Advanced glycation end products induced IL-6 and VEGF-A production and apoptosis in osteocyte-like MLO-Y4 cells by activating RAGE and ERK1/2, P38 and STAT3 signalling pathways

Abstract

Advanced glycation end products (AGEs) are involved in osteopenia in people with diabetes and the elderly. Interleukin-6 (IL-6) and vascular endothelial growth factor-A (VEGF-A) are potent regulators of bone metabo- lism, and in bone tissue, osteocytes are an important source of these regulators. However, whether AGEs can directly regulate IL-6 and VEGF-A secretion by osteocytes is unknown. In this study, we evaluated the effect of AGEs on IL-6 and VEGF- A production as well as apoptosis in osteocyte-like MLO-Y4 cells. We also studied the involvement of receptor for advanced glycation end products (RAGE) and the role of extracellular signal- regulated kinases 1 and 2 (ERK1/2), P38 and signal transducer and activator of transcription 3 (STAT3) signalling pathways. We found that 100 μg/ml AGEs significantly induced apoptosis and up-regulated the expression of IL-6 and VEGF-A in MLO-Y4 cells. Additionally, AGEs significantly activated the ERK1/2, P38 and STAT3 signalling pathways. The ERK1/2 inhibitor U0126, the P38 inhibitor SB239063 and the STAT3 inhibitor S3I-201 all attenuated the effects of AGEs on MLO-Y4 cell apoptosis and IL-6 and VEGF-A secretion. Moreover, activation of the three signalling pathways was abolished by their respective inhibitors. Additionally, the AGEs- induced effects, including increased apoptosis, up-regulated expression of IL-6 and VEGF-A and activation of the three signalling pathways, were all abolished by pre-treating the osteocytes with the RAGE antagonist FPS-ZM1. Together, these data convince us that AGEs can activate the ERK1/2, P38 and STAT3 signalling pathways via RAGE and that their activation involves the AGEs-induced up-regulation of IL-6 and VEGF-A production as well as apoptosis in osteocytes. These results highlight the role of osteocytes in the regulation of bone metabolism by AGEs.

1. Introduction

Osteoporosis, a systemic skeletal disorder characterized by de- creased bone mineral density (BMD), can compromise bone strength and increase the risk of bone fractures [1]. In turn, bone fractures, particularly those occurring in the vertebra and hip, often reduce quality of life and can even lead to morbidity and mortality [2]. In the past few decades, advanced glycation end products (AGEs) have been demonstrated to be critical mediators of the pathogenesis and devel- opment of osteoporosis [3].

AGEs are a kind of irreversible Amadori products that result from protein glycation, a type of non-enzymatic post-translational mod- ification. Accelerated AGEs generation and accumulation are found in osteoporosis [4]. Recent studies demonstrate that protein glycation could affect bone remodeling [5]. Protein cross-linking is one of the most important processes of AGEs formation. Consequently, increased amounts of AGEs in bone could lead to more matrix protein cross- linking, which partially contributes to their deleterious effects on the biomechanical properties of bone [6]. AGEs can interact with several receptors to increase oxidative stress and inflammation [7], and among these receptors, the receptor for advanced glycation end products (RAGE) is the best characterized [8]. AGEs can bind to RAGE, resulting in functional alterations to osteoblasts and osteoclasts [9,10]; these alterations contribute to the development of bone disease.

Osteocytes are the most abundant cells in mammalian bone tissue [11] and are a major regulator of skeletal activity. Furthermore, they are involved in the development of osteoporosis. They not only regulate the behaviours of osteoblasts and osteoclasts [12] but also participate directly in the pathogenesis of osteoporosis. Specifically, osteocyte apoptosis causes cumulative and irrevocable effects that inhibit the lacuno-canalicular network and prevent the repair of bone microcracks; these actions directly induce bone fragility, which plays a major role in osteoporosis [13]. In addition, many cytokines, including interleukin-6 (IL-6) and vascular endothelial growth factor-A (VEGF-A), are involved in the regulation of bone turnover via osteocytes [14,15].

Based on these data, we suppose that AGEs might be involved in the development of osteoporosis by acting directly upon osteocytes. Unfortunately, little is known about the role of AGEs in osteocytes. Thus, this study aimed to determine whether AGEs can induce apop- tosis and increase IL-6 and VEGF-A production in osteocyte-like MLO- Y4 cells and to further explore the probable mechanisms underlying this role. We eventually found that AGEs could activate the ERK1/2, P38 and STAT3 signalling pathways via RAGE and that their activation is involved in the AGEs-induced up-regulation of IL-6 and VEGF-A pro- duction as well as apoptosis in MLO-Y4 cells. These results highlight the role of osteocytes in the regulation of bone metabolism by AGEs.

2. Materials and methods

2.1. Chemicals and reagents

Foetal bovine serum (FBS), α-modified essential medium and calf serum were bought from HyClone (HyClone Lab). Bovine serum al- bumin (BSA) was purchased from Sigma (St Louis, MO, USA). TRI Reagent® was purchased from Molecular Research Center (Invitrogen, Carlsbad, CA, USA). The PrimeScript™ RT Reagent Kit with gDNA Eraser (Perfect Real-Time) and SYBR® Premix Ex Taq™ II were pur- chased from Takara Biotechnology (DRR820A; Takara, Japan). A total protein extraction kit was purchased from KeyGen Biotech (KeyGen Biotech, China). SDS-PAGE gels were purchased from Bio-Rad (Hercules, CA, USA). An ECL kit was purchased from Pierce (Rockford, IL, USA). The RAGE antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), and the other antibodies were purchased from Abcam (Cambs, England). An annexin V-fluorescein isothiocyanate (FITC)/PI apoptosis detection kit was purchased from Dojindo (Dojindo, Japan). All other chemicals and reagents were ob- tained from commercial suppliers and were of analytical grade.

2.2. Preparation of AGEs

AGEs were prepared as follows [16]. Briefly, 50 mg/ml BSA was incubated under sterile conditions at 37 °C in the dark with 500 mM ribose in 400 mM sodium phosphate buffer (pH 7.4) for 7 days. To re- move the unincorporated ribose and terminate the reaction, all aliquots were dialyzed extensively against PBS (pH 7.4) at 4 °C overnight. Next, the solutions were lyophilized, and the lyophilized powder was weighed and dissolved in PBS (pH 7.4) at a final concentration of 50 μg/μl. Then, the final solution was filtered using a Millipore Express 0.22-um filter for sterilization and stored at 4 °C for future use.

2.3. AGEs fluorescence test

As AGEs are fluorescent, a small portion of the obtained solution was subjected to testing using a luminescence spectrometer antigen [17]. Furthermore, this cell line has the properties of osteocytes and is a useful tool for examining osteocyte function [18]. In this study, MLO-Y4 cells were cultured in flasks with an appropriate amount of α- modified essential medium supplemented with 100 U/ml penicillin (Sigma, St. Louis, MO, USA), 100 g/ml streptomycin (Sigma), 5% FBS and 5% calf serum at 37 °C in a humidified atmosphere with 95% air and 5% CO2. The medium was changed every 2 days.

2.5. Apoptosis detection

Cells were seed into 6 well-culture plates at a concentration of 5 × 105 cells/ml and were treated with AGEs. Apoptosis was detected using an annexin V-fluorescein isothiocyanate (FITC)/PI apoptosis de- tection kit and a FACSCalibur™ flow cytometer (BD Biosciences), and the data were analysed using CellQuest™ software, version 3.3 (BD Biosciences). Data are shown in quadrants positioned on annexin V/PI plots to distinguish living cells (annexin V−/PI−), early apoptotic cells (annexin V+/PI−) and late apoptotic cells (annexin+/PI+).

2.6. Quantitative real-time RT-polymerase chain reaction (qPCR)

Gene expression levels of IL-6, VEGF-A and RAGE were determined by qPCR. First, the total RNA was extracted from cells by using TRIzol reagent; then, a thermal cycler (FTC2000, Fenglin Biotechnology Co., Shanghai, China) was utilized to reverse transcribe 1 μg of RNA into cDNA using the PrimeScript RT Reagent Kit. Finally, 2 μl of cDNA and the SYBR Premix RX Taq™ II Kit were used to perform real-time RT-PCR assays on an ABI PRISM 7300 Fast Real-Time PCR System. The primer information is shown in Table 1. Gene expression analyses were per- formed, and the relative expression levels were calculated and nor- malized to that of GAPDH by using the 2−△△Ct method.

2.7. ELISA

Osteocytes were seeded in 6-well plates with the appropriate medium. After incubation, the supernatants were collected and cen- trifuged at 1000 revolutions per second for 20 min. IL-6 and VEGF-A levels in the supernatants were quantified using specific ELISA kits and an ELISA reader (HTS700 +, USA) according to the manufacturer’s in- structions. Each sample was measured three times.

2.8. Western blots

Cells washed with PBS were lysed, and the total protein was ob- tained by using a total protein extraction kit. The protein samples were diluted with SDS sample buffer at a ratio of 1:5 and then boiled for 8 min. Equal amounts of protein (20–50 μg) for each sample were re- solved by SDS-PAGE and transferred onto PVDF membranes. Next, 5% BSA in TBST (10 mm Tris-HCl, pH 8.0; 150 mm NaCl; 0.05% Tween20) was used to block nonspecific binding for 1 h; then, the membranes were incubated at 4 °C overnight with specific primary antibodies against ERK1/2, phospho-ERK1/2, P38, phospho-P38, STAT3, phospho-STAT3 and RAGE. The next day, the membranes were (HTS7000 +, USA); the excitation wavelength was set to 340 nm, the emission wavelength was set to 420 nm, and the slit widths were set to 2.5 nm.

2.4. Cell culture

MLO-Y4 cells were a generous gift from Prof. Lynda F. Bonewald (Department of Oral Biology, University of Missouri-Kansas City, Kansas City, MO, USA). This cell line was first established from trans- genic mice created using an osteocalcin promoter that drives the large T rewarmed at room temperature for half an hour and washed with TBST. The washed membranes were incubated with specific secondary anti- bodies for 1 h at room temperature and washed again with TBST. Finally, an ECL kit was used to detect the proteins. Each protein was assayed three times using different samples.

2.9. Statistical analysis

All experiments were repeated three times, and the obtained data were subjected to one-way ANOVA and S-N-K multiple comparison tests. p < 0.05 was considered significant. All analyses were per- formed using IBM SPSS Statistics 20.0 software.

3. Results

3.1. Influence of AGEs on levels of IL-6 and VEGF-A and apoptosis

To test the effects of AGEs on the mRNA levels of IL-6 and VEGF-A, qPCR was used. As shown in Fig. 1A, AGEs significantly induced mRNA levels of IL-6 at concentrations of 50, 100 and 200 μg/ml (p < 0.05) and mRNA levels of VEGF-A at 100 and 200 μg/ml (p < 0.05); of the different concentrations, 100 μg/ml was the optimum concentration for inducing these effects of AGEs. Next, AGEs were added to a flask at the 100 μg/ml optimum concentration and incubated for different times (0.5 h, 1 h, 2 h, 4 h, 8 h and 24 h). Fig. 1A shows that the levels of IL-6 and VEGF-A mRNA gradually increased with the AGEs incubation times; the IL-6 mRNA level reached its peak at 4 h, and the VEGF-A mRNA level reached its peak at 4 h and 8 h and then gradually de- creased.

The levels of IL-6 and VEGF-A secretion were measured by ELISA. As shown in Fig. 1B, AGEs promoted the secretion of IL-6 and VEGF-A in a time-dependent manner; both concentrations peaked at 24 h. AGEs significantly increased the secretion of IL-6 at 4 h, 8 h and 24 h (p < 0.05), whereas VEGF-A secretion was significantly increased at 8 h and 24 h (p < 0.05).Apoptosis in MLO-Y4 cells was measured by flow cytometry, as shown in Fig. 1C. After 72 h of incubation, the AGEs group had a sig- nificantly greater number of apoptotic cells than that in the control group (p < 0.05).

3.2. Effects of AGEs on RAGE levels

RAGE mRNA levels were evaluated by qPCR. As shown in Fig. 2B, AGEs significantly increased the mRNA levels of RAGE, and 100 μg/ml was the optimum concentration for inducing this effect. Next, AGEs were added to a flask at the 100 μg/ml optimum concentration and incubated for different times (0.5 h, 1 h, 2 h, 4 h, 8 h and 24 h). As shown in Fig. 2A, the levels of RAGE mRNA gradually increased with the AGEs incubation times; RAGE mRNA levels reached their peak at 4 h and 8 h. When pre-incubated with FPS-ZM1, a RAGE antagonist, the RAGE mRNA levels were significantly decreased (p < 0.05).
RAGE protein levels were measured by western blot. As shown in Fig. 2B, AGEs significantly increased the RAGE protein levels at the 2-h time point (p < 0.05); then, the RAGE protein levels gradually in- creased and reached their peak at the 24-h time point (almost 8-fold the control value). In addition, we tested whether AGEs could increase RAGE levels via positive feedback by using FPS-ZM1 to block the binding of AGEs to RAGE. As shown in Fig. 2B, after FPS-ZM1 in- cubation, RAGE levels were significantly decreased (p < 0.05); these results validate the positive feedback regulation.

3.3. Effects of AGEs on the activation of ERK1/2, P38 and STAT3 signalling pathways

Activation of the ERK1/2, P38 and STAT3 signalling pathways was measured by western blot, and their activation level was assessed by using the ratios of P-ERK/ERK, P-P38/P38 and P-STAT3/STAT3. As shown in Fig. 3, the ratio of P-ERK/ERK was significantly increased at 5 min (p < 0.05), whereas the ratio of P-STAT3/STAT3 was sig- nificantly increased at 10 min (p < 0.05); both of these ratios peaked at 3 h. The ratio of P-P38/P38 was significantly increased at 5 min (p < 0.05) and reached its peak at 10 min. Then, the ratio of P-P38/ P38 gradually decreased, but at 3 h, it was still significantly higher than that of the controls (p < 0.05).

3.4. RAGE and ERK1/2, P38 and STAT3 signalling pathways are involved in the AGEs-induced production of IL-6 and VEGF-A as well as apoptosis in osteocyte-like MLO-Y4 cells

RAGE is expressed by many cell types, and it plays a role in their functions [19,20]. Here, we used the specific RAGE antagonist FPS-ZM1 to determine the role of RAGE in IL-6 and VEGF-A production as well as apoptosis in MLO-Y4 cells. As shown in Fig. 4A, pre-treatment with FPS- ZM1 significantly attenuated IL-6 and VEGF-A mRNA expression (p < 0.05). Furthermore, we validated the role of RAGE in the secre- tion of IL-6 and VEGF-A, as shown in Fig. 4B. Pre-treatment with FPS- ZM1 significantly inhibited the secretion of IL-6 and VEGF-A (p < 0.05). In addition, the role of RAGE in apoptosis was studied. As shown in Fig. 4C, AGEs obviously induced apoptosis (almost 11.2% of cells were apoptotic), whereas FPS-ZM1 significantly attenuated the effects of AGEs (almost 4% of cells were apoptotic, p < 0.05).

To determine the involvement of ERK1/2, P38 and STAT3 activa- tion in IL-6 and VEGF-A production as well as apoptosis in MLO-Y4 cells, their respective inhibitors, U0126, SB239063 and S3I-201, were administered 1 h before AGEs treatment. As shown in Fig. 4, the ad- dition of the specific inhibitors significantly blunted the AGEs-induced increase in IL-6 and VEGF-A production (Fig. 4A, Fig. 4B, p < 0.05) and apoptosis (Fig. 4C, p < 0.05).

3.5. AGEs activate ERK1/2, P38 and STAT3 signalling pathways via RAGE

To evaluate the role of RAGE in ERK1/2, P38 and STAT3 activation, FPS-ZM1 was used to pre-treat osteocyte-like MLO-Y4 cells 1 h before AGEs treatment. As shown in Fig. 5, FPS-ZM1 significantly inhibited the AGEs-induced activation of ERK1/2, P38 and STAT3 (p < 0.05).

4. Discussion

In our study, we demonstrate that AGEs can activate the ERK1/2, P38 and STAT3 signalling pathways and successfully induce IL-6 and VEGF-A secretion and apoptosis via RAGE activation in MLO-Y4 cells. Our results show that AGEs can induce MLO-Y4 cell apoptosis. The ability of AGEs to induce apoptosis has been demonstrated in many cell types, including osteoblasts [20] and adipose stem cells [21]. However, the mechanisms underlying this action have not yet been clearly de- fined. K. Tanaka et al. [22] reported apoptosis-related DNA fragments that were induced by AGEs in osteocytes. However, this study did not investigate the mechanisms underlying these effects. In this study, we found that approximately 11.27% of MLO-Y4 cells were apoptotic when incubated with AGEs for 72 h, whereas only 2.88% of cells were apoptotic in the control group. Osteocyte apoptosis has been suggested to be involved in bone turnover regulation, and the osteoporotic state is associated with reduced osteocyte viability [23]. Osteocyte apoptosis also plays a major role in glucocorticoid-induced osteoporosis [13]. Thus, we hypothesized that AGEs are involved in bone turnover reg- ulation via osteocyte apoptosis. In the elderly and people with diabetes, AGEs are present at high levels in tissues, including bone; these high levels might partly account for senile osteoporosis and diabetic bone disease. In addition, AGEs promote IL-6 secretion in osteocytes. In our study, AGEs induced IL-6 production in a time- and dose-dependent manner, and these results are consistent with those reported in os- teoarthritis chondrocytes [24]. IL-6 is an important regulator of bone remodeling. IL-6 has dual effects on bone turnover and is thought to play a key destructive role in rheumatoid arthritis, osteoporosis and bone cancers [25]. Therefore, we believe that IL-6 is one of the most important factors involved in AGEs-related bone diseases, such as os- teoporosis. In addition, VEGF-A secretion is induced by AGEs in MLO- Y4 cells. Like IL-6, VEGF-A increased in a time- and dose-dependent manner in this study. VEGF-A secretion induced by AGEs is observed in many cells. Fu-yuan Hong et al. [26] reported an increase in VEGF-A production caused by AGEs in peritoneal mesothelial cells. It is known that VEGF-A is an important factor involved in angiogenesis, bone formation and resorption [27,28]. Previous research reports that VEGF- A can stimulate osteoclast activity [29,30]. In this study, an increase in VEGF-A secretion due to AGEs stimulation in MLO-Y4 cells was dis- covered; these results suggest that AGEs may contribute to bone re- sorption via promoting VEGF-A production in osteocytes.

We further observed that RAGE was essential for AGEs-induced apoptosis and up-regulation of IL-6 and VEGF-A secretion in MLO-Y4 cells. RAGE is expressed in many cell types and is involved in the function of AGEs. Furthermore, RAGE is involved in many diseases, and its antagonist may serve as a potential treatment for these diseases, which include Alzheimer's disease and atherosclerosis [27,28]. In bone tissue, RAGE has been identified in osteocytes [22]. As is well known, osteocytes are the most abundant cell type and act as a regulator of bone turnover. However, few studies of the role of RAGE in osteocytes have been conducted. So far, only one study of the role of RAGE in osteocyte function [22] has been published, and it showed that RAGE is
involved in RANKL and sclerostin secretion. In the current study, our data showed that RAGE is involved in AGEs-induced apoptosis and IL-6 and VEGF-A up-regulation in MLO-Y4 cells. In addition to our results, Zhou et al. [19] have demonstrated that RAGE-deficient osteoclasts exhibit impaired maturation and reduced bone resorption activity; moreoever, it has been proven that RAGE contributes to the effects of AGEs on osteoblast proliferation and function [20]. Furthermore, our data indicated that AGEs increased RAGE expression via a positive feedback mechanism, and these results are consistent with a previous study [31]. Although RAGE has been demonstrated to be expressed in many cell types, it is normally present at low levels under healthy conditions. However, in states of chronic inflammation, such as dia- betes, RAGE is up-regulated [32]. AGEs bind to RAGE, potentially
causing the release of the transcription factor NF-κB [31]. RAGE ex- pression is up-regulated by NF-κB; therefore, as RAGE binds to its li- gands, it continues to up-regulate NF-κB, which continues to increase the expression of RAGE in a positive feedback loop [31]. This effect could amplify the biological function of AGEs to some degree. In this study, this positive feedback mechanism could promoted the AGEs-in- duced production of IL-6 and VEGF-A as well as apoptosis. Apoptosis and increased levels of IL-6 and VEGF-A play a role in bone destruction. Therefore, we can infer that RAGE likely serves as a potent regulator of bone metabolism and that its involvement in osteocyte function affects this regulation.

We then further explored the mechanisms underlying AGEs reg- ulation in osteocytes. As noted, the activation of several intracellular signal pathways, such as ERK1/2, P38 and STAT, is involved in the function of RAGE [33–35]. However, no studies of the signalling pathways linked with RAGE in osteocytes could be found. Previous studies show that the ERK1/2, P38 and STAT3 pathways are involve in the secretion of VEGF-A and IL-6 [26,36,37]. Thus, we hypothesize that the activation of the three pathways may be involved in the regulation of apoptosis and the up-regulation of IL-6 and VEGF-A via RAGE in osteocytes. In fact, our results show the rapid phosphorylation of ERK1/ 2, P38 and STAT3 after MLO-Y4 cells are stimulated with AGEs; this phosphorylation indicates the activation of the three pathways. Fur- thermore, three pathways inhibitors, U0126, SB239063 and S3I-201, were added before AGEs stimulation, and this pre-treatment sig- nificantly decreased the number of apoptotic cells and inhibited the increase in IL-6 and VEGF-A. These results suggest that these three signalling pathways are involved in AGEs-induced apoptosis and up- regulation of IL-6 and VEGF-A. Furthermore, the RAGE antagonist FPS- ZM1 was used to pre-treat osteocytes, which were then treated with AGEs. The results show that FPS-ZM1 significantly decreased the
activation levels of the three pathways, blunted the increasing levels of IL-6 and VEGF-A and reduced the numbers of apoptotic cells. These data indicate that AGEs activate the ERK1/2, p38 and STAT3 pathways via RAGE and that these three pathways are involved in the increased rate of apoptosis and the up-regulation of IL-6 and VEGF-A in osteo- cytes.

This study has some shortcomings. MLO-Y4 cells are not identical to osteocytes in vivo, and osteocytes in vivo may not accumulate the same AGEs that we used. Therefore, further in vivo studies are needed to validate our findings.In sum, we conclude that AGEs can activate the ERK1/2, P38 and STAT3 signalling pathways via RAGE and that their activation is in- volved in the AGEs-induced up-regulation of IL-6 and VEGF-A as well as apoptosis in osteocytes. These results highlight the role of osteocytes in the regulation of bone metabolism by AGEs.