Patchouli Alcohol Modulates the Pregnancy X Receptor/Toll-like Receptor 4/Nuclear Factor Kappa B Axis to Suppress Osteoclastogenesis

Abstract

The incidence of osteoporosis, which is primarily characterized by plethoric osteoclast (OC) formation and severe bone loss, has increased in recent years. Millions of people worldwide, especially postmenopausal women, suffer from osteoporosis. The drugs commonly used to treat osteoporosis still exist many disadvantages, but natural extracts provide options for the treatment of osteoporosis. Therefore, the identification of cost-effective natural compounds is important. Patchouli alcohol (PA), a natural compound extracted from Pogostemon cablin that exerts anti-inflammatory effects, is used as a treatment for gastroenteritis. However, no research on the use of Patchouli alcohol in osteoporosis has been reported. We found that PA dose-dependently inhibited the receptor activator of nuclear factor kappa-B ligand (RANKL)-induced formation and function of OCs without cytotoxicity. Furthermore, these inhibitory effects were reflected in the significant effect of PA on the NF-κB signaling pathway, as PA suppressed the transcription factors NFATc1 and c-Fos. We also determined that PA activated expression of the nuclear receptor pregnane X receptor (PXR) and promoted the PXR/Toll-like receptor 4 (TLR4) axis to inhibit the nuclear import of NF-κB (p50 and p65). Additionally, PA exerted therapeutic effects against osteoporosis in ovariectomized (OVX) mice, supporting the use of PA as a treatment for osteoporosis in the future.

Keywords: PXR; nuclear factor κB; osteoclast; osteoporosis; patchouli alcohol; receptor activator for nuclear factor κB ligand.

FIGURE 1

Osteoclastogenesis was impaired by PA. (A) Representative images of TRAcP-positive cells after stimulation with PA at different on the indicated days (magnification = ×100). (B) TRAcP-positive cells in 96-well plates stimulated with PA at different concentrations on the indicated days were counted and analyzed (C, E) Representative images of TRAcP-positive cells after stimulation with 10 μM PA on the indicated days (magnification = ×100). (D) TRAcP-positive cells in 96-well plates stimulated with 10 μM PA on the indicated days were counted and analyzed. (F) A CCK-8 assay was performed to detect the cytotoxicity of PA against BMMs.

FIGURE 2

Bone resorption was impaired by PA. (A) Representative confocal images of F-actin and nuclei in OCs treated with or without 5 μM or 10 μM PA and subjected to immunofluorescence staining (scale bar = 200 μm). (B, C) Quantification of the average OC area and mean nuclear number in OCs. (D) Representative images of the bone resorption area in bone slices after OCs were treated with RANKL in the presence or absence of 5 μM or 10 μM PA (scale bar = 200 μm). (E) Quantification of the bone resorption area in bone slices.

FIGURE 3

PA inhibited RANKL-induced OC marker genes. (A–D) Real-time PCR was used to detect the mRNA expression levels of tartrate-resistant acid phosphatase (TRAcP/acp5), cathepsin K (Ctsk), c-Fos, and nuclear factor of activated T cells, cytoplasmic 1 (NFATc1) in the presence or absence of 5 μM or 10 μM PA. All of the data were normalized to data for the housekeeping gene Hprt. (E) Western blotting was performed to measure the protein content of V-ATPase-d2, CTSK, integrin β3, NFATc1, and c-Fos at day 0, day 1, day 3, and day 5 after stimulation with GST-rRANKL (50 ng/ml) with or without PA (10 μM). (F–J) Quantitative analysis of protein expression in the PA-stimulated and control groups at different times. The expression of all proteins was normalized to β-actin expression. The data represent the mean ± SD. Significant differences are indicated as follows: *p < 0.05, **p < 0.01, and ***p < 0.001.

FIGURE 4

PA suppressed RANKL-induced NF-κB signaling. (A) NF-κB luciferase activity in BMMs treated with PA at different concentrations. (B) Western blotting was performed to measure the protein content of p-P65 and IκB-α after 0, 10, 20, 30, and 60 min of stimulation with GST-rRANKL (50 ng/ml) with or without PA (10 μM). (C,D) Quantitative analysis of protein expression in the PA-stimulated and control groups at different times. The expression of all proteins was normalized to β-actin expression. (E) Representative confocal images of p65 in OCs treated with or without 10 μM PA subjected to immunofluorescence staining (scale bar = 200 μm). The nuclear translocation of p65 is indicated in this figure. (F) Quantitative analysis of the mean p65 fluorescence intensity in the nucleus after sample processing as described in (E).

FIGURE 5

PA activated PXR. (A) Nonbonding interactions between PA and PXR (Compound CID of PA: 10955174). (B) 3D structure of PXR and its binding site for PA. (C) Hydrogen bonds between PA and neighboring amino acid residues in PXR. (D) Real-time PCR was used to detect the mRNA expression levels of PXR after RANKL-induced OC differentiation for different durations. All data were normalized to data for the housekeeping gene Hprt. (E) Western blotting was performed to measure the protein content of PXR after 0, 10, 20, 30, and 60 min of stimulation with GST-rRANKL (50 ng/ml) with or without PA (10 μM). (F) Quantitative analysis of PXR protein expression in the PA-stimulated and control groups at different times. The expression of all proteins was normalized to β-actin expression.

FIGURE 6

PA activated its target, PXR, to suppress downstream NF-κB signaling. (A) Representative images of TRAcP-positive cells after transfection with siRNAs in the presence of PA (10 μM) (magnification = ×100). (B) TRAcP-positive cells transfected with siRNAs in the presence of PA (10 μM) in 96-well plates were counted and analyzed. (C) Representative images of the bone resorption area on hydroxyapatite-coated plates upon transfection with siRNAs in the presence of PA (10 μM) (magnification = ×100). (D) Quantification of the bone resorption area on hydroxyapatite-coated plates. (E) Western blotting was performed to measure the p65 protein content in the nucleus after transfection with siRNAs in the presence of PA (10 μM) for 1 h. (F) Quantitative analysis of p65 protein expression in the nucleus after transfection with siRNAs in the presence of PA (10 μM) for 1 h. The expression of all proteins was normalized to histone H3 expression. (G) Representative confocal images showing p65 after transfection with siRNAs in the presence of RANKL and PA (10 μM) under immunofluorescence staining (scale bar = 200 μm).

FIGURE 7

PA downregulated the TLR4/Myd88/TRAF6 axis. (A) The fold change in TLR4 mRNA expression after treatment with or without PA for 0, 1, 2, three or 6 h determined using real-time PCR. (B) Western blotting was performed to measure the TLR4, MyD88, and TRAF6 protein content after transfection with siRNAs in the presence of PA (10 μM) for 1 h. (C–E) Quantitative analysis of TLR4, MyD88, and TRAF6 protein expression after transfection with siRNAs in the presence of PA (10 μM) for 1 h. The expression of all proteins was normalized to β-actin expression.

FIGURE 8

PA inhibited OCs in the OVX mouse model. (A) 3D computer reconstruction of the femur and tibias in each group captured by a micro-CT instrument. (B–E) The relevant bone microstructure-related parameters were quantitatively measured: BV/TV, Tb. Sp., Tb. N, and Tb. Th. (F) Representative images of tibias stained with H and E and TRAcP. Scale bar = 500 mm; scale bar = 100 mm in the enlarged pictures. (G, H) Quantitative analyses of the OC number per bone surface (N.Oc/BS).

FIGURE 9

Schematic diagram showing the mechanism by which PA inhibits osteoclastogenesis.