Material and instruments

Macroporous silica gel (250 Å, 5.0 μm) was from Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences (Lanzhou, China). Reference standards of angiotensin II (Ang II), valsartan, losartan, candesartan, salbutamol, atenolol, emodin, aloe-emodin, and chrysophanol were supplied by Macklin Biochemical Co., Ltd (Shanghai, China). Reference standard of PD123319 was purchased from TargetMol (Boston, MA, USA). Rhei Radix et Rhizoma was obtained from Beijing Tongrentang Pharmacy (Xi’an, China).

Analysis of Rhei Radix et Rhizoma extract and the binding interactions were performed on the high-performance liquid chromatography (HPLC) system (LC-2030C 3D Plus, Shimadzu, Japan). The compounds of interest were identified by HPLC-tandem mass spectrometry using Shimadzu LCMS-8045 (Shimadzu, Japan) and confirmed by comparing their mass spectra with the data of the corresponding reference standards.

Expression and immobilization of AT1R and AT2R

By a reported method [15], we expressed the Halo-tagged AT1R and AT2R in the E. coli system. Briefly, Halo-tagged AT1R or AT2R were engineered in the pReceiver-B02 plasmid. Subsequent expressions were performed by incubating the vector for 20 h in E. coli BL21 (DE3) at 37 °C using Luria-Bertani (LB) solid medium with ampicillin resistance (100 μg/mL). A single colony was collected and transferred into the LB liquid medium (50 mL with 100 μg/mL ampicillin) for an extra 12 h incubation. The cells with ampicillin resistance were used to scale up the expression in an auto-induction medium (600 mL) by performing the incubation for 10 h. Repeating the protocol three times, we achieved 8.0 g cell pellets by centrifuging the culture (1800 mL) for 20 min. We resuspended the pellets in phosphate buffer (80 mL, 20 mM, pH 7.4) to disrupt the cell membranes by ultrasonic assay. Centrifuging the suspension for 20 min at 12,000 rpm, we collected the supernatant as the cell lysates containing AT1R or AT2R for further analysis by SDS-PAGE.

We accomplished the immobilization of AT1R and AT2R by a reported method [16]. Aminopropyl silica gel was obtained using 3-aminopropyltriethoxysilane as a silylating reagent. The subsequent reaction was carried out by adding aminopropyl silica gel (1.0 g, 1.0 eq) into reaction mixture solution obtained by dissolving 6-bromohexanoic acid (33.55 mg, 1.2 eq) and N,N-diisopropyl ethylamine (DIEA, 74 μL, 3.0 eq) in dimethylformamide (DMF, 8.0 mL). After the addition of N,N,N′,N′-tetramethyluronium hexafluorophosphate (HATU, 65.18 mg, 1.2 eq) into the mixture, the acylation reaction was allowed to proceed for 4.0 h at room temperature under a vigorously agitating condition. The activated gel was collected by vacuum filtrating the reaction mixture. The collected gels were thoroughly rinsed and straightforwardly mixed with 100 mL cell lysate containing AT1R or AT2R to react for 60 min. The immobilized receptors were collected by centrifugation and rinsed three times by phosphate buffer (20 mM, pH 7.4) to remove unreacted or non-specific adsorption proteins. The previously mentioned functionalized stationary phase was then packed into a chromatographic column tube (4.6 × 50 mm) for further investigation.

Characterization of the immobilized receptors

The immobilized AT1R or AT2R was characterized using X-ray photoelectron spectroscopy (XPS), immunofluorescence technique, and ligand-receptor interaction analysis. XPS was used to analyze the surface elements of silica gels, aminopropyl silica gels, 6-bromohexanoic acid-coated gels, and the immobilized receptors. Immunofluorescence was used to characterize the antibody-binding activity and specificity of the immobilized receptors to the control stationary phases. This aim was achieved by incubating the control stationary phases including 6-bromohexanoic acid modified gels and the immobilized receptors with their primary antibodies at 4 ℃ overnight. Subsequent incubation with anti-rabbit green fluorescent secondary antibody was carried out for 2 h against the light with the removal of the nonspecific adsorption by tris(hydroxymethyl)aminomethane buffered saline with tween (TBST) washing. Green signals were recorded by a fluorescence microscope after the gels were totally rinsed by TBST.

Ligand-binding specificity was characterized by determining the retention times of control drugs like salbutamol and atenolol (non-specific ligands of AT1R and AT2R), canonical ligands including valsartan, losartan, and candesartan (specific ligands of AT1R), and PD123319 (specific ligand of AT2R) as well as frontal analysis determination of their association constants with the receptors. The void time of the chromatographic system was tested using sodium nitrite as it has been widely accepted as a non-retained compound on an immobilized protein column. The mobile phase was 20 mM phosphate buffer (pH 7.4) at a flow rate of 1 mL/min at 25 ℃. Frontal analysis was performed on immobilized AT1R column using mobile phases containing diverse concentrations of ligands: 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, and 5.0 μM for valsartan; 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, and 5.0 μM for losartan; and 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, and 10.0 μM for candesartan. Likewise, the ligand concentrations were 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, and 10.0 μM for PD123319; 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, and 10.0 μM for valsartan; 0.05, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0, and 20.0 μM for losartan, when they were applied on the immobilized AT2R column. Each concentration was triplicated to acquire the breakthrough curves for calculating the binding parameters by Eq. (1) [17]:

$$\frac_}}=\frac_}_}[C]}+\frac_}}$$

(1)

where mlapp is the apparent adsorption amount. mL denotes the number of binding sites. KA represents the association constant, while [C] is the ligand concentration.

Screening of bioactive compounds from Rhei Radix et Rhizoma

The herbal extract was prepared using the heating-reflux method. Briefly, the herb was dried at 60 ℃ prior to being ground as powder (80 meshes). One gram of the dried powder was suspended in 50 mL methanol for 30 min. Subsequent extraction was performed by refluxing the suspension with a duration of 1.0 h. The extracts were combined and condensed to 25.0 mg/mL by rotary evaporation. The condensed extract was treated with a 0.45 μm nylon membrane before chromatographic analysis on AT1R and AT2R columns. The mobile phase was 20 mM phosphate buffer at 0.2 mL/min flow rate. The peaks detected at 254 nm with retention times greater than the void time were considered as the potential bioactive compounds.

The retained compounds were collected for additional analysis via HPLC-MS/MS. The separation was carried out on an Agilent Eclipse XDB-C18 column (3.5 μm, 2.1 mm × 150 mm). The mobile phase was the mixture of water and methanol with an addition of 0.1% formic acid (V: V) at a flow rate of 0.7 mL/min. For the separation of the retained compounds on the AT1R column, the ratio of methanol/water was 75:25 (V: V), while it was adjusted to 85:15 (V: V) for AT2R-targeting compound separation. The optimized mass spectrometry conditions included: ionizing mode-negative; drying gas-N2; flow rate of drying gas-10.0 L/min; temperature-300 ℃; and mass scan range-m/z 50 to 1500 Da. Once the compounds of interest were identified by comparing their mass patterns with the data of the reference standards, frontal analysis was designed to assess their drug-like properties by calculating the binding affinity with AT1R and AT2R.

Bioactivity of the screened compounds

Human skin fibroblasts (HSF) were purchased from Shanghai Gaining Biotechnology Co. (Shanghai, China) and cultured in Dulbecco's Modified Eagle (DMEM) media (Servicebio, Wuhan, China) containing 10% fetal bovine serum (Every Green, Zhejiang, China). The HSF cell lines incubated at 37 °C in the atmosphere containing 5% CO2 were used for assessing the anti-scar bioactivity of the screened compounds. The pathological cells induced by 100 nM Ang-II were cultured in 6-well dishes for a 12 h incubation prior to being exposed to the positive control drugs and the screened compounds with 20 μM concentration, the dose of the reagents was based on results from pre-experiments. After another 12 h incubation, we harvested the cells in radio immunoprecipitation assay (RIPA) lysis buffer containing phenylmethanesulfonyl fluoride (PSMF) for examining the expression of AT1R, AT2R, and the cytokines associated with their signaling pathway. The concentration of the total proteins in the cell lysates was determined by a bicinchoninic acid (BCA) kit (Beyotime, Shanghai, China), while the protein and the cytokine expressions were tested by western blot analysis.

Animal grouping and experimental design

All animal experiments were authorized by the Animal Care and Use Committee of Northwest University in China (No. NWU-AWC-20221104L). Male New Zealand white rabbits (2.5 ± 0.2 kg) were obtained from Chengdu Dossy Experimental Animal Co., Ltd. (SCXK (Chuan) 2019-031). Rabbits were housed in a controlled environment (22 ± 2 °C), and 12 h light/dark cycles with free access to food and tap water.

Hypertrophic scar rabbits were modeled according to a previously reported method [18]. Briefly, a 10-mm biopsy punch was used to make six full-thickness lesions down to bare cartilage on each ear of the rabbits. The wounds were ordered randomly and each was ten millimeters in diameter. We also removed the perichondrium, dermis, and epidermis in each wound under sterile conditions. Before the above surgery, all rabbits were treated with sodium pentobarbital (30 mg/kg) intravenously to alleviate their pains. We covered the wounds with sterile gauze for the 1st day. On the 7th day after the surgery, the scars were randomly separated into 4 groups (n = 12): Group 1 served as the model group and received a daily thin coating of basic ointment; groups 2 and 3 were treated with 5% aloe-emodin ointment and 5% emodin ointment, respectively; group 4 were used as positive control group taking 5% valsartan ointment. The dose of the ointment serving to every scar was 5 mg. We performed the above topical treatment once daily for 4 weeks and unwounded rabbits were gathered as a normal group. Animals were sacrificed postoperative for 35 days. The scar tissue was collected and separated into two halves: one was preserved for 3 days in a 4% paraformaldehyde for histopathological analysis, while the other half was separated with the removal of cartilage for Western blot tests. We homogenized the scars tissues using RIPA-containing PSMF solution and collected the supernatant for protein concentration determination.

Histological analysis

We fixed the scar tissues with 4% paraformaldehyde for 3 days, embedded them in paraffin, and sectioned them with a dermatome. Hematoxylin-eosin (H&E) staining was used to determine the scar elevation index (SEI) [19]. Photography was carried out under the Nikon SMZ25 microscope system, SEI measurement was performed using NIS-Elements Imaging Software Version 4.50. We also observed the collagen deposition by Masson’s trichrome staining, resulting in the collagen fibers blue, the cytoplasm light pink, and the nucleus dark brown, respectively. We analyzed the band density, thickness, placement, and collagen deposition using light microscopy.

Western blot analysis

The cell lysates and the tissue lysates were separated by 13% SDS-PAGE to transfer the bands of interest onto nitrocellulose filter (NC) membranes. The blotted membranes were blocked for 60 min with 5% nonfat milk (Sangon, Shanghai, China) at room temperature. The blocked membranes were exposed to the appropriate primary antibodies for AT1R (Proteintech, Wuhan, China), AT2R (Abcam, Cambridge, UK), matrix metalloproteinase-1 (MMP-1), tissue inhibitors of matrix metalloproteinase-1 (TIMP-1, Immunoway, Plano, TX, USA), TGF-β1, IL-6, NF-κB1 (Sangon, Shanghai, China), collagen I, collagen III, and GAPDH (Servicebio, Wuhan, China) at 4 ℃ followed by overnight incubation. With the removal of nonspecific adsorption by 1 × TBST, we used horseradish peroxidase-conjugated anti-mouse or anti-rabbit (Sangon Biotech, Shanghai, China) secondary antibodies to visualize the specific proteins. The band intensities were quantified by Image J 1.53 k software (National Institutes of Health, USA).

Statistical analysis

One-way ANOVA followed by Tukey tests were used for data analysis and the data were expressed as mean ± SEM. The experiments were independently repeated in triplicate to confirm the findings. Statistical significance was set at P < 0.05.