Involvement of AKT/β-catenin in Enhanced Osteogenesis by Cannabidiol in Bone Stromal Cells Harvested from Human Tori

Abstract

Cannabidiol (CBD), a major cannabinoid of Cannabis sativa, has been utilized for several medical purposes, such as, an anti-resorptive activity that may benefit treatment of alveolar bone destruction in periodontitis. This study aimed to investigate osteogenic effects of CBD in primary bone stromal cells harvested from jaw tori of healthy patients. To determine cytotoxicity of CBD, the bone stromal cells were treated with various doses of CBD for 24 or 48 h and then analyzed by an alamarBlue^® assay. No cytotoxicity was found in these cells treated with CBD up to 10 µM. Enhanced osteogenic differentiation and biomineralization of the cells treated with non-toxic doses of CBD were determined by alkaline phosphatase staining and Alizarin Red and von Kossa staining, respectively, and confirmed by upregulated mRNA expressions of runt-related transcription factor 2 (RUNX2), bone sialoprotein (BSP), and Osterix. Treatment with CBD significantly enhanced osteogenic differentiation and biomineralization and upregulated mRNA expressions of RUNX2, BSP, and Osterix (p < 0.05). Pretreatment with MK-2206, an AKT inhibitor, or with CWP232228, a β-catenin antagonist, significantly decreased the increased Alizarin Red staining and the upregulated expressions of BSP and Osterix (p < 0.05), suggesting involvement of AKT/β-catenin in osteogenic induction upon treatment with CBD.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-03763-5.

Introduction

Cannabis sativa has been widely used in medicine for centuries. Its two principal components are ∆⁹-tetrahydrocannabinol (THC), with a distinctive intoxicated effect, and cannabidiol (CBD) that does not cause any intoxication¹. CBD is currently one of the most studied cannabinoids because it possesses a wide range of therapeutic effects, including anti-inflammatory, immunomodulatory, and analgesic², antinociceptive and antiproliferative properties³. Regarding the association between CBD and bone biology, a previous study has shown that the type 2 cannabinoid receptor, a prime target for CBD, regulates enhanced bone loss in ovariectomized mice⁴. Correspondingly, another study has demonstrated that CBD can reduce bone resorption, while enhancing healing of bone fracture⁵. In addition, CBD is considered an effective treatment for arthritis via an oral administration⁶.

Periodontitis, a common chronic inflammatory disease of periodontal tissues, is associated with accumulated bacterial plaque biofilm that induces a myriad of host inflammatory responses, eventually leading to destruction and loss of tooth supporting structures, particularly alveolar bone⁷. Bone remodeling is a physiological process to maintain human bone mass by a homeostatic balance between bone resorption and deposition by osteoclasts and osteoblasts, respectively. This balance is disrupted during the progression of periodontitis by shifting towards more bone resorption than deposition, resulting in continuous destruction of alveolar bone. Consequently, it is of our great interest to search for a phytoactive compound that exerts its osteogenic effect on human alveolar bone stromal cells, resulting in enhanced bone formation. Because CBD can exert the anti-resorptive effect, while possibly promoting osteoblast function, CBD could be potentially used as an adjunctive treatment for periodontitis, where promotion of osteogenesis is required to cope with increased alveolar bone loss.

Osteoblasts, differentiated from mesenchymal stromal cells, synthesize organic matrix, followed by induction of mineralization on that matrix⁸. The classical or “canonical” osteogenic differentiation pathway, controlled by wingless-related integration site (Wnt) and bone morphogenic proteins, is involved with the β-catenin signaling⁹. This pathway activates runt-related factor 2 (RUNX2) and osterix (OSX), which induce expressions of osteogenic genes, such as, bone sialoprotein (BSP)⁹. Moreover, activation of Ak strain transforming (AKT) results in accumulation of β-catenin in the cytoplasm that later moves into the nucleus to induce osteogenic gene transcription¹⁰, which finally promotes osteoblast differentiation. In contrast to the canonical pathway, the non-canonical pathway, also activated by Wnt ligands, does not involve β-catenin, but instead involves the Ca²⁺/nuclear factor-κB (NF-κB) pathway¹¹. With the promotion of bone fracture healing by CBD as aforementioned⁵, it was hypothesized in this study that CBD could induce osteogenesis of primary human bone stromal cells harvested from jaw tori via an AKT/β-catenin-dependent mechanism. Therefore, this study aimed to determine the osteogenic effect of CBD as well as its underlying molecular mechanism.

Methods

Chemicals and antibodies

CBD (C₂₁H₃₀O₂) with the molecular weight of 314.46 Da was obtained from Cerilliant^® Corporation (#C-045; Round Rock, TX, USA) and already dissolved in absolute methanol at a stock concentration of 1 mg/mL¹². Ascorbic acid (#A4544), β-glycerophosphate (#G9422), dexamethasone, and dimethyl sulfoxide (DMSO; #D4902) were from Sigma-Aldrich, Co. (St. Louis, MO, USA), while MK-2206 (#HY-10358), an AKT inhibitor, and CWP232228 (#HY-18959), an antagonist of β-catenin function, were from MedChemExpress (Monmouth Junction, NJ, USA). Anti-phospho-AKT (serine 473; #4060) and anti-total AKT (#4691) antibodies were bought from Cell Signaling Technology, Inc. (Danvers, MA, USA); anti-β-catenin antibody was from Abcam (#ab32572; Cambridge, UK); anti-β-actin (#sc-47778) and anti-Lamin A/C (#sc-7292) antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Culture of human bone stromal cells

The bone stromal cells were harvested from jaw tori of twelve healthy patients (9 males and 3 females; mean age = 68.5 ± 12.17 years) during a surgical removal of tori prepared for prosthodontic treatment under local anesthesia. The protocol of this study was approved by Mae Fah Luang University Ethics Committee on Human Research, Mae Fah Luang University (EC 23134-19). All experiments involving human participants were performed in accordance with the Declaration of Helsinki’s guidelines and regulations. Written informed consent was obtained from each participant prior to collection of bony specimens. Isolation of the bone stromal cell was performed as previously described¹³. Briefly, the specimens were transported to the laboratory in a sterile container, containing serum-free Dulbecco’s modified Eagle medium (DMEM; #11885-084; Gibco by Life Technologies, Ltd., Paisley, UK), supplemented with 1% (v/v) penicillin/streptomycin (#15140-122; Gibco) and 1% (v/v) amphotericin B (#15290026; Gibco). The specimens were washed with HEPES-buffered saline several times to remove blood cells until the buffer was clear, transferred to a sterile Petri dish, and sequentially digested by incubating in 1 mg/mL of Collagenase/Dispase^® solution (#10269638001; Roche Diagnostics GmbH, Mannheim, Germany) in DMEM twice, in 0.25% EDTA trypsin (#25200072; Gibco) once, and in the Collagenase/Dispase^® solution another time at 37 °C for 30 min each incubation. Subsequent to incubation with each digestion solution, the specimens were flushed and their supernatant was collected and pooled in the same 50-mL tube, containing an equal volume of DMEM, supplemented with 10% (v/v) fetal bovine serum (FBS; #A5256701; Gibco). The supernatant was centrifuged at 1400g for 5 min and was carefully removed leaving the cell pellet to be resuspended in DMEM, supplemented with 10% (v/v) FBS and 1% (v/v) penicillin/streptomycin, in a 75-cm² culture flask. Cells were incubated at 37 °C in a humidified incubator with 5% CO₂, and medium replacement was done every 2–3 days.

Characterization of human bone stromal cells

The bone stromal cells from passages 2 to 5, used to determine the osteogenic effects of CBD, were first characterized for their potential to differentiate and mineralize extracellular matrix after treatment with osteoblast induction medium, as mentioned below. ALP, Alizarin Red, and von Kossa staining together with mRNA expressions of osteogenic genes were analyzed by standard cell staining methods and reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR), respectively, using a specific oligonucleotide primer pair for RUNX2, ALP, BSP, or COL1A1¹⁴, and for glyceraldehyde 3-phosphate dehydrogenase (GAPDH), as a housekeeping gene¹⁴. The oligonucleotide sequences of primers are summarized in Table 1.

Table 1.

Oligonucleotide primers used in this study.

Gene	Forward 5′- -3′	Reverse 5′- -3′	Accession Number
RUNX2	GCCTTCAAGGTGGTAGCCC	CGTTACCCGCCATGACAGTA	NM_001024630.4
ALP	ACTGGTACTCAGACAACGAGAT	ACGTCAATGTCCCTGATGTTATG	NM_001369804.2
BSP	GCAGTAGTGACTCATCCGAAGAA	GCCTCAGAGTCTTCATCTTCATTC	NM_004967.4
COL1A1	GATTCCCTGGACCTAAAGGTGC	AGCCTCTCCATCTTTGCCAGCA	XM_054315082.1
OSX	TGCTTGAGGAGGAAGTTCAC	CTGCTTTGCCCAGAGTTGTT	XM_054371078.1
GAPDH	GTCTCCTCTGACTTCAACAGC	ACCACCATGTTGCTGTAGCCAA	NM_002046.7

Cell viability assay

The bone stromal cells were seeded in 96-well sterile black plates with clear bottom at 1 × 10⁴ cells/well in 100 µL of DMEM, supplemented with 10% (v/v) FBS and 1% (v/v) penicillin/streptomycin, and incubated in a humidified incubator with 5% CO₂ overnight. The medium was then removed, and the cells were incubated in the medium, containing CBD at various concentrations from 0.01, 0.03, 0.1, 0.3, 1, 3, 10, to 30 µM, at 37 °C for 24 or 48 h. Cell viability was evaluated by a fluorometric assay to measure the cell metabolic activity using alamarBlue dye^® (#BUF012B; Bio-Rad Laboratories, Hercules, CA, USA). After treatment with CBD for indicated timepoints, a 10-µL volume of alamarBlue dye^® was added to each well and incubated at 37 °C for 3 h. The fluorescence signal was measured by a multimode microplate reader (EnVision^® 2105, Perkin Elmer, Shelton, CT, USA) with excitation and emission wavelengths at 530 and 590 nm, respectively. The signal from the cells treated with 1% (v/v) methanol as a vehicle control was set at 100% of cell viability.

Cell treatment

The bone stromal cells were seeded in 24-well culture plates at 1 × 10⁵ cells/well or in 6-well culture plates at 2 × 10⁵ cells/well and incubated at 37 °C in a humidified incubator with 5% CO₂. At 80% cell confluence, the medium was removed. In the experiment for analysis of mRNA expressions, the cells were incubated in complete medium, i.e., DMEM with 10% (v/v) FBS, 1% (v/v) streptomycin/penicillin, 50 µg/ml of ascorbic acid, and 10 mM β-glycerophosphate. The reason for RT-qPCR experiments, conducted in the cells incubated in the complete medium that did not contain dexamethasone, was because treatment with dexamethasone already induced osteogenic gene expressions until the inducible effect of CBD on osteogenic gene upregulation could not be observed (see a supplementary RT-qPCR file). In the staining experiments for analysis of differentiation and mineralization, the cells were incubated in the complete medium or in the complete medium together with 0.1 µM dexamethasone, and called osteogenic medium¹⁵. Note that treatment with dexamethasone can upregulate RUNX2 expression and enhance its activity in bone marrow stromal cells, resulting in enhanced osteogenesis¹⁶. For the inhibitor studies, the staining experiments were conducted in the cells, incubated only in the osteogenic medium to simplify the experimental design.

For the inhibitor experiments, the bone stromal cells cultured in the complete or osteogenic medium were pretreated with MK-2206 at 0.03, 0.3, or 3 µM, with CWP232228 at 0.05 or 0.1 µM, or with 0.1% (v/v) DMSO, as a vehicle control for the two inhibitors, for 30 min prior to treatment with non-toxic doses of CBD. The medium was replenished every 3 days. The cytotoxicity of various doses of MK-2206 and of CWP232228 in the presence of treatment with CBD at 1 µM for 24 or 48 h was first determined by the alamarBlue dye^® assay, and the non-toxic doses of MK-2206 up to 3 µM and those of CWP232228 up to 0.1 µM were selected for the inhibitor studies in Figs. 4 and 5, and 6 (see a supplementary cytotoxicity file).

Fig. 4.

(A) Representative immunoblots showing phospho-AKT and total AKT expressions in the bone stromal cells treated with indicated doses of CBD for 2 days. (B) A bar graph showing significant increases in phospho-AKT/total AKT ratios in CBD-treated cells, compared to that in the methanol-treated cells, set to 1. (C) Representative immunoblots showing phospho-AKT and total AKT expressions in the cells pretreated with MK-2206 at 0.03, 0.3, or 3 µM for 30 min prior to treatment with CBD at 1 µM for 2 days. (D) A bar graph showing significant decreases in phospho-AKT/total AKT ratios in the inhibitor-treated cells, compared to that in the CBD-treated cells, set to 1. (E) Representative images of ALP, Alizarin Red, and von Kossa staining in the bone stromal cells pretreated with MK-2206 or DMSO prior to treatment with CBD at 1 µM for 7 and 14 days, respectively. Scale bars = 3.5 mm. Bar graphs showing the mean relative areas of positive staining for ALP (F), Alizarin Red (G), and von Kossa (H). (I) A bar graph demonstrating the mean degrees of mRNA expressions for RUNX2, BSP, and OSX, normalized by that of GAPDH, in the bone stromal cells pretreated with MK-2206 or DMSO prior to treatment with CBD at 1 µM for 2 days, compared to those in the methanol-treated cells, set to 1. Error bars = standard deviation; n = 3; *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 5.

(A) Representative immunoblots showing β-catenin expression in the cytosolic and nuclear extracts of bone stromal cells upon treatment with indicated doses of CBD for 2 days. β-actin and Lamin A/C were used as an internal control for the cytosolic and nuclear extract, respectively. Bar graphs showing significant increases in β-catenin/β-actin ratios (B) and β-catenin/Lamin A/C ratios (C) in the CBD-treated cells, compared to those in the methanol-treated cells, set to 1. (D) Immunofluorescence images showing induction of β-catenin expression localized in the cytoplasm and nucleus of CBD-treated cells for 2 days. Scale bars = 50 μm. (E) Representative immunoblots showing β-catenin and Lamin A/C expressions in the nuclear extract of cells pretreated with CWP232228 at 0.1 µM for 30 min prior to treatment with CBD at 1 µM for 2 days. (F) A bar graph showing a significant decrease in β-catenin/Lamin A/C ratio in the inhibitor-treated cells, compared to that in the CBD-treated or the CBD- and DMSO-treated cells. (G) Representative images of ALP, Alizarin Red, and von Kossa staining in the bone stromal cells pretreated with CWP232228 or DMSO prior to treatment with CBD 1 µM for 7 and 14 days, respectively. Scale bars = 3.5 mm. Bar graphs showing the mean relative areas of positive staining for ALP (H), Alizarin Red (I), and von Kossa (J). (K) A bar graph demonstrating the mean degrees of mRNA expressions for RUNX2, BSP, and OSX, normalized by that of GAPDH, in the bone stromal cells pretreated with CWP232228 or DMSO prior to treatment with CBD at 1 µM for 2 days, compared to those of the methanol-treated cells, set to 1. Error bars = standard deviation; n = 3; *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 6.

(A) Representative immunoblots showing β-catenin and Lamin A/C expressions in the nuclear extract of the bone stromal cells pretreated with MK-2206 at 0.03, 0.3, 3 µM or DMSO prior to treatment with CBD at 1 µM for 2 days. (B) A bar graph demonstrating the mean β-catenin/Lamin A/C ratios in the inhibitor-treated cells, compared to that in the CBD-treated cells, set to 1. Error bars = standard deviation; n = 3; **p < 0.01. (C) Representative immunoblots showing β-catenin and Lamin A/C expressions in the nuclear extract of the bone stromal cells only treated with MK-2206 at 0.03, 0.3, 3 µM or DMSO for 2 days. (D) A bar graph demonstrating the mean β-catenin/Lamin A/C ratios in the inhibitor-treated cells, compared to that in CBD-untreated cells, set to 1. Error bars = standard deviation; n = 3.

Cell staining methods

After being treated for 7 days, the bone stromal cells were fixed with 4% (v/v) paraformaldehyde in phosphate-buffered saline (PBS) at 4 °C for 15 min, followed by washing with PBS twice. For ALP staining, a 0.5-mL volume of CHAP buffer, containing 100 nM Tris, pH 9.5, 100 mM NaCl, and 50 mM MgCl₂ in distilled water, was added for 30 min at room temperature in the dark. After removal of the buffer, the cells were stained with 0.5 mL of BCIP/NBT (5-bromo-4-chloro-3-indolyl phosphate 4-toluidine salt #11383221001/4-nitroblue tetrazolium chloride solution #11383213001; Roche Diagnostics GmbH) solution in distilled water for 30 min at room temperature in the dark, followed by washing twice with PBS for 5 min each.

After cell treatment for 14 days, mineralization in culture was determined by Alizarin Red and von Kossa staining to monitor calcium and phosphate ion deposits, respectively. For Alizarin Red staining, the fixed cells were stained with 2% (w/v) Alizarin Red solution (#A5533; Sigma-Aldrich, Co.), pH 4.1–4.3, for 45 min at room temperature. Thereafter, the cells were rinsed several times with distilled water to remove the remaining stains, followed by rinsing with PBS to stop the reaction. For von Kossa staining, a 0.5-mL volume of 1% (w/v) silver nitrate solution (#1.01512.0100) was added and incubated for 3 min at room temperature in the dark. The solution was then aspirated, and the cells were rinsed three times with distilled water, followed by addition of a 0.5-mL volume of 5% (w/v) sodium carbonate (#1.06392.1000) in 10% (v/v) formaldehyde for 3 min at room temperature in the dark. After washing with distilled water three times, a 0.5-mL volume of 5% (w/v) disodium thiosulphate (#217263) solution was added for 3 min at room temperature in the dark, followed by washing with distilled water three times. The digital images of ALP, Alizarin Red and von Kossa staining were captured by a stereomicroscope (Olympus SZ61; Olympus, Inc., Tokyo, Japan). The percentage of positive staining area in each condition was analyzed by ImageJ software (NIH, Bethesda, MD, USA).

RNA extraction and RT-qPCR

After cell treatment for 2 days, total RNA was harvested using the RNAspin Mini Isolation kit (#25050072; Cytiva^™ Amersham, Buckinghamshire, UK) according to the manufacturer’s protocol. The amounts of total RNA were quantified by a NanoDrop^™ One spectrophotometer (ThermoFisher Scientific, Madison, WI, USA) at 260 nm and 280 nm wavelengths. The 260/280 ratios of all samples were found to be at around 2.0 (data not shown). Thereafter, one µg of total RNA from each sample was converted into complementary DNA (cDNA) using the RevertAid cDNA Synthesis kit (#K1632; ThermoFisher Scientific, Vilnius, Lithuania). qPCR was performed using the SensiFAST^™ SYBR NO-ROX kit (#BIO-98005; Meridian Bioscience^®, Cincinnati, OH, USA) with a specific primer pair for RUNX2, ALP, BSP, COL1A1, OSX, or GAPDH (Table 1). PCR was conducted for 40 cycles with the denaturing, annealing, and polymerizing temperatures at 95 °C, 60 °C, and 72 °C, respectively, using the CFX Opus 96 Real-time PCR system (Bio-Rad Laboratories). The melting curve analysis was performed after PCR and showed no non-specific primer dimer (data not shown). Expressions of the osteogenic genes were normalized by that of GAPDH, as a relative Ct (ΔCt). In Fig. 1, the relative gene expression (ΔΔCt) was obtained by comparing ΔCt between the cells cultured in osteogenic medium and those in complete medium, set to 1.0. In Figs. 3 and 4, and 5, the ΔΔCt was obtained by comparing ΔCt of CBD-treated, of CBD- and inhibitor-treated, and of CBD- and DMSO-treated samples with that of the methanol control, set to 1.0.

Fig. 1.

Osteogenesis of primary bone stromal cells from human tori. (A) Representative images showing enhanced staining of alkaline phosphatase (ALP), Alizarin Red, and von Kossa upon being incubated in osteogenic medium for 7 and 14 days, respectively, compared with complete medium. Scale bars = 3.5 mm. (B) A bar graph demonstrating significantly upregulated mRNA expressions of runt-related transcription factor 2 (RUNX2), alkaline phosphatase (ALP), bone sialoprotein (BSP), and collagen type I alpha 1 (COL1A1), normalized by expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH), at day 2. Error bars = standard deviation; n = 3; *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 3.

Representative images of ALP staining (A), Alizarin Red staining (C), and von Kossa staining (E) in the bone stromal cells cultured in complete or osteogenic medium in the presence or absence of treatment with CBD at 1, 3, or 10 µM for 7 and 14 days, respectively. Scale bars = 3.5 mm. Bar graphs showing the mean relative areas of positive staining for ALP (B), Alizarin Red (D), and von Kossa (F), compared with those of the methanol-treated cells cultured in complete medium, set to 1. (G) Bar graphs demonstrating the mean degrees of mRNA expressions for RUNX2, BSP, and osterix (OSX), normalized by that of GAPDH, in the bone stromal cells treated with CBD at 1, 3, or 10 µM for 2 days, compared with those in the methanol-treated cells cultured in complete medium, set to 1. Error bars = standard deviation; n = 3; *p < 0.05; **p < 0.01; ***p < 0.001.

Protein extraction and Immunoblotting

After cell treatment for 2 days, whole cell lysates were harvested using RIPA buffer¹², supplemented with cOmplete Mini protease inhibitor cocktails (#11836170001; Roche Diagnostics GmbH), while nuclear and cytosolic protein fractions were extracted using an NE-PER^® nuclear and cytoplasmic extraction kit (#78833; Pierce, Rockford, IL, USA), following the manufacturer’s protocol. A 10-µg quantity of nuclear protein and a 20-µg quantity of cytosolic or total protein were separated by 10% SDS-PAGE and transferred to nitrocellulose membrane. The membrane was blocked with 5% non-fat dry milk in 0.1% (v/v) Tween-20 (Bio-Rad Laboratories) in Tris-buffered saline (TBS) and incubated with anti-phospho-AKT (1:2,000), anti-total AKT (1:1,000), anti-β-catenin (1:5,000), anti-β-actin (1:1,000), or anti-Lamin A/C (1:500) antibody at 4 °C overnight. The membrane was incubated with horseradish peroxidase-conjugated secondary antibody (1:2000) and reacted with LumiGLO Reserve^™ Chemiluminescent substrate (#5430-0050; SeraCare Life Sciences, Inc., Milford, MA, USA). Immunoreactivity signals were captured by the ChemiDoc^™ MP Imaging system (Bio-Rad Laboratories). Band intensities of phospho-AKT and those of cytosolic and nuclear β-catenin were measured by Scion Image program (Scion Corporation, Frederick, MD, USA) and normalized by those of total AKT, cytosolic β-actin, and nuclear Lamin A/C, respectively. The degrees of normalized expressions were reported as ratios in the experimental samples relative to the methanol control or the CBD-treated sample, set to 1.0.

Immunofluorescence

To localize β-catenin in the cytoplasm or in the nucleus, the bone stromal cells were seeded on coverslips in a 48-well culture plate and treated with indicated doses of CBD for 2 days. The treated cells were fixed and reacted with rabbit anti-β-catenin (1:250) antibody overnight, followed by incubation with NorthernLights™ 557 anti-rabbit IgG antibody (1:2,000; #NL004; R&D Systems, Inc., Minneapolis, MN, USA) and 1 µM DAPI (#40011; Biotium, Inc., Hayward, CA, USA). Subsequently, the coverslips were mounted using fluorescence mounting medium (#002627; Fluoromount G, Electron Microscopy Sciences, Hatfield, PA, USA), and fluorescence images were visualized and captured by a fluorescence microscope (Nikon Eclipse Ni-E, Nikon Corp., Kanagawa, Japan).

Statistical analysis

Each experiment was conducted in triplicate for each of the three distinct cell lines, isolated from three different donors. All data were found to be normally distributed by Shapiro-Wilk test. One-way ANOVA and Post-hoc analysis were used to determine statistically significant differences between CBD-treated groups and the methanol control group or between inhibitor-pretreated groups and the CBD-treated group at a p-value less than 0.05. The statistical analyses were performed using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA).

Results

Characterization of osteoblast precursors

To first characterize the differentiation and mineralization potentials of primary bone stromal cells isolated from human tori, they were cultured in either complete or osteogenic medium for 14 days. It was found that the intensities of ALP, Alizarin Red, and von Kossa staining were enhanced in these cells cultured in the osteogenic medium, compared with those cultured in the complete medium (Fig. 1A). By RT-qPCR, significantly upregulated mRNA expressions of RUNX2, ALP, BSP, and COL1A1 were found in these cells cultured in the osteogenic medium, compared with those cultured in the complete medium (p = 0.0028, p = 0.00001, p = 0.0345, and p = 0.00001, respectively; Fig. 1B).

Treatment with CBD enhances differentiation, mineralization in culture, and osteogenic gene expressions in human bone stromal cells

The viability of bone stromal cells upon treatment with various doses of CBD was first checked by an alamarBlue^® assay. Treatment with CBD at any concentrations from 0.01 to 10 µM for 24–48 h was found not to be toxic to the bone stromal cells, compared with those treated with 1% (v/v) methanol control (Fig. 2A or B). However, treatment with CBD at 30 µM significantly decreased the mean percentages of cell viability at 24 and 48 h (p = 0.00001 and p = 0.00001, respectively; Fig. 2A and B). These findings agree with the cell viability results of our recent study in primary human gingival fibroblasts¹². Therefore, the non-toxic doses of CBD at 1, 3, and 10 µM were chosen for subsequent experiments to demonstrate the osteogenic effects of CBD.

Fig. 2.

Bar graphs demonstrating the mean percentages of cell viability upon treatment with CBD at indicated doses (0.01-30 µM) for 24 h (A) or 48 h (B), compared with 1% (v/v) methanol-treated cells, set to 100%. Error bars = standard deviation; n = 3; ***p < 0.001.

Treatment with CBD at 10 µM enhanced ALP staining in the bone stromal cells cultured in either complete or osteogenic medium, compared with the methanol control (Fig. 3A). However, by ImageJ analysis, the mean relative area of positive ALP staining was only found to be significantly increased in the bone stromal cells cultured in the osteogenic medium and treated with CBD at 10 µM, compared with the methanol control (p = 0.040; Fig. 3B). As expected, the intensities of ALP staining were stronger in the cells cultured in the osteogenic medium than those cultured in the complete medium (Fig. 3A), consistent with greater mean relative areas of positive ALP staining in the cells cultured in the osteogenic medium than those cultured in the complete medium (Fig. 3B).

Moreover, treatment with CBD enhanced mineralization of the bone stromal cells cultured in the osteogenic medium in a dose-dependent manner, as shown by Alizarin Red (Fig. 3C) and von Kossa (Fig. 3E) staining, compared with the methanol control. Significant increases in the mean relative areas of positive staining were found in the bone stromal cells treated with CBD at 3 or 10 µM for Alizarin Red (p = 0.024 or p = 0.0134; Fig. 3D) and for von Kossa (p = 0.037 or p = 0.031; Fig. 3F). As with ALP staining, the intensities of Alizarin Red and von Kossa staining were stronger in the cells cultured in the osteogenic medium than those cultured in the complete medium (Fig. 3C and E, respectively), consistent with greater mean relative areas of positive staining for Alizarin Red (Fig. 3D) and von Kossa (Fig. 3F) in the cells cultured in the osteogenic medium than those cultured in the complete medium. Note that we did not test whether or not treatment with CBD alone without the addition of dexamethasone, ascorbic acid, and β-glycerophosphate was able to induce osteogenic differentiation and mineralization in culture, but treatment with CBD at 10 µM in the presence of ascorbic acid and β-glycerophosphate (or complete medium) slightly increased ALP, Alizarin Red, and von Kossa staining (Fig. 3A, C, and E, respectively).

To further verify the differentiation and mineralization effects of treatment with CBD in human bone stromal cells, the degrees of osteogenic mRNA expressions were determined by RT-qPCR. Compared with the methanol control, treatment with CBD at 1 µM significantly induced mRNA expressions of RUNX2 (p = 0.0100), BSP (p = 0.00001), and OSX (p = 0.0009; Fig. 3G), that at 3 µM significantly induced mRNA expressions of RUNX2 (p = 0.0073) and OSX (p = 0.0005; Fig. 3G), and that at 10 µM significantly induced mRNA expressions of RUNX2 (p = 0.0475), BSP (p = 0.0428), and OSX (p = 0.0007; Fig. 3G).

Involvement of AKT/β-catenin in enhanced osteogenesis upon treatment with CBD in human bone stromal cells

The involvement of AKT activation by phosphorylation was first determined for the underlying mechanisms of osteogenic enhancement by treatment with CBD. Treatment with CBD at 1, 3, or 10 µM induced a phosphorylated form of AKT (phospho-AKT) at the serine 473 residue in the whole cell lysates of bone stromal cells (Fig. 4A), consistent with significant increases in the band intensities of phospho-AKT, normalized by those of total AKT (phospho-AKT/total AKT), upon treatment with CBD at 1 (p = 0.0357), 3 (p = 0.0299), or 10 (p = 0.0023) µM (Fig. 4B). The efficacy of MK-2206 was then examined, which demonstrated a dose-dependent decrease in phospho-AKT expression by pretreatment of the bone stromal cells with indicated doses of MK-2206 (Fig. 4C). By densitometry, the mean degree of phospho-AKT expression, normalized by that of total AKT expression, was found to be significantly reduced by pretreatment with MK-2206 at 1, 3, or 10 µM (p = 0.00001; Fig. 4D). Subsequently, the inhibitory effects of pretreatment with MK-2206 on enhancement of differentiation, mineralization in culture, and osteogenic gene expressions upon treatment with CBD at 1 µM in the bone stromal cells were determined. Pretreatment with MK-2206 was found to decrease only the enhanced Alizarin Red staining upon CBD treatment in a dose-dependent fashion (Fig. 4E), with significant decreases in the mean relative area of positive staining by the dose of MK-2206 at 3 µM, compared with CBD treatment alone (p = 0.0480) or with the DMSO vehicle control (p = 0.0498; Fig. 4G). However, there was no apparent decrease in enhanced ALP or von Kossa staining by pretreatment with MK-2206 at any doses (Fig. 4E), corresponding with the non-significant differences in the mean relative areas of positive staining between MK-2206-pretreated and then CBD-treated samples and CBD treatment alone (Fig. 4F for ALP staining; Fig. 4H for von Kossa staining). In addition, pretreatment with MK-2206 at 3 µM significantly reduced the upregulated expressions of BSP and OSX, but not RUNX2, upon treatment with CBD at 1 µM (p = 0.037 and p = 0.038; Fig. 4I) or with both CBD and DMSO (p = 0.019 and p = 0.016; Fig. 4I).

To further explore the molecular mechanisms of osteogenic induction pathway by treatment with CBD, involvement of β-catenin was investigated. Treatment with indicated doses of CBD induced expression of β-catenin in both cytosolic and nuclear extracts (Fig. 5A). By densitometry, the mean degree of β-catenin expression relative to that of β-actin expression (β-catenin/β-actin) in the cytosolic extract was significantly increased by treatment with CBD at 3 (p = 0.0008) or 10 (p = 0.0015) µM (Fig. 5B), while the mean degree of β-catenin expression relative to that of Lamin A/C expression (β-catenin/Lamin A/C) in the nuclear extract was significantly increased by treatment with CBD at 10 µM (p = 0.0107; Fig. 5C). By immunofluorescence, β-catenin expression was found to be more intense in CBD-treated bone stromal cells than in the methanol-treated cells (Fig. 5D). The efficacy of CWP232228, an antagonist that inhibits β-catenin function in the nucleus, was examined. Pretreatment with CWP232228 at 0.1 µM reduced the upregulated expression of β-catenin in the nuclear extract upon treatment with CBD at 1 µM (Fig. 5E) with significant decreases found when compared with CBD treatment alone (p = 0.0050) or with the DMSO control (p = 0.0222; Fig. 5F). As with the inhibitory effect of pretreatment with MK-2206 on enhanced Alizarin Red staining by treatment with CBD, pretreatment with indicated doses of CWP232228 diminished the enhanced Alizarin Red staining in a dose-dependent manner (Fig. 5G) with significant decreases found by pretreatment with CWP232228 at 0.1 µM when compared with CBD treatment alone (p = 0.0127) or with the DMSO control (p = 0.0169; Fig. 5I). However, no discernible reduction in enhanced ALP or von Kossa staining was observed by pretreatment with CWP232228 at 0.05 or 0.1 µM (Fig. 5G), corresponding with non-significant differences in the mean relative areas of positive staining between CWP232228-pretreated and then CBD-treated samples and CBD treatment alone (Fig. 5H for ALP staining; Fig. 5J for von Kossa staining). Moreover, pretreatment with CWP232228 at 0.1 µM significantly inhibited the upregulated expressions of BSP and OSX, but not RUNX2, upon treatment with CBD at 1 µM (p = 0.0001 and p = 0.0482; Fig. 5K) or with both CBD and DMSO (p = 0.0014 and p = 0.0454; Fig. 5K).

To demonstrate β-catenin as a downstream molecule of AKT activation, as reported in the SW620 colorectal cancer cell line¹⁷, expression of β-catenin in the nuclear extract was investigated after pretreatment with the AKT inhibitor. Pretreatment with indicated doses of MK-2206 partially reduced the upregulated β-catenin expression upon treatment with CBD at 1 µM in the nuclear extract in a dose-dependent fashion (Fig. 6A). A significant decrease in the β-catenin expression was found by pretreatment with MK-2206 at 3 µM, when compared with CBD treatment alone (p = 0.0057; Fig. 6B). Notably, there were no noticeable differences observed for the degrees of β-catenin accumulation in the nuclear extract of CBD-untreated bone stromal cells, treated with different doses of MK-2206 (Fig. 6C), consistent with no significant differences found for the ratios of β-catenin to Lamin A/C expression among different samples (Fig. 6D). The non-significance differences imply that the inhibitory effect on accumulation of nuclear β-catenin upon AKT inactivation is more pronounced in the cells treated with CBD.

Discussion

Our in vitro study demonstrated that treatment with non-toxic doses of CBD up to 10 µM significantly enhanced differentiation, mineralization in culture, and osteogenic gene expressions, as assayed by ALP staining, Alizarin Red and von Kossa staining, and RT-qPCR, respectively, only in osteogenic medium but not in complete medium, indicating that CBD enhances osteogenesis of alveolar bone stromal cells. These findings agree with induction of mineralization by treatment with CBD in several osteoblastic cell lines, including SaOS-2, MG63, and U2OS¹⁸. Nevertheless, primary human bone stromal cells tested in this study should be a better representative of physiological responses of enhanced osteogenesis than the aforementioned immortalized osteoblastic cell lines. Note that large variations observed for the inducible effects of CBD on differentiation and mineralization in culture (Figs. 3, 4 and 5) that are not found in the previous study¹⁸, using the immortalized bone cell lines, may have resulted from use of primary human bone cells isolated from different donors.

Furthermore, expressions of active phospho-AKT and β-catenin were significantly enhanced by treatment with CBD in the bone stromal cells, particularly an increased accumulation of β-catenin in their nuclear fraction. The enhancement of phospho-AKT expression by CBD treatment in the bone stromal cells is in line with the previous results, demonstrating that CBD can induce apoptosis in a human breast cancer cell line, MCF-7, through FAK/MAPK/AKT/NF-κB signaling¹⁹, while the upregulated expression of β-catenin by treatment with CBD is similar to the inductive effect of CBD on β-catenin expression in a rat pheochromocytoma cell line, PC-12²⁰. The discovery that CBD promotes osteogenesis via the AKT/β-catenin pathway may offer new therapeutic targets for bone regeneration research.

Since treatment with CBD at 1 µM already enhanced both phospho-AKT and β-catenin expressions; therefore, this dose was subsequently selected for the inhibition studies. Note that the observed reduction of phosphorylated AKT expression upon treatment with CBD at 10 µM (Fig. 4B), although expression of total AKT remained unchanged (Fig. 4A), may be because this dose is close to the cytotoxic dose at 30 µM (Fig. 2). To first verify the efficiency of MK-2206 and CWP232228, it was demonstrated that pretreatment with MK-2206 suppressed the upregulated expression of phospho-AKT upon treatment with CBD in a dose-dependent manner, whereas that with CWP232228 at 0.1 µM significantly decreased the CBD-enhanced nuclear accumulation of β-catenin. In addition, pretreatment with MK-2206 partially reduced the nuclear accumulation of β-catenin upon treatment with CBD, implying that β-catenin acts as a downstream molecule of AKT activation for CBD-enhanced osteogenesis in the bone stromal cells. The connection between AKT and β-catenin observed in this study is concordant with the previous findings, showing that A Disintegrin and Metalloprotease 9 activates the Wnt/β-catenin signaling through AKT activation in a colorectal cancer cell line, SW620¹⁷.

However, it is noteworthy that pretreatment with either MK-2206 or CWP232228 significantly decreased only enhanced Alizarin Red staining or upregulation of BSP and OSX genes, but not the RUNX2 gene, as an early osteogenic marker²¹, upon CBD treatment. These findings suggest an involvement of AKT and β-catenin during the late stage of CBD-enhanced osteogenesis of the bone stromal cells. Expression of BSP, whose product is a major non-collagenous bone sialoprotein, plays a central role in de novo bone formation by enhancing calcification, which can be reflected by increased Alizarin Red staining in vitro, during the late stage of osteogenesis²², while OSX is known to involve in late osteoblast maturation and activity in addition to its early osteoblast differentiation effect²³. Notably, OSX can transcriptionally upregulate BSP gene expression²⁴, emphasizing the interconnection between OSX and BSP. In addition to the involvement of AKT and β-catenin in CBD-enhanced osteogenesis of the bone stromal cells, implications of downstream transcription factors in the non-canonical osteogenic pathway, such as, NF-κB¹¹, which have not yet been determined in this study, warrants further investigations.

Another limitation of this study is a lack of an assessment to demonstrate the direct impact of treatment with CBD on AKT phosphorylation and β-catenin transcriptional activity at an earlier timepoint than 2 days, e.g., 1 h, because the observed increases in β-catenin expression in the nuclear and cytosolic fractions upon treatment with CBD (Fig. 5A-C) may have resulted from overall upregulation of β-catenin expression rather than increased transcriptional activity. Therefore, the short-term treatment with CBD should be further conducted in a future study. Moreover, a lack of investigation into the enhanced osteogenesis of CBD in an animal model to validate the in vitro results prior to being tested in clinical trials for treatment of periodontitis is considered another limitation. Last but not least, the long-term effects and a safety profile as well as the suitable dosages and synergistic effects of CBD with other bone regeneration materials commercially available in dentistry still remain unexplored.

However, the enhanced mineralization of CBD in the bone stromal cells harvested from tori shown in our in vitro study is consistent with the in vivo findings in extraoral bones^4,25, which have demonstrated that CBD treatment by intraperitoneal administration improved bone mineral density and increased bone matrix in osteopenic femurs and vertebrae, while decreased osteoclastic activity in ovariectomized rats, as a model of postmenopausal osteoporosis. It is worthwhile to note that CBD, but not THC, enhances the biomechanical properties of healing rat mid-femoral fractures for 8 weeks²⁶. Furthermore, in a case report of two postmenopausal women with osteopenia, oral administration of CBD is associated with reduction in serum markers of bone turnover²⁷. Recently in a randomized controlled trial, it has been shown that CBD can attenuate carboxyl-terminal collagen crosslinks, a marker of bone resorption²⁸. With these in vivo findings and our in vitro results, it is proposed that CBD possesses an induced osteogenic effect together with a reduced osteoclastic activity.

Besides the enhanced differentiation and mineralization effects of CBD in the bone stromal cells, CBD exerts its other biological effects in other oral mesenchymal cells, such as, human dental pulp cells²⁹, gingival mesenchymal cells³⁰, primary human gingival fibroblasts¹². Treatment with CBD at 1–5 µM and that at 3–10 µM enhanced proliferation of human dental pulp cells²⁹ and human gingival fibroblasts¹², while that at 0.1-2 µM and that at 1–10 µM enhanced osteogenic differentiation of human dental mesenchymal stem cells³¹ and the bone stromal cells shown in this study, respectively. Note a little difference in the doses of CBD used to exert its various biological activities in different cell types, which may have been explained by distinct degrees of cellular responses to CBD treatment.

Conclusion

This in vitro study demonstrates the induced differentiation and mineralization effects of CBD in primary alveolar bone stromal cells isolated from human tori, involving the AKT and β-catenin signaling molecules. It is probable that CBD would be beneficial for future clinical trials in dentistry, especially for bone regeneration as a novel therapy for patients affected with periodontitis. However, additional in vivo experiments are still required to determine an appropriate dose of CBD and its sustained release form within nanoparticles, incorporated into the natural or synthetic scaffold. Moreover, involvement of other signaling molecules should be further explored.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

A. M. and S.K. were responsible for conceptualization, project administration, resources and funding acquisition, and an original draft. M. T., K. M., P. D., J. P., P. S., and S. S. investigated and analysed formal data. A. M. and E. D. were responsible for data curation. A. M. prepared six figures and one table. S. K. was responsible for project supervision.All authors reviewed and edited the manuscript.

Funding

Financial support from Intramural Endowment Fund, School of Dentistry, Mae Fah Luang University, Chiang Rai, to A.M. and S.K. (#681B08010) and from Research Grant for New Scholar, National Research Council of Thailand (NRCT), #N42A660843, 2023, to A.M. is gratefully acknowledged.

Data availability

The datasets generated and/or analyzed in the current study are not publicly available because we do not have consent from all patients, whose bone specimens were provided for all experiments conducted in this study, to publicly open their raw data. However, these data are still available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Conflict of interest

All authors declare no potential conflict of interest.