Isopsoralen suppresses receptor activator of nuclear factor kappa-β ligand-induced osteoclastogenesis by inhibiting the NF-κB signaling

Reagents

Isopsoralen (purity >98%, Fig. 1A) and the MTT kit were purchased from Solebao Company (Beijing, China). Alpha-modified Eagle’s medium (α-MEM), fetal bovine serum (FBS), penicillin, and streptomycin were purchased from Gibco (Rockville, MD, USA). TRIzol reagent was purchased from Tiangen (Beijing, China). The SYBR Green Master Mix was purchased from Imgenex (Littleton, CO, USA). Recombinant M-CSF and RANKL were obtained from R&D Systems (Minneapolis, MN, USA). The TRAP staining kit was purchased from Sigma Aldrich (St. Louis, MO, USA). Anti-P65, anti-phospho-P65, anti-IκB α, and anti-phospho-IκB α antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-MMP-9, anti-P50/P105, and anti-phospho-P50/P65 antibodies were purchased from Abcam (Cambridge, UK). Anti-GAPDH antibody was obtained from ABclonal Technology (Wuhan, Hubei, China). Anti-NFATc1 antibody was obtained from AiFang Biological (Changsha, Hunan, China). The anti-CTSK antibody was obtained from Proteintech Group (Wuhan, Hubei, China).

Ethical use of animal

Male C57BL/6 mice, 4 to 5 weeks-old, were purchased from the Comparative Medical Center of Yangzhou University. All the animal experiments were designed and approved by the Yangzhou University Laboratory Animal Ethics Committee (yzu-lcyxy-n059). All animals were raised in the Yangzhou University Center for Comparative Medicine, with different groups raised in different cages. The experimental environment was managed by the Animal Center: a room temperature of 26 °C and 12 h light-dark cycles were maintained, and the breeding environment was cleaned regularly by professionals. All mice were euthanized with anesthetic at the end of the experimental. After freezing treatment, the experimental animal carcasses were sealed and handed over to the Animal Carcass Processing Center of Yangzhou University for centralized harmless treatment. There were no redundant experimental animals in this experiment.

Cell culture and cell viability assay

Bone marrow monocytes (BMMCs) were extracted from whole bone marrow of the mice as described previously (Hu et al., 2008). In brief, femurs of four to five week-old male C57BL/6 mice were taken. After the bone marrow cavity was exposed, the medium was aspirated with a sterile one mL syringe, and the bone marrow cavity was thoroughly flushed until it turned white. After collecting the bone marrow cavity rinse solution, the rinse solution was filtered into a new 50mL centrifuge tube using a 100 µm filter. After centrifugation at 2000 rpm for 5 min, the cells were resuspended using primary osteoclast medium supplemented with 30ng/mL M-CSF (Yang et al., 2022; Liu et al., 2017), and the cells were evenly spread into a 24-well plate. Unadherent cells were washed with PBS after 48 h, and the remaining cells were BMMCs. BMMCs were supplemented with 10% (v/v) FBS, 1% (w/v) penicillin/streptomycin and 30 ng/mL M-CSF in α-MEM complete medium in an incubator at 37 °C and 5% CO₂ until ready for further applications. To verify the cytotoxic effect of isopsoralen on BMMCs, cell proliferation and viability were measured using an MTT assay. BMMCs were seeded in 96-well plates at a density of 3 × 10³ cells/well and incubated with different concentrations of isopsoralen (10, 30, 50, and 100 µM) and M-CSF (30 ng/mL) for 24, 48, and 72 h. MTT solution (10 µL/well) was added to each well, and cells were incubated for 4 h. The absorbance was measured at 490 nm using a microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).

In vitro osteoclast formation assay

Whether isopsoralen affects the differentiation of BMMCs was observed using TRAP staining. After the BMMCs were extracted into 24-well plates, and the cells had proliferated to 70% confluence, the differentiation of BMMCs was stimulated with 50 ng/mL RANKL (Momiuchi et al., 2021; Yin et al., 2019) and cultured with different concentrations (10, 20, and 30 µM) of isopsoralen to generate mature multinucleated osteoclasts. The cells were then fixed with 4% paraformaldehyde for 1 min, washed with sterile double-distilled water, and TRAP incubation solution was added, followed by incubation at 37 °C for 60 min in the dark. Cells that stained dark red, containing three or more nuclei, were mature osteoclasts.

Quantitative real-time PCR (qRT-PCR)

Real-time quantitative PCR was used to study the expression of specific genes involved in osteoclast differentiation induced by isopsoralen. BMMCs were lysed with TRIzol and total RNA was extracted. One microgram of total RNA was reverse-transcribed into single-stranded cDNA using oligo-dT primers (Promega Corporation, Madison, WI, USA). qRT-PCR was performed using the SYBR Green Master Mix. The relative mRNA levels were consistent with the expression levels of the housekeeping gene, GAPDH. All data were analyzed using the 2^−ΔΔCT method. Primer sequences were shown in Table 1.

Western blot

In order to analyze the effect of protein related to osteoclast differentiation, BMMCs were extracted from 24-well plates. After cell proliferation reached 70%, BMMCs differentiation was stimulated by 50 ng/ml RANKL and cultured with isopsoralen at different concentrations (10, 20, and 30 µM). After reaching this time point, cell lysates were collected. Whole cell lysates were separated by SDS-PAGE and transferred to nitrocellulose membranes. Nonspecific binding was blocked with 5% nonfat milk for 1 h, and the membrane was incubated with the appropriate primary antibody overnight at 4 °C. The membrane was washed three times with TBS-Tween (TBST) and incubated with the appropriate HRP-conjugated secondary antibody for 1.5 h. The color was developed and exposed, and the gray value of the bands was analyzed using Image J software with ECL reagents and a chemoluminescence imaging system.

Bioinformatics analysis

After predicting the target genes of isopsoralen using the default settings of STITCH, Cytoscape 3.8.2 software was used to construct the interaction network of isopsoralen and its target genes. The enrich KEGG language package in the R-Studio software was used to enrich and analyze the signaling pathways of isopsoralen target genes. We extracted the data from the GSE176265 database (this database uses RANKL to induce primary mouse mononuclear macrophages, and collects cells for RNA-seq on the 0th, 1st, 2nd, and 3rd days of induction), we converted the gene ID, DEGs were screened according to P < 0.05 and —LogFC—>2. KEGG signaling pathway enrichment analysis of DEGs during osteoclast-induced differentiation was performed using R-Studio software. Subsequently, the Venn diagram online tool, Venny 2.1 (http://bioinfogp.cnb.csic.es/tools/venny/index.html) was used to obtain isopsoralen-targeted genes shared with DEGs during the osteoclast-induced differentiation KEGG pathway analysis. The enrichment information for the top five KEGG pathways with the smallest p-value was presented using GOplot, an R language package that combines the visualization of expression data with functional analysis. The genes involved in all five KEGG pathways were regarded as core genes. We searched the osteoclast differentiation signaling pathway map on the KEGG official website to identify the signaling pathways involved in the core genes.

Statistical analysis methods

All basic experiments were performed independently at least three times. SPSS software (version 22.0) was used for statistical analysis, and GraphPad Prism 8 software was used to draw graphs. Quantitative data were described as mean ± standard deviation, independent samples t-test was used for comparison between two groups, and one-way analysis of variance was used for comparison between multiple groups. Statistical significance was set to p < 0.05*, p < 0.01**, or p < 0.001***.

Effects of isopsoralen on the proliferation of BMMCs

To probe the effect of isopsoralen on the proliferation and viability of BMMCs, we used an MTT kit to detect cell proliferation and viability (Fig. 1B). The results show that treatment with isopsoralen (<50 µM) for 24, 48, or 72 h did not reduce the viability of BMMCs (P > 0.05). When the concentration of isopsoralen was greater than or equal to 50 µM, cell viability decreases gradually with increasing isopsoralen concentration (P < 0.05). The results of this experiment determined the maximum concentration of isopsoralen required for the subsequent detection of the effect of isopsoralen on RANKL-induced osteoclast differentiation.

Isopsoralen inhibits RANKL-induced osteoclast differentiation

To explore whether isopsoralen could inhibit osteoclast differentiation, we plated BMMCs in 24-well plates in the presence of 0, 10, 20, and 30 µM isopsoralen by adding 30 ng/mL M-CSF; BMMCs were treated with 50 ng/mL RANKL. In vitro osteoclast differentiation was assessed by TRAP staining after five days, and TRAP-positive cells were quantified. Compared to the RANKL-induced group, isopsoralen reduced the formation of mature osteoclasts in a dose-dependent manner (Figs. 2A and 2B).

To further verify the effect of isopsoralen on osteoclast differentiation, BMMC cells treated as described above. After five days of induction, cell proteins were extracted for Western blot experiments. After five days of induction, isopsoralen had a stronger inhibitory effect on the expression of NFATc1, CTSK, and MMP9 proteins in osteoclasts at increasing concentrations (Fig. 2C). We also extracted cellular RNA to further verify that isopsoralen inhibited the mRNA expression of NFATc1, CTSK, and MMP9 in osteoclasts (Fig. 2D). The results were similar to those of Western blot experiments. Isopsoralen inhibited the expression of osteoclast-related genes induced by RANKL in a concentration-dependent manner.

Target genes, interaction networks and KEGG analysis of isopsoralen

To explore the mechanism by which isopsoralen inhibits osteoclast differentiation, we used the default search method of the STITCH database to find 30 proteins and compounds that may act on isopsoralen, 28 of which are isopsoralen target genes. Cytoscape 3.8.2 software was used to establish an interaction network (Fig. 3A). The first layer, from the inside to the outside, was the interaction between isopsoralen and protein, and the second layer was the interaction between protein and protein or other compounds. Cytoscape 3.8.2 software analysis obtained the degree, closeness, and betweenness of each target gene in the interaction network. The degree of freedom represents the degree to which a node is related to all other nodes in the network, closeness centrality represents the distance between a node and other nodes in the network (Yu et al., 2020), and spacing centrality represents the shortest node between the other two nodes (Lee, Choi & Cho, 2018). The next step of visualization and analysis was performed according to the degree of freedom (Fig. 3B) (the darker the color and the larger the circle, the higher the degree of freedom). HDAC1, NFκB1, and NFκB2 are the three genes with the highest degrees of freedom.

We used the language package in the R-Studio software to perform KEGG signaling pathway enrichment analysis for isopsoralen target genes and draw bubble plots (Fig. 3C). A total of 87 signaling pathways (P < 0.05) were enriched. Ranking them from small to large, the top five were the NF-κB signaling pathway, Toll-like receptor signaling pathway, C-type lectin receptor signaling pathway, osteoclast differentiation signaling pathway, and Epstein-Barr virus infection signaling pathway.

DEGs and KEGG analysis of induced differentiation of osteoclasts

We downloaded the GSE176265 dataset (transcriptomic analysis of osteoclastogenesis) from the GEO database and analyzed RNA-Seq data on days 0 and 3 after induction of BMMCs with RANKL through the language package in R-Studio software. After setting the screening criteria at P < 0.05, and —LogFC—>2, we obtained 174 differentially expressed genes, of which 111 were downregulated and 63 were upregulated; they were heatmapped using the pheatmap R language package (Fig. 3D).

We used the language package in the R-Studio software to analyze the KEGG signaling pathway enrichment of DEGs during osteoclast-induced differentiation and drew a bubble plot (Fig. 3E). A total of 41 signaling pathways (P < 0.05) were enriched. Ranking the p-value from small to large, the top five are the neutrophil extracellular trap formation signaling pathway, systemic lupus erythematosus signaling pathway, cell cycle signaling pathway, alcoholism signaling pathway and programmed necrosis signaling pathway.

KEGG pathway shared by isopsoralen target genes and DEGs during osteoclast-induced differentiation

We obtained 17 isopsoralen target genes and DEGs in the osteoclast-induced differentiation process using the Venn Diagram online tool from the previously obtained isopsoralen target gene KEGG pathway and KEGG pathway of osteoclast-induced differentiation process DEGs. According to the order of p-value from small to large (Table 2), the first five were the EB virus infection signaling pathway, TNF signaling pathway, IL-17 signaling pathway, human T-cell leukemia virus 1 infection signaling pathway and apoptosis signaling pathway.

Isopsoralen inhibits RANKL-induced osteoclastogenesis by inhibiting NF-κB signaling pathway

We analyzed the top five shared KEGG signaling pathways and related genes through GOplot R package visualization and enrichment analysis to obtain the NFKB1 (P50), TNF, RELA (P65), NFKBIA (IκBα), IKBKB (IKK–β), and CHUK (IKK–α) genes that are involved in all five KEGG pathways, and therefore, the core genes (Fig. 4A). After bringing the above six core genes into the osteoclast signaling pathway map, we found that the focus was on the NF-κB signaling pathway (Fig. 4B); therefore, we predicted that isopsoralen inhibits osteoclasts generation by inhibiting the NF-κB signaling pathway.

After RANKL binds to its receptor, RANK, it recruits TRAF6 to its cytoplasmic domain, which subsequently activates signaling pathways such as MAPK and NF-κB to induce the expression of osteoclastogenesis genes (Lee et al., 2005). We examined the effect of isopsoralen on activation of the NF-κB signaling pathway. The analysis of experimental results shows (Fig. 4C) that the levels of p-P65, p-P50, and p-IκB α in RANKL-treated BMMCs were significantly increased, while isopsoralen significantly decreased p-P65, p-P65 and p-IκB α in a time-dependent manner. These data suggested that isopsoralen inhibits RANKL-induced osteoclastogenesis by inhibiting the NF-κB signaling pathway.