcl-2 were significantly up-regulated in pancreas tissues from the rats with taurocholate-induced SAP (Fig. 8). However, administration of PNS further significantly decreased Bcl-2 expression and increased Caspase-3 expression in taurocholate-induced SAP rats. These results demonstrated that PNS significantly activated Caspase-3 expression and enhanced Bcl-2 expression in SAP model rats.

The effects of PNS treatment on the apoptosis of pancreatic cells from taurocholate-exposed SAP rats were explored via TUNEL assays. In the rats exposed to taurocholate, the apoptotic rate of pancreatic cells was significantly higher than that in the control group (Fig. 9). Notably, compared to the taurocholate group, the PNS treatment group exhibited a significantly increased apoptotic rate of pancreatic cells (P < 0.05). The differences suggested that PNS increased the apoptosis in pancreatic cells in the SAP model rats.

In pancreatitis assessment, pancreatic edema is one of the key factors [29, 37]. After injecting 5% sodium taurocholate into the biliary-pancreatic duct of rats in our model, the pancreas/body weight ratio was significantly increased (Fig. 10). Furthermore, serum amylase activity in the SAP rats was significantly increased. Notably, PNS treatment largely decreased the ratio of pancreas to body weight as well as serum amylase activity, showing a positive effect on pancreatic injury and edema.

Previous studies indicated that intestinal barrier injury might lead to the multiple organ dysfunction syndrome [38]. We examined plasma D-lactic acid and DAO with the two markers indicating intestinal barrier dysfunction caused by SAP. We determined the two markers by ELISA and serum endotoxin (Fig. 11). In the SAP model group, the plasma DAO, D-lactic acid, and serum endotoxin were significantly increased compared to the control group. Notably, compared with the SAP group, the PNS group exhibited the significantly decreased plasma D-lactic acid, DAO and endotoxin (P < 0.05). Our results showed that administration of PNS significantly reduced plasma D-lactic acid, endotoxin, and DAO in SAP rats.

In order to evaluate the effects of PNS on local pancreatic and ileal injury, the morphology of the pancreas in the rats of all the three groups was observed (Fig. 12). The SAP group exhibited the severe edema and the high destruction degree of histoarchitecture of acini cells and pancreatic pathological scores in the SAP group were significantly increased compared to the model group. After establishing SAP models, villi and crypt structures were damaged to some extent and hair became thinner and shorter. Heavy inflammatory cell infiltration in the intrinsic membrane was observed and lymphatic dilatation and edema occurred. In the PNS treatment group, the injury to ileal and pancreatic tissues was attenuated and pancreatic and ileal injury scores were significantly reduced. These data suggested that PNS treatment attenuated the injury to pancreatic and ileal tissues.

The survival rate of the rats with taurocholate-induced SAP was significantly decreased compared to the control group. The decreasing trend in the survival rate induced by taurocholate-induced SAP was significantly attenuated by the pre-treatment with PNS compared to the untreated taurocholate-induced SAP group (Fig. 13).