CGGD full ingredient formula granules were purchased from Beijing Tcmages Pharmaceutical Co., Ltd. (Beijing, China). CGGD consists of seven herbs including Bupleurum chinense DC. (batch number: 18013961), Neolitsea cassia (L.) Kosterm. (batch number: 18000871), Zingiber officinale Roscoe (batch number: 1809811), Trichosanthes kirilowii Maxim. (batch number: 18012241), Scutellaria baicalensis Georgi (batch number: 1809691), Ostrea gigas Thunberg (batch number: 18017841), and Glycyrrhiza uralensis Fisch. ex DC. (batch number: 18010991) according to “shang han lun,” and the mixed proportion of respective herbs is illustrated in Table 1. The voucher specimens were deposited in the Institute of Acute Abdominal Diseases of Integrated Traditional Chinese and Western Medicine, Tianjin Nankai Hospital. Saikosaponin A, saikosaponin D (identified for Bupleurum chinense DC.), cinnamaldehyde, cinnamic acid (identified for Neolitsea cassia (L.) Kosterm.), 6-gingerol, 8-gingerol, 10-gingerol, zingerone (identified for Zingiber officinale Roscoe), γ-aminobutyric acid (identified for Trichosanthes kirilowii Maxim.), baicalin, wogonoside, wogonin (identified for Scutellaria baicalensis Georgi), liquiritin, liquiritin apioside, isoliquiritin apioside, and glycyrrhizic acid (identified for Glycyrrhiza uralensis Fisch. ex DC.) (the purities of all standards were higher than 98% by HPLC analysis) were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). Anti-α-SMA (A5228) was obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, United States). Anti-collagen I (ABT123) was obtained from Millipore (Billerica, MA, United States). Anti-LC3B was obtained from Novus Biologicals (Littleton, CO, United States). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH, 5174), Atg5 (12994), Beclin-1 (3495), P62 (5114), p-ERK1/2 (4370s), p-P38 (9211s), p-JNK (9255s), p-AKT (4060s), p-AMPK (2535), and p-mTOR (2971) antibodies were obtained from Cell Signaling Technology (Beverly, MA, United States). Fibronectin (ab2413), MMP2 (ab37150), and TIMP2 (ab1828) antibodies were purchased from Abcam (Cambridge, MA, United States). Rapamycin was purchased from Selleck Chemicals (Shanghai, China). SP600125 was provided by MCE (Shanghai, China).

The CGGD formula granules were pulverized into homogenous powder (through a 65-mesh sieve), and then the powder (0.1 g) was extracted with methanol (1:100 g/mL) in an ultrasonic water bath at 40°C for 30 min. The solution was centrifuged at 7,000 rpm for 10 min and then filtered with a 0.22 μm filter. Aliquot (10 μL) of the solution was injected into HPLC-DAD (Thermo Scientific Ultimate 3000) for analysis. An Agilent Eclipse Plus C18 column (4.6 mm × 250 mm, 5 μm) was used at a column temperature of 30°C. The sample was eluted with the mobile phase of A (0.1% (v/v) formic acid) and B (acetonitrile) under the following gradient program: 5–10% B for 0–2 min, 10–14% B for 2–5 min, 14–18% B for 5–8 min, 18–20% B for 8–10 min, 20–21% B for 10–13 min, 21–25% B for 13–17 min, 25–25% B for 17–30 min, 25–38% B for 30–33 min, 38–40% B for 33–43 min, 40–50% B for 43–50 min, 50% B for 50–55 min, 50–95% B for 55–60 min, and 95% B for 60–62 min. The analytes were detected at the wavelength of 258 nm. The flow rate was 1.0 ml/min. Between two analytes, the column was re-equilibrated with 15% B for 10 min.

Thirty male Wistar rats (weight: 170–190 g) were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The rats were raised under specific pathogen-free conditions: 12 h light/dark cycle at 22–24°C. The animals were randomly divided into three groups (n = 10 per group): control group, CP group, and CP treated with CGGD group. All experiments were conducted in accordance with the Chinese Guide for the Care and Use of Laboratory Animals as well as the methods for the management of experimental animals. The animal experiments were approved by the Animal Ethics Committee of Tianjin Nankai Hospital (Approval No. NKYY-DWLL-2019-003).

The CP model was induced by caudal vein injection of dibutyltin dichloride (DBTC) according to previously published protocols (Zhang et al., 2017). Briefly, DBTC (Sigma-Aldrich, China) was first dissolved in Part Ⅰ (absolute ethyl alcohol) and then mixed with Part Ⅱ (glycerol) and Part Ⅲ (DMSO). The proportion of three solutions was 1:2:2. The CP group and CGGD group were injected with DBTC solution at a dose of 7 mg/kg (Inoue et al., 2002). The control group was treated with the same volume of solvent (ethanol:glycerol:DMSO, 1:2:2). One day after DBTC injection, the CGGD group rats received CGGD (1.44 g/kg body weight/dose, 144 mg/ml concentration) by gavage once daily, according to the body weight conversion between rats and humans, that is, all the granules’ weight 5.77 g/60 kg (human body weight) × 15 (conversion coefficient for humans and rats) = 1.44 g/kg per day. The rats in the control group and CP group were given the same volume of distilled water by gavage once daily. After 4 weeks, the animals were anesthetized and abdominal aortic blood was taken for serum collection, and the serum was stored at −80°C until use. The pancreatic tissues were harvested, and part of each was immediately stored at −80°C, while the other part of each was fixed in neutral phosphate formaldehyde for further experiments.

The formaldehyde-fixed pancreatic tissues were embedded in paraffin wax and cut into 5 μm sections. For morphological assessment, tissue sections were stained with hematoxylin and eosin (H&E). The results of histological changes were evaluated by two pathologists and scored on four indicators, including inflammation, abnormal architecture, glandular atrophy, and fibrosis, in accordance with previous reports (Niina et al., 2014). The collagen fiber deposition was also assessed by sirius red staining (Zhao et al., 2010). The photographs were taken by Leica microscope DMI4000B.

The serum amylase and lipase were detected by kits (C016-1-1, A054-2-1) from Nanjing Jiang Bioengineering Institute (Nanjing, China). The experimental operation was carried out according to the instructions.

The pancreatic tissue sections were detected with Bioss Kit-SP0023 (Beijing, China) according to the manufacturer’s instructions. After blocking with goat serum, the sections were incubated with primary antibodies, α-SMA (dilution 1:500) and LC3B (1:300), overnight at 4°C. Then, the nucleus was counterstained with hematoxylin. After dehydration and mounting, the slides were photographed by a microscope.

Pancreatic stellate cells (PSCs) were isolated from the rat pancreas as described previously (Li et al., 2018a). PSCs cultured for three generations were used for experiments.

Twenty male Wistar rats weighing 170–190 g were randomly divided into two groups, with 10 rats in each group. The CGGD group rats were orally administered for three consecutive days (1.44 g/kg body weight/dose), once a day. The control group rats were given distilled water. On the third day, the rats were anesthetized 2 h after final administration, the blood was collected from the abdominal aorta, and the serum was obtained by centrifugation at 3,000 rpm for 10 min. After 56°C water bath for 30 min, the serum was filtered with a 0.22 μm cellulose acetate membrane and stored at −80°C for use.

Total RNA from pancreatic tissues was extracted by Trizol reagent (Promega, China). Then, the total RNA of each sample was reverse-transcribed into cDNA using the First Strand cDNA Synthesis Kit (cat. no. K1622; Thermo, United States). RT-PCR was performed using the DyNAmo Flash SYBR Green Kit (cat. no. F-415L; Thermo) and the Applied Biosystems 7500 Fast Real-Time PCR System (Thermo, Singapore). Primer sequences for RT-PCR analysis are listed in Table 2. The value of the target genes’ mRNA was expressed as a relative intensity normalized to the endogenous control GAPDH.

Total protein was extracted from pancreatic tissues using RIPA lysis buffer adding protease inhibitor and phosphatase inhibitor (Millipore, United States). The protein concentrations were measured by the BCA Protein Assay Kit (Pierce, United States). Protein samples were separated by SDS-PAGE and then electro-transferred to the polyvinylidene fluoride membrane (Millipore, United States). The membranes were blocked with 5% non-fat milk and probed with primary antibodies at 4°C overnight and subsequently secondary antibodies (1:5,000) for 1.5 h at room temperature. The protein bands on membranes were visualized by the Chemiluminescent Substrate Kit (Pierce, United States) and quantified using the Chemidoc XRS System (Bio-Rad, CA).

A total of 5×10⁴ cells/well were seeded in a six-well plate. After cells adhered to the plate, PSCs were incubated with control serum (CONs) and CGGD serum (CGGDs) for 24 h, or pretreated with 10 μM rapamycin or 20 μM SP600125 separately for 1 h, and then combined with CONs or CGGDs for 24 h. The cells were fixed with neutral formalin and followed by staining with α-SMA (dilution 1:500) and LC3B (dilution 1:300). The nucleus was counterstained with DAPI. The images were captured using fluorescent microscopy.

All data are presented as mean ± standard derivation (SD). Statistical analyses were performed using one-way ANOVA or two-way ANOVA and Tukey’s test analysis by Prism software (GraphPad). A value of p < 0.05 was considered to be statistically significant.

According to the Pharmacopoeia of China, one or more standard ingredients were selected for each herbal medicine. The chromatograms of eight mixed standards and the extracts of CGGD formula granules are illustrated in Figure 1. The result showed that the contents of the investigated analytes were as follows: 1.94 ± 0.02 mg/g of liquiritin apioside, 1.67 ± 0.01 mg/g of liquiritin, 0.46 ± 0.01 mg/g of isoliquiritin apioside, 22.34 ± 0.89 mg/g of baicalin, 0.26 ± 0.00 mg/g of cinnamic acid, 4.42 ± 0.17 mg/g of wogonoside, 4.91 ± 0.19 mg/g of glycyrrhizic acid, and 0.61 ± 0.02 mg/g of wogonin (n = 3). But the contents of saikosaponin A, saikosaponin D, γ-aminobutyric acid, cinnamaldehyde, 6-gingerol, 8-gingerol, 10-gingerol, and zingerone were not detected under the condition.

Throughout the experiment, the body weight of all rats was monitored. In the first 7 days after injection of DBTC, most rats exhibited significant illness including ruffled fur, loss of activity, and jaundice in their tails. But the rat activity in the CP + CGGD group was a bit better than that of CP rats. Thereafter, these features improved gradually following the recovery time, and the common phenotypic characteristics in CP and CP + CGGD groups showed no differences. As shown in Figure 2A, in the CP group, the rat growth rate slowed down during the experimental period, and the body weight was decreased significantly compared with the control ones. However, the change caused by CP was not restored by treatment with CGGD. The serum amylase and lipase reflect the secretory function of the pancreas. In the three groups, they were similar, respectively, with no statistical difference found (Figure 2B). Light microscopic investigations showed that control group rats had a normal histological architecture of the pancreas. In the CP group, DBTC induced severe histological changes, including inflammatory cell infiltrates, abnormal architecture, glandular atrophy, and fibrosis (Figure 2C). Treatment with CGGD significantly attenuated the markers of pancreatic damage compared with the CP group.

Sirius red staining displayed the collagen deposition in the pancreas. As shown in Figure 3A, there were more red-stained areas in CP rats, while treatment with CGGD significantly decreased the COL content (sirius red–positive area). Activated PSCs specifically express α-SMA and collagen, leading to pancreatic fibrosis. As shown in Figure 3B, the activation marker of PSC, α-SMA, was increased significantly in the CP group compared with the control group. CGGD treatment reduced the α-SMA expression compared to the CP ones. The mRNA levels of α-SMA, COL I, FN, MMP2, and TIMP2 in CP were all increased, while CGGD treatment decreased the expression of these markers significantly (Figure 3C). The changes in protein levels of α-SMA, COL I, and FN were consistent with the mRNA levels (Figure 3D).

It is known that autophagy participates in PSC activation. Immunohistochemistry staining showed that the LC3B expression level was upregulated in the CP group compared with the control group, and the level of LC3B was downregulated markedly by CGGD (Figure 4A). Meanwhile, the mRNA expression levels of LC3B, Atg5, and Beclin-1 were significantly increased in the CP group, and these levels were suppressed in the CP + CGGD group (Figure 4B). Moreover, western blotting analysis demonstrated that the expressions of Atg5, Beclin-1, and LC3B were basically consistent with the corresponding RT-PCR results (Figure 4C).

The mTOR expression in pancreatic tissues was detected by western blot. The protein levels of mTOR and p-mTOR were suppressed in the CP group, and CGGD treatment increased these protein expressions (Figure 4D).

In order to determine the effect of CGGDs on PSCs, different proportions of control serum (CONs) and CGGDs were added to PSCs for 24 h, and the target markers were detected. The mRNA level of FN was decreased by 20% CGGDs, but the levels of α-SMA and COL I showed no significant change compared with the 20% CONs (Figure 5A). The mRNA and protein levels of α-SMA, COL I, and FN, were all decreased significantly by 50% CGGDs compared with the 50% CONs (Figures 5A–C). These results showed that CGGDs inhibited PSC activation in a dose-dependent manner.

Compared with the 50% CONs, the mRNA expressions of Atg5, Beclin-1, and LC3B were all downregulated significantly by 50% CGGDs (Figure 5D). The western blot results were similar to the mRNA results. The P62 expression level was increased in the CGGDs group compared with the CONs group, indicating the lower level of autophagy (Figure 5E). The above results reveal a suppressive effect of CGGDs on PSC autophagy and activation.

mTOR is a downstream target for MAPK, PI3K/AKT, and AMPK, the activation of which suppresses autophagy. After PSCs were treated with CGGDs for 24 h, the levels of p-mTOR and p-JNK were markedly increased in the presence of 20% and 50% CGGDs compared with the corresponding CONs. However, the levels of p-ERK1/2, p-P38, p-AKT, and p-AMPK were not changed in the absence or presence of CGGDs (Figure 6A).

Furthermore, the mTOR inhibitor rapamycin and JNK inhibitor SP600125 were applied to evaluate the role of mTOR and JNK in CGGDs-induced inhibition of PSC activation. PSCs were pretreated with or without rapamycin for 1 h and then incubated with 50% CGGDs or 50% CONs for 24 h. Western blot results indicated that rapamycin suppressed the level of p-mTOR in CONs + Rapa– and CGGDs + Rapa–treated cells compared with CONs and CGGDs ones, respectively. The expression of p-mTOR was much higher in the CGGDs + Rapa group than the CONs + Rapa group. Moreover, LC3B-II and COL I were both increased in CGGDs/CONs + Rapa groups, and the increased expression caused by Rapa (CONs + Rapa) can be decreased by CGGDs (CGGDs + Rapa) significantly (Figure 6B). The immunofluorescence staining showed similar results to Figure 6B, with increasing α-SMA and LC3B in CONs + Rapa and CGGDs + Rapa groups, and CGGDs suppressed the levels of α-SMA and LC3B that were elevated by Rapa (Figure 6D). These results indicated that CGGDs could inhibit PSC autophagy and activation by mTOR.

On pretreatment with the JNK inhibitor SP600125, the expression of p-JNK was reduced significantly, while the level of LC3B was increased significantly (CONs + SP600125 vs. CONs and CGGDs + SP600125 vs. CGGDs). CGGDs elevated the downregulated protein level of p-JNK induced by SP600125 and, meanwhile, reduced the high expression level of LC3B caused by SP600125 (CGGDs + SP600125 vs. CONs + SP600125) (Figure 6C). With the treatment of CGGDs, SP600125 exhibited promoting effect of COL I. But compared with CONs + SP600125 treatment, CGGDs + SP600125 treatment showed no influence on the COL I protein expression. Moreover, as shown in Figure 6D, the α-SMA expression was not different in CONs, CONs + SP600125, and CGGDs + SP600125 groups. However, the level of α-SMA in the CGGDs + SP600125 group was higher than that in the CGGDs group. The immunofluorescence results of LC3B were consistent with the western blot ones. These results revealed CGGDs could suppress PSC autophagy by JNK. Furthermore, the inhibitory effect of CGGDs on PSC activation is partly through the JNK pathway.