A and B, Scanning electron microscopy detail of the random fiber arrangement of the scaffolds (reproduced with permission from Nanofiber Solutions Inc).
After overnight incubation with poly(L)-lysine/laminin solution, scaffolds were overlaid with 2.5 to 3.5 × 105 rabbit adipose mesenchymal cells and incubated for 48 hours to allow for cellular adhesion. DNA content was measured with DNA-QF reagent (Sigma), and cell number and growth area were determined as described in the Methods section. Error bars indicate means (SDs). A-MEM indicates advanced minimum essential media; PCL indicates polycaprolactone; PDO, polydioxanone; PHBV, poly-3-hydroxybutyrate-co-3-hydroxyvalerate; PLCL, poly(L-lactide-co-caprolactone); PLGA, poly(lactic-co-glycolic acid); PS, polystyrene.
A total of 2.5 to 3.5 × 105 cells/scaffold were incubated with regular Advanced Minimum Essential Media (A-MEM) media or A-MEM plus chondrogenic supplement for 28 days, digested with papain to remove proteins and analyzed for DNA and sulfated glycosaminoglycan (sGAG) content as described in the Methods section. Because scaffold volumes differed owing to material thickness (Figure 1A), data were normalized to unit volume. A, DNA content analysis. B, Fold change in normalized sGAG content (sGAG per DNA) for scaffolds. PCL indicates polycaprolactone; PDO, polydioxanone; PHBV, poly-3-hydroxybutyrate-co-3-hydroxyvalerate; PLCL, poly(L-lactide-co-caprolactone); PLGA, poly(lactic-co-glycolic acid); PS, polystyrene. incubated with chondrogenic supplement relative to those incubated with media alone. Asterisks indicate statistical significance.
Scaffolds were fixed in 10% formalin, paraffin-embedded, and sectioned into 5-µm slices. After deparaffinization with xylene-ethanol solution, the slides were stained with nuclear fast red for DNA and Alcian blue (Sigma) to reveal chondrogenic matrix. Error bars indicate means (SDs). PCL indicates polycaprolactone; PDO, polydioxanone; PHBV, poly-3-hydroxybutyrate-co-3-hydroxyvalerate; PLGA, poly(lactic-co-glycolic acid); PS, polystyrene.
eFigure. Overall experimental layout
Customize your JAMA Network experience by selecting one or more topics from the list below.
San-Marina S, Sharma A, Voss SG, Janus JR, Hamilton GS. Assessment of Scaffolding Properties for Chondrogenic Differentiation of Adipose-Derived Mesenchymal Stem Cells in Nasal Reconstruction. JAMA Facial Plast Surg. 2017;19(2):108–114. doi:10.1001/jamafacial.2016.1200
Copyright 2016 American Medical Association. All Rights Reserved.
Are there differences in chondrogenic differentiation of adipose mesenchymal stem cells grown on different electro spun scaffold matrices with similar porosity and fiber alignment features?
Coating scaffolds with poly(L)-lysine/laminin was necessary for efficient cell saturation of scaffold surfaces. Polydioxanone and poly-3-hydroxybutyrate-co-3-hydroxyvalerate (PHBV)-polycaprolactone scaffolds supported chondrogenic fate commitment better than PHBV, poly(L-lactide-co-caprolactone) poly(lactic-co-glycolic acid), and polystyrene based on soluble sulfated glycosaminoglycan analysis and microscopic observation of chondrogenic matrix deposition.
Materials, reagents and protocols were identified for future consideration in tissue engineering experiments for nasal reconstruction using pro-chondrogenically differentiated adipose mesenchymal stem cells.
Nasal reconstruction in patients who are missing a significant amount of structural nasal support remains a difficult challenge. One challenge is the deficiency of cartilage left within the nose as a consequence of rhinectomy or a midline destructive disease. Historically, the standard donor source for large quantities of native cartilage has been costal cartilage.
To enable the development of protocols for new mesenchymal stem cell technologies as alternative procedures with reduced donor site morbidity, risk of infection and extrusion.
Design, Setting, and Materials
We examined 6 popular scaffold materials in current practice in terms of their biodegradability in tissue culture, effect on adipose-derived mesenchymal stem cell growth, and chondrogenic fate commitment. Various biomaterials of matching size, porosity, and fiber alignment were synthesized by electrospinning and overlaid with rabbit adipose-derived mesenchymal cells in media supplemented or not with chondrogenic factors. Experiments were performed in vitro using as end points biomarkers for cell growth and chondrogenic differentiation. Polydioxanone (PDO), poly-3-hydroxybutyrate-co-3-hydroxyvalerate (PHBV), PHBV-polycaprolactone, poly(L-lactide-co-caprolactone), poly(lactic-co-glycolic acid), and polystyrene scaffolds of 60% to 70% porosity and random fiber alignment were coated with poly(L)-lysine/laminin to promote cell adhesion and incubated for 28 days with 2.5 to 3.5 × 105 rabbit adipose mesenchymal cells.
Main Outcomes and Measures
Cell growth was measured by fluorometric DNA quantitation and chondrogenic differentiation of stem cells by spectrophotometric sulfated glycosaminoglycan (sGAG) assay. Microscopic visualization of cell growth and matrix deposition on formalin-fixed, paraffin-embedded tissue sections was performed, respectively, with nuclear fast red and Alcian blue.
Of 6 scaffold materials tested using rabbit apidose mesenchymal cells, uncoated scaffolds promoted limited cell adhesion but coating with poly(L)-lysine/laminin enabled efficient cell saturation of scaffold surfaces, albeit with limited involvement of scaffold interiors. Similar growth rates were observed under these conditions, based on DNA content analysis. However, PDO and PHBV/PCL scaffolds supported chondrogenic fate commitment better than other materials, based on soluble sGAG analysis and microscopic observation of chondrogenic matrix deposition. The mean (SD) sGAG scaffold values expressed as fold increase over control were PDO, 2.26 (0.88), PHBV/PCL, 2.09 (0.83), PLCL, 1.36 (0.39), PLGA, 1.34 (0.77), PHBV, 1.07 (0.31), and PS, 0.38 (0.14).
Conclusions and Relevance
These results establish materials, reagents, and protocols for tissue engineering for nasal reconstruction using single-layer, chondrogenically differentiated, adipose-derived mesenchymal stem cells. Stackable, scaffold-supported, multisheet bioengineered tissue may be generated using these protocols.
Level of Evidence
Nasal reconstruction is an extremely challenging surgery, especially when a significant amount of the nasal structural framework is missing. Septal cartilage remains the ideal source of material for nasal reconstruction. It is thin, elastic, and typically strong. However, in patients with catastrophic nasal deformities it is almost always missing. Auricular and costal cartilages have been used to recreate the nasal framework, but both donor sources have some inherent drawbacks. Auricular cartilage is curved, soft, friable, and in relatively short supply. Costal cartilage is abundant but is often brittle and calcified in older patients. If improperly carved, costal cartilage is prone to warping, if not immediately then over time. Therefore, the ideal reconstructive material for patients who need large quantities of cartilage would be a block of bioengineered cartilage with mechanical quality similar to that of nasal septal cartilage.1
Recent advances in regenerative medicine have begun to lay the groundwork for growing cartilage with histologic, functional, and mechanical qualities of native nasal septal cartilage.2-6 Typically, the process consists of attaching cells to a scaffold and allowing them to expand and secrete matrix components while the scaffold itself dissolves. Some of the desired characteristics of an ideal chondrogenic scaffold are listed in the Box.1,7
Optimum cellular adhesion.
Aids in chondrogenic differentiation.
Made of minimally immunogenic material(s).
Dissolves in vivo to be completely replaced by native hyaline cartilage.
To date, preferred cellular sources for bioengineered cartilage have been mature chondrocytes harvested from donor sites and expanded ex vivo and mesenchymal-derived stem cells treated with growth factors to promote chondrocytic differentiation.8 Induced pluripotent stem cells or other pluripotent cell types have been used to a lesser extent.9 The primary challenge is that, in tissue culture, mature chondrocytes grown on flat surfaces adopt fibroblast-like morphologic characteristics and undergo dedifferentiation.10 Similarly, adipose-derived mesenchymal stem cells require specialized tissue culture conditions to differentiate into chondrocytes, such as inclusion of alginate beads or formation of high-density pellets.11 Thus, for chondrocytes and mesenchymal-derived stem cells the choice of a scaffold surface with appropriate mechanical strength and surface topography is critical for promoting optimal chondrogenesis.12,13
Cartilage scaffold engineering has been an active area of research, with 2 dominant streams: 3-dimensional (3-D) printing and electrospinning. While 3-D printing is able to generate novel 3-D structures, electrospinning offers the additional advantage of better simulating the extracellular environment on a nanoscale.14 Electrospinning uses a high voltage applied to a liquid droplet, resulting in repulsive electrostatic forces. These repulsive forces deform the droplet into a fine stream of liquid, which rapidly dries in to a nanometer-scale fiber. Research into electrospun scaffolds has generated a range of materials with advantageous properties for growing chondrocytes and adipose-derived stem cells,1,2,4,6,7,15 and data for prioritizing in vivo scaffold usage for chondrocytes are starting to emerge.16 However, to our knowledge, a direct head-to-head comparison between existing materials to determine their suitability for cartilage bioengineering using adipose-derived stem cells has not been performed. In this study, we used adipose-derived mesenchymal stem cells to compare 6 different types of popular electrospun scaffold materials with closely matched design parameters and to determine the material effect on cellular adhesion, cartilage formation, and scaffold dissolution. Adipose-derived cells were selected because of their established ability to differentiate into cartilage, the abundant supply of material, and the less controversial nature of their use when compared with those of embryonic origin.
Institutional review board approval was not needed for the study because no human material was used. Rabbit mesenchymal stem cells (Institutional Animal Care and Use Committee No. A52714) were prepared by digestion with collagenase type 1 (C0130-1G; Sigma) as previously described.21 Briefly, 2 to 4 g of subcutaneous fat from the neck area were minced with 3 to 5 mL of 0.1% to 0.15% collagenase 1 solution in advanced minimum essential media (A-MEM) (Life Technologies) containing 10% fetal bovine serum (FBS), 1% glutamax and 1.0 to 1.5 mg/mL penicillin-streptomycin antibiotics. Following 1.0 to 1.5 hours incubation at 37°C with occasional shaking, cells and tissue debris were pelleted by 5 minutes of centrifugation at 500g. The pellet was washed once with Dulbecco phosphate-buffered saline, strained through a 70-µm sieve to remove large particulate matter, and recovered by centrifugation for 5-minute at 500g. The resuspended pellets containing a variety of cell types, including adipose stem cells, were then transferred to appropriate size tissue culture containers and incubated overnight with 5 to 20 mL of A-MEM supplemented with 10% FBS plus glutamax and penicillin-streptomycin solution. The following day, fresh media were added. Adhered cells were allowed to reach 60% to 80% confluence prior to cryogenic storage. Preparations from 3 rabbits were pooled. Fresh aliquots containing approximately 1 × 106 cells were thawed out on the day of the experiment.
Scaffolds were sterilized in 70% ethanol under UV radiation and then washed twice with cold phosphate-buffered saline (PBS). To promote cellular adhesion, sterile scaffolds were incubated overnight with 200 µL of 0.01% poly(L)-lysine containing 4 µg/mL mouse laminin and washed with PBS. This protocol was developed to promote adhesion of neuronal cells on coverslips and was successfully adapted in our laboratory to culture adipose mesenchymal stem cells on synthetic scaffold surfaces. They were then placed in 96 wells, overlaid with 250 µL of 1.0 to 1.5 × 106 cells/mL rabbit adipose mesenchymal cells, and incubated for 48 hours to allow for cellular adhesion. The overall experimental layout is shown in the eFigure in the Supplement. In experiments in which the goal was to optimize cell adhesion, the scaffolds were directly fixed with 150 µL of 10% formalin after 48 hours of incubation. In experiments in which the effect of chondrogenic supplements was investigated, scaffolds were incubated for 28 days with changes of 200 µL of media containing chondrogenic factors every other day. Basic media for the cellular retention experiments consisted of A-MEM. For chondrogenesis experiments, a chondrogenic supplement from Life Technologies was added to basic media. All media contained 10% FBS, 2 mM L-alanyl-L-glutamine (Glutamax; Life Technologies), and 10.0 U/mL of penicillin, 10.0 µg/mL of streptomycin, and 0.25 µg/mL of Fungizone antimycotic (Life Technologies). After 28 days the scaffolds were either fixed in 10% formalin and processed by paraffin embedding and sectioning or minced and processed for sulfated glycosaminoglycan (sGAG) and DNA analyses. For microscopic analysis, fixed tissue was paraffin-embedded and cut into 5-µm sections and processed for biological staining. Slides were deparaffinized with a series of xylene and ethanol washes as follows: 100% xylene (2 × 3 minutes), 1:1 xylene-ethanol (1 × 3 minutes), 100% ethanol (2 × 3 minutes), 95% ethanol (1 × 3 minutes), 75% ethanol (1 × 3 minutes), and finally 50% ethanol (1 × 3 minutes) followed by rehydration in a water bath. Cellular content was routinely monitored by hematoxylin-eosin staining. Alcian blue (Sigma, A5268) and nuclear fast red (Sigma-F4648) staining was performed as previously described (PromoCell). For sGAG and DNA content analysis, scaffolds were digested overnight with 1.0 mL of a papain solution containing 0.2M of sodium phosphate buffer (pH 6.4), 8 mg/mL of sodium acetate, 4 mg/mL of disodium EDTA, 0.8 mg/mL of cysteine–hydrochloride, and 250 µL of a papain suspension, containing about 5 mg of enzyme (Sigma, P3125), according to the Blyscan protocol (Biocolor). sGAG was measured with Blyscan reagent by colorimetric absorption at 656 nm and DNA with DNA-QF reagent (Sigma) by fluorescence spectrometry with excitation and emission wavelengths of 360 nm and 460 nm, respectively. Cell number calculations were performed using the formula: 3 × 109 bp × 2 (diploid) × 660 (average molecular weight of 1 bp) × 1.67 × 10−12 pg (weight in Daltons) = 6.6 pg/diploid primary cell.
Data collected from duplicate experiments with 2 to 3 scaffolds per condition are illustrated as means (SDs). Comparisons of group means were performed with t test for unpaired samples at P ≤ .05 or lower for biological significance. Data analysis was performed using GraphPad 2016 software.
Scaffolds generated by electrospinning were 60% to 70% porous and had a random fiber distribution but no set pore size. While all scaffolds were 10 mm in diameter, PDO, poly-3-hydroxybutyrate-co-3-hydroxyvalerate (PHBV), and PHBV-PCL scaffolds were 1 mm thick and poly(L-lactide-co-caprolactone) (PLCL), poly(lactic-co-glycolic acid) (PLGA), and polystyrene (PS) scaffolds were 2 mm thick (Table and Figure 1). In initial experiments, cells applied directly to sterile scaffolds and maintained for 48 hours in incubation media proliferated poorly owing to lack of adhesion. Furthermore, uncoated and/or uncellularized PDO, PLCL, PLGA, and PS scaffolds underwent different degrees of fragmentation as part of their dissolution process. In the hope of improving cell adhesion while decreasing fragmentation, scaffolds were coated overnight with a poly(L)-lysine/laminin solution. This step improved cell adhesion to near-complete saturation of the available scaffold area (Figure 2). Furthermore, coating decreased fragmentation. After 28 days in media only the PLCL scaffold underwent significant fragmentation.
Having achieved optimal cell adhesion, we next investigated chondrogenic differentiation on each scaffold after 28 days in cell culture in the presence of the chondrogenic supplement. After digestion with papain solution to remove proteins, scaffolds were analyzed for DNA as a proxy for cell growth and for sGAG as an indicator of matrix deposition (Figure 3). DNA data normalized to scaffold volume showed that 1-mm scaffolds (PDO, PHBV, and PHBV-PCL) contained more DNA per unit volume than the 2-mm scaffolds (PLCL, PLGA, and PS), suggesting that cellular proliferation occurred mostly at the surface rather than inside the scaffold body. DNA content analysis also showed that, with the exception of the PS scaffold, no statistically significant differences in cell proliferation were observed between incubations performed with media or chondrogenic supplement (Figure 3A). Thus, replicative senescence, the process by which adipose-derived mesenchymal stem cells cultured for extended periods eventually fail to expand,17 likely did not play an important role in the observed results. Highest mean (SD) sGAG values expressed as fold increase over control were observed for PDO (2.26 [0.88]) and PHBV-PCL (2.09 [0.83]) followed by PLCL (1.36 [0.39]), PLGA (1.34 [0.77]), PHBV (1.07 [0.31]), and PS (0.38 [0.14]) scaffolds. In the case of the PS scaffold, the amount of sGAG actually decreased at the end of incubation relative to the control (Figure 3B), suggesting that cells cultured on this material may have undergone dedifferentiation. To obtain confirmation, cells were directly cultured on 96-well polystyrene plates and assayed for sGAG synthesis. A similar pattern of reduced sGAG production was observed (Figure 3B) in keeping with previous observations that PS promotes chondrogenic dedifferentiation.18-20
To obtain visual confirmation of matrix deposition, scaffolds were sectioned into 5-µm slices and stained with nuclear fast red for nucleic acids and Alcian blue for glycoproteins-glycosaminoglycan residues (Figure 4). PLCL scaffolds fragmented during processing and were unavailable for staining. Of the 5 scaffolds left, PDO, PHBV-PCL, and PHBV showed the greatest relative amount of matrix deposition on incubation with the chondrogenic supplement. Cells grown on the PS scaffold did not show evidence of matrix deposition, in agreement with the results of the soluble sGAG assay.
The phenomenon of adipose-derived mesenchymal stem cell differentiation has been extensively studied since it was first recognized in 2001 by Zuk et al.21 Since then, a variety of materials and biochemical approaches have been considered to maximize chondrogenic differentiation. An important concern in stem cell protocols for targeted differentiation is to provide a scaffolding material with optimal topographical and physicochemical properties. Both elasticity modulus and surface features of the seeding surface can have an impact on the direction of differentiation (ie, chondrogenic, osteogenic, adipogenic, or neurogenic) as well as its extent.12,13,22 Because of the high rate of innovation in materials science, attention has been focused primarily on testing new polymers rather than on head-to-head comparative studies of existing scaffolds. In this study, electrospun scaffolds with a standardized narrow range of porosities and fiber alignment were generated based on previously identified advantageous properties, such as mechanical stiffness, biocompatibility, biodegradability, and potential for cartilage engineering. PLGA is a biocompatible suture material that has been used extensively in cartilage tissue engineering.6 PLCL has been used in vascular tissue engineering and has mechanical and elastic properties that have also shown promise in cartilage bioengineering.23,24 PDO is another material with low immunogenicity used to create slowly degrading surgical sutures. This material has been used in vascular scaffolds and hybrid chondrogenic scaffolds.25,26 PDO sutures are regularly used in cartilage fixation in rhinoplasty. Polystyrene has numerous in vivo applications and is routinely used as a scaffold for 2-D cell culture techniques. Electrospun 3-D polystyrene scaffolds provide a comparative standard for other scaffolding materials. PHBV is a slowly degrading biocompatible bioplastic most commonly created by bacteria. PHBV is a brittle and hydrophobic substance that has been shown to promote chondrogenesis.27,28 Cellular attachment to modified PHBV scaffolds is controversial. While in some studies hybrid PHBV scaffolds with increased hydrophilicity reportedly improved cellular adhesion, other studies showed that increased hydrophilicity does not result in better adhesion.29 PHBV-PCL hybrid scaffolds were chosen to address the usefulness of hybrid scaffolds with increased hydrophilicity. PCL also confers improved mechanical properties to PHBV while possibly retaining better biodegradability relative to pure PHBV scaffolds.29,30 In results consistent with observations that scaffolds inherently provide poor adhesion surfaces and thus impede cell growth, we found that few cells were retained to uncoated scaffolds after 48 hours of incubation. In our study, the PHBV-PCL hybrid scaffold was not superior in this respect to other scaffolds. In all cases, coating with poly(L)lysine/laminin maximized cell adhesion and growth and decreased scaffold fragmentation. This is a novel discovery that we hope will translate into scaffold improvement.
Murphy and O’Brien31 have shown that pore size is an important consideration in scaffold design because a balance must be achieved between allowing optimal cell attachment and facilitating cell migration into the scaffold core. Thus, a high fiber density will increase cell adhesion while at the same time decrease pore size and therefore limit cell growth. In the present study, a random fiber alignment, rather than a set pore size design, was chosen to generate scaffolds with a controlled range of 60% to 70% porosity. Data suggest that cellular proliferation inside the scaffold was minimal under these conditions, as the amount of DNA normalized to scaffold volume was lower for 2-mm scaffolds than for 1-mm scaffolds (Figure 4).
It is hard to gauge the role this random fiber pattern played on the observed results. Lamellar surface patterns were shown to promote osteogenic differentiation while hexagonal patterns promoted adipogenic differentiation.22 For chondrogenic differentiation, exposed residues may be at least as important as surface features because exposed hyaluronic acid residues in polymerized hyaluronic acid methacrylate hydrogels increased by approximately 40-fold the chondrogenic commitment of mesenchymal stem cells over inert poly(ethylene glycol) (PEG) hydrogels.32 Furthermore, the necessity of providing a rigid matrix is challenged by observations that tissue stretching is a principal determinant of chondrogenic commitment for mesenchymal stem cells.19 In our study, DNA content analysis showed that poly(L)-lysine/laminin coating minimized materials differences with respect to cell growth after but did not interfere with their effect on chondrogenic fate commitment. PDO and PHBV-PCL scaffolds had the highest sGAG content relative to controls and histologically displayed relatively large amounts of structured matrix deposition. A similar distinction between the ability of different synthetic materials to promote sGAG synthesis by porcine chondrocytes was previously reported.16 Recent observations challenge long-held beliefs that biochemical supplements and serum-based growth factors “do the heavy lifting of induced differentiation.”13 Increasingly, polymeric substrates are likewise shown to affect stem cell fate decisions by their binding affinity, molecular flexibility, stiffness, or degradability. In experiments in which small functional groups were attached to the surface of inert PEG hydrogels, carbonyl residues, such as those present on PDO and PHBV polymers, were shown to promote chondrogenic fate commitment of mesenchymal stem cells while exposed phosphates preponderantly enabled osteogenic differentiation.33 Histological analyses confirmed that PDO and PHBV-PCL scaffolds preeminently supported chondrogenic differentiation resulting in a greater extent of matrix organization relative to other scaffolding materials (Figure 4).
Future studies are necessary to determine the effects of extending the length of chondrogenic differentiation and supplementation with growth factors on cell proliferation and matrix deposition. Furthermore, it is important to optimize pore size to maximize cell growth.31
These results provide a useful head-to-head comparison of some of the principal scaffold materials in common use today and demonstrate the interaction between surface topology, physicochemical materials properties, and soluble factors in chondrogenic differentiation of adipose-derived mesenchymal stem cells. Scaffold materials and reagents have been identified that may serve as useful biomaterials toward the goal of producing stackable sheets of cartilage tissue for nasal reconstructive surgery.
Corresponding Author: Serban San-Marina, MD, PhD, Department of Otolaryngology, Mayo Clinic, 200 1st St SW, Rochester, MN 55902 (firstname.lastname@example.org).
Accepted for Publication: July 25, 2016.
Published Online: October 13, 2016. doi:10.1001/jamafacial.2016.1200
Author Contributions: Drs San-Marina and Sharma had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. They contributed equally to the manuscript.
Concept and design: San Marina, Sharma, Janus, Hamilton.
Acquisition, analysis, or interpretation of data: San Marina, Sharma, Voss.
Drafting of the manuscript: San Marina, Sharma, Janus, Hamilton.
Critical revision of the manuscript for important intellectual content: San Marina, Voss, Janus, Hamilton.
Statistical analysis: San Marina.
Obtaining funding: Sharma, Janus, Hamilton.
Administrative, technical, or material support: All authors.
Study supervision: San Marina, Sharma, Janus.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by an American Academy of Facial Plastic and Reconstructive Surgery (AAFPRS) Clinical Investigator Award to Dr Hamilton.
Role of the Funder/Sponsor: The AAFPRS had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: