enriched hydrogels

1 downloads 12 Views 412KB Size Report
Bioactive glasses in the form of particulates have been added to hydrogels to enhance mineralizability upon incuba- tion in simulated body fluid (SBF), which ...

High-resolution synchrotron X-ray analysis of bioglass-enriched hydrogels Svetlana Gorodzha,1 Timothy E. L. Douglas,2 Sangram K. Samal,3 Rainer Detsch,4 Katarzyna Cholewa-Kowalska,5 Kevin Braeckmans,3 Aldo R. Boccaccini,4 Andre G. Skirtach,2 Venera Weinhardt,6 Tilo Baumbach,7 Maria A. Surmeneva,1 Roman A. Surmenev1,8 1

Department of Experimental Physics, National Research Tomsk Polytechnic University, Russia Department of Molecular Biotechnology, Coupure Links 653, Ghent University, Belgium 3 Laboratory of General Biochemistry & Physical Pharmacy, Ghent University, Belgium 4 Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstr. 6, Erlangen 91058, Germany 5 Department of Glass Technology and Amorphous Coatings, AGH University of Science and Technology, Krakow, Poland 6 Centre for Organismal Studies, University of Heidelberg, Heidelberg, Germany 7 Laboratory for Applications of Synchrotron Radiation, Karlsruhe Institute of Technology, Karlsruhe, Germany 8 Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Stuttgart, Germany 2

Received 18 September 2015; revised 7 December 2015; accepted 5 January 2016 Published online 00 Month 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35642 Abstract: Enrichment of hydrogels with inorganic particles improves their suitability for bone regeneration by enhancing their mechanical properties, mineralizability, and bioactivity as well as adhesion, proliferation, and differentiation of boneforming cells, while maintaining injectability. Low aggregation and homogeneous distribution maximize particle surface area, promoting mineralization, cell–particle interactions, and homogenous tissue regeneration. Hence, determination of the size and distribution of particles/particle agglomerates in the hydrogel is desirable. Commonly used techniques have drawbacks. Highresolution techniques (e.g., SEM) require drying. Distribution in the dry state is not representative of the wet state. Techniques in the wet state (histology, mCT) are of lower resolution. Here, self-gelling, injectable composites of Gellan Gum (GG) hydrogel

and two different types of sol–gel-derived bioactive glass (bioglass) particles were analyzed in the wet state using Synchrotron X-ray radiation, enabling high-resolution determination of particle size and spatial distribution. The lower detection limit volume was 9 3 1025 mm3. Bioglass particle suspensions were also studied using zeta potential measurements and Coulter analysis. Aggregation of bioglass particles in the GG hydrogels occurred and aggregate distribution was inhomogeneous. Bioglass promoted attachment of rat mesenchymal stem cells C 2016 Wiley Periodicals, Inc. J Biomed (rMSC) and mineralization. V Mater Res Part A: 00A:000–000, 2016.

Key Words: X-ray, bioglass, composite, hydrogel, particle

How to cite this article: Gorodzha S, Douglas TEL, Samal SK, Detsch R, Cholewa-Kowalska K, Braeckmans K, Boccaccini AR, Skirtach AG, Weinhardt V, Baumbach T, Surmeneva MA, Surmenev RA. 2016. High-resolution synchrotron X-ray analysis of bioglass-enriched hydrogels. J Biomed Mater Res Part A 2015:00A:000–000.


Hydrogel–inorganic particle composites have recently been gaining interest as materials for bone regeneration. Enrichment with inorganic particles improves hydrogels’ mechanical properties,1–9 mineralizability with calcium phosphate (CaP),10,11 and biological properties, for example, bioactivity and adhesion, proliferation, and differentiation of boneforming cells,8,9,12 while maintaining injectability. Bioactive glasses in the form of particulates have been added to hydrogels to enhance mineralizability upon incubation in simulated body fluid (SBF), which contains mineral ions at concentrations similar to those found in blood plasma. Addition of bioglass particles to a variety of noninjectable

hydrogels resulted in enhanced apatite formation after incubation in SBF.13–18 Furthermore, addition of bioglass particles to the anionic, calcium-binding polysaccharide gellan gum (GG), which has been investigated as a material for regeneration of skin, cartilage, and invertebral discs,19–23 can promote GG hydrogel formation and increase compressive strength.24 This is presumably due to the release of such ions as Ca21 from bioglass particles, which ionically cross-link the anionic GG polymer chains. The resulting GG–bioglass composite hydrogels were self-gelling, injectable, and promoted mineralization upon incubation in SBF. Two types of GG–bioglass composite hydrogels displayed antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA). These

Correspondence to: T. E. L. Douglas; e-mail: [email protected] Contract grant sponsor: BMBF X-Regio Project



bioglasses, hereafter referred to as A2 and S2, were produced by a sol–gel method25 and had high and low CaO contents, namely 54 and 16 mol %, respectively. Incorporation of A2 and S2 bioglass particles in GG hydrogels also increased the proliferation of rat mesenchymal stem cells. All these aforementioned features are desirable for applications in bone regeneration. It is known that inorganic particles dispersed in hydrogels may form aggregates, and hence efforts have been made to reduce aggregation by the use of surfactants,26 or by simultaneous formation of particles inside hydrogels during hydrogel formation.27 Hence, knowledge of the size and distribution of particles/particle agglomerates in the hydrogel is desirable. Such methods as scanning electron microscopy (SEM), which are used to visualize particles with a higher resolution, require drying. Distribution and size in the dry state and in the wet state may differ greatly. Methods which can be used to study particle distribution in the wet state include sectioning and histological staining, both followed by imaging with light microscopy. With the development of super-resolution light microscopes, biocompatibility of materials can be addressed on the cellular level. However, both SEM and light microscopy provide only two-dimensional images (for optically non-transparent materials in case of light microscopes), and thus analysis of particle distribution is limited to the sample surface. To image biomaterials in three dimensions, methods such as electron tomography or micro-computed tomography (lCT) can be used. While lCT has a lower resolution than electron tomography, it enables analysis of large three-dimensional volumes. For example, volumes up to 4 3 4 3 4 mm3 can be investigated with 1 mm resolution (in comparison to 700 3 500 3 200 nm3 in electron microscopy.28 Also, lCT provides sufficient contrast for low-Z materials, such as hydrogels.29 In this study, the aforementioned self-gelling, injectable GG-bioglass hydrogel composites developed in previous work were analyzed in the wet state using synchrotron X-ray radiation to determine size and spatial distribution of bioglass particles/aggregates in the GG hydrogel. Synchrotron X-ray radiation, chosen due to its nondestructive character, can be used with hydrogels in the wet state (i.e., not with dried hydrogels), and has a higher resolution than other techniques as mentioned above. Synchrotron X-ray imaging has been widely used in biomaterials research to image bone formation around titanium-based implants.30–34 However, to our best knowledge, this is the first report on use of this technique to study hydrogel–particle composites in the biomaterials field. To better interpret bioglass particle aggregation in GG hydrogels, the bioglass particles themselves were subjected to physicochemical characterization. Coulter analysis was performed to study particle size distribution. Zeta potential was also measured. Furthermore, rat mesenchymal stem cells (rMSC) were seeded onto GG–bioglass hydrogel composites. Previous work has demonstrated the cytocompatibility of GG-A2 and GG-S2 hydrogel composites for rMSC, and the ability of rMSC to proliferate on the surface of the composites.24 In



TABLE I. Bioactive Glasses A2 and S2: Composition Molar % component Bioglass type A2 S2




40 80

54 16

6 4

this study, rMSC were visualized with fluorescence microscopy to study cell distribution on the surface of GG–bioglass composites and to establish if there was a link between cell and bioglass particle distribution. MATERIALS AND METHODS

GG hydrogel–bioglass composite production Bioglass A2 and S2 were produced by a sol–gel technique25 (see Table I for composition). Self-gelling GG–bioglass hydrogel composites (1% w/v bioglass, 0.7% w/v GG) were generated as described in previous work.24 Briefly, composite components were sterilized prior to composite fabrication by autoclaving (10 min, 1158C, 1.2 bars). Four milliliters of a suspension of A2 or S2 bioglass particles in 18.75 mM MgCl2 solution were prepared in an ultrasound bath for 5 min. This suspension was preheated to 408C and mixed with a 16 ml 0.875% (w/v) GG solution, which had also been preheated to 408C, and mixed immediately and intensively with a spatula for 5 s. The resulting homogeneous mixture was then immediately cast on a Petri dish of diameter 50 mm at room temperature. After 15 min, when the GG–hydrogel composite hydrogels had set, cylinders of diameter 6 mm and height of 1 cm were cut out with a hole punch. Synchrotron X-ray imaging The cylindrical GG–bioglass hydrogel composite samples of height 1 cm and diameter 6 mm were placed in 0.5 ml vessels filled with water to prevent evaporation. Samples were subjected to X-ray imaging at the Topo-Tomo beamline, ANKA light source facility, Karlsruhe Institute of Technology, Germany. From the spectrum of the bending magnet (white beam) 10 keV monochromatic X-ray beam was selected by the double multilayer monochromator. The detector system consisted of Andor Neo camera (sensor size 5.5 megapixels, 6.5 lm pixel size), BAM microscope 3.63 and the scintillator LAG 50 lm thickness resulting in the effective pixel size of 1.8 lm and field of view of 10 3 6.7 mm. Volumes were obtained by X-ray computed tomography: 1501 projection images were taken over an angular range of 3608 with exposure time 2 ms. The dark current of the camera and reference images without sample were taken to reduce artefacts. Zeta potential and particle size distribution measurements A2 and S2 particles were suspended in phosphate-buffered saline (PBS, GIBCO, Merelbele, Belgium) or double-distilled water (ddH2O) at a concentration of 1 mg/ml for zeta potential measurements. Zeta potential was measured using a Zetasizer Nano Series (Malvern Instruments, Hoeilaart, Belgium). The particle suspensions were transferred to DTS 1060 disposable folded capillary cells. Measurements were



TABLE II. Zeta Potential of Bioactive Glasses A2 and S2 in Media ddH2O, ddH2O with pH Stabilized at 7.5 by Addition of HCl, and Phosphate-Buffered Saline (PBS) and Final pH Values of Media Measured Bioglass type


A2 A2

ddH2O ddH2O

A2 S2 S2

PBS ddH2O ddH2O



Final pH of medium 11.6 7.5 (stabilized using HCl) 9.5 9.0 7.5 (stabilized using HCl) 7.7

Zeta potential [mV]; mean 6SD 21.68 6 0.40 28.08 6 0.43 217.32 60.86 219.02 6 0.69 219.8 6 0.48 216.72 6 1.31

carried out at a temperature of 258C. Because of possible reactions of the bioglass preparations with the medium of dispersion (ddH2O, PBS), the pH was also measured as it may influence the zeta potential. For bioglass suspensions in ddH2O, measurements were made both after adjustment of pH to 7.5 by addition of 1M HCl and without pH adjustment. The potential was determined five times (each measurement being the average of 100 runs). The mean values and standard deviations were calculated and are shown in Table II. The instrument automatically calculates zeta potential according to the Smoluchowski equation. The size distribution of the microparticles was studied by the electrical sensing zone method with a Beckman–Coulter MultisizerTM 4 COULTER COUNTERV. For each Coulter Counter experiment, 20 lL of microparticle suspension, prepared by suspending bioglass powders in PBS at a concentration of 1 mg/mL, was diluted in 10 mL PBS. Fifty microliters of this dilution was applied in the Multisizer, which was equipped with a 20 lm aperture tube, which permits sizing down to 0.4–16 mm equivalent spherical diameter. R

Isolation of rat mesenchymal stem cells (rMSC), cell seeding, and visualization rMSC were derived from the bone marrow from of femora and tibiae of Lewis rats was performed as described by Lamers et al. 35 and in a previous publication.24 Briefly, bone marrow was expelled and disaggregated. rMSC and hematopoietic cells in media were placed on cell culture plastic. When cells grew to confluence, there was a coculture, which consisted predominantly of rMSC with some hematopoietic stem cells. Hematopoietic cells were undesirable and were removed by media change after 3 days and rinsing with PBS. Media was subsequently refreshed every 3 days until subconfluency was attained. rMSC differentiation ability was confirmed by induction of the osteogenic, adipogenic, and chondrogenic phenotypes (data not shown). To visualize rMSC on GG–bioglass hydrogel samples, samples were incubated in PBS for 7 days, then 1 3 105 rMSC were seeded and cultured for 14 days. rMSC distribution, cytoskeleton formation, and mineral formation were visualized by fluorescence microscopy after staining as described in detail in previous publications.36–38 Briefly, adherent cells were fixed with 3.7 vol % paraformaldehyde for 10 min and permeabil-

ized with 0.1 vol % Triton X-100 (in PBS) for 10 min at room temperature. Two different staining protocols were used. In the first protocol, green fluorescent Sytox (Molecular ProbesV) and red fluorescent Rhodamine Phalloidin (Molecular ProbesV) were used for cell staining. To detect the cytoskeleton, the cells were incubated for 60 min with phalloidin (diluted 1:50 by volume) at room temperature followed by incubation with 1 lg/ml Sytox for 5 min. Samples were washed and left in PBS for microscopic imaging. In the second protocol, to detect the cytoskeleton, cells were incubated for 60 min with red fluorescent Alexa FluorV Phalloidin (Molecular ProbesV) (diluted 1:50 by volume) at room temperature followed by incubation with 1 mg/mL blue fluorescent DAPI (40 ,6-diamidino-2-phenylindole dihydrochloride—Roche) for 5 min to visualize cell nuclei. To visualize mineral, samples were subjected to the OsteoImageTM Mineralization Assay (Lonza) for 30 min at room temperature (stock solution was diluted 1:100 (v/v)). Finally, samples were rinsed and stored in PBS for microscopic imaging (FM, DMI 6000B, Leica). R





Synchrotron X-ray imaging of GG–bioglass hydrogel composites and zeta potential and size distribution measurements of bioglass particle suspensions 3D reconstructions of selected regions of the GG-S2 and GGA2 hydrogel–bioglass composites are presented in Figure 1. It was clear that aggregates of different sizes were present, and that these aggregates were not uniformly distributed throughout the GG hydrogels. A higher number of larger aggregates appeared to be present in GG-A2 hydrogel–bioglass composites. Results of quantitative analysis of size distributions of aggregates are shown in Figure 2 and confirmed the nonuniformity of aggregate size. The most commonly occurring A2 and S2 aggregates were of volume 0.0002 mm3 (200,000 mm3) corresponding to an approximate diameter of 70 mm, assuming that the aggregates have a spherical shape. Numerous aggregates in the volume range 0.0002–0.005 mm3 (200,000–5,000,000 mm3) were detected, corresponding to approximate diameters in the range 70–210 mm. The largest S2 aggregates were of volume 0.5 mm3 (5 3 109 mm3), corresponding to an approximate diameter of 980 mm, assuming a spherical shape. In contrast, the largest A2 aggregates had a considerably higher volume. The largest particle detected was of the approximate volume 7 mm3 (70 3 109 mm3), corresponding to the approximate diameter of 2370 mm, assuming a spherical shape. Results of Coulter analysis are shown in Figure 3. The A2 particle distribution displayed two maxima at 0.14 and 8 mm3, while the S2 particle distribution displayed a single maximum at 0.6 mm3. These results prove that aggregation of bioglass particles in the GG hydrogels occurred. It should be pointed out that aggregates smaller than 0.00006 mm3, (corresponding to approximate diameter 50 mm), were not detected by synchrotron X-ray radiation. Possibly, individual particles or small aggregates of volume