Artesunate prevents knee intraarticular adhesion via PRKR-like ER kinase (PERK) signal pathway

ART was provided by Li Shizhen Pharmaceutical Co., Ltd. (Hubei, China). Fibroblasts were cultured in 6-well plates, 96-well plates, or 10-cm-diameter culture dishes. GSK2606414, a PERK signaling pathway inhibitor, was purchased from Medchem Express Co., Ltd. (Shanghai, China). Q-VD-Oph, a caspase signaling pathway inhibitor, was purchased from Medchem Express Co., Ltd. (Shanghai, China). When the cells were grown to nearly 80%, they were rinsed with phosphate buffer saline (PBS) for three times. Human fibroblasts were treated with ART. The control group was incubated with PBS. Subsequently, experimental and control cells were collected for subsequent cell experiments.

The Cell Count Kit-8 (CCK-8, Dojindo, Tokyo, Japan) was employed for testing fibroblast viability. The fifth to seventh generations of fibroblasts were taken, and the cells were planted in two 96-well plates. In one group, when the cells are grown to about 80%, the cells will be treated with ART at different concentrations (0, 5 μM, 10 μM, 20 μM, 40 μM, and 80 μM). Twenty-four hours later, the metabolites were first removed through PBS buffer washing, and then DMEM 100 μl and 10 μl CCK-8 were added to further culture the cells for 2 h. In the other group, when grown to 80%, the cells will be treated with 10 μM for 0, 12, 24, 36, 48, 60, and 72 h. Their optical densities were measured at 450 nm using a Microplate absorbance reader (Bio-Tek, Elx 800, USA). Cell viability was calculated based on the reference manual.

Viable cells were collected and planted in 6-well plates, and then the cells were placed in the incubator overnight. Three well cells in a 6-well plate were classified as a control group, and the remaining were cultured for 12 h in the incubator by being treated with 10 μM. Then, the cells were harvested and rinsed three times with an appropriate amount of PBS buffer solution. To grow suspension cells, the cells were suspended in 1 ml binding buffer× 1, and 100 μl of cell suspension was transferred to a test tube. Subsequently, 5 μl FITC Annexin V and 5 μl PI were added to each tube. Finally, a 400 μl 1× binding buffer was added to a test tube, and the cells were stored at ambient temperature in a light proof way. The mixture containing FITC Annexin V and PI was incubated for 15–20 min, and the apoptosis of the mixture was finally detected by flow cytometry.

Different concentrations of ART treated cells and cells of the same concentration of ART were harvested for different treatment times to conduct Western blot analysis. First, the collected fibroblasts were treated on ice with a radioimmune-deposition (RIPA) buffer (Beyotime, Hangzhou, China). Subsequently, sonicate and centrifuge processing were performed to harvest proteins. Lastly, the concentration of protein was detected using BCA Protein Assay Kit (Beyotime, Hangzhou, China). Further, 40 μg of each protein lysate were separated on the Western blot at a 5–12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA) for 1.5 h at 200 mA. Next, the PDVF membrane was blocked for 2 h at ambient temperature with 5% skimmed milk. Then, PVDF membrane was washed with tris-buffered saline and Tween20 (TBST) solution. Following the reference manual, PVDF membranes were incubated with primary and secondary antibodies continuously. The primary antibodies were used, including anti-cleaved-poly ADP-ribose polymerase (cleaved PARP), anti-cleaved caspase3, anti-Bax, anti-Bcl2, anti-78 kDa glucose-regulated protein (GRP78), anti-CHOP, anti-PERK, anti-phospho-PERK (p-PERK), anti-eukaryotic translation initiation factor 2α (eIF2α), anti-phospho-eIF2α (p-eIF2α), and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and secondary antibodies including antimouse or antirabbit IgG. All of the above antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA); the concentration of the antibodies used in this study was 1 μg/ml.

Tunnel staining was employed to test apoptosis of human fibroblasts (KeyGEN, Nanjing, China). The experimental methods rigorously followed the manufacturer’s instructions. Fluorescence microscope was used to observe the features of apoptosis. Fibroblasts stained by TUNEL were considered apoptotic. The total number of fibroblasts was calculated using DAPI staining.

The animal experiment was approved by the Animal Ethics Committee of Yangzhou University, and all the rabbits received humanitarian care. Twenty-four mature 4-month-old male New Zealand rabbits weighed 3.8 kg for this study. Rabbits were split into four groups at random (six in each group), including ART (15 mg/kg) group, (30 mg/kg) group, (60 mg/kg) group, and control group. Before the experiment, rabbits were kept in captivity for 1 week to adapt to the laboratory environment.

An intraarticular adhesion animal model was built based on previous scholars’ studies [20]. New Zealand rabbits were first given an intravenous injection of 2% pentobarbital (1.5 ml/kg) to achieve general anesthesia. Subsequently, the hair around the knee joint was shaved, the exposed skin was disinfected with iodophor, and the femoral condyle was then exposed. The cortical bone of nearly 10 mm × 10 mm was removed on both sides of the femoral condyle, and the articular cartilage was protected carefully. After the operation area was disinfected and hemostasis, the wound would be sutured. Next, the skin disinfection in the surgical area was achieved, and local injection concentrations of 15, 30, and 60 mg/kg ART were applied in the corresponding experimental group. The control group was injected with an equal amount of physiological saline. The drug was injected once per 2 days for five times. Furthermore, after the surgery, the cefazolin sodium was intramuscularly injected for 3 days to prevent postoperative infection of the rabbit; the surgical knee joint was externally fixed for 4 weeks in the Kirkner wire bending position.

Histological analysis was conducted 4 weeks after surgery. One rabbit was taken from each group at random. Subsequently, the rabbits were euthanized, the knee joint was opened along the original incision, and then the knee joint was removed (including all connective tissue and fibrous adhesion scar). The samples were fixed in 10% formalin for 1 week and then decalcified for 2 weeks. The tissue was embedded in paraffin, and transverse sections 5 mm perpendicular to the femoral axis were removed and stained with hematoxylin-eosin (HE) and Masson’s trichrome, respectively. Intraarticular scar adhesion and intraarticular collagen tissue were further evaluated under a × 400 magnification of an optical microscope.

The expressions of GRP78 (10 μg/ml) and CHOP (10 μg/ml) of the sections were detected using immunohistochemical staining. After deparaffinization and rehydration in gradient ethanol, the sodium citrate was added to the distilled water to pretreat the sections. After sealing with 10% goat serum at ambient temperature for 30 min, these sections were incubated with anti-GRP78 antibody and anti-CHOP antibody in an environment of 4 °C. They were then incubated at ambient temperature for 2 h and then analyzed with a coating tool to detect antibody binding. Subsequently, the sections were stained with hematoxylin. Finally, these sections could be observed under an optical microscope.

SPSS 19.0 statistical software was employed to analyze the data. The independent sample t test was used in our experiment. All data were expressed as the mean ± standard deviation (S.D) values. Experiments were repeated three times. Statistical significance was defined as P < 0.05.

Several important information can be acquired from HE staining images. In the experimental group, the scar tissues of 15 mg/kg, 30 mg/kg, and 60 mg/kg was less than those of the control group, and the 60 mg/kg group was still lower than the 15 mg/kg and 30 mg/kg group (Fig. 1a). The number of fibroblasts was measured by image pro plus, and it was found that the number of fibroblasts gradually decreased as the concentration of ART rose (Fig. 1b). Masson’s trichrome staining image displayed that, compared with the control group, collagen density in the experimental group was significantly lower than the control group. The trend is similar to that of HE staining (Fig. 1c, d). Based on the above results, we can preliminarily conclude that local application of ART can reduce the scar adhesion in the joint.

In order to further explore the mechanism, we examined the expression of markers GRP78 and CHOP by immunohistochemical analysis. Immunohistochemical analysis and the results of quantitative analysis of immunohistochemistry showed that GRP78 and CHOP in the cytoplasm of fibroblasts in the ART-treated group were significantly higher than those in the control group which showed a concentration-dependent manner (Fig. 2a–d). We can conclude that ERS was involved in ART-mediated delayed intraarticular adhesion.

Fibroblasts treated with different concentrations of ART and cultured with the same concentration of ART for different time were detected by cck-8 method. It was found that ART could inhibit the viability of fibroblasts in a concentration-dependent and time-dependent manner (Fig. 3a, b). To determine whether ART affects cell viability through apoptosis, we use tunnel staining and flow cytometry, and the results showed that ART could induce the apoptosis of fibroblasts (Fig. 3c–f). In addition, cleaved-PARP were used as apoptosis marker proteins; Western blot further showed that ART can induce fibroblast apoptosis (Fig. 3g, h). We use Q-VD-OPh, a caspase signaling pathway inhibitor, which demonstrated the cytoprotective effect of caspase inhibitors by Western blot quantitative analysis (Fig. 3i, j). In summary, ART can reduce cell viability through apoptosis.

After different concentrations of ART treatment, it was found that the proapoptotic proteins CHOP and Bax were upregulated in expression, while the anti-apoptotic Bcl-2 expression was downregulated, and the ratio of Bax/Bcl-2 increased (Fig. 4a, b). Figure 4c, d showed that GRP78, the ratio of p-PERK/PERK and p-eIF2α/eIF2α also showed an increasing trend and the effect is concentration-dependent. On the other hand, we treated fibroblasts with ART at a concentration of 10 μM for 0 h, 12 h, 24 h, and 48 h. The results showed that PERK pathway-related proteins were activated and gradually increased in a time-dependent manner (Fig. 4e, f). To further investigate the role of PERK signal pathway in ART-induced fibroblast apoptosis, we use the PERK signal pathway inhibitor GSK2606414. The results showed that after the PERK pathway was inhibited, the expression of CHOP, GRP78, and Bax/Bcl-2 was decreased (Fig. 4g, h). CCK-8 results were similar (Fig. 4i); it was reported that after the inhibition of PERK pathway, the cell viability was higher than that of the normal drug treatment group. These results indicate that the ART could induce fibroblast apoptosis by activating the PERK signal pathway.