Home
Search
Collections
Journals
About
Contact us
My IOPscience
Incorporation of sol–gel bioactive glass into PLGA improves mechanical properties and bioactivity of composite scaffolds and results in their osteoinductive properties
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Biomed. Mater. 9 065001 (http://iopscience.iop.org/1748-605X/9/6/065001) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 129.82.28.124 This content was downloaded on 28/10/2014 at 11:15
Please note that terms and conditions apply.
Biomedical Materials Biomed. Mater. 9 (2014) 065001 (15pp)
doi:10.1088/1748-6041/9/6/065001
Incorporation of sol–gel bioactive glass into PLGA improves mechanical properties and bioactivity of composite scaffolds and results in their osteoinductive properties J Filipowska1, J Pawlik2, K Cholewa-Kowalska2, G Tylko1, E Pamula3, L Niedzwiedzki4, M Szuta5, M Laczka2 and A M Osyczka1 1
Department of Cell Biology and Imaging, Faculty of Biology and Earth Sciences, Jagiellonian University, Gronostajowa 9, 30-387 Krakow, Poland 2 Department of Glass Technology and Amorphous Coatings, Faculty of Materials Science and Ceramics, AGH-University of Science and Technology, Mickiewicza Ave. 30, 30-059 Krakow, Poland 3 Department of Biomaterials, Faculty of Materials Science and Ceramics, AGH-University of Science and Technology, Mickiewicza Ave. 30, 30-059 Krakow, Poland 4 Department of Orthopedics and Physiotherapy, Institute of Physiotherapy, Faculty of Health Care, School of Medicine, Jagiellonian University, Michalowskiego 12, 31-126 Krakow, Poland 5 Department of Maxillofacial Surgery, Faculty of Medicine, School of Medicine, Jagiellonian University, The Rydygier’s Hospital, os. Złotej Jesieni 1, 31-826, Krakow, Poland E-mail: [email protected] Received 11 February 2014, revised 29 July 2014 Accepted for publication 31 July 2014 Published 20 October 2014 Abstract
In this study, 3D porous bioactive composite scaffolds were produced and evaluated for their physico-chemical and biological properties. Polymer poly-L-lactide-co-glycolide (PLGA) matrix scaffolds were modified with sol–gel-derived bioactive glasses (SBGs) of CaO–SiO2– P2O5 systems. We hypothesized that SBG incorporation into PLGA matrix would improve the chemical and biological activity of composite materials as well as their mechanical properties. We applied two bioactive glasses, designated as S2 or A2, differing in the content of SiO2 and CaO (i.e. 80 mol% SiO2, 16 mol% CaO for S2 and 40 mol% SiO2, 52 mol% CaO for A2). The composites were characterized for their porosity, bioactivity, microstructure and mechanical properties. The osteoinductive properties of these composites were evaluated in human bone marrow stromal cell (hBMSC) cultures grown in either standard growth medium or treated with recombinant human bone morphogenetic protein-2 (rhBMP-2) or dexamethasone (Dex). After incubation in simulated body fluid, calcium phosphate precipitates formed inside the pores of both A2-PLGA and S2-PLGA scaffolds. The compressive strength of the latter was increased slightly compared to PLGA. Both composites promoted superior hBMSC attachment to the material surface and stimulated the expression of several osteogenic markers in hBMSC compared to cells grown on unmodified PLGA. There were also marked differences in the response of hBMSC to composite scaffolds, depending on chemical compositions of the scaffolds and culture treatments. Compared to silica-rich S2-PLGA, hBMSC grown on calcium-rich A2-PLGA were overall less responsive to rhBMP-2 or Dex and the osteoinductive properties of these A2-PLGA scaffolds seemed partially dependent on their ability to induce BMP signaling in untreated hBMSC. Thus, beyond the ability of currently studied composites to enhance hBMSC osteogenesis, it may become possible to modulate the osteogenic response of hBMSC, depending on the chemistry of SBGs incorporated into polymer matrix.
1748-6041/14/065001+15$33.00
1
© 2014 IOP Publishing Ltd Printed in the UK
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Keywords: sol–gel bioactive glass, PLGA, composites, bone marrow-derived mesenchymal stem cells (hBMSC), in vitro osteogenesis, 3D culture, osteoinduction (Some figures may appear in colour only in the online journal)
1. Introduction
sites [21]. Similar capabilities are expected from osteoinductive biomaterials [22]. Even though hBMSC cultures may poorly respond to BMPs [23], these cells are the most common source of osteoprogenitors for bone tissue engineering [24]. If the engineered construct is combined with cells in vitro, BMSCs must be prompted with growth factors and/or hormones to initiate osteogenesis. Certain properties of the construct, such as appropriate stiffness [25], high calcium content [26] or the presence of HA [27, 28] may also direct BMSCs onto the osteogenic path. Considering the above, we have assumed that the bone-targeting scaffolds should induce hBMSC in vitro osteogenesis on their own and/or enhance osteogenic response of hBMSCs to the osteogenic supplements, especially rhBMP-2. Previously, we reported the development of composite 2D surfaces made of PLGA and SBG that supported BMP-stimulated osteogenesis of hBMSCs [29]. We have now fabricated similar composites in the form of scaffolds and subjected them to hBMSC cultures with and without rhBMP-2 or dexamethasone (Dex). The former has been identified as a relatively weak inducer of hBMSC osteogenesis in vitro [30]. In contrast, Dex is a well-recognized potent inducer of both rodent and human BMSC osteogenesis in vitro [31, 32] and served in our studies as a positive control for comparison of rhBMP-2-induced effects. Moreover, we have examined the bioactivity of our composite scaffolds by studying their internal and external microstructure after incubation in the SBF solution. We have also evaluated SBG–PLGA scaffold mechanical properties, examining changes in compressive strength and compressive modulus after incubation of these 3D materials in the SBF solution. Beyond the high bioactivity and enhanced mechanical properties of the experimental composites, we show that the composite scaffolds direct hBMSCs onto an osteogenic path without additional stimuli and enhance BMP-mediated responses.
One of the most challenging tasks in the modern field of biomedical engineering is to tailor material properties to specific medical needs and provide a device with ‘healing’ features; ideally, such an implant would heal or regenerate the tissue without any follow-up pharmacological and surgical treatments. Composite materials address some of these challenges, as they combine and tailor properties of particular composite components [1, 2]. The primary goal of bone tissue engineering is to design osteoconductive or, ideally, osteoinductive materials. The former will support bone tissue formation and remodeling by providing a proper surface or scaffold for osteoblastic cells to attach, proliferate and differentiate [3]. The latter indicates a ‘smart biomaterial’ that will stimulate bone tissue to regenerate. Osteogenic scaffolds should mimic cancellous bone morphology and function in order to optimize integration with surrounding tissue. That is why the requirements of these materials include a controllable degradation rate that is synchronized with bone tissue regeneration [4, 5], mechanical properties that promote regeneration of a particular skeletal site [6, 7] and biological activity that supports or promotes osteogenesis [8, 9]. Several composite materials have been reported to be suitable for bone tissue engineering, but only a few have been defined as osteoinductive [10–14]. In fact, there is a limited number of material and biological components that display osteoinductive properties, such as natural or synthetic hydroxyapatite (HA), demineralized bone matrix or natural or recombinant bone morphogenetic proteins [15]. Although poly-L-lactide-co-glycolide (PLGA) has been approved by the FDA for a clinical use, its application for bone tissue engineering is limited by its poor osteoconductivity and suboptimal mechanical properties for application in load-bearing sites [16]. It is currently very common to combine PLGA with other types of materials, e.g. bioactive glasses, to obtain composites exhibiting a balanced set of properties. Thus, we have tested the hypothesis that the sol–gel-derived bioactive glasses (SBGs) we previously developed and described [17– 19] may serve as osteoinductive components of PLGA-based composite scaffolds for human bone-marrow-derived mesenchymal stem cell (hBMSC)-based bone cell therapies. The phenomenon of osteoinductivity can rely on (1) material factors such as material bioactivity, porosity, phase composition and crystallinity and (2) biological factors such as osteoinductive growth factors and hormones [20]. The latter also depend on the implantation site and the species-specific biological responses. To date, bone morphogenetic proteins (BMPs) are the best known osteoinductive molecules found in bone. They are capable of inducing bone formation either in natural or recombinant forms when implanted in heterotopic
2. Materials and methods 2.1. Materials
PLGA was synthesized via a ring opening process in the presence of low toxicity zirconium acetyloacetonate as a copolymerization initiator [33]. The molar ratio of L-lactide to glycolide in the copolymer was 85 : 15, and the molecular masses of PLGA were Mn=80 kDa and Mw=152 kDa. Bioactive glasses (A2 and S2) of the composition of A2: 40% SiO2–52% CaO–6%P2O5 [mol%] and S2: 80% SiO2–16% CaO–4%P2O5 [mol%] were produced using the sol–gel method as described previously [17]. Composite scaffold fabrication has been previously described in detail [29]. Briefly, PLGA scaffolds modified 2
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Mononuclear cells were seeded to T-75 flasks (BD Falcon) in a standard cell growth medium composed of alpha-MEM, 10% mesenchymal stem cell qualified fetal bovine serum (MSCqualified FBS) and 1% antibiotics (penicillin and streptomycin). Media were initially changed at day 7 of the primary culture, then every 2–3 days. Cells were grown in primary culture until they reached 80–90% confluence. hBMSCs were then detached from the bottom of tissue culture flasks with 0.25% Trypsin-EDTA and either loaded onto the scaffolds or further expanded in T-75 flasks. All experimental cultures were established with hBMSCs at passages 1–6.
with A2 or S2 21 vol% SBGs (designated as A2-PLGA and S2-PLGA) were fabricated by the addition of porogen (NaCl) into the copolymer and bioactive glass suspension in methylene chloride. The mixed slurry was then packed into cylindrical vials and dried in air (24 h) and in vacuum (48 h). The vials with NaCl/SBG–PLGA mixture were cut into slices of approximately 3 mm thickness and extensively washed in ultra-highquality (UHQ) water. Finally, the samples were dried in air and vacuum for at least 24 h. PLGA reference scaffolds were prepared using a similar protocol to serve as controls. 2.2. Materials evaluation
2.3.2. Cell seeding and cultures on the experimental scaffolds. Before the cell seeding procedure, SBG–PLGA
2.2.1. Porosity. The porosity (P) of scaffolds was calculated
according to the equation, where ρs, the apparent density of the scaffold, was calculated by dividing the scaffold weight by the scaffold volume, and ρp is the density of PLGA (1.3 g cm−3). The porosity of the 3D scaffolds was verified by the mercuric porosimetry (Poremaster 60 Quantachrome). The reported porosity of scaffolds represents the average of three samples of each type analyzed.
experimental scaffolds were sterilized in 70% ethanol (water solution) for 48 h, and then washed three times with PBS to remove ethanol traces from the scaffold pores. HBMSCs were suspended at a density of (1.0–1.4) × 105 cells per 120–150 µl of standard cell growth medium supplemented with 40 µl of fibrinogen (stock solution of 1.75 mg ml−1 PBS) and 1.6 µl of thrombin (stock solution of 25 U ml−1 in PBS) to initiate clot formations. Sterilized scaffolds (12 mm in diameter and 2–4 mm height) were placed into separate wells of 24-well plates, and each scaffold was loaded with a 150 µl aliquot of hBMSC suspension. Scaffolds were then incubated for 40 min at 37 °C in a humidified 5% CO2 atmosphere to allow the fibrinogen polymerization. Then, 1.5 ml of standard cell growth medium was added per well. After an initial 24 h culturing, the scaffolds were placed in a new 24-well plate and fresh growth medium was added. Cells in scaffolds were either grown in a standard growth medium or an osteogenic medium that consisted of a growth medium supplemented with 50 µg ml−1 ascorbate-2-phosphate and either 100 ng ml−1 rhBMP-2 or 10-7M dexamethasone (Dex). Following day 1, media were changed every 2–3 days.
2.2.2. Bioactivity. The bioactivity of the composites was
assessed by an in vitro simulated body fluid (SBF) test according to the method described by Kokubo [34]. SBF was prepared by dissolving reagent-grade chemicals in de-ionized water and buffered to pH 7.4 at 37 °C with TRIS tris(hydroxymethyl aminomethane) and HCl. The samples were immersed in SBF and incubated at 37 °C for 14 and 21 days. The samples were then washed in UHQ water and air and vacuum dried to a constant weight. 2.2.3. Microstructure. The internal (cross-section) and external morphology of the scaffolds before and after SBF incubation, as well as the chemical composition of the 3D scaffolds, were examined using SEM/EDAX analyses (Nova 200, NanoSEM, FEI, USA).
2.4. Cell culture evaluation 2.4.1. Cell viability assay. Cell viability was examined with the CellTiter96Aqueous One Solution Cell Proliferation Assay (MTS, Promega). After 7 or 21 days of culture, scaffolds were washed 3 times with PBS, transferred to a fresh 24-well plate and each scaffold was covered with 0.5 ml solution of 10% MTS reagent in PBS. The colorimetric reaction was developed in the dark at 37 °C in a humidified 5% CO2 atmosphere. Afterward, the MTS solutions were collected from the 24-well plate, 200 µl aliquots were transferred to separate wells in a 96-well plate and the absorbance was measured at 490 nm in a plate reader (ELx 808IU, UltraMicroplate Reader, Biokom). The absorbance units obtained from the MTS test were converted to the live cell number based on a standard curve prepared in our laboratory using hBMSCs seeded at different densities and cultured overnight. After 24 h, the MTS test was performed. This reflects the changes in the MTS absorbance units versus the live cell number.
2.2.4. Mechanical properties. The compressive strength and
compressive modulus values of as-prepared scaffolds and scaffolds incubated for 1 month in SBF were determined using a universal testing machine (Zwick Z2.5, Germany). Samples (N=3 for each type) were compressed with a constant deformation rate of 1 mm min-1. The average values represent the compressive strength at 10% strain and compressive modulus of scaffolds. 2.3. Cell culture 2.3.1. hBMSC isolation and culture expansion. Unless stated
otherwise, all cell culture reagents were purchased from Life Technologies. Adult human bone marrow stromal cells (hBMSCs) were harvested from the iliac crest of adult patients (32–78 years old, both genders) according to the approved Institutional Review Board protocol (KBET/17/L/2007). The mononuclear cell fraction was isolated using Ficoll-Paque (GE Healthcare), as described in the manufacturer’s protocol.
2.4.2. Scanning electron microscopy (SEM) and confocal microscopy analyses. For SEM analyses, hBMSCs were 3
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Table 1. Primer sequences and product lengths.
Primer sequence Gene name
Forward (5’–3’)
Reverse (5’–3’)
Length
Tata box-binding protein Osteopontin Bone morphogenetic protein-2 Osteocalcin Osteoprotegrin Collagen I Bone sialoprotein
GGAGCTGTGATGTGAAGTTTCCTA TGGAAAGCGAGGAGTTGAATG TGCTAGTAACTTTTGGCCATGATG
CCAGGAAATAACTCTGGCTCATAAC CATCCAGCTGACTCGTTTCATAA GCGTTTCCGCTCTTTGTGTT
91bp 117bp 86bp
AAGAGACCCAGGCGCTACCT GTCAAGCAGGAGTGCAATCG GTCTAGACATGTTCAGCTTTGTGGA AACGAAGAAAGCGAAGCAGAA
AACTCGTCACAGTCCGGATTG TAGCGCCCTTCCTTGCATT CTTGGTCTCGTCACAGATCACGTCAT TCTGCCTCTGTGCTGTTGGT
110bp 59bp 245bp 77bp
were washed three times with PBS, then cut into halves and placed in the Eppendorf tubes. Any PBS remaining inside the scaffolds was aspirated and 300 µl aliquots of cell lysis buffer supplemented with protease inhibitors were added to individual scaffolds. Subsequently, the proteins were extracted from the 3D cultures by repeated vortexing and freeze/thaw cycles. Protein lysates were clarified from cell debris by centrifugation. Clarified protein lysates were stored at −80 °C until the western blot analyses. Protein concentrations were determined with a Micro BCA protein assay kit (Thermo Scientific). Equal amounts of protein lysates from cultured hBMSCs on individual scaffolds were separated on NuPAGE 4–12% bis-Tris gels under reducing conditions and then transferred to PVDF membranes. Membranes were probed overnight with primary anti-human Smad1 or anti-human phospho-Smad 1/5/8 antibodies (Cell Signaling Technology) and then with horseradish peroxidase-linked rabbit anti-human secondary antibodies (GE Healthcare). The peroxidase-based signal was detected using Western Lightning Chemiluminescence Reagent Plus (GE Healthcare) and captured on Hyperfilm ECL chemiluminescent films (Perkin-Elmer).
cultured for 9 days on experimental scaffolds in standard growth medium. Then, scaffolds were washed 3 times with PBS, treated with 2.5% glutaraldehyde solution in cacodylate buffer for 2 h, washed in cacodylate buffer and fixed for 1 h in 0.25% OsO4 in cacodylate buffer. Samples were dehydrated in increasing concentrations of ethanol. Finally, the samples were dried in the CO2 critical point dryer, attached to microscope adhesive holders and covered with a thin gold layer. The SEM images were taken with the JEOL JSM5410 scanning electron microscope (JEOL, Tokyo, Japan). For confocal microscopy imaging, hBMSCs were cultured for 7 days on experimental scaffolds in a growth medium supplemented with ascorbate-2-phosphate and dexamethasone. Then cells were washed 3 times with PBS and fixed in 4–8% formalin solution in PBS for 20 min. Cell cytoplasm was stained with 5% eosin (Eosin Y, ethanol solution, Thermo Scientific) and cell nuclei with DAPI diluted 1 : 1000 in PBS. The confocal images were taken using the Zeiss Axiovert 200M microscope combined with LSM510 Meta Confocal head (Carl Zeiss, MicroImaging GmbH, Jena, Germany), using the Zeiss 20 × /0.8 Plan-Apochromat objective.
2.4.4. Western blot analyses. Whole cell extracts were obtained
2.4.5. Quantitative RT-PCR. Gene expression studies were performed at culture days 2, 10 and 20 on experimental scaffolds. Briefly, the total RNA was extracted from cultures on individual experimental scaffolds using the SPEEDTOOLS TOTAL RNA EXTRACTION KIT (Biotools) according to the manufacturer’s protocol. The aliquots of 0.5 µg RNA were reverse transcribed to cDNA (SuperScript III First Strand Synthesis System, Life Technologies). PCR reaction mixtures (15 µl total volume) contained 1 µl cDNA, gene-specific primers, SYBR®Green I, AmpliTaq Gold® DNA Polymerase and the reaction buffer as recommended by the manufacturer. Primer sequences are shown in table 1. Taq-Man Gene Expression Assays were used for the analyses of Runx2 (Hs00231692_m1 RUNX2) and Tata box-binding protein (Hs99999910_m1 TBP) mRNA levels. Each reaction mixture (10 µl total volume) contained 1 µl cDNA, specific TaqMan probe and TaqMan Gene Expression Master Mix as recommended by the manufacturer. All PCR reactions were run in StepOnePlus Real-Time PCR apparatus (Applied Biosystems) and the ddCt method was used for the analyses of gene expression levels. Gene expression levels of the housekeeping gene, Tata box-binding protein, were used in each reaction to normalize the values.
as described by Osyczka and Leboy [30]. Briefly, hBMSCs were examined two days post-seeding and subsequent culture in the experimental scaffolds. Scaffolds containing hBMSCs
2.4.6. Evaluation of collagen production and mineralization of extracellular matrix. For the evaluation of total collagen protein
2.4.3. ALP activity measurements. ALP activity was mea-
sured kinetically at day 7 of the hBMSC cultures on experimental scaffolds, as originally described by Osyczka and Leboy [30]. Briefly, cell viability was determined using the MTS assay described above, and scaffolds were washed extensively with PBS prior to performing the ALP activity measurements. Scaffolds were then cut into halves, placed into 1.5 ml centrifuge tubes and covered with 200 µl of cell digestion buffer. Samples were extensively vortexed and centrifuged, followed by a freeze/thaw cycle. Finally, the thawed scaffolds were incubated at 37 °C for 30 min and again extensively vortexed and centrifuged to remove cell debris. Clarified protein extracts were used for reactions with the alkaline phosphatase substrate, p-nitrophenol phosphate (pNp) (Sigma Aldrich). The changes in the absorbance at 405 nm were measured for 6 min at 1 min intervals. The total ALP activity was expressed as nmol pNp/min/total volume of the protein extract. The ALP activity was then normalized to the number of viable cells estimated from the MTS assay.
4
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Figure 1. SEM and EDAX analyses of the external and internal surfaces of (a) PLGA, (b) S2-PLGA and (c) A2-PLGA scaffolds. The middle panel shows representative high-magnification SEM images of the scaffold external surface area. Scale bars are indicated at the bottom panel of each image. EDAX results are representative (average) for the studied surface.
levels and mineralization of extracellular matrix (ECM), hBMSCs were cultured for 20 days on the experimental scaffolds in osteogenic conditions and analyzed as described before [25]. Briefly, at culture day 20, hBMSCs were assayed for viability with MTS, and then the scaffolds were extensively washed with PBS and stained for collagen with Sirius Red and for minerals with Alizarin Red S. The dyes were then extracted and measured colorimetrically at 405 nm (AR) or 490 nm (SR).
v.10, StatSoft software. One-way or multivariate ANOVA and post-hoc Tukey’s tests were used to calculate statistically significant differences at p
Search
Collections
Journals
About
Contact us
My IOPscience
Incorporation of sol–gel bioactive glass into PLGA improves mechanical properties and bioactivity of composite scaffolds and results in their osteoinductive properties
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Biomed. Mater. 9 065001 (http://iopscience.iop.org/1748-605X/9/6/065001) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 129.82.28.124 This content was downloaded on 28/10/2014 at 11:15
Please note that terms and conditions apply.
Biomedical Materials Biomed. Mater. 9 (2014) 065001 (15pp)
doi:10.1088/1748-6041/9/6/065001
Incorporation of sol–gel bioactive glass into PLGA improves mechanical properties and bioactivity of composite scaffolds and results in their osteoinductive properties J Filipowska1, J Pawlik2, K Cholewa-Kowalska2, G Tylko1, E Pamula3, L Niedzwiedzki4, M Szuta5, M Laczka2 and A M Osyczka1 1
Department of Cell Biology and Imaging, Faculty of Biology and Earth Sciences, Jagiellonian University, Gronostajowa 9, 30-387 Krakow, Poland 2 Department of Glass Technology and Amorphous Coatings, Faculty of Materials Science and Ceramics, AGH-University of Science and Technology, Mickiewicza Ave. 30, 30-059 Krakow, Poland 3 Department of Biomaterials, Faculty of Materials Science and Ceramics, AGH-University of Science and Technology, Mickiewicza Ave. 30, 30-059 Krakow, Poland 4 Department of Orthopedics and Physiotherapy, Institute of Physiotherapy, Faculty of Health Care, School of Medicine, Jagiellonian University, Michalowskiego 12, 31-126 Krakow, Poland 5 Department of Maxillofacial Surgery, Faculty of Medicine, School of Medicine, Jagiellonian University, The Rydygier’s Hospital, os. Złotej Jesieni 1, 31-826, Krakow, Poland E-mail: [email protected] Received 11 February 2014, revised 29 July 2014 Accepted for publication 31 July 2014 Published 20 October 2014 Abstract
In this study, 3D porous bioactive composite scaffolds were produced and evaluated for their physico-chemical and biological properties. Polymer poly-L-lactide-co-glycolide (PLGA) matrix scaffolds were modified with sol–gel-derived bioactive glasses (SBGs) of CaO–SiO2– P2O5 systems. We hypothesized that SBG incorporation into PLGA matrix would improve the chemical and biological activity of composite materials as well as their mechanical properties. We applied two bioactive glasses, designated as S2 or A2, differing in the content of SiO2 and CaO (i.e. 80 mol% SiO2, 16 mol% CaO for S2 and 40 mol% SiO2, 52 mol% CaO for A2). The composites were characterized for their porosity, bioactivity, microstructure and mechanical properties. The osteoinductive properties of these composites were evaluated in human bone marrow stromal cell (hBMSC) cultures grown in either standard growth medium or treated with recombinant human bone morphogenetic protein-2 (rhBMP-2) or dexamethasone (Dex). After incubation in simulated body fluid, calcium phosphate precipitates formed inside the pores of both A2-PLGA and S2-PLGA scaffolds. The compressive strength of the latter was increased slightly compared to PLGA. Both composites promoted superior hBMSC attachment to the material surface and stimulated the expression of several osteogenic markers in hBMSC compared to cells grown on unmodified PLGA. There were also marked differences in the response of hBMSC to composite scaffolds, depending on chemical compositions of the scaffolds and culture treatments. Compared to silica-rich S2-PLGA, hBMSC grown on calcium-rich A2-PLGA were overall less responsive to rhBMP-2 or Dex and the osteoinductive properties of these A2-PLGA scaffolds seemed partially dependent on their ability to induce BMP signaling in untreated hBMSC. Thus, beyond the ability of currently studied composites to enhance hBMSC osteogenesis, it may become possible to modulate the osteogenic response of hBMSC, depending on the chemistry of SBGs incorporated into polymer matrix.
1748-6041/14/065001+15$33.00
1
© 2014 IOP Publishing Ltd Printed in the UK
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Keywords: sol–gel bioactive glass, PLGA, composites, bone marrow-derived mesenchymal stem cells (hBMSC), in vitro osteogenesis, 3D culture, osteoinduction (Some figures may appear in colour only in the online journal)
1. Introduction
sites [21]. Similar capabilities are expected from osteoinductive biomaterials [22]. Even though hBMSC cultures may poorly respond to BMPs [23], these cells are the most common source of osteoprogenitors for bone tissue engineering [24]. If the engineered construct is combined with cells in vitro, BMSCs must be prompted with growth factors and/or hormones to initiate osteogenesis. Certain properties of the construct, such as appropriate stiffness [25], high calcium content [26] or the presence of HA [27, 28] may also direct BMSCs onto the osteogenic path. Considering the above, we have assumed that the bone-targeting scaffolds should induce hBMSC in vitro osteogenesis on their own and/or enhance osteogenic response of hBMSCs to the osteogenic supplements, especially rhBMP-2. Previously, we reported the development of composite 2D surfaces made of PLGA and SBG that supported BMP-stimulated osteogenesis of hBMSCs [29]. We have now fabricated similar composites in the form of scaffolds and subjected them to hBMSC cultures with and without rhBMP-2 or dexamethasone (Dex). The former has been identified as a relatively weak inducer of hBMSC osteogenesis in vitro [30]. In contrast, Dex is a well-recognized potent inducer of both rodent and human BMSC osteogenesis in vitro [31, 32] and served in our studies as a positive control for comparison of rhBMP-2-induced effects. Moreover, we have examined the bioactivity of our composite scaffolds by studying their internal and external microstructure after incubation in the SBF solution. We have also evaluated SBG–PLGA scaffold mechanical properties, examining changes in compressive strength and compressive modulus after incubation of these 3D materials in the SBF solution. Beyond the high bioactivity and enhanced mechanical properties of the experimental composites, we show that the composite scaffolds direct hBMSCs onto an osteogenic path without additional stimuli and enhance BMP-mediated responses.
One of the most challenging tasks in the modern field of biomedical engineering is to tailor material properties to specific medical needs and provide a device with ‘healing’ features; ideally, such an implant would heal or regenerate the tissue without any follow-up pharmacological and surgical treatments. Composite materials address some of these challenges, as they combine and tailor properties of particular composite components [1, 2]. The primary goal of bone tissue engineering is to design osteoconductive or, ideally, osteoinductive materials. The former will support bone tissue formation and remodeling by providing a proper surface or scaffold for osteoblastic cells to attach, proliferate and differentiate [3]. The latter indicates a ‘smart biomaterial’ that will stimulate bone tissue to regenerate. Osteogenic scaffolds should mimic cancellous bone morphology and function in order to optimize integration with surrounding tissue. That is why the requirements of these materials include a controllable degradation rate that is synchronized with bone tissue regeneration [4, 5], mechanical properties that promote regeneration of a particular skeletal site [6, 7] and biological activity that supports or promotes osteogenesis [8, 9]. Several composite materials have been reported to be suitable for bone tissue engineering, but only a few have been defined as osteoinductive [10–14]. In fact, there is a limited number of material and biological components that display osteoinductive properties, such as natural or synthetic hydroxyapatite (HA), demineralized bone matrix or natural or recombinant bone morphogenetic proteins [15]. Although poly-L-lactide-co-glycolide (PLGA) has been approved by the FDA for a clinical use, its application for bone tissue engineering is limited by its poor osteoconductivity and suboptimal mechanical properties for application in load-bearing sites [16]. It is currently very common to combine PLGA with other types of materials, e.g. bioactive glasses, to obtain composites exhibiting a balanced set of properties. Thus, we have tested the hypothesis that the sol–gel-derived bioactive glasses (SBGs) we previously developed and described [17– 19] may serve as osteoinductive components of PLGA-based composite scaffolds for human bone-marrow-derived mesenchymal stem cell (hBMSC)-based bone cell therapies. The phenomenon of osteoinductivity can rely on (1) material factors such as material bioactivity, porosity, phase composition and crystallinity and (2) biological factors such as osteoinductive growth factors and hormones [20]. The latter also depend on the implantation site and the species-specific biological responses. To date, bone morphogenetic proteins (BMPs) are the best known osteoinductive molecules found in bone. They are capable of inducing bone formation either in natural or recombinant forms when implanted in heterotopic
2. Materials and methods 2.1. Materials
PLGA was synthesized via a ring opening process in the presence of low toxicity zirconium acetyloacetonate as a copolymerization initiator [33]. The molar ratio of L-lactide to glycolide in the copolymer was 85 : 15, and the molecular masses of PLGA were Mn=80 kDa and Mw=152 kDa. Bioactive glasses (A2 and S2) of the composition of A2: 40% SiO2–52% CaO–6%P2O5 [mol%] and S2: 80% SiO2–16% CaO–4%P2O5 [mol%] were produced using the sol–gel method as described previously [17]. Composite scaffold fabrication has been previously described in detail [29]. Briefly, PLGA scaffolds modified 2
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Mononuclear cells were seeded to T-75 flasks (BD Falcon) in a standard cell growth medium composed of alpha-MEM, 10% mesenchymal stem cell qualified fetal bovine serum (MSCqualified FBS) and 1% antibiotics (penicillin and streptomycin). Media were initially changed at day 7 of the primary culture, then every 2–3 days. Cells were grown in primary culture until they reached 80–90% confluence. hBMSCs were then detached from the bottom of tissue culture flasks with 0.25% Trypsin-EDTA and either loaded onto the scaffolds or further expanded in T-75 flasks. All experimental cultures were established with hBMSCs at passages 1–6.
with A2 or S2 21 vol% SBGs (designated as A2-PLGA and S2-PLGA) were fabricated by the addition of porogen (NaCl) into the copolymer and bioactive glass suspension in methylene chloride. The mixed slurry was then packed into cylindrical vials and dried in air (24 h) and in vacuum (48 h). The vials with NaCl/SBG–PLGA mixture were cut into slices of approximately 3 mm thickness and extensively washed in ultra-highquality (UHQ) water. Finally, the samples were dried in air and vacuum for at least 24 h. PLGA reference scaffolds were prepared using a similar protocol to serve as controls. 2.2. Materials evaluation
2.3.2. Cell seeding and cultures on the experimental scaffolds. Before the cell seeding procedure, SBG–PLGA
2.2.1. Porosity. The porosity (P) of scaffolds was calculated
according to the equation, where ρs, the apparent density of the scaffold, was calculated by dividing the scaffold weight by the scaffold volume, and ρp is the density of PLGA (1.3 g cm−3). The porosity of the 3D scaffolds was verified by the mercuric porosimetry (Poremaster 60 Quantachrome). The reported porosity of scaffolds represents the average of three samples of each type analyzed.
experimental scaffolds were sterilized in 70% ethanol (water solution) for 48 h, and then washed three times with PBS to remove ethanol traces from the scaffold pores. HBMSCs were suspended at a density of (1.0–1.4) × 105 cells per 120–150 µl of standard cell growth medium supplemented with 40 µl of fibrinogen (stock solution of 1.75 mg ml−1 PBS) and 1.6 µl of thrombin (stock solution of 25 U ml−1 in PBS) to initiate clot formations. Sterilized scaffolds (12 mm in diameter and 2–4 mm height) were placed into separate wells of 24-well plates, and each scaffold was loaded with a 150 µl aliquot of hBMSC suspension. Scaffolds were then incubated for 40 min at 37 °C in a humidified 5% CO2 atmosphere to allow the fibrinogen polymerization. Then, 1.5 ml of standard cell growth medium was added per well. After an initial 24 h culturing, the scaffolds were placed in a new 24-well plate and fresh growth medium was added. Cells in scaffolds were either grown in a standard growth medium or an osteogenic medium that consisted of a growth medium supplemented with 50 µg ml−1 ascorbate-2-phosphate and either 100 ng ml−1 rhBMP-2 or 10-7M dexamethasone (Dex). Following day 1, media were changed every 2–3 days.
2.2.2. Bioactivity. The bioactivity of the composites was
assessed by an in vitro simulated body fluid (SBF) test according to the method described by Kokubo [34]. SBF was prepared by dissolving reagent-grade chemicals in de-ionized water and buffered to pH 7.4 at 37 °C with TRIS tris(hydroxymethyl aminomethane) and HCl. The samples were immersed in SBF and incubated at 37 °C for 14 and 21 days. The samples were then washed in UHQ water and air and vacuum dried to a constant weight. 2.2.3. Microstructure. The internal (cross-section) and external morphology of the scaffolds before and after SBF incubation, as well as the chemical composition of the 3D scaffolds, were examined using SEM/EDAX analyses (Nova 200, NanoSEM, FEI, USA).
2.4. Cell culture evaluation 2.4.1. Cell viability assay. Cell viability was examined with the CellTiter96Aqueous One Solution Cell Proliferation Assay (MTS, Promega). After 7 or 21 days of culture, scaffolds were washed 3 times with PBS, transferred to a fresh 24-well plate and each scaffold was covered with 0.5 ml solution of 10% MTS reagent in PBS. The colorimetric reaction was developed in the dark at 37 °C in a humidified 5% CO2 atmosphere. Afterward, the MTS solutions were collected from the 24-well plate, 200 µl aliquots were transferred to separate wells in a 96-well plate and the absorbance was measured at 490 nm in a plate reader (ELx 808IU, UltraMicroplate Reader, Biokom). The absorbance units obtained from the MTS test were converted to the live cell number based on a standard curve prepared in our laboratory using hBMSCs seeded at different densities and cultured overnight. After 24 h, the MTS test was performed. This reflects the changes in the MTS absorbance units versus the live cell number.
2.2.4. Mechanical properties. The compressive strength and
compressive modulus values of as-prepared scaffolds and scaffolds incubated for 1 month in SBF were determined using a universal testing machine (Zwick Z2.5, Germany). Samples (N=3 for each type) were compressed with a constant deformation rate of 1 mm min-1. The average values represent the compressive strength at 10% strain and compressive modulus of scaffolds. 2.3. Cell culture 2.3.1. hBMSC isolation and culture expansion. Unless stated
otherwise, all cell culture reagents were purchased from Life Technologies. Adult human bone marrow stromal cells (hBMSCs) were harvested from the iliac crest of adult patients (32–78 years old, both genders) according to the approved Institutional Review Board protocol (KBET/17/L/2007). The mononuclear cell fraction was isolated using Ficoll-Paque (GE Healthcare), as described in the manufacturer’s protocol.
2.4.2. Scanning electron microscopy (SEM) and confocal microscopy analyses. For SEM analyses, hBMSCs were 3
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Table 1. Primer sequences and product lengths.
Primer sequence Gene name
Forward (5’–3’)
Reverse (5’–3’)
Length
Tata box-binding protein Osteopontin Bone morphogenetic protein-2 Osteocalcin Osteoprotegrin Collagen I Bone sialoprotein
GGAGCTGTGATGTGAAGTTTCCTA TGGAAAGCGAGGAGTTGAATG TGCTAGTAACTTTTGGCCATGATG
CCAGGAAATAACTCTGGCTCATAAC CATCCAGCTGACTCGTTTCATAA GCGTTTCCGCTCTTTGTGTT
91bp 117bp 86bp
AAGAGACCCAGGCGCTACCT GTCAAGCAGGAGTGCAATCG GTCTAGACATGTTCAGCTTTGTGGA AACGAAGAAAGCGAAGCAGAA
AACTCGTCACAGTCCGGATTG TAGCGCCCTTCCTTGCATT CTTGGTCTCGTCACAGATCACGTCAT TCTGCCTCTGTGCTGTTGGT
110bp 59bp 245bp 77bp
were washed three times with PBS, then cut into halves and placed in the Eppendorf tubes. Any PBS remaining inside the scaffolds was aspirated and 300 µl aliquots of cell lysis buffer supplemented with protease inhibitors were added to individual scaffolds. Subsequently, the proteins were extracted from the 3D cultures by repeated vortexing and freeze/thaw cycles. Protein lysates were clarified from cell debris by centrifugation. Clarified protein lysates were stored at −80 °C until the western blot analyses. Protein concentrations were determined with a Micro BCA protein assay kit (Thermo Scientific). Equal amounts of protein lysates from cultured hBMSCs on individual scaffolds were separated on NuPAGE 4–12% bis-Tris gels under reducing conditions and then transferred to PVDF membranes. Membranes were probed overnight with primary anti-human Smad1 or anti-human phospho-Smad 1/5/8 antibodies (Cell Signaling Technology) and then with horseradish peroxidase-linked rabbit anti-human secondary antibodies (GE Healthcare). The peroxidase-based signal was detected using Western Lightning Chemiluminescence Reagent Plus (GE Healthcare) and captured on Hyperfilm ECL chemiluminescent films (Perkin-Elmer).
cultured for 9 days on experimental scaffolds in standard growth medium. Then, scaffolds were washed 3 times with PBS, treated with 2.5% glutaraldehyde solution in cacodylate buffer for 2 h, washed in cacodylate buffer and fixed for 1 h in 0.25% OsO4 in cacodylate buffer. Samples were dehydrated in increasing concentrations of ethanol. Finally, the samples were dried in the CO2 critical point dryer, attached to microscope adhesive holders and covered with a thin gold layer. The SEM images were taken with the JEOL JSM5410 scanning electron microscope (JEOL, Tokyo, Japan). For confocal microscopy imaging, hBMSCs were cultured for 7 days on experimental scaffolds in a growth medium supplemented with ascorbate-2-phosphate and dexamethasone. Then cells were washed 3 times with PBS and fixed in 4–8% formalin solution in PBS for 20 min. Cell cytoplasm was stained with 5% eosin (Eosin Y, ethanol solution, Thermo Scientific) and cell nuclei with DAPI diluted 1 : 1000 in PBS. The confocal images were taken using the Zeiss Axiovert 200M microscope combined with LSM510 Meta Confocal head (Carl Zeiss, MicroImaging GmbH, Jena, Germany), using the Zeiss 20 × /0.8 Plan-Apochromat objective.
2.4.4. Western blot analyses. Whole cell extracts were obtained
2.4.5. Quantitative RT-PCR. Gene expression studies were performed at culture days 2, 10 and 20 on experimental scaffolds. Briefly, the total RNA was extracted from cultures on individual experimental scaffolds using the SPEEDTOOLS TOTAL RNA EXTRACTION KIT (Biotools) according to the manufacturer’s protocol. The aliquots of 0.5 µg RNA were reverse transcribed to cDNA (SuperScript III First Strand Synthesis System, Life Technologies). PCR reaction mixtures (15 µl total volume) contained 1 µl cDNA, gene-specific primers, SYBR®Green I, AmpliTaq Gold® DNA Polymerase and the reaction buffer as recommended by the manufacturer. Primer sequences are shown in table 1. Taq-Man Gene Expression Assays were used for the analyses of Runx2 (Hs00231692_m1 RUNX2) and Tata box-binding protein (Hs99999910_m1 TBP) mRNA levels. Each reaction mixture (10 µl total volume) contained 1 µl cDNA, specific TaqMan probe and TaqMan Gene Expression Master Mix as recommended by the manufacturer. All PCR reactions were run in StepOnePlus Real-Time PCR apparatus (Applied Biosystems) and the ddCt method was used for the analyses of gene expression levels. Gene expression levels of the housekeeping gene, Tata box-binding protein, were used in each reaction to normalize the values.
as described by Osyczka and Leboy [30]. Briefly, hBMSCs were examined two days post-seeding and subsequent culture in the experimental scaffolds. Scaffolds containing hBMSCs
2.4.6. Evaluation of collagen production and mineralization of extracellular matrix. For the evaluation of total collagen protein
2.4.3. ALP activity measurements. ALP activity was mea-
sured kinetically at day 7 of the hBMSC cultures on experimental scaffolds, as originally described by Osyczka and Leboy [30]. Briefly, cell viability was determined using the MTS assay described above, and scaffolds were washed extensively with PBS prior to performing the ALP activity measurements. Scaffolds were then cut into halves, placed into 1.5 ml centrifuge tubes and covered with 200 µl of cell digestion buffer. Samples were extensively vortexed and centrifuged, followed by a freeze/thaw cycle. Finally, the thawed scaffolds were incubated at 37 °C for 30 min and again extensively vortexed and centrifuged to remove cell debris. Clarified protein extracts were used for reactions with the alkaline phosphatase substrate, p-nitrophenol phosphate (pNp) (Sigma Aldrich). The changes in the absorbance at 405 nm were measured for 6 min at 1 min intervals. The total ALP activity was expressed as nmol pNp/min/total volume of the protein extract. The ALP activity was then normalized to the number of viable cells estimated from the MTS assay.
4
J Filipowska et al
Biomed. Mater. 9 (2014) 065001
Figure 1. SEM and EDAX analyses of the external and internal surfaces of (a) PLGA, (b) S2-PLGA and (c) A2-PLGA scaffolds. The middle panel shows representative high-magnification SEM images of the scaffold external surface area. Scale bars are indicated at the bottom panel of each image. EDAX results are representative (average) for the studied surface.
levels and mineralization of extracellular matrix (ECM), hBMSCs were cultured for 20 days on the experimental scaffolds in osteogenic conditions and analyzed as described before [25]. Briefly, at culture day 20, hBMSCs were assayed for viability with MTS, and then the scaffolds were extensively washed with PBS and stained for collagen with Sirius Red and for minerals with Alizarin Red S. The dyes were then extracted and measured colorimetrically at 405 nm (AR) or 490 nm (SR).
v.10, StatSoft software. One-way or multivariate ANOVA and post-hoc Tukey’s tests were used to calculate statistically significant differences at p
