Hybrid chitosan/gelatin/nanohydroxyapatite (CS/Gel/nHA) scaffolds have attracted considerable interest in tissue engineering (TE) of mineralized tissues. The present study aimed to investigate the potential of CS/Gel/nHA scaffolds loaded with dental pulp stem cells (DPSCs) to induce odontogenic differentiation and in vitro biomineralization.
CS/Gel/nHA scaffolds were synthesized by freeze-drying, seeded with DPSCs, and characterized with flow cytometry. Scanning Electron Microscopy (SEM), live/dead staining, and MTT assays were used to evaluate cell morphology and viability; real-time PCR for odontogenesis-related gene expression analysis; SEM-EDS (Energy Dispersive X-ray spectroscopy), and X-ray Diffraction analysis (XRD) for structural and chemical characterization of the mineralized constructs, respectively.
CS/Gel/nHA scaffolds supported viability and proliferation of DPSCs over 14 days in culture. Gene expression patterns indicated pronounced odontogenic shift of DPSCs, evidenced by upregulation of DSPP, BMP-2, ALP, and the transcription factors RunX2 and Osterix . SEM-EDS showed the production of a nanocrystalline mineralized matrix inside the cell-based and - to a lesser extent - the cell-free constructs, with a time-dependent production of net-like nanocrystals (appr. 25−30 nm in diameter). XRD analysis gave the crystallite size ( D = 50 nm) but could not distinguish between the initially incorporated and the biologically produced nHA.
This is the first study validating the potential of CS/Gel/nHA scaffolds to support viability and proliferation of DPSCs, and to provide a biomimetic microenvironment favoring odontogenic differentiation and in vitro biomineralization without the addition of any inductive factors, including dexamethasone and/or growth/morphogenetic factors. These results reveal a promising strategy towards TE of mineralized dental tissues.
Numerous studies have provided evidence that restorative dental procedures are closely linked to several biological and technical complications, as well as adverse reactions, leading to unpredictable therapeutic efficiency [ ]. The latter justifies the rationale for pursuing regeneration approaches for dentin – the bulk of a tooth's structure – as the ultimate goal of modern dental tissue engineering (TE) strategies. In recent years, TE has achieved regeneration of the dentin/pulp complex in preclinical [ , ] and clinical models [ ] or even regeneration of bio-roots [ , ] and fully functional teeth in animal models [ , ].
TE aims to regenerate tissues and organs by using cell and biomaterial-based approaches. Towards this direction, the selection of suitable scaffold materials, providing the three-dimensional microenvironment mimicking the extracellular matrix (ECM), is of crucial importance. Different scaffold materials have been employed for mineralized tissue regeneration applications, including porous bioceramics ( e.g. hydroxyapatite/HA, and β-tricalcium phosphate/β-TCP), natural molecules ( e.g. collagen and chitosan), as well as synthetic polymers ( e.g. polyglycolic acid/ PGA and polylactic acid/ PLA) [ , ]. Each material offers variable chemical and structural characteristics, degradation properties, and versatility in handling; nevertheless, only a few have been used effectively for the targeted regeneration of mineralized dental tissues [ ].
Dentin is a complex structure comprising various tissues with different functions and specificities regarding extracellular matrix (ECM) content, and crystallographic profiles. It includes tubular dentin (orthodentin) that represents the main structure produced during dentinogenesis by meta-mitotic odontoblasts, with cell bodies located outside the predentin/dentin layer at the pulp periphery and cell processes extending inside the dentin tubules. After tooth eruption, response to mild trauma ( e.g. carious decay or abrasion), leads to the production of the reactionary (tertiary) dentin by the odontoblastic or subjacent pre-odontoblastic (Hoel’s) cells, that may appear either as tubular or atubular orthodentin or as less organized bone-like tissue (osteodentin). In case of severe trauma (deep caries, restorative procedures etc. ) leading to disruption of the odontoblastic layer, a less mineralized, bone-like tissue called reparative dentin (osteodentin), is produced as a result of the activity of pulp stem/progenitor cells [ ]. Dentin primarily consists of a mineral phase (70 wt %), an organic phase (20 wt %), and water (10 wt %). The organic phase primarily comprises collagen I (90% in intertubular dentin), and other non-collagenous proteins (NCP), including the Small Integrin-Binding Ligand N-linked Glycoproteins (SIBLINGs), such as DSPP, DMP-1, BSP, OPN and MEPE [ ], the Small Leucine-rich proteoglycans (SLRPS), such as biglycan, and decorin and fibromodulin, non-phosphorylated proteins, such as osteocalcin, growth factors, enzymes, phospholipids, proteolipids etc. Despite some similarities in structure, dentin – in contrast to bone – is not vascularized and has little or no remodelling capacity, providing an excellent model to study biomineralization processes of skeletal tissues.
Natural blends combining chitosan (CS) and gelatin (Gel) have been applied as promising scaffold materials for mineralized tissue (primarily bone) regeneration [ ]. The use of CS as a bioactive compound is desirable due to its high biocompatibility, efficient biodegradability by enzymes into oligosaccharides that are rapidly resorbed, and the “Generally Recognized as Safe” (GRAS) status assigned by the US Food and Drug Administration [ ]. CS is capable of forming complexes with inorganic ceramics like hydroxyapatite (HA), as well as with organic compounds, like extracellular matrix (ECM) proteins, growth factors, or other biomaterials [ ]. In specific, blending CS with Gel is an important pathway to improve its biological and mechanical properties. Gelatin is a protein derived from collagen and contains the Arg–Gly–Asp (RGD) sequence found in the natural ECM, which is promoting initial integrin-mediated cell attachment, a crucial step to facilitate cell spreading and proliferation within the biomimetic matrix [ ]. Gelatin has been widely used in the biomedical field because of its merits, including biological origin, biodegradability, hydrogel properties, and commercial availability at comparatively low cost [ , ].
The incorporation of HA into biomimetic porous CS/Gel blends has also been found to enhance the mechanical properties, promote initial cell adhesion, and support the long-term cell growth [ ]. Maintaining higher cell proliferation and further inducing cell osteogenic differentiation were some of the assets of the incorporation of HA in the form of nanoparticles (nHA) inside CS/Gel scaffolds [ ]. HA is bioactive, osteoconductive, non-toxic, and non-immunogenic and its crystal structure is similar to that of bone mineral [ ]. Recently, HA nanoparticles (nHA) were used in CS/Gel scaffolds to increase biocompatibility and bioactivity for bone tissue engineering [ ]. Although several reports on the chemical and biological properties of CS/Gel/nHA blends demonstrate the increasing interest of these biopolymers in bone tissue regeneration, it is still not known whether they could be used as suitable scaffold materials to stimulate the differentiation of dental pulp stem cells (DPSCs) into odontoblastic lineages to induce mineralized dental tissue regeneration.
DPSCs represent a promising source of oral Mesenchymal Stem Cells (MSCs), with great importance in regenerative dentistry [ ]. These cells have been extensively studied for their in vitro multilineage differentiation potential towards osteo/odontogenic, adipogenic, chondrogenic, neurogenic, angiogenic, and myogenic lineages [ ], while in vivo studies confirm their ability to reconstitute functional dentin/pulp-like complexes [ ]. Previous research has shown that an exogenous application of Bone Morphogenic Protein-2 (BMP-2) or Dentin Matrix Protein (DMP-1) in DPSCs seeded in bioceramic scaffolds promoted the biomineralization and the formation of a nanocrystalline HA-rich dentin-like matrix [ ]. However, no reports so far have investigated the possibility of combining DPSCs with biomimetic hybrid CS/Gel/nHA blends towards mineralized dental tissue regeneration.
Based on the above, this study aims to evaluate the potential of hybrid CS/Gel/nHA scaffolds produced by freeze-drying to act as a porous microenvironment promoting attachment, viability, odontogenic differentiation and in vitro biomineralization of DPSCs.
Materials & methods
Synthesis of chitosan/gelatin/nano-hydroxyapatite (CS/Gel/nHA) scaffolds
The scaffolds were prepared by mixing a chitosan/gelatin blend with a nano-hydroxyapatite (nHA) suspension in ultrapure water according to modified protocols based on Ji et al. [ ] and Georgopoulou et al. [ ]. Chitosan, gelatin, nano-hydroxyapatite were purchased from Sigma-Aldrich. First, we dissolved 4% w/v low molecular weight (50−190 kDa) chitosan (CS) in solution of 1% v/v acetic acid and 4% w/v gelatin (Gel) in ultrapure demineralized water, stirred for 2 h at 50 °C and produced a 40%:60% CS:Gel blend. Then, we prepared a 10% w/v rod-like nano-hydroxyapatite suspension in ultrapure water, poured together with the CS/Gel blend at a concentration of 55% w/v, and stirred for 2 h at 50 °C. In this way we produced a composite material with a relative concentration of 55% w/v nHA to the CS/Gel blend. A volume of 400 μL of the above mixture was cast into each well of a 24-well plate, transferred to freeze at −20 °C overnight, and lyophilized for 24 h at −40 °C. Lyophilized scaffolds were crosslinked with 0.1% v/v glutaraldehyde, and lyophilized for another 24 h at −40 °C. Finally, scaffolds were neutralized by rinsing them with 0.1 N NaOH, then with ultrapure water until pH was neutral, and finally with PBS overnight prior their placing in a vented oven for 24 h at 37 °C.
Characterization of the CS/Gel scaffolds
Scanning electron microscopy (SEM)
For the morphological characterization of the scaffolds loaded with cells, Field-emission (FE) scanning electron microscopy (SEM) (FE-SEM, JEOL JSM-7000F) including energy-dispersive X-ray spectroscopy (EDS) analysis was performed. Briefly, the specimens were washed with PBS, fixed with 3% glutaraldehyde in 0.1M sodium cacodylate buffer (pH = 7.4), and dehydrated in an ascending series of ethanol/water (35, 50, 70, 95 and 100 % ethanol). Finally, specimens were dried in a critical point drier (Baltec CPD 030), sputter-coated with a 20 nm thick layer of gold-palladium (Baltec SCD 050) and observed under the SEM at an accelerating voltage of 15–20 kV. The average pore size of the scaffolds was determined by direct geometric analysis on the SEM images. Energy-dispersive X-ray spectroscopy (EDS) analysis (FE-SEM, JEOL JSM-7000F) was performed on a piece of CS/Gel/nHA scaffold which was placed on a carbon tape-coated stub. The sample was then platinum-coated for 1 min at 20 mA. The XRD patterns of the CS/Gel/nHA scaffolds were collected as described in the Subsection 2.7 .
Fourier transform infrared spectroscopic analysis (FTIR)
The Fourier transform infrared spectroscopy analysis (FTIR) of the CS/Gel/nHA scaffolds was recorded using an optical spectrometer (Nicolet 6700, Thermo Electron Corporation) in the region 400–4000 cm −1 . The spectral data were collected, and the numerical values were transferred to the software Origin for graphical representation.
The ability of lyophilized CS/Gel/nHA scaffolds to absorb water was determined by swelling them in phosphate buffer saline (PBS, pH 7.4). The dry samples were weighted and then placed in PBS for 30 min. The excess of water was removed with filter paper, and their weight was measured immediately. The percentage of water uptake was computed by the following formula:
o is the weight of dried scaffolds, while W
1 is the weight of wet scaffolds after 30 min in PBS. The values were expressed as the mean ± standard error (SE; n = 6).
Establishment of DPSC cultures
The enzymatic dissociation method was employed to generate DPSCs cultures from extracted third molars, as previously described [ ]. Each culture was established from one wisdom tooth per donor. This study was carried out in accordance with the recommendations of the Institutional Ethics Committee. The protocol was approved by the same Committee (number 46/20-3-2019). All subjects signed informed consent according to the Declaration of Helsinki. Disinfection with iodine was applied to the third molars, and a cut was made around the cementum-enamel junction to expose the pulp chamber. Then, the pulp tissue was retrieved, thoroughly minced, and digested in a solution of 3 mg/mL collagenase I and 4 mg/mL dispase II (Invitrogen, Karlsruhe, Germany) for 45 min at 37 °C. The cells were cultured in α-MEM (Minimum Essential Media) medium (Invitrogen), supplemented with 15% fetal bovine serum (FBS, Invitrogen), 100 mM L-ascorbic acid phosphate (Sigma-Aldrich, Steinheim, Germany) and antibiotics/antimycotics (=Complete Culture Medium; CCM) before been incubated at 37 °C in 5% CO 2 . All experiments were conducted with DPSCs, cultured in passage 2 to 6, collected from a minimum of three donors.
Immunophenotypic characterization of DPSC cultures
DPSCs were analyzed by flow cytometry for mesenchymal (STRO-1, CD146, CD90/Thy-1, CD73), endothelial (CD105/endoglin), embryonic (SSEA-4) and hematopoietic (CD34, CD45) stem cell (SC) markers, as previously described [ ]. Single-cell suspensions, obtained by culture trypsinization, were labeled for surface markers with the following fluorochrome-conjugated antibodies: STRO-1-FITC (fluorescein isothiocyanate), CD146-PE (phycoerythrin), CD90-FITC, CD73-PE, CD105-FITC, SSEA-4-FITC, CD34-APC (allophycocyanin), CD45-PE (all from BioLegend, Fell, Germany). After staining and washing of the cells with a FACS buffer (PBS + 1% BSA + 0.1% NaN 3 ), the labeled cells were analyzed using a Guava®easyCyte 8 H T Benchtop Flow Cytometer (Merck Millipore, Billerica, Massachusetts, U.S.A.). A total of 50,000 events/sample were acquired. Data were analyzed using the software Summit 5.1 (version 5.1 for Windows, Beckman Coulter, Inc., Krefeld, Germany).
Seeding of DPSCs into the CS/Gel/nHA scaffolds and analysis of cell morphology and viability
The synthesized CS/Gel/nHA scaffolds were placed in 24-well plates and exposed to UV-A light irradiation for 5 min for complete disinfection. Then, the scaffolds were incubated for 24 h with CCM at 37 °C and 5% CO 2 , aiming at their initial wetting with the serum-containing medium while achieving pH stabilization between 7.2 and 7.4. After reaching the desirable pH, each scaffold was spotted with 200 μL CCM containing 2 × 10 6 DPSCs and incubated for 1 h at 37 °C and 5% CO 2 to allow initial cell attachment. Afterwards, each spotted scaffold containing well was filled with 1 mL CCM. Medium change was performed every other day.
After 1, 3, 7, and 14 days cell viability was evaluated by live/dead staining fluorescent staining, applying calcein AM and ethidium homodimer (EthD-1) fluorescent dyes for living and dead cells respectively, and visualized by confocal microscopy. Meanwhile, the cell viability/proliferation was further assessed using a metabolic-based test (MTT assay) at the same time-points. SEM was employed to evaluate cell morphology within the CS/Gel/nHA scaffolds.
Evaluation of the morphological characteristics of cell-seeded scaffolds by SEM
For SEM observation, the DPSCs/scaffold constructs were washed with PBS and prepared as described in Section 2.2.1 .
Assessment of cell viability/proliferation based on metabolic activity
An MTT [3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay was used to assess cell viability/proliferation inside the CS/Gel/nHA scaffolds. Each well/scaffold was spotted with 5·10 5 cells as described in paragraph 2.5. After 1, 2, 3, 7 and 14 d, MTT (0.5 mg/mL in CCM) was added to each cell/scaffold construct and incubated for 4 h at 37 °C and 5% CO 2 . The MTT-insoluble formazan was then dissolved by DMSO that was applied for 4 h to the constructs at 37 °C. A wavelength of 545 nm and a reference filter of 630 nm were applied to measure the absorbance against blank (DMSO) by a microplate reader (Epock, Biotek, Biotek Instruments, Inc., Vermont, USA). Scaffolds without cells were treated and incubated under the same conditions to be used as control. Thereby, the optical density (OD) values of the control scaffolds were subtracted from values obtained by the corresponding cell-seeded CS/Gel/nHA scaffolds.
Evaluation of the ratio of living to dead cell by confocal microscopy
The cell-seeded CS/Gel/nHA scaffolds were double-stained calcein AM and EthD-1 fluorescent dyes. Stained cell/scaffold constructs were observed under a confocal microscope (Leica Microsystems, Wetzlar, Germany). Approximately 20 depth-dependent serial sections were obtained, and the projection (z-stacked) images were composed. The quantification of the percentage of living and dead cells was performed with the Image J color pixel counter plugin.
Evaluation of the odontogenic differentiation potential of DPSC-seeded CS/Gel/nHA scaffolds by real-time reverse-transcription polymerase chain reaction analysis (real-time PCR)
CS/Gel/nHA scaffolds were spotted with DPSCs as described in Section 2.5 . Total RNA isolation from each construct was performed after 1, 3, 7, and 14 d. Briefly, CCM was removed from each well containing the DPSC-seeded scaffolds, and after two washes with PBS, 600 μL collagenase I (4 mg/mL, Invitrogen) were added. The plates were incubated for 45 min in an incubator at 37 °C and 5% CO 2 . Then, the supernatant of each well was collected and placed into a falcon tube. Subsequently, 400 μL of a 0.25% trypsin/mM EDTA solution (Invitrogen) were added in each well containing the DPSC-seeded scaffolds and incubated for another 7 min at 37 °C and 5% CO 2 . Afterwards, the supernatant of each well was collected and pooled with the supernatant of the same sample collected after collagenase I treatment at the first step. The collected supernatants after both enzymatic treatments were centrifuged at 150 g for 5 min, followed by washing with PBS, and another centrifugation step under the same conditions. Finally, the supernatant was removed, and the remaining cell pellet was exposed to 700 μL RA1 Lysis Buffer, followed by RNA isolation by the Nucleospin RNA isolation kit (Macherey Nagel, Düren, Germany) according to the manufacturer’s instructions. All collected RNA samples were stored at −80 °C for the qPCR analysis.
Reverse transcription was conducted using a superscript first-strand synthesis kit (Takara, Takara Bio USA, Inc., Mountain View, CA), according to manufacturer’s instructions. Reactions were performed using SYBR-Select PCR Master Mix (Applied Biosystems, Foster City, CA) in a Step One Plus thermal cycler (Applied Biosystems). The reactions’ protocol started with two initial incubation steps at 50 °C for 2 min and at 95 °C for 2 min and were followed by 40 cycles of PCR, comprising denaturation for 15 s at 95 °C, and annealing/extension for 1 min at 60 °C. The Primer-Blast software from the NCBI nucleotide sequence database ( http://www.ncbi.nlm.nih.gov/BLAST ) was used to design primers for the following genes: DSPP, BMP-2, RUNX2, ALP, and Osterix ( Table 1 ). The results were adjusted by amplification efficiency (LinRegPCR) and normalized against two housekeeping genes (succinate dehydrogenase complex, subunit A, flavoprotein-SDH-A; beta-2-microglobulin-B2M).
|Gene symbol||Forward (5’-3’)||Reverse (5’-3’)|| Amplicon
Mineralized tissue characterization inside the CS/Gel/nHA scaffolds
CS/Gel/nHA scaffolds were spotted with DPSCs as described in Section 2.5 and cultured with CCM, supplemented with 1.8 mM KH 2 PO 4 (monopotassium phosphate) and 5 mM beta-glycerophosphate, to provide the phosphate sources necessary for the in vitro biomineralization. Medium change was performed every other day for a total of 48 d. Cell-free scaffolds were used as controls for this assay to assess chemical calcium phosphate (CaP) precipitation due to the addition of phosphates to the culture medium. After 14, 28, and 48 d, samples were processed for SEM examination combined with energy-dispersive X-ray spectroscopy (SEM-EDS), as described in Section 2.1 , to visualize the porous structure of DPSC-seeded scaffolds and to provide the elemental composition of the formed calcium phosphate phases, respectively. For the microscopic (SEM) and spectroscopic (EDS) analyses of the samples of the in vitro biomineralization experiments, an ESEM Quanta 400 FEG instrument (FEI), equipped with EDS spectroscopy (EDS; Genesis 4000, SUTW-Si (Li) detector) operating in a high vacuum with gold/palladium-sputtered samples were used. In addition, XRD measurements were carried out to crystallographically investigate the inorganic part of the DPSC-seeded scaffolds (biomineralization process). The porous samples, ground into a fine powder, were deposited on a silicon single crystal sample holder to minimize scattering and investigated in Bragg-Brentano geometry with a Bruker D8 Advance instrument (Cu K α radiation, 1.54 Å, 40 kV and 40 mA). All samples were rotated and measured from 5−90° 2 θ with a step size of 0.01° and a counting time of 0.6 s at each step. The phase analysis was performed with the software Diffrac Suite EVA V1.2 from Bruker using the patterns of hydroxyapatite HAP (#09-0432) from the ICDD database as references. After the instrumental characterization of the diffractometer by measuring a standard powder sample LaB 6 from NIST (SRM 660b; a (LaB 6 ) =4.15689 Å), Rietveld refinement with the program package TOPAS 4.2 from Bruker enabled the determination of the lattice parameters a and c (hexagonal crystal system of HAP) and crystallite size D of the formed calcium phosphate particles.
All assays were performed in three independent biological experiments (n = 3) with two to four technical replicates each. The experimental data were analyzed using one-way ANOVA, followed by Tukey’s post-hoc test for multiple comparisons between groups. Normal distribution was confirmed by D'Agostino & Pearson normality tests. Data were expressed as means ± standard deviation (SD) or ± standard error (SE). For the above-mentioned analyses the GraphPad Prism 6.0 (GraphPad Software, Inc.; La Jolla, CA 92037 USA) was used (* p < 0.05; ** p < 0.01; n.s. denotes statistically non-significant).