Direct Deposited Porous Scaffolds of Calcium Phosphate Cement With Alginate

1

2

3

4

567

8

1 0

1112131415

16171819202122

2 3

46

47

48

49

50

51

52

53

54

55

56

57

Acta Biomaterialia xxx (2011) xxx–xxx

ACTBIO 1725 No. of Pages 9, Model 5G

30 April 2011

Contents lists available at ScienceDirect

Acta Biomaterialia

journal homepage: www.elsevier .com/locate /actabiomat

Direct deposited porous scaffolds of calcium phosphate cement with alginatefor drug delivery and bone tissue engineering

Gil-Su Lee a,b, Jeong-Hui Park a,b, Ueon Sang Shin b, Hae-Won Kim a,b,c,⇑a Department of Nanobiomedical Science and WCU Research Center, Dankook University Graduate School, Yongin, South Koreab Institute of Tissue Regeneration Engineering, Dankook University, Yongin, South Koreac Department of Biomaterials Science, School of Dentistry, Dankook University, Yongin, South Korea

a r t i c l e i n f o a b s t r a c t

242526272829303132333435

Article history:Received 23 November 2010Received in revised form 7 April 2011Accepted 12 April 2011Available online xxxx

Keywords:Porous scaffoldsSelf-setting cementsCalcium phosphatesProtein deliveryBone regeneration

363738394041424344

1742-7061/$ - see front matter � 2011 Acta Materialdoi:10.1016/j.actbio.2011.04.008

⇑ Corresponding author at: Department of NanobResearch Center, Dankook University Graduate Schoo+82 41 550 1926.

E-mail address: [email protected] (H.-W. Kim).

Please cite this article in press as: Lee G-S et al.tissue engineering. Acta Biomater (2011), doi:1

This study reports the preparation of novel porous scaffolds of calcium phosphate cement (CPC) com-bined with alginate, and their potential usefulness as a three-dimensional (3-D) matrix for drug deliveryand tissue engineering of bone. An a-tricalcium phosphate-based powder was mixed with sodium algi-nate solution and then directly injected into a fibrous structure in a Ca-containing bath. A rapid hardeningreaction of the alginate with Ca2+ helps to shape the composite into a fibrous form with diameters of hun-dreds of micrometers, and subsequent pressing in a mold allows the formation of 3-D porous scaffoldswith different porosity levels. After transformation of the CPC into a calcium-deficient hydroxyapatitephase in simulated biological fluid the scaffold was shown to retain its mechanical stability. Duringthe process biological proteins, such as bovine serum albumin and lysozyme, used as model proteins,were observed to be effectively loaded onto and released from the scaffolds for up to more than a month,proving the efficacy of the scaffolds as a drug delivering matrix. Mesenchymal stem cells (MSC) were iso-lated from rat bone marrow and then cultured on the CPC–alginate porous scaffolds to investigate theability to be populated by cells and their subsequent differentiation along the osteogenic lineage. Itwas shown that MSC increasingly actively populated and also permeated into the porous network withtime of culture. In particular, cells cultured within a scaffold with a relatively high porosity level showedfavorable proliferation and osteogenic differentiation. An in vivo pilot study of the CPC–alginate porousscaffolds after implantation into the rat calvarium for 6 weeks revealed the formation of new bone tissuewithin the scaffold, closing the defect almost completely. Based on these results, the newly developedCPC–alginate porous scaffolds could be potentially useful as a 3-D matrix for drug delivery and tissueengineering of bone.

� 2011 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

45

58

59

60

61

62

63

64

65

66

67

68

69

1. Introduction

Rapid setting cements are very useful in bone tissue regenera-tion, as either a direct filling or an injectable material. Calciumphosphate cements (CPCs) have been one of the most widely stud-ied bioactive ceramics for this purpose [1,2]. Although some chal-lenges, such as the mechanical properties, the introduction ofmacropores and control of the dissolution rate, remain to be im-proved, many fascinating properties of CPCs make them a usefulchoice in the treatment of bone defects [2,3]. CPCs have been foundto be cell and tissue compatible, and they self-set, making themuseful as an injectable material requiring minimally invasive sur-

70

71

72

73

74

75

ia Inc. Published by Elsevier Ltd. A

iomedical Science and WCUl, Yongin, South Korea. Tel.:

Direct deposited porous scaffo0.1016/j.actbio.2011.04.008

gery, and, in addition, they can carry therapeutic molecules withinthe formulation [4,5].

Scaffolds with a three-dimensional (3-D) porous network pro-vide effective matrix conditions for bone tissue engineering [6–8].Tissue cells are ex vivo cultured with the scaffolds to better mimicthe structure and function of native tissues than the materials orcells alone [8,9]. In the course of ex vivo engineering of tissues thecontrolled release of therapeutic molecules such as growth factorsis favored, to modulate cellular function and speed up boneformation. CPC-based materials have also been considered goodcandidates for the delivery of therapeutics carried within theirstructure, because they self-hard under mild conditions, with thetherapeutics being safely incorporated, and retain a sustainable re-lease profile [4]. To apply CPC-based materials to bone tissue engi-neering their development as 3-D scaffolds which support cellproliferation and cell–material composite construction is necessary.

To this end we here aim to develop a novel cell scaffoldingmaterial made of CPCs in combination with sodium alginate. In

ll rights reserved.

lds of calcium phosphate cement with alginate for drug delivery and bone

http://dx.doi.org/10.1016/j.actbio.2011.04.008

mailto:[email protected]


http://www.sciencedirect.com/science/journal/17427061

http://www.elsevier.com/locate/actabiomat


76

77

78

79

80

81

82

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

123

124

125

126

127

128

129

130

131

132

133

2 G.-S. Lee et al. / Acta Biomaterialia xxx (2011) xxx–xxx


30 April 2011

particular, a fibrous network was formulated by directly depositingthe composite suspension under a Ca-containing solution.The deposited suspension rapidly sets to form a gelled networkin the presence of alginate and is cross-linked by Ca2+ ions[10,11]. The hardened porous scaffold is considered to be cell com-patible and useful for bone tissue engineering. Moreover, the scaf-fold is considered to be able to load and deliver bioactive moleculescontained within the structure. The processing techniques to de-velop the CPC–alginate porous scaffolds are described and thein vitro cellular responses of mesenchymal stem cells (MSC) fromrat bone marrow to them have been investigated, prior to theirapplication in bone tissue engineering. An in vivo pilot study wasalso performed to evaluate tissue compatibility, and the drug deliv-ery potential of the scaffold was assessed using two different mod-el proteins.

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

Fig. 1. Schematic drawing of the processing set-up to prepare calcium phosphatecement (CPC)/alginate composite (CPA) fibrous network for use as a 3-D tissueengineering scaffold. CPC–alginate suspension was injected through a needle usingan air pump regulator into a cylindrical mold containing CaCl2 (150 mM) whichpromotes rapid setting of the composite suspension.

2. Materials and methods

2.1. Preparation of the composite suspension

The experimental a-tricalcium phosphate (a-TCP)-based ce-ment powder was prepared as described in a previous report[12]. Commercial calcium carbonate and anhydrous dicalciumphosphate (both from Aldrich) were mixed and thermally reactedat 1400 �C for 3 h, then air quenched, which resulted in completereaction to form the a-TCP phase [12]. The powders were ballmilled and sieved down to 45 lm, and then kept under vacuumfor further use. The average particle size of the a-TCP was4.79 lm, measured using a particle size analyzer (Saturn DigiSizer5200, Micromeritics, USA). Sodium alginate (Aldrich) solution wasprepared in 5% Na2HPO4 (in distilled, deionized water) at a concen-tration of 2 wt.%. The cement powder was mixed with the alginatesolution at an appropriate ratio to prepare the composite suspen-sion for dispensation through a nozzle.

2.2. Direct dispensation into scaffolds

The possibility of the direct deposition of the composite suspen-sions was examined by varying the mixing ratio of cement powder/alginate solution (from 1.0 to 2.5 by weight). Above a ratio of 2.0the suspension was too viscous to dispense through the nozzle,therefore ratios of 1.0–2.0 were used for the experiments. Themixed suspension was placed in a syringe and then dispensed intoa Ca-containing bath (150 mM CaCl2) in order to rapidly solidifythe deposit, as schematically illustrated in Fig. 1. The dispensationpressure was adjusted to 500 kPa using a regulator (IEI, AD2000C).The size was controlled by means of different needle gauges (23–27 G). After dispensing the materials into 10 ml of Ca-containingsolution within a cylindrical mold (u = 10 mm) the fibrous depos-its were further pressed down manually to produce a disc-shapedscaffold of specific height. The process of depositing scaffolds with-in the Ca-containing bath took 1 min. The height of the scaffoldwas varied in order to obtain different levels of porosity (1.2 mmfor low, 1.5 mm for medium and 2.0 mm for high porosity). Like-wise, the amount (weight) of scaffold material to be dispensedwas varied (0.5 g for low, 0.4 g for medium and 0.3 g for highporosity) while the height of the scaffolds was kept constant(3 mm) to give different levels of porosity. The as-hardened scaf-folds were used for further in vitro cell assays and in vivo animalstudies without further treatment, such as soaking in water.

2.3. Characterization of 3-D porous scaffolds

The composite scaffolds obtained were thoroughly washed indistilled water and then immersed in simulated body fluid (SBF

Please cite this article in press as: Lee G-S et al. Direct deposited porous scaffotissue engineering. Acta Biomater (2011), doi:10.1016/j.actbio.2011.04.008

containing 284.0 mM Na+, 10 mM K+, 3.0 mM Mg2+, 5.0 mM Ca2+,295.6 mM Cl–, 8.4 mM HCO�3 , 2.0 mM HPO2�

4 , 1.0 mM SO2�4 ) at

37 �C for periods of up to 7 days. Samples were washed and driedunder vacuum and the morphology was examined by scanningelectron microscopy (SEM) (Hitachi S-3000H). Composition changewas monitored by energy dispersive spectroscopy (EDS) (BrukerSNE-3000 M) in a scanning electron microscope. An X-ray diffrac-tometer (Rigaku Ultima IV) was used to detect changes in the crys-talline phases of the scaffolds after them. The pore structure ofscaffolds with different porosities was analyzed by micro-com-puted tomography (lCT) (Skyscan model 1172). A disc (u10 � 3 mm) of each sample was placed with the top and bottomsurfaces parallel to the scanning plane. Scanning was with a11 Mp X-ray camera and 758 files were acquired with an imagepixel size of 19.92 lm. The surface charge of the a-TCP particleswas determined by measurement of the zeta potential (ZetasizerZEN3600, Malvern Instruments). The a-TCP particles were sieved(40 lm) and dispersed in distilled water at 1 mg ml–1 and the zetapotential measured at room temperature and pH 7.0 using a dis-posable capillary cell (DTS1060C) and Zetasizer software (v.6.20). The measurement was repeated on three different samples.

The elastic modulus of the scaffolds was measured by means ofdynamic mechanical analysis (DMA) (DMA25, Metravib, France).Samples with three different porosities were prepared with thedimensions 5 mm diameter � 10 mm height to which a dynamiccompression load was applied. The storage modulus of the sampleswas recorded. Three samples were tested for each group.

2.4. Assas of protein delivery capacity

Protein release from the CPC–alginate porous scaffold was as-sessed using bovine serum albumin (BSA) and lysozyme as themodel proteins. Loading of each protein was carried out in two dif-ferent ways: one was to add the protein to the alginate solutionand then mix this with CPC powder, which was subsequentlydeposited in a protein-containing porous scaffold (‘‘loading I’’);the other was to add the protein to the CPC suspension, whichwas incubated for 1 h with gentle agitation, and then the solutionwas mixed with alginate solution which was then deposited in aporous scaffold (‘‘loading II’’). Protein content in each scaffold sam-ple was set at 33.3 lg mg scaffold–1. 1 g of the protein-containingporous scaffold was used for the protein release test. This was



174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

G.-S. Lee et al. / Acta Biomaterialia xxx (2011) xxx–xxx 3


30 April 2011

based on a pilot study that showed that the deposition of 1.4 g ofthe suspension resulted in the production of 1 g (± 0.039 g) of finalscaffold sample. Each sample was immersed in 10 ml of phosphatebuffered saline (PBS), pH 7, and 37 �C for up to 28 days. At eachmeasurement time (1, 2, 3, 6 and 24 h, and 2, 3, 7, 10, 14, 21 and28 days) the scaffold was removed and the remaining medium as-sessed by the BCA (bicinchoninic acid) method. At each measure-ment point the medium was refreshed.

2.5. In vitro osteoblast culture and proliferation

For the cell response tests CPC–alginate scaffolds having threedifferent levels of porosity (low 13.6%, medium 34.0%, and high53.7%) were prepared. Mesenchymal stem cells (MSC) derivedfrom rat bone marrow were harvested from the femora and tibiaeof 5-week-old male rats [13]. The femora and tibiae were rapidlydissected and placed in a-minimal essential medium (a-MEM).The sectioned bone was treated with collagenase and dispase solu-tion for 30 min and then the bone marrow was flushed out andcentrifuged at 1500 r.p.m. The pellet was disrupted and culturedunder normal culture conditions, in a-MEM supplemented with10% fetal bovine serum, containing antibiotic/antimycotic solution(10,000 U penicillin, 10,000 lg streptomycin, and 25 lg amphoter-icin B/m, Gibco) at 37 �C in an atmosphere of 5% CO2/95% air. After5 days of culture, non-adherent cells were removed and supple-mented with new medium. Cells were maintained under normalculture conditions and underwent three passages before use forfurther in vitro assays.

A scaffold sample was placed into each well of 24-well plate anda suspension of 1 � 105 cells was seeded on each sample. Cellswere incubated for up to 14 days under the influence of osteogenicfactors (50 lg ml–1 ascorbic acid, 100 nM dexamethasone and10 mM b-glycerophosphate). Cell proliferation was then measuredusing the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphe-nyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) method. When theMTS reagent (tetrazolium salt) is applied to living cells it is reducedby the cells to a colored formazan product which is soluble in cul-ture medium. The quantity of formazan product, which is directlyproportional to the number of living cells, was measured at anabsorbance of 490 nm using an ELISA plate reader (iMARK, BioRad).Three replicate samples were tested by MTS assay.

2.6. Alkaline phosphatase determination

As an a priori index for in vitro osteogenic differentiation of theMSC during culture on the CPC–alginate composite scaffolds alka-line phosphatase (ALP) activity was determined. After culture for 7and 14 days in osteogenic medium the cell layer was harvested andtreated with 0.1% Triton X-100 cell lysis medium and further dis-rupted by sequential freezing and thawing. The total protein con-tent was assayed using a commercial DC protein assay kit(BioRad), and the aliquot of the reaction sample was determinedafter normalization to the total protein content. The ALP activityof the cells was determined using an ALP assay kit (procedureNo. ALP-10, Sigma) [13]. The p-nitrophenol produced in the pres-ence of ALP was determined by its absorbance at 405 nm. Threereplicate samples were tested for ALP activity.

Cell test data are represented as means ± standard deviation(SD), and statistical analysis was carried out by one-way analysisof variance (ANOVA). Statistical significance was considered atP < 0.05.

2.7. In vivo pilot study on bone compatibility

10-week-old male Sprague–Dawley rats were used for thein vivo study. The surgical protocol was in accordance with the


guidelines of the Animal Care and Use Committee of Dankook Uni-versity, South Korea. Animals were anesthetized by means of intra-muscular injection using ketamine (80 mg kg–1) and xylazine(10 mg kg–1). An incision was made in the anterior region of thecalvarium and a 5 mm diameter critical sized full thickness bonedefect was prepared using a trephine drill under continuous sterilesaline irrigation. For the in vivo test scaffolds with the dimensions5 mm diameter � 2 mm height were prepared using a differentsized mold and the amount of composite deposited was also ad-justed to produce scaffolds with high porosity (53.7%). The pre-pared scaffolds were implanted within the calvarium defect.Defects without implanted scaffolds were used as negative con-trols. Soft tissues were sutured to achieve primary closure. Sixweeks after implantation the animals were killed. The area of theoriginal surgical defect and the surrounding tissues were removeden bloc and fixed in 10% neutral formalin solution and then decal-cified. Tissues were embedded in a paraffin block and then serialsectioned using a microtome (Leica™). The 4–6 lm thickness sec-tions were mounted on microscope slides. Slides with tissue sec-tions were deparaffinized and hydrated through series of xyleneand alcohol. The tissue slides were stained with hematoxylin andeosin (H&E) and Masson’s trichrome (MT) and viewed under anoptical microscope for histological observation.

3. Results and discussion

3.1. Scaffold fabrication and morphology

Upon deposition of the mixture suspension into a Ca-containingbath rapid hardening occurred due to the reaction of sodium algi-nate with Ca2+ ions. Therefore, the as-deposited shape could bemaintained during the process, resulting in a 3-D network. In fact,without alginate in the CPC composite rapid hardening was notpossible, resulting in complete disintegration of the scaffolds dur-ing dispensation. With the powder to liquid ratios used here (1.0–2.0), necessary to allow injectability through the nozzle, the algi-nate-free CPC suspensions were observed not to harden, even afterseveral hours. However, this hurdle could be overcome by the useof alginate, due to its rapid crosslinking reaction with Ca2+ ions inthe deposition bath. Therefore, the addition of alginate to the CPCcomposite and the use of an appropriate powder to liquid ratio arethe essential processing considerations in terms of obtaining bothinjectability and hardenability.

The diameter of fibers could be tuned by changing the needlegauge and the composition of the suspension, as shown inFig. 2a. Here we could obtain fiber diameters in the range �200–600 lm with different gauge needles (23, 25, 26 and 27 G) andcompositions (CPA20 and CPA15). Fiber diameter decreased asthe ratio of CPC powder to alginate liquid was increased (from1.5 to 2.0) and the needle gauge increased (from 23 to 27 G, corre-sponding to a decrease in needle inner diameter from 0.32 to0.16 mm). The piled up fibrous network was further shaped intoa 3-D scaffold by applying a compressive load. By varying the levelof compression the porosity of the scaffolds was easily controlled.Here we varied the porosity of the composite scaffolds at low,medium and high levels.

In fact, compared with other types of bioceramics, there hasbeen little development of porous scaffolds based on CPC compos-ites. A recent review highlighted the importance of CPC-based scaf-folds as potential drug delivering systems because of their self-hardening property [14,15]. Some studies on scaffold fabricationincorporated biopolymer phases such as chitosan and poly(lacticacid)/alginate within CPCs and applied conventional scaffoldingmethods [6,16]. Compared with those previous works, the scaffoldsdeveloped here are produced by a novel methodology, direct depo-



296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

Fig. 2. (a) Fiber diameter variations (from �200–600 lm) with change in needle gauge (23, 25, 26 and 27 G) and the composition of the suspension (CPA20 and CPA15).Diameter decreases with increasing needle gauge (decreasing needle diameter) and increasing the ratio of CPC powder to alginate liquid. Further compression of the piled upfibrous network facilitated shaping into a 3-D scaffold, the porosity of which, depending on the compression level, was easily controlled. (b) Pore structure of the CPC–alginate3-D porous scaffolds revealed by micro-computed tomography (lCT). 3-D reconstructed and 2-D cross-section images are shown. The measured porosities were 13.6%, 34.0%,and 53.7%, respectively, for the low, medium and high porosity scaffolds. Porosity data are presented as means ± 1 standard deviation for three different samples.



30 April 2011

sition of suspensions and formation of 3-D structures. Using thisprocess the pore configuration, including stem size and porosity,are controllable, and the scaffolds can be formulated into complexshapes by packing appropriate amounts within a designed mold.Although here we randomly piled up the deposits to form a 3-Dstructure improvements to the process such as direct writing willbe possible to produce well-defined 3-D configurations. This re-mains a further interesting research area.

The pore structure of the 3-D composite scaffolds was revealedby lCT, as shown in Fig. 2b. In the case of the low porosity scaffold(porosity �14%) some pores appeared to be clogged due to com-pression. The medium porosity scaffold (porosity �34%) had great-er pore space and better pore interconnection. The pores in thehigh porosity scaffold (porosity �54%) were highly spaced andinterconnected, providing suitable 3-D pore channels for cellmigration and tissue perfusion [17,18].

The CPC–alginate 3-D porous scaffolds were immersed in simu-lated body fluid and the phase transformation investigated interms of surface microstructure and phase analysis. The micro-structure of the scaffolds after varying immersion times is shownin Fig. 3a. Before immersion (‘0d’) the surface was dense, withCPC particles embedded within the alginate matrix. After immer-sion for 1 day (‘1d’) some tiny crystallites started to form on thesurface. By day 3 (‘3d’) the crystalline phase grew to form an even


network of platelet-like crystallites. After 7 days (‘7d’) the forma-tion of nanocrystals was even greater, with micron sized islandshaving formed, which become layered and merged together. Thephase changes of CPC and the CPC–alginate scaffold were moni-tored by XRD during the immersion test. Initially only a-TCP peakswere noted (closed circles). With immersion HA appeared as a newphase (asterisks), the intensity of HA peaks increasing withincreasing immersion time, suggesting phase transformation ofa-TCP to HA [19]. By day 7 only HA phase was observed, suggestingcomplete transformation. The EDS analysis supported the phaseconversion of a-TCP to HA, as deduced from the change in Ca/P ra-tio, which increased from 1.517 (similar to a-TCP) to 1.659 (similarto HA). Based on this phase evolution of the CPC–alginate undersimulated body fluid conditions, the scaffold developed here isconsidered to retain good bioactivity and to provide favorable sub-strate conditions for bone-associated cells to grow and developinto tissue, as the scaffold in body fluid will transform into bonemineral like HA phase, which should play a significant role in bio-logical reactions [20,21].

While the CPC–alginate scaffolds showed pore configurationsand compositions that are favorable for use in tissue engineering,the mechanical properties of the scaffolds need to be improvedfor load-bearing applications. Although the scaffolds were shownto be relatively brittle, the elastic moduli of the scaffolds measured



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

Fig. 3. Immersion tests of the CPC–alginate 3-D porous scaffolds in simulated body fluid for different times. (a) SEM microstructure change before (0d) and after immersionfor days 1, 3 and 7 (days are noted in each image). Tiny crystallites started to appear on day 1, which grew considerably over day 3 and 7, with development of highly facetednanocrystallines. (b) XRD patterns of the scaffolds during the immersion tests. The initial a-TCP phase started to transform into HA phase on day 1, which continued todevelop with immersion time to reach almost complete transformation into HA at day 7. (c) EDS analysis of the Ca and P atomic composition of the transformed phase,showing the increment in Ca/P ratio from 1.517 to 1.659, supporting the phase transformation of a-TCP to HA.



30 April 2011

by dynamic mechanical analysis were in the range similar to thatof trabecular bone (96 ± 63 MPa for high, 398 ± 63 MPa for mediumand 573 ± 87 MPa for low porosity scaffolds versus 50–500 MPa fortrabecular bone) [22], suggesting possible application for boneregeneration, however, mainly in non-load-bearing areas. How-ever, considering the porosity of our scaffolds the values observedare relatively lower than those reported for HA-based bioceramicscaffolds [23–25]. This is mainly due to the intrinsic poor mechan-ical properties of the CPC-based materials. Further improvementsin the mechanical properties are thus needed for load-bearingapplications, which will be further study.

375

376

377

378

379

380

381

382

383

384

Fig. 4. Proliferation of mesenchymal stem cells (MSC) derived from rat bonemarrow upon the 3-D porous scaffolds with different porosity levels (low, mediumand high) during culture for 3, 7 and 14 days. Significant differences were noticedbetween the scaffolds at each culture time (⁄P < 0.05 vs. low porosity and #P < 0.05vs. medium porosity, ANOVA, n = 4).

3.2. MSC culture and osteogenic differentiation

To address the possibility of using the CPC–alginate porous scaf-folds for bone tissue engineering we first investigated the re-sponses of MSC derived from rat bone marrow to scaffolds withdifferent levels of porosity. 1 � 105 MSC were seeded on the scaf-folds and cultured for up to 14 days in osteogenic medium contain-ing ascorbic acid, dexamethasone and b-glycerophosphate. Cellgrowth was measured as mitochondrial activity of the cells (MTSassay), as shown in Fig. 4. MSC proliferated well on all three typesof scaffolds, showing an ongoing increase in MTS level with culturetime. At the initial time point of 3 days cell proliferation was signif-icantly higher in the high porosity scaffold compared with the lowand medium porosity scaffolds (P < 0.05), and this was maintainedfor up to 7 days. By day 14 the difference between the scaffoldswith medium and high porosity was reduced.

The cell morphology on the fibrous scaffolds was observed bySEM at different magnifications during culture for 7 and 14 days,as shown in Fig. 5 (low Fig. 5a, medium Fig. 5b and high porosityFig. 5c). Cells on day 7 showed an elongated morphology with goodadherence to the underlying fiber stems. By day 14 cells showed


more extensive cytoskeletal processes covering the stem surfacealmost completely. When we observed the cross-sectioned internalsurface large number of cells were found deep in the pore channels,particularly in the scaffolds with high porosity. Based on these re-sults for cell proliferation and morphology the MSCs were demon-strated to favor the underlying CPC–alginate matrices with goodadherence, active filopodial protrusions and an increase in numberwith prolonged culture.

Stem cell differentiation along the osteogenic lineage upon thecomposite scaffolds was investigated by determining ALP activity.



385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

Fig. 5. Morphology observed by SEM of cells grown on the CPC–alginate porousscaffolds with different levels of porosity during culture for 7 and 14 days: (a) lowporosity; (b) medium porosity; (c) high porosity. Cells on day 7 adhered well to theunderlying fibrous stems, showing a highly elongated cell shape with manyfilopodia. By day 14 the cells had proliferated, were in contact with each other andcovered the surface of stems almost completely. This cell growth and morphologywas similarly observed for all porous scaffolds. Moreover, many cells were foundinside the scaffold (deep in the pore channels), particularly in the samples with highporosity. Bar scale 100 lm.

Fig. 6. Specific alkaline phosphatase activity of cells cultured upon the 3-D porousscaffolds with different porosity levels (low, medium and high) during culture for 7,14 and 21 days. Significant differences were noticed between the scaffolds at eachculture time (⁄P < 0.05 vs. low porosity and #P < 0.05 vs. medium porosity, ANOVA,n = 4).



30 April 2011

Cells were cultured on the scaffolds with different levels of poros-ity for up to 21 days in osteogenic medium and the ALP level mea-sured, as shown in Fig. 6. The ALP activity showed ongoingincreases with culture time up to 21 days for all scaffolds. Thisincrement was greater for the scaffold with high porosity than itwas for the others. The results demonstrate that the MSC culturedupon the 3-D porous scaffolds were stimulated to differentiate


along the osteogenic lineage and it was greater in the scaffold withhigh porosity. The cell proliferation and ALP activity results suggestthe use of a scaffold with high porosity to obtain better in vitroMSC function upon the 3-D fibrous matrix, which will ultimatelybe useful for the production of ex vivo tissue engineered constructsfor bone tissue engineering.

The scaffold with high porosity is believed to provide goodspaces and substrate conditions for cells to migrate and proliferatethree-dimensionally, making it easier for cell movement and func-tion through the open spaces [17,18]. However, clear reasons whythe cells favor more open scaffolds cannot be elucidated from ourexperiments and results. The conditions used here are somewhatlimited in design, i.e. 2-D static culture conditions, so that the geo-metrical effects of the scaffolds cannot be fully considered. Thisultimately warrants further study on 3-D dynamic cultures, suchas flow perfusion systems. At least from these in vitro results onMSC behavior, the CPC–alginate porous scaffolds with high poros-ity are considered to have potential as a 3-D scaffold for bone tis-sue engineering. Based on the cellular compatibility tests webriefly performed an experiment on in vivo tissue responses tothe scaffolds, such as bone in-growth, as a pilot study.

3.3. In vivo pilot study

The CPC–alginate scaffold with high porosity was implanted ina rat calvarium with a critical size defect of u = 5 mm [25,26]. At6 weeks post-implantation tissue samples were harvested andexamined by lCT, as shown in Fig. 7. A first look at the implantedsamples with X-rays showed a weak image of a 2-D fibrous net-work in the scaffold compared with an almost clear image in theblank (negative control) (Fig. 7a). The 2-D lCT image showed thatthe defect region was fully filled in the scaffold sample, but re-mained almost unfilled in the control (Fig. 7b). The reconstructed3-D lCT images showed the 3-D structure of the porous scaffoldand bone in-growth within the defect region. Bone regenerationhardly occurred in the negative control, suggesting a critical sizebone defect [25–27]. On the other hand, in the scaffold samplenew bone in-growth could not be clearly differentiated from theremaining material.

Histological staining of explants containing the porous scaffoldat 6 weeks post-operation revealed the quality of tissue and boneformation. Fig. 8 shows (Fig. 8a and b) H&E and (Fig. 8c and d)



Fig. 7. (a) X-ray image and (b, c) micro-computed tomography (lCT) of the calvarium defect recovered 6 weeks after implantation of the CPC–alginate scaffold with highporosity, and a blank defect used as a negative control. (b) 2-D and (c, d) 3-D constructed images. The scaffold was implanted in a rat calvarium with a critical size defect of /= 5 mm, and at 6 weeks post-implantation tissue samples were harvested and examined by lCT. The defect region showed �100% coverage, associated with both the scaffoldand newly formed bone. Dotted bar scale 1 mm.

Fig. 8. Histological staining of the tissues generated by the scaffold with high porosity at 6 weeks post-operation: (a, b) hematoxylin and eosin (HE) stain; (c, d) Masson’strichrome (MT) stain at different magnifications. Arrows indicate defect margins (a). OB, old bone; NB, new bone; S, scaffold (b). Connective bone tissue filled the pore channelof the scaffold throughout the defect region (a), and newly formed tissue lined the fiber stems of the scaffold (b). Pale or dark blue areas stained with MT demonstrate theformation of bony tissue, the major bone extracellular matrix.



30 April 2011

Please cite this article in press as: Lee G-S et al. Direct deposited porous scaffolds of calcium phosphate cement with alginate for drug delivery and bonetissue engineering. Acta Biomater (2011), doi:10.1016/j.actbio.2011.04.008


432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

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

Fig. 9. Release profiles of the proteins BSA and lysozyme from the CPC–alginateporous scaffold. The protein loading method was varied in two ways: one was toadd the protein to the alginate solution and then mix this with CPC powder, whichwas subsequently deposited in a protein-containing porous scaffold (‘‘loading I’’);the other was to add the protein to the CPC suspension, which was incubated for 1 hwith gentle agitation, and then the solution was mixed with alginate solution whichwas then deposited in a porous scaffold (‘‘loading II’’). The protein content in eachscaffold was set at 33.3 lg mg scaffold–1. Release tests were carried out inphosphate-buffered saline, pH 7 (PBS), for up to 14 days. The protein-containingscaffolds were immersed in PBS. At each measurement time point the scaffold wasremoved and the remaining medium assessed by the BCA method. In the case ofBSA there was no significant difference in the release profile dependent on themethod of drug loading, however, the release rate of lysozyme was significantlyreduced after ‘loading II’ compared with ‘loading I’. ⁄P < 0.05 for lysozyme loading IIvs. lysozyme loading I and #P < 0.05 for lysozyme loading II vs. BSA loading II,ANOVA, n = 3).

Table 1Zeta potential measurement of the CPC powder and the proteins used in the study.

Sample CPC powder Lysozyme BSA

Zeta potential at pH 7 �18.14 2.53 �16.84



30 April 2011

MT staining at different magnifications. No observable inflamma-tory responses to or tissue rejection of the implanted scaffoldswere found. Connective bone tissue was shown to fill the porechannel of the scaffold throughout the defect region (Fig. 8a). Amagnified image showed newly formed tissue (dark red) liningthe fiber stems of the scaffold (pale red) (Fig. 8b). MT staining dem-onstrated the formation of bony tissue, with the extracellular ma-trix appearing pale or dark blue (Fig. 8c and d).

In the histological images MT staining was even found withinthe CPC–alginate scaffold framework, demonstrating possiblereplacement of the scaffold by cells and tissues. Although a largeportion of the scaffolds appeared not to be biodegraded duringthe 6 weeks of implantation in rat calvarium, the results supportpossible in vivo degradation of the composite, either CPC or algi-nate or both. In general, CPC derived from a-TCP have HA as themajor phase in vivo, so they are degraded very slowly due to thetransformed HA, remaining present sometimes over years in vivo[28]. On the other hand, alginate degradation is known to occurrelatively quickly, possibly in weeks to months even in vitro,depending on the alginate concentration and degree of cross-link-ing [29]. At least from this study, the CPC–alginate scaffolds devel-oped are considered to maintain their structural integrity for6 weeks in rat calvarium, although there were indications of possi-ble biodegradation. Although biodegradation should be controlledin concert with the rate of tissue regeneration, the use of CPC basedon brushite instead of HA may be favored when faster degradationis required [30,31].

The in vivo tissue responses in rat calvarium showed that theCPC–alginate porous scaffolds developed here have good tissuecompatibility and even a level of regenerative potential of bone tis-sue, suggesting potential as an implantable material for boneregeneration. Based on the cell and tissue compatibility, furtherstudy is needed into preparing constructs of stem cells in this novelporous scaffold and in vivo implantation of the constructs for bonetissue engineering. To obtain further stimulation of cellular re-sponses within the scaffolds we performed a feasibility study ofthe protein delivery potential of the scaffolds.

3.4. Protein delivery potential

BSA and lysozyme, which are negatively and positively charged,respectively, at pH 7 were used as model proteins [32]. The initialprotein loading was carried out by two different methods. One wasto add the protein in alginate solution and then mixed with CPCpowder, which was deposited into a porous scaffold. The othermethod was to add the protein to the CPC suspension which wasincubated for 1 h with gentle agitation and then the protein–CPCsuspension was mixed with alginate solution which was subse-quently deposited into a porous scaffold. The protein initiallyadded to each 1 g of scaffold sample was set at 33.3 mg.

Fig. 9 shows the release profiles of the proteins in PBS measuredfor up to 28 days. Initially (within 12 h) up to about 20–30% of thelysozyme and BSA was released for all loading conditions, whichmay be proteins loosely bound on the surface and in direct contactwith the solution. After this initial burst the release of both pro-teins was sustained at a reduced rate with time for up to 28 days.The release profiles of BSA did not vary much between the differentmethods of loading (closed squares vs. open squares). However, inthe case of lysozyme the release rate was greatly reduced when thelysozyme was first added to the CPC powder (‘‘loading II’’) com-pared with when the protein was first added to the alginate (‘‘load-ing I’’). While the release of lysozyme was higher than that of BSAfor the ‘‘loading I’’ method, the trend was reversed for the ‘‘loadingII’’ method. Therefore, it is considered that interaction of the pro-teins with the components of scaffold, particularly CPC, is different,i.e. CPC may better retard the release of lysozyme than that of BSA


due to a strong affinity or chemical bonding [33,34]. To elucidateone possible reason on this behavior we measured the surfacecharge of the a-TCP powder and the proteins by means of zeta po-tential measurement, shown in Table 1. While a-TCP and BSA werenegatively charged, lysozyme was highly positively charged at pH7. Based on this, it is considered that electrostatic attraction to theCPC powder was higher for lysozyme than BSA. Although alginateis also negatively charged and thus has some ionic interaction withlysozyme, its hydrogel nature may mean it has a more open struc-ture than CPC, allowing water permeation and providing paths forprotein release from the structure. Moreover, the different degra-dation behaviors of alginate and CPC, discussed above, could alsoaffect the rate of release of lysozyme. Based on this result it isanticipated that this CPC–alginate scaffold may be effective inthe delivery of positively charged growth factors, such as basicfibroblast growth factor [35]. Further investigation is needed onthis. These results on in vitro protein release demonstrate that bio-logical molecules are easily and safely loaded within the scaffoldand the potential role of the porous scaffolds for sustainable pro-tein release, particularly positively charged ones, for up to at leasta month.

The results shown here confirm that the CPC–alginate porousscaffolds are cell compatible and bioactive. The rapid settinghydrogel-like structure, together with its moldability to defectivebone sites, supports the benefit of CPC–alginate scaffolds as 3-D



521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552553554555556557558559560561562563564565566567568569570571572573574

575576577578579580581



30 April 2011

matrices for stem cells, recruiting them into osteogenic develop-ment. Ongoing studies on the use of CPC–alginate composite scaf-folds as cell delivery devices for bone tissue engineering areunderway, in combination with tissue cells and stimulatory pro-teins within the structure.

582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613

4. Conclusions

CPC combined with alginate solution was effectively formedinto a porous scaffold by means of fiber deposition into a Ca-con-taining solution. The CPC–alginate scaffolds self-hardened, weremoldable into various shapes and the porosity was tunable. Thescaffolds were shown to have favorable 3-D matrix characteristicsfor pre-osteoblastic cell adherence and proliferation, as well astheir differentiation into osteoblasts. Implantation of a scaffoldinto a rat calvarium defect provided evidence of tissue compatibil-ity and the ability for bone regeneration through the pore geome-try. Moreover, the scaffolds showed an ability to safely loadbiological proteins (BSA and lysozyme) during preparation and torelease them in vitro for over a month. CPC–alginate scaffoldscan be further developed into tissue engineered constructs deliver-ing biological molecules stimulating bone regeneration.

Acknowledgements

This work was supported by a Priority Research Centers Pro-gram (no. 2009-0093829) and a WCU Program (no. R31-10069)through the National Research Foundation funded by the Ministryof Education, Science and Technology.

614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649

Appendix A. Figures with essential colour discrimination

Certain figures in this article, particularly Figures 1–3, 8, and 9,are difficult to interpret in black and white. The full colour imagescan be found in the on-line version, at doi: 10.1016/j.actbio.2011.04.008.

References

[1] Brown WE, Chow LC. A new calcium phosphate setting cement. J Dent Res1983;62:672.

[2] Gruninger SE, Siew C, Chow LC, O’Young A, Ts’ao NK, Brown WE. Evaluation ofthe biocompatibility of a new calcium phosphate setting cement. J Dent Res1986;63:200.

[3] Bohner M. Reactivity of calcium phosphate cements. J Mater Chem2007;17:3980–6.

[4] Ginebra MP, Traykova T, Planell JA. Calcium phosphate cements: competitivedrug carriers for the musculoskeletal system? Biomater 2006;27(10):2171–7.

[5] Blom EJ, Klein-Nulend J, Yin L, van Waas MAJ, Burger EH. Transforming growthfactor-b1 in calcium phosphate cement stimulates bone regeneration. J DentRes 2000;79:255.

[6] Moreau JL, Xu HHK. Mesenchymal stem cell proliferation and differentiationon an injectable calcium phosphate–chitosan composite scaffold. Biomater2009;30:2675–82.

[7] Freed LE, Vunjak-Novakovic G, Biron RJ, Eagles DB, Lesnoy DC, Barlow SK, et al.Biodegradable polymer scaffolds for tissue engineering. Biotechnology1994;7:689–93.

[8] Hutmacher DW. Polymeric scaffolds in tissue engineering bone and cartilage.Biomater 2000;21:2529–43.

[9] Langer R, Vacanti JP. Tissue engineering. Science 1993;260:920–6.[10] Levenberg S, Langer R. Advances in tissue engineering. Curr Top Dev Biol

2004;61:113–34.
650

[11] Ishikawa K, Miyamoto Y, Kon M, Nagayama M, Asaoka K. Non-decay type fast-setting calcium phosphate cement: composite with sodium alginate. Biomater1995;16:527–32.

[12] Oh SA, Lee GS, Park JH, Kim HW. Osteoclastic cell behaviors affected by thealfa-tricalcium phosphate based bone cements. J Mater Sci Mater Med2010;21:3019–27.

[13] Oh SA, Kim SH, Won JE, Kim JJ, Shin US, Kim HW. Effects on growth andosteogenic differentiation of mesenchymal stem cells by the zinc-added sol–gel bioactive glass granules, J Tiss Eng. 2010;2010:article ID 475260.

[14] Ginebra M, Espanol MT, Montufar E, Perez R, Mestres G. New processingapproaches in calcium phosphate cements and their applications inregenerative medicine. Acta Biomater 2010;6:2863–73.

[15] Espanol M, Perez RA, Montufar EB, et al. Intrinsic porosity of calciumphosphate cements and its significance for drug delivery and tissueengineering applications. Acta Biomater 2009;5:2752–62.

[16] Qi X, Ye J, Wang Y. Alginate/poly(lactic–co-glycolic acid)/calcium phosphatecement scaffold with oriented pore structure for bone tissue engineering. JBiomed Mater Res A 2009;89:980–7.

[17] Kim HW, Shin SY, Kim HE, Lee YM, Chung CP. Bone formation on the apatite-coated zirconia porous scaffolds within a rabbit calvarial defect. J BiomaterAppl 2008;22:485–504.

[18] Okamoto M, Dohi Y, Ohgushi H, Shimaoka H, Ikeuchi M, Matsushima A, et al.Influence of the porosity of hydroxyapatite ceramics on in vitro and in vivobone formation by cultured rat bone marrow stromal cells. J Mater Sci MaterMed 2006;17:327–36.

[19] Li Y, Zhang X, de Groot K. Hydrolysis and phase transition of alpha-tricalciumphosphate. Biomater 1997;18:737–41.

[20] Ginebra MP, Fernandez E, De Maeyer EAP, Verbeeck RMH, Boltong MG,Ginebra J, et al. Setting reaction and hardening of an apatite calcium phosphatecement. J Dent Res 1997;76:905–12.

[21] Kurashina K, Kurita H, Hirano M, Kotani A, Klein CPAT, Groot K. In vivo study ofcalcium phosphate cements: implantation of an a-tricalcium phosphate/dicalcium phosphate dibasic/tetracalcium phosphate monoxide cement paste.Biomater 1997;18:539–43.

[22] Hench LL. Bioceramics. J Am Ceram Soc 1998;81:1705–28.[23] Zhang Y, Xu HHK, Takagi S, Chow LC. In-situ hardening hydroxyapatite-based

scaffold for bone repair. J Mater Sci Mater Med 2006;17:437–45.[24] Ramay HR, Zhang M. Preparation of porous hydroxyapatite scaffolds by

combination of the gel-casting and polymer sponge methods. Biomater2003;24:3293–302.

[25] Furlaneto FA, Nagata MJ, Fucini SE, Deliberador TM, Okamoto T, Messora MR.Bone healing in critical-size defects treated with bioactive glass/calciumsulfate: a histologic and histometric study in rat calvaria. Clin Oral ImplantsRes 2007;18:311–8.

[26] Jegal SH, Park JH, Kim JH, Kim TH, Shin US, Kim TI, et al. Functional compositenanofibers of poly(lactide–co-caprolactone) containing gelatin–apatite bonemimetic precipitate for bone regeneration. Acta Biomater 2011;7:1609–17.

[27] Hong KS, Kim EC, Bang SH, et al. Bone regeneration by bioactive hybridmembrane containing FGF2 within rat calvarium. J Biomed Mater Res A2010;94:1184–94.

[28] Miyamoto Y, Ishikawa K, Takechi M, Toh T, Yoshida Y, Nagayama M, et al.Tissue response to fast-setting calcium phosphate cement in vivo. J BiomedMater Res 1997;37:457–64.

[29] Barralet JE, Wang L, Lawson M, Triffitt JT, Cooper PR, Shelton RM. Comparisonof bone marrow cell growth on 2D and 3-D alginate hydrogels. J Mater SciMater Med 2005;16:515–9.

[30] Bohner M, heiss F, Apelt D, Hirsiger W, Hoouriet R, Rizzoli G, Frei E, Auer JA,von Rechenberg B. Compositional changes of a dicalcium phosphate dihydratecement after implantation in sheep. Biomaterials 2003;24:3463–74.

[31] Sergey V, Dorozhkin SV. Calcium orthophosphate cements for biomedicalapplication. J Mater Sci Mater Med 2008;43:3028–57.

[32] Rezwan K, Meier LP, Gauckler LJ. Lysozyme and bovine serum albuminadsorption on uncoated silica and AlOOH-coated silica particles: the influenceof positively and negatively charged oxide surface coatings. Biomaterials2005;26:4351–7.

[33] Zhu XD, Fan HS, Li DX, Xiao YM, Zhang XD. Protein adsorption and zetapotentials of a biphasic calcium phosphate ceramic under various conditions. JBiomed Mater Res B 2007;82:65–73.

[34] Zhu XD, Fan HS, Zhao CY, Lu J, Ikoma T, Tanaka J, et al. Competitive adsorptionof bovine serum albumin and lysozyme on characterized calcium phosphatesby polyacrylamide gel electrophoresis method. J Mater Sci Mater Med2007;18:2243–9.

[35] Yun YR, Won JE, Jeon E, Lee J, Kang W, Jo H, et al. Fibroblast growth factors:biology, function, and application for tissue regeneration. J Tiss Eng 2010;article ID 218142, doi:10.4061/2010/218142.



Direct Deposited Porous Scaffolds of Calcium Phosphate Cement With Alginate

Documents