Astragalus polysaccharide attenuates LPS‐related inflammatory osteolysis by suppressing osteoclastogenesis by reducing the MAPK signalling pathway

Abstract Bacterial products can stimulate inflammatory reaction and activate immune cells to enhance the production of inflammatory cytokines, and finally promote osteoclasts recruitment and activity, leading to bone destruction. Unfortunately, effective preventive and treatment measures for inflammatory osteolysis are limited and usually confuse the orthopedist. Astragalus polysaccharide (APS), the main extractive of Astragali Radix, has been widely used for treating inflammatory diseases. In the current study, in vitro and in vivo experimental results demonstrated that APS notably inhibited osteoclast formation and differentiation dose‐dependently. Moreover, we found that APS down‐regulated RANKL‐related osteoclastogenesis and levels of osteoclast marker genes, such as NFATC1, TRAP, c‐FOS and cathepsin K. Further underlying mechanism investigation revealed that APS attenuated activity of MAPK signalling pathways (eg ERK, JNK and p38) and ROS production induced by RANKL. Additionally, APS was also found to suppress LPS‐related inflammatory osteolysis by decreasing inflammatory factors' production in vivo. Overall, our findings demonstrate that APS effectively down‐regulates inflammatory osteolysis due to osteoclast differentiation and has the potential to become an effective treatment of the disorders associated with osteoclast.

infection. [9][10][11][12] LPS is considered a potent inducer of inflammatory reactions and a pro-inflammatory cytokine, stimulating macrophages, fibroblasts and other cells to secrete various pathogenic cytokines responsible for inflammation, such as IL-1β, IL-6 and TNFα. Moreover, LPS can also correspondingly activate the NF-kB and MAPK signal pathways to facilitate osteoclastogenesis, resulting in destructive osteolysis. 9,11,12 In addition to this indirect effect of LPS through inflammatory factors, it has been reported that LPS correspondingly activates TLR4 signalling cascades, further promoting the expression of CXCR4 and TRAF6, and ultimately likewise activates the MAPKs signalling pathway that subsequently leads to osteoclast overactivation. 10,13 Therefore, based on the rationale above, suppressing osteoclast formation and differentiation by inhibiting inflammation can be a promising strategy for treating LPS-induced inflammatory osteolysis.
In recent years, plant-derived natural products and their derivatives have been demonstrated to be valuable sources to explore new therapeutic approaches for clinical diseases due to their specific pharmacological activities. Astragali Radix (Huang Qi), the dried roots of Astragalus membranaceus (Fisch.) Bge. are one of the most well known herbal medicines widely used to treat distinct diseases in China. 14,15 Astragali Radix extracts contain numerous bioactive ingredients, among which Astragalus polysaccharide (APS) is one of the most critical bioactive extract components. 14,16 APS was found to exhibit various bioactivities, including antioxidant, 17,18 antiinflammatory, 14,17 immunoregulatory, 19 hypoglycaemic 15 and antitumour activities. 20 Na Dong et al demonstrated that APS alleviated the MAPK and NF-kB signalling pathway gene expression levels and ultimately correspondingly reduced the production of inflammatory factors interleukin 6 (IL-6), IL-Iβ and TNFα induced by LPS. 21 However, the effects of APS on inflammatory osteolysis caused by osteoclasts remain unrevealed.
In this work, it has been suggested that APS might potentially become a novel therapeutic approach for inflammatory osteolysis by suppressing osteoclast formation. We thus aimed to investigate the potential efficacy of APS on osteoclast-associated osteolytic bone conditions and further elucidate the underlying molecular roles of APS in osteoclasts formation.

| Materials and reagents
The highly pure APS (>98%) was obtained from Solarbio Science & Technology Company, and a 5 mg/mL stock solution of APS was prepared in PBS. DMEM, α-MEM and FBS were obtained from Thermo Fisher Scientific. Recombinant murine RANKL and M-CSF were obtained from R&D Systems Reagents. The complete DMEM and α-MEM induction mediums consisting of M-CSF (25 ng/mL) and RANKL (50 ng/mL) were prepared for RAW264.7 cells and bone marrow-derived monocytes (BMMs), respectively. Specific antibodies for c-Fos, NFATc1, P38 phosphorylated (p) P38 and β-actin were purchased from Cell Signaling Technology. ELISA kits for mouse IL-6, IL-1β, and TNFα and specific antibodies for CTSK, JNK, p-JNK, ERK and p-ERK were obtained from Bioss Antibodies.
The TRAP Kit, Actin Cytoskeleton/Focal Adhesion Staining Kit and LPS were purchased from Sigma-Aldrich. The Cell Counting Kit-8 (CCK-8) was purchased from Beyotime Biotechnology.

RAW264.7 cells were obtained from the American Type Culture
Collection and cultured in a T25 flask with a complete DMEM consisting of FBS (10%), penicillin (100 U/mL) and streptomycin (100 µg/mL). Six-week-old C57BL/6J mice were used to isolate BMMs following the protocols consent by the Animal Ethics Committee of the Chongqing Medical University. The long bones were dissected to remove all soft tissues, and the bone marrow was flushed from the femur and tibia. Cells were incubated for 24 hours in α-MEM supplemented with 25 ng/mL M-CSF, 10% FBS, penicillin (100 units/ mL) and streptomycin (100 mg/mL). Subsequently, non-adherent cells were removed. After that, adherent BMMs were cultured in a T25 flask with a complete α-MEM consisting of M-CSF (25 ng/mL) in an incubator at a constant temperature of 37°C with 5% CO 2 .
After 24 hours pre-incubation, the cells were administrated by APS (0, 1, 5, 10, 50 or 100 µg/mL) for 24 hours, 48 hours and 72 hours, respectively. Ten microliters of the CCK-8 working solution were added afterwards. The plate was incubated at 37°C for another 2 hours avoiding light. A microplate reader (Multiscan Spectrum; Thermo Labsystems) was applied to determine the absorbance at 450 nm. RAW264.7 or BMMs (5 × 10 3 per well) were placed into 96-well plates for 12 hours and then cultured in the complete induction medium. The RAW264.7 or BMMs were administrated with APS or blank. The medium of cell culture was changed every two days.

| Osteoclastogenesis assays
After a five-day incubation, all the RAW264.7 or BMMs were fixed with 4% paraformaldehyde (PFA) for 30 minutes and washed with PBS in triplicate. A light microscope was then applied to count TRAP-positive multinucleated cells (nuclei ≥3). TRAP staining was performed on the cells using a TRAP kit following the manufacturer's procedures. In the next moment, a light microscope was used again to count TRAP-positive multinucleated cells (nuclei ≥3). As above, for LPS-induced osteoclastogenesis, RAW264.7 and BMMs were cultured in the complete induction medium for 24 hours, then cultured in a complete medium consisting of LPS (100 ng/mL) and were treated with APS or blank PBS. In the end, the osteoclasts were counted.

| In vitro bone resorption assay
To measure osteoclast activity, RAW264.7 (1 × 10 5 per well) was cultured on 6-well plates in the complete induction medium until the generation of mature osteoclasts. The identical numbers of mature osteoclasts (1 × 10 4 cells per well) were placed into hydroxyapatite-coated 96-well plates (Corning Osteoassay). The induced cells were treated with APS or blank PBS. After a twoday incubation, sodium hypochlorite was used to bleach the wells three times (10 minutes per cycle) to remove cells, followed by the resorbed area measurement. The percentage area of hydroxyapatite surface resorbed by the osteoclasts was quantified by utilizing ImageJ software (NIH).

| RNA isolation and RT-qPCR
RAW264.7 and BMMs (1 × 10 5 per well) were placed in 6-well plates for 24 hours and then cultured in the complete induction medium.
The cells were administrated with APS or blank PBS. After a five-day (BMMs) or three-day (RAW264.7) incubation, total RNAs were isolated from cells by utilizing TRIzol reagent (Thermo Fisher Scientific) based on the manufacturer's instructions. One microgram of total RNAs was applied to synthesize cDNA using reverse transcriptase (Takara Bio Inc). SYBR Premix Ex Taq II (Takara Bio Inc) was implemented to conduct qRT-PCR in a PCR detection system (Bio-Rad).
The primer sequences are summarized in Table 1. The levels of the target genes were determined relative to the levels of GAPDH's messenger RNA (mRNA).

| Western blot analysis
RAW264.7 and BMMs (1 × 10 5 per well) were placed in 6-well plates for 24 hours and then cultured with APS or blank PBS in the complete induction medium for five days (BMMs) or three days (RAW264.7).
The RIPA lysis buffer containing a cocktail combination of protease inhibitor and phosphatase inhibitor was applied to extract total cellular proteins from the cells abovementioned. The obtained proteins were then resolved by SDS-PAGE and electron transferred to PVDF membranes (GE Healthcare). All membranes were blocked in a TBST solution consisting of 5% non-fat milk for 1 hour before incubating with primary antibodies overnight. The membranes were then rinsed with TBST in triplicate and immersed in the secondary antibody at room temperature for 2 hours. The reactivity of the antibody was then determined by an enhanced chemiluminescence reagent (GE Healthcare). The protein bands' images were captured by the Chemi Doc XRS+Imaging System (Bio-Rad) followed by the analyses using the ImageJ software. β-actin was applied as an internal control.

Gene
Forward Reverse The sequences of primers for qPCR

| Intracellular reactive oxygen species detection
Intracellular ROS levels were investigated by a DCFDA cellular ROS detection assay kit (Abcam). BMMs (5 × 10 3 per well) were cultured in 96-well plates in the complete induction medium and were treated with APS for 72 hours. The cells were then incubated with DCFDA at 37°C, avoiding light, for 30 minutes. The images were taken by fluorescence microscopy.

| Enzyme-linked immunosorbent assay
For the mouse blood sample, the blood coagulated naturally at room temperature for 20 minutes and was centrifuged for about 20 minutes (2000 rpm) at 4°C. After that, the ELISA test was performed on the collected supernatant, and the corresponding ELISA kits were applied to determine the levels of IL-6, IL-1β and TNFα in mouse blood samples following the manufacturer's instructions.

| LPS-stimulated mouse model of calvarial osteolysis
Animal EDTA for 2 weeks. Finally, H&E staining was performed, and the obtained results were used to determine inflammatory osteolysis in vivo.

| Statistical analysis
Unless otherwise indicated, all experiments were carried out in triplicate independently. Data are exhibited as mean ± standard deviation. The Student's t test was applied to conduct difference analyses between the indicated groups. GraphPad Prism 8 (GraphPad Software) was used to carry out all the analyses. *P values less than .05, **P values less than .01 and ***P values less than .001 were determined to be statistically significant.

| APS inhibited RANKL-stimulated osteoclast formation and fusion
First of all, to assess the effects of APS on the osteoclast formation

| APS attenuated bone resorption activity in vitro
Subsequently, a hydroxyapatite resorption experiment was conducted to evaluate the influence of APS on the function of osteoclast. The hydroxyapatite resorption areas shown in Figure 2A,C were apparently reduced after treating osteoclasts with high-dose APS (80 µg/mL).
These data indicated that APS attenuated osteoclastic bone resorption activity in vitro in a concentration-dependent fashion.

| APS reduced RANKL-stimulated osteoclastspecific gene levels, osteoclastic differentiation and function-associated protein levels
We

| APS inhibited intracellular ROS production stimulated by RANKL
As mentioned earlier, APS is an antioxidant. Therefore, to explore the underlying mechanism of APS-dependent osteoclastogenesis  (Figure 6D,E). Therefore, we consider that APS may inhibit osteoclast formation by inhibiting ROS production.

| D ISCUSS I ON
Chronic gram-negative bacterial bone diseases, including infection of orthopaedic implants, periodontitis, septic arthritis, septic arthritis and osteomyelitis, are often caused by enhanced osteoclastic activity and osteolysis. 22,23 At present, it is evident that the molecular mechanism of bacteria-induced osteolysis is related to the activation of immune cells by bacterial endotoxin, which ultimately leads to the formation of osteoclasts. 12,24,25 Unfortunately, to date, available treatment approaches for inflammatory osteolysis are still limited. It is well known that bisphosphonate, an antiresorptive agent with a definite curative effect, is widely applied to treat osteolytic bone diseases, 26,27 while it may result in some additional complications, such as osteonecrosis of the jaw and atypical fractures, eventually causing orthopedists' confusion. 26,27 Hence, the exploration of novel and effective anti-osteolysis agents is still highly required. been known as a potent stimulus of the progression of inflammatory osteolysis for a long time, so that it is widely used to establish an animal inflammatory bone erosion model. 28,29 In this study, LPS was utilized to cause inflammatory osteolysis in a mouse model of calvaria. As the main extractive of Astragali Radix, APS has been widely used for treating inflammatory diseases. 21,30,31 As osteoclast plays a critical role in inflammatory osteolysis, 1,32 the current study aimed at investigating if APS has the ability to inhibit osteoclastogenesis in murine models of inflammatory osteolysis. We found that The underlying mechanism of APS's inhibition on osteoclast formation was further investigated. RANKL is an indispensable critical cytokine that participates in osteoclast formation, differentiation and function. 29,33 After stimulation by RANKL, TRAF2/6 signalling is activated by RANK, resulting in the phosphorylation of three MAPK pathways (JNK, P38 and ERK) mediated by MEK1/2, MKK7 and MKK6. 28,33,34 Furthermore, an increased level of c-Fos is essential for the production of NFATc1 to regulate osteoclast differentiation. 35,36 NFATc1 is recognized as a well-characterized essential transcription factor required to activate osteoclastic-specific genes, including CTSK, TRAP, DC-STAMP, MMP9 and ATP6V0D2. These genes play substantial roles in the proliferation, differentiation, maturation and osteoclastic bone resorption of osteoclasts. 35,37,38 Our study shows that APS suppresses osteoclast formation and osteolysis via its broad-spectrum inhibition on MAPK signalling The ROS-induced osteoclast differentiation mechanism may be related to the effects of ROS on activating the downstream NF-kB signalling pathway and further activate osteoclast-related genes, such as NFATc1. 41,44,45 In summary, our findings demonstrate that APS acts as an inhibitor to reduce the activation of MAPK and ROS signalling pathways during osteoclastogenesis associated with RANKL induction (Figure 8).

| CON CLUS IONS
In conclusion, the current study proves that APS inhibits osteoclastogenesis and osteoclast function induced by RANKL/LPS in vitro and LPS-related inflammatory osteolysis in vivo. It has been clarified that APS exhibits suppressive effects on the MAPK and ROS signalling pathways, ROS production and the downstream factors responsible for osteoclastogenesis. In addition, APS also reduces LPS-induced inflammatory osteolysis in vivo by decreasing inflammatory factors' production. Therefore, our findings indicate that APS has the potential to become a therapeutic candidate for osteolytic-related diseases.

ACK N OWLED G EM ENTS
This study was funded by grants from the National Natural Science Foundation of China (NSFC), Grant number 81672167.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.