Effect of ion release from Cu-doped 45S5 Bioglass® on 3D endothelial cell morphogenesis

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 Acta Biomaterialia xxx (2015) xxx–xxx 1

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat 5 6

4

Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis

7

Christoph Stähli a, Mark James-Bhasin a, Alexander Hoppe b, Aldo R. Boccaccini b, Showan N. Nazhat a,⇑

3

8 9 10 11 1 2 3 6 14 15 16 17 18 19 20 21 22 23 24 25

a b

Department of Mining and Materials Engineering, McGill University, 3610 University Street, Montreal, QC H3A 0C5, Canada Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany

a r t i c l e

i n f o

Article history: Received 24 October 2014 Received in revised form 10 February 2015 Accepted 5 March 2015 Available online xxxx Keywords: Bioactive glass Copper Bioinorganics Capillary-like network Angiogenesis

a b s t r a c t Both silicate-based bioactive glasses and copper ions have demonstrated angiogenic activity and therefore represent promising bioinorganic agents for the promotion of vascularization in tissue-engineered scaffolds. This study examined the effect of ionic release products from 45S5 BioglassÒ doped with 0 and 2.5 wt.% CuO (BG and Cu-BG respectively) on the formation of capillary-like networks by SVEC410 endothelial cells (ECs) seeded in a three-dimensional (3D) type I collagen matrix. Copper and silicon release following 24 h dissolution increased non-proportionally with Cu-BG concentration in cell culture medium, while calcium levels were decreased below the initial medium concentration. EC network length, connectivity, branching, quantified by means of a 3D morphometric image analysis method, as well as proliferation and metabolic activity were reduced in a dose-dependent fashion by BG and CuBG ionic release products. This reduction was less prominent for BG compared to an equivalent concentration of Cu-BG, which was attributed to a lower extent of silicon release and calcium consumption. Moreover, a CuCl2 dose equivalent to the highest concentration of Cu-BG exhibited no effect on ECs. In conclusion, while the previously reported pro-angiogenic activity of both BioglassÒ and copper may not be reflected in a direct response of ECs, this study provides a maximum glass concentration for non-harmful angiogenic stimulation to be examined in future work. Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

45 46

1. Introduction

47

The success of tissue engineering strategies is dependent on the formation of an effective blood supply to ensure cell viability following scaffold engraftment. Approaches such as endothelial cell (EC) transplantation or the delivery of angiogenic growth factors have been proposed in order to promote vascularization [1,2]. Bioinorganics are gaining increasing interest as angiogenic agents due to their higher cost-effectiveness and a more sustained activity following administration compared to growth factors [3–5]. Since the invention of 45S5 BioglassÒ, silicate-based bioactive glasses have been clinically used as bone regenerative materials for several decades, owing to their ability to bond to bone [6]. Specifically, upon exposure to biological fluid, glass modifying ions are leached out while the silicate network is hydrolyzed, resulting in the formation of a silica-rich gel layer on the glass surface.

48 49 50 51 52 53 54 55 56 57 58 59 60

⇑ Corresponding author. Tel.: +1 514 398 5524; fax: +1 514 398 4492. E-mail addresses: [email protected] (C. Stähli), [email protected] (M. James-Bhasin), [email protected] (A. Hoppe), [email protected] (A.R. Boccaccini), showan.nazhat@ mcgill.ca (S.N. Nazhat).

Subsequently, a carbonated hydroxyapatite surface layer is precipitated, which supports osteoblast attachment and bone growth [7,8]. More recently, the specific effects of ionic dissolution products from bioactive glasses on osteogenic and angiogenic processes have gained increasing interest [9]. For example, ions released from 45S5 BioglassÒ were shown to increase osteoblast proliferation, insulin-like growth factor II production and to regulate expression levels of numerous genes including cell cycle regulators and matrix metalloproteinases (MMPs) [10,11]. Moreover, 45S5 BioglassÒ dissolution products stimulated vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) production from fibroblasts in vitro, possibly through an increase in interleukin-6 or tumor necrosis factor-a release [12], and in turn increased EC proliferation and the formation of tubular structures [12–14] (reviewed in [15]). In vivo, BioglassÒ enhanced de novo vascularization in polymeric scaffolds [16,17] while a BioglassÒ granula matrix was shown to support axial vascularization in the arteriovenous loop model [18]. The doping of bioactive glasses with small quantities of physiological trace elements, such as cobalt, zinc or magnesium, may further enhance their biological response [3]. In particular, copper ions were shown to stimulate EC proliferation, capillary-like

http://dx.doi.org/10.1016/j.actbio.2015.03.009 1742-7061/Ó 2015 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 2 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113

114 115 116 117 118 119 120 121 122

C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

network formation and in vivo scaffold vascularization [19–23]. The angiogenic activity of copper has been attributed to an up-regulation of hypoxia inducible factor (HIF)-1a [24], VEGF [25] as well as endothelial nitric oxide synthase expression [20]. Moreover, copper plays a role in extracellular matrix (ECM) degradation and remodeling by regulating gene expression, protein levels and activities of MMPs [26–29]. However, exceeding optimal copper ion concentrations may result in cytotoxicity due to the generation of free radical species, and brings about the risk of neurodegenerative diseases [30,31]. Therefore, the doping of bioactive glasses with small amounts of copper provides a promising approach for controlled and sustained ion release. The incorporation of copper into mesoporous high-silicate glass scaffolds as well as a calcium silicate bioceramic was shown to enhance (HIF)-1a and VEGF expression and, as a result, EC morphogenesis [32,33]. Moreover, copper-doped borosilicate glasses stimulated in vivo angiogenesis and bone regeneration [34,35]. Recently, copper-doped 45S5 BioglassÒ-derived scaffolds were developed and characterized in terms of their dissolution reactions [36]. In order to optimize these materials toward ultimately successful and non-toxic angiogenic stimulation, a careful examination of the dose-dependent effects of the different released ions on ECs is required. In this study, three-dimensional (3D) EC cultures were exposed to constant concentrations of ionic release products from both copper- and non-doped 45S5 BioglassÒ, along with an equivalent copper ion dose, in order to distinguish between effects caused by copper and other ions. A recently developed image analysis method [37] was applied which allowed for a complete 3D morphometric characterization of EC capillary-like networks in type I collagen hydrogel matrices, along with an examination of EC proliferation and metabolic activity. 2. Materials and methods 2.1. Glass production Melt-derived 45S5 bioactive glasses doped with 0 or 2.5 wt.% CuO (BG and Cu-BG respectively; Table 1) were produced as described previously [36]. Briefly, SiO2, Na2CO3, CaCO3, (Merck, Germany), Ca3(PO4)2 and CuCO3
Cu(OH)2 (Sigma–Aldrich, Germany) were homogenized and melted at 1450 °C for 45 min. The glass was then fritted and milled to a final particle size of d50 = 5 lm.

123

2.2. Determination of glass dissolution

124

BG and Cu-BG powders were incubated for 2, 6, 24 or 72 h in triplicates in flat-bottom glass vials (2.5 cm in diameter) containing 8 mL Dulbecco’s modified Eagle’s medium (DMEM; ATCC, USA) supplemented with 1  penicillin–streptomycin–glutamine mixture (PSG; Gibco, Canada) and 10 lg/mL gentamicin (Life Technologies, Canada). The medium supernatant was then removed and filtered (pore size = 0.22 lm). Aliquots of 1 mL were diluted 1:10 in deionized water (DW) containing 4% HNO3 and 1% H2O2 (w/v) and digested at approximately 100 °C for 1 h in a closed container (in order to avoid evaporation). Copper, silicon and calcium ion concentrations were determined using an

125 126 127 128 129 130 131 132 133 134

Table 1 Glass formulations. Glass code

BG Cu-BG

Components (wt.%) SiO2

Na2O

P 2 O5

CaO

CuO

45 45

24.5 24.5

6 6

24.5 22

– 2.5

inductively coupled plasma – optical emission spectrophotometer (ICP-OES; Thermo Scientific iCAP 6500). Values were calibrated against certified standards serially diluted to 100, 10, 1, and 0.1 ppm. The presence of the culture medium in the samples was verified to have no effect on the slope of the calibration. Moreover, silicate leaching from glass vials was verified to be negligible compared to release from BioglassÒ. For endothelial cell treatment, triplicate media samples after 24 h dissolution were pooled in order to achieve average concentrations of ionic release products and diluted 3-fold in DMEM supplemented with final concentrations of 1  PSG, 10 lg/mL gentamicin and 10% fetal bovine serum (FBS).

135

2.3. Cell culture in 3D collagen gels in the presence of glass dissolution products

147

An immortalized murine endothelial cell line (SVEC4-10, purchased from American Type Culture Collection, USA; CRL-2181) was cultured in DMEM, supplemented as specified in paragraph 2.2, in a humidified atmosphere of 5% CO2 at 37 °C. To produce cellular collagen gels, a 1:4 mixture of 10  DMEM (Sigma, Canada) and 2.00 mg/ml rat-tail collagen type I (First Link Ltd., UK) was neutralized by adding 5 M NaOH (Fisher Scientific, Canada). SVEC4-10 cells between passages 30 and 40 were added at a density of 3.2  105 cells/mL while the solution was kept on ice. Following thorough mixing, 50 lL of solution was pipetted to flat bottom 96-well plates and incubated for 30 min at 37 °C for collagen gelation. Subsequently, all gels were first incubated with 100 lL complete growth medium (containing no glass release products) for 30 min in order to achieve a consistent pH. Then, the media were replaced with 100 lL media containing ionic release products from BG or Cu-BG, as described in paragraph 2.2, or CuCl2 at a concentration matching copper ions released from 6.4 mg/mL Cu-BG for 24 h (10.5 ppm; Cu eq). All media were renewed with identical media after 60 min, in order to reduce the dilution effect resulting from the collagen gel volume, and subsequently every 24 h.

149

2.4. Morphometric characterization of capillary-like networks

170

Collagen gels at days 1, 3 and 5 (n = 6) were washed with Dulbecco’s phosphate-buffered saline (DPBS; Wisent, Canada) and fixed in 3.7% formaldehyde (methanol-free; Polysciences Inc., Canada) for 30 min. Then, gels were carefully removed from the wells, extracted in acetone at 20 °C and washed in DPBS. F-actin was stained by incubation in 1% bovine serum albumin solution for 30 min and subsequently in 5 units/mL Alexa FluorÒ 633 Phalloidin (Life Technologies, Canada) for 60 min. Following staining, gels were washed with DPBS (2  10 min) and transferred to glass dishes (35 mm diameter; MatTek, USA) for observation through CLSM (LSM 5, Carl Zeiss, Germany). 3D image stacks were acquired through the entire thickness in the center of the gels (Fig. S1a) using HeNe laser excitation (633 nm) with a 10  objective at a pinhole of 90 lm, a z-distance of 10 lm and a pixel size of 2.5 lm. EC network structures in 3D stacks were analyzed through a macro-automated succession of image processing operations using ImageJ software (version 1.47v; Rasband, W.S., U.S. National Institutes of Health, USA, http://imagej.nih.gov/ij, 1997–2014), as illustrated in Fig. S1 (previously reported in [37]). First, images were corrected for background fluorescence (ImageJ function ‘‘Subtract background’’) and binarized to extract EC structures (‘‘Threshold’’). Subsequently, individual round cells with no pronounced protrusions were removed using the ‘‘Analyze particles’’ filter (circularity > 0.6) and structures smaller than 100 voxels were excluded by the ‘‘3D object counter’’ plugin [38]. Planar EC structures on the gel surface (clearly distinguishable from 3D

171

Ò

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 Bioglass on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

136 137 138 139 140 141 142 143 144 145 146

148

150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169

172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

205

structures within the gel) were manually removed. The resulting network structures were skeletonized (‘‘Skeletonize (2D/3D)’’ plugin [39]) and the total network length, the number and length of discrete structures and branches as well as the number of nodes were determined (‘‘Analyze skeleton ( 2D/3D)’’ plugin [40]). Data were normalized by the analyzed gel volume, using the average gel height determined from the distance between the lowermost and uppermost ECs in the stacks. Illustrations displaying non-vertical projections were generated using the ‘‘Volume Viewer’’ plugin [41].

206

2.5. Determination of the number of EC nuclei

207

217

Fixed gels were incubated in 1 lM Hoechst 33342 solution (Invitrogen, Canada) for 15 min to stain EC nuclei and washed twice in DPBS. 3D image stacks were acquired through the entire thickness in the center of the gels (Fig. S1a) by CLSM using diode laser excitation (405 nm) with a 10  objective at a pinhole of 104 lm, a z-distance of 10 lm and a pixel size of 2.5 lm. Following manual removal of planar EC structures on the gel surface, the total number of EC nuclei in the image stacks was determined using a 3D maxima finding algorithm (‘‘ Find peaks’’ ImageJ plugin [42]) where threshold, minimum size and saddle parameters were optimized by visual comparison with raw images.

218

2.6. Measurement of cell metabolic activity

219

229

EC metabolic activity in collagen gels was monitored at days 1, 3 and 5 (n = 4). The culture medium was replaced with 150 lL of 20% AlamarBlue™ reagent (Invitrogen, Canada) and, after 4 h of incubation at 37 °C, 100 lL aliquots were collected and transferred to a black 96-well plate for analysis. The fluorescent intensity of reduced AlamarBlue™ was measured using a Mitras LB 940 microplate reader (Berthold Technologies, Germany) equipped with a 555/580 nm filter pair. Background fluorescence measured in medium incubated with acellular gels was subtracted from all values. Finally, the remaining AlamarBlue™ solution was replaced with 100 lL culture medium for ongoing treatment.

230

2.7. Statistical analysis

231

Data for each assay time point were analyzed for statistical significance between conditions using a one-way ANOVA test followed by Tukey means comparison (Origin Pro v8, OriginLab, MA, USA).

197 198 199 200 201 202 203 204

208 209 210 211 212 213 214 215 216

220 221 222 223 224 225 226 227 228

232 233 234

235

3. Results

236

3.1. Ion release from BG in cell culture medium

237

Copper, silicon and calcium concentrations resulting from BG and Cu-BG dissolution in cell culture medium, and following 3-fold dilution in plain medium, are shown in Fig. 1. Media from 0.4, 1.6 and 6.4 mg/mL Cu-BG as well as 6.4 mg/mL BG were used for treatment of ECs. Copper ion release increased with glass concentration and dissolution time but appeared to approach saturation at the highest glass concentrations as well as at longer dissolution times. Silicon release increased steadily with glass concentrations, but reached a maximal value with increasing dissolution times. BG, at a concentration of 6.4 mg/mL resulted in an approximately 30% lower silicon concentration compared to the same amount of Cu-BG. A calcium concentration of approximately 61 ppm was measured in plain cell culture medium. While only the lowest glass concentrations, as well as the shortest dissolution duration, showed a slight calcium release, higher glass concentrations and dissolution times resulted in calcium consumption. Specifically,

238 239 240 241 242 243 244 245 246 247 248 249 250 251 252

3

Fig. 1. (a) Copper, (b) silicon and (c) calcium ion concentrations released from BGs during dissolution for different time periods in cell culture medium following 3-fold dilution in plain cell culture medium. Copper release appeared to reach a saturation level with increasing glass concentration. Both copper and silicon release increased non-proportionally with duration time. Calcium levels present in cell culture medium (61 ppm) were reduced by increasing glass concentrations and duration times. Silicon release and calcium reduction were more prominent for Cu-BG compared to an identical concentration of BG (hollow square symbols). (Error bars: SD, n = 3).

the incubation of 25.6 mg/mL Cu-BG for 24 h caused a 65% reduction in calcium concentration in medium (before 3-fold dilution), suggesting the consumption of calcium in a precipitate.

253

3.2. Effects of BG release products on EC cultures in collagen

256

3.2.1. Morphometric analysis of 3D capillary-like networks Fig. 2 shows the actin skeleton and EC nuclei in typical capillary-like network structures. The evolution of networks over time in non-glass control medium as well as in medium containing Cu-BG ionic release products is shown in Fig. 3. Network length, connectivity and branching were quantified through a 3D morphometric network characterization method (illustrated in Fig. S1). The growth in total network length was reduced in a dose-dependent fashion by BG and Cu-BG release products (Fig. 4a and b). Compared to an identical concentration of nondoped BG, Cu-BG (which exhibited higher silicate release and calcium consumption) showed a stronger reduction in network length. A CuCl2 dose equivalent to the copper ion concentration released from 6.4 mg/mL Cu-BG (10.5 ppm; Cu eq) showed no effect compared to control medium.

257

Ò

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 Bioglass on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

254 255

258 259 260 261 262 263 264 265 266 267 268 269 270 271

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 4

C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

Fig. 2. EC network structures in 3D collagen hydrogel matrices at day 5 in culture. The cell morphology is visualized by phalloidin staining of F-actin (red; a and c) while cell nuclei are revealed by Hoechst-staining (blue; b and c).

Fig. 3. Maximum intensity projections of CLSM image stacks showing EC assembly into capillary-like networks in normal medium and in medium containing ions released during 24 h Cu-BG dissolution. 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291

The length distribution of discrete network structures revealed a significant increase over time in the proportion of longer structures (Fig. 4c and Fig. S2a), thus demonstrating an increase in network connectivity. Conversely, ionic release products from 6.4 mg/ mL of both BG and Cu-BG resulted in lower proportions of long structures (Fig. 4c) and a significantly lower average length per discrete structure (Fig. S2b). The length distribution of individual branches (either isolated linear structures or segments demarcated by one or two nodes) is presented in Fig. S3a–c and average length values are shown in Fig. S3d. Very similar branch length distributions were observed at all time points, indicating a dynamic balance between linear growth and branching. At day 1, a slightly lower proportion of long branches were measured in the presence of ionic release products from 6.4 mg/mL BG and Cu-BG, which mainly reflected shorter linear structures as revealed in Fig. 3. At day 5, in contrast, BG and Cu-BG resulted in higher proportions of long branches and significantly higher average length values, which reflected overall longer distances between nodes, and thus less dense networks. The number of nodes per network length, which provides a measure of network branching, increased over

time and was significantly reduced by BG and Cu-BG, where Cu-BG showed a significantly stronger reduction at day 5 compared to the same quantity of BG (Fig. 4d).

292

3.2.2. Determination of nuclei number The number of fluorescently labeled EC nuclei was determined through automated 3D image analysis as a measure of EC proliferation during network formation (Fig. 5a). While neither BG nor CuBG release products had an effect on cell number at day 1, both BG and, to a greater extent, Cu-BG significantly reduced cell numbers at day 3 and 5 in culture, in parallel with the observed reduction in network length, connectivity and branching.

295

3.2.3. Cell metabolic activity The metabolic activity of ECs during network formation, determined by AlamarBlue™ reduction (Fig. 5b), was significantly decreased at day 1 and 3 in the presence of ionic release products from BG and, to a greater extent, Cu-BG. At day 5, in contrast, only Cu-BG had a significant effect on metabolic activity.

303

Ò

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 Bioglass on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

293 294

296 297 298 299 300 301 302

304 305 306 307 308

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

5

Fig. 4. (a) Total network length per unit volume in response to ionic release products from BG and Cu-BG as well as a CuCl2 dose equivalent to copper ions released from 6.4 mg/mL Cu-BG (Cu eq). (b) Expansion of the y-scale at day 1. The network length was significantly reduced at all time points by 1.6 and 6.4 mg/mL Cu-BG and BG, but not by CuCl2. (c) Cumulative length distribution of discrete network structures at day 1 and 3. The proportions of long structures increased over time as EC structures merged into larger networks, while more short structures were present with 6.4 mg/mL of both BG and Cu-BG. (The length distribution at day 5 in culture as well as the average length values are presented in Fig. S2). (d) The number of nodes per network length increased over time and was significantly reduced by BG and Cu-BG, but not by CuCl2. (Error bars: SD, n = 6).

Fig. 5. (a) The number of EC nuclei per unit volume as well as (b) EC metabolic activity increased over time and were significantly reduced by ionic release products from both BG and Cu-BG, but not by CuCl2. (Error bars: SD, n = 6 and n = 4 for (a) and (b) respectively.)

4. Discussion

309

45S5 BioglassÒ is a clinically successful bone regenerative material and has recently been shown to possess pro-angiogenic activity [12,17]. Moreover, bioactive glasses provide a promising material for the sustained release of copper ions, which may further improve angiogenic stimulation if supplied at an optimal concentration [22,32]. Therefore, this study examined the effect of copper-doped 45S5 BioglassÒ on 3D capillary-like network formation by endothelial cells in type I collagen matrices. BG and Cu-BG dissolution in DMEM resulted in copper and silicon release, but with a calcium decrease to levels lower than initially present in DMEM. Calcium consumption is likely a consequence of a formation of a calcium phosphate precipitate on the glass surface. Specifically, the formation of an initially amorphous calcium phosphate layer is commonly observed upon BioglassÒ exposure to biological fluid [43] and has been indicated by Fourier transform infrared spectroscopy following one day of dissolution of the present 45S5 BioglassÒ powders in simulated body fluid [44]. Interestingly, only higher quantities of BG and Cu-BG led to calcium consumption, while slight calcium release, above the DMEM concentration, was observed for 0.4 and 1.6 mg/mL Cu-BG. This finding may be explained by the reciprocal relationship between dissolution reactions and pH changes. Specifically, BioglassÒ dissolution results in a pH increase due to immediate ion exchange between H+ and glass modifiers, especially labile Na+ ions [45,46]. A higher pH, in turn, favors the precipitation of a calcium phosphate surface layer [45]. Therefore, only conditions above a threshold concentration of approximately 1.6 mg/mL may have resulted in a high enough pH to enable calcium phosphate deposition and thus calcium consumption. Previously, positive calcium release has been shown during the dissolution of BioglassÒ particles in the 300–700 lm range in DMEM [10,11]. However, slight calcium depletion due to

310

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 6 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406

C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

BioglassÒ dissolution in DMEM has also been observed and was more prominent with smaller particles (in the herein used 10 lm range) [47], which is in line with the observation that a higher surface area leads to a higher rate of calcium phosphate precipitation [48]. Copper release increased with glass concentration up to 6.4 mg/ mL but remained approximately constant at higher glass concentrations. Moreover, only a minimal increase in both copper and silicon release was observed between 24 and 72 h dissolution. Similarly, copper and silicon release from Cu-BG derived scaffolds were previously shown to reach maximum levels after approximately 7 days dissolution [36]. Since silicate is a network former, silicon release provides a measure for the overall extent of glass dissolution. Therefore, Cu-BG dissolution under the present conditions appears to be terminated within the first few days, which is likely due to a reduction in reaction rates with higher solution concentrations and/or the formation of a calcium phosphate surface layer acting as a diffusion barrier. Silicate release as well as calcium consumption were higher in Cu-BG compared to BG. This finding is possibly due to a weakening of the glass network, and thus a higher reactivity of Cu-BG, as indicated by a lower glass transition temperature [36]. The quantification of dynamic morphological changes by ECs in a 3D ECM substitute has been widely applied to examine pro- or anti-angiogenic stimuli. In particular, in vitro-reconstituted hydrogels of type I collagen, the most abundant ECM component, enable the formation of capillary-like EC networks that resemble the initial stages of in vivo neovascularization [49,50]. An automated 3D image analysis method, based on previously reported techniques [51,52], was recently developed [37] in order to quantify morphological parameters of EC networks including length, connectivity and branching, which are critical to an ultimately functional blood supply. In contrast to 2D image processing methods, this technique applied 3D CLSM image stacks and therefore provided a more accurate network quantification by taking into account the vertical component of tubular structures and by separating structures positioned on top of each other. In this study, EC network length, connectivity and branching were reduced in the presence of ionic release products from increasing amounts of Cu-BG as well as by BG, but were not altered by a copper dose equivalent to the highest Cu-BG concentration. This trend was closely matched by EC metabolic activity and proliferation, which play an important role in neovascularization [53]. However, at day 1, while both network length and metabolic activity were significantly reduced by the highest glass concentration, nuclei number was not affected. Adverse effects on ECs are not expected to be reflected in proliferation after only one day in culture. Conversely, at day 5, network length and nuclei number were reduced by 6.4 mg/mL BG although the corresponding metabolic activity was not altered. This discrepancy may be due to the high cell density in non-glass treated cultures, possibly combined with a depletion in nutrients, thus resulting in a lower metabolic activity per cell. In general, the observed adverse effects on ECs were not due to copper ion release, but reflected the extent of silicon release and calcium depletion in the medium. Specifically, silicon release, calcium consumption as well as all examined measures of cellular response resulting from 6.4 mg/mL BG were in between those of 6.4 and 1.6 mg/mL Cu-BG. Further, pH changes resulting from the altered silicon and calcium ionic concentrations may have affected EC functions. An increase in pH upon 45S5 BioglassÒ dissolution has previously been correlated with a reduction in EC proliferation [54]. The present study only examined the EC response to ionic release products after a 24 h dissolution. Nevertheless, the complementary ion release data suggest that ionic release products from

higher glass concentrations would further reduce EC functions while ions from 72 h dissolution would have an only slightly different effect compared to 24 h dissolution. Although the lowest Cu-BG concentration investigated in this study showed no adverse effects on ECs, no positive stimulation on ECs was observed. The pro-angiogenic potential of BioglassÒ may not be reflected in EC cultures, but is likely related to the up-regulation of angiogenic factors in other cell types such as fibroblasts or osteoblasts, as previously demonstrated in co-culture models [12,14,55]. In addition, Cu2+ ions in the range of 10 ppm did not alter EC network parameters in the present model, but were previously shown to up-regulate human umbilical vein EC (HUVEC) proliferation and the formation of tubular structures [19,22]. However, adverse effects of the same concentrations of Cu2+ on HUVECs have also been observed [56,57]. These conflicting results may be related to the role of culture medium composition on copper-mediated angiogenic or cytotoxic effects [57]. Furthermore, a potential role of MMP production, which is required for the digestion of type I collagen and ECM invasion and which was shown to be regulated by copper ions [26,28,58– 60], should be examined in future studies. In this work, 3D EC network formation was investigated in the presence of constant concentrations of BG and Cu-BG ionic release products as well as a CuCl2 concentration equivalent to copper release from Cu-BG, which allowed for the distinction of coppermediated effects from the effects of other ions. Specifically, none of the adverse effects of Cu-BG were attributable to copper ions. Moreover, 0.4 mg/mL Cu-BG exhibited no harmful effects and yielded a copper dose slightly above the physiological level in human blood plasma (1.5 ppm [61,62]). This range of Cu-BG concentration should therefore be examined in future studies for potential pro-angiogenic activity, possibly reflected in the expression of angiogenic factors from cell types other than endothelial cells. Nevertheless, the incorporation of more than 2.5 wt.% CuO into BioglassÒ may be suitable in order to exploit the full angiogenic potential of copper without inducing cytotoxic effects resulting from glass dissolution. Finally, the quantification of ion release in this work provides a basis for future models applying direct EC exposure to simultaneously dissolving BioglassÒ particles.

407

5. Conclusions

446

This study offered a detailed examination of the effects of ionic release products from copper-doped and non-doped 45S5 BioglassÒ on 3D EC network formation in type I collagen matrices. Ion quantification following BioglassÒ dissolution revealed the consumption of calcium ions present in cell culture medium, likely due to the formation of a calcium phosphate layer. Copper and silicon release as well as calcium consumption showed a non-linear increase with glass concentration in medium and with dissolution time, while BG resulted in lower silicon release and calcium consumption compared to an identical quantity of Cu-BG. Higher amounts of silicon release and calcium consumption resulting from both BG and Cu-BG dissolution were associated with a reduced EC network length, connectivity, branching as well as proliferation and metabolic activity, whereas a Cu2+ ion dose equivalent to 6.4 mg/mL Cu-BG had no effect on ECs. Previously reported pro-angiogenic activity of both copper and 45S5 BioglassÒ was not observed in the current model and is therefore likely attributable to the expression of angiogenic factors from cell types other than endothelial cells. The lowest Cu-BG concentration investigated in this study (0.4 mg/mL) exhibited no adverse effects on ECs and should therefore be examined in future studies, possibly applying co-culture models, while the doping of BioglassÒ with more than 2.5 wt.% CuO should be considered.

447

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx 470

Acknowledgements

471

481

This study was supported by the Canadian Natural Sciences and Engineering Research Council, Canada Foundation for Innovation: Leaders Opportunity Funds, Québec Ministère de l’Enseignement supérieur, de la Recherche, de la Science et de la Technologie (MESRST) and McGill University Faculty of Engineering Hatch Faculty Fellowship. Partial funding from the Bavarian-QuébecProject ‘‘Mesenchymal stem cell seeded nanocomposite constructs for bone tissue engineering’’ through the Québec MESRST and the Bavarian Research Alliance is gratefully acknowledged. The authors thank Ranjan Roy and Andrew Golsztajn for their assistance in this work.

482

Appendix A. Figures with essential color discrimination

483

486

Certain figures in this article, particularly Figs. 1–3 are difficult to interpret in black and white. The full color images can be found in the on-line version, at http://dx.doi.org/10.1016/j.actbio.2015. 03.009.

487

Appendix B. Supplementary data

488 490

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2015.03. 009.

491

References

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539

[1] Jain RK, Au P, Tam J, Duda DG, Fukumura D. Engineering vascularized tissue. Nat Biotechnol 2005;23:821–3. [2] Hopkins SP, Bulgrin JP, Sims RL, Bowman B, Donovan DL, Schmidt SP. Controlled delivery of vascular endothelial growth factor promotes neovascularization and maintains limb function in a rabbit model of ischemia. J Vasc Surg 1998;27:886–95. [3] Hoppe A, Güldal NS, Boccaccini AR. A review of the biological response to ionic dissolution products from bioactive glasses and glass-ceramics. Biomaterials 2011;32:2757–74. [4] Habibovic P, Barralet JE. Bioinorganics and biomaterials: bone repair. Acta Biomater 2011;7:3013–26. [5] Gbureck U, Holzel T, Doillon CJ, Muller FA, Barralet JE. Direct printing of bioceramic implants with spatially localized angiogenic factors. Adv Mater 2007;19:795–800. [6] Hench LL. The story of BioglassÒ. J Mater Sci Mater Med 2006;17:967–78. [7] Hench LL. Bioceramics – from concept to clinic. J Am Ceram Soc 1991;74:1487–510. [8] Hench LL, Greenspan D. Interactions between bioactive glass and collagen: a review and new perspectives. J Austr Ceram Soc 2013;49:1–40. [9] Hench LL. Genetic design of bioactive glass. J Eur Ceram Soc 2009;29:1257–65. [10] Xynos ID, Edgar AJ, Buttery LDK, Hench LL, Polak JM. Ionic products of bioactive glass dissolution increase proliferation of human osteoblasts and induce insulin-like growth factor II mRNA expression and protein synthesis. Biochem Biophys Res Commun 2000;276:461–5. [11] Xynos ID, Edgar AJ, Buttery LDK, Hench LL, Polak JM. Gene-expression profiling of human osteoblasts following treatment with the ionic products of BioglassÒ 45S5 dissolution. J Biomed Mater Res 2001;55:151–7. [12] Day RM. Bioactive glass stimulates the secretion of angiogenic growth factors and angiogenesis in vitro. Tissue Eng 2005;11:768–77. [13] Gerhardt L-C, Widdows KL, Erol MM, Burch CW, Sanz-Herrera JA, Ochoa I, et al. The pro-angiogenic properties of multi-functional bioactive glass composite scaffolds. Biomaterials 2011;32:4096–108. [14] Leu A, Leach JK. Proangiogenic potential of a collagen/bioactive glass substrate. Pharm Res 2008;25:1222–9. [15] Gorustovich AA, Roether JA, Boccaccini AR. Effect of bioactive glasses on angiogenesis: a review of in vitro and in vivo evidences. Tissue Eng Part B 2010;16:199–207. [16] Day RM, Maquet V, Boccaccini AR, Jérôme R, Forbes A. In vitro and in vivo analysis of macroporous biodegradable poly(D,L-lactide-co-glycolide) scaffolds containing bioactive glass. J Biomed Mater Res A 2005;75A:778–87. [17] Leach JK, Kaigler D, Wang Z, Krebsbach PH, Mooney DJ. Coating of VEGFreleasing scaffolds with bioactive glass for angiogenesis and bone regeneration. Biomaterials 2006;27:3249–55. [18] Arkudas A, Balzer A, Buehrer G, Arnold I, Hoppe A, Detsch R, et al. Evaluation of angiogenesis of bioactive glass in the arteriovenous loop model. Tissue Engineering – Part C 2013;19:479–86. [19] Hu GF. Copper stimulates proliferation of human endothelial cells under culture. J Cell Biochem 1998;69:326–35.

472 473 474 475 476 477 478 479 480

484 485

489

7

[20] Li S, Xie H, Li S, Kang YJ. Copper stimulates growth of human umbilical vein endothelial cells in a vascular endothelial growth factor-independent pathway. Exp Biol Med 2012;237:77–82. [21] Giavaresi G, Torricelli P, Fornasari PM, Giardino R, Barbucci R, Leone G. Blood vessel formation after soft-tissue implantation of hyaluronan based hydrogel supplemented with copper ions. Biomaterials 2005;26:3001–8. [22] Gérard C, Bordeleau LJ, Barralet J, Doillon CJ. The stimulation of angiogenesis and collagen deposition by copper. Biomaterials 2009;31:824–31. [23] Barralet J, Gbureck U, Habibovic P, Vorndran E, Gerard C, Doillon CJ. Angiogenesis in calcium phosphate scaffolds by inorganic copper ion release. Tissue Eng Part A 2009;15:1601–9. [24] Feng W, Ye F, Xue W, Zhou Z, Kang YJ. Copper regulation of hypoxia-inducible factor-1 activity. Mol Pharmacol 2009;75:174–82. [25] Sen CK, Khanna S, Venojarvi M, Trikha P, Ellison EC, Hunt TK, et al. Copperinduced vascular endothelial growth factor expression and wound healing. Am J Physiol Heart Circ Physiol 2002;282:H1821–7. [26] Philips N, Hwang H, Chauhan S, Leonardi D, Gonzalez S. Stimulation of cell proliferation and expression of matrixmetalloproteinase-1 and interluekin-8 genes in dermal fibroblasts by copper. Connect Tissue Res 2010;51:224–9. [27] Simeon A, Emonard H, Hornebeck W, Maquart FX. The tripeptide-copper complex glycyl-L-histidyl-L-lysine-Cu2+ stimulates matrix metalloproteinase2 expression by fibroblast cultures. Life Sci 2000;67:2257–65. [28] Elsherif L, Jiang YC, Saari JT, Kang YJ. Dietary copper restriction-induced changes in myocardial gene expression and the effect of copper repletion. Exp Biol Med 2004;229:616–22. [29] Adamson IYR, Vincent R, Bakowska J. Differential production of metalloproteinases after instilling various urban air particle samples to rat lung. Exp Lung Res 2003;29:375–88. [30] Gaetke LM, Chow CK. Copper toxicity, oxidative stress, and antioxidant nutrients. Toxicology 2003;189:147–63. [31] Hung YH, Bush AI, Cherny RA. Copper in the brain and Alzheimer’s disease. J Biol Inorg Chem 2010;15:61–76. [32] Wu C, Zhou Y, Xu M, Han P, Chen L, Chang J, et al. Copper-containing mesoporous bioactive glass scaffolds with multifunctional properties of angiogenesis capacity, osteostimulation and antibacterial activity. Biomaterials 2013;34:422–33. [33] Kong N, Lin K, Li H, Chang J. Synergy effects of copper and silicon ions on stimulation of vascularization by copper-doped calcium silicate. J Mater Chem B 2014;2:1100–10. [34] Zhao S, Wang H, Zhang Y, Huang W, Rahaman MN, Liu Z, et al. Copper-doped borosilicate bioactive glass scaffolds with improved angiogenic and osteogenic capacity for repairing osseous defects. Acta Biomater 2015;14:185–96. [35] Wang H, Zhao SC, Zhou J, Shen YQ, Huang WH, Zhang CQ, et al. Evaluation of borate bioactive glass scaffolds as a controlled delivery system for copper ions in stimulating osteogenesis and angiogenesis in bone healing. J Mater Chem B 2014;2:8547–57. [36] Hoppe A, Meszaros R, Stähli C, Romeis S, Schmidt J, Peukert W, et al. In vitro reactivity of Cu doped 45S5 BioglassÒ derived scaffolds for bone tissue engineering. J Mater Chem B 2013;1:5659–74. [37] Stähli C, James-Bhasin M, Nazhat SN. Three-dimensional endothelial cell morphogenesis under controlled ion release from copper-doped phosphate glass. J Controlled Release 2015;200:222–32. [38] Bolte S, Cordelieres FP. A guided tour into subcellular colocalization analysis in light microscopy. J Microsc – Oxford 2006;224:213–32. [39] Zhang TY, Suen CY. A fast parallel algorithm for thinning digital patterns. Commun ACM 1984;27:236–9. [40] Arganda-Carreras I, Fernández-González R, Muñoz-Barrutia A, Ortiz-DeSolorzano C. 3D reconstruction of histological sections: application to mammary gland tissue. Microsc Res Tech 2010;73:1019–29. [41] Barthel, KU. Volume Viewer, Available from: [Last accessed: 2014[. [42] Herbert, A. GDSC ImageJ Plugins, Available from: [Last accessed: 2014[. [43] Hench LL, Polak JM, Xynos ID, Buttery LDK. Bioactive materials to control cell cycle. Mater Res Innovations 2000;3:313–23. [44] Mackovic M, Hoppe A, Detsch R, Mohn D, Stark WJ, Spiecker E, et al. Bioactive glass (type 45S5) nanoparticles: in vitro reactivity on nanoscale and biocompatibility. J Nanopart Res 2012;14. [45] Cerruti M, Greenspan D, Powers K. Effect of pH and ionic strength on the reactivity of BioglassÒ 45S5. Biomaterials 2005;26:1665–74. [46] DeAza PN, Guitian F, Merlos A, LoraTamayo E, DeAza S. Bioceramics – simulated body fluid interfaces: pH and its influence of hydroxyapatite formation. J Mater Sci Mater Med 1996;7:399–402. [47] Sepulveda P, Jones JR, Hench LL. In vitro dissolution of melt-derived 45S5 and sol-gel derived 58S bioactive glasses. J Biomed Mater Res 2002;61:301–11. [48] Cerruti MG, Greenspan D, Powers K. An analytical model for the dissolution of different particle size samples of BioglassÒ in TRIS-buffered solution. Biomaterials 2005;26:4903–11. [49] Goodwin AM. In vitro assays of angiogenesis for assessment of angiogenic and anti-angiogenic agents. Microvasc Res 2007;74:172–83. [50] Kamei M, Saunders WB, Bayless KJ, Dye L, Davis GE, Weinstein BM. Endothelial tubes assemble from intracellular vacuoles in vivo. Nature 2006;442:453–6. [51] Fuchs S, Jiang X, Schmidt H, Dohle E, Ghanaati S, Orth C, et al. Dynamic processes involved in the pre-vascularization of silk fibroin constructs for bone regeneration using outgrowth endothelial cells. Biomaterials 2009;30:1329–38.

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625

ACTBIO 3621

No. of Pages 8, Model 5G

13 March 2015 8 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643

C. Stähli et al. / Acta Biomaterialia xxx (2015) xxx–xxx

[52] Peters K, Schmidt H, Kamp G, Unger RE, Hoffmann B, Stenzel F, et al. Softwaresupported quantification of angiogenesis in an in vitro culture system. Examples of applications in studies of basic research, biocompatibility and drug discovery. In: Zubar RV, editor. Trends in angiogenesis research. New York: Nova Biomedical Books; 2005. p. 103–23. [53] Carmeliet P. Mechanisms of angiogenesis and arteriogenesis. Nat Med 2000;6:389–95. [54] Aina V, Malavasi G, Pla AF, Munaron L, Morterra C. Zinc-containing bioactive glasses: surface reactivity and behaviour towards endothelial cells. Acta Biomater 2009;5:1211–22. [55] Keshaw H, Forbes A, Day RM. Release of angiogenic growth factors from cells encapsulated in alginate beads with bioactive glass. Biomaterials 2005;26:4171–9. [56] Yu HS, Pastor SA, Lam KW, Yee RW. Ascorbate-enhanced copper toxicity on bovine corneal endothelial-cells in vitro. Curr Eye Res 1990;9:177–82. [57] Stähli C, Muja N, Nazhat SN. Controlled copper ion release from phosphatebased glasses improves human umbilical vein endothelial cell survival in a reduced nutrient environment. Tissue Eng Part A 2013;19:548–57.

[58] Juin A, Billottet C, Moreau V, Destaing O, Albiges-Rizo C, Rosenbaum J, et al. Physiological type I collagen organization induces the formation of a novel class of linear invadosomes. Mol Biol Cell 2012;23:297–309. [59] Ohuchi E, Imai K, Fujii Y, Sato H, Seiki M, Okada Y. Membrane type 1 matrix metalloproteinase digests interstitial collagens and other extracellular matrix macromolecules. J Biol Chem 1997;272:2446–51. [60] Schmidt-Hansen B, Ornas D, Grigorian M, Klingelhofer J, Tulchinsky E, Lukanidin E, et al. Extracellular S100A4(mts1) stimulates invasive growth of mouse endothelial cells and modulates MMP-13 matrix metalloproteinase activity. Oncogene 2004;23:5487–95. [61] Abiaka C, Olusi S, Al-Awadhi A. Reference ranges of copper and zinc and the prevalence of their deficiencies in an Arab population aged 15–80 years. Biol Trace Elem Res 2003;91:33–43. [62] Sanchez C, Lopez-Jurado M, Aranda P, Llopis J. Plasma levels of copper, manganese and selenium in an adult population in southern Spain: influence of age, obesity and lifestyle factors. Sci Total Environ 2010;408: 1014–20.

Please cite this article in press as: Stähli C et al. Effect of ion release from Cu-doped 45S5 BioglassÒ on 3D endothelial cell morphogenesis. Acta Biomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.009

644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661