Composite hydrogel scaffolds incorporating ...

Biomaterials xxx (2013) 1e10

Contents lists available at ScienceDirect

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

Composite hydrogel scaffolds incorporating decellularized adipose tissue for soft tissue engineering with adipose-derived stem cells Hoi Ki Cheung a, b, Tim Tian Y. Han a, b, Dale M. Marecak a, b, John F. Watkins c, Brian G. Amsden a, b, Lauren E. Flynn a, b, d, * a

Department of Chemical Engineering, Queen’s University, 19 Division Street, Kingston, Ontario, Canada K7L 3N6 Human Mobility Research Centre, Kingston General Hospital, 76 Stuart Street, Kingston, Ontario, Canada K7L 2V7 Department of Surgery, Queen’s University, 166 Brock Street, Kingston, Ontario, Canada K7L 5G2 d Department of Biomedical and Molecular Sciences, Queen’s University, 18 Stuart Street, Kingston, Ontario, Canada K7L 3N6 b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 31 October 2013 Accepted 21 November 2013 Available online xxx

An injectable tissue-engineered adipose substitute that could be used to deliver adipose-derived stem cells (ASCs), filling irregular defects and stimulating natural soft tissue regeneration, would have significant value in plastic and reconstructive surgery. With this focus, the primary aim of the current study was to characterize the response of human ASCs encapsulated within three-dimensional bioscaffolds incorporating decellularized adipose tissue (DAT) as a bioactive matrix within photo-cross-linkable methacrylated glycol chitosan (MGC) or methacrylated chondroitin sulphate (MCS) delivery vehicles. Stable MGC- and MCS-based composite scaffolds were fabricated containing up to 5 wt% cryomilled DAT through initiation with long-wavelength ultraviolet light. The encapsulation strategy allows for tuning of the 3-D microenvironment and provides an effective method of cell delivery with high seeding efficiency and uniformity, which could be adapted as a minimally-invasive in situ approach. Through in vitro cell culture studies, human ASCs were assessed over 14 days in terms of viability, glycerol-3-phosphate dehydrogenase (GPDH) enzyme activity, adipogenic gene expression and intracellular lipid accumulation. In all of the composites, the DAT functioned as a cell-supportive matrix that enhanced ASC viability, retention and adipogenesis within the gels. The choice of hydrogel also influenced the cell response, with significantly higher viability and adipogenic differentiation observed in the MCS composites containing 5 wt% DAT. In vivo analysis in a subcutaneous Wistar rat model at 1, 4 and 12 weeks showed superior implant integration and adipogenesis in the MCS-based composites, with allogenic ASCs promoting cell infiltration, angiogenesis and ultimately, fat formation. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Adipose tissue engineering Mesenchymal stem cell Scaffold Extracellular matrix Chitosan Chondroitin sulphate

1. Introduction Adipose tissue engineering represents a promising alternative for reconstructive and cosmetic applications in plastic surgery to restore injury- or age-related soft tissue loss within the subcutaneous layer [1,2]. A common strategy is to seed threedimensional scaffolds with regenerative cell populations, such as adipose-derived stem cells (ASCs), to create tissue substitutes that can be used to induce stable and predictable fat formation [3,4]. Ideally, the scaffold should mimic the properties of the native extracellular matrix (ECM), supporting cell adhesion and viability,

* Corresponding author. Department of Chemical Engineering, Queen’s University, 19 Division Street, Kingston, Ontario, Canada K7L 3N6. Tel.: þ1 613 533 6000x79177; fax: þ1 613 533 6637. E-mail address: lauren.fl[email protected] (L.E. Flynn).

and maintaining the structural integrity of the construct as it is gradually replaced with healthy adipose tissue [5]. Many studies to date have focused on designing implantable scaffolds comprised of synthetic or naturally-derived polymers that have a pre-defined shape and volume [6]. More recent efforts in this application have included the development of injectable biomaterials, including hydrogels, particulates and microcarriers, which offer a more minimally-invasive strategy for soft tissue augmentation in the clinic [7,8]. In all of these approaches, it has become apparent that many different factors can influence adipogenesis in cell-based therapies with ASCs, including complex cell-ECM and cellecell interactions within the engineered microenvironment. These results highlight the importance of rational design of delivery vehicles for ASCs in a tissue-specific manner to enable long-term soft tissue regeneration. As a step toward engineering adipose tissue replacements for use in plastic and reconstructive surgery, our group has been

0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.11.067

Please cite this article in press as: Cheung HK, et al., Composite hydrogel scaffolds incorporating decellularized adipose tissue for soft tissue engineering with adipose-derived stem cells, Biomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.11.067

2

H.K. Cheung et al. / Biomaterials xxx (2013) 1e10

developing implantable and injectable biomaterials derived from decellularized adipose tissue (DAT), including 3-D scaffolds, microcarriers, and foams [9e11]. Interestingly, we have shown that the decellularized ECM of human adipose tissue provides a highly conducive milieu for fat formation both in vitro and in vivo. Other groups investigating scaffolds incorporating ECM from adipose tissue have reported similar findings [12e14], including recent work with temperature sensitive adipose-derived hydrogels [15,16]. Moreover, we have shown that the DAT can naturally induce the adipogenic differentiation of human ASCs in culture without medium supplementation with exogenous differentiation factors. For clinical applications in soft tissue augmentation, our aim in the current study was to develop injectable bioscaffolds that integrate the unique pro-adipogenic properties of the DAT within a customizable hydrogel delivery vehicle, to engineer composite biomaterials that have highly tunable characteristics. More specifically, methacrylated glycol chitosan (MGC) and methacrylated chondroitin sulphate (MCS) were investigated as photo-crosslinkable hydrogel carriers for the encapsulation of human ASCs within a cryomilled DAT matrix. Chitosan is a linear polysaccharide of N-acetyl-glucosamine and N-glucosamine subunits obtained from the deacetylation of chitin derived from crustaceans, which has been explored in a range of tissue-engineering applications due to its innate wound healing and antibacterial properties [17,18]. The use of chitosan as a biomaterial, however, is hampered by its poor water solubility at physiologic pH. Glycol chitosan, formed by the reaction of ethylene oxide with chitosan, is both O and N-glycolated [18], making it soluble in aqueous media from pH 2 to 12. In contrast, chondroitin sulphate is a sulphated glycosaminoglycan (GAG) that consists of repeating D-glucuronic acid and N-acetyl galactosamine disaccharide units. It is a highly water-soluble structural component of the human ECM that has been indicated to play important roles during embryogenesis and wound healing [19e21]. Both chitosan and chondroitin sulphate are readily chemically modifiable and can be rendered photo-cross-linkable through methacrylation to enable in situ gelation with lowintensity UV light under mild conditions. To investigate the potential of our composite scaffold approach for adipose regeneration, fabrication methods were established to synthesize stable MGC- and MCS-based bioscaffolds containing 0, 3 or 5% (w/v) cryomilled DAT, with the base hydrogels formulated to have matching mechanical properties. Extensive in vitro cell culture experimentation was conducted to characterize the viability and adipogenic differentiation of human ASCs encapsulated within the composites. Further, the in vivo biocompatibility of the MGC- and MCS-based composites containing 5% DAT was assessed in a subcutaneous Wistar rat model at 1, 4 and 12 weeks, both with and without incorporation of allogenic rat ASCs. 2. Materials and methods 2.1. Materials Chondroitin sulphate (weight average molecular weight of approximately 15 kDa) was generously donated by Stellar Pharmaceuticals Inc. (London, ON, Canada). Unless otherwise stated, all chemicals were purchased from SigmaeAldrich Canada Ltd. (Oakville, ON, Canada) and used as received.

2.2.1. Cryo-milling of decellularized adipose tissue Following decellularization, the DAT was snap frozen in liquid nitrogen, lyophilized, and finely minced with sharp scissors. The minced DAT was then cryomilled using a laboratory ball mill (Sartorius Stedim Mikro-Dismembrator, Goettingen, Germany) at 2500 rpm for 2 min. The milled DAT was stored at 4 C until further use. Immediately prior to composite scaffold fabrication, the DAT was sterilized by exposure to UV light for 60 min. 2.3. Polymer purification and reaction The chemical compositions of all glycol chitosan (GC) and chondroitin sulphate (CS) derivatives were assessed by 1H NMR spectroscopy using a Bruker Avance-600 Ultrashield spectrometer with a 5 mm TBI S3 probe and data collection with Bruker’s XWIN NMR software. The samples were prepared by dissolving the polymer overnight in deuterium oxide at a concentration of 20 mg/mL and adjusted to pH 10 using 1 M sodium hydroxide. All GC-based polymers were analyzed at 90 C and all CS-based polymers were analyzed at room temperature. 2.3.1. Purification and methacrylation of glycol chitosan The GC (weight average molecular weight of approximately 120 kDa, from Wako Chemical USA Inc., Richmond, VA, USA) was purified using an established protocol [22]. Briefly, the polymer was dissolved in deionized water, filtered, and dialyzed against distilled water using 50 kDa dialysis tubing (Spectrum Laboratories Inc., Rancho Dominguez, CA, USA). The purified GC was frozen and lyophilized. Subsequently, the GC was methacrylated using glycidyl methacrylate via an established protocol [22]. Purified GC was dissolved in deionized water to obtain a 2% (w/v) solution that was adjusted to pH 9.0 using 1 M sodium hydroxide. Glycidyl methacrylate at 3% (v/v) was added to achieve an initial molar ratio of glycidyl methacrylate to free amine (per mol glycol chitosan residue) of 0.1. The solution was allowed to react for 24 h before the reaction was neutralized (pH 7) with 1 M hydrochloric acid. The methacrylated glycol chitosan (MGC) solution was dialyzed twice against deionized water with 12 kDa dialysis tubing (Fisher Scientific, Oakville, Canada) for 2 h. The purified MGC was frozen and lyophilized. 2.3.2. Methacrylation of chondroitin sulphate The methods for the methacrylation of the CS using methacrylate anhydride were adapted from the approach described by Li et al. [23]. CS was dissolved at 0.2 g/ mL in 1.0 M sodium phosphate. A 1:1 blend of methacrylic anhydride and dimethylsulfoxide (DMSO) was prepared and added to the dissolved CS in a volumetric ratio of 0.06:1. After 2 h incubation at room temperature, the pH was adjusted to pH 10 with 5 M sodium hydroxide, and then allowed to react for an additional 22 h. The solution was then dialyzed against deionized water for 12 h with 3.5 kDa dialysis tubing (Fisher Scientific, Hanover Park, IL, USA), and the purified methacrylated chondroitin sulphate (MCS) solution was frozen and lyophilized. 2.3.3. Hydrogel characterization MGC and MCS hydrogels were fabricated via photo-polymerization based on previously-established methods [22]. To prepare hydrogels with varying Young’s moduli, the prepolymer concentration in solution was adjusted prior to photo-crosslinking, with the objective of obtaining MGC and MCS concentrations yielding hydrogels with similar moduli. Specifically, MGC hydrogels were made from MGC solution concentrations of 1.5, 2, 2.5 and 3% (w/v) and compared against MCS hydrogels prepared from MCS solution concentrations of 8, 10 and 15% (w/v) (4 replicate samples (n) with 3 measurements of each sample (N)). A 5 mg/mL solution of Irgacure 2959 (2-hydroxy-40 -(2-hydroxyethoxy)-2-methylpropiophenone) photoinitiator in deionized water was added to the prepolymer solution to achieve a final concentration of 0.05% (w/v). The gels (100 mL) were fabricated in a 16-well chamber glass slide (Nalgene Nunc International, Rochester, New York) and photo-cross-linked with long-wavelength ultraviolet light (320e390 nm, EXFO Lite, EFOS Corporation, Mississauga, Canada) at an intensity of 10 mW/cm2 for 3 min, to form hydrogels with a diameter of 7 mm and a height of 3 mm. The indentation method described in the study by Hayes et al. was used to measure the Young’s modulus of fully hydrated gels in deionized water (n ¼ 4, N ¼ 3) [24]. A TA XT plus texture analyzer (Texture Technologies Corp., New York) equipped with flat-ended cylindrical indenters (3 and 7 mm diameter) was used to apply force at a rate of 0.05 mm/s over a distance of 0.5 mm. The Poisson ratio and the Young’s modulus for each gel were calculated using equations derived from Hayes et al., as described previously [24].

2.2. Adipose tissue procurement and processing Adipose tissue samples were collected from female patients undergoing elective breast reduction or abdominoplasty surgery at the Kingston General Hospital or Hotel Dieu Hospital in Kingston, ON, Canada. The samples were transported to the lab on ice in sterile, cation-free phosphate buffered saline (D-PBS; Thermo Scientific HyClone, Fisher Scientific, Oakville, ON, Canada), supplemented with 20 mg/mL bovine serum albumin (BSA). The donor age, weight and height were recorded. Within 2 h, the tissues were processed for ASC isolation or decellularization, using published methods [9]. This study was reviewed and approved by the research ethics board at Queen’s University (REB# CHEM-002-07).

2.4. Fabrication of composite hydrogel scaffolds Composite MGC-DAT and MCS-DAT hydrogels were fabricated via photopolymerization, using methods adapted from the previous section [22]. Based on the initial mechanical characterization studies of the hydrogels, the MGC was dissolved at a concentration of 1.5% (w/v) and the MCS at a concentration of 10% (w/v) for all subsequent studies. The Irgacure 2959 photoinitiator solution (5 mg/mL) was sterile syringe filtered and added to achieve a final concentration of 0.05% (w/v). Prepolymer solutions were prepared containing 0, 3 or 5% (w/v) milled DAT and mixed by gentle pipetting. 100 mL of each solution was transferred into a 16-well


H.K. Cheung et al. / Biomaterials xxx (2013) 1e10 chamber glass slide (Nalgene Nunc International, Rochester, New York) and photocross-linked as described previously. 2.5. Composite scaffold physical characterization Photo-cross-linked MGC and MCS scaffolds incorporating 0, 3 or 5% (w/v) milled DAT were fabricated for analysis of Young’s modulus and sol content as described below. The number of replicate samples of each scaffold type per trial (n) is indicated, as well as the number of times each study was repeated (N) to confirm the data. 2.5.1. Mechanical properties As described, the indentation method of Hayes et al. was used to measure the Young’s modulus of the fully hydrated composite scaffolds in deionized water (n ¼ 4, N ¼ 3) [24], using a TA XT plus texture analyzer (Texture Technologies Corp., New York) with flat-ended cylindrical indenters (3 and 7 mm diameter). 2.5.2. Sol content To measure the sol content, immediately following cross-linking the scaffolds (n ¼ 4, N ¼ 3) were frozen in liquid nitrogen and lyophilized. The initial dry mass was recorded (mo). To extract the sol content, the gels were incubated three times in fresh deionized water for 3 h, and then snap frozen in liquid nitrogen, lyophilized, and re-weighed (mgel). To account for any DAT loss from the composite scaffolds during washing, the wash water was filtered and the residue was lyophilized and weighed (mDATloss). The sol content was calculated using the following equation:

Sol ¼

mo mgel þ mDATloss mo

100%

3

WA, USA) with established protocols [27]. ASCs cultured on TCPS 6-well plates in adipogenic differentiation medium (2-D positive control) and proliferation medium (2-D negative control), as well as ASCs cultured on milled DAT alone in differentiation medium (3-D positive control), were included at every time point. The GPDH activity levels in all samples were normalized to the total cytosolic protein content measured using the Bio-Rad Protein Assay (Bio-Rad Laboratories, Inc.) with an albumin standard. One unit was defined as the activity required to oxidize 1 mmole of NADH in 1 min. 2.6.5. End-point RT-PCR analysis Expression of the adipogenic genes peroxisome-proliferator activated receptorg (PPARg), CCAAT/enhancer binding protein-a (C/EPBa), and lipoprotein lipase (LPL) was assessed in the MGC- and MCS-based scaffolds with 0% and 5% DAT at 7 and 14 days after adipogenic induction (n ¼ 2, N ¼ 3) by end-point RT-PCR, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the housekeeping gene [28], using published methods and primer sets [9]. The 2-D and 3-D controls described in the previous section were included in every trial. The PCR products were analyzed by gel electrophoresis on 5% agarose gels, stained with ethidium bromide and detected using ultraviolet light (G:box Chemi HR16, Syngene, Cambridge, UK). Densitometry analysis was performed using ImageJ analysis software. 2.6.6. Oil red O staining Oil red O staining was used to qualitatively assess ASC adipogenesis in terms of intracellular lipid accumulation in the MGC- and MCS-based scaffolds containing 0% and 5% DAT at 14 days after adipogenic induction (n ¼ 3, N ¼ 3), using established methods [27]. As described previously, the milled DAT was pre-labeled with Alexa FluorÒ 350 carboxylic acid, succinimidyl ester before composite scaffold fabrication. At each time point, the scaffolds were fixed overnight in 10% neutral buffered formalin, rinsed in PBS and incubated in the oil red O working dye solution for 10 min. The samples were rinsed extensively with water to remove excess dye, and visualized using an Olympus FV 1000 laser scanning confocal microscope.

2.6. In vitro characterization of the ASC-seeded composite bioscaffolds 2.6.1. Adipose-derived stem cell culture As discussed, human ASCs were isolated from fresh adipose tissue samples and expanded on tissue culture polystyrene (TCPS) using established protocols [25]. Fresh proliferation medium comprised of DMEM:Ham’s F-12 medium supplemented with 10% fetal bovine serum (FBS) (Thermo Scientific Hyclone, Cat. # SH30396, Fisher Scientific) and 100 U/mL penicillin and 0.1 mg/mL streptomycin (1% penestrep) (Gibco Cat. # 15140-122, Life Technologies, Burlington, Canada) was provided every 2e3 days and the cells were passaged at 80% confluence. Passage 2 (P2) ASCs were used for all encapsulation experiments. For each trial of an assay, all scaffold groups and TCPS controls were seeded with ASCs sourced from a single donor to eliminate potential cell donor variability. All studies were repeated with cells from 3 different donors (N ¼ 3) to verify the trends. 2.6.2. ASC photo-encapsulation and scaffold culture For all cell culture studies, the prepolymer solutions were prepared in proliferation medium. P2 ASCs were encapsulated within the hydrogels through suspension in the prepolymer solutions at a concentration of 1 107 cells/mL, and then photo-cross-linking using the methods described above. Immediately following encapsulation, the cell-seeded gels were transferred into 6-well plates and cultured in proliferation medium (37 C, 5% CO2). After 24 h, adipogenic differentiation was induced using a defined serum-free adipogenic differentiation medium comprised of DMEM:Ham’s F12 nutrient mixture supplemented with 33 mM biotin, 17 mM pantothenate, 66 nM human insulin, 1 nM triiodothyronine (T3), 10 mg/mL transferrin, 100 nM hydrocortisone, 100 U/mL penicillin and 0.1 mg/mL streptomycin [26]. For the first 72 h, 1 mg/mL of troglitazone and 0.25 mM isobutylmethylxanthine (IBMX) was included in the differentiation medium. Fresh medium was provided to all samples every 2e3 days. 2.6.3. Cell viability A LIVE/DEADÒ Viability Assay kit (Life Technologies Inc., Burlington, Canada) was used to assess cell viability in the MGC- and MCS-based scaffolds with 0%, 3% or 5% DAT at 24 h after encapsulation, as well as at 7 and 14 days after the induction of differentiation (n ¼ 3, N ¼ 3). To facilitate visualization using confocal microscopy, the DAT was pre-labeled with an amine-reactive Alexa FluorÒ 350 carboxylic acid, succinimidyl ester (Life Technologies) before composite scaffold fabrication [27]. Imaging was conducted with an Olympus FV 1000 laser scanning confocal microscope using a mosaic stitching to capture the complete cross-sectional area of the gel at a specific depth. For each scaffold, five layers were scanned, with each layer separated by 100e200 mm. ImageJ analysis software was used to calculate the number of live and dead cells in each layer. 2.6.4. Glycerol-3-phosphate dehydrogenase (GPDH) activity To quantify ASC adipogenesis within the scaffolds, intracellular GPDH enzyme activity was measured in the MGC- or MCS-based scaffolds containing 0%, 3% or 5% DAT at 3, 7 and 14 days after the induction of adipogenic differentiation (n ¼ 3, N ¼ 3) using a GPDH Activity Kit (Kamiya Biomedical Corporation, Cat # KT-010, Seattle,

2.7. In vivo characterization of the composite bioscaffolds The in vivo response to the composite MGC þ 5% DAT and MCS þ 5% DAT scaffolds was assessed at 1, 4 and 12 weeks in an established subcutaneous Wistar rat model (12-week old females, Charles River Laboratories) [11]. Scaffolds seeded with 1 106 allogenic rat ASCs isolated from the epididymal fat pad of 12-week old male Wistar rats and unseeded control scaffolds were analyzed in triplicate at each time point, with a total of 9 rats included in the study. All studies followed the Canadian Council on Animal Care (CCAC) guidelines for the care and use of laboratory animals and were approved by the Queen’s University Animal Care Committee (Protocol # Flynn-2009-059 & Flynn-2010-053). The rats were sacrificed by CO2 overdose and the scaffolds were explanted for histological analysis using Masson’s trichrome staining. 2.8. Statistical analysis All numerical data is expressed as the mean standard deviation (SD). Statistical analyses were performed using GraphPad PrismÒ version 6 (GraphPad Software, San Diego, CA), by one-way ANOVA with a Tukey’s post-hoc test and p < 0.05.

3. Results 3.1. Hydrogel characterization N-methacrylate glycol chitosan (MGC) and O-methacrylate chondroitin sulfate (MCS) were prepared with a degree of methacrylation of 14 and 16%, respectively. In assessing the mechanical properties of the hydrogels alone, a general upward trend in Young’s modulus was observed for both prepolymers with respect to an increase in polymer concentration (Fig. 1), with MGC exhibiting a greater increase in modulus for a given increase in concentration. The Young’s modulus of a hydrogel made from 1.5% (w/ v) MGC solution (30.1 4.0 kPa) was comparable to that made from 10% (w/v) MCS solution (37.1 5.0 kPa). The Poisson ratio remained relatively constant between concentrations and gel type, with the ratio ranging from 0.129 to 0.149 in MGC gels and 0.130 to 0.137 in MCS gels, consistent with those reported in our previous study with MGC gels [22]. These concentrations were therefore used for all MGC- and MCS-based hydrogels for the remainder of this study, to control for the potential effects of hydrogel stiffness on the ASC response.


4


concentration, no statistically significant differences were found in the moduli between the scaffolds fabricated with MGC versus MCS (Fig. 2B). However, incorporating DAT increased the stiffness of the scaffolds, with significant differences observed in the mean values for each DAT concentration (0%, 3%, 5%) for both MGC and MCS. Sol content was measured to assess the average amount of unreacted polymer in each photo-cross-linked scaffold group (Fig. 2C). Similar sol content was found for both the MGC- and MCS-based scaffolds, with decreasing amounts as the DAT concentration increased. 3.3. ASC viability

Fig. 1. Young’s moduli of photo-cross-linked MGC and MCS for varying polymer concentrations (n ¼ 4, N ¼ 3). Hydrogels synthesized with a 1.5% (w/v) MGC solution (30.1 4.0 kPa) had comparable properties to those made with a 10% (w/v) MCS solution (37.1 5.0 kPa). * No statistical difference in modulus values (p < 0.05).

3.2. Physical characterization of the composite scaffolds Stable composite hydrogels were fabricated through photocross-linking MGC or MCS containing 0%, 3%, or 5% milled DAT dispersed through the hydrogel phase (Fig. 2A). For each DAT

Cell viability was assessed using confocal microscopy at 24 h, as well as after 7 days and 14 days in culture in adipogenic differentiation medium (Fig. 3A). At 24 h after encapsulation, the average viability ranged from 68 to 76%. Incorporating DAT enhanced cell viability in both the MGC- and MCS-based hydrogels, although the differences in viability between the scaffold groups were not statistically significant at 24 h. However, there was a significant decrease in cell viability in the MGC and MCS without DAT from 24 h to 14 days, and incorporating DAT promoted long-term viability in the composites (Fig. 3A). Further, at 14 days, viability in the MCS þ 5% DAT scaffolds (69.7 6.2%) was significantly higher than all other groups. The viability data was analyzed to determine the average viable cell count per xy plane across the depth of the gel during the culture

Fig. 2. (A) Representative images of the scaffolds immediately following photo-cross-linking. Macroscopically, there were no significant differences in the composite scaffolds containing 3% or 5% DAT. (B) Young’s moduli of fully-hydrated MGC- and MCS-based scaffolds (n ¼ 4, N ¼ 3). The degree of substitution (DOS) of the base hydrogel was 14% for the MGC and 16% for the MCS. (C) Sol content (n ¼ 4, N ¼ 3) for the hydrogels and composites.



5

Fig. 3. (A) ASC viability after photo-encapsulation and culturing in adipogenic differentiation medium. *Statistically different at 14 d as compared to 24 h **Statistically different than MGC þ 5% DAT and ***Statistically different than all other groups at 14 days. (B) Average number of viable cells per xy plane. *All groups statistically different at 14 d as compared to 24 h **Statistically different than corresponding hydrogel-alone group at that time point. All data is expressed as the mean SD. (n ¼ 3, N ¼ 3) (p < 0.05).

period (Fig. 3B). For all of the scaffold groups, there was a significant difference in the cell counts at 24 h as compared to 14 days, consistent with a gradual cell loss over time. Incorporating the DAT enhanced viable cell retention, with statistically higher average cell counts in the MGC and MCS composites incorporating 5% DAT as compared to the hydrogels alone at 7 and 14 days (Fig. 3B). 3.4. GPDH enzyme activity The adipogenic differentiation of the ASCs was quantitatively assessed in the MGC- and MCS-based scaffolds, as well as on milled DAT and TCPS controls at 72 h, 7 days and 14 days after adipogenic induction (Fig. 4). GPDH activity was enhanced in all scaffold groups as compared to the 2-D positive controls, and the levels generally increased from 72 h to 14 days, consistent with a progression in differentiation. No significant activity was detected in the negative controls cultured on TCPS in proliferation medium. Incorporating DAT enhanced GPDH activity at all time points in the hydrogel groups. More specifically, a significant difference was observed in the MGC þ 5% DAT scaffolds as compared to the MGC þ 0% DAT scaffolds at 7 and 14 days. Similarly, the MCS þ 5% DAT scaffolds were significantly different than the hydrogel alone at 7 days, as well as all other groups at 14 days. Overall, there was a trend for higher GPDH activity in the MCS-based scaffolds relative to the MGC-based scaffolds. Considering all of the scaffold groups including the DAT controls, the lowest GPDH activity levels were consistently observed in the MGC scaffolds without DAT at all time points.

3.5. Adipogenic gene expression End-point RT-PCR analysis with densitometry was used to semiquantitatively assess the expression of the adipogenic master regulators PPARg and C/EBPa, as well as the later marker lipoprotein lipase (LPL), with GAPDH as the housekeeping gene in the MGCand MCS-based scaffolds containing 0% or 5% DAT, along with the DAT and TCPS controls (Fig. 5). At both 7 and 14 days, only GAPDH was detected in the non-induced TCPS controls cultured in proliferation medium (results not shown). The intensity of the gene of interest bands were normalized to those of GAPDH for each sample, and these values were used to calculate the relative expression levels for each of three adipogenic genes compared to the induced TCPS positive control samples at 7 days (Fig. 5). For all of the groups, the relative expression levels of the three adipogenic markers increased from 7 to 14 days. Consistent with the GPDH enzyme activity results, adipogenic gene expression was enhanced in the MCS-based scaffolds relative to the MGC-based scaffolds, and incorporating DAT had a positive effect on adipogenesis. Moreover, in comparing all of the groups, the MCS þ 5% DAT composite scaffolds had the highest levels of PPARg, C/EBPa and LPL gene expression at both time points. 3.6. Intracellular lipid accumulation Intracellular lipid accumulation in the MGC- and MCS-based scaffolds containing 0% or 5% DAT was visualized after oil red O

Fig. 4. GPDH enzyme activity after the induction of adipogenic differentiation in the scaffold groups and induced TCPS positive controls. *All scaffold groups were statistically different at 14 days as compared to 72 h, consistent with a progression in differentiation. **The composites including 5% DAT were significantly different than the corresponding hydrogel alone scaffolds (0% DAT) at the same time point. ***Significantly different than all other groups at 14 days. All data is expressed as the mean SD. (n ¼ 3, N ¼ 3) (p < 0.05).


6


Fig. 5. End-point RT-PCR analysis of adipogenic gene expression at 7 and 14 days after adipogenic induction using GAPDH as the endogenous housekeeping gene. Representative bands are shown for each of the sample groups (n ¼ 2, N ¼ 3). The relative levels of expression were semi-quantitatively measured by densitometry with the induced TCPS controls at 7 days as the calibrator. All data is expressed as the mean SD.

staining by confocal microscopy at 14 days after the induction of differentiation (Fig. 6). In all of the scaffolds, ASCs containing red intracellular lipid droplets, characteristic of cells undergoing adipogenic differentiation, were detected. In the MGC and MCS hydrogels without DAT, individual cells containing lipid were distributed throughout the scaffolds. Incorporating milled DAT within both types of composites qualitatively enhanced the level of adipogenic differentiation, with clusters of differentiating ASCs containing extensive intracellular lipid observed in close contact with the DAT.

microenvironment points to the need for greater integration of the fields of materials science and cell biology to be able to more closely mimic the events that occur during normal tissue development and thereby enhance regeneration [29]. In the current study, we sought to engineer optimized bioscaffolds for adipogenic applications by combining the favorable features of synthetic hydrogels with the unique bioactive properties of human adipose tissue ECM. This

3.7. In vivo characterization of the composite scaffolds Representative Masson’s trichrome staining of the MGC- and MCS-based composites with 5% DAT at 1, 4 and 12 weeks postimplantation are shown in Figs. 7 and 8 respectively. Macroscopically, the MCS-based composites degraded more rapidly than the MGC-based scaffolds, with the MCS supporting more extensive cellular infiltration and enhanced integration over the course of the 12-week study. At 12 weeks, it was estimated that approximately 50% of the initial implant volume for the MGC-based scaffolds remained, as compared to 20e25% of the MCS-based scaffolds. Both types of composites were surrounded by a fibrous capsule, which qualitatively decreased in thickness and increased in vascularization as the scaffolds remodeled. Seeding the composites with 1 106 allogenic rat ASCs enhanced the rate of scaffold degradation, as well as inflammatory cell infiltration and angiogenesis in the scaffold regions. By 12 weeks, the majority of the DAT within the seeded scaffolds had degraded, with large fragments of the MGC (Fig. 7) and small fragments of the MCS (Fig. 8) remaining. A small number of adipocytes were found at the periphery of the MGC þ 5% DAT scaffolds at 12 weeks (Fig. 7). In contrast, more extensive adipogenesis was observed in the MCS-based scaffolds, with large numbers of mature adipocytes visualized in the central regions of the seeded implants at 12 weeks (Fig. 8). 4. Discussion While progress has been made in biomaterials-based approaches for soft tissue engineering, the complexity of the native cellular

Fig. 6. Representative oil red O staining of intracellular lipid accumulation at 14 days after the induction of adipogenic differentiation in the MGC- and MCS-based scaffolds containing 0% or 5% DAT. Clusters of differentiating cells containing lipid droplets (red) were observed around the milled DAT (blue) encapsulated within the composite scaffolds.



7

Fig. 7. Representative Masson’s trichrome staining of the MGC þ 5% DAT composites at 1, 4 and 12 weeks post-implantation in the subcutaneous Wistar rat model. Both unseeded scaffolds and scaffolds seeded with 1 106 allogenic rat ASCs were compared. All images show the interface between the host tissues (at top of images) and the implant, with the MGC staining purple/red and the encapsulated DAT appearing blue. At 4 weeks, enhanced degradation and cellular infiltration in the ASC-seeded implants was noted. By 12 weeks, there was significant resorption of the DAT in the seeded scaffolds, and a small number of adipocytes observed at the periphery of the implant region (arrows). Scale bars represent 100 microns.

composite approach allows for a high degree of tunability in terms of the compositional and mechanical properties of the constructs, and also enables the homogenous distribution of the stem cell population throughout the scaffold by encapsulation under mild conditions [30,31]. This strategy could be applied to create implantable scaffolds in a range of 3-D shapes and volumes, with the geometry controlled through the use of patient-specific moulds [32]. Moreover, it could also be adapted as a more minimally-invasive injectable approach with in situ cross-linking, which would be useful for the reconstruction of irregular soft tissue defects [33]. MGC and MCS are attractive materials for cell encapsulation and delivery because these highly-hydrated and porous hydrogels can facilitate nutrient and waste transfer by diffusion, to promote longterm cell viability following encapsulation. Further, the bulk mechanical properties of the hydrogels can be tuned to mimic native soft tissues and prevent frictional irritation following implantation [34,35]. This is of particular importance for applications in volume augmentation, where mechanical mismatching can result in scar

tissue formation that can cause poor implant integration, an unnatural feel and appearance, and potentially implant migration or collapse [36,37]. Both polymers were methacrylated to render them susceptible to photo-polymerization, which provides a rapid and effective polymer cross-linking method that can be used for cell entrapment with low cytotoxicity [38]. Conducting a simultaneous assessment of MGC and MCS provided an opportunity to explore the influence of the hydrogel carrier within the composites, without the confounding effects of cell donor variability. The ASC response can be profoundly influenced by the cell donor characteristics, as well as the isolation and culture techniques employed in the analysis [39]. Consequently, it can be challenging at times to directly compare the results obtained by different research groups. From our experience, there is great value in assessing multiple scaffold groups within a single study in order to identify the key design characteristics for the engineered matrices and ultimately optimize the 3-D cellular microenvironment.


8


Fig. 8. Representative Masson’s trichrome staining of the MCS þ 5% DAT composites at 1, 4 and 12 weeks post-implantation in the subcutaneous Wistar rat model. Both unseeded scaffolds and scaffolds seeded with 1 106 allogenic rat ASCs were compared. All images show the interface between the host tissues (at top of images) and the implant, with the MCS staining light blue/clear and the encapsulated DAT appearing dark blue. Extensive cellular infiltration into scaffolds was noted as early as 1 week, with enhanced microscopic degradation relative to the MGC-based scaffolds. Seeding the scaffolds with ASCs enhanced cellularity within the implant region at all time points. At 12 weeks, mature adipocytes (arrows) were observed in the implants, with more significant adipogenesis in the ASC-seeded group. Scale bars represent 100 microns.

In terms of the two polymers studied, the cationic MGC and anionic MCS are both very hydrophilic due to their charge characteristics. Importantly, we were able to match the biomechanical properties of the two base hydrogels by adjusting their concentrations, to eliminate the possibility of differences in the hydrogel stiffness influencing the observed cell responses [40]. While chitosan-based biomaterials can stimulate wound healing, a potential advantage to the application of MCS is that sulphated GAGs are known to function as reservoirs for growth factors due to the natural electrostatic interactions that occur between the positivelycharged proteins and negatively-charged sulphate groups [41]. Within the native ECM, the contributions of sulphated GAGs to directing cell responses, including proliferation, migration and differentiation, are mediated through their ability to bind and present growth factors [42]. This natural capacity may have contributed to the more favorable responses observed in terms of adipogenesis with the MCS-based bioscaffolds.

While the MGC and MCS hydrogels have many advantages in terms of injectable cell delivery vehicles, both materials lack the biological cues to promote cell attachment and long-term viability. To address this limitation, a common strategy is to modify polymers with cell-adhesive peptides such as the integrin-binding sequence RGD [43]. However, a composite approach integrating decellularized ECM components within the hydrogel presents an interesting alternative that has not yet been extensively studied. In previous work, composite bioscaffolds for adipose tissue engineering comprised of decellularized human placenta with cross-linked hyaluronan were explored, with the combined scaffolds showing enhanced adipogenesis relative to scaffolds comprised of the decellularized placenta alone [44]. In the current study we sought to adapt this general strategy and apply a more tissue-specific regenerative approach. In particular, unlike the decellularized placenta, our recent studies have demonstrated that the decellularized ECM of human adipose tissue has both adipo-inductive and



adipo-conductive properties that provide a strong rationale for its selection as a bioactive matrix for soft tissue augmentation and reconstruction [9e11]. The adipogenic differentiation of the ASCs on the DAT is likely mediated through both biochemical and biomechanical signaling events and there is evidence that these bioscaffolds may recapitulate the native ASC niche [11,45]. From a translational standpoint, human adipose tissue discarded from elective surgery represents an abundant and readily available source of human ECM. Cryomilling of the DAT enabled homogenous dispersion within the hydrogel phase for injectable delivery prior to photo-crosslinking to generate stable 3-D composite scaffolds that were easy to handle. Interestingly, a decrease in sol content was observed for the scaffolds with higher DAT concentrations, indicating enhanced crosslinking efficiency. Based on these results, it is likely that the proteins in the DAT reacted with the methacrylate groups on the hydrogel or with other proteins during the free-radical polymerization. This is supported by previous work with collagen scaffolds that demonstrated the involvement of aromatic groups on amino acids such as tyrosine and phenylalanine during photo-cross-linking [46]. Incorporating the DAT within the hydrogels enhanced longterm cell viability and retention by providing the ASCs with a cell-adhesive matrix, enriched in collagen and laminin [9]. Similar viability was observed for the base MGC and MCS hydrogels. However, the viability in the composite MCS with 5% DAT scaffolds was higher than all other groups at 14 days, indicating that the higher DAT concentration and sulphated GAG matrix were favorable for cell survival. In addition, the inclusion of the DAT within the hydrogels significantly enhanced ASC adipogenesis, with very consistent results observed for all cell donors in terms of GPDH enzyme activity, adipogenic gene expression and intracellular lipid accumulation. The MCS with 5% DAT scaffolds induced very high levels of adipogenic differentiation, supporting the selection of these composites for adipose regeneration. The encapsulation of the ASCs within a microenvironment combining the collagenbased DAT embedded within a sulphated GAG matrix may provide the closest mimic of the natural cell-ECM interactions that occur within tissues [47]. While CS-based biomaterials have principally been studied in the context of cartilage regeneration [48e 50], studies in adipogenic cell lines have shown that soluble and cell-associated CS synthesis is upregulated during adipogenesis, suggesting that the MCS hydrogels could play a direct role in mediating the cell response [51]. Cell density is another important factor that may have influenced ASC adipogenesis within the composite bioscaffolds. In particular, the cell encapsulation strategy provides a highly efficient means of entrapping the cells within the scaffold during seeding [44]. While there was some cell loss over time, a relatively high density was maintained in the composites over 14 days. In contrast, cells seeded on DAT scaffolds without a hydrogel carrier are more prone to cell washout during routine culturing, which can lead to regional variability in the number of cells distributed throughout the matrix [9]. From this standpoint, the encapsulation approach is advantageous in that higher cell densities may be favorable in terms of supporting long-term viability and proliferation within the 3-D scaffolds. Moreover, a high level of cell-to-cell contact is recognized as being a key mediator of adipogenesis [52e54]. While not a direct focus of the current study, another potential factor could be the influence of cell shape within the engineered microenvironments, as a more rounded cellular morphology has been shown to help direct mesenchymal stem cell differentiation toward the adipogenic lineage [55]. While in vitro studies are extremely useful for elucidating the key bioscaffold properties that can help direct stem cell proliferation and adipogenic differentiation, ultimately any biomaterial for

9

tissue-engineering applications must be tested in an appropriate animal model [56]. Our in vivo analysis corroborated our cell culture studies indicating that the MCS-based composites promoted adipogenesis. In contrast to our previous studies with the DAT bioscaffolds alone, seeding with allogenic rASCs was shown to enhance fat formation within the scaffolds at 12 weeks [11]. This difference may be related to the improved cell seeding efficiency with the encapsulation approach resulting in a larger exogenous ASC population at the time of implantation. Based on the promising results with the MCS-based composites, our future studies will include refining the biodegradation characteristics of the MCS, as well as optimizing DAT incorporation and the cell seeding density, to promote long-term volume retention following implantation for soft tissue augmentation. Further, we will explore the application of the composite technology for in situ cell encapsulation as a minimally-invasive injectable approach. 5. Conclusions Composite bioscaffolds incorporating decellularized tissue as a bioactive matrix prepared in situ through photo-cross-linking an injectable prepolymer carrier represent promising alternatives for tissue-specific stem cell delivery. MGC or MCS can be applied as cross-linkable delivery vehicles for cell encapsulation in combination with cryomilled cell-adhesive ECM, to support long-term cell viability within the engineered bioscaffolds. In our studies, the bulk mechanical characteristics of the constructs were tuned by varying the polymer formulation. Photo-cross-linking methods were applied to yield highly-hydrated scaffolds containing decellularized adipose ECM for soft tissue implantation. The incorporation of DAT within the hydrogel phase significantly enhanced ASC adipogenic differentiation, as assessed through GPDH enzyme activity, adipogenic gene expression and intracellular lipid accumulation. Factors that may have influenced the observed stem cell differentiation response include a microenvironment that provides tissue-specific biochemical and biomechanical cell-ECM interactions, as well as the promotion of cellecell interactions favorable for adipogenesis. The results indicate that a composite scaffold approach has great potential for adipose tissue engineering, which could be adapted into a minimally-invasive technology for the reconstruction of irregular soft tissue defects. Overall, the general encapsulation strategy could be applied with other decellularized ECM sources for the development of tissue-specific regenerative therapies, such as for cartilage or muscle regeneration. Acknowledgments Funding for this study was provided by the Canadian Institutes of Health Research (CIHR), with infrastructure support from the NSERC Research Tools and Instruments (RTI) program, the Canada Foundation for Innovation (CFI) and the Ontario Ministry of Research and Innovation (MRI). The authors would like to acknowledge Drs. K. Meathrel, J. Davidson, D. McKay, and Mrs. K. Martin for clinical collaborationsto support this work. Thank you to Dr. S. D. Waldman for access to his microplate reader. Mr. C. Brown, Mr. S. Young and Dr. J. Bianco are acknowledged for their technical assistance with the rat implantation surgeries. In addition, we would like to thank Dr. Françoise Sauriol for NMR guidance. References [1] Gomillion CT, Burg KJL. Stem cells and adipose tissue engineering. Biomaterials 2006;27(36):6052e63. [2] Vallee M, Cote JF, Fradette J. Adipose-tissue engineering: taking advantage of the properties of human adipose-derived stem/stromal cells. Pathol Biol (Paris) 2009;57(4):309e17.


10


[3] Bauer-Kreisel P, Goepferich A, Blunk T. Cell-delivery therapeutics for adipose tissue regeneration. Adv Drug Deliv Rev 2010;62(7e8):798e813. [4] Choi JH, Gimble JM, Lee K, Marra KG, Rubin JP, Yoo JJ, et al. Adipose tissue engineering for soft tissue regeneration. Tissue Eng Part B Rev 2010;16(4): 413e26. [5] Flynn L, Woodhouse KA. Adipose tissue engineering with cells in engineered matrices. Organogenesis 2008;4(4):228e35. [6] Hemmrich K, von Heimburg D. Biomaterials for adipose tissue engineering. Expert Rev Med Dev 2006;3(5):635e45. [7] Young DA, Christman KL. Injectable biomaterials for adipose tissue engineering. Biomed Mater 2012;7(2):024104. [8] Turner AE, Flynn LE. Design and characterization of tissue-specific extracellular matrix-derived microcarriers. Tissue Eng Part C Methods 2012;18(3): 186e97. [9] Flynn LE. The use of decellularized adipose tissue to provide an inductive microenvironment for the adipogenic differentiation of human adiposederived stem cells. Biomaterials 2010;31(17):4715e24. [10] Turner AE, Yu C, Bianco J, Watkins JF, Flynn LE. The performance of decellularized adipose tissue microcarriers as an inductive substrate for human adipose-derived stem cells. Biomaterials 2012;33(18):4490e9. [11] Yu C, Bianco J, Brown C, Fuetterer L, Watkins JE, Samani A, et al. Porous decellularized adipose tissue foams for soft tissue regeneration. Biomaterials 2013;34(13):3290e302. [12] Uriel S, Huang JJ, Moya ML, Francis ME, Wang R, Chang SY, et al. The role of adipose protein derived hydrogels in adipogenesis. Biomaterials 2008;29(27): 3712e9. [13] Brown BN, Freund JM, Han L, Rubin JP, Reing JE, Jeffries EM, et al. Comparison of three methods for the derivation of a biologic scaffold composed of adipose tissue extracellular matrix. Tissue Eng Part C Methods 2011;17(4):411e21. [14] Wang L, Johnson JA, Zhang Q, Beahm EK. Combining decellularized human adipose tissue extracellular matrix and adipose-derived stem cells for adipose tissue engineering. Acta Biomater 2013;9(11):8921e31. [15] Young DA, Ibrahim DO, Hu D, Christman KL. Injectable hydrogel scaffold from decellularized human lipoaspirate. Acta Biomater 2011;7(3):1040e9. [16] Poon CJ, Pereira ECMV, Sinha S, Palmer JA, Woods AA, Morrison WA, et al. Preparation of an adipogenic hydrogel from subcutaneous adipose tissue. Acta Biomater 2013;9(3):5609e20. [17] Kim IY, Seo SJ, Moon HS, Yoo MK, Park IY, Kim BC, et al. Chitosan and its derivatives for tissue engineering applications. Biotechnol Adv 2008;26(1): 1e21. [18] Knight DK, Shapka SN, Amsden BG. Structure, depolymerization, and cytocompatibility evaluation of glycol chitosan. J Biomed Mater Res A 2007;83(3): 787e98. [19] Ruoslahti E. Structure and biology of proteoglycans. Annu Rev Cell Biol 1988;4:229e55. [20] Funderburg FM, Markwald RR. Conditioning of native substrates by chondroitin sulfate proteoglycans during cardiac mesenchymal cell migration. J Cell Biol 1986;103(6 Pt 1):2475e87. [21] Simeon A, Wegrowski Y, Bontemps Y, Maquart FX. Expression of glycosaminoglycans and small proteoglycans in wounds: modulation by the tripeptidecopper complex glycyl-L-histidyl-L-lysine-Cu(2þ). J Invest Dermatol 2000;115(6):962e8. [22] Amsden BG, Sukarto A, Knight DK, Shapka SN. Methacrylated glycol chitosan as a photopolymerizable biomaterial. Biomacromolecules 2007;8(12):3758e66. [23] Li Q, Williams CG, Sun DD, Wang J, Leong K, Elisseeff JH. Photocrosslinkable polysaccharides based on chondroitin sulfate. J Biomed Mater Res A 2004;68(1):28e33. [24] Hayes WC, Herrmann G, Mockros LF, Keer LM. Mathematical-analysis for indentation tests of articular-cartilage. J Biomech 1972;5(5):541. [25] Flynn L, Semple JL, Woodhouse KA. Decellularized placental matrices for adipose tissue engineering. J Biomed Mater Res A 2006;79A(2):359e69. [26] Hauner H, Skurk T, Wabitsch M. Cultures of human adipose precursor cells. Methods Mol Biol 2001;155:239e47. [27] Flynn L, Prestwich GD, Semple JL, Woodhouse KA. Adipose tissue engineering with naturally derived scaffolds and adipose-derived stem cells. Biomaterials 2007;28(26):3834e42. [28] Gorzelniak K, Janke J, Engeli S, Sharma AM. Validation of endogenous controls for gene expression studies in human adipocytes and preadipocytes. Horm Metab Res 2001;33(10):625e7. [29] Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat Biotechnol 2005;23(1):47e55. [30] Peppas NA, Huang Y, Torres-Lugo M, Ward JH, Zhang J. Physicochemical, foundations and structural design of hydrogels in medicine and biology. Annu Rev Biomed Eng 2000;2:9e29.

[31] Williams CG, Malik AN, Kim TK, Manson PN, Elisseeff JH. Variable cytocompatibility of six cell lines with photoinitiators used for polymerizing hydrogels and cell encapsulation. Biomaterials 2005;26(11):1211e8. [32] Patrick CW. Breast tissue engineering. Annu Rev Biomed Eng 2004;6:109e30. [33] Smeds KA, Grinstaff MW. Photocrosslinkable polysaccharides for in situ hydrogel formation. J Biomed Mater Res A 2001;54(1):115e21. [34] Alhadlaq A, Tang M, Mao JJ. Engineered adipose tissue from human mesenchymal stem cells maintains predefined shape and dimension: implications in soft tissue augmentation and reconstruction. Tissue Eng 2005;11(3e4):556e66. [35] Ramires PA, Miccoli MA, Panzarini E, Dini L, Protopapa C. In vitro and in vivo biocompatibility evaluation of a polyalkylimide hydrogel for soft tissue augmentation. J Biomed Mater Res B 2005;72B(2):230e8. [36] Krueger EA, Wilkins EG, Strawderman M, Cederna P, Goldfarb S, Vicini FA, et al. Complications and patient satisfaction following expander/implant breast reconstruction with and without radiotherapy. Int J Radiat Oncol 2001;49(3):713e21. [37] Christensen L, Breiting V, Janssen M, Vuust J, Hogdall E. Adverse reactions to injectable soft tissue permanent fillers. Aesthet Plast Surg 2005;29(1):34e48. [38] Bryant SJ, Davis-Arehart KA, Luo N, Shoemaker RK, Arthur JA, Anseth KS. Synthesis and characterization of photopolymerized multifunctional hydrogels: water-soluble poly(vinyl alcohol) and chondroitin sulfate macromers for chondrocyte encapsulation. Macromolecules 2004;37(18):6726e33. [39] Bourin P, Bunnell BA, Casteilla L, Dominici M, Katz AJ, March KL, et al. Stromal cells from the adipose tissue-derived stromal vascular fraction and culture expanded adipose tissue-derived stromal/stem cells: a joint statement of the International Federation for Adipose Therapeutics and Science (IFATS) and the International Society for Cellular Therapy (ISCT). Cytotherapy 2013;15(6): 641e8. [40] Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell lineage specification. Cell 2006;126(4):677e89. [41] Lim JJ, Temenoff JS. The effect of desulfation of chondroitin sulfate on interactions with positively charged growth factors and upregulation of cartilaginous markers in encapsulated MSCs. Biomaterials 2013;34(21):5007e18. [42] Hardingham TE, Fosang AJ. Proteoglycans e many forms and many functions. Faseb J 1992;6(3):861e70. [43] Jeon O, Alsberg E. Photofunctionalization of alginate hydrogels to promote adhesion and proliferation of human mesenchymal stem cells. Tissue Eng Part A 2013;19(11e12):1424e32. [44] Flynn LE, Prestwich GD, Semple JL, Woodhouse KA. Proliferation and differentiation of adipose-derived stem cells on naturally derived scaffolds. Biomaterials 2008;29(12):1862e71. [45] Young DA, Choi YS, Engler AJ, Christman KL. Stimulation of adipogenesis of adult adipose-derived stem cells using substrates that mimic the stiffness of adipose tissue. Biomaterials 2013;34(34):8581e8. [46] Fathima NN, Suresh R, Rao JR, Nair BU. Effect of UV irradiation on the physicochemical properties of collagen stabilized using aldehydes. J Appl Polym Sci 2007;104(6):3642e8. [47] Badylak SE. The extracellular matrix as a scaffold for tissue reconstruction. Semin Cell Dev Biol 2002;13(5):377e83. [48] Wang DA, Varghese S, Sharma B, Strehin I, Fermanian S, Gorham J, et al. Multifunctional chondroitin sulphate for cartilage tissue-biomaterial integration. Nat Mater 2007;6(5):385e92. [49] Varghese S, Hwang NS, Canver AC, Theprungsirikul P, Lin DW, Elisseeff J. Chondroitin sulfate based niches for chondrogenic differentiation of mesenchymal stem cells. Matrix Biol 2008;27(1):12e21. [50] Silva JM, Georgi N, Costa R, Sher P, Reis RL, Van Blitterswijk CA, et al. Nanostructured 3D constructs based on chitosan and chondroitin sulphate multilayers for cartilage tissue engineering. PLos One 2013;8(2). [51] Calvo JC, Rodbard D, Katki A, Chernick S, Yanagishita M. Differentiation of 3t3L1 preadipocytes with 3-isobutyl-1-methylxanthine and dexamethasone stimulates cell-associated and soluble chondroitin 4-sulfate proteoglycans. J Biol Chem 1991;266(17):11237e44. [52] Lu HX, Guo LK, Wozniak MJ, Kawazoe N, Tateishi T, Zhang XD, et al. Effect of cell density on adipogenic differentiation of mesenchymal stem cells. Biochem Biophs Res Commun 2009;381(3):322e7. [53] Tang J, Peng R, Ding JD. The regulation of stem cell differentiation by cell-cell contact on micropatterned material surfaces. Biomaterials 2010;31(9):2470e6. [54] Gobaa S, Hoehnel S, Roccio M, Negro A, Kobel S, Lutolf MP. Artificial niche microarrays for probing single stem cell fate in high throughput. Nat Methods 2011;8(11):949e55. [55] McBeath R, Pirone DM, Nelson CM, Bhadriraju K, Chen CS. Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev Cell 2004;6(4):483e95. [56] Patrick CW, Uthamanthil R, Beahm E, Frye C. Animal models for adipose tissue engineering. Tissue Eng Part B Rev 2008;14(2):167e78.
