PLGA+HA/βTCP Scaffold Incorporating Simvastatin: A Promising Biomaterial for Bone Tissue Engineering

Bone tissue has biological and mechanical characteristics that require complex clinical strategies to allow the reconstruction of its structure and function. For this reason, bone tissue engineering has been proposed to reconstruct bone tissue in oral and maxillofacial approaches.¹  The purpose of tissue engineering is to mimic the 3 main components of living tissues, which are the cells, extracellular matrix (ECM), and signaling molecules. Therefore, the main requirements for an effective bone tissue engineering strategy are as follows: (1) sufficient number of viable bone-forming cells, (2) scaffold capable of supporting these cells and providing adequate blood supply, and (3) substances able to induce osteogenic differentiation of mesenchymal stem cells.² 

Mesenchymal stem cells are a relevant option for bone tissue engineering once they are able to differentiate into different cell lines,³  including osteoblasts and endothelial cells. The activity of these cells is regulated by signaling molecules in the human body and by the micro- and nano-interactions with ECM.²  For this reason, the development of scaffolds requires the reproducibility of the characteristics and performance of the specific ECM for each tissue.² 

Regarding scaffold synthesizing, poly-d,l-lactic-co-glycolic copolymer (PLGA) is a biocompatible biomaterial, easily synthesized, and biodegradable in nontoxic byproducts.^4–6  Although PLGA is not considered osteoinductive,^7,8  it allows the incorporation and release of biomolecules with substantivity.^8,9  Hence, the degradation mechanism of PLGA is extremely useful for the controlled release of signaling molecules.²  Nevertheless, PLGA has demonstrated reduced cell adhesion and proliferation in response to its hydrophobicity.^10,11  Notably, because of their biocompatibility and chemical similarity to the mineral component of mammalian bones, hydroxyapatite (HA) and β-tricalcium phosphate (βTCP) are used to complement the properties of copolymers.^2,6,12,13  The use of HA and βTCP in conjunction with copolymers compensates its low mechanical properties and brittleness. Moreover, these ceramics may increase the hydrophilicity of polymeric scaffolds.¹⁴  Therefore, porous PLGA+HA/βTCP composite scaffolds should minimize the hydrophobicity of PLGA and the mechanical weakness of HA/βTCP biphasic ceramic.^6,13 

Simvastatin (SIM) is an inhibitor of 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in the cholesterol synthesis pathway, used to treat hypercholesterolemia,¹⁵  that interestingly acts on the osteogenic differentiation.¹⁶  Despite the promising results of SIM to promote bone regeneration,^16–19  an efficient scaffold that incorporates the adequate dose of SIM and reproduces the properties of the bone ECM to permit bone tissue engineering has not been developed. Thus, this study aimed to synthesize, characterize, and evaluate the degradation and biocompatibility of PLGA+HA/βTCP scaffolds incorporating SIM. The null hypothesis of this study was that PLGA+HA/βTCP scaffold incorporating SIM would not be a promising biomaterial for bone tissue engineering.

Scaffolds were produced with PLGA and biphasic ceramic (HA/βTCP) obtained by the solvent evaporation technique and leaching of sucrose particles. The PLGA composite was obtained from 50% (m/m) polylactic acid (Resomer LT 706S, Evonik Boehringer Ing, Pharma GmbH & Co, Germany) and 50% (m/m) lactide-co-glycolide (Resomer LG 824S, Evonik Boehringer Ing, Pharma GmbH & Co). Copolymers were solubilized in 10% (m/v) chloroform PA (Synth) at room temperature under constant stirring for 24 hours. After complete dissolution of the copolymers, biphasic ceramic was added in the ratio 1:1 (m/m) between polymer and ceramic. Then, sucrose particles (75% m/m, Merck) up to 500 μm diameter were added to the solution. The blend was placed in cylindrical molds until total solvent evaporation. Samples were sectioned to obtain cylinders of 5 mm in diameter and 1 mm in height. About 50 samples weighting approximately 0.015–0.025 g were obtained for each group. Sucrose was removed using polyvinyl alcohol baths for 24 hours, followed by 24-hour baths with distilled water under constant stirring. Samples with SIM (M = 418.57, Sigma-Aldrich) received 1% (m:m) of this medication solubilized in chloroform with the polymer, and the subsequent steps followed the same method described above. Samples were sterilized by ethylene oxide.

PLGA+HA/βTCP and PLGA+HA/βTCP+SIM scaffolds were characterized and analyzed for rate and quality of degradation.^4,20  The cylinders were immersed in phosphate-buffered saline (PBS) at 37°C under 35 rpm (TE-424 incubator, Tecnal) for 7, 14, 21, and 28 days. Nondegraded samples (day 0) were taken as reference. After drying in laminar flow hood for 24 hours, the following tests were performed in triplicate for each experimental time:

• Mass variation: Samples were weighted in analytical balance (Shimadzu Corporation, Kyoto, Japan) to verify the mass changes after degradation.

• Scanning electron microscopy (SEM): Samples were coated with gold-palladium and analyzed by electron microscope (TM3030, Hitachi) at 10 kV. For each sample, 4 records with different magnifications (×50, ×100, ×500, ×1000) were obtained. SEM images were obtained to analyze scaffolds morphology.

• Analysis of porosity: From the ×50 images obtained by SEM, the 25 major porous areas (cm²) were calculated for each experimental time using ImageJ software (National Institutes of Health) to verify if there are variations on porosity after degradation experiment.

• Fourier transform infrared spectroscopy (FTIR): FTIR analyses were performed with a spectrophotometer (Cary 660 Series FTIR Spectrometer, Agilent Technologies Inc) with a horizontal attenuated total reflectance accessory (ZnSe). Samples were placed directly on the crystal, and the mean of 20 scans in the range of 4000–650 cm⁻¹ and resolution of 4 cm⁻¹ was performed for each sample. FTIR was carried out to determine the chemical composition of the samples.

• Differential scanning calorimetry (DSC): Thermograms were obtained using DSC-60 (Shimadzu Corporation). Approximately 1.5 mg of each sample was sealed in aluminum melting pot and subjected to a heating rate of 10°C/min under a flow-nitrogen atmosphere of 100 mL/min. The temperature range analyzed was 25–350°C in a single run. The DSC cell was calibrated with indium and zinc. The obtained data were processed using TA-60 software (Shimadzu Corporation). DSC analysis measures the amount of energy absorbed or released by a sample when it is heated or cooled, providing quantitative and qualitative data on endothermic processes.

• Thermogravimetry (TG): Analyses were performed using a thermobalance TGA-50 (Shimadzu Corporation). Approximately 5 mg of each sample was placed in open platinum crucibles, heated at a rate of 10°C/min¹  through a temperature range up to 350°C and under a nitrogen flow rate of 100 mL/min. The instrument was previously calibrated with a standard reference of calcium oxalate. The obtained data were processed using TA-60 software (Shimadzu Corporation). The derived thermogravimetric curves (DrTG) were used to identify the maximum degradation temperature (T_deg.max). TG analyses are performed to determine the temperature and weight change of decomposition reactions and the water content or the residual solvents in a material.

Scaffolds hydrophilicity was evaluated through the drop test. Thus, 30 μL Dulbecco's modified Eagle medium (DMEM, Gibco, Life Technologies) were dripped onto each scaffold, and the behavior of the drop was evaluated through standardized photographs (EOS T5, macro lens EF 100 mm, Canon) using a tripod and maintaining the same conditions of light and distance to the object. The images were obtained at the time of dripping and 1 hour later. The contact angles were measured using ImageJ software (National Institutes of Health). Wettability tests are important to predict the interaction of the biomaterial with organic fluids.

The toxicity of the scaffolds in cell cultures was evaluated by MTS (3-[4,5-dimethylthiazol-2-yl]-5-[3-carboxymethoxyphenyl]-2-[4-sulfophenyl]-2H-tetrazolium, inner salt, CellTiter 96, Promega). Preosteoblasts MC3T3-E1 subclone 4 2593 (American Type Culture Collection) and stem cells from human exfoliated deciduous teeth (SHED, Curityba Biotech) were cultured at 2 × 10⁴ cells per well in 96-well plates at 37°C and 5% CO₂. Modified αMEM (Nutricell) was used to seed MC3T3-E1, and DMEM (Gibco, Life Technologies) was used for SHED. All media were supplemented with 10% fetal bovine serum (FBS, Gibco, Life Technologies). For the experiments, after 24 hours of cell seeding, samples were added into each well. The positive control consisted of cellular culture with no sample, and the negative control was the culture media only. The MTS colorimetric test was performed in quadruplicate at days 1, 3, and 7. After each experimental period, samples and media were gently removed, and wells were washed with PBS. Then, 15 μL MTS solution was dispensed onto the wells with 75 μL media, and the plate was incubated for 4 hours at 37°C and 5% CO₂. The absorbance was measured by spectrophotometer (SpectraMax M2, Molecular Devices) at 570 nm, and the percentages of viable cells were calculated in relation to the positive controls.

An independent statistician analyzed all datasets with statistical software (GraphPad Prism 8 Software Inc). Qualitative and descriptive analyses were performed for samples processed in SEM, FTIR, and hydrophilicity tests. Statistical analyzes were performed for the results of mass variation and cytotoxicity using 1-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. For mass variation analyses, results were compared among different experimental time evaluating each group separately. For porosity analyses and DSC and TG numerical results, the t test was applied. Results were compared among groups evaluating each experimental time separately. The level of significance was set at P < .05.

Mass variation results are recorded in Figure 1a. PLGA+HA/βTCP scaffolds at days 21 and 28 were bulkier than at days 0, 7, and 14 (ANOVA, 0–21: P < .0001; 0–28: P < .0001; 7–21: P = .0005; 7–28: P < .0001; 14–21: P = .0411; 14–28: P = .0003), whereas no significant difference was found for mass determination of PLGA+HA/βTCP+SIM. According to Figure 1b, PLGA+HA/βTCP and PLGA+HA/βTCP+SIM demonstrated similar porosity, except for day 21 of degradation (t test, P = .0021), when PLGA+HA/βTCP demonstrated greater porosity. The presence of micro- (100–500 μm), meso- (500–1000 μm), and macropores (≥1000 μm)²¹  were observed in the structure of the polymer with particles of HA/βTCP homogeneously dispersed (Figure 2) in both PLGA+HA/βTCP and PLGA+HA/βTCP+SIM. There were no microcracks in the polymer or ceramic structures for both scaffolds (Figure 2). Regarding degradation, the physical structure of the scaffolds remained practically unaltered at days 0, 7, and 28 (Figure 2).

According to Figure 3, an equivalent pattern was observed for PLGA+HA/βTCP (black line) and PLGA+HA/βTCP+SIM (red line) scaffolds. There was a prevalence of 1750 cm⁻¹ absorption bands, which represents double stranded between carbon and oxygen (C=O). Bands between 500–1500 cm⁻¹ that include esters (C–O) and hydrocarbons (CH₂ and CH) of PLGA and PO₄⁻³ of apatites were observed. Some traces of CH, CH₃, and CH₂ hydrocarbons around 3000 cm⁻¹ were detected. The results at all experimental times were similar.

The DSC thermograms (Figure 4a) followed a similar pattern for PLGA+HA/βTCP (black line) and PLGA+HA/βTCP+SIM (red line) scaffolds, with very close numerical temperature values. The equipment was not able to detect the glass transition temperature of the samples. The endothermic peaks occurred at about 150–160°C, returning more intensely after 325°C. Because many of the curves did not end their course up to the 350°C analyzed, the temperature values of the second peaks were not possible to calculate. The endothermic peaks were higher for PLGA+HA/βTCP (156.96 ± 1.11°C) than PLGA+HA/βTCP+SIM (153.99 ± 0.65°C) (t test, P = .0161). The TG graphs (Figure 4b) were also very similar over the characterization and degradation period for both PLGA+HA/βTCP and PLGA+HA/βTCP+SIM scaffolds. TG curves present the variation of mass (%) and the derivative of the mass loss (TGA or DrTGA) as a function of temperature (°C). The midpoints (T_deg.max.) remained stable over the degradation days for PLGA+HA/βTCP (346.63 ± 11.34) and PLGA+HA/βTCP+SIM (333.82 ± 17.03) (t test, P = .1991). The mass loss was also similar for PLGA+HA/βTCP (−2.50 ± 0.19 mg) and PLGA+HA/βTCP+SIM (−2.60 ± 0.28 mg) (t test, P = .5263).

PLGA+HA/βTCP+SIM scaffolds were able to immediately absorb the DMEM, whereas PLGA+HA/βTCP scaffolds maintained the drop at an obtuse angle. After 1 hour, the greatest diffusion of DMEM was observed for scaffolds embedding SIM (Figure 5).

PLGA+HA/βTCP and PLGA+HA/βTCP+SIM scaffolds were biocompatible for osteoblasts (MC3T3-E1) and mesenchymal stem cells (SHED) (Figure 6) because the percentage of viable cells were more than 70% (ISO 10993-5:2009).²²  For osteoblasts, PLGA+HA/βTCP+SIM exhibited more viable cells than other groups at day 1 (ANOVA, P = .0006). At days 3 and 7, there were no significant differences among groups (ANOVA, P = .1600 and P = .8420, respectively). For mesenchymal stem cells, there were no significant differences between PLGA+HA/βTCP and PLGA+HA/βTCP+SIM at days 1, 3, and 7; but controls were different at all analyzed times (ANOVA, P = .0002, P = .0061, and P = .0014, respectively; Figure 6).

Bone tissue engineering using bone-forming cells, porous scaffolds, and osteoinductor agents may lead to an alternative to improve surgical techniques that aim to reconstruct bone defects. Despite the promising results of several studies using bone tissue engineering for oral and maxillofacial area,^12,23,24  bone tissue engineering is still a challenge because of the difficulty in obtaining the adequate scaffold to mimic bone ECM for cells and blood supply and to deliver osteogenic molecules efficiently with no impairment to the surrounding tissues. For these reasons, we proposed a scaffold for bone tissue engineering based on PLGA+HA/βTCP incorporating SIM. We also characterized and evaluated the degradation and biocompatibility of this scaffold. In summary, we produced PLGA+HA/βTCP scaffold incorporating SIM using an affordable and simple method that could be easily applied on a large scale. We demonstrated PLGA+HA/βTCP scaffold incorporating SIM presented adequate composition and porosity for bone tissue engineering. In addition, the proposed scaffold maintained its structural, chemical, and thermal integrity over a period of 28 days of degradation. Moreover, PLGA+HA/βTCP scaffold incorporating SIM was biocompatible for osteoblasts and mesenchymal stem cells.

Herein, the porous PLGA+HA/βTC scaffold incorporating SIM was designed to gather the useful properties of each biomaterial. Thus, HA/βTCP ceramic releases calcium and phosphate ions, whereas PLGA incorporates bioactive substances to form a controlled-release system for osteoinduction. Additionally, HA/βTCP increases the hydrophilicity of PLGA,¹⁴  a property that is related to cell adhesion and proliferation.^10,11  Therefore, the proposed composite scaffold should minimize the hydrophobicity of PLGA and the mechanical weaknesses of HA/βTCP.^6,13  Interestingly, the incorporation of SIM into the PLGA+HA/βTCP scaffold also improved the hydrophilicity of PLGA, as we observed in the wettability test. In bone tissue engineering, hydrophobic scaffold impairs the cells adhesion and protein adsorption onto its surface,^2,18  compromising the bone reconstruction.

After some time in the presence of water, PLGA initiates the degradation by hydrolysis, primarily losing the mechanical properties and, later, mass.²⁵  Herein, the period analyzed did not allowed the verification of mass loss. After 28 days of degradation, the physical structure of the scaffolds remained essentially unaltered, which was already expected and desired. PLGA copolymers with high concentration of lactic acid, as in the present study (82:18), are less hydrophilic and absorb less water, leading to a slower degradation of the polymer chains.²⁶  We chose this PLGA composition because scaffolds for bone tissue engineering should be slowly replaced by newly formed bone tissue, maintaining adequate mechanical resistance to the masticatory forces. Furthermore, slow degradation rates allow the gradual release of the incorporated biomolecules, avoiding the bulk degradation, which could impair the physico-chemical properties of the biomaterial, mainly in terms of mechanical properties,¹⁸  besides the cytotoxicity effect because of the high dose of delivered SIM.

To allow bone tissue engineering, 3-dimensional porous scaffolds with internal structure mimetizing the native bone tissue are preferred because they have mechanical strength to resist along the time, help to conduct and support cells during proliferation and differentiation processes, and allow neovascularization, which is fundamental for the maintenance and integration of the scaffold in the host tissue.^1,2  We observed the pores maximum diameter ranged from 112 to 5583 μm, demonstrating the presence of macro-, meso-, and micropores²¹  in the polymer structure, with HA and βTCP particles homogeneously dispersed. There is an indication that pore sizes around 200–1500 μm more closely mimic the native trabecular bone.²⁷  However, histologic analysis revealed that macropores, such as we observed, allowed more cellular growth and uniform mineral deposition.²¹  The uniformity and interconnectivity of the pores in the scaffolds detected in this study are fundamental for bone tissue engineering approaches, because these properties facilitate the penetration of fluids, cell culture medium, growth factors, and cells through the material and could promote the tissue formation in an organized network.^6,13,28  Moreover, the drug release rate from scaffolds is influenced by pore size and interconnectivity, besides polymer composition, hydrophobicity, crystallinity, and degradability.¹⁸ 

SIM is a small molecule that belongs to the statin group. It is known as coenzyme A reductase inhibitor, mainly used to decrease serum cholesterol levels, whereas it is interestingly linked to new bone formation by increasing the expression of BMP-2 in bone cells.²⁹  Besides the promising results of SIM for bone formation,^16–19  its action on osteogenic differentiation depends on the incorporated dose into the scaffolds and the ability of the scaffolds to promote its gradual release, allowing local osteogenic stimulus over prolonged periods of time.²³  The incorporation of SIM into the scaffold occurs because of the presence of polymer. FTIR analysis showed no band shift among the samples, indicating the drug was successfully encapsulated by the polymer in the present study. High doses (>1 mg/mL) of SIM promote cell death and exacerbated inflammatory responses in animal and human studies.^23,30,31  Herein, 680 μg/mL SIM was embedded into the scaffolds. This concentration was biocompatible to osteoblasts and mesenchymal stem cells. Additionally, the incorporation of SIM improved the scaffolds wettability.

As previously mentioned, PLGA+HA/βTC scaffold incorporating SIM was biocompatible. This information is related to the scaffold composition and to the concentration of the incorporated SIM. Mendes et al,³²  investigating the in vitro osteoinductive effect of 2% SIM incorporated in PLGA scaffolds on mesenchymal stem cells, observed a cytotoxic effect of the drug used at that concentration. On the other hand, SIM concentrations similar to the one used in the present study have shown relevant impact on osteogenic differentiation while minor cytotoxic effect were revealed.³³  It is important to highlight that our study used cell lineages related to the bone tissue engineering, reinforcing the relevance of the results found.

The main limitation of the present study is the short period of degradation analyzed, because the tests were performed for no longer than 28 days on the biomaterials behavior, whereas PLGA remains basically unaltered for a period of about 90 days.^23,24,34  Thus, we suggest future studies to evaluate the degradation of PLGA+HA/βTCP scaffold incorporating SIM for longer periods of time. In addition, the present study has all the limitations of a preliminary in vitro study, where no conclusions can be transferred for clinical situations. However, the in vitro preliminary results were promising for future research, especially regarding osteogenic differentiation of mesenchymal stem cells. Consequently, because PLGA+HA/βTCP scaffold incorporating SIM presented suitable results regarding physico-chemical and biological properties for bone tissue engineering, further studies should include osteogenic differentiation assays to evaluate the ability of the SIM-incorporated scaffolds to differentiate mesenchymal stem cells into osteoblasts-like cells, and further in vivo and clinical studies must be performed to elucidate the biological behavior of the proposed scaffold incorporating simvastatin.

In summary, PLGA+HA/βTCP scaffold incorporating SIM presented adequate structural, chemical, thermal, and biological properties for bone tissue engineering. Adequate scaffold composition, characteristics of porous surface, hydrophilicity, and biocompatibility were achieved. Moreover, the manufacturing technique proposed here is simple, reproducible, and affordable. These results are extremely promising for future applications of bone tissue engineering. Likewise, further research is warranted to evaluate the ability of PLGA+HA/βTCP scaffolds incorporating SIM to induce the osteogenic differentiation in vitro, as well as bone and vessel formation in vivo.

Abbreviations

ANOVA:

analysis of variance

βTCP:

β-tricalcium phosphate

DrTG:

derived thermogravimetric curve

DSC:

differential scanning calorimetry

ECM:

extracellular matrix

FTIR:

Fourier transform infrared spectroscopy

HA:

hydroxyapatite

HMG-CoA:

3-hydroxy-3-methyl-glutaryl coenzyme A

MTS:

3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium

PBS:

phosphate-buffered saline

PLGA:

poly(lactic-co-glycolic acid)

SEM:

scanning electron microscopy

SHED:

stem cells from human exfoliated deciduous teeth

SIM:

sinvastatin

T_deg.max:

maximum degradation temperature

TG:

thermogravimetry

This project was supported by International Team for Implantology Foundation Grant 1113_2015. The authors thank Curityba Biotech (Curitiba, PR, Brazil) for providing the mesenchymal stem cells. The authors acknowledge Prof Gislaine Fongaro, MSc, PhD, Department of Microbiology, Federal University of Santa Catarina, Brazil, for reviewing the methodology and the statistical analyses of the results and conclusion of this study. The authors are grateful to the Laboratório Central de Microscopia Eletrônica at the Universidade Federal de Santa Catarina (UFSC) for providing their equipment and infrastructure to perform some experimental analyses. The authors thank the University Hospital Sterilization Service at UFSC for samples sterilization.

The authors declare no conflicts of interest.