Chrysophanic acid shifts the differentiation tendency of BMSCs to prevent alcohol‐induced osteonecrosis of the femoral head

Abstract Objectives Osteonecrosis of the femoral head (ONFH), largely caused by alcohol abuse, is a refractory bone disease characterized by the impaired capacity of osteogenic differentiation of bone mesenchymal stem cells (BMSCs), as well as the disordered adipocyte accumulation. Chrysophanic acid (CPA) is a natural anthraquinone which has lipid regulation and bone protection capacity. The aim of this study was to reveal the potential function of CPA and the underlying mechanisms for the alcohol‐induced ONFH. Materials and Methods The effects of alcohol and CPA on BMSCs were investigated by cell proliferation, induced differentiation assays and immunofluorescent staining. Meanwhile, the function of PI3K/AKT and AMPK pathway was investigated in the process of osteogenic and adipogenic differentiation, respectively. Furthermore, we established the rat model of alcohol‐induced ONFH to reveal the pharmacotherapeutic effect of CPA in vivo using radiographical and histopathological methods. Results In vitro, alcohol significantly inhibited the proliferation and osteogenic differentiation of BMSCs but stimulated the adipogenic differentiation. However, CPA could counteract the anti‐osteogenesis of alcohol partly via PI3K/AKT pathway and retard the promotion of alcohol‐induced adipogenesis via AMPK pathway. In vivo, radiographical and histopathological findings showed that CPA could alleviate alcohol‐induced ONFH and substantially restore the bone volume. Conclusions We demonstrated that CPA ameliorated alcohol‐induced ONFH possibly via regulating the differentiation tendency of BMSCs. Hence, CPA may become a beneficial herb extract to alleviate alcohol‐induced ONFH.

non-traumatic ONFH have been revealed, including corticosteroids medication, alcohol (AL) abuse, lipid metabolism disorders and autoimmune diseases. 3 Although alcohol drinking is one of the most common aetiologies for ONFH, especially in the eastern Asia, the exact pathway of alcohol-induced interruption on the bone homoeostasis of the femoral head needs further elucidation. 4,5 BMSCs have high proliferative potential and the ability to differentiate into osteoblasts, chondrocytes and adipocytes. 6 BMSCs are considered as the precursor cells of osteoblasts, thus critically influencing bone growth, regeneration and remodelling. 7 BMSCs pools are impaired, and the osteoblasts show significant abnormalities in patients with ONFH. 8 Therefore, some researchers hypothesized that ONFH might be a disease of BMSCs. Intriguingly, accumulating reports indicated that alcohol impairs bone homoeostasis via a direct inhibition on BMSCs. Particularly, alcohol significantly inhibits proliferation and DNA synthesis of the osteoprogenitor cells. 9 Moreover, alcohol critically retards the formation of mineralization nodes and the osteogenic differentiation of BMSCs. 10 On the contrary, alcohol tends to induce BMSCs into adipocytes. 11 Intriguingly, intramedullary fat accumulation and trabecular sparsity are essential pathological features of ONFH. 12,13 Therefore, the disturbed differentiation tendency of

BMSCs into osteoblasts and adipocytes might substantially contribute
to the aetiological foundation of the alcohol-induced ONFH.
Recently, an increasing interest has risen for the treatment of intractable bone diseases with traditional Chinese medicine (TCM). 14 Our previous studies have indicated that muscone, cordycepin and osthole have beneficial effects for ONFH. [15][16][17] CPA, also known as chrysophanol and 1,8-dihydroxy-3-methyl-anthraquinone, can be extracted from various herbs, including Rumex crispus, Polygonum multiflorum and Cassia obtusifolia. 18,19 CPA has been revealed to possess many beneficial pharmacological effects, such as lipid regulation, neuroprotection, anti-cancer, anti-inflammation, hepatoprotection and anti-obesity function. [19][20][21][22][23] It is worth noting that the ethanol extract of Rumex crispus root increases alkaline phosphatase (ALP) and bone nodule formation in MG-63 cells, 24 and the water extract of Rumex crispus root inhibits osteoclast differentiation via RANKL pathway and promotes osteoblast mineralization via extracellular signal-regulated kinase (ERK) phosphorylation. 24 Moreover, CPA remarkably inhibits adipogenesis via AMPK pathway both in vivo and in vitro. 25 Therefore, we hypothesized that CPA might exert a potential protection for the alcohol-induced ONFH via regulating the differentiation tendency of BMSCs. In this study, for the first time, we demonstrated that CPA could modulate the differentiation tendency of BMSCs, alleviate the morphological features of alcohol-induced ONFH and substantially restore the bone volume within the femoral head.

| Cell culture
We extracted BMSCs from tibias and femurs of 4-week-old Sprague-Dawley rats based on the previous protocol. 17,26 BMSCs were incubated in α-minimum essential medium (α-MEM) (Gibco) with 10% foetal bovine serum (FBS) (Gibco), 1% penicillin/streptomycin (Invitrogen) at 37°C with 5% CO 2 in a humidified atmosphere. The cells used in the whole study were between the 3 rd to the 7 th passage.

| Osteogenesis assay
1 × 10 5 of BMSCs were seeded in 24-well plates. When cells came to 80% confluence, osteogenic medium (Cyagen) was used to induce osteogenic differentiation and refreshed every 2 days. Alizarin red staining (ARS) was conducted and quantified at Day 21. ALP staining and its activity detection were performed at Day 7 and Day 14. All images were captured by a LEICA microscope, and the ALP activity was detected at 560 nm by a multifunctional microplate reader.

| Adipogenesis assay
2 × 10 5 cells were seeded in 6-well plates. When cells reached 80% confluence, adipogenic medium (Cyagen) was used to induce adipogenic differentiation and refreshed according to the manufacturer's protocol. Oil Red O staining was conducted and quantified at Day 21.

| Quantitative real-time polymerase chain reaction assay (QPCR)
Bone mesenchymal stem cells (BMSCs) were incubated in respective medium for 48 h, and then, total RNA was extracted based on the manufacturer's protocol (EZBioscience). Reverse transcriptase reactions included the purified total RNA and 50 nM RT primer, with M-MLV reverse transcriptase (EZBioscience) used. QPCR was performed by an ABI 7900T system. GAPDH was set as the reference gene of RNA expression. The primers are listed in Table 1.

| Western blotting
In brief, BMSCs were lysed by cell RIPA lysis buffer (Beyotime) with Scanning densitometry was used to quantify the signals.

| Micro-CT scanning
All rats were sacrificed under the general anaesthesia, and bilateral femoral heads were harvested and fixed in 4% paraformaldehyde. All femoral heads were scanned by a micro-CT scanner (SkyScan 1176).
The CTVol software was used to manage and reconstruct the images.

| Histological and immunohistochemical staining
Femoral heads were decalcified, embedded in paraffin, sectioned into 5 μm thick slices and stained with haematoxylin and eosin (H&E). For immunohistochemical staining, slides were deparaffinized, antigen retrieved, blocked and incubated with the primary antibodies of COL I and OCN, and the relevant biotinylated secondary antibodies. Finally, slides were stained with DAB and further counterstained with haematoxylin.

| Apoptosis assay
Cellular apoptosis within the femoral head was detected by TdTmediated dUTP nick-end labelling (TUNEL) technique. When the section was deparaffinized and antigen retrieved, the TUNEL staining was undertaken based on the manufacture's protocols (Beyotime).

| Statistical analysis
All data were shown as means ± standard error of the mean (SEM).
Statistical differences were determined by the Bonferroni's post hoc test (one-way ANOVA) or Student's t test in SPSS 18 (IBM). * P < .05, ** P < .01, and *** P < .001 were considered to have statistical significance. and yielded the best rescue effect on Day 5 and Day 7. Therefore, we used 10 μΜ CPA in the following experiment ( Figure 1A).

| CPA abolished the inhibition of alcohol on osteogenic differentiation of BMSCs
Subsequently, the osteogenesis assay indicated that alcohol could remarkably inhibit the ALP activity at Day 7 and Day 14, while co-treatment with CPA could substantially abolish the inhibition of alcohol ( Figure 1B).
As the osteogenic-associated markers, the gene expression of COL I, OCN and OPN of BMSCs was distinctly decreased by alcohol, while CPA could reverse the inhibition and promote genes expression to a much higher level ( Figure 1C-1E). Moreover, the protein expression of OCN and Runx2 were remarkably decreased by alcohol; however, the inhibitory effect was distinctly rescued by supplementary CPA ( Figure 1F). Immunofluorescence staining demonstrated that the intensity of COL I and OCN was apparently decreased by alcohol; however, CPA could reverse the alcohol-induced inhibition on COL I and OCN expression of BMSCs (Figure 2A and 2B).
Finally, the ARS and ALP staining indicated that both mineralization nodes and ALP activity were significantly decreased by alcohol, while CPA treatment could critically restore the capacity of osteogenesis when BMSCs were cocultured in 50 mM ethanol. ( Figure 3A).

PI3K/AKT signalling pathway is vital to cell development and differ-
entiation. The results demonstrated that 50 mM alcohol dramatically decreased the level of PI3K and phosphorylated AKT. However, cotreatment with CPA could restore the expression of PI3K and p-AKT within BMSCs against alcohol ( Figure 3B). LY294002, a selective inhibitor, was used to block PI3K/AKT pathway to detect its role in the osteoprotective effect of CPA. It was revealed that 500 nM LY294002 distinctly decreased the upregulation of PI3K and p-AKT when alcohol-treated BMSCs were further cultured with CPA ( Figure 3C). Next, ARS and ALP staining were conducted to observe the role of PI3K/AKT signalling in morphological performance. The results showed that LY294002 dramatically abrogated the preventive effect of CPA on the alcohol-induced loss of extracellular mineralization ( Figure 3D).

| CPA retarded the alcohol-irritated adipogenesis via AMPK pathway
AMPK signalling pathway has a key role in the fatty acid metabolism.
Western blot demonstrated that alcohol remarkably decreased the level of p-AMPK ( Figure 4F). However, co-treatment with CPA could reverse the inhibition of alcohol on p-AMPK in BMSCs ( Figure 4F). The result indicated that DMP dramatically abolished the preventive effect of CPA on the alcohol-induced extracellular adipogenesis in BMSCs ( Figure 4H).

| CPA alleviated alcohol-induced ONFH in the rat model
The experimental procedure was illustrated in Figure 5A. of BMSCs and the potential of osteogenic differentiation were obviously decreased. 30,[33][34][35] In the present study, the results showed that the proliferation of BMSCs was inhibited by alcohol, and the osteogenic differentiation of BMSCs was dramatically retarded.
Abundant evidence has accumulated indicating that PI3K/AKT pathway is a central regulator in the osteogenesis of BMSCs. [36][37][38][39] Notably, PI3K/AKT pathway is involved in the osteogenic differentiation of alcohol-induced bone loss in our previous work as well. 28 Our results indicated that both levels of PI3K and phosphorylated CPA is a quite safe drug, even the high dose (30 μg/mL) of which has no significant disturbance in chromosomal aberrations, 49 and the accumulation of CPA metabolites is quite low in the liver and kidney. 50 Our results also demonstrated that CPA was well tolerated as evidenced by no changes in daily behaviour, survival rate and body weight in rats.
In summary, CPA is convincingly demonstrated to be beneficial to the alcohol-induced ONFH, largely due to its capacity of shifting the differentiation tendency of BMSCs, and to ameliorate radiological and histological features of the femoral head in rats. Hence, CPA may become a potential pharmacotherapeutic herb extract for the prevention of the alcohol-induced ONFH.

ACK N OWLED G EM ENTS
This study was supported in part by the National Natural Science Foundation of China (81672143, 81802247 and 81902237) and Shanghai Jiao Tong University, School of Medicine (DLY201616).

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interests. Chen and Y. Gao wrote the manuscript. C. Zhang and Y. Gao critically revised the manuscript. All authors approved the final manuscript.

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

F I G U R E 7
The protective effects of CPA against alcohol-induced ONFH in the rat model. There was significantly decreased new bone formation in alcohol-treated group revealed by the fluorochrome labelling, while the pharmacotherapy with CPA critically restored bone formation capacity O RCI D Youshui Gao https://orcid.org/0000-0001-9242-2486