Insulin growth factor binding protein-3 enhances dental implant osseointegration against methylglyoxal-induced bone deterioration in a rat model

MC3T3 E1 cell culture and MGO treatment

MC3T3 E1 cells (DCRL-2593; American Tissue Type Collection, Manassas, VA, USA) were cultured in α-MEM (Gibco BRL, Grand Island, NY, USA) supplemented with 2 mM glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, and 10% fetal bovine serum in a humidified 5% CO₂ atmosphere at 37°C and sub-cultured at a 1:4 ratio. The mineralization experiments for Alizarin red staining and alkaline phosphatase (ALP) activity were performed with MC3T3 E1 cells cultured in 50 μg/mL ascorbic acid, 10 mM β-glycerophosphate, and 100 nM dexamethasone, as previously described [33]. To confirm the response of MGO-treated MC3T3 E1 cells, 80% confluent cells were exposed to MGO (400 μM); fresh medium was replaced 2 hours later, and the cells were cultured for 15 days.

Preparation of Ch-GNPs

Ch-GNPs were prepared by a simple graft-on technique as previously described [27]. Briefly, 2 mL of 0.33% chitosan solution and 0.1 M hydrochloric acid (HCl) was mixed with 1 mL of a 10 mM freshly prepared chloroauric acid (HAuCl₄) solution and stirred for 1 hour. Later, the prepared solution was constantly mixed with 0.1 M ice-cold, freshly prepared sodium borohydrate. A rapid change to a red wine color indicated the formation of Ch-GNPs. The Ch-GNPs were collected by ultracentrifugation at 35,000 × g at 4°C for 30 minutes. The Ch-GNP stock solution was used in triple-distilled water for further experimentation.

Preparation of DNA complexes

Complexes of Ch-GNPs and plasmid DNA (pcDNA3.1 IGFBP-3 and LacZ) were prepared as previously described [23]. In brief, the Ch-GNPs and plasmid DNA (pcDNA3.1 IGFBP-3, pcDNA3.1 LacZ, Invitrogen, Carlsbad, CA, USA) were used to prepare the complexes in water, and the Ch-GNP solution (40 μg) from the stock solution (50 mg/mL) of nanoparticles was mixed with 20 μg of DNA.

Loading of Ch-GNP/DNA complexes on Ti surfaces and mini-screws

The cDNA of LacZ and IGFBP-3 was cloned in plasmid DNA. The Ch-GNP/DNA complexes were deposited on cleaned Ti surfaces (6×6×0.1 cm for a 100-mm cell dish), using the dipping technique at room temperature. The Ch-GNP/DNA solution was mixed with 500 μL of serum and antibiotic-free medium, and the surface was coated with it, followed by drying at room temperature. The Ti surface coated with Ch-GNPs/IGFBP-3 was placed in a 100-mm-diameter cell culture plate, the same number of cells (5×10⁵) were seeded on the plate, and 1 mL of cell culture medium was added. After 3 hours, the cell culture medium was replaced with a fresh medium. For in vivo analysis, dental implants were prepared as previously described [22]. Briefly, commercially available pure, cylindrically shaped Ti square thread screws (4.5 mm in length and 0.85 mm in diameter) were used as dental implants. For coating with Ch-GNPs/IGFBP-3 and LacZ, the implants were immersed 10 times in a nanoparticle DNA solution and frozen at −40°C. The total coating amount of the DNA was 20 μg each. Ch-GNP/LacZ-coated implants were used as a control.

Western blot analysis

A previously described method was used for western blot analysis [27]. In brief, total proteins were extracted from the MC3T3 E1 cells with a lysis buffer containing 150 mM NaCl, 5 mM EDTA, 50 mM Tri-HCl (pH 8.0), 1% NP 40, 1 mM pepstatin, 1 mM aprotinin, and 0.1 mM leupeptin. Proteins in the cells were quantified by the Bradford dye-binding procedure (Bio-Rad, Hercules, CA, USA). The samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (8% to 15%) under denaturing conditions and transferred to a Hybond-P membrane (Amersham, Arlington, IL, USA). Specific primary antibodies were used at a ratio of 1:1,000 to 1:2,000 and incubated at 4°C overnight, and then incubated with a horseradish peroxidase-IgG-conjugated secondary antibody at room temperature for 1 hour. A chemiluminescence detection reagent was used to detect the signals according to the manufacturer's protocol (Amersham Pharmacia Biotech, London, UK) with a LAS-4000 CCD imaging system (Fujifilm, Tokyo, Japan).

Alizarin red staining and ALP activity

Alizarin red staining was performed in the MC3T3 E1 cells after 3, 6, 9, 12, and 15 days of differentiation induction culture. After the indicated times of cell culture, the cells were washed with phosphate-buffered saline (PBS), air-dried, and fixed in 95% ice-cold ethanol at −20°C for 30 minutes. After fixation, the cells were stained with 40 mM of Alizarin red stain (pH 4.2) at room temperature for 1 hour. The plates were washed with deionized water 5 times and then rinsed with PBS (without magnesium and calcium) for 15 minutes.

ALP activity was measured in MC3T3 E1 cells collected in cold PBS at 3, 6, 9, 12, and 15 days and sonicated with a cell disruptor (Heat System-Ultrasonics, Plainview, NJ, USA) in an ice bath. ALP activity in the supernatant was measured using the SensoLyte pNPP Alkaline Phosphatase Assay Kit (AnaSpec, Inc., Fremont, CA, USA) according to the manufacturer's protocol [34].

Determination of ROS generation

ROS generation of MC3T3 E1 cells was measured by the Muse Oxidative stress kit using the Muse cell analyzer (Merck Millipore, KGaA, Darmstadt, Germany) as a fluorescent-based analytical technique. The manufacturer-specific protocol was followed for the assay. In brief, MC3T3 E1 cells were treated with LacZ and IGFBP-3 for 1 hour prior to MGO (400 μM) treatment and incubated for 24 or 48 hours. Samples of 1×10⁷ cells/mL were prepared in 1× assay buffer and treated with an oxidative stress reagent based on dihydroethidium; this reagent is used to detect ROS that are oxidized with superoxide anion to produce the DNA-binding fluorophore ethidium bromide, which intercalates with DNA, resulting in red fluorescence.

Animals and surgical procedures

The Animal Ethical Committee of Jeonbuk National University (CBNU-2019-00299) approved the protocol for the use of animals in the study. Six-week-old Sprague-Dawley male rats were used in the experiment. The animals were randomly assigned to 4 groups: no MGO administration (control; n = 12), MGO administration (MGO; n = 12), MGO administration with Ch-GNPs/LacZ (MGO-LacZ; n = 12), and MGO administration with Ch-GNPs/IGFBP-3 (MGO-IGFBP-3; n = 12). The dosage of MGO was determined according to previous studies [35]. First, to observe changes in bone quality by MGO in vivo, PBS was injected into the control group and 75 mg/kg of MGO was injected into the MGO group twice a week for a total of 6 weeks. All surgical procedures were performed under general anesthesia induced with zolazepam (Zoletil 50; Virbac Carros, France) and xylazine hydrochloride (Rompun; Bayer Korea, Seoul, Korea). The lower first molar was extracted carefully to avoid damage to the extraction socket. The animals were given intramuscular injections of amikacin for up to 3 days. At 1 week after tooth extraction, 75 mg/kg of MGO was intraperitoneally injected into both groups twice a week for a total of 10 weeks. At 4 weeks after tooth extraction, the specified implants were inserted into the indicated groups. After 3 and 6 weeks, the rats were euthanatized by cervical dislocation under general anesthesia, and samples were collected for examination.

Micro-computed tomography analysis

Micro-computed tomography (μCT) was performed with an anode electrical current of 100 kV at a resolution of 18 μm using a model 1076 apparatus (Skyscan, Kontich, Belgium). After anesthesia, the femur and mandibles were scanned with μCT to detect dynamic changes in the tissues and peri-implant tissue at 3 and 6 weeks. The regions of interest (ROIs) that included the trabecular compartment around the femur were selected. The ROI of the alveolar bone was manually established in the interradicular septum bone of the right mandibular first molar (M1) without an implant. The coronal and horizontal planes of M1 were confirmed by 2-dimensional images, which were generated by DataViewer (Skyscan). First, in the coronal plane passing through the center of the buccal and lingual roots, 2 horizontal surfaces were selected that individually passed through the alveolar ridge crest and apex of the buccal root. Second, on the horizontal plane of the M1 tooth, the interalveolar septum was selected by drawing a contour from the center of one root canal to another root canal by avoiding the roots and other structures. After scanning, 3-dimensional (3D) models were generated by CTVol (Skyscan), and the bone volume and density around the implants were analyzed using CTAn (Skyscan), which was also used to examine the μCT datasets for new bone growth. The collective sum of all ROI layers over a continuous set of cross-sectional image slices represented the volume of interest in the regenerated bone. Furthermore, new bone around the hole and bone mineral density (BMD) were calculated by phantom and Hounsfield units (HU) (low phantom [0.25] 1157.7907 HU and high phantom in μCT HU [0.75] 3233.3492 HU). Binary thresholds (gray-scale index, implant area: 160 mm × 255 mm; new bone area: 100 mm × 143 mm; and total bone area: 70 mm × 120 mm) were used to create the 3D images.

Histology and immunohistochemical staining

The mandibles were isolated and fixed in 10% neutral-buffered formalin solution. After fixation, the tissues were decalcified in 15% EDTA and 0.1 M Tris (pH 7.0). After decalcification, the implant was removed. The tissues were dehydrated with different percentage of alcohol, cleared in xylene, and embedded in paraffin. Tissue sections of 8 μm were mounted on glass slides and stained with hematoxylin and eosin (H&E) and immunohistochemical (IHC) stains. IHC staining was performed to detect the expression of IGFBP-3, BMP-2, BMP-7, osteoprotegerin (OPG), receptor activator of nuclear factor-κB ligand (RANKL), and receptor for advanced glycation end products (RAGE) using an immunohistochemistry accessory kit (Bethyl Laboratories, Montgomery, TX, USA). The primary antibodies were used at 1:200 dilutions according to the protocol. The slides were visualized microscopically (Carl Zeiss, Ostalbkreis, Germany). The levels of antibody expression were measured with ImageJ (National Institutes of Health, Bethesda, MD, USA) software.

Statistical analysis

All results were analyzed independently. All values are presented as the mean±standard deviation of 3 independent experiments. Statistical significance was assessed by the unpaired t-test. P values <0.05 were considered to indicate statistical significance. This in vivo study was conducted in groups containing 6 rats each.

MGO impairs bone formation in MC3T3 E1 cells

MC3T3 E1 cells were treated with MGO (400 μM) in osteogenic medium for up to 15 days for the detection of osteogenic differentiation and mineralization. The cells treated with MGO demonstrated decreased Alizarin red staining in a time-dependent manner compared to the mock osteogenic medium (Figure 1A). Similarly, ALP activity was also significantly decreased by MGO treatment at 9, 12, and 15 days compared to the mock osteogenic medium (Figure 1B). The expression of osteogenic differentiation proteins (BMP-2, BMP-7, and OPG) was also decreased by MGO, whereas the expression of RANKL was increased by MGO treatment in a time-dependent manner (Figure 1C).

Figure 1
MGO impairs the function of bone formation in MC3T3 E1 cells. (A) Alizarin red staining after treatment of MC3T3 E1 cells with MGO for 3, 6, 9, 12, and 15 days. (B) Effects of MGO on ALP activity in MC3T3 E1 cells at 3, 6, 9, 12, and 15 days. MC3T3 E1 cells were treated with OM with 5 mM β-glycerol phosphate, 100 μM ascorbic acid, and 10 nM dexamethasone for the indicated times and ALP was measured. (C) Protein expression of BMP-2, BMP-7, RANKL, and OPG analyzed by western blots after treatment with MGO. (D) Effects of MGO on the expression of inflammatory proteins and anti-oxidant enzymes analyzed by western blots. (E) Expression of IGFBP-3 in MC3T3 E1 cells induced with or without MGO. Each value was reported as the mean±standard deviation of 3 experiments.
MGO: methylglyoxal, ALP: alkaline phosphatase, BMP: bone morphogenetic protein, OM: osteogenic medium, RANKL: receptor activator of nuclear factor-κB ligand, OPG: osteoprotegerin, IGFBP-3: insulin growth factor binding protein-3.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide

MGO changes the expression of inflammatory, anti-oxidant, and IGFBP-3 proteins

Inflammation-related molecules (tumor necrosis factor-alpha [TNF-α], interleukin [IL]-6, and RAGE) were detected after the treatment of MC3T3 E1 cells with MGO. The cells demonstrated an increased expression of these molecules relative to controls (Figure 1D). Further, anti-oxidant activity was also evaluated by analyzing anti-oxidant molecules (Cu/Zn-superoxide dismutase [SOD] and Mn-SOD). These SOD enzymes were reduced in cells treated with MGO (Figure 1D). Similarly, the expression of IGFBP-3 protein also decreased after treatment with MGO (Figure 1E).

IGFBP-3 increases osteogenic differentiation and mineralization in MGO-treated MC3T3 E1 cells

The effect of IGFBP-3 in the response to MGO (400 μM) was examined in MC3T3-E1 cells after treatment with the conjugated Ch-GNP/IGFBP-3 complexes. Ch-GNP/LacZ complexes were used as the Ch-GNP control vector. The cells treated with Ch-GNPs/LacZ and MGO demonstrated decreased Alizarin red staining in a time-dependent manner. In contrast, the Ch-GNP/IGFBP-3-treated cells showed increased Alizarin red staining at 3, 9, and 15 days compared to the control Ch-GNP/LacZ cells, even when the cells were induced with MGO (Figure 2A). Correspondingly, ALP activity was also significantly increased by the Ch-GNP/IGFBP-3 complexes at 6 and 9 days compared to Ch-GNPs/LacZ (Figure 2B). The expression of IGFBP-3 protein significantly increased up to 15 days after the cells were treated with conjugated Ch-GNPs/IGFBP-3 (Figure 2C). MC3T3 E1 cells treated with the conjugated Ch-GNP/IGFBP-3 complexes demonstrated higher expression levels of BMP-2, BMP-7, and OPG than the Ch-GNP/LacZ cells, even though the cells were induced with MGO. RANKL expression was decreased by Ch-GNPs/IGFBP-3 treatment (Figure 2D).

Figure 2
IGFBP-3 increased bone formation in MGO-treated MC3T3 E1 cells. (A) Analysis of osteogenic differentiation in response to Ch-GNPs/IGFBP-3 in MC3T3 E1 cells induced with MGO and measurement of calcium deposition for matrix mineralization by Alizarin red staining. (B) Measurement of ALP activity by Ch-GNPs/IGFBP-3 in MC-3T3 E1 cells induced with MGO. (C) Analysis of protein expression of IGFBP-3 by western blots in MC3T3 E1 cells after treatment with Ch-GNPs/IGFBP-3 and induction with MGO. (D) Expression of osteogenic-related proteins (BMP-2, BMP-7, and OPG) and an osteoclastic protein (RANKL). (E) Expression of inflammatory (TNF-α, IL-6, and RAGE) and anti-oxidant proteins. (F) Measurement of ROS formation by the Muse oxidative stress assay. (G) Graphical representation of ROS formation. Each value was reported as the mean±standard deviation of 3 experiments.
IGFBP-3: insulin growth factor binding protein-3, MGO: methylglyoxal, Ch-GNP: chitosan gold nanoparticle, MGO: methylglyoxal, BMP: bone morphogenetic protein, OPG: osteoprotegerin, RANKL: receptor activator of nuclear factor-κB ligand, TNF: tumor necrosis factor, IL: interleukin, RAGE: receptor for advanced glycation end products, ROS: reactive oxygen species.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide

IGFBP-3 activates anti-inflammatory and anti-oxidant expression in MGO-induced MC3T3 E1 cells

MGO-induced Ch-GNPs/LacZ cells demonstrated increased expressions of TNF-α, IL-6, and RAGE at the indicated times (Figure 2E). High expression of the anti-oxidant molecules Cu/Zn-SOD and Mn-SOD was seen in IGFBP-3-overexpressing MC3T3 E1 cells, even though the cells were induced with MGO (Figure 2E). The levels of ROS were analyzed 24 and 48 hours after the cells were induced with MGO. The ROS levels were significantly reduced in Ch-GNP/IGFBP-3 cells compared to the Ch-GNPs/LacZ cells in a time-dependent manner (Figure 2F and G).

MGO reduces femur and alveolar bone formation in MGO-administered rats

μCT analysis and H&E staining were performed after the administration of MGO to rats. μCT examination of the femur 3 and 6 weeks after MGO administration demonstrated lower femoral cortical BMD, trabecular BMD, trabecular bone volume/total volume (BV/TV), trabecular number, trabecular thickness, and higher values for the femoral trabecular space compared to the controls (Figure 3A). However, μCT examination of the alveolar bone showed lower BMD and BV/TV in the MGO-administered group than in the control group (without MGO administered) at only 6 weeks (Figure 3B). The H&E staining results also confirmed a higher level of bone loss in the MGO-administered group at 3 and 6 weeks (Figure 3C).

Figure 3
MGO impairs femur and alveolar bone formation in MGO-administered rats. (A) μCT (2-dimensional) imaging of rat femur after the administration of MGO (75 mg/kg) for 3 and 6 weeks. Analysis of femoral cortical BMD, trabecular BMD, BV/TV, trabecular number, thickness, and space. (B) Detection of changes in rat alveolar bone by 2-dimensonal μCT after the administration of MGO and analysis of mandible BMD and BV/TV. M1: the mandibular first molar; M: mesial root; B: buccal root; D: distal root; L: lingual root; and T1 and T2: the 2 parallel horizontal planes passing through the alveolar ridge and the apex of the buccal muscle separately. (C) H&E staining of rat mandible with or without the administration of MGO. Data are expressed as mean±SEM at 3 and 6 weeks (n=6).
MGO: methylglyoxal, μCT, micro-computed tomography, BMD: bone mineral density, BV: bone volume, TV: total volume, H&E: hematoxylin and eosin, SEM: standard error of the mean.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide

IGFBP-3 restores mandibular bone deterioration in MGO-administered rats

To examine the role of IGFBP-3 in the osseointegration of dental implants in the MGO-administered rat model, the first molar of the rat mandibles was extracted and the recovery of bone deterioration was checked 3 and 6 weeks after MGO administration (Figure 4A). Bone formation by Ch-GNPs/IGFBP-3 near the implant sites was higher than that of Ch-GNPs/LacZ in a time-dependent manner (Figure 4B). The 3D μCT images showed comparatively increased new bone formation, BMD, and BV in the Ch-GNP/IGFBP-3 group (Figure 4C). Furthermore, H&E staining also confirmed the recovery of bone deterioration in response to Ch-GNP/IGFBP-3-coated implants in MGO-administered rat mandibles compared to the Ch-GNP/LacZ group at 3 and 6 weeks (Figure 4D).

Figure 4
IGFBP-3 restores MGO-induced bone deterioration in MGO-administered rats. (A) Two-dimensional μCT analysis of Ch-GNP/LacZ- and Ch-GNs/IGFBP-3-coated implants 3 and 6 weeks after the administration of MGO. (B) Three-dimensional μCT analysis of Ch-GNP/LacZ- and Ch-GNPs/IGFBP-3-coated implants. (C) Analysis of BMD and BV/TV. (D) H&E staining of the mandible around the implant surface 3 and 6 weeks after implantation. Data are expressed as mean ± SEM at 3 and 6 weeks (n=6).
IGFBP-3: insulin growth factor binding protein-3, MGO: methylglyoxal, μCT, micro-computed tomography, Ch-GNP: chitosan gold nanoparticle, BMD: bone mineral density, BV: bone volume, TV: total volume, H&E: hematoxylin and eosin, SEM: standard error of the mean.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide

Ch-GNP/IGFBP-3-coated implants decrease inflammatory molecules and increase osteogenic differentiation molecules in MGO-administered rats

IHC analysis demonstrated that the expression of IGFBP-3 increased at 3 and 6 weeks in the Ch-GNP/IGFBP-3-coated implant group (Figure 5A). These results indicated that the Ch-GNPs carried the reporter gene to the implantation site. Levels of inflammation-related molecules (TNF-α, IL-6, and RAGE) were increased in the Ch-GNP/LacZ-coated implant group 3 and 6 weeks after the administration of MGO, whereas levels of these molecules decreased in the Ch-GNP/IGFBP-3-coated implant group even with MGO administration (Figure 5A and B).

Figure 5
Ch-GNP/IGFBP-3-coated implants inhibit inflammatory proteins in MGO-administered rats. (A, B) Analysis of IGFBP-3, TNF-α, IL-6, and RAGE by IHC staining after 3 and 6 weeks. The intensity specific for the brown color correlates with the protein level.
Ch-GNP: chitosan gold nanoparticle, IGFBP-3: insulin growth factor binding protein-3, MGO: methylglyoxal, TNF: tumor necrosis factor, IL: interleukin, RAGE: receptor for advanced glycation end products, IHC: immunohistochemistry.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide

The expression of osteogenic differentiation molecules near the implant site was identified by IHC. The expression of BMP-2, BMP-7, and OPG was higher in the MGO-administered Ch-GNP/IGFBP-3-coated implant group than in the Ch-GNP/LacZ-coated implant group at 3 and 6 weeks (Figure 6A and B). The osteoclast-related molecule RANKL was more highly expressed in the MGO-administered Ch-GNPs/LacZ-coated implant group than in the Ch-GNPs/IGFBP-3-coated implant group in a time-dependent manner (Figure 6B).

Figure 6
Ch-GNP/IGFBP-3-coated implants increase osteogenic differentiation in MGO-administered rats. (A, B) IHC staining of BMP-2, BMP-7, OPG, and RANKL for the detection of osteoblast differentiation. The intensity specific for the brown color correlates with the protein level.
Ch-GNP: chitosan gold nanoparticle, IGFBP-3: insulin growth factor binding protein-3, MGO: methylglyoxal, IHC: immunohistochemistry, BMP: bone morphogenetic protein, OPG: osteoprotegerin, RANKL: receptor activator of nuclear factor-κB ligand.

^*P<0.05.

• Click for larger image
• Download as PowerPoint slide