Biomaterials 32 (2011) 6006e6016

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Large-scale production of murine embryonic stem cell-derived osteoblasts and chondrocytes on microcarriers in serum-free media
Roz Alfred a, Jaymi T. Taiani b, Roman J. Krawetz c, Akihiro Yamashita c, Derrick E. Rancourt c, Michael S. Kallos a, *
a b c

Pharmaceutical Production Research Facility (PPRF), Schulich School of Engineering, University of Calgary, Calgary, AB T2N 1N4, Canada Department of Medical Sciences, Faculty of Medicine, University of Calgary, 3330 Hospital Dr. N.W., Calgary, AB T2N 4N1, Canada Department of Biochemistry and Molecular Biology, Faculty of Medicine, University of Calgary, 3330 Hospital Dr. N.W., Calgary, AB T2N 4N1, Canada

a r t i c l e i n f o
Article history: Received 21 March 2011 Accepted 5 April 2011 Available online 26 May 2011 Keywords: Osteoblasts Chondrocytes Stirred suspension bioreactors Microcarriers Serum-free cultivation

a b s t r a c t
The generation of tissue-engineered constructs from stem cells for the treatment of musculoskeletal diseases may have immense impact in regenerative medicine, but there are difculties associated with stem cell culture and differentiation, including the use of serum. Here we present serum-free protocols for the successful production of murine embryonic stem cell (mESC) derived osteoblasts and chondrocytes on CultiSpher S macroporous microcarriers in stirred suspension bioreactors. Various inoculum forms and agitation rates were investigated. Produced osteogenic cells were implanted ectopically into SCID mice and orthotopically into a murine burr-hole fracture model. Osterix, osteocalcin and collagen type I were upregulated in osteogenic cultures, while aggrecan and collagen type II were upregulated in chondrogenic cultures. Histological analysis using alizarin red S, von Kossa and alcian blue staining conrmed the presence of osteoblasts and chondrocytes, respectively in cultured microcarriers and excised tissue. Finally, implantation of derived cells into a mouse fracture model revealed cellular integration without any tumor formation. Overall, microcarriers may provide a supportive scaffold for ESC expansion and differentiation in a serum-free bioprocess for in vivo implantation. These ndings lay the groundwork for the development of clinical therapies for musculoskeletal injuries and diseases using hESCs and iPS cells. 2011 Elsevier Ltd. All rights reserved.

1. Introduction The growing demand for organ replacements coupled with a severe shortage of donors has led to the emergence of alternative therapies [1e3], including tissue engineering, which involves the production of cells, tissues or organs to replace or restore function in damaged or injured tissue. Cellular therapies, including therapies for musculoskeletal diseases, will require at least 107e109 cells per treatment [4]. The demand for clinically relevant numbers of cells necessitates the development of a robust, controllable system to scale-up the expansion of stem cells and their derived progeny. Current therapies for musculoskeletal diseases, including osteoporosis, osteoarthritis and non-union fractures, commonly focus on the use of palliatives, which only serve to alleviate pain [5]. Other treatment protocols involve the use of autograft and allograft
* Corresponding author. Present address: Pharmaceutical Production Research Facility, University of Calgary, 2500 University Drive N.W., Calgary, AB T2N 1N4, Canada. Tel.: 1 403 220 8107; fax: 1 403 284 4852. E-mail address: mskallos@ucalgary.ca (M.S. Kallos). 0142-9612/$ e see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2011.04.015

transplantations for skeletal defects, which are effective but have several limitations, including tissue availability, donor site morbidity and host immune rejection. There are currently no cures for these diseases. Although the benets of cell-based bone tissue engineering has been known for some time [6e10], several issues are yet to be resolved before clinical applications of this technology can be realized. One of these issues is the production of relevant numbers of the correct type of cells. The unlimited self-renewal capabilities of embryonic stem cells (ESCs) coupled with their ability to differentiate into every cell type in the body makes them attractive for use in regenerative medicine therapies [11]. However, successful downstream clinical use of these cells will require large-scale generation in a robust and tightly regulated fashion [12]. Produced cells may eventually be transplanted as aggregates, dissociated single cells or incorporated into a scaffold. Conventional methods for production of osteoblasts and chondrocytes from embryonic stem cells involve the use of static tissue cultures (e.g., dishes, cell factories) [13e21]. The available surface area for propagation of cells in static tissue culture is limited and

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

6007

passaging and harvesting protocols are time consuming and laborintensive. Static tissue culture systems also lack continuous monitoring and tight control of the culture microenvironment, which could lead to spontaneous stem cell differentiation and/or culture heterogeneity, and typically employ the use of animal-derived products [13e21]. Furthermore, the production of mineralized extracellular matrix from ESC-derived osteogenic cells creates tremendous difculties when attempting to harvest the cells from adherent static cultures, and leads to extensive cell loss. Exposure of ESCs to serum-containing media poses a major concern for downstream clinical applications because of the effects on ESC immunogenicity and potential pathogenicity. Serum is subject to considerable lot-to-lot variability and may contain unidentied pathogens, prions and viruses [22]. Moreover, for clinical applications, it will be important to avoid immune-based rejection responses including xenogeneic responses induced by the presence of animal-derived components [23]. Finally, a three dimensional (3D) culture system more closely resembles the physiological environment and has been shown to promote lineage-specic differentiation of several types of cells [24]. In considering the high demand for alternative treatments for musculoskeletal diseases and the limited supply of cells, as well as concerns surrounding differentiation efciencies, it is clear that the development of a robust, reproducible bioprocesses for the production of clinically relevant quantities of osteoblasts and chondrocytes is necessary. First introduced by van Wezel in 1967 to mass produce viral vaccines and biological cell products using mammalian cells [25], microcarriers have been successfully used for the large-scale production of a variety of biological products [26]. Microcarriers provide immense surface areas for attachment and expansion of anchorage-dependent cells in stirred tank reactors. The use of microcarriers offers a number of advantages including lower total costs, ease of harvesting and downstream processing procedures, ease of scale-up, and an overall reduction in the space required for a given-sized operation due to the high growth surface area per unit volume of reactor [27]. Microcarriers have typically been used in the production of viral vaccines using VERO cells, and the production of recombinant therapeutics using Chinese Hamster Ovary (CHO) and Baby Hamster Kidney (BHK) cells [28e30]. In addition to their capacity to serve as scaffolds for the propagation of anchorage-dependent and anchorage-preferred cells, microcarriers are also efcacious to employ as delivery vehicles for implantation of the cells in vivo. They have been used for the culture of pancreatic islet cells [31] mesenchymal stem cells [32e34] and broblast cells [35e37] amongst others. Despite the potential benets of producing clinically relevant quantities of osteoblasts and chondrocytes for cellular therapies, to date however, there have been no studies on the production of ESCderived osteoblasts or chondrocytes in a feeder-free, serum-free system using microcarrier scaffolds. To this end, we set out to develop an effective system for the large-scale production of ESCderived cells on microcarriers under feeder-free and serum-free conditions in stirred suspension bioreactors.

2.2. Culture medium 2.2.1. Maintenance medium Murine ESC maintenance medium consisted of Dulbeccos Modied Eagle Medium with high glucose (DMEM, Invitrogen, Carlsbad, California USA. Cat# 10569-010) supplemented with 15% Knockout-Serum Replacement (KSR, Invitrogen, Cat# 10828-018), 1% v/v non-essential amino acids (NEAA, Invitrogen, Carlsbad, California USA Cat# 11140-050), 0.2% (v/v) b-mercaptoethanol (Invitrogen, Carlsbad, California USA Cat# 21985-023) and 1% v/v penicillin/streptomycin (Pen/ Strep, Invitrogen, Cat# 15140-122). 2.2.2. Osteogenic medium Osteogenic medium (OM) consisted of maintenance medium supplemented with 105 M dexamethasone (SigmaeAldrich, Ontario Canada. Cat# D4902), 50 mg/mL ascorbic acid (SigmaeAldrich, Ontario Canada. Cat# A4403), and 5 mM b-glycerophosphate (SigmaeAldrich, Ontario Canada. Cat# G9891). 2.2.3. Chondrogenic medium There were two types of chondrogenic media which were used in separate cultures; (i) chondrogenic medium 1 (CM1) consisted of DMEM (Invitrogen, Carlsbad, California USA. Cat# 12100-061), supplemented with 15% Knockout-Serum Replacement (KSR, Invitrogen, Cat# 10828-018), 1% non-essential amino acids (NEAA, Invitrogen, Carlsbad, California USA. Cat# 11140-050), 1% penicillin/streptomycin (Pen/Strep, Invitrogen, Cat# 15140-122), 1 mM Sodium Pyruvate (100, Gibco, Carlsbad, California USA. #11140-050), 100 mM b-mercaptoethanol (550, Gibco #21985-0230), 50 mg/mL Ascorbic acid (SigmaeAldrich, Ontario Canada), 10 ng/mL BMP-2 (BMP-2, PeproTech # 120-02), 10 ng/mL TGFb1 (TGFb1, PeproTech Rocky Hill, NJ, USA. Cat# 100-21C) and 1% of an insulinetransferrineselenium complex (ITS, Invitrogen, Ontario Canada #51500-056), and (ii) chondrogenic medium 2 (CM2) consisted of all of the components in CM1 minus the TGFb1. 2.3. Mouse embryonic stem cell cultivation 2.3.1. Microcarrier preparation and cell expansion For microcarrier preparation, 0.1 g CultiSpher S (Percell Biolytica, storp, Sweden), a highly cross-linked type A porcine-derived gelatin macroporous microcarrier was weighed dry and initially hydrated separately in 50 mL of Ca2, Mg2 e free phosphate buffer saline (PBS, Invitrogen) overnight at room temperature. The supernatant was then removed and the microcarriers were washed twice in fresh PBS and sterilized by autoclaving at 120  C for 30 min. After sterilization, microcarriers were equilibrated in 20 mL of culture medium with LIF in NDS 125 mL bioreactors (NDS Technologies, Palo Alto, California USA, Cat# 264501-125) at 37  C and 5% CO2. After 1 h, medium volume was increased to 50 mL and cultivation in bioreactors was performed at 60 rpm, 5% CO2 and 37  C. ESCs previously thawed and expanded in T-75 tissue culture vessels were used to seed sterilized microcarriers at a 5:1 cell to bead ratio. Cultures were then intermittently stirred (3 min of stirring at 60 rpm followed by 27 min off) during the rst 8 h of culture. The medium was topped-up to a nal volume of 100 mL and stirred continuously at 60 rpm for 6 days. Medium was changed every second day during the expansion phase. 2.3.2. Differentiation to osteoblasts and chondrocytes Osteogenic cultures consisted of maintenance medium replaced with osteogenic medium; (i) a single cell inoculum without microcarriers (SC-MCs), agitated at 100 rpm, (ii) single cell inoculum plus empty sterilized microcarriers (SC MCs) agitated at 60 rpm, (iii) cell-loaded inoculum microcarriers from day 6 expansion, induced into differentiation without dissociation (CL MCs) agitated at 60 and 100 rpm. For all conditions, differentiation was carried out for 30 days. For chondrogenic differentiation, maintenance medium was replaced with CM1 and bioreactors were seeded with day 6 cell-loaded microcarriers (CL MCs) and agitated at 60 or 100 rpm. On day 5, CM1 was replaced with CM2 for the remainder of the differentiation. For all conditions, differentiation proceeded for a total of 30 days. 2.4. Cell harvest To harvest cells from microcarrier cultures, 2.0 mL duplicate samples were retrieved from spinners and placed in 15 mL centrifuge tubes. The supernatant was removed and the microcarriers rinsed twice with PBS and trypsinized with 0.25% Trypsin-EDTA (Invitrogen, Cat# 25200-056) for 15 min, with mechanical dissociation every 5 min. CultiSpher S microcarriers completely degraded upon trypsinization. The resulting cell suspension was preserved for later analyses. 2.5. Immunocytochemistry and ow cytometry

2. Materials and methods 2.1. Mouse embryonic stem cell line Unless otherwise stated, all cell-handling procedures were conducted in a sterile laminar ow hood using aseptic techniques and all ESC cultures were carried out in a humidied, 37  C and 5% CO2 incubator. Murine D3 ESCs (ATCC, American Type Culture Collection, Rockville, MD, USA) were used in all cultures. All bioreactor cultures were performed in duplicate and all cell counts were performed on duplicate samples.

For immunocytochemistry, cells were washed twice with PBS and xed with 4% Paraformaldehyde (Sigma) in PBS (pH 7.4) for 15 min at room temperature. After washing 3 with PBS, cells were permeabilized with 0.5% Saponin (Sigma, G-7900) in 1 PBS/1% BSA for 15 min at room temperature and then blocked with 3% bovine serum albumin (SigmaeAldrich, Ontario, Canada. Cat.# A9418) in 1 PBS for 30 min. The cells

6008

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016 processor for various histological analyses. Initially, samples were embedded into molds, creating parafn blocks. The blocks were then cut on a Leica 2125 microtome (Leica Microsystems, Ontario Canada) at 4 mm and placed on slides for staining and analyses. Stained slides were visualized under a Leica DMIL inverted microscope. Differentiated samples were assessed for calcium and mineral deposition by alizarin red and von Kossa staining for osteogenic differentiation, and alcian blue for cartilage. All conditions were tested to determine whether mineralization was enhanced or impeded by agitation or inoculum form. Samples tested included; (i) cell-loaded osteogenic induced microcarriers agitated at 100 rpm, (ii) cell-loaded osteogenic induced microcarriers, agitated at 60 rpm, (iii) single cell osteogenic induced microcarriers, agitated at 60 rpm and (iv) cell-loaded chondrogenic induced microcarriers, agitated at 60 or 100 rpm. Positive controls used for histological analyses were bovine bone tissue samples. Chondrogenic-differentiated cell-loaded CultiSpher S microcarrier samples were used as negative controls for mineral deposition, while the osteogenic-differentiated cell-loaded CultiSpher S microcarrier samples were used as negative controls for ECM deposition and cartilage formation. 2.6.1. von Kossa von Kossa staining was used to visualize mineralization. The von Kossa stain does not react with calcium but rather reacts with phosphate in the presence of acidic materials [38]. Mineralization was determined by black or brown depositions from the reaction between silver ions with phosphate ions. 2.6.2. Massons trichrome Massons Trichrome stain was used to examine the types of tissue present within the excised tissue. Red colored tissue was identied as skeletal muscle tissue, blue or green coloration was identied as collagen and bone.

were then incubated overnight at 4  C with primary antibodies against osterix, osteocalcin and collagen I for osteoblasts, and aggrecan and collagen II for chondrocytes (all from Santa Cruz Biotechnology, Santa Cruz, CA). The next day, the cells were washed 3 with PBS for 5 min and incubated in the dark with secondary antibodies for 1 h at room temperature. Secondary antibodies used were FITC conjugated donkey anti-goat antibodies (Santa Cruz Biotechnology, Santa Cruz, CA or Jackson ImmunoResearch Laboratories). Flow cytometry analysis was carried out on harvested differentiated cells using a FACS Calibur instrument and CellQuest software from Becton Dickinson. Each sample was run twice, with 10,000 events registered per run. The data was gated between a forward scatter (FSC-H) of 200 and 800, as to exclude cell debris and clumps of cells respectively. Controls were stained with secondary antibodies only. Immunoassaying of intact colonized microcarriers was performed and visualized using a Carl Zeiss laser scanning confocal microscope. For immunoassaying, intact colonized microcarriers were collected from spinners and let to settle for a few minutes in a 15 mL centrifuge tube. The supernatant was collected and disposed. The beads were then preserved in 4% PFA overnight. The next day, microcarriers were washed twice in 1 PBS and permeabilized with a 0.5% saponin in 1 PBS overnight at 4  C. Samples were later stained overnight with primary antibodies against osterix, osteocalcin and collagen I for osteoblasts, and Aggrecan and Collagen II for chondrocytes (all from Santa Cruz Biotechnology, Santa Cruz, CA). A FITC conjugated secondary antibody was used the next day to stain the cells. After staining, the loaded microcarriers were then mounted on slide with spacers using a mountant (9:1, glycerol: PBS) and the sides were sealed with clear nail polish and later visualized. 2.6. Histological analyses of differentiated colonized microcarriers At the end of the differentiation, day 29 cell-loaded CultiSpher S microcarriers were processed on the Tissue-Tek VIP 6 (Torrance, California USA) automated

Fig. 1. Process schematic shows culture, differentiation and transplantation of embryonic stem cell-derived osteoblasts and chondrocytes into animal models. Upon isolation, ESCs were scaled-up in suspension bioreactors in serum-free medium and then differentiated to osteoblasts or chondrocytes in suspension bioreactors with or without the use of microcarriers to obtain relevant numbers of cells. Differentiated cells were then harvested and transplanted as full osteogenic aggregates without microcarriers (from SC-MCs) or as cell-loaded microcarrier scaffolds (CL MCs).

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016 2.6.3. Alcian blue staining Alcian blue dye was used to identify the presence of glycosaminoglycans (GAG) in the chondrogenic cell-loaded microcarriers (CL MCs). These samples were prepared for histology as described above. Cartilage nodule formation was evaluated on prepared slides from day 29 of chondrogenic culture. 2.6.4. Alizarin red S staining To prepare cells for Alizarin Red staining, differentiation media was rinsed off each well with 2.0 mL PBS. After rinsing, 2.0 mL of buffered formalin was added to each well and incubated for 1 h at room temperature. The formalin was then discarded and the wells were rinsed twice with de-ionized (DI) water. After rinsing, 2.0 mL of 1% Alizarin Red S solution was added to the wells and incubated for 20 min at room temperature and the wells were rinsed twice with 2.0 mL of DI water. Slides were captured using a Zeiss inverted microscope.

6009

3. Results 3.1. Effect of inoculum form and agitation rate on differentiation ESCs were expanded rst in static tissue culture asks for 3 days (Day 9 to 6) and then in bioreactors in serum fee medium during the next 6 days (Day 6 to Day 0) as shown in Fig. 1. Fig. 2 (aed) shows the morphology of cells during the expansion period on CultiSpher S microcarriers, and the growth kinetics are represented in Fig. 2 (e). Following expansion, cultures were induced into osteogenic or chondrogenic differentiation (Day 0), while the cells were in the exponential growth phase. The effects of inoculum form and agitation on differentiation in serum-free medium were investigated. To investigate the effect of inoculum form, one set of bioreactors were seeded with cell-loaded microcarriers and agitated at 60 rpm (CL MCs). Here, maintenance medium was completely replaced with osteogenic medium. Another set of bioreactors was lled with empty sterilized microcarriers, equilibrated with osteogenic differentiation media, and then seeded with a single cell inoculum (SC MCs). Here again, cultures were agitated at 60 rpm. The effect of agitation rate was also investigated for osteogenic and chondrogenic-differentiated cultures. Here, bioreactors with osteogenic or chondrogenic differentiation media were inoculated with cell-loaded microcarriers and cultures were agitated either at 60 rpm (0.35 Pa maximum shear stress) or 100 rpm (0.60 Pa maximum shear stress). Samples were collected at various time intervals to assess the effect of agitation on differentiation. 3.2. Culture morphology The appearance of the cultures is shown over the course of the 30 days in Fig. 3 (aeo). Upon osteogenic induction, differentiated cells began to mineralize, as seen in Fig. 3 (aec, feh, kem). By day 10 of induction, more agglomeration could be observed in the single cell inoculum (Fig. 3 f), compared to the cell-loaded inoculum (Fig. 3 g). This is not unusual as ESCs have been known to aggregate when cultured in suspension [39e42]. The presence of more cells on the surface of the microcarriers at the start of differentiation, as indicated by the black arrows in Fig. 3a may have

2.7. In vivo implantation of esc-derived cells To assess the tumorigenicity of the differentiated cells in vivo, day 15 osteogenic aggregates (obtained from the SC-MC inoculum) as well as day 15 osteogenic cellloaded microcarriers (CL MCs) were separately injected subcutaneously into severe combined immuno-decient (SCID) mice (Taconic, Albany, New York. USA), using a 26 gauge syringe. Cells were not removed from the aggregates or microcarriers. Mice were housed in the single-barrier animal facility at the University of Calgary. The animals were euthanized after 6 weeks, and a second set of mice were euthanized after 8 weeks post-implantation. The tissue at the point of implantation was dissected and xed in 10% formalin for histological analysis. After dehydration in ascending concentrations of ethanol and xylene, the specimens were embedded in parafn, and sections were then deparafnized, hydrated, and stained for histological analyses. Tumorigenicity of the cells was also assessed by implanting osteogenic cellloaded microcarriers into burr-hole fractures created in the tibiae of osteoporotic (OVX) mice. One burr-hole fracture was induced in the left proximal tibiae in 5 Sv129 strain mice (strain-matched to the ESCs), as described by Taiani et al. (2010). Each mouse was rst ovariectomized, then the fracture was induced and then cell-loaded microcarriers were immediately implanted into the fracture. For implantation, approximately twenty day 25 cell-loaded CultiSpher S microcarriers were lifted from a Petri dish using a scalpel blade, transferred to the injury site and then pushed down into the burr-hole with a 27-gauge needle. Bone histomorphometry was assessed at day 0 and at 1 week, 2 weeks and 4 weeks following implantation using in vivo micro-CT imaging at an isotropic resolution of 15 mm (Micro-CT 35, Scanco Medical, Brttisellen, Switzerland). All surgical and animal care procedures were approved by the University of Calgary Animal Care Committee. 2.7.1. Statistical analysis Statistical analyses of the growth kinetics and FACS analyses data were conducted using the one-way ANOVA (analysis of variance). Results were considered statistically signicant if p < 0.05.

Fig. 2. Photomicrographs represent day 0 and day 4 progression of expansion for murine ESCs cultured on CultiSpher S microcarriers. Top row (a, b) represents day 0 brighteld and DAPI images, respectively, while bottom row (c, d) represents day 4 brighteld and DAPI images, respectively. A representative growth curve for murine embryonic stem cells, cultured on CultiSpher S microcarriers in serum-free media is shown in (e). For all conditions, bioreactors were inoculated with single cells at 60,000 cells/mL, agitated at 60 rpm and incubated at 37 C and 5% CO2. Scale bars represent 125 mm.

6010

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

Fig. 3. Progression of osteogenic and chondrogenic differentiation for murine embryonic stem cells induced to differentiate in osteogenic and chondrogenic media on CultiSpher S microcarriers. Photomicrographs represent day 0 (top row), day 10 (middle row) and day 30 (bottom row), murine ESCs induction to osteoblasts or chondrocytes. Cultures were induced into osteogenic differentiation as a single cell inoculum with microcarriers (SC MCs) agitated at 60 rpm (a, f & k), as a cell-loaded inoculum (CL MCs) agitated at 60 rpm (b, g & l) or as a cell-loaded inoculum (CL MCs) agitated at 100 rpm (c, h & m). Chondrogenic-differentiated cultures were induced either as a cell-loaded inoculum (CL MCs) agitated at 100 rpm (d, i & n) or at 60 rpm (e, j & o). Fluorescent Activated Cell Sorting Analyses (FACS) for days 5, 14 and 29 osteoblasts and chondrocytes, harvested from CultiSpher S microcarriers, indicate over 70% expression of markers for osteogenic and chondrogenic differentiation including osterix, collagen I and osteocalcin in all osteogenic culture conditions (p, b & r), while aggrecan and collagen II and were expressed in chondrogenic cultures as shown in (s & t). Graphs represent days 5, 14 and 29 progression of osterix (p), collagen type 1 (q), osteocalcin (r) osteogenic-differentiated cells. Expression of aggrecan (s) and collagen type II (t) were also assessed in chondrogenic-differentiated cells agitated at 60 or 100 rpm.

contributed to agglomeration within the cultures. There was an effect of agitation rate that became more apparent by day 30. Specically, the cultures at 100 rpm had smaller cell clusters and a narrower size distribution than the same cultures at 60 rpm. Chondrogenic induction proceeded with cell-loaded microcarriers (CL MCs) agitated at 60 rpm and at 100 rpm. By day 30 of cultivation, cellular outgrowths could be observed in the cultures (Fig. 3 n, o). However, the size of the outgrowths in the 60 rpm cultures were larger than those in cultures agitated at 100 rpm, suggesting that higher agitation rates led to better size control. 3.3. Flow cytometry and immunocytochemistry of harvested cells from CultiSpher S microcarrier cultures Murine embryonic stem cells induced into osteogenic differentiation on CultiSpher S macroporous microcarriers were completely dissociated into single cells after treatment with trypsin, which dissolved the microcarriers. Flow cytometry analyses of osteogenic cells harvested at various time points of differentiation showed high expression (over 70%) of osterix, osteocalcin and collagen type

I, as shown in Fig. 3 (per). Osterix, an early transcription factor in osteogenic differentiation, was expressed on day 5 of differentiation and peaked by day 14 for all conditions tested. Collagen type I and osteocalcin, the major noncollagenous component of bone matrix were expressed by day 14 of differentiation. Markers of osteogenic differentiation, including osterix, collagen type I and osteocalcin, for both forms of inoculation (SC MCs or CL MCs), showed no statistically relevant difference in the fraction of cells expressing the osteogenic markers as seen in Fig. 3 (p, q & r). Analyses of osteogenic or chondrogenic cultures agitated at 100 rpm versus 60 rpm, showed no statistically relevant difference in the number of cells expressing the markers analyzed, as seen in Fig. 3 (p, q & r) for osteoblasts or Fig. 3 (s & t) for chondrocytes. Whole intact osteogenic and chondrogenic cell-loaded microcarriers were probed for markers of osteogenic and chondrogenic differentiation at different time points, and visualized under a Leica inverted microscope as shown in Fig. 4. During osteogenic differentiation, results suggest expression of osterix was evident by day 5 of osteogenic differentiation, while osterix, osteocalcin and collagen type I, were expressed by day 29 of differentiation.

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

6011

Fig. 4. Immunocytochemical analyses for bioreactor ESC-derived osteoblasts and chondrocytes. Photomicrographs represent brighteld and uorescent images of intact cell-loaded microcarriers (CL MCs) analyzed for markers of osteogenic induction including osterix, osteocalcin and collagen type I (aef). Suspended osteoblasts cells were also harvested from cell-loaded microcarriers (gei) and immunoassayed for markers of osteogenic differentiation including osterix (g), osteocalcin (h) and collagen type I (i). Immunocytochemistry was also performed on days 5, 14 and 29 whole intact chondrogenic aggregates, and stained with markers for chondrogenic differentiation including aggrecan and collagen type II. Suspended chondrogenic cells were also harvested from microcarriers and immunoassayed for markers of chondrogenic differentiation including aggrecan (p) and collagen type II (g). Control samples (r) were stained only with secondary antibodies. For immunocytochemical analyses, intact cell-loaded microcarriers were cultivated in bioreactors, inoculated at 60,000 cells/mL and agitated at 60 rpm. For harvested and dissociated cultures, suspended cells were seeded into 24 well plates at a density of 60,000 cells/mL in differentiation medium and incubated at 37 and 5% CO2. Cultures were analyzed after 48 h. Scale bars represent 200 mm.

Expression of aggrecan and collagen type II, markers, which are indicative of a fully mature state of chondrogenesis, were observed by day 14 of differentiation. Similar results were observed when osteogenic and chondrogenic cell-loaded microcarrier cultures were dissociated and immunocytochemistry performed on dissociated cell suspension as shown in Fig. 4 (gei). 3.4. Histological analyses of cell-loaded CultiSpher S microcarrier culture von Kossa and Alizarin Red S analyses of induced osteoblasts cells (Fig. 5) revealed that mineralization had occurred for all the conditions tested including; cell-loaded osteogenic induced microcarriers (CL MCs) agitated at 100 rpm (Fig. 5 (a & e)), cellloaded osteogenic induced microcarriers (CL MCs) agitated at 60 rpm (Fig. 5 (b & f)) and single cell plus microcarrier (SC MC) osteogenic induction, agitated at 60 rpm (Fig. 5 (c & g)). A qualitative comparison of mineralization by von Kossa and Alizarin Red S histological analyses for calcium deposition, suggests no difference in the levels of differentiation between both inoculum forms.

Positive controls for von Kossa and Alizarin Red S, shown in Fig. 5 (d & h), respectively, were bovine bone tissue samples. Histological analyses were also performed on chondrogenicdifferentiated samples for the presence of GAGs by staining with alcian blue. Day 29 chondrogenic-differentiated cells showed evidence of cartilage formation, staining positive for alcian blue as shown in Fig. 5 (i & j) for cell-loaded chondrocytes agitated at 100 rpm and 60 rpm, respectively. Bovine articular cartilage tissue was used as positive controls (5 l), while negative controls (osteogenic cell-loaded microcarriers) showed no evidence of GAGs, (5 k). A qualitative comparison of chondrogenic-differentiated cultures agitated at 100 rpm compared to cultures at agitated at 60 rpm, showed no difference in the level of alcian blue expressed within samples as seen in Fig. 5 (i vs. j). 3.5. Histological analyses of excised tissue after ectopic implantation To determine osteogenic potential in vivo, aggregates of osteoblasts were harvested on day 15 from the bioreactors inoculated

6012

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

Fig. 5. von Kossa (top row) and alizarin red S (middle row) histological analyses of cell-loaded microcarrier cultures show mineral deposition (black deposits) in osteogenic cultures. Photomicrographs represent von Kossa and alizarin Red S analyses of: osteoblasts cell-loaded microcarriers, agitated at 100 rpm (a & e) respectively; osteoblasts cell-loaded microcarriers, agitated at 60 rpm (b & f) respectively, and single cell osteogenic induced microcarriers, agitated at 60 rpm (c & g) respectively. Positive controls for von Kossa and alizarin red (d & h). Bottom row represents chondrogenesis in cell-loaded microcarrier (CL MC) cultures agitated at 100 rpm (i) or at 60 rpm (j). Tissue formation indicated by dark blue coloration in chondrogenic cultures. Positive controls for alcian blue is represented in (l) while (k) represents negative controls for alcian blue. Scale bars represent 200 mm. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article).

with single cells without microcarriers (SC-MC) and implanted subcutaneously into SCID mice. Ectopic implantation was also performed on a separate set of mice using day 15 cell-loaded microcarriers (CL MCs) undergoing osteogenic differentiation. After 6e8 weeks the animals were euthanized and tissue at the point of injection was dissected and preserved in 10% formalin for histology. Tissue samples were excised from the site of implantation and analyzed as shown in Fig. 6. Mineralization was evidenced by black mineral deposition in tissue samples stained with von Kossa (6b), while red nodule formation was visible in tissue samples stained with Alizarin red S (6c). Histological analyses were also performed on tissue samples using Massons Trichrome and H&E staining to determine the types of tissue present in the excised tissue samples. Massons Trichrome produces a red coloration for keratin and muscle bers and a blue or green coloration for collagen and bone. Analyses of samples transplanted with osteoblasts aggregates cultured without microcarriers (SC-MCs), revealed muscle ber tissue formation primarily by day 15 of differentiation, with little collagen deposition as shown in Fig. 6d. Tissue samples harvested from SCID mice implanted with cell-loaded microcarriers (CL MCs) showed evidence of collagen deposition by the cells from 6 weeks, with the effects becoming more noticeable after 8 weeks of transplantation as seen in Fig. 6 (e & f).

Finally, Hematoxylin and eosin (H&E) analyses of excised tissue revealed formation of muscle tissue, 8 weeks after implantation of osteogenic aggregates as shown in Fig. 6 g. However, when cells were expanded, differentiated, and then transplanted into SCID mice while still on microcarriers (SC MCs or CL MCs), H&E tissue analyses revealed formation of bone tissue in excised tissue. Bone tissue formation was evident after 6 weeks when the rst tissue was excised and was more evident after 8 weeks of transplantation as shown in Fig. 6 (h & i) respectively.

3.6. Orthotopic implantation of differentiated cells into burr-hole fractures For this study, the burr-hole fracture model previously reported by Taiani et al. [39] was generated in ve 8-week old female ovariectomized Sv129 mice (the mice were strain-matched to the D3 ESCs used in this study). Day 25 cell-loaded CultiSpher S microcarriers (CL MCs) were used for implantation into the burrhole fractures. Micro-CT imaging performed at Day 0, and at Weeks 1, 2 and 4 post-implantation showed formation of new cortical and trabecular bone at the fracture site as seen in Fig. 7 (aed). Importantly, no abnormal disruption of the bone architecture, which could indicate tumor formation, was detected in any of the 5 mice

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

6013

Fig. 6. Sample tissue excised from mice 8 weeks after subcutaneous implantation (a). von Kossa and alizarin Red S analyses of excised tissue samples from SCID mice after 8 weeks of transplantation (b & c) respectively, show mineral deposition within tissue samples. Massons Trichrome histological analyses of excised tissue show evidence of bone formation in osteogenic tissue. Photomicrographs represent excised tissue, 8 weeks after implantation with osteogenic aggregates, without microcarriers (d), excised tissue, 6 and 8 weeks after implantation with osteoblasts cell-loaded microcarriers (e & f) respectively. Bottom row represents H&E histological analyses of excised tissue 8 weeks after transplantation with osteogenic aggregates without microcarriers (g), 6 and 8 weeks after transplantation with osteoblasts cell-loaded microcarriers (h & i). Aggregate cultures without microcarriers were inoculated at 60,000 cells/mL while microcarrier cultures were inoculated at 4000 cells/mL and incubated for 30 days at 37  C and 5% CO2. Black arrows indicate collagen deposition and bone formation, while green arrows indicate muscle bers. Scale bars represent 200 mm. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article).

for up to 4 weeks after implantation of the cells. Furthermore, new cortical bone had bridged the burr-hole defect in all mice. 4. Discussion Though the importance of developing effective systems to scaleup production of osteoblasts and chondrocytes to meet the rising demand of these cells for cell therapy has been realized for a while, thus far, only a few studies have explored the use of bioreactors for the tissue engineering of bone constructs from human bone marrow stromal cells [43,44]. Others have investigated murine embryonic stem cell encapsulation and differentiation to osteoblasts and chondrocytes [15,45e50]. Investigations have also been performed on endochondral ossication of cartilage formed by ESCs seeded on scaffolds [51], while others have explored production of tissue-engineered cartilage with and without EB formation [52,53]. This is the rst time however, that mESCs have been differentiated to osteoblasts and chondrocytes on CultiSpher S microcarriers in bioreactors under feeder-free and serum-free conditions.

In the current study, stirred suspension bioreactors provided an effective controllable method for differentiating ESCs to osteoblasts and chondrocytes on a biodegradable, macroporous, collagen-based microcarrier. CultiSpher S microcarriers offer several advantages over other microcarrier types including effective harvesting, shear protection and biodegradability. Unlike other types of microcarriers that are primarily glass or plastic, usually coated, CultiSpher S beads are made up of a highly cross-linked type A porcine-derived gelatin matrix. The ability to effectively harvest cells from microcarriers is an important aspect of microcarrier cell cultures especially during scale-up. The macroporosity and gelatinous composition of CultiSpher S beads enables easy cell harvest as the microcarrier completely disintegrates, leaving behind a single cell suspension. Histological analysis was also possible as a result of easy sectioning of the samples. In addition, CultiSpher S beads are FDA approved and have recently been used for in vivo dermis regeneration in humans [54,55]. Therefore, CultiSpher S microcarriers are a promising scaffold for use in the production of murine ESC-derived cells and eventual implantation of derived cells for the treatment of degenerative diseases.

6014

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

Fig. 7. Representative micro-CT images of orthotopic implantation of day 25 osteoblast-loaded microcarriers in a burr-hole fracture in one of the ve mice. Images represent day 0 (a), week 1, (b), week 2 (c) and week 4 (d) from the same mouse.

When the effect of inoculum form (single cells microcarriers versus cell-loaded microcarriers) on differentiation was assessed, immunocytochemical analyses as well as histological analyses showed mineralization could be observed under all conditions. This is important especially for cell cultures that require passaging as aggregates, as CultiSpher microcarriers may be passaged after complete, partial or no dissociation at all. Alizarin red staining was used to examine the presence of calcium deposition [56], as the presence of phosphate (as determined by von Kossa) does not necessarily imply that calcium is present in tissue [38]. Further, when two different hydrodynamic shear forces (60 versus 100 rpm) were applied to the cultures, von Kossa and alizarin red S analyses of cultures suggests there were no differences in differentiation efciencies between the cultures. The size of clumps was also controlled with increased shear, which is in agreement with previous studies that have found that agitation rates could be used to control the size of ESC aggregates in bioreactor cultures [42,57,58]. Overall, these ndings suggest that ESCs may be induced into osteogenic differentiation with the same efcacy when cultures are differentiated as single cells inoculated onto microcarriers (SC MCs) or when differentiation is initiated on previously colonized microcarriers (CL MCs). Differentiation on CultiSpher S microcarriers led to cells retaining a high expression (>70%) of osteogenic markers in serum-free medium. In addition, higher agitation rates resulted in better control of cellular outgrowths in cultures. CultiSpher S microcarriers presents a great advantage for cultivation of cell cultures at high agitation rates (>100 rpm), as their macroporous conguration protects cells against shear damage. This is a desirable characteristic for scaffolds used in differentiating cell cultures, particularly in agitated cultures, as previous studies of ESC differentiation have found that hydrodynamic shear can lead to the maintenance of pluripotency [39,59]. In addition, the conguration of cells grown in CultiSpher S miocrocarriers allows for effective differentiation without the formation of layers that have plagued osteogenic differentiation of ESCs as micromass aggregates in suspension bioreactors [59]. The

study by Yamashita (2010) found that cell culture methods had signicant inuence on ESC differentiation to osteoblasts. Histological analyses of osteogenic aggregates produced by micromass cultivation in the study revealed there was an inner core of mineralized cells that were surrounded by a layer of non- mineralized cells lying underneath a layer of extracellular matrix. The use of CultiSpher S microcarriers with average pore diameters of about 20 mm, may prevent development of these layers and hence enhance bone regeneration in vivo. Though the above described methodologies were effective at differentiating ESCs to osteoblasts and chondrocytes, the efciencies of differentiation as quantied by ow cytometry analyses for markers of differentiation was only 80% at best. This was found to be the case for all cultures tested. It may be possible to improve differentiation outcomes by investigating the use of different types of bioreactors or improving the scaffold design to better enhance differentiation [60e62]. However, incorporating sorting techniques, including sorting by uorescent activated cell sorting (FACS) or magnetic activated cell sorting (MACS), may further purify cultures prior to transplantation, or select for a population of progenitors which will enhance differentiation efciencies. When tissue samples were harvested 6 and 8 weeks after ectopic implantation of osteoblast cell-loaded CultiSpher S microcarriers into SCID mice and the results compared to tissue samples implanted with osteogenic aggregates only, results showed evidence of collagen deposition by the cell-loaded microcarriers as early as 6 weeks which became more evident after 8 weeks of implantation. However, analyses of samples transplanted with osteoblast aggregates only revealed muscle ber tissue formation even 8 weeks after transplantation. This suggests that formation of bone tissue was enhanced with the use of a collagen-based scaffold. Orthotopic implantation of cell-loaded microcarriers into a burr-hole fracture resulted in the formation of new bone at the fracture site. Although this type of fracture has been shown to heal without therapeutic intervention [39], it is important to note that no abnormal disruption of the bone architecture was observed

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016

6015

following implantation of the osteoblast-loaded CultiSpher S microcarriers. The formation of large soft tissue masses that cause disruption of the bone architecture has been observed by Taiani et al. (2011, submitted) following the implantation of some types of ESC-derived skeletal cells into the burr-hole fracture. No such disruptions were observed in the current study, and in contrast, fractures were observed to heal normally. This is a signicant nding, particularly in the eld of regenerative medicine using embryonic stem cells. Over the past decade, there has been signicant skepticism regarding the case for using ESC-derived cells for tissue engineering and regenerative medicine applications because of the issue with tumor formation upon implantation. One possible explanation for this is the fact that the development of tumors appears to be linked to the use of animal based serum in the cell culture media. Implantation of ESC-derived cells cultured in bioreactors in serum-containing medium and later implantation into a burr-hole defect in bone has been shown to result in tumor formation. However, when the cells were implanted after differentiation in static tissue culture, no tumors were observed [39]. In the current study, we have demonstrated that the elimination of animal based serum in culture media may inhibit formation of tumors post-implantation. Finally, human embryonic stem cells (hESCs) are generally passaged in small clumps, as complete dissociation of hESCs during passaging has been shown to result in cell death [40,63]. The use of CultiSpher S microcarriers presents a great advantage for eventual human applications as these microcarriers could be dissociated into small clumps using collagenase, without complete dissociation into single cells. Overall, cultivation of ESCs on CultiSpher S microcarriers in suspension bioreactors would allow for ESC-derived cells to be produced in a robust and controllable environment, while protecting the cells from hydrodynamic shear. 5. Conclusions Here, we have shown successful production of ESC-derived osteoblasts and chondrocytes in suspension bioreactors. Results suggest effective differentiation to osteoblasts and chondrocytes on a macroporous gelatin based microcarrier. The biodegradable nature of CultiSpher S offers great potential for tissue engineering bone and cartilage as it provides a suitable environment for expansion and differentiation of ESCs and degrades upon transplantation, leaving only regenerated tissue where there was once an injury. When the effect of inoculum form (single cell versus cellloaded microcarrier inoculum) on the efciency of differentiation was investigated, no differences were observed. In addition, differentiation was effective at various agitation rates. Finally, orthotopic transplantation of derived cells did not result in tumor formation. The methodologies developed in this research may provide a path forward in the generation and transplantation of relevant numbers of cells for use in musculoskeletal regenerative therapies. Acknowledgments RMA was awarded scholarships from the Province of Alberta (Queen Elizabeth II Scholarship). RMA, JTT and RK received funding from the Skeletal Regenerative Medicine Team funded by the Canadian Institutes of Health Research (CIHR). RMA and JTT received funding from the Biomedical Engineering Graduate Program at the University of Calgary. JTT was funded through NSERC and AHFMR. AY was supported by AHFMR and JSPS Fellowships. MSK and DER received funding through an NSERC Collaborative Health Research Project, and DER is an AHFMR Senior Scholar. We would like to thank

Dr John Matyas from the Faculty of Veterinary Medicine, University of Calgary, Calgary, AB, Canada, T2N 4N1 for use of the Confocal Microscope and Maureen Bukhari from the University of Calgary Faculty of Veterinary medicine for the histopathology work performed on specimens. References
[1] Navarro-Alvarez N, Soto-Gutierrez A, Kobayashi N. Hepatic stem cells and liver development. In: Rajasekhar VK, Vemuri MC, editors. Regulatory networks in stem cells. New York: Humana Press; 2009. p. 439e72. [2] Tsang VL. Three-dimensional tissue fabrication: application in hepatic tissue engineering. In: Ma PX, Elisseeff JH, editors. Scaffolding in tissue engineering. CRC Press. Taylor and Francis Group; 2006. p. 609e21. [3] Dekel B, Reisner Y. Progenitor cell therapy for kidney regeneration. In: Battler A, Leor J, editors. Stem cell and gene-based therapy: frontiers in regenerative medicine. London: Springer-Verlag; 2006. p. 209e23. [4] Mason C, Dunnill P. Quantities of cells used for regenerative medicine and some implications for clinicians and bioprocessors. Regen Med 2009;4(2):153e7. [5] Palaniswamy C, Selvaraj DR, Rao V, Patel U. Newer therapies for osteoporosis. Am J Ther 2010;17(2):197e200. [6] Zhou J, Lin H, Fang T, Li X, Dai W, Uemura T, et al. The repair of large segmental bone defects in the rabbit with vascularized tissue engineered bone. Biomaterials 2010;31(6):1171e9. [7] Bianco P, Cancedda FD, Riminucci M, Cancedda R. Bone formation via cartilage models: the borderline chondrocyte. Matrix Biol 1998;17(3):185e92. [8] Kim WS, Vacanti CA, Upton J, Vacanti JP. Bone defect repair with tissueengineered cartilage. Plast Reconstr Surg 1994;94(5):580e4. [9] Vacanti CA, Kim W, Upton J, Vacanti MP, Mooney D, Schloo B, et al. Tissueengineered growth of bone and cartilage. Transplant Proc 1993;25(1 Pt 2): 1019e21. [10] Rotter N, Haisch A, Bucheler M. Cartilage and bone tissue engineering for reconstructive head and neck surgery. Eur Arch Otorhinolaryngol 2005; 262(7):539e45. [11] Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282(5391):1145e7. [12] Thomson H. Bioprocessing of embryonic stem cells for drug discovery. Trends Biotechnol 2007;25(5):224e30. [13] Titorencu I, Jinga V, Constantinescu E, Gafencu A, Ciohodaru C, Manolescu I, et al. Proliferation, differentiation and characterization of osteoblasts from human BM mesenchymal cells. Cytotherapy 2007;9(7):682e96. [14] Duplomb L, Dagouassat M, Jourdon P, Heymann D. Differentiation of osteoblasts from mouse embryonic stem cells without generation of embryoid body. In Vitro Cell Dev Biol Anim 2007;43(1):21e4. [15] Hwang YS, Randle WL, Bielby RC, Polak JM, Mantalaris A. Enhanced derivation of osteogenic cells from murine embryonic stem cells after treatment with HepG2-conditioned medium and modulation of the embryoid body formation period: application to skeletal tissue engineering. Tissue Eng 2006;12(6): 1381e92. [16] Zur Nieden NI, Kempka G, Rancourt DE, Ahr HJ. Induction of chondro-, osteoand adipogenesis in embryonic stem cells by bone morphogenetic protein-2: effect of cofactors on differentiating lineages. BMC Dev Biol 2005;5:1e15. [17] Jukes JM, Both SK, van Blitterswijk CA, de Boer J. Potential of embryonic stem cells for in vivo bone regeneration. Regen Med 2008;3(6):783e5. [18] Kramer J, Hargus G, Rohwedel J. Derivation and characterization of chondrocytes from embryonic stem cells in vitro. Methods Mol Biol 2006;330:171e90. [19] Garreta E, Genove E, Borros S, Semino CE. Osteogenic differentiation of mouse embryonic stem cells and mouse embryonic broblasts in a three-dimensional self-assembling peptide scaffold. Tissue Eng 2006;12(8):2215e27. [20] Bourne S, Polak JM, Hughes SP, Buttery LD. Osteogenic differentiation of mouse embryonic stem cells: differential gene expression analysis by cDNA microarray and purication of osteoblasts by cadherin-11 magnetically activated cell sorting. Tissue Eng 2004;10(5e6):796e806. [21] Buttery LD, Bourne S, Xynos JD, Wood H, Hughes FJ, Hughes SP, et al. Differentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng 2001;7(1):89e99. [22] Rajala K, Hakala H, Panula S, Aivio S, Pihlajamaki H, Suuronen R, et al. Testing of nine different xeno-free culture media for human embryonic stem cell cultures. Hum Reprod 2007;22(5):1231e8. [23] Hisamatsu-Sakamoto M, Sakamoto N, Rosenberg AS. Embryonic stem cells cultured in serum-free medium acquire bovine apolipoprotein B-100 from feeder cell layers and serum replacement medium. Stem Cells 2008;26(1):72e8. [24] Abbott A. Cell culture: biologys new dimension. Nature 2003;424(6951): 870e2. [25] Van Wezel AL. Growth of cell-strains and primary cells on micro-carriers in homogeneous culture. Nature 1967;216(5110):64e5. [26] Tree JA, Richardson C, Fooks AR, Clegg JC, Looby D. Comparison of large-scale mammalian cell culture systems with egg culture for the production of inuenza virus A vaccine strains. Vaccine 2001 May 14;19(25e26):3444e50. [27] GE healthcare. In: GE, editor. Handbooks from GE-healthcare. microcarrier cell culture e principles and methods. Piscataway: GE Healthcare; 2005. p. 1e161.

6016

R. Alfred et al. / Biomaterials 32 (2011) 6006e6016 [45] Hwang YS, Cho J, Tay F, Heng JY, Ho R, Kazarian SG, et al. The use of murine embryonic stem cells, alginate encapsulation, and rotary microgravity bioreactor in bone tissue engineering. Biomaterials 2009;30(4):499e507. [46] Hwang NS, Varghese S, Zhang Z, Elisseeff J. Chondrogenic differentiation of human embryonic stem cell-derived cells in arginine-glycine-aspartatemodied hydrogels. Tissue Eng 2006;12(9):2695e706. [47] Hwang NS, Varghese S, Theprungsirikul P, Canver A, Elisseeff J. Enhanced chondrogenic differentiation of murine embryonic stem cells in hydrogels with glucosamine. Biomaterials 2006;27(36):6015e23. [48] Hwang NS, Kim MS, Sampattavanich S, Baek JH, Zhang Z, Elisseeff J. Effects of three-dimensional culture and growth factors on the chondrogenic differentiation of murine embryonic stem cells. Stem Cells 2006;24(2):284e91. [49] Tielens S, Declercq H, Gorski T, Lippens E, Schacht E, Cornelissen M. Gelatinbased microcarriers as embryonic stem cell delivery system in bone tissue engineering: an in-vitro study. Biomacromolecules 2007;8(3):825e32. [50] Declercq HA, Gorski TL, Tielens SP, Schacht EH, Cornelissen MJ. Encapsulation of osteoblast seeded microcarriers into injectable, photopolymerizable threedimensional scaffolds based on d, l-lactide and epsilon-caprolactone. Biomacromolecules 2005;6(3):1608e14. [51] Jukes JM, Both SK, Leusink A, Sterk LM, van Blitterswijk CA, de Boer J. Endochondral bone tissue engineering using embryonic stem cells. Proc Natl Acad Sci USA 2008;105(19):6840e5. [52] Nakagawa T, Lee SY, Reddi AH. Induction of chondrogenesis from human embryonic stem cells without embryoid body formation by bone morphogenetic protein 7 and transforming growth factor beta1. Arthritis Rheum 2009;60(12):3686e92. [53] Xie F, Zhang W, Chen F, Zhou G, Cui L, Liu W, et al. Chondrogenic differentiation of mouse embryonic stem cells promoted by mature chondrocytes. Sci China C Life Sci 2008;51(9):774e8. [54] Huss FR, Nyman E, Bolin JS, Kratz G. Use of macroporous gelatine spheres as a biodegradable scaffold for guided tissue regeneration of healthy dermis in humans: an in vivo study. J Plast Reconstr Aesthet Surg 2010;63(5):848e57. [55] Huss FR, Nyman E, Gustafson CJ, Gisselfalt K, Liljensten E, Kratz G. Characterization of a new degradable polymer scaffold for regeneration of the dermis: in vitro and in vivo human studies. Organogenesis 2008;4(3):195e200. [56] Sila-Asna M, Bunyaratvej A, Maeda S, Kitaguchi H, Bunyaratavej N. Osteoblast differentiation and bone formation gene expression in strontium-inducing bone marrow mesenchymal stem cell. Kobe J Med Sci 2007;53(1e2):25e35. [57] Gilbertson JA, Sen A, Behie LA, Kallos MS. Scaled-up production of mammalian neural precursor cell aggregates in computer-controlled suspension bioreactors. Biotechnol Bioeng 2006;94(4):783e92. [58] Baghbaderani BA, Behie LA, Sen A, Mukhida K, Hong M, Mendez I. Expansion of human neural precursor cells in large-scale bioreactors for the treatment of neurodegenerative disorders. Biotechnol Prog 2008;24(4):859e70. [59] Yamashita A, Nishikawa S, Rancourt DE. Microenvironment modulates osteogenic cell lineage commitment in differentiated embryonic stem cells. PLoS One 2010;5(3):e9663. 1e10. [60] Yeatts AB, Fisher JP. Bone tissue engineering bioreactors: dynamic culture and the inuence of shear stress. Bone 2011 Feb 1;48(2):171e81. [61] Zhang ZY, Teoh SH, Teo EY, Khoon Chong MS, Shin CW, Tien FT, et al. A comparison of bioreactors for culture of fetal mesenchymal stem cells for bone tissue engineering. Biomaterials 2010;31(33):8684e95. [62] El Haj AJ, Cartmell SH. Bioreactors for bone tissue engineering. Proc Inst Mech Eng H 2010;224(12):1523e32. [63] Li Y, Powell S, Brunette E, Lebkowski J, Mandalam R. Expansion of human embryonic stem cells in dened serum-free medium devoid of animal-derived products. Biotechnol Bioeng 2005;91(6):688e98.

[28] Berry JM, Barnabe N, Coombs KM, Butler M. Production of reovirus type-1 and type-3 from vero cells grown on solid and macroporous microcarriers. Biotechnol Bioeng 1999;62(1):12e9. [29] Frazzati-Gallina NM, Paoli RL, Mourao-Fuches RM, Jorge SA, Pereira CA. Higher production of rabies virus in serum-free medium cell cultures on microcarriers. J Biotechnol 2001;92(1):67e72. [30] Junker BH, Wu F, Wang S, Waterbury J, Hunt G, Hennessey J, et al. Evaluation of a microcarrier process for large-scale cultivation of attenuated hepatitis A. Cytotechnology 1992;9(1e3):173e87. [31] Del Guerra S, Bracci C, Nilsson K, Belcourt A, Kessler L, Lupi R, et al. Entrapment of dispersed pancreatic islet cells in CultiSpher-S macroporous gelatin microcarriers: preparation, in vitro characterization, and microencapsulation. Biotechnol Bioeng 2001;75(6):741e4. [32] Eibes G, Dos Santos F, Andrade PZ, Boura JS, Abecasis MM, Silva CL, et al. Maximizing the ex-vivo expansion of human mesenchymal stem cells using a microcarrier-based stirred culture system. J Biotechnol 2010;146(4): 194e7. [33] Bouf C, Thomas O, Bony C, Giteau A, Venier-Julienne MC, Jorgensen C, et al. The role of pharmacologically active microcarriers releasing TGF-beta3 in cartilage formation in vivo by mesenchymal stem cells. Biomaterials 2010; 31(25):6485e93. [34] Yang Y, Rossi FM, Putnins EE. Ex vivo expansion of rat bone marrow mesenchymal stromal cells on microcarrier beads in spin culture. Biomaterials 2007;28(20):3110e20. [35] Sommar P, Pettersson S, Ness C, Johnson H, Kratz G, Junker JP. Engineering three-dimensional cartilage- and bone-like tissues using human dermal broblasts and macroporous gelatine microcarriers. J Plast Reconstr Aesthet Surg 2010;63(6):1036e46. [36] Phillips BW, Lim RY, Tan TT, Rust WL, Crook JM. Efcient expansion of clinicalgrade human broblasts on microcarriers: cells suitable for ex vivo expansion of clinical-grade hESCs. J Biotechnol 2008;134(1e2):79e87. [37] Zhang L, Zhang Y, Yan C, Yu J. The culture of chicken embryo broblast cells on microcarriers to produce infectious bursal disease virus. Appl Biochem Biotechnol 1997;62(2e3):291e302. [38] Bonewald LF, Harris SE, Rosser J, Dallas MR, Dallas SL, Camacho NP, et al. Von Kossa staining alone is not sufcient to conrm that mineralization in vitro represents bone formation. Calcif Tissue Int 2003;72(5):537e47. [39] Taiani JT, Krawetz RJ, Zur Nieden NI, Elizabeth Wu Y, Kallos MS, Matyas JR, et al. Reduced differentiation efciency of murine embryonic stem cells in stirred suspension bioreactors. Stem Cells Dev 2010;19(7):989e98. [40] Krawetz R, Taiani JT, Liu S, Meng G, Li X, Kallos MS, et al. Large-scale expansion of pluripotent human embryonic stem cells in stirred-suspension bioreactors. Tissue Eng Part C Methods 2010;16(4):573e82. [41] Alfred R, Gareau T, Krawetz R, Rancourt D, Kallos MS. Serum-free scaled up expansion and differentiation of murine embryonic stem cells to osteoblasts in suspension bioreactors. Biotechnol Bioeng 2010;106(5):829e40. [42] Cormier JT, Zur Nieden NI, Rancourt DE, Kallos MS. Expansion of undifferentiated murine embryonic stem cells as aggregates in suspension culture bioreactors. Tissue Eng 2006;12(11):3233e45. [43] Janssen FW, Van Dijkhuizen-Radersma R, Van Oorschot A, Oostra J, de Bruijn JD, Van Blitterswijk CA. Human tissue-engineered bone produced in clinically relevant amounts using a semi-automated perfusion bioreactor system: a preliminary study. J Tissue Eng Regen Med 2010;4(1):12e24. [44] Chen X, Xu H, Wan C, McCaigue M, Li G. Bioreactor expansion of human adult bone marrow-derived mesenchymal stem cells. Stem Cells 2006;24(9): 2052e9 (Dayton, Ohio).