Liraglutide Promotes the Osteogenic Differentiation in MC3T3-E1 Cells via Regulating the Expression of Smad2/3 Through PI3K/Akt and Wnt/β-Catenin Pathways

Abstract

Diabetes is a worldwide health problem with increasing prevalence. Some reports indicate the interplay between bone and glucose metabolism. The imbalance between bone resorption and formation resulted in the structural integrity and strength of bone. Glucagon-like peptide-1 (GLP-1) and its agonists (Liraglutide) have an anabolic action on bone remodeling by stimulating osteoblast differentiation as well as increasing osteoblast longevity. However, the underlying mechanisms remain elusive. We detected the presence of GLP-1 receptor (GLP-1R) in MC3T3-E1 cells via immunocytochemistry assay. Alkaline phosphatase activity assay, alizarin red stain, quantitative real-time polymerase chain reaction, and western blot were employed to detect the effect of Liraglutide on osteogenic differentiation. Liraglutide promoted the expression of GLP-1R in a dosage- and time-dependent manner, and it enhanced the osteogenic differentiation in MC3T3-E1 cells. Liraglutide application improved the levels of Smad2/3 and p-Smad2/3; however, the silencing of Smad2/3 blocked the osteogenic differentiation induced by Liraglutide. What is more, the application of PI3K and Wnt inhibitors inhibited the upregulation of Akt, p-Akt, β-catenin, Smad2/3, and p-Smad2/3 induced by Liraglutide. Liraglutide facilitated the osteogenic differentiation via the regulation of Smad2/3 via PI3K/AKT and Wnt/β-catenin pathways. These data revealed a new mechanism of Liraglutide inducing osteogenic differentiation and provided theory evidence to maintain normal bone metabolism during diabetes therapy.

Introduction

Diabetes continues to affect an increasing proportion of the population worldwide. Some evidence shows defective bone quality and a higher risk of fractures in the diabetic conditions (Vestergaard, 2014). So bone also exerts an important target organ for diabetes treatment. Clinical studies indicate that some diabetic patients receiving hypoglucemic treatments are at high risk for bone fractures (Mannucci and Dicembrini, 2015). The imbalance between osteoclast and osteoblast activity is the key pathogenesis of bone metabolic disorder (Vellucci et al., 2018). Thus, restoration of this balance and prevention of bone destruction by inhibiting bone-resorbing osteoclast or inducing bone-forming osteoblast functions highlight potential new means for therapeutic intervention.

Incretin hormone is a kind of peptide that is secreted in the gastrointestinal tract in response to nutrient intake and it potentiates insulin release in a glucose-dependent manner (Nauck and Meier, 2018). Some evidence showed that incretin hormones have an anabolic action on bone remodeling by stimulating osteoblast differentiation as well as increasing osteoblast longevity (Luo, et al., 2016; Ramsey and Isales, 2017). Glucagon-like peptide-1 (GLP-1) is a key incretin hormone secreted from gut L cells. GLP-1 stimulates insulin secretion by pancreatic β cells in a glucose-dependent way and suppresses glucagon release from α cells (Kazafeos, 2011). GLP-1 exerts its activities via the interaction with GLP-1 receptor (GLP-1R), which is widely distributed in bone (Campbell and Drucker, 2013). However, GLP is rapidly degraded and inactivated by the ubiquitous protease dipeptidyl peptidase IV (DPP-IV) in the circulation. So GLP-1R agonists (GLP-1RA) are synthesized for clinical use due to an extended half-life with their resistance to degradation by DPP-IV (Ryan et al., 2011). Liraglutide belongs to a synthetic GLP-1RA with a prolonged half-life (Coulter et al., 2018). Liraglutide shares 97% homology with the structure of human GLP-1 (Gallwitz, 2015). It is synthesized on the basis of a GLP-1 backbone, where arginine replaces lysine 34 and a palmitoyl fatty acid is added at position 26 via a glutamic acid spacer (Russell-Jones, 2009). Liraglutide is widely used in the treatment of diabetes (Lambadiari et al., 2018) and obesity to decrease weight gain (Camkurt et al., 2018). What is more, Liraglutide is reported to relieve cardiac dilated function (Boyle et al., 2018) and reduce oxidative stress (Rizzo et al., 2015) in diabetic patients.

Previous research reveals that Liraglutide is involved in facilitating bone formation and suppressing bone resorption. In diabetic Goto-Kakizaki rats (Sun et al., 2015), type 1 diabetic mice (Mansur et al., 2015), ovariectomized rats (Lu et al., 2015), and type 2 diabetes and obesity (Iepsen et al., 2015), Liraglutide significantly prevents deterioration of the quality of the bone matrix. Although some studies have been performed to determine the beneficial anabolic effects of Liraglutide on skeletal health, the downstream molecular mechanisms underlying the osteogenic effects still remain elusive. It was reported that Liraglutide facilitates osteogenic proliferation and differentiation of MC3T3-E1 cells via PI3K/AKT, ERK1/2, and cAMP/PKA pathways (Wu et al., 2017). It is also reported that Liraglutide attenuates the osteoblastic differentiation through AMPK/mTOR signaling in MC3T3-E1 cells (Hu et al., 2016). Further, Liraglutide represses serum deprivation-induced apoptosis in MC3T3-E1 cells through cAMP/PKA/β-catenin and PI3K/AKT/GSK3β pathways (Wu et al., 2018).

Transforming growth factor-β (TGF-β) is directly or indirectly involved in a variety of cellular functions, including cell proliferation, differentiation, and apoptosis (Yan et al., 2018). Smad proteins are the major mediators of the TGF-β pathway to regulate the transcription of their targets (Chen et al., 2018). Smad2/3, as critical components of TGF-β signaling, are closely associated with the regulation of bone formation (de Kroon et al., 2017). In our study, we aimed at exploring the relationship between Liraglutide and the TGF-β/Smad pathway to reveal the new mechanism of Liraglutide functioning. During the study, we found that Liraglutide regulated GLP-1R expression in a dosage- and time-dependent manner in MC3T3-E1 cells. We also found that Liraglutide modulated an increase in the alkaline phosphatase (ALP) activity and bone matrix mineralization, and it promoted the expression of osteogenic markers runt-related transcription factor-2 (RUNX2) and osteocalcin (OCN). Liraglutide enhanced the levels of Smad2/3, p-Smad2/Smad2, and p-Smad3/Smad3; however, the silencing of Smad2/3 blocked the osteogenic differentiation induced by Liraglutide. The inhibitors of PI3K and Wnt suppressed the expression of Smad2/3 and β-catenin induced by Liraglutide. Taken together, these data revealed that Liraglutide was involved in the osteogenic differentiation via regulating the expression of Smad2/3, and Liraglutide modulated the levels of Smad2/3 through PI3K/AKT and Wnt/β-catenin pathways. These findings provided new evidence for the essential function of Liragultide in bone metabolism therapy.

Materials and Methods

Cell culture

Osteocytic cell line MC3T3-E1 was purchased from Yingniurui Biological Medicine Technology Co Ltd. (Wuxi, Jiangsu, China). MC3T3-E1 cells were cultured with alpha minimal essential medium (α-MEM; Hyclone) supplemented with 10% fetal bovine serum (FBS; Hyclone), 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified incubator of 5% CO₂. The culture medium was changed every 2–3 days.

Cell proliferation ability detection by cell counting kit 8

MC3T3-E1 cells were seeded into a 96-well plate at the density of 2 × 10³ cells/well. After 24, 48, and 72 h, 10 μL of cell counting kit 8 (CCK8) stock solution were added to each well at each time point, and the plates were further incubated for 4 h at 37°C. The OD values were detected at the wavelength of 450 nm in a microplate reader (Thermo). Survival rate (%) = OD_{experimental group}/OD_{control group} × 100%.

Cell cycle distribution analyzed by flow cytometry

MC3T3-E1 cells were seeded into a six-well plate at 5 × 10⁶ cells/well. After 24 h, cells were cultured with 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide. The control group was administered with the same volume of physiological saline. After 72 h of incubation, cells were washed three times with phosphate-buffered saline (PBS). After digestion and centrifugation, the cells were resuspended with 0.5 mL PBS and treated with 3.5 mL 70% pre-cooling ethanol overnight at 4°C. The next day, the cells were washed three times with PBS, and they were treated with 1 mL of propidium iodide (KeyGEN Biotech, Nanjing, China) containing 0.2 mg of RNase A for 15 min at 37°C. Cell cycle was tested by a flow cytometer (BectonDickinson).

Immunocytochemistry

MC3T3-E1 cells were seeded into a six-well plate with cover slips at 1 × 10⁴ cells/well. After 24 h, the cells were cultured with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide. The control group was administered with the same volume of physiological saline. After 48 h of incubation, the cells were fixed with 4% paraformaldehyde for 30 min, and they were permeabilized with 0.3% Triton X-100 in PBS at room temperature for 10 min. After being washed with PBS, the cells were blocked with 5% bovine serum albumin for 30 min at 37°C. Then, the cells were incubated with rabbit polyclonal anti-GLP-1R antibody (ABclonal Biotechnology Co., Ltd.; 1:500 dilution) at 4°C overnight. The next morning, after being washed with PBS, the cells were probed with Alexa Fluor 488-labeled sheep anti-rabbit IgG (Invitrogen; 1:10,000 dilution) for 2 h at room temperature. After PBS washes, the nuclear was stained with 4′,6′-diamino-2-phenylindole (DAPI, 1 mg/mL; Sigma) for 10 min at room temperature. Images from five non-overlapping fields were captured by using an inverted laser confocal microscope (Leica, Germany).

Quantitative real-time polymerase chain reaction

MC3T3-E1 cells were seeded into six-well plates. After 24 h of incubation, the cells were treated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide for 3, 7, 14, and 21 days, respectively. The control group was incubated with the α-MEM supplemented with 10% FBS. Total RNA was extracted with TRIzol reagent (CWBIOtech, Beijing, China). RNA was reverse transcribed with GoTaq^®Reverse Transcription System (Promega, Madison, WI). Polymerase chain reaction (PCR) was performed with GoTaq^®Green Master Mix (Promega) in a Light Cycler instrument (BioRad). The relative expression of target genes was normalized to GAPDH with the 2^−ΔΔCT method (Schmittgen and Livak, 2008). The primers for quantitative real-time PCR (qRT-PCR) were listed as follows: GLP-1R forward primer: 5′-tctccaaactgaaggctaat-3′, reverse primer: 5′-gtccatcacaaaggcaaa-3′; RUNX2 forward primer: 5′-ggactgggtatggtttgtat-3′, reverse primer: 5′-gctgaagaggctgtttga-3′; OCN forward primer: 5′-accacatcggctttcagg-3′, reverse primer: 5′-catagggctgggaggtca-3′; GAPDH forward primer: 5′-gcaagttcaacggcacag-3′, reverse primer: 5′-cgccagtagactccacgac-3′.

Western blot

MC3T3-E1 cells at logarithmic phase were seeded into six-well plates. After 24 h, the cells were cultured with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide for 3, 7, 14, and 21 days, respectively. The control group was incubated with the α-MEM supplemented with 10% FBS. Proteins were extracted by using RIPA buffer (Sigma). Protein concentration was determined by BCA kit (Beyotime Biotechnology, Shanghai, China). Proteins were subjected to SDS-PAGE and transferred onto polyvinylidene fluoride membranes (Millipore, Bedford, MA). After being blocked with 5% bovine serum albumin for 1 h at room temperature, the membranes were incubated with rabbit anti-GLP-1R (1:500), anti-Runx2 (1:500), anti-OCN (1:500), anti-GAPDH (1:1000) (ABclonal Biotechnology Co., Ltd.), anti-Smad2 (1:500), anti-Smad3 (1:500), anti-p-Smad2 (1:300), or anti-p-Smad3 (1:300) (Abcam) antibodies at 4°C overnight, respectively. The next day, membranes were washed with TBST and probed with the corresponding horseradish peroxidase conjugated goat anti-rabbit immunoglobulin G secondary antibody (1:5000; Abcam) for 2 h at room temperature. After TBST washes, membranes were developed with an enhanced chemiluminescence kit (PerkinElmer). The relative expression of target proteins was calculated with Quantity One Software (BioRad) via the normalization to GAPDH.

ALP activity assay and alizarin red stain analysis

MC3T3-E1 cells were seeded into 96-well plates for ALP activity assay. Cells were incubated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide for 3, 7, 14, and 21 days, respectively. The control group was incubated with the α-MEM supplemented with 10% FBS; it was incubated with the same amount of physiological saline. Cells were harvested and lysed with protein lysis buffer (Beyotime Biotechnology), and 30 μL samples were quantified with an ALP Detection Kit (Jiancheng Biotech Institute, Nanjing, China) at a wavelength of 520 nm according to the manufacturer's protocol. ALP activity was normalized to total protein and expressed as IU/mg.

MC3T3-E1 cells were seeded into a six-well plate for mineralization assay. Cells were incubated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. The control group was incubated with the α-MEM supplemented with 10% FBS. After 21 days, cells were washed with PBS and fixed with 70% pre-cooling ethanol for 1 h at room temperature. After PBS washes, cells were stained with 1% alizarin red solution (Sigma) at 37°C for 1 h. Then, the cells were washed with distilled water and captured. Alizarin red was precipitated with 10% cetylpyridinium chloride (Sigma) for 30 min at room temperature. The calcium deposition was determined with a microplate spectrophotometer at the wavelength of 562 nm. All values were normalized to the total protein content.

Cell transfection and inhibition study

MC3T3-E1 cells were seeded into six-well plates for cell transfection. After 24 h, cells were divided into eight groups: SiCon, SiCon+Li, SiSmad2, SiSmad2+Li, SiSmad3, SiSmad3+Li, SiSmad2+SiSmad3, and SiSmad2+SiSmad3+Li. The cells were transfected with siRNA targeted negative control (SiCon), Smad2 (SiSmad2), Smad3 (SiSmad3), or Smad2+Smad3 (SiSmad2+SiSmad3), respectively. After 48 h, SiCon+Li, SiSmad2+Li, SiSmad3+Li, and SiSmad2+SiSmad3+Li groups were treated with osteogenic medium containing 10⁻⁷ M Liraglutide. Cells were isolated for qRT-PCR and western blot, ALP activity, and alizarin red stain analysis. The siRNA silencing efficiency was determined by qRT-PCR and western blot. The primers for qRT-PCR were listed as follows: Smad3 forward primer: 5′-accacccagaatatcaaca-3′, reverse primer: 5′-actcgcacaagtctacgg-3′; Smad2 forward primer: 5′-aacgggacaattagatgag-3′, reverse primer: 5′-ccacagtcggcagtagat-3′. The antibodies for western blot were rabbit anti-Smad2, anti-Smad3, anti-GAPDH, and anti-RUNX2 antibodies.

MC3T3-E1 cells were seeded into six-well plates and divided into six groups: control, Li, DKK-1, DKK-1+Li, LY294002, and LY294002+Li. The Li group was administered with osteogenic medium containing 10⁻⁷ M Liraglutide. The DKK-1 group was treated with DKK-1 at the final concentration of 0.2 μg/mL. The DKK-1+Li group was pretreated with DKK-1 (0.2 μg/mL) for 3 h and then cultured with 10⁻⁷ M Liraglutide. LY294002 cells were incubated with LY294002 at the final concentration of 10 μM. The DKK-1+Li group was pretreated with LY294002 (10 μM) for 1.5 h and administered with osteogenic medium containing 10⁻⁷ M Liraglutide. The control group was given the same amount of vehicle. After 3 days, cells were collected for qRT-PCR and western blot. The antibodies for western blot were rabbit anti-AKT (1:1000), anti-p-AKT (1:500), anti-β-catenin (1:500), anti-Smad2, anti-Smad3, anti-p-Smad2, and anti-p-Smad3 antibodies (Abcam).

Statistical analysis

All experiments were independently repeated three times. Data were presented as the mean ± standard deviation and analyzed by using SPSS software version 18.0 (SPSS, Inc.). Statistical analysis was performed by using Student's t-test and one-way ANOVA followed by post hoc methods. p < 0.05 was considered statistically significant.

Results

Liraglutide has no effect on MC3T3-E1 cell proliferation

To explore the effect of Liraglutide on cell proliferation, we performed CCK8 assay and cell cycle testing. In Figure 1A, no differences about cell viability were observed among groups treated with 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide for 24 h (p > 0.05). Further, different incubation time (24, 48, and 72 h) had no effect on cell viability (p > 0.05). Then, we further investigated the effect of different concentrations of Liraglutide (0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M) on cell cycle in Figure 2B. After 72 h of incubation, the percentage of cell numbers in G1, S, and G2/M from four groups (0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide) displayed no difference (p > 0.05). These results showed that Liraglutide (0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M) had no effect on MC3T3-E1 cell proliferation after 24, 48, and 72 h of incubation.

FIG. 1.

Liraglutide has no effect on cell proliferation of MC3T3-E1 cells. (A) Liraglutide had no effect on cell viability detected by CCK8 assay. (B) Liraglutide did not affect cell cycle analyzed by flow cytometry. (C) Liraglutide did not affect the cell cycle distribution. MC3T3-E1 cells were incubated with 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. After 24, 48, or 72 h, cells were collected for CCK8 assay and flow cytometry analysis. CCK8, cell counting kit 8.

FIG. 2.

Liraglutide stimulates the expression of GLP-1R in MC3T3-E1 cells. (A) GLP-1R was present in MC3T3-E1 cells. MC3T3-E1 cells were seeded into a six-well plate and divided into four groups. MC3T3-E1 cells in 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide groups were incubated with osteogenic medium containing 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. The control group was administered with α-MEM supplemented with 10% FBS. After 48 h, cells were collected for immunocytochemistry. Liraglutide enhanced the mRNA (B) and protein (C) level of GLP-1R in a concentration-dependent manner. (D) The band density of GLP-1R was calculated with Quantity One Software. MC3T3-E1 cells were seeded into a six-well plate and divided into five groups. MC3T3-E1 cells in 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide groups were incubated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. The control group was incubated with the normal medium. After 3, 7, 14, or 21 days of incubation, cells were collected for RNA and protein extraction. Compared with the control group, *p < 0.05; compared with 0 M group, ^# p < 0.05. Liraglutide promoted the mRNA (E) and protein (F) levels of GLP-1R in a time-dependent way. MC3T3-E1 cells were incubated with osteogenic medium containing 10⁻⁷ M Liraglutide. The control group was incubated with normal medium. After 3, 7, 14, or 21 days of incubation, cells were collected for RNA and protein extraction. Compared with the control group, *p < 0.05; compared with day 3 of incubation, ^# p < 0.05. α-MEM, alpha minimal essential medium; FBS, fetal bovine serum; GLP-1R, glucagon-like peptide-1 receptor.

Liraglutide promotes GLP-1R expression in a dosage- and time-dependent manner

Immunocytochemistry was employed to explore the presence of GLP-1R in MC3T3-E1 cells. Figure 2A showed that GLP-1R was localized in the whole cells, and the signals of GLP-1R were enhanced after Liraglutide incubation. To examine the effect of Liraglutide on GLP-1R expression in MC3T3-E1 cells, we incubated MC3T3-E1 cells with different concentrations of Liraglutide (0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M) for 3, 7, 14, or 21 days, respectively. In Figure 2B, the mRNA levels of GLP-1R were improved with an increase of Liraglutide concentration (0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M) under the same incubation time. What is more, similar trends were observed on the protein levels of GLP-1R in Figure 2C and D. Based on the earlier results, 10⁻⁷ M Liraglutide was selected as the optimal concentration for further study. As shown in Figure 2E and F, the mRNA and protein levels of GLP-1R were the highest on day 14 at the concentration of 10⁻⁷ M Liraglutide. Thus, we concluded that GLP-1R was present in MC3T3-E1 and Liraglutide promoted GLP-1R expression in a dosage- and time-dependent manner in MC3T3-E1 cells.

Liraglutide induces the osteogenic differentiation of MC3T3-E1

To explore the effect of Liraglutide on the osteogenic differentiation of MC3T3-E1, we performed the ALP activity and alizarin red stain analysis, and we detected the expression of osteogenic markers RUNX2 and OCN. In Figure 3A, compared with the control group, the addition of osteogenic medium (0 M Liraglutide) enhanced the ALP activity. Compared with untreated time-matched controls, Liraglutide significantly stimulated the ALP activity on 3, 7, 14, and 21 days of Liraglutide incubation. The effect of Liraglutide on the mineralization of MC3T3-E1 cells was examined by alizarin red staining in Figure 3B and C. After continuous treatment for 21 days, the number of mineralized nodules and the content of calcium deposition were significantly increased in the presence of 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, with a maximal effect at 10⁻⁷ M. In Figure 3D and E, the mRNA levels of RUNX2 and OCN were upregulated with an increase of the Liraglutide concentration. Similar trends about the protein levels of RUNX2 and OCN were observed in Figure 3F–H. In Figure 3I, under 10⁻⁷ M Liraglutide treatment, the mRNA and protein levels of RUNX2 at 7 days was the highest; however, OCN at 21 days was the highest. Similar trends on the protein levels of RUNX2 and OCN were observed in Figure 3J. So Lireglutide stimulated the osteogenic differentiation of MC3T3-E1.

FIG. 3.

Liraglutide induces the osteogenic differentiation of MC3T3-E1. (A) Liraglutide enhanced the ALP activity. (B, C) Liraglutide promoted the mineralization of MC3T3-E1 determined by alizarin red staining. (D, E). Liraglutide induced the mRNA expression of RUNX2 and OCN. (F–H) Liraglutide regulated the protein levels of RUNX2 and OCN. MC3T3-E1 cells were divided into five groups and incubated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. The control group was incubated with normal medium. After 3, 7, 14, or 21 days of incubation, cells were collected for RNA and protein extraction. Compared with the control group, *p < 0.05; compared with 0 M group, ^# p < 0.05. (I, J). Liraglutide (10⁻⁷ M) regulated the mRNA level of RUNX2 and OCN. MC3T3-E1 cells were incubated with osteogenic medium containing 10⁻⁷ M Liraglutide. The control group was incubated with normal medium. After 3, 7, 14, or 21 days of incubation, cells were collected for RNA and protein extraction. Compared with the control group, *p < 0.05; compared with day 3 of incubation, ^# p < 0.05. ALP, alkaline phosphatase; OCN, osteocalcin; RUNX2, runt-related transcription factor-2.

Liregluatide promotes the osteogenic differentiation of MC3T3-E1 via regulating Smad2/3

To examine the relationship between Liraglutide and the TGF-β/Smad pathway, we incubated cells with a range concentration of Liraglutide to examine the expression of Smad2/3 in Figure 4. In Figure 4A, compared with control and 0 M groups, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide significantly upregulated the mRNA of Smad2 and Smad3 (p < 0.05). Similarly, Liraglutide promoted the protein levels of Smad2/3 in Figure 4B–D. What is more, the p-Smad2 and p-Smad3 expression was induced, and the p-Smad2/Smad2 and p-Smad3/Smad3 were also upregulated in a dosage-dependent manner. That is to say, Liraglute upregulated the expression of Smad2/3 and the ratio of p-Smad/Smad in MC3T3-E1 cells.

FIG. 4.

Liraglutide modulates the expression of Smad2/3 in MC3T3-E cells. (A) Liraglutide promoted the mRNA levels of Smad2/3 in MC3T3-E1 cells. (B–D) Liraglutide regulated the expression of Smad2/3 and the phosphorylation of p-Smad2/3. MC3T3-E1 cells were divided into five groups and incubated with osteogenic medium containing 0, 10⁻⁹, 10⁻⁸, or 10⁻⁷ M Liraglutide, respectively. The control group was incubated with the normal medium. After 3 days of incubation, cells were collected for qRT-PCR analysis and western blot. Compared with the control group, *p < 0.05; compared with 0 M group, ^# p < 0.05. qRT-PCR, quantitative real-time polymerase chain reaction.

To further determine the relationship between Smad2/3 and Liraglutide-induced osteogenic differentiation, we knocked down the levels of Smad2/3. In Figure 5, we found that the transfection of siRNA targeted to Smad2 and Smad3 significantly repressed the mRNA and protein levels of Smad2/3, which provided good evidence for follow-up experiments. In Figure 6A, the addition of Liraglutide significantly enhanced the ALP activity (SiCon+Li > SiCon, SiSmad2+Li > SiSmad2, SiSmad3+Li > SiSmad3, SiSmad2+SiSmad3+Li > SiSmad2+SiSmad3). However, compared with SiCon+Li, the ALP activity in SiSmad2+Li, SiSmad3+Li, and SiSmad2+SiSmad3+Li was obviously blocked, with the maximum effect in the SiSmad2+SiSmad3+Li group. Similarly, SiSmad2, SiSmad3, or SiSmad2+SiSmad3 transfection suppressed Liraglutide-induced upregulation on RUNX2 mRNA and protein levels (Fig. 6B–D). After continuous treatment for 21 days, compared with the SiCon+Li group, the number of mineralized nodules and the content of calcium deposition after siRNA transfection under the condition of Liraglutide were significantly decreased (p < 0.05, Fig. 6E, F). Taken together, the silencing of Smad2/3 blocked the osteogenic differentiation induced by Liraglutide in MC3T3-E1.

FIG. 5.

SiRNA transfection successfully blocked the expression of Smad2/3. (A) SiSmad2/3 transfection blocked the mRNA levels of Smad2/3. (B) SiSmad2/3 transfection blocked the protein levels of Smad2/3. (C) The relative expression of Smad2/3 was calculated by the normalization to GAPDH. MC3T3-E1 cells were transfected with siRNA-Smad2, siRNA-Smad3, siRNA-Smad2+siRNA-Smad3, or SiCon, respectively. The control group was administered with the same amount of vehicle. After 48 h, RNA and protein was extracted for qRT-PCR and western blot. Compared with the SiCon group, *p < 0.05.

FIG. 6.

Liraglutide induces the osteogenic differentiation of MC3T3-E1 via regulating Smad2/3. (A) The silencing of Smad2 or Smad3 suppressed the ALP activity increase induced by Liraglutide. (B–D) The knockdown of Smad2 and Smad3 affected the mRNA and protein levels of RUNX2 induced by Liraglutide. (E, F) The silencing of Smad2 and Smad3 repressed the mineralized nodules and calcium deposition. MC3T3-E1 cells were divided into eight groups. Cells were transfected with siRNA targeted negative control (SiCon), Smad2 (SiSmad2), Smad3 (SiSmad3), or Smad2+Smad3 (SiSmad2+SiSmad3), respectively. After 48 h, SiCon+Li, SiSmad2+Li, SiSmad3+Li, and SiSmad2+SiSmad3+Li groups were treated with osteogenic medium containing 10⁻⁷ M Liraglutide. After another 3 days, RNA and proteins were extracted for Smad2/3 expression detection. After 7 days, cells were collected for ALP activity and RUNX2 expression analysis. After 21 days, cells were collected for the mineralized nodules and calcium deposition determination. *p < 0.05. Compared with Li group.

Liraglutide modulates the expression of Smad2/3 via Wnt/β-catenin and PI3K/AKT pathways

To explore the mechanism by which Liraglutide modulates the expression of Smad2/3 in MC3T3-E1 cells, we employed the PI3K and Wnt inhibitors. In Figure 7A and 7B, the application of Liraglutide improved the expression of AKT and the phosphorylation of AKT (p-AKT) (Li > control, LY+Li > LY). However, the pretreatment of LY significantly reduced the enhancement of AKT and p-AKT induced by Liraglutide (LY+Li < Li). In Figure 7C and D, the relative protein expression of β-catenin in the Li group was higher than that in the control group. However, compared with the Li group, the addition of DKK-1 or LY obviously repressed the upregulation of β-catenin caused by Liraglutide (p < 0.05). Similar trends were observed on the levels of Smad2, p-Smad2, p-Smad2/Smad2, smad3, p-Smad3, and p-Smad3/Smad3 (Fig. 7C, F). In Figure 7C, we further examined the effect of inhibitors on the mRNA of Smad2/3 in Figure 7G. The application of DKK-1 or LY under the incubation of Liraglutide significantly abolished the upregulation of Smad2/3 mRNA induced by Liraglutide. These data indicated that Liraglutide regulated the expression of Smad2/3 through the activation of Wnt/β-catenin and PI3K/AKT pathways.

FIG. 7.

Liraglutide modulates the expression of Smad2/3 via Wnt/β-catenin and PI3K/AKT pathways. (A, B) LY blocked the expression of AKT and p-AKT induced by Liraglutide. (C–F) DKK-1 and LY repressed the protein levels of β-catenin, Smad2/3, p-Smad2/3, p-Smad2/Smad2, and p-Smad3/Smad3 induced by Liraglutide. (G) DKK-1 and LY repressed the mRNA level of Smad2/3. MC3T3-E1 cells were divided into six groups: control, Li, DKK-1, DKK-1+Li, LY294002, and DKK-1+Li. The Li group was administered with osteogenic medium containing 10⁻⁷ M Liraglutide. The DKK-1 group was treated with DKK-1 at the final concentration of 0.2 μg/mL. The DKK-1+Li group was pretreated with DKK-1 (0.2 μg/mL) for 3 h and then cultured with 10⁻⁷ M Liraglutide. LY294002 cells were incubated with LY294002 at the final concentration of 10 μM. The DKK-1+Li group was pretreated with LY294002 (10 μM) for 1.5 h and administered with osteogenic medium containing 10⁻⁷ M Liraglutide. The control group was given the same amount of vehicle. After 3 days, cells were collected for qRT-PCR and western blot. *p < 0.05. Compared with Li group, ^# p < 0.05.

Discussion

Overactive bone resorption or dysfunction of bone formation causes the imbalance between bone resorption and formation and results in the structural integrity and strength of bone (Sanghani-Kerai et al., 2018). Bone metabolism disorder affects a large proportion of the elderly population and is associated with high morbidity and economic burden. GLP-1 and GLP-1RA play a positive important role in facilitating bone formation and suppressing bone resorption (Nauck and Meier, 2018). Recently, several studies indicated that the drug Liraglutide exerts osteogenic and anti-resorptive effects (Ramsey and Isales, 2017). Our current study investigated the effects of Liraglutide on the osteogenic differentiation and explored the underlying mechanism by which Liraglutide promoted the osteogenic differentiation in MC3T3-E1 cells. These findings revealed a new mechanism by which Liraglutide promoted osteogenic differentiation and provided theory evidence to maintain normal bone metabolism during diabetes therapy.

Previous studies showed that GLP-1 exerts its biological activities via binding to the GLP-1R (Kazafeos, 2011). In our research, GLP-1R was confirmed to be present in MC3T3-E1 cells by immunocytochemistry, qRT-PCR, and western blot. These results were in agreement with most previous studies (Aoyama et al., 2014; Wu et al., 2018), but they were contrary to the data reported by Nuche-Berenguer and colleagues (Nuche-Berenguer et al., 2010). The different results might be caused by differences in primers (Wu et al., 2017); however, the exact reasons need more evidence. Wu and colleagues showed that GLP-1R was localized in the surface of MC3T3-E1 cells, but we detected that the signals of GLP-1R were in the whole cells. Further investigations need to be performed in the follow-up study. In addition, the application of Liraglutide significantly improved the mRNA and protein levels of GLP-1R in MC3T3-E1 cells in a dosage- and time-dependent manner. These data provided new evidence about the close relationship between GLP-1R and Liraglutide.

Osteogenic markers such as ALP activity, mineralized nodule, and the expression of RUNX2 and OCN were examined to evaluate the ability of MC3T3-E1 differentiation. The function of Liraglutide on cell differentiation remains elusive. Some reports indicated that Liraglutide facilitates the osteogenic differentiation in MC3T3-E1 cells (Sun et al., 2015; Wu et al., 2017); however, some scholars found that Liraglutide blocks the osteogenic differentiation (Hu et al., 2016). In our study, intracellular ALP activity and the number of mineralized nodules were significantly increased after Liraglutide incubation in a dosage-dependent manner. The mRNA and protein levels of RUNX2 and OCN were also upregulated by Liraglutide. The highest level of RUNX2 was at 7 days of Liraglutide incubation; however, the highest level of OCN appeared at 21 days. The process of osteoblast differentiation consists of three steps: cell proliferation, extracellular matrix formation and maturation, and bone matrix mineralization. During bone matrix maturation, ALP, an early-stage phenotypic marker in the osteoblastic differentiation process (Yu et al., 2017a), increases. The role of RUNX2 is to promote early immature differentiation of osteoblasts, and RUNX2 is the earliest and most specific marker of bone formation (Gu et al., 2018). During bone matrix mineralization, OCN expression increases and the formation of bone nodules indicates bone matrix mineralization (Blair et al., 2017). These data revealed that Liraglutide significantly promoted MC3T3-E1 cells differentiation in different stages, which was in agreement with the first opinion.

The TGF-β/Smad signaling plays a direct and significant role in regulating osteoblast differentiation (Yu et al., 2017b). Smad proteins (Smad2/3/4) are the major transcription factors of TGF-β1 signaling and are involved in TGF-β-related cell responses and pathophysiological functions (Budi et al., 2017). A previous study revealed that Smad2/3 played an important role in the regulation of bone formation (Heo et al., 2018), and the expression of Smad2/3 and p-Smad2/3 was elevated during cementoblast differentiation and mineralization (Li et al., 2015). In our study, we found that Liraglutide significantly upregulated the expression of Smad2/3, p-Smad2/3 and the ratio of p-Smad2/Smad2 and p-Smad3/Smad3, and the silencing of Smad2/3 obviously blocked the osteogenic differentiation induced by Liraglutide. That is to say, Smad2/3 is essential for Liraglutide-induced osteoblastic differentiation. In combination with a previous study, we concluded that Liraglutide facilitated the osteoblastic differentiation via regulating the expression of Smad2/3, which provided new insights for exploring the mechanism of Liraglutide functioning on osteoblastic differentiation.

It is reported that Liraglutide facilitates osteogenic differentiation via the PI3K/AKT pathway (Wu et al., 2017). In our study, we found that the application of Liraglutide promoted the levels of AKT and p-AKT, and the pretreatment of PI3K inhibitor LY294002 suppressed the upregulation of AKT and p-AKT. These results suggested that LY294002 successfully blocked the PI3K/AKT pathway. In PI3K class II-α knock-out mice, the TGF-β-induced Smad2/3 signaling was blocked (Aki et al., 2015). The application of LY294002 extended the duration of Smad2/3 activation and repressed the phosphorylation of Smad2/3 (Yu et al., 2015). In our study, LY294002 also repressed the levels of Smad2/3 and p-Smad2/3 under Liraglutide incubation. Thus, Liraglutide modulated the activation of Smad2/3 via the PI3K/AKT pathway to be involved in osteoblast-mediated bone formation.

Many studies revealed that the Wnt/β-catenin signaling pathways are involved in every bone metabolism process (Yang et al., 2011). In our study, the Wnt inhibitor DKK-1 suppressed the mRNA and protein levels of β-catenin, suggesting that DKK-1 successfully blocked the Wnt/β-catenin pathway. During myofibroblast differentiation, Wnt3α modulated the promoter element binding to Smad2/3 (Yang et al., 2011). Axin and GSK3β in the Wnt/β-catenin pathway stabilized Smad3 protein to modulate the TGF-β pathway (Guo et al., 2008). Smad3 induces the activation of Wnt/β-catenin signaling in mouse osteoblastic cells and promotes the osteoblast differentiation (Liu et al., 2016). These findings revealed the close relationship between Wnt/β-catenin and TGF-β/Smad pathways. In our study, the addition of DKK-1 repressed the activation of Smad2/3 induced by Liraglutide. That is to say, Liraglutide modulated the expression of Smad2/3 via the Wnt/β-catenin pathway for osteoblastic differentiation. In our study, we also found that LY294002 blocked the upregulation of β-catenin induced by Liraglutide, suggesting the crosstalk between Wnt/β-catenin and PI3K/AKT pathways.

Conclusions

We concluded that Liraglutide upregulated the levels of Smad2/3 and p-Smad2/3 via PI3K/AKT and Wnt/β-catenin pathways, in turn promoting osteoblastic differentiation in MC3T3-E1 cells. These findings revealed new mechanisms by which Liraglutide induced osteoblastic differentiation, and they provided theory evidence to maintain normal bone metabolism during diabetes therapy.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Aki

, Yoshioka

, Okamoto

, Takuwa

, and Takuwa

(2015). Phosphatidylinositol 3-kinase class II alpha-isoform PI3K-C2alpha is required for transforming growth factor beta-induced Smad signaling in endothelial cells. J Biol Chem, 290, 6086–6105.

Aoyama

, Watari

, Podyma-Inoue

K.A.

, Yanagishita

, and Ono

(2014). Expression of glucagon-like peptide-1 receptor and glucosedependent insulinotropic polypeptide receptor is regulated by the glucose concentration in mouse osteoblastic MC3T3-E1 cells. Int J Mol Med, 34, 475–482.

Blair

H.C.

, Larrouture

Q.C.

, Li

, Lin

, Beer-Stoltz

, Liu

, et al. (2017). Osteoblast differentiation and bone matrix formation in vivo and in vitro. Tissue Eng Part B Rev, 23, 268–280.

Boyle

J.G.

, Livingstone

, and Petrie

J.R.

(2018). Cardiovascular benefits of GLP-1 agonists in type 2 diabetes: a comparative review. Clin Sci (Lond), 132, 1699–1709.

Budi

E.H.

, Duan

, and Derynck

(2017). Transforming growth factor-beta receptors and Smads: regulatory complexity and functional versatility. Trends Cell Biol, 27, 658–672.

Camkurt

M.A.

, Lavagnino

, Zhang

X.Y.

, and Teixeira

A.L.

(2018). Liraglutide for psychiatric disorders: clinical evidence and challenges. Horm Mol Biol Clin Investig. [Epub ahead of print]; DOI: 10.1515/hmbci-2018-0031.

Campbell

J.E.

, and Drucker

D.J.

(2013). Pharmacology, physiology, and mechanisms of incretin hormone action. Cell Metab, 17, 819–837.

Chen

, Yang

, Lu

D.W.

, Zhao

, Feng

Y.L.

, Chen

, et al. (2018). Central role of dysregulation of TGF-beta/Smad in CKD progression and potential targets of its treatment. Biomed Pharmacother, 101, 670–681.

Coulter

A.A.

, Rebello

C.J.

, and Greenway

F.L.

(2018). Centrally acting agents for obesity: past, present, and future. Drugs, 78, 1113–1132.

10.

de Kroon

L.M.

, Narcisi

, van den Akker

G.G.

, Vitters

E.L.

, Blaney Davidson

E.N.

, van Osch

G.J.

, et al. (2017). SMAD3 and SMAD4 have a more dominant role than SMAD2 in TGFbeta-induced chondrogenic differentiation of bone marrow-derived mesenchymal stem cells. Sci Rep, 7, 43164.

11.

Gallwitz

(2015). GLP-1 receptor agonists in type 2 diabetes and beyond—new insights 2015. Eur Endocrinol, 11, 21–25.

12.

, Bai

, and Zhang

(2018). Osteogenic stimulation of human dental pulp stem cells with a novel gelatin-hydroxyapatite-tricalcium phosphate scaffold. J Biomed Mater Res A, 106, 1851–1861.

13.

Guo

, Ramirez

, Waddell

D.S.

, Li

, Liu

, and Wang

X.F.

(2008). Axin and GSK3- control Smad3 protein stability and modulate TGF-signaling. Genes Dev, 22, 106–120.

14.

Heo

S.Y.

, Ko

S.C.

, Nam

S.Y.

, Oh

, Kim

Y.M.

, Kim

J.I.

, et al. (2018). Fish bone peptide promotes osteogenic differentiation of MC3T3-E1 pre-osteoblasts through upregulation of MAPKs and Smad pathways activated BMP-2 receptor. Cell Biochem Funct, 36, 137–146.

15.

X.K.

, Yin

X.H.

, Zhang

H.Q.

, Guo

C.F.

, and Tang

M.X.

(2016). Liraglutide attenuates the osteoblastic differentiation of MC3T3E1 cells by modulating AMPK/mTOR signaling. Mol Med Rep, 14, 3662–3668.

16.

Iepsen

E.W.

, Torekov

S.S.

, and Holst

J.J.

(2015). Liraglutide for Type 2 diabetes and obesity: a 2015 update. Expert Rev Cardiovasc Ther, 13, 753–767.

17.

Kazafeos

(2011). Incretin effect: GLP-1, GIP, DPP4. Diabetes Res Clin Pract, 93 Suppl 1, S32–S36.

18.

Lambadiari

, Pavlidis

, Kousathana

, Varoudi

, Vlastos

, Maratou

, et al. (2018). Effects of 6-month treatment with the glucagon like peptide-1 analogue liraglutide on arterial stiffness, left ventricular myocardial deformation and oxidative stress in subjects with newly diagnosed type 2 diabetes. Cardiovasc Diabetol, 17, 8.

19.

, Zhu

, and Xiao

(2015). Multi-walled carbon nanotubes promote cementoblast differentiation and mineralization through the TGF-beta/Smad signaling pathway. Int J Mol Sci, 16, 3188–3201.

20.

Liu

, Hao

, Wang

, and Su

(2016). miR-23b targets Smad 3 and ameliorates the LPS-inhibited osteogenic differentiation in preosteoblast MC3T3-E1 cells. J Toxicol Sci, 41, 185–193.

21.

, Sun

, Yu

, Wang

, Liu

, Zhao

, et al. (2015). Glucagon-like peptide-1 receptor agonist Liraglutide has anabolic bone effects in ovariectomized rats without diabetes. PLoS One, 10, e0132744.

22.

Luo

, Liu

, and Lu

(2016). Glucagon-like peptide-1(GLP-1) receptor agonists: potential to reduce fracture risk in diabetic patients?. Br J Clin Pharmacol, 81, 78–88.

23.

Mannucci

, and Dicembrini

(2015). Drugs for type 2 diabetes: role in the regulation of bone metabolism. Clin Cases Miner Bone Metab, 12, 130–134.

24.

Mansur

S.A.

, Mieczkowska

, Bouvard

, Flatt

P.R.

, Chappard

, Irwin

, et al. (2015). Stable incretin mimetics counter rapid deterioration of bone quality in type 1 diabetes mellitus. J Cell Physiol, 230, 3009–3018.

25.

Nauck

M.A.

, and Meier

J.J.

(2018). Incretin hormones: their role in health and disease. Diabetes Obes Metab, 20 Suppl 1, 5–21.

26.

Nuche-Berenguer

, Portal-Nunez

, Moreno

, Gonzalez

, Acitores

, Lopez-Herradon

, et al. (2010). Presence of a functional receptor for GLP-1 in osteoblastic cells, independent of the cAMP-linked GLP-1 receptor. J Cell Physiol, 225, 585–592.

27.

Ramsey

, and Isales

C.M.

(2017). Intestinal incretins and the regulation of bone physiology. Adv Exp Med Biol, 1033, 13–33.

28.

Rizzo

, Abate

, Chandalia

, Rizvi

A.A.

, Giglio

R.V.

, Nikolic

, et al. (2015). Liraglutide reduces oxidative stress and restores heme oxygenase-1 and ghrelin levels in patients with type 2 diabetes: a prospective pilot study. J Clin Endocrinol Metab, 100, 603–606.

29.

Russell-Jones

(2009). Molecular, pharmacological and clinical aspects of liraglutide, a once-daily human GLP-1 analogue. Mol Cell Endocrinol, 297, 137–140.

30.

Ryan

G.J.

, Foster

K.T.

, and Jobe

L.J.

(2011). Review of the therapeutic uses of liraglutide. Clin Ther, 33, 793–811.

31.

Sanghani-Kerai

, McCreary

, Lancashire

, Osagie

, Coathup

, and Blunn

(2018). Stem cell interventions for bone healing: fractures and osteoporosis. Curr Stem Cell Res Ther, 13, 369–377.

32.

Schmittgen

T.D.

, and Livak

K.J.

(2008). Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc, 3, 1101–1108.

33.

Sun

H.X.

, Lu

, Luo

, Zhao

, and Liu

J.M.

(2015). Liraglutide, the glucagon-like peptide-1 receptor agonist, has anabolic bone effects in diabetic Goto-Kakizaki rats. J Diabetes, 7, 584–588.

34.

Vellucci

, Terenzi

, Kanis

J.A.

, Kress

H.G.

, Mediati

R.D.

, Reginster

J.Y.

, et al. (2018). Understanding osteoporotic pain and its pharmacological treatment. Osteoporos Int, 29, 1477–1491.

35.

Vestergaard

(2014). Diabetes and bone fracture: risk factors for old and young. Diabetologia, 57, 2007–2008.

36.

, Li

, Xue

, and Li

(2017). Liraglutide, a glucagon-like peptide-1 receptor agonist, facilitates osteogenic proliferation and differentiation in MC3T3-E1 cells through phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT), extracellular signal-related kinase (ERK)1/2, and cAMP/protein kinase A (PKA) signaling pathways involving beta-catenin. Exp Cell Res, 360, 281–291.

37.

, Li

, Xue

, and Li

(2018). Liraglutide inhibits the apoptosis of MC3T3-E1 cells induced by serum deprivation through cAMP/PKA/beta-catenin and PI3K/AKT/GSK3beta signaling pathways. Mol Cells, 41, 234–243.

38.

Yan

, Xiong

, and Chen

Y.G.

(2018). Feedback regulation of TGF-beta signaling. Acta Biochim Biophys Sin (Shanghai), 50, 37–50.

39.

Yang

, Yang

, Tu

, Zheng

, Cai

, and Wang

(2011). Strontium enhances osteogenic differentiation of mesenchymal stem cells and in vivo bone formation by activating Wnt/catenin signaling. Stem Cells, 29, 981–991.

40.

, Tang

D.Z.

, Li

S.Y.

, Wu

, and Chen

(2017a). Daidzein promotes proliferation and differentiation in osteoblastic OCT1 cells via activation of the BMP-2/Smads pathway. Pharmazie, 72, 35–40.

41.

, Xu

, Li

, Xie

, Xi

, Wang

, et al. (2017b). Zinc-modified calcium silicate coatings promote osteogenic differentiation through TGF-beta/Smad pathway and osseointegration in osteopenic rabbits. Sci Rep, 7, 3440.

42.

J.S.

, Ramasamy

T.S.

, Murphy

, Holt

M.K.

, Czapiewski

, Wei

S.K.

, et al. (2015). PI3K/mTORC2 regulates TGF-beta/activin signalling by modulating Smad2/3 activity via linker phosphorylation. Nat Commun, 6, 7212.