Patrascu Et Al. Polyglycolic Acid-Hyaluronan Scaffolds Loaded with Bone Marrow-Derived Mesenchymal Stem Cells Show

Home , Diaphysis

Polyglycolic acid-hyaluronan scaffolds loaded with bone marrow- derived mesenchymal stem cells show chondrogenic differentiation in vitro and cartilage repair in the rabbit model

Jenel M. Patrascu,1* Jan Philipp Kruger,€ 2* Hademar G. Boss,€ 1 Anna-Katharina Ketzmar,2 Undine Freymann,2 Michael Sittinger,3,4 Michael Notter,5 Michaela Endres,2,3 Christian Kaps2 1Department of Orthopaedic Surgery, V. Babes University of Medicine and Pharmacy, Timisoara, Romania 2TransTissue Technologies GmbH, Chariteplatz 1, 10117 Berlin, Germany 3Department of Rheumatology, Tissue Engineering Laboratory, Charite Campus Mitte, Charite - Universitatsmedizin€ Berlin, Chariteplatz 1, 10117, Berlin, Germany 4Berlin-Brandenburg Center for Regenerative Therapies, Charite-Universit atsmedizin€ Berlin, Augustenburger Platz 1, 13353 Berlin 5Department of Hematology and Oncology, Charite-Universitatsmedizin€ Berlin, Hindenburgdamm 30, 12200 Berlin, Germany

Received 5 November 2012; revised 28 January 2013; accepted 6 March 2013 Published online 10 May 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.32944

Abstract: In cartilage repair, scaffold-assisted one-step chondrogenic marker genes like type II collagen and formation approaches are used to improve the microfracture (Mfx) tech- of hyaline-like cartilaginous tissue in MSC-laden PGA-HA nique. Since the number of progenitors in Mfx is low and may implants. Histological evaluation of rabbit repair tissue forma- further decrease with age, aim of our study was to analyze the tion after 30 and 45 days showed formation of repair tissue, chondrogenic potential of freeze-dried polyglycolic acid-hyalur- rich in chondrocytic cells and of a hyaline-like appearance. onan (PGA-HA) implants preloaded with mesenchymal Controls showed no articular resurfacing, tissue repair in the stem cells (MSCs) in vitro and in a rabbit articular cartilage subchondral zone and fibrin formation. These results suggest defect model. Human bone marrow-derived MSC from iliac that MSC-laden PGA-HA scaffolds have chondrogenic potential crest were cultured in freeze-dried PGA-HA implants for and are a promising option for stem cell-mediated cartilage chondrogenic differentiation. In a pilot study, implants were regeneration. VC 2013 Wiley Periodicals, Inc. J Biomed Mater Res Part loaded with autologous rabbit MSC and used to cover 5 mm 3 B: Appl Biomater 101B: 1310–1320, 2013. 6 mm full-thickness femoral articular cartilage defects (n 5 4). Untreated defects (n 5 3) served as controls. Gene expression Key Words: cartilage repair, cartilage regeneration, cell-free analysis and histology showed induction of typical implant, microfracture, stem cells

How to cite this article: Patrascu JM, Kruger€ JP, Boss€ HG, Ketzmar A-K, Freymann U, Sittinger M, Notter M, Endres M, Kaps C. 2013. Polyglycolic acid-hyaluronan scaffolds loaded with bone marrow-derived mesenchymal stem cells show chondrogenic differentiation in vitro and cartilage repair in the rabbit model. J Biomed Mater Res Part B 2013:101B:1310–1320.

INTRODUCTION or Mfxs. In Mfx, repair tissue formation by mesenchymal In cartilage repair, a variety of techniques evolved that aim stem cells (MSCs) may be stimulated by growth and differ- at cartilage resurfacing and=or regeneration. These techni- entiation factors from the subchondral bone and=or the sy- ques comprise debridement, bone marrow stimulation (e.g., novial fluid.5,8,9 Although clinical studies demonstrated that abrasion, drilling, and microfracturing), osteochondral graft- the Mfx technique shows good results in the short and mid- ing or mosaicplasty, and autologous chondrocyte implanta- term with up to 5 years follow-up,10 the repair tissue tion with or without the use of scaffolds.1–5 The induced by Mfxs has been shown to be of mostly fibrocarti- microfracture (Mfx) technique is a frequently used and cost laginous appearance with limited short-term durability.11 effective first-line treatment option for the repair of focal Apparently, the Mfx treatment shows good short-term cartilage defects.6,7 In Mfx, bleeding of the subchondral results, but clinical results may be variable in the mid to bone is induced by the introduction of multiple perforations long terms. Apart from that, the technique may be limited

*These authors contributed equally to this work. JPK, UF, ME, and CK are employees of TransTissue Technologies GmbH (TTT). TTT develops scaffold-based cartilage repair treatment strategies and developed the PGA-HA scaffold. CK is consultant of BioTissue AG that distributes the PGA-HA based chondrotissue cartilage implant. MS is consultant of BioTissue Technologies GmbH that produces the PGA-HA scaffold. Correspondence to: C. Kaps; e-mail: [email protected] Contract grant sponsor: European Union, EU-FP7 program; contract grand number: TissueGEN: HEALTH-F4-2011-278955

1310 VC 2013 WILEY PERIODICALS, INC. ORIGINAL RESEARCH REPORT by its indication for relatively small focal defects with an 100 U=mL penicillin, 100 mg=mL streptomycin, and 10% intact defect shoulder surrounding the defect. human serum and transferred to cell culture flasks. Medium Recently, scaffold-assisted one-step approaches evolved was exchanged after 72 h and then every 2 days thereafter. that may improve cartilage repair by enhancing repair tissue Cells were allowed to grow to 90% confluence, detached by formation in microfracturing and may extend the indication 0.05% trypsin–ethylenediaminetetraacetic acid (Biochrom) even to partly unshouldered defects by covering the micro- in phosphate-buffered saline (PBS; Biochrom) and used fractured defect with, for example, porcine collagen scaffolds directly for further analysis. with fibrin12,13 or resorbable polyglycolic acid-hyaluronan (PGA-HA) scaffolds.14,15 PGA-HA scaffolds have been shown Fluorescence-activated cell sorting to recruit MSC into the scaffold and guide them toward car- For characterization of MSC, typical MSC-related cell surface tilage repair, with improvement of cartilage repair tissue antigens were analyzed. MSC (2.5 3 105 cells, passage 0; n formation compared to microfracturing alone.16 Clinically, in 5 3 donors) were washed in PBS=0.5% bovine serum albu- a group of 52 patients, implantation of PGA-HA scaffolds for min and incubated with monoclonal mouse anti-human anti- treatment of cartilage defects improved the patients’ situa- bodies CD34-phycoerythrin (PE), CD45-fluorescein- tion as shown by the Knee injury and Osteoarthritis Out- isothiocyanate (FITC), CD73-PE, CD90-FITC, CD105-FITC, come Score (KOOS) at 1 year follow-up and formed hyaline- and CD166-PE, conjugated with FITC or PE for 15 min on like cartilage tissue as assessed by second-look biopsies.17 ice. Staining of cell surface antigens was analyzed using the However, from the technical and cell biology point of view, fluorescence-activated cell sorting Calibur equipped with variability in the mid- to long-term outcome may be, for CELLQUEST software (Becton Dickinson). Apoptotic cells example, due to different depths of the Mfxs,18 use of the were excluded from analysis using propidium iodide. CD34 drilling or the Mfx procedure19 and maybe more impor- stained cells served as isotypic control. tantly due to a reduced availability of MSC with chondrogenic capacity in the subchondral bone marrow. In Three-dimensional tissue culture of human MSC particular, with increasing age, the number of MSC in the in polymer scaffolds bone marrow is reduced dramatically20 and in early osteo- The chondrotissue matrix was prepared as described previ- arthritis, for instance, there may be a reduced potential of ously.29 In brief, resorbable PGA scaffolds (BioTissue AG, cells to induce hyaline-like cartilage repair.21 To overcome Zurich, Switzerland) of 8 mm 3 8mm3 0.5 mm were these hurdles, the enrichment of scaffolds with autologous immersed with 32 mL hyaluronic acid (HA, OstenilVR ; TRB bone marrow concentrate or cells are currently investigated Chemedica AG, Germany) and subsequently freeze-dried for to improve and=or extend the Mfx technique and=or carti- 16 h using a lyophilisator (Leybold-Heraeus, Germany). MSC lage repair.22,23 A recent report about bone marrow concen- (passage 0, n 5 3 donors) were seeded into the PGA-HA trate has shown that cells within the concentrate grow on scaffolds at a density of 2 3 107 viable cells=mL by resus- the HA-based scaffold and are able to undergo chondrogenic pending 0.64 3 106 cells in 21 mL cell culture medium and differentiation.24 In addition, expanded bone marrow and=or 11 mL fibrinogen (Tissucol, Baxter International). Polymer- adipose derived stem cells have been shown to differentiate ization of the fibrinogen was achieved by adding thrombin along the chondrogenic lineage in, for example, platelet-rich (10%, v=v in PBS; Tissucol) and incubation at 37 C for 15 plasma (PRP)-derived scaffolds, porous poly(ethylene min. glycol) diglycidyl ether-cross-linked HA, poly-L-lactide-co- epsilon-caprolactone (PLCL)/chitosan scaffolds, or RGD- Differentiation of human MSC in polymer scaffolds polyhydroxyalkanoate scaffolds.25–28 We hypothesize that For chondrogenic differentiation, MSC-PGA-HA scaffolds (n bone marrow-derived MSC show chondrogenic differentia- 5 4 per donor; n 5 3 donors) were cultured up to 14 days tion in clinically applicable PGA-HA scaffolds in vitro and in DME medium containing 1% Insulin–Transferrin–Sele- that implantation of these cells in PGA-HA scaffolds into nium 1 1, 1 mM sodium pyruvate, 0.35 mM L-proline, 0.17 full-thickness articular cartilage defects leads to hyaline-like mM L-ascorbic acid-2-phosphate, and 0.1 mM dexamethasone cartilage repair in the rabbit model. (all Sigma-Aldrich, St. Louis, MO) and 10 ng=mL transforming growth factor-b3 (TGFB3; Peprotech). MSC-PGA-HA scaf- MATERIALS AND METHODS folds in DME medium without TGFB3 served as controls. Isolation and culture of human MSC Medium was exchanged every 2–3 days. Human adult MSC were isolated from iliac crest bone marrow aspirates of healthy donors (n 5 11, 3 female, 8 male, Real-time polymerase chain reaction age 43–72). Bone marrow samples were derived from Total RNA was isolated from MSC monolayer cultures (pas- donors who were examined to exclude hematopoietic neo- sage 0, n 5 5 donors) at 90% confluence and from MSC cul- plasmas and were histologically diagnosed as normal. The tured in PGA-HA-scaffolds (n 5 6) as described.30 study was approved by the ethical committee of the Subsequently, total RNA (3 mg) was reversely transcribed Charite-Universit€atsmedizin Berlin. In brief, aspirates (1 mL) with the iScript cDNA Synthesis Kit according to the manu- were suspended in 21 mL Dulbecco’s modified Eagle’s facturer’s instructions (BioRad, Munchen,€ Germany). The rel- (DME) medium (Biochrom, Berlin, Germany) containing 2 ative expression level of the housekeeping gene ng=mL basic fibroblast growth factor (Tebu-bio, Germany), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2013 VOL 101B, ISSUE 7 1311 TABLE I. Oligonucleotide Sequences

Base Gene Name Accession no. Oligonucleotide (50 ! 30) (Up=Down) Pairs Collagen type I NM_000088 CGA TGG CTG CAC GAG TCA CAC=CAG GTT GGG ATG GAG 180 GGA GTT TAC Collagen type II NM_001844 CCG GGC AGA GGG CAA TAG CAG GTT=CAA TGA TGG GGA 128 GGC GTG AG Collagen type IX NM_001853 AAT CAG GCT CTC GAA GCT CAT AAA A=CCT GCC ACA CCC CCG 100 CTC CTT CAT Cartilage oligomeric NM_000095 CCG GAG GGT GAC GCG CAG ATT GA=TGC CCT CGA AGT CCA 133 matrix protein CGC CAT TGA A Aggrecan NM_001135 GGC TGC TGT CCC CGT AGA AGA=GGG AGG CCA AGT AGG AAG GAT 163 Link-protein NM_001884 GCG TCC GCT ACC CCA TCT CTA=GCG CTC TAA GGG CAC ATT CAG TT 145 Osteocalcin NM_199173 GAG CCC CAG TTC CCC TAC CC=GCC TCC TGA AAG CCG ATG TG 103 Fatty acid-binding NM_001442 CCT TAG ATG GGG GTG TCC TGG TA=AAC GTC CCT TGG CTT 156 protein 4 ATG CTC TC Glyceraldehyde- NM_002046 GGC GAT GCT GGC GCT GAG TAC=TGG TCC ACA CCC ATG ACG A 149 3-phospate dehydrogenase used to normalize samples. Real-time reverse-transcriptase an intramuscular injection of 3 mg=kg xylazine and 10.0 polymerase chain reaction (PCR) using i-Cycler PCR System mg=kg ketamine. Anesthesia was maintained by inhalation

(BioRad) was performed with 1 mL of each cDNA sample in of 1.5–2.0% isoflurane delivered in oxygen and air (FiO2 triplicate using the SYBR green PCR Core Kit (Applied Bio- 0.4). All surgical procedures were performed under aseptic systems, Foster City, CA). Relative quantification of marker conditions. For bone marrow harvest, a 3 cm anterolateral genes (Table I) expression was performed and is given as incision was made at the hip to prepare the diaphysis of the percentage of the GAPDH product. femoral bone. Bone marrow of 5 mL was aspirated from the diaphysis using an 18-gauge needle and a syringe. Four Histology and immune-histochemistry weeks after bone marrow harvest and MSC preparation as VR MSC in PGA-HA scaffolds were embedded in optimal cutting described above, the PGA-HA implant (chondrotissue , Bio- temperature compound, frozen and cryosectioned (6 mm). Tissue AG, Zurich, Switzerland) was cut into 6 mm 3 7mm 7 Sections (n 5 9) were digested for 30 min. with 50 U=mL 3 1.1 mm and loaded with 2 3 10 =mL MSC in cell culture hyaluronidase (Sigma-Aldrich) at room temperature, followed medium. To prepare the cartilage defects, the joint was by staining with rabbit anti-human type II collagen antibodies opened by an anteromedial approach and the femoral con- (Acris, Hiddenhausen, Germany) for 40 min. Colorimetrical dyles were exposed. Degenerative joint disease and skeletal detection was done with 3-amino-9-ethylcarbazole (EnVi- abnormalities were excluded by visual inspection and full- TM sion ; Dako, Glostrup, Denmark) and counterstaining with thickness cartilage trochlear defects of 5 mm 3 6 mm were hematoxylin (Merk, Darmstadt, Germany). Proteoglycans created in the weight bearing cartilage using a scalpel and a and=or collagens were visualized by staining with Alcian blue sharp spoon. For covering of the defect, cell-laden PGA-HA 8GS (Roth, Karlsruhe, Germany; n 5 9) and safranin O (n 5 implants (n 5 4) were introduced using the press-fit tech- 9). Osteogenic differentiation was assessed by staining of nique. Cartilage defects without covering with cell-laden mineralized matrix components according to von Kossa (n 5 implants served as controls (n 5 3). Wound closure was 3). Adipogenic differentiation and intracellular lipid droplets performed in layers and protected by a plaster. Postopera- were visualized using Oil Red O (n 5 3). tively, prophylactic maintenance of analgesia was achieved by daily application of 20 mg=kg amoxyclin and calvulanate Statistical analysis for 5 days. At 30 days and 45 days, rabbits were anesthe- For statistical analysis of gene expression data (n 5 9per tized and sacrificed by administration of 100 mg=kg thio- point in time; n 5 3 independent donors with gene expres- pentone and 2 mmol=kg potassium chloride intravenously. sion analysis in triplicates), the Kolmogorov–Smirnov method Joints were fixed in formalin, decalcified and embedded in was applied for testing normal distribution of the data. The paraffin. Sections (6 mm) were stained with eosin and sec- parametric t-test and the nonparametric Mann–Whitney rank tions through the center of the defect (n 5 3 per defect) sum test were applied. Differences were considered signifi- were evaluated histologically. cant at p < 0.05 and a fold change of >3or<23. RESULTS Implantation of MSC-PGA-HA scaffolds in rabbit Cell surface antigen pattern of human bone cartilage defects and histology marrow-derived MSC Seven New Zealand White rabbits (3.5–4.0 kg) were used in Directly seeded, low expanded (passage 0) iliac crest bone this study. For anesthesia, rabbits were premedicated with marrow-derived MSC showed typical cell surface antigens

1312 PATRASCU ET AL. PROGENITOR-LOADED POLYGLYCOLIC ACID-HYALURONAN SCAFFOLDS FOR CARTILAGE REPAIR ORIGINAL RESEARCH REPORT

days showed formation of a cartilaginous extracellular matrix with deposition of proteoglycan [Fig. 2(B,D), arrowheads] and type II collagen [Fig. 2(F), arrowhead]. To confirm chondrogenic differentiation of MSC in PGA- HA scaffolds, semiquantitative gene expression analysis of genes coding for typical cartilage matrix molecules was performed (Fig. 3). Treatment of MSC cultured in PGA-HA scaffolds with TGFB3 significantly (p < 0.05) induced the expression of the chondrogenic marker genes type II collagen (from mean 62% of the GAPDH expression level found in untreated controls to mean 8201% of the GAPDH expression level found in TGFB3 treated MSC), type IX collagen (from 2% to 191%), cartilage oligomeric matrix protein (COMP, from 8.9% to 180%), aggrecan (from 0.3% to 7.8%), and cartilage link protein (from 14% to 531%), compared to untreated MSC in PGA-HA scaffolds. Compared to monolayer cultures of MSC, the three-dimensional assembly and tissue culture of MSC in PGA-HA scaffolds significantly (p < 0.05) induced the chondrogenic marker genes type II collagen (from 2.3% found in monolayer to 62% in PGA-HA scaffolds) and COMP (from 1.6% to 8.9%), even in the absence of the chondrogenic inducer TGFB3.

Adipogenic or osteogenic differentiation of human bone marrow-derived MSC in PGA-HA scaffolds Since MSC show a multipotential differentiation capacity toward cartilage as well as—unwanted in cartilage repair— bone and fat, the osteogenic and adipogenic differentiation potential of MSC in PGA-HA scaffolds was evaluated by his-

FIGURE 1. Cell surface antigen profile of adult human bone marrow- tological staining and gene expression analysis of typical derived MSC. Flow cytometric analysis shows that low expanded adipogenic and osteogenic markers (Fig. 4). At day 14, MSC MSC (passage 0) are positive for the antigens CD73, CD90, CD105, cultured in PGA-HA scaffolds with or without TGFB3 and CD166, while cells were negative for CD34 and CD45. [Color fig- showed no mineralization of the extracellular matrix and a ure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] low expression of the osteogenic marker gene osteocalcin [Fig. 4(A)]. MSC-PGA-HA cultures with or without TGFB3 treatment showed no lipid droplet formation. Compared to known from MSC and progenitor cell. MSC showed a MSC grown in monolayer cultures, MSC in PGA-HA scaffolds homogenous population and were negative for CD34 and showed significantly increased levels of the adipogenic CD45, but presented the antigens CD73, CD90, CD105, and marker gene fatty acid-binding protein 4 (FABP4,from CD166 (Fig. 1). 172% in monolayer to 5568% in MSC-PGA-HA without TGFB3). Treatment of MSC-PGA-HA cultures with TGFB3 sig- Chondrogenic differentiation of human bone nificantly decreased (from 5568% in untreated controls to marrow-derived MSC in PGA-HA scaffolds 115% in MSC-PGA-Ha treated with TGFB3) FABP4 expres- To test whether human progenitor cells are capable of sion levels [Fig. 4(B)]. undergoing chondrogenic differentiation in PGA-HA scaffolds, MSC were cultured for up to 14 days in PGA-HA scaffolds using standard chondrogenic conditions and Implantation of MSC-laden PGA-HA scaffolds in rabbit stimulated with TGFB3. Histological analysis of MSC-PGA-HA articular defects cultures (Fig. 2) revealed that control cultures not treated Rectangular full-thickness trochlear cartilage defects of 5 with TGFB3 showed virtually no deposition of proteoglycan mm 3 6 mm were left untreated [Fig. 5(A)] or filled press [Fig. 2(A,C)] or type II collagen [Fig. 2(E)]. MSC [Fig. 2(A), fit with MSC-laden PGA-HA scaffolds [Fig. 5(B)]. At 45 days, arrowhead] were embedded in a HA-fibrin matrix [Fig. 2(A), rabbits showed no lameness or abnormal behavior. There asterisk], distributed evenly within the PGA-HA scaffold and were no clinical signs of inflammation, infection or allergic scaffold fibers were partly fragmented, showing first signs reaction. The exposed joints showed no signs of synovial of degradation [Fig. 2(A), double arrowhead]. Scaffold fibers inflammation or irritation. The synovial fluid was clear and appeared blue when stained with alcian blue [Fig. 2(A)], appeared normal. In contrast to untreated controls, the and stained red upon staining with safranin O [Fig. 2(B)]. MSC-PGA-HA treated cartilage defects showed macroscopi- MSC-PGA-HA cultures that were treated with TGFB3 for 14 cally cartilaginous repair tissue formation with a smooth

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2013 VOL 101B, ISSUE 7 1313 FIGURE 2. Chondrogenic differentiation of MSC-laden PGA-HA scaffolds in vitro. Untreated controls showed no deposition of proteoglycan as assessed by Alcian blue (A; *HA–ﬁbrin matrix; arrowhead MSC; double arrowheads PGA ﬁber) and Safranin O staining (C). Immunostaining for type II collagen was negative (E). MSC-laden PGA-scaffolds treated with TGFB3 for 14 days showed deposition of proteoglycan according to Alcian blue (B, arrowheads) and Safranin O staining (C, arrowheads). The newly formed extracellular matrix was rich in type II collagen (F, arrowheads).

surface and almost complete filling of the defect (data not cells that were scattered through the repair tissue [Fig. 5(F), shown). arrows]. At day 45, the cartilage repair tissue covered the At days 30 and 45, untreated controls showed some car- defect and showed chondrocytic cells, with in part columnar tilaginous repair tissue formation adjacent to the subchon- distribution [Fig. 5(G,H), arrows]. The repair tissue was of a dral bone [Fig. 5(C,D), black arrows]. The chondral area was hyaline-like to hyaline appearance and the subchondral predominantly filled with fibrin that showed few cells [Fig. bone showed tidemark formation [Fig. 5(G), double arrow]. 5(C,D), asterisk]. At day 30, the subchondral bone showed signs of bone remodeling [Fig. 5(C), double arrowhead], DISCUSSION while the subchondral bone was clearly delineated at day In the present study, we demonstrated that human bone 45 [Fig. 5(D), double arrowhead]. marrow-derived MSCs have the capacity to undergo chon- Defects treated with MSC-laden PGA-HA scaffolds drogenic lineage development in vitro when embedded in showed formation of cell-rich cartilaginous repair tissue at resorbable PGA-HA scaffolds. Transplantation of autologous day 30 that partially filled the defect [Fig. 5(E,F)]. The tissue MSC-laden PGA-HA scaffolds into full-thickness articular car- showed good bonding to the remodeling subchondral bone tilage defects resulted in hyaline-like cartilage repair in the and appeared unstructured with round-shaped chondrocytic rabbit model.

1314 PATRASCU ET AL. PROGENITOR-LOADED POLYGLYCOLIC ACID-HYALURONAN SCAFFOLDS FOR CARTILAGE REPAIR ORIGINAL RESEARCH REPORT

FIGURE 3. Semiquantitative real-time gene expression analysis of MSC-laden PGA-HA scaffolds treated with TGFB3. The expression level of typical chondrogenic marker genes such as types I, II and IX collagens, aggrecan, cartilage oligomeric matrix protein (COMP), and cartilage link-protein was calculated as percentage of the expression level of the housekeeping gene GAPDH. The bars show the mean (n 5 3) and the SD. *Signiﬁcantly different (p < 0.05) compared to MSC-PGA-HA without TGFB3 treatment. #Signiﬁcantly different (p < 0.05) compared to monolayer MSC.

In cartilage repair, mesenchymal progenitor cells play a clinical study in young athletes with a 3-year follow-up that key role in bone marrow stimulating treatment strategies showed fibrocartilage and surface fibrillation in 8 out of 14 like drilling or microfracturing. The Mfx technique is a fre- biopsies.11 Therefore, a variety of one-step cartilage repair quently used first line cartilage repair option and induces techniques evolved that aim at the improvement of the Mfx- the formation of cartilage repair tissue by perforating the mediated cartilage repair tissue formation. These techniques subchondral bone. These Mfxs allow mesenchymal progeni- utilize different scaffolds and blood derivatives to cover the tor cells, which have a considerable chondrogenic potential, microfractured defect. The autologous matrix-induced chon- to populate the defect and form cartilaginous tissue.5,31 In drogenesis technique uses a porcine collagen type I=III Mfx-mediated cartilage repair, unstructured repair tissue membrane to cover the microfractured defect, followed by formation is often apparent that is predominantly of a fibro- the injection of a serum–fibrin mixture and=or PRP gels.32– cartilaginous type. This may be underlined by a recent 34 The use of the PGA-HA scaffold, also used in the current

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2013 VOL 101B, ISSUE 7 1315 FIGURE 4. Absence of osteogenic and adipogenic differentiation in chondrogenic MSC-laden PGA-HA scaffolds. MSC-laden PGA-HA scaffolds treated with TGFB3 showed no deposition of mineralized matrix as assessed by von Kossa staining, expression level of the osteogenic marker gene osteocalcin were low in MSC cultured in monolayer and in MSC-laden PGA-HA scaffolds with and without TGFB3 treatment (A). Oil red O staining showed no lipid droplets or vacuoles ﬁlled with lipids in MSC-laden PGA-HA scaffolds. The adipogenic marker gene fatty acid-binding protein 4 (FABP4) was elevated in MSC-laden PGA-HA scaffolds compared to MSC in monolayer (#p < 0.05), while TGFB3 repressed FABP4 (*p < 0.05) compared to the expression level found MSC without TGFB3 treatment (B). [Color ﬁgure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

study in vitro, in the treatment of chondral cartilage defects was more pronounced in the presence of the chondrogenic have been shown to be safe and effective, and is accompa- inducer TGFB3, the PGA-HA scaffold alone induced key car- nied by complete defect filling and hyaline-like cartilage tilage marker genes in MSC. This is in line with our previ- repair tissue formation.12,14,15 In the ovine model, covering ous studies that showed that the particular hyaluronic of microfractured defects with PGA-HA scaffolds immersed acid—that is a component of the PGA-HA scaffold—induces in autologous serum significantly improved the formation of or initiates the chondrogenic differentiation sequence in cartilage repair tissue with type II collagen compared to equine and human MSCs.29,35 In addition, the PGA-HA scaf- Mfx treatment alone.16 Recently, in a case series with 52 fold alone, without TGFB3, showed cartilage repair tissue patients suffering from tibial and femoral cartilage defects, formation in the rabbit full thickness cartilage defect model the PGA-HA scaffold was combined with PRP and showed when augmented with autologous bone marrow-derived hyaline-like cartilage formation as well as significant and cells. Therefore, it is suggested that the PGA-HA scaffold clinically meaningful patients’ improvement at 12 months alone has a chondrogenesis inducing effect on MSC, in vitro follow-up as assessed by the KOOS.17 The textile PGA-HA and in vivo. scaffold seems to have good chondrogenesis supporting or Bone marrow-derived stem and=or progenitor cells have even inducing properties by potentially helping to enrich a multilineage differentiation capacity and have been shown mesenchymal progenitor cells in the defect and guiding to undergo chondrogenic differentiation upon stimulation them toward cartilage formation.16,29 This is in concordance with, for example, all TGFB isoforms in pellet cultures.36,37 with our findings that show chondrogenic differentiation of There seems to be no difference in the chondrogenic poten- human MSCs in PGA-HA scaffolds in vitro. Although the tial regarding collagen deposition induced by different TGFB chondrogenic differentiation of MSC in PGA-HA scaffolds subtypes in human MSC.38 Recently, human bone marrow-

1316 PATRASCU ET AL. PROGENITOR-LOADED POLYGLYCOLIC ACID-HYALURONAN SCAFFOLDS FOR CARTILAGE REPAIR ORIGINAL RESEARCH REPORT

FIGURE 5. Articular cartilage repair with MSC-laden PGA-HA scaffolds in the rabbit cartilage defect model. Large full-thickness cartilage trochlear defects of 5 mm 3 6 mm were prepared (A) and covered with MSC-laden PGA-HA scaffolds using the press-fit technique (B). Untreated control defects showed some cartilaginous repair tissue (arrowhead) adjacent to the remodeling subchondral bone (double arrowhead). The chondral defect area was filled with fibrin (asterisks), at day 30 (C) and day 45 (D). At day 30, defects treated with MSC-laden PGA-HA scaffolds (E and F) showed unstructured cartilaginous repair tissue that partially filled the defect and that was rich in chondrocytic cells (F, arrowheads). At day 45 after MSC-laden PGA-HA implantation (G and H), the defects were covered with hyaline-like cartilage repair tissue showing chondrocytic cells partially in a columnar distribution (G and H, arrowheads). The subchondral bone showed the formation of a clearly delineated tidemark (G, double arrowheads). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2013 VOL 101B, ISSUE 7 1317 derived MSC have been shown to undergo chondrogenic dif- autologous bone marrow concentrate seems to be more ferentiation in chitosan-based (chitosan-poly-butylene ter- attractive as a source for bone marrow MSC in one-step car- ephtalate adipate) scaffolds upon TGFB3 treatment.39 Rat tilage repair procedures.22,46 Recently, it has been shown MSC have been shown to undergo chondrogenic differentia- that cells present in bone marrow concentrate are able to tion when assembled three-dimensionally in silk fibroin or undergo chondrogenic differentiation in vitro when seeded silk fibroin–chitosan scaffolds, as assessed by PCR analysis onto a HA-based scaffold.24 Nevertheless, for clinical appli- of cartilage marker genes.40 These studies are in line with cation, a potential drawback is that the number of stem our current study that showed cartilaginous tissue forma- cells in bone marrow concentrate is low (0.04%),22 suggest- tion in vitro with extracellular cartilage matrix formation ing that grafts with expanded bone marrow-derived MSC and induction of typical cartilage marker genes, after stimu- may be more effective in cartilage repair. Comparative clini- lation of less expanded human bone marrow-derived pro- cal trials have to show whether bone marrow concentrate genitors with TGFB3 in PGA-HA scaffolds. Interestingly, a and=or expanded MSC lead to a better outcome in cartilage recent report showed that human MSC in poly-lactic acid repair compared to Mfx alone or scaffold-assisted Mfx (PLLA) scaffolds formed extracellular matrix that contained approaches. Previously, we have shown in the ovine model aggrecan and collagens type I and X,41 which are known that the PGA-HA scaffold alone leads to cartilage repair and from fibrous and=or hypertrophic cartilage tissue. Obviously, improvement of the Mfx techniques, when used to cover the type of scaffold and chondrogenic stimulus is important microfractured defects.16,29 In the current study, we did not for the induction of a hyaline-like cartilage tissue, as shown include an empty PGA-HA scaffold group. This is a clear li- here for the PGA-HA scaffold. mitation of the study and we cannot conclude from the data In the rabbit model, autologous progenitors from bone that augmentation of the PGA-HA scaffold with MSC marrow embedded in a collagen gel were transplanted into improves cartilage repair compared to Mfx or to implanta- large (3 mm 3 6 mm) full-thickness cartilage defects for up tion a cell-free PGA-HA scaffold. A further limitation may be to 24 weeks. The cells developed into chondrocytes that we used human and not rabbit MSC to show that stem throughout the defect and formed hyaline-like cartilage cells undergo proper chondrogenic differentiation in the repair tissue.42 For osteochondral repair, rabbit bone mar- PGA-HA scaffold. Consequently, the data obtained from the row-derived MSC were seeded in sponge-like PLCL scaffolds in vitro and in vivo studies are not comparable and may and implanted into full-thickness (diameter 4.5 mm, depth therefore not prove the intrinsic chondrogenic potential of 5 mm) osteochondral femoral condylar defects. After 3–6 human MSC in PGA-HA scaffolds, in vivo. In particular for months, hyaline cartilage-like repair tissue was evident, clinical approaches, a limitation is that rabbit or animal joint while empty defects and defects treated with the PLCL scaf- defect models do not resemble the human situation. There- folds alone showed limited repair.43 Another approach fore, clinical pilot studies and=or clinical trials are needed, favored the use of bone marrow-derived MSC and polylac- before recommending such an approach for clinical use. tic-co-glycolic acid scaffolds (PLGA) that were pretreated In conclusion, we have shown that bone marrow-derived with TGB3 for osteochondral repair in the rabbit model. MSC show chondrogenic differentiation in PGA-HA scaffolds Twelve weeks after transplantation, MSC-PLGA grafts pre- in vitro and that repair of large full-thickness cartilage treated with TGFB3 showed repair tissue that was not dif- defects was feasible in the rabbit model by implanting MSC- ferent from normal, hyaline cartilage. In addition, the TGFB3 laden PGA-HA scaffolds. These findings suggest that PGA-HA treated grafts showed improved cartilage regeneration com- scaffolds are suited for regenerative medicine cartilage pared to MSC-PLGA grafts that were not pretreated with the repair approaches based on bone marrow-derived stem growth factor.44 As shown here, autologous MSC in PGA-HA and=or progenitor cells. scaffolds, not treated with TGFB3, developed hyaline-like cartilage repair tissue that covered larger defects of 5 mm ACKNOWLEDGMENT 3 6 mm after 45 days. However, clinical studies have to The authors are very grateful to Samuel Vetterlein for the show that these particular scaffolds loaded with bone mar- excellent technical assistance. row-derived cells or bone marrow do repair cartilage defects. Further limitations of this study are the lack of REFERENCES immune-histochemical proof of hyaline cartilage formation 1. Brittberg M, Lindahl A, Nilsson A, Ohlsson C, Isaksson O, Peter- with type II collagen deposition and the relatively short fol- son L. Treatment of deep cartilage defects in the knee with auto- low-up. logous chondrocyte transplantation. N Engl J Med 1994;331(14):889–895. Meanwhile, the use of autologous bone marrow-derived 2. Hubbard MJ. Articular debridement versus washout for degenera- progenitors for cartilage repair was challenged clinically and tion of the medial femoral condyle. A five-year study. J Bone has been shown to be safe in 41 patients with a mean fol- Joint Surg Br 1996;78(2):217–219. 45 low-up of 75 months, and some case reports suggest the 3. Kreuz PC, Muller S, Freymann U, Erggelet C, Niemeyer P, Kaps C, effectiveness of bone marrow-derived progenitor cells for Hirschmuller A. Repair of focal cartilage defects with scaffold- cartilage repair in traumatic, osteoarthritic, and osteogenesis assisted autologous chondrocyte grafts: Clinical and biomechani- 46–50 cal results 48 months after transplantation. Am J Sports Med imperfecta defects. It has to be taken into account that 2011;39(8):1697–1705. using expanded bone marrow-derived MSC is a two-step 4. Matsusue Y, Yamamuro T, Hama H. Arthroscopic multiple osteo- procedure that is time and cost intensive. Therefore, chondral transplantation to the chondral defect in the knee

1318 PATRASCU ET AL. PROGENITOR-LOADED POLYGLYCOLIC ACID-HYALURONAN SCAFFOLDS FOR CARTILAGE REPAIR ORIGINAL RESEARCH REPORT

associated with anterior cruciate ligament disruption. Arthroscopy marrow: In vitro characterisation of mesenchymal stem cells from 1993;9(3):318–321. iliac crest and subchondral bone. Injury 2010;41(11):1172–1177. 5. Steadman JR, Rodkey WG, Briggs KK, Rodrigo JJ. The microfrac- 23. Shi J, Zhang X, Zeng X, Zhu J, Pi Y, Zhou C, Ao Y. One-step artic- ture technic in the management of complete cartilage defects in ular cartilage repair: Combination of in situ bone marrow stem the knee joint. Orthopade 1999;28(1):26–32. cells with cell-free poly(L-lactic-co-glycolic acid) scaffold in a rab- 6. Steadman JR, Briggs KK, Rodrigo JJ, Kocher MS, Gill TJ, Rodkey bit model. Orthopedics 2012;35(5):e665–e671. WG. Outcomes of microfracture for traumatic chondral defects of 24. Cavallo C, Desando G, Columbaro M, Ferrari A, Zini N, Facchini the knee: Average 11-year follow-up. Arthroscopy 2003;19(5):477– A, Grigolo B. Chondrogenic differentiation of bone marrow con- 484. centrate grown onto a hylauronan scaffold: Rationale for its use 7. Steadman JR, Rodkey WG, Rodrigo JJ. Microfracture: Surgical in the treatment of cartilage lesions J Biomed Mater Res A 2012 technique and rehabilitation to treat chondral defects. Clin Orthop [Epub ahead of print]. Relat Res 2001(391 Suppl):S362–S369. 25. Xie X, Wang Y, Zhao C, Guo S, Liu S, Jia W, Tuan RS, Zhang C. 8. Endres M, Neumann K, Haupl T, Erggelet C, Ringe J, Sittinger M, Comparative evaluation of MSCs from bone marrow and adipose Kaps C. Synovial fluid recruits human mesenchymal progenitors tissue seeded in PRP-derived scaffold for cartilage regeneration. from subchondral spongious bone marrow. J Orthop Res Biomaterials 2012;33(29):7008–7018. 2007;25(10):1299–1307. 26. Yang Z, Wu Y, Li C, Zhang T, Zou Y, Hui JH, Ge Z, Lee EH. 9. Shapiro F, Koide S, Glimcher MJ. Cell origin and differentiation in Improved mesenchymal stem cells attachment and in vitro carti- the repair of full-thickness defects of articular cartilage. J Bone lage tissue formation on chitosan-modified poly(L-lactide-co-epsi- Joint Surg Am 1993;75(4):532–553. lon-caprolactone) scaffold. Tissue Eng Part A 2012;18(3-4):242– 10. Kreuz PC, Erggelet C, Steinwachs MR, Krause SJ, Lahm A, Nie- 251. meyer P, Ghanem N, Uhl M, Sudkamp N. Is microfracture of 27. Yoon IS, Chung CW, Sung JH, Cho HJ, Kim JS, Shim WS, Shim chondral defects in the knee associated with different results in CK, Chung SJ, Kim DD. Proliferation and chondrogenic differen- patients aged 40 years or younger? Arthroscopy tiation of human adipose-derived mesenchymal stem cells in po- 2006;22(11):1180–1186. rous hyaluronic acid scaffold. J Biosci Bioeng 2011;112(4):402– 11. Gudas R, Kalesinskas RJ, Kimtys V, Stankevicius E, Toliusis V, 408. Bernotavicius G, Smailys A. A prospective randomized clinical 28. You M, Peng G, Li J, Ma P, Wang Z, Shu W, Peng S, Chen GQ. study of mosaic osteochondral autologous transplantation versus Chondrogenic differentiation of human bone marrow mesenchy- microfracture for the treatment of osteochondral defects in the mal stem cells on polyhydroxyalkanoate (PHA) scaffolds coated knee joint in young athletes. Arthroscopy 2005;21(9):1066–1075. with PHA granule binding protein PhaP fused with RGD peptide. 12. Dhollander AA, Verdonk PC, Lambrecht S, Almqvist KF, Elewaut Biomaterials 2011;32(9):2305–2313. D, Verbruggen G, Verdonk R. The combination of microfracture 29. Erggelet C, Neumann K, Endres M, Haberstroh K, Sittinger M, and a cell-free polymer-based implant immersed with autologous Kaps C. Regeneration of ovine articular cartilage defects by cell- serum for cartilage defect coverage. Knee Surg Sports Traumatol free polymer-based implants. Biomaterials 2007;28(36):5570–5580. Arthrosc 2011. 30. Chomczynski P. A reagent for the single-step simultaneous isola- 13. Gille J, Schuseil E, Wimmer J, Gellissen J, Schulz AP, Behrens P. tion of RNA, DNA and proteins from cell and tissue samples. Bio- Mid-term results of autologous matrix-induced chondrogenesis techniques 1993;15(3):532–534, 536–537. for treatment of focal cartilage defects in the knee. Knee Surg 31. Neumann K, Dehne T, Endres M, Erggelet C, Kaps C, Ringe J, Sit- Sports Traumatol Arthrosc 2010;18(11):1456–1464. tinger M. Chondrogenic differentiation capacity of human mesen- 14. Patrascu JM, Freymann U, Kaps C, Poenaru DV. Repair of a post- chymal progenitor cells derived from subchondral cortico- traumatic cartilage defect with a cell-free polymer-based cartilage spongious bone. J Orthop Res 2008;26(11):1449–1456. implant: A follow-up at two years by MRI and histological review. 32. Behrens P. Matrix-coupled microfracture—A new concept for car- J Bone Joint Surg Br 2010;92(8):1160–1163. tilage defect repair. Arthroskopie 2005;18:193–197. 15. Zantop T, Petersen W. Arthroscopic implantation of a matrix to 33. Benthien JP, Behrens P. The treatment of chondral and osteo- cover large chondral defect during microfracture. Arthroscopy chondral defects of the knee with autologous matrix-induced 2009;25(11):1354–1360. chondrogenesis (AMIC): Method description and recent develop- 16. Erggelet C, Endres M, Neumann K, Morawietz L, Ringe J, Haber- ments. Knee Surg Sports Traumatol Arthrosc 2011;19(8):1316– stroh K, Sittinger M, Kaps C. Formation of cartilage repair tissue 1319. in articular cartilage defects pretreated with microfracture and 34. Dhollander AA, De Neve F, Almqvist KF, Verdonk R, Lambrecht S, covered with cell-free polymer-based implants. J Orthop Res Elewaut D, Verbruggen G, Verdonk PC. Autologous matrix- 2009;27(10):1353–1360. induced chondrogenesis combined with platelet-rich plasma gel: 17. Siclari A, Mascaro G, Gentili C, Cancedda R, Boux E. A cell-free Technical description and a five pilot patients report. Knee Surg scaffold-based cartilage repair provides improved function hya- Sports Traumatol Arthrosc 2011;19(4):536–542. line-like repair at one year. Clin Orthop Relat Res 2012;470(3):910– 35. Hegewald AA, Ringe J, Bartel J, Kruger I, Notter M, Barnewitz D, 919. Kaps C, Sittinger M. Hyaluronic acid and autologous synovial 18. Chen H, Hoemann CD, Sun J, Chevrier A, McKee MD, Shive MS, fluid induce chondrogenic differentiation of equine mesenchymal Hurtig M, Buschmann MD. Depth of subchondral perforation stem cells: A preliminary study. Tissue Cell 2004;36(6):431–438. influences the outcome of bone marrow stimulation cartilage 36. Barry F, Boynton RE, Liu B, Murphy JM. Chondrogenic differentia- repair. J Orthop Res 2011;29(8):1178–1184. tion of mesenchymal stem cells from bone marrow: Differentia- 19. Chen H, Sun J, Hoemann CD, Lascau-Coman V, Ouyang W, tion-dependent gene expression of matrix components. Exp Cell McKee MD, Shive MS, Buschmann MD. Drilling and microfracture Res 2001;268(2):189–200. lead to different bone structure and necrosis during bone-marrow 37. Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. In stimulation for cartilage repair. J Orthop Res 2009;27(11):1432– vitro chondrogenesis of bone marrow-derived mesenchymal pro- 1438. genitor cells. Exp Cell Res 1998;238(1):265–2672. 20. Caplan AI. The mesengenic process. Clin Plast Surg 1994;21(3): 38. Cals FL, Hellingman CA, Koevoet W, Baatenburg de Jong RJ, van 429–435. Osch GJ. Effects of transforming growth factor-beta subtypes on 21. Kaul G, Cucchiarini M, Remberger K, Kohn D, Madry H. Failed car- in vitro cartilage production and mineralization of human bone tilage repair for early osteoarthritis defects: A biochemical, histo- marrow stromal-derived mesenchymal stem cells. J Tissue Eng logical and immunohistochemical analysis of the repair tissue Regen Med 2012;6(1):68–76. after treatment with marrow-stimulation techniques. Knee Surg 39. Alves da Silva ML, Martins A, Costa-Pinto AR, Correlo VM, Sol P, Sports Traumatol Arthrosc 2012;20(11):2315–2324. Bhattacharya M, Faria S, Reis RL, Neves NM. Chondrogenic differ- 22. de Girolamo L, Bertolini G, Cervellin M, Sozzi G, Volpi P. Treat- entiation of human bone marrow mesenchymal stem cells in chi- ment of chondral defects of the knee with one step matrix- tosan-based scaffolds using a flow-perfusion bioreactor. J Tissue assisted technique enhanced by autologous concentrated bone Eng Regen Med 2011;5(9):722–732.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2013 VOL 101B, ISSUE 7 1319 40. Bhardwaj N, Kundu SC. Chondrogenic differentiation of rat MSCs Yoshiya S, Hattori K, Ohgushi H. Safety of autologous bone mar- on porous scaffolds of silk fibroin=chitosan blends. Biomaterials row-derived mesenchymal stem cell transplantation for cartilage 2012;33(10):2848–2857. repair in 41 patients with 45 joints followed for up to 11 years 41. Izal I, Aranda P, Sanz-Ramos P, Ripalda P, Mora G, Granero-Molto and 5 months. J Tissue Eng Regen Med 2011;5(2):146–150. F, Deplaine H, Gomez-Ribelles JL, Ferrer GG, Acosta V, Ochoa I, 46. Gigante A, Calcagno S, Cecconi S, Ramazzotti D, Manzotti S, Enea Garcıa-Aznar JM, Andreu EJ, Monleon-Pradas M, DoblareM, D. Use of collagen scaffold and autologous bone marrow concen- Prosper F. Culture of human bone marrow-derived mesenchymal trate as a one-step cartilage repair in the knee: Histological results stem cells on of poly(L-lactic acid) scaffolds: Potential application of second-look biopsies at 1 year follow-up. Int J Immunopathol for the tissue engineering of cartilage. Knee Surg Sports Trauma- Pharmacol 2011;24(1 Suppl 2):69–72. tol Arthrosc 2012 [Epub ahead of print]. 47. Giordano A, Galderisi U, Marino IR. From the laboratory bench to 42. Wakitani S, Goto T, Pineda SJ, Young RG, Mansour JM, Caplan the patient’s bedside: An update on clinical trials with mesenchy- AI, Goldberg VM. Mesenchymal cell-based repair of large, full- mal stem cells. J Cell Physiol 2007;211(1):27–35. thickness defects of articular cartilage. J Bone Joint Surg Am 48. Kuroda R, Ishida K, Matsumoto T, Akisue T, Fujioka H, Mizuno K, 1994;76(4):579–592. Ohgushi H, Wakitani S, Kurosaka M. Treatment of a full-thickness 43. Xie J, Han Z, Naito M, Maeyama A, Kim SH, Kim YH, Matsuda T. articular cartilage defect in the femoral condyle of an athlete with Articular cartilage tissue engineering based on a mechano-active autologous bone-marrow stromal cells. Osteoarthritis Cartilage scaffold made of poly(L-lactide-co-epsilon-caprolactone): In vivo 2007;15(2):226–231. performance in adult rabbits. J Biomed Mater Res B Appl Bio- mater 2010;94(1):80–88. 49. Wakitani S, Imoto K, Yamamoto T, Saito M, Murata N, Yoneda M. 44. Han SH, Kim YH, Park MS, Kim IA, Shin JW, Yang WI, Jee KS, Human autologous culture expanded bone marrow mesenchymal Park KD, Ryu GH, Lee JW. Histological and biomechanical proper- cell transplantation for repair of cartilage defects in osteoarthritic ties of regenerated articular cartilage using chondrogenic bone knees. Osteoarthritis Cartilage 2002;10(3):199–206. marrow stromal cells with a PLGA scaffold in vivo. J Biomed 50. Wakitani S, Mitsuoka T, Nakamura N, Toritsuka Y, Nakamura Y, Mater Res A 2008;87(4):850–861. Horibe S. Autologous bone marrow stromal cell transplantation 45. Wakitani S, Okabe T, Horibe S, Mitsuoka T, Saito M, Koyama T, for repair of full-thickness articular cartilage defects in human Nawata M, Tensho K, Kato H, Uematsu K, Kuroda R, Kurosaka M, patellae: Two case reports. Cell Transplant 2004;13(5):595–600.

1320 PATRASCU ET AL. PROGENITOR-LOADED POLYGLYCOLIC ACID-HYALURONAN SCAFFOLDS FOR CARTILAGE REPAIR