Osteoarthritis and Cartilage 23 (2015) 1307e1315

Engineering of hyaline cartilage with a calcified zone using bonemarrow stromal cells

W.D. Lee y, M.B. Hurtig z, R.M. Pilliar y x, W.L. Stanford y k ¶# **, R.A. Kandel y yy *

y Institute of Biomaterials and Biomedical Engineering, University of Toronto, 164 College St., Toronto, Ontario M5S 3G9, Canadaz Ontario Veterinary College, University of Guelph, 50 McGilvray Street, Guelph, Ontario N1G 2W1, Canadax Faculty of Dentistry, University of Toronto, 124 Edward St., Toronto, Ontario M5G 1G6, Canadak Sprott Centre for Stem Cell Research, Ottawa Hospital Research Institute, University of Ottawa, 501 Smyth Road, Ottawa, ON K1H 8L6, Canada¶ Department of Cellular & Molecular Medicine, University of Ottawa, 501 Smyth Road, Box 511, Ottawa, Ontario K1H 8L6, Canada# Department of Biochemistry, Microbiology and Immunology, University of Ottawa, 501 Smyth Road, Box 511, Ottawa, Ontario K1H 8L6, Canadayy Department of Pathology and Laboratory Medicine, Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, University of Toronto, 600 UniversityAve., Toronto, Ontario M5G 1X5, Canada

a r t i c l e i n f o

Article history:Received 7 January 2015Accepted 8 April 2015

Keywords:Tissue engineeringMesenchymal stromal cellsCartilageebone interfaceOsteochondral constructs

* Address correspondence and reprint requests toPathology and Laboratory Medicine, Mount Sinai HosToronto, Ontario M5G 1X5, Canada. Tel: 1-(416)-586-** Address correspondence and reprint requests to:for Stem Cell Research, Ottawa Health Research InstituOttawa, Ontario K1H 8L6, Canada. Tel: 1-(613)-737-88Fax: 1-(613)-739-6294.

E-mail addresses: wstanford@ohri.ca (W.L. Stan(R.A. Kandel).

http://dx.doi.org/10.1016/j.joca.2015.04.0101063-4584/© 2015 Osteoarthritis Research Society In

s u m m a r y

Objective: In healthy joints, a zone of calcified cartilage (ZCC) provides the mechanical integration be-tween articular cartilage and subchondral bone. Recapitulation of this architectural feature should serveto resist the constant shear force from the movement of the joint and prevent the delamination of tissue-engineered cartilage. Previous approaches to create the ZCC at the cartilageesubstrate interface haverelied on strategic use of exogenous scaffolds and adhesives, which are susceptible to failure by degra-dation and wear. In contrast, we report a successful scaffold-free engineering of ZCC to integrate tissue-engineered cartilage and a porous biodegradable bone substitute, using sheep bone marrow stromal cells(BMSCs) as the cell source for both cartilaginous zones.Design: BMSCs were predifferentiated to chondrocytes, harvested and then grown on a porous calciumpolyphosphate substrate in the presence of triiodothyronine (T3). T3 was withdrawn, and additionalpredifferentiated chondrocytes were placed on top of the construct and grown for 21 days.Results: This protocol yielded two distinct zones: hyaline cartilage that accumulated proteoglycans andcollagen type II, and calcified cartilage adjacent to the substrate that additionally accumulated mineral andcollagen typeX. Constructswith the calcified interfacehad comparable compressive strength tonative sheeposteochondral tissue and higher interfacial shear strength compared to control without a calcified zone.Conclusion: This protocol improves on the existing scaffold-free approaches to cartilage tissue engi-neering by incorporating a calcified zone. Since this protocol employs no xenogeneic material, it will beappropriate for use in preclinical large-animal studies.

© 2015 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Introduction

In a synovial joint, articular cartilage bears both compressive andshear forces to facilitate weight bearing and movement. Once

: R.A. Kandel, Department ofpital, 600 University Avenue,8516; Fax: 1-(416)-586-8719.W.L. Stanford, Sprott Centrete, 501 Smyth Road, Box 511,99x75495;

ford), rkandel@mtsinai.on.ca

ternational. Published by Elsevier L

damaged, these mechanical forces contribute to the progression ofcartilage damage, resulting in loss of cartilage and subchondral boneremodeling. Tissue engineering is widely regarded as a promisingtechnology to generate constructs that could replace damagedcartilage and bone1. A wide range of osteochondral tissue engineer-ing strategies are under investigation2, including our effort to pro-duce biphasic constructs that consist of scaffold-free cartilage tissueformed by articular chondrocytes or bone marrow stromal cells(BMSCs) on porous bone substitutes (calcium polyphosphate)3,4.Calcium polyphosphate is a biodegradable ceramic with excellentload-bearing and osseointegrative properties5,6. In our design,ingrowth of bone into calcium polyphosphate holds the biphasicconstruct inplace and provides support against compressive force on

td. All rights reserved.

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e13151308

tissue-engineered cartilage. However, to resist shear force, a me-chanically competent interface between the cartilage and its sub-strate is needed. Preventing cartilage delamination is a commonchallenge shared by many cartilage repair strategies7,8.

Healthy cartilage tissue interfaces with bone in vivo through azone of calcified cartilage (ZCC), which enhances mechanical inte-gration by redistributing forces acting at the interface9. BMSCs aresuitable for creating calcified cartilage, because they can be differ-entiated to produce cartilage tissues that mineralize their matrix10.While chondrocytes from the deep zone of articular cartilage canalso yield calcified cartilage in vitro11, BMSCs can be obtainedwithout donor site morbidity and expanded to clinically relevantnumbers. Furthermore, in vivo studies have shown that BMSC-derived cartilage can maintain their non-calcified phenotype afterbeing implanted in a joint cavity12,13, demonstrating the potentialof BMSCs to produce both calcified and non-calcified cartilage.However, to use BMSCs as a single source of cells to create cartilagetissues with a calcified cartilaginous interface, limiting minerali-zation of BMSC-derived cartilage to the interface will be crucial forthe success of these implants.

Using BMSCs as the cell source, we report the first successfulengineering of an osteochondral-like construct that incorporates aZCC at the cartilage-calcium polyphosphate interface without theuse of a scaffold as the cartilage tissue forms on top of a substrate. Astepwise culturing strategy enabled the selective activation ofBMSC-derived chondrocytes to induce mineralization only at theinterface, while avoiding mineralization in the contiguous cartilagetissue, thus conferring localized calcified and hyaline zones to theengineered cartilage tissue. This approach is suitable for clinical useas the BMSCs were expanded in media supplemented with autol-ogous serum and the construct cultured in serum-free, definedmedia, so this strategy can be directly translated to in vivo pre-clinical studies.

Materials and methods

Chondrogenic predifferentiation of BMSCs in membrane cultures

Sheep BMSCs were isolated and expanded from bone marrowaspirates taken from ewes aged between 2 and 5 years as previ-ously described4 using autologous serum (see SupplementaryMethods). To differentiate BMSCs to chondrocytes, 2.0 � 106

BMSCs were seeded onto 12 mm cell culture inserts (0.2 mm poresize; Millipore, Billerica, MA) that were coated with collagen typeIV (50 mg/insert; SigmaeAldrich, Oakville, ON, Canada). Cells werecultured in a defined chondrogenic media composed of high-glucose Dulbecco's modified Eagle medium (DMEM; Life Technol-ogies, Burlington, ON, Canada), 1� insulin-transferrin-seleniumplus (ITSþ) cell culture supplement (BD Biosciences, Bedford,MA), 100 nM dexamethasone (SigmaeAldrich), 100 mg/mL ascorbicacid (SigmaeAldrich) and 10 ng/mL transforming growth factor-b3(TGF-b3; R&D Systems, Minneapolis, MN). For the first 72 h,chondrogenic media was additionally supplemented with 10 mMblebbistatin (Cayman Chemical, Ann Arbor, MI) to inhibit sponta-neous detachment of cells. Media was changed every 2e3 days.After 3 weeks of culture, cells (herein referred to as prediffer-entiated chondrocytes) were isolated by digesting the tissue in 0.5%(w/v) collagenase (Roche Diagnostics, Indianapolis, IN) for 2 h at37�C.

Preparation of porous calcium polyphosphate substrates with ahydroxyapatite coating

Porous calcium polyphosphate disks of 4 mm diameter and2 mm height were prepared from gravity-sintered 75e150 mm

calcium polyphosphate particles as previously described14. Diskswere incubated in phosphate-buffered saline (PBS) for 1 week.Then, a thin, submicron hydroxyapatite layer was deposited overthe disks using a solegel processing method. This hydroxyapatitelayer was intended to serve as a ‘barrier’ coating inhibiting the rateof release of calcium polyphosphate degradation products duringin vitro cartilage formation15. The sol was prepared with triethylphosphite and calcium nitrate tetrahydrate (SigmaeAldrich) aspreviously described16. Disks were dipped into the sol gel side-ways for 8 s and withdrawn at a speed of 20 cm/min using astepper motor-driven actuator, air-dried for 10 min and annealedfor 15 min at 210�C. To ensure a continuous barrier coating, coateddisks were dipped again, annealed at 500�C for 20 min andgradually cooled. The coated disks were placed in Tygon tubing tocreate a well-like structure and subsequently sterilized by g-irra-diation (2.5 MRad).

Optimizing the mineralizing culture condition for predifferentiatedchondrocytes

To find a culture condition in which predifferentiated chon-drocytes would form calcified cartilage tissue, predifferentiatedchondrocytes were placed on the top surface of porous calcium pol-yphosphate substrates (1.5�106 cells/substrate)andcultured inbasalconstruct media composed of DMEM, 1� ITSþ, 100 mg/mL ascorbicacid, 50 mg/mL L-proline and 10 mM b-glycerophosphate (SigmaeAldrich). Todeterminewhich conditionwould inducemineralization,100 nM dexamethasone, 3 nM triiodothyronine (T3) or 100 nM reti-noic acid (all SigmaeAldrich) were added to cultures. The cultureswere harvested at various times up to 21 days for further study.

Tissue culture of multiphasic constructs

Predifferentiated chondrocytes were placed on the top surfaceof porous calcium polyphosphate substrates (1.0 � 106 cells/sub-strate) and cultured in basal construct media, with or without 3 nMT3. T3 was withdrawn at day 4. At day 5, additional prediffer-entiated chondrocyteswere placed on top of the existing constructs(1.5 � 106 cells/construct). The multi-layered constructs weremaintained under the same culture conditions for up to 21 days.This protocol is shown in Fig. 1.

Visualization and quantification of alkaline phosphate (ALP) activity

To visualize ALP activity, tissues were removed from their sub-strates and fixed in 10% neutral buffered formalin for 1 h. Then,samples were incubated in 30% (w/v) sucrose overnight at 4�C andsnap-frozen in Tissue-Tek OCT compound (Sakura Finetek, Tor-rance, CA). 6 mm cross-section cryosections were cut and mountedon silanized glass slides. Sections were incubated in azo dye(Naphthol AS-MX phosphate and Fast Blue BB salt, both Sigma-eAldrich) for 10 min, counterstained with eosin, dried andmounted with coverslips. To quantify the ALP activity, cells wereisolated from tissues by digestion in 0.5% collagenase for 2 h andlysed by sonication for 15 min in 0.2 M Tris pH 7.4 buffer. Activitylevels were quantified with p-nitrophenol phosphate (Sigma-eAldrich) and measuring the absorbance at 405 nm against astandard curve of p-nitrophenol11. Values were normalized to thelysate's DNA content (see Biochemical Analysis).

Gene expression analysis

Predifferentiated chondrocytes (1.5 � 106 cells/substrate) werecultured on porous calcium polyphosphate substrates in basalconstruct media with 4 days of 3 nM T3. T3 was removed after 4

Fig. 1. Line diagram of the tissue culture protocol used for forming scaffold-free multiphasic osteochondral-like constructs.

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e1315 1309

days, and the cells were grown in basal construct media only for theremainder of the 21 days of culture. Controls were cells that wereuntreated and grown in the same media. Tissues were removedfrom their substrates at day 4 and 21 and homogenized in 1 mLTRIzol reagent (Life Technologies) by bead milling. Total RNA wasthen extracted and reverse-transcribed using SuperScript VILOcDNA Synthesis Kit (Life Technologies). Quantitative PCR was per-formed using a LightCycler 96 Real-Time PCR System (Roche Di-agnostics) and SYBR GreenER master mix (Life Technologies) withgene-specific primers (Supplementary Table 1).

Micro-computed tomography (mCT) imaging

Constructs were imaged using a SkyScan 1174v2 mCT scanner(Bruker, Belgium). Scanning was performed at 50 kV and 800 mAthrough a 0.25 mm aluminum shield with a voxel size of 6.9 mm.After reconstruction, cross-sectional tomographs were obtainedwith the software provided by the manufacturer.

Histological analysis of whole constructs

Constructs were fixed in 10% neutral buffered formalin over-night and infiltrated using the Osteo-Bed bone embedding kit(Polysciences Inc., Warrington, PA). Sections were cut and groundto ~50 mm thickness, stained with toluidine blue and light greenand visualized under light microscopy.

Histological analysis of cartilaginous tissues

Tissue mineral and proteoglycan accumulation was visualizedby staining histological sections (see Supplementary Methods)with von Kossa and toluidine blue. Orientation of collagen fibrilswas visualized by polarized light microscopy of trichrome-stained sections. Visualization for collagens types I, II and Xwere visualized by immunofluorescence (see SupplementaryMethods).

Biochemical analysis of extracellular matrix and mineralaccumulation

Tissues were digested in 40 mg/mL papain (SigmaeAldrich) andDNA, glycosaminoglycan and hydroxyproline contents were quan-tified as previously described17. To measure mineral accumulation,the tissues were lyophilized, dry weights measured, and thendigested in 3 N hydrochloric acid at 90�C for 2 h. The pH wasadjusted to 4.0 with 1.5 M acetate buffer. Phosphate content wasdetermined using the heteropoly blue assay and measuring theabsorbance at 620 nm. Calcium content was determined using theo-cresolphthalein complexone assay and measuring the absor-bance at 570 nm.

Mechanical testing of multiphasic constructs

On day 21 of culture, the equilibrium compressive modulus ofthe multiphasic construct's cartilaginous tissue was determined by

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e13151310

stress relaxation testing on the Mach-1 mechanical tester (Bio-Momentum, Laval, Canada) with a 0.65 mm-diameter indenter aspreviously described4. Interfacial shear strength was determinedby applying a force at the interface region of these samples using aspecially designed jig attached to an Instron universal testing ma-chine as previously described15 (see Supplementary Methods).

Statistical testing

All values are expressed as mean ± 95% confidence intervals (CI).Each experiment was carried out with three donor animals (N ¼ 3).Univariate analysis of variance and Tukey post-hoc testing wereused to analyze ALP activity levels on days 0e4, accumulation ofGAG, total collagen and mineral contents, as well as the compres-sive moduli of the constructs. The interfacial shear strength data ofthe constructs were analyzed using t test. To account for the wideanimal-to-animal variability, ALP activity levels on days 7, 10 and 14were each analyzed using paired t test. Similarly, gene expressionlevels on days 4 and 21 were log-transformed and analyzed usingpaired t test.

Results

Generation of calcified cartilage at the calcium polyphosphateinterface with predifferentiated chondrocytes

To engineer a multiphasic construct that incorporates a ZCC atthe cartilage-calcium polyphosphate interface using prediffer-entiated chondrocytes, we first sought to establish a culture

Fig. 2. Predifferentiated chondrocytes formed cartilaginous tissues on porous calcium podifferentiated chondrocytes cultured for 7 days with 3 nM triiodothyronine (T3) or 100 nMdetected. *P ¼ 0.043 between þT3 eDex and þT3 þDex; P ¼ 0.045 between þT3 eDex anformed by culturing 2 � 106 predifferentiated chondrocytes on the substrates with 3 nM T3Asterisk denotes the original location of the substrate. Scale bar ¼ 200 mm. (C) Cross-sectinumbers of seeded cells: 2 � 106 (left), 1.5 � 106 (middle) and 1.0 � 106 (right) prediffermineralized zone (green) and the substrate (gray). Arrows indicate the lack of separation be

condition in which predifferentiated chondrocytes could becultured on porous calcium polyphosphate substrates to formmineralized cartilage tissue. Triiodothyronine (T3)18, retinoic acid19

and dexamethasone10 have all been identified as potential inducersof chondrocyte mineralization in vitro. We cultured prediffer-entiated chondrocytes on calcium polyphosphate substrates witheither 3 nM T3 or 100 mM retinoic acid treatment in the presence orabsence of dexamethasone for 1 week, and then quantified the ALPactivity of the cells as an indicator of chondrocyte mineralizationpotential20,21. ALP activity was the highest in predifferentiatedchondrocytes cultured in the absence of dexamethasone with theT3 treatment [Fig. 2(A)]. While ALP activity was also increased withthe retinoic acid treatment, continued culture did not yield carti-laginous tissues on calcium polyphosphate after 3 weeks (data notshown). Therefore, treatment of predifferentiated chondrocyteswith T3 was selected for further study.

When 2 � 106 predifferentiated chondrocytes were cultured onporous calcium polyphosphate substrates for 3 weeks, themineralized zone was found in the top of the tissue [Fig. 2(B)].Between this mineralized zone and the substrate was an inter-mediary zone of non-mineralized cartilage, composed of cells withround morphology and extracellular matrix rich in proteoglycans.Changing the concentration of T3 treatment did not change thelocation of the mineralized zone (data not shown). However,when the number of initially seeded predifferentiated chon-drocytes was decreased, the distance between the cartila-geesubstrate interface and the zone of mineralized cartilage alsodecreased [Fig. 2(C and D)]. At an initial seeding of 1 � 106 pre-differentiated chondrocytes, the zone of mineralized cartilage

lyphosphate substrates with a zone of mineralized cartilage. (A) ALP activity of pre-retinoic acid (RA) with or without dexamethasone (Dex). Mean ± 95% CI, n ¼ 3, nd: notd þRA eDex. (B) Von Kossa and toluidine blue-stained sections of cartilaginous tissuefor 3 weeks, showing mineralized zone occurring at the superficial aspect of cartilage.onal mCT tomographs and (D) and histology of 3-week-old constructs with decreasingentiated chondrocytes. In (D) vertical bars denote non-mineralized cartilage (purple),tween the mineralized zone and the substrate. Scale bar ¼ 1 mm (C) and 200 mm (D).

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e1315 1311

formed consistently adjacent to the interface [Fig. 2(C and D),arrows].

Short-term treatment of predifferentiated chondrocytes with T3 wassufficient to stimulate terminal differentiation

ALP activity levels of predifferentiated chondrocytes werecharacterized over a 2-week culture period. T3-treated cells had asignificantly increased ALP activity by day 4 [Fig. 3(A)]. Even thoughthe T3 treatment was withdrawn at day 4, the ALP activity level ofcells at day 7 was significantly higher than those that had beencontinuously treated with T3, and were comparable at days 10 and14 [Fig. 3(B)].

To confirm that the 4-day T3 treatment was sufficient toinduce terminal differentiation of predifferentiated chon-drocytes, gene expression levels were examined. As early as 4days, collagen type X (Col10a1) expression was significantlyhigher in the T3-treated cells compared to the untreated control[Fig. 3(G)]. By 21 days, T3-treated cells expressed less collagentype II (Col2a1), aggrecan (Agc1) and Sox9, but greater Col10a1and osteopontin (Spp1) compared to the non-treated controls

Fig. 3. Short-term T3 treatment of predifferentiated chondrocytes was sufficient to stimulatethe first 4 days. In selected cultures, T3 treatment was withdrawn after 4 days (WD). (A) ALPWD, compared to cells in which T3 treatment was continued (þ) for up to 14 days. *P ¼ 0.01controls (�). *P ¼ 0.013 (Col2a1 on day 21); P ¼ 0.031 (Agc1 on day 21); P ¼ 0.009 (Sox9 on(Spp1 on day 21). Data points are color-coded for each animal throughout. Results are expr

[Fig. 3(CeJ)]. This confirmed that the 4-day T3 treatment ofpredifferentiated chondrocytes had altered the differentiation ofthese cells.

T3-treated predifferentiated chondrocytes did not induce ALPactivity in non-T3-treated predifferentiated chondrocytes

Based on this observation, a two-stage culture protocol wasdeveloped (Fig. 1). To grow an interfacial zone of mineralizedcartilage, 1.0 � 106 predifferentiated chondrocytes were firstseeded on porous calcium polyphosphate substrates, which wasthe number of cells that had been previously determined to beoptimal for creating the mineralized cartilage at the interface, andcultured in the presence of T3 [Fig. 2(C and D)]. T3 was withdrawnat 4 days of culture, then 1.5 � 106 predifferentiated chondrocyteswere seeded on top of the existing tissue 1 day later and cultured inthe presence of b-glycerophosphate for up to 21 days to generate azone of non-mineralized cartilage. To confirm that the T3-treatedcells would not stimulate the overlaid, non-T3-treated cells toalso mineralize, the tissues on calcium polyphosphate were har-vested 5 days after the addition of the top layer of cells (10 days of

terminal differentiation. Predifferentiated chondrocytes were treated with 3 nM T3 foractivity during the first 4 days. *P < 0.001 between days 0e4 & 2-4. (B) ALP activity of4. (CeJ) Gene expression analysis of WD on days 4 and 21 compared to non-T3-treatedday 21); P ¼ 0.003 (Col10a1 on days 4 & 21); P ¼ 0.025 (Spp1 on day 4) and P ¼ 0.002essed as mean ± 95% CI, n ¼ 3.

Fig. 4. Layer of predifferentiated chondrocytes treated with T3 did not induce theoverlaid layer of non-treated predifferentiated chondrocytes to activate ALP activity.1 � 106 cells were cultured on porous calcium polyphosphate substrates with (A) orwithout (B) 3 nM T3 for 4 days. On day 5, 1.5 � 106 cells were seeded on top of thesecells/tissue and cultured without T3. Cryosections of tissues from day 10 multiphasicconstructs were stained for ALP activity. The blue staining indicative of ALP was seenonly in the mineralizing layer (vertical bar). Asterisk denotes the location of where thesubstrate would have been located prior to its removal, scale bar ¼ 200 mm.

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e13151312

culture in total) and cryosectioned. Histological analysis demon-strated that ALP activity was present only in the calcifying bottomlayer (Fig. 4), demonstrating the lack of mineralization potential inthe top layer even though b-glycerophosphate was present.

Fig. 5. A two-step culture protocol of predifferentiated chondrocytes on porous calcium poly(A) Day 21 constructs with 4-day T3 treatment were processed undecalcified, sectioned an(brown). (BeI) Tissues were excised from the substrate and examined histologically for thestaining for the accumulation of different types of collagen (CeE, GeI). Constructs with T3-trdenoted as eT3. Scale bars ¼ 200 mm. (J) Differential collagen fiber orientations on tissues

Fig. 6. Accumulation of extracellular matrix in T3-treated (þT3) tissues was less comparedAccumulated proteoglycan and (B) total collagenwere quantified and normalized by their DNand Native. (C) Accumulation of mineral in tissue was quantified by measuring the calciumdata from sheep cartilage explants of femoral condyles. Results are expressed as mean ± 9

Characterization of the multiphasic constructs

Constructs were grown for 21 days using the two-stage cultureprotocol with or without the 4-day T3 treatment of the interfaciallayer and characterized using various methods. Histological ex-amination of cartilage tissue on the calcium polyphosphate showedtwo tissue zones. The upper zone of non-mineralized cartilage washyaline cartilage, rich in type II collagen and proteoglycan [Fig. 5(A,B, D)]. The interfacial zone was calcified cartilage, the mineralstaining with von Kossa staining and the extracellular matrixstaining with toluidine blue and type X collagen [Fig. 5(A, B, E)].Ingrowth of tissue into upper parts of the porous calcium poly-phosphate substrate was also observed, with the mineral clustersdirectly in contact with the calcium polyphosphate particles[Fig. 5(A)]. No type I collagenwas detected in either zone [Fig. 5(C)].In contrast, control constructs not treated with T3 did not show abizonal composition as there was no ZCC, and there was an unevendistribution of extracellular matrix [Fig. 5(FeI)]. Differentialcollagen alignment in the tissue was detected using polarized lightmicroscopy. In the superficial aspect, collagen was aligned parallelto the surface. Elsewhere, collagenwas observed both pericellularlyand aligned orthogonal to the surface. Collagen, parallel to thesurface, was also observed in areaswithin the bottom layer of tissue[Fig. 5(J)]. T3-treated constructs accumulated less proteoglycans[Fig. 6(A)] and total collagens [Fig. 6(B)] compared to untreatedconstructs, but the proteoglycan content was comparable to nativesheep cartilage. Tissues on T3-treated constructs accumulated

phosphate substrates produced cartilage tissue with a mineralized zone at the interface.d stained to reveal the cartilaginous (purple), mineral (green) and substrate particlesdistribution of mineral and proteoglycan accumulation (B, F), by immunofluorescenteated interfacial layer are denoted as þT3, and untreated (no T3) control constructs arewas seen by polarized light microscopy. Scale bar ¼ 100 mm.

to those in untreated (eT3) tissues, but comparable to native articular cartilage. (A)A content. (A) *P ¼ 0.001, (B) *P ¼ 0.012 between þT3 and eT3; P ¼ 0.037 between þT3and phosphate contents, normalized by tissue dry weight. **P < 0.001. Native denotes5% CI, n ¼ 6.

Fig. 7. T3-treated constructs (þT3) exhibited comparable compressive strength as native sheep osteochondral explants and stronger shear strength than untreated constructs (eT3).(A) Equilibrium compressive modulus was measured by stress relaxation test with a 0.65 mm-diameter indenter situated at the center of the tissue. *P ¼ 0.008, **P < 0.001. (B, C) Ashear load was applied to the cartilageesubstrate interface until failure. All substrates were identical in shape. The interfacial shear strength testing could not be performed onosteochondral explants due to slippage of the cartilage tissue. (B) *P ¼ 0.003; (C) *P ¼ 0.016. Results are expressed as mean ± 95% CI, n ¼ 9 for constructs and 6 for explants.

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e1315 1313

calcium and phosphate, consistent with calcification, while un-treated constructs did not [Fig. 6(C)].

T3-treated tissues had a comparable equilibrium compressivemodulus to that of osteochondral tissue obtained fromnative sheepfemoral condyles [Fig. 7(A), P ¼ 0.15]. The untreated constructs hada significantly higher equilibrium compressive modulus comparedto that of the T3-treated tissues, possibly due to their greaterextracellular matrix accumulation [Fig. 6(A, B)]. However, T3-treated constructs could withstand a stronger shear force[Fig. 7(B)] and absorb more energy [Fig. 7(C)] than untreated con-structs. Histological assessment of the failed constructs confirmedthat failure occurred at the interface (data not shown). Thisdemonstrated that the mineralized tissue at the interface [Fig. 5(A)]enhanced the interfacial shear strength.

Discussion

A mechanically competent interface between the cartilage andits substrate is required to resist the shear force generated by themovement of the joint. Previous studies have sought to recreate theZCC that exists between hyaline cartilage and bone to reinforce thisinterface with limited success22. To overcome this problem, wesuccessfully tissue-engineered osteochondral-like constructs withBMSCs as the single cell source, in which cartilage tissue and aporous bone substitute substrate were formed with an interfacialZCC without using any exogenous scaffolds in the cartilage tissue.The two-stage culture protocol enabled the selective stimulation ofpredifferentiated chondrocytes to form calcified cartilage only atthe interface. Treatment of the initially seeded cells with T3induced the terminal differentiation of cells. This resulted in a ZCC,which, in addition to mineral, also contained collagens type II and Xand proteoglycans. These features were seen only in the part of thetissue that was treatedwithmineralization-inducing T3, which wasadjacent to (and integrated with) the porous calcium poly-phosphate substrate. The interfacial shear strength of constructswith the calcified cartilaginous interfacewas increased significantlycompared to those without it, functionally validating its intendedeffect. The hyaline quality of non-mineralized cartilage wasdemonstrated by the accumulation of collagen type II and the lackof mineral or collagen type I. Further study is required to determinewhether this osteochondral-like tissue formation will occur onsubstrates other than hydroxyapatite-coated calcium poly-phosphate substrate.

Thus far, approaches to induce selective mineralization ofBMSC-derived cartilage and create this interface include delivery ofbioactive signals either locally via microspheres23 or by using gra-dients generated by a bioreactor24 to undifferentiated BMSCsseeded in scaffolds. In another approach, BMSCs predifferentiatedto chondrocytes and osteoblasts are seeded into appropriate

locations within scaffolds25,26. For both strategies, use of exogenousscaffolds is necessary, which has the potential to hinder hostintegration and tissue regeneration27. Matching the rate of scaffolddegeneration to the rate of tissue accumulation remains a challengewhen employing scaffolds22,28. Synthetic materials also accumulatewear over time, andmay provide a focal point for shear failure if notreplaced with new tissue. On the other hand, if the rate of scaffolddegradation exceeds the rate of cellular remodeling, the mechani-cal integrity is compromised and the construct would fail underload. A scaffold-free approach obviates this issue altogether; how-ever, this necessitated an alternative strategy to specify the locationof the calcified interface.

The development of our two-stage culture protocol was firstenabled by the observation that BMSC-derived predifferentiatedchondrocytes formed mineralized cartilage at the tissue's superfi-cial aspect with mineral-inducing stimulus. Given that T3 wasdelivered to the tissue through culture media, and that the locationof calcified cartilage did not changewith tissue thickness, this effectwas likely due to T3 diffusion into the tissue. As there was noscaffold present, selective delivery of T3 to the interface was diffi-cult to achieve with media alone. To circumvent this, we seeded asmall number of cells on the calcium polyphosphate in the pres-ence of T3. This generated thin cartilaginous tissue, forcingmineralization to take place juxtaposed to the porous calciumpolyphosphate substrate. This enabled the subsequent formation ofa continuous layer of non-mineralized cartilage formed by the samecell type in the absence of T3. The presence of collagen types II, Xand proteoglycans in the calcified interface, as well as the absenceof collagen type I, is an important finding in our study. Takentogether with the finding that T3-treated cells expressed hyper-trophic chondrocyte makers, this demonstrates that the interface iscartilaginous, not osseous, successfully recapitulating characteris-tics of the native articular cartilage-subchondral bone interface.

Histology showed that the layer of non-mineralized cartilagesuccessfully fused to the underlying T3-treated tissue. Interestingly,fissures were observed if the intended interfacial tissue (i.e., thefirst layer) was not treated with T3, and the total thickness ofcartilage tissues on these constructs was increased, with an unevendistribution of the extracellular matrix. There are two possibleexplanations for these results. One is that the manner that theinterfacial layer grows may affect integration, as BMSC-derivedchondrocytes share many characteristics with proliferating carti-lage29 that grows appositionally. In development, Indian hedgehog,FGF and BMP signaling pathways regulate this appositionalgrowth30; however, in an in vitro system, absence of such complexregulation may allow the interfacial layer to expand laterally. Thepresence of tubing around the constructs prevented lateralexpansion by confinement, which could have resulted in lateralmechanical stress and caused the tissue to buckle and shear. This

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e13151314

was not the case for T3-stimulated interfacial layer, whose behaviormay more resemble terminal hypertrophic chondrocytes, as evi-denced by their mineralized matrix. Terminal hypertrophic chon-drocytes do not proliferate and eventually undergo apoptosis uponvascularization of matrix and bone formation31. Alternatively, theT3 treatment may have affected tissue fusion. Pellets of humanMSCs undergoing chondrogenesis have been shown to accumulatetenascin along the outer surface that prevented them from fusingwith one another32. The fissure between the two layers of non-mineralizing cartilage may stem from a similar boundary-settingphenomenon, whether driven by tenascin accumulation orotherwise.

Sheep gait studies estimate the maximum anterioposteriorshear stress in a stifle joint to be about 500 kPa33,34, which isequivalent to approximately 9 N of peak shear force in our study.However, using our methodology the peak shear force of nativecartilage interface could not be measured, as it withstood themaximum load (80 N) without shear failure. In healthy cartilage,collagen fibrils run orthogonally through the tidemark plane, whichwould confer a high resistance against failure in our mode of study.Although our multiphasic construct possesses inferior interfacialshear strength compared to articular cartilage, it still represented afour-fold improvement over the non-mineralized control, andsheep gait studies suggest that this improvement is adequate forpreventing delamination.

Importantly, this protocol can be expanded to generate large,anatomically shaped osteochondral-like constructs as re-placements for damaged joint tissues. Protocols have been devel-oped to generate ~108 BMSCswith chondrogenic potential in a goodmanufacturing practices-complaint manner35,36, which can be usedto generate sufficiently large numbers of chondrocytes to cover alarge articular surface. We have previously shown that calciumpolyphosphate powders can be used in additive manufacturingmethods to generate substrates of required shape and size37. Cal-cium polyphosphate formed in this way exhibited the same degreeof osseointegration after implantation as those produced conven-tionally38. As such, preclinical studies are warranted to determine ifthese osteochondral-like constructs with a mineralized cartilagi-nous interface could repair joint defects in vivo.

Conclusion

Using BMSCs as a single cell source, an osteochondral-likeconstruct comprised of hyaline cartilage and porous bone substi-tute was created with a ZCC interfacing them. The novel, two-stepcell culture strategy employed no exogenous scaffolds, just a sub-strate on which to form tissue. The presence of calcified cartilageincreased the shear load that the construct could withstand at theinterface. As no xenogeneic material was used at any point duringculture, this strategy can be directly translated into preclinicalstudies to determine whether these constructs can repair jointdefects in vivo.

Author contributions

Study conception and design: WDL, WLS, RAK.Provision of study materials: MBH, RMP.Analysis and interpretation of data: WDL, RMP, RAK.Drafting of article: WDL, WLS, RAK.Critical revision of the article: WDL, RMP, WLS, RAK.Final approval of the article: WDL, WLS, RAK.

Role of the funding sourcesNone of the listed bodies played any role in this study.

Conflict of interestNo competing financial interests exist for any of the listed authors.

Acknowledgments

We thank Eugene Hu for preparing the porous calcium poly-phosphate substrates. We also thank Dr Jian Wang and NancyValiquette for their technical assistance. This work was supportedby the United States Department of Defense (to R.A.K., W81XWH-10-1-0787) and the Natural Sciences and Engineering ResearchCouncil of Canada (to W.L.S., RGPIN 293170-11). W.L.S. is supportedby a Canadian Research Chair.

Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.joca.2015.04.010.

References

1. Swieszkowski W, Tuan BHS, Kurzydlowski KJ, Hutmacher DW.Repair and regeneration of osteochondral defects in thearticular joints. Biomol Eng 2007;24:489e95.

2. Shimomura K, Moriguchi Y, Murawski CD, Yoshikawa H,Nakamura N. Osteochondral tissue engineering with biphasicscaffold: current strategies and techniques. Tissue Eng Part BRev 2014;20:468e76.

3. Kandel RA, Grynpas M, Pilliar R, Lee J, Wang J, Waldman S,et al. Repair of osteochondral defects with biphasic cartilage-calcium polyphosphate constructs in a sheep model. Bio-materials 2006;27:4120e31.

4. Lee WD, Hurtig MB, Kandel RA, Stanford WL. Membrane cul-ture of bone marrow stromal cells yields better tissue thanpellet culture for engineering cartilage-bone substitutebiphasic constructs in a two-step process. Tissue Eng Part CMethods 2011;17:939e48.

5. Grynpas MD, Pilliar RM, Kandel RA, Renlund R, Filiaggi M,Dumitriu M. Porous calcium polyphosphate scaffolds for bonesubstitute applications in vivo studies. Biomaterials 2002;23:2063e70.

6. Pilliar RM, Kandel RA, Grynpas MD, Hu Y. Porous calciumpolyphosphate as load-bearing bone substitutes: in vivo study.J Biomed Mater Res B Appl Biomater 2013;101:1e8.

7. Vahdati A, Wagner DR. Implant size and mechanical propertiesinfluence the failure of the adhesive bond between cartilageimplants and native tissue in a finite element analysis.J Biomech 2013;46:1554e60.

8. Harris JD, Siston RA, Brophy RH, Lattermann C, Carey JL,Flanigan DC. Failures, re-operations, and complications afterautologous chondrocyte implantationea systematic review.Osteoarthritis Cartilage 2011;19:779e91.

9. Hoemann CD, Lafantaisie-Favreau C-H, Lascau-Coman V,Chen G, Guzm�an-Morales J. The cartilage-bone interface.J Knee Surg 2012;25:85e97.

10. Mueller MB, Tuan RS. Functional characterization of hyper-trophy in chondrogenesis of human mesenchymal stem cells.Arthritis Rheum 2008;58:1377e88.

11. Allan KS, Pilliar RM, Wang J, Grynpas MD, Kandel RA. Forma-tion of biphasic constructs containing cartilage with a calcifiedzone interface. Tissue Eng 2007;13:167e77.

12. Zscharnack M, Hepp P, Richter R, Aigner T, Schulz R,Somerson J, et al. Repair of chronic osteochondral defects usingpredifferentiated mesenchymal stem cells in an ovine model.Am J Sports Med 2010;38:1857e69.

W.D. Lee et al. / Osteoarthritis and Cartilage 23 (2015) 1307e1315 1315

13. Shimomura K, Moriguchi Y, Ando W, Nansai R, Fujie H,Hart DA, et al. Osteochondral repair using a scaffold-free tis-sue-engineered construct derived from synovial mesenchymalstem cells and a hydroxyapatite-based artificial bone. TissueEng Part A 2014;20(17e18):2291e304.

14. Waldman SD, Grynpas MD, Pilliar RM, Kandel RA. Character-ization of cartilagenous tissue formed on calcium poly-phosphate substrates in vitro. J Biomed Mater Res 2002;62:323e30.

15. St-Pierre J-P, Gan L, Wang J, Pilliar RM, Grynpas MD,Kandel RA. The incorporation of a zone of calcified cartilageimproves the interfacial shear strength between in vitro-formed cartilage and the underlying substrate. Acta Biomater2012;8:1603e15.

16. Gan L, Pilliar R. Calcium phosphate sol-gel-derived thin filmson porous-surfaced implants for enhanced osteoconductivity.Part I: synthesis and characterization. Biomaterials 2004;25:5303e12.

17. Taylor DW, Ahmed N, Gan L, Gross AE, Kandel RA. Proteogly-can and collagen accumulation by passaged chondrocytes canbe enhanced through side-by-side culture with primarychondrocytes. Tissue Eng Part A 2010;16:643e51.

18. Alini M, Kofsky Y, Wu W, Pidoux I, Poole AR. In serum-freeculture thyroid hormones can induce full expression ofchondrocyte hypertrophy leading to matrix calcification.J Bone Miner Res 1996;11:105e13.

19. Iwamoto M, Shapiro IM, Yagami K, Boskey AL, Leboy PS,Adams SL, et al. Retinoic acid induces rapid mineralization andexpression of mineralization-related genes in chondrocytes.Exp Cell Res 1993;207:413e20.

20. Miao D, Scutt A. Histochemical localization of alkaline phos-phatase activity in decalcified bone and cartilage. J HistochemCytochem 2002;50:333e40.

21. Jiang J, Leong NL, Mung JC, Hidaka C, Lu HH. Interaction be-tween zonal populations of articular chondrocytes suppresseschondrocyte mineralization and this process is mediated byPTHrP. Osteoarthritis Cartilage 2008;16:70e82.

22. Huey DJ, Hu JC, Athanasiou KA. Unlike bone, cartilage regen-eration remains elusive. Science 2012;338:917e21.

23. Dormer NH, Singh M, Wang L, Berkland CJ, Detamore MS.Osteochondral interface tissue engineering using macroscopicgradients of bioactive signals. Ann Biomed Eng 2010;38:2167e82.

24. Chen K, Ng KS, Ravi S, Goh JC, Toh SL. In vitro generation ofwhole osteochondral constructs using rabbit bone marrowstromal cells, employing a two-chambered co-culture welldesign. J Tissue Eng Regen Med 2013, http://dx.doi.org/10.1002/term.1716 [Epub ahead of print].

25. Cheng H-W, Luk KDK, Cheung KMC, Chan BP. In vitro gener-ation of an osteochondral interface from mesenchymal stemcell-collagen microspheres. Biomaterials 2011;32:1526e35.

26. Grayson WL, Bhumiratana S, Grace Chao PH, Hung CT, Vunjak-Novakovic G. Spatial regulation of human mesenchymal stemcell differentiation in engineered osteochondral constructs:effects of pre-differentiation, soluble factors and mediumperfusion. Osteoarthritis Cartilage 2010;18:714e23.

27. Castro NJ, Hacking SA, Zhang LG. Recent progress in interfacialtissue engineering approaches for osteochondral defects. AnnBiomed Eng 2012;40:1628e40.

28. Dhote V, Vernerey FJ. Mathematical model of the role ofdegradation on matrix development in hydrogel scaffold.Biomech Model Mechanobiol 2014;13:167e83.

29. Dickhut A, Pelttari K, Janicki P, Wagner W, Eckstein V,Egermann M, et al. Calcification or dedifferentiation: require-ment to lock mesenchymal stem cells in a desired differenti-ation stage. J Cell Physiol 2009;219:219e26.

30. Kronenberg HM. Developmental regulation of the growthplate. Nature 2003;423:332e6.

31. Farnum CE, Wilsman NJ. Cellular turnover at the chondro-osseous junction of growth plate cartilage: analysis by serialsections at the light microscopical level. J Orthop Res 1989;7:654e66.

32. Bhumiratana S, Eton RE, Oungoulian SR, Wan LQ, Ateshian GA,Vunjak-Novakovic G. Large, stratified, and mechanically func-tional human cartilage grown in vitro by mesenchymalcondensation. Proc Natl Acad Sci USA 2014;111:6940e5.

33. Taylor WR, Poepplau BM, K€onig C, Ehrig RM, Zachow S,Duda GN, et al. Themedial-lateral force distribution in the ovinestifle joint during walking. J Orthop Res 2011;29:567e71.

34. Lee-Shee NK, Dickey JP, Hurtig MB. Contact mechanics of theovine stifle during simulated early stance in gait. An in vitrostudy using robotics. Vet Comp Orthop Traumatol 2007;20:70e2.

35. Chase LG, Yang S, Zachar V, Yang Z, Lakshmipathy U,Bradford J, et al. Development and characterization of a clini-cally compliant xeno-free culture medium in goodmanufacturing practice for human multipotent mesenchymalstem cells. Stem Cells Transl Med 2012;1:750e8.

36. Gottipamula S, Muttigi MS, Chaansa S, Ashwin KM, Priya N,Kolkundkar U, et al. Large-scale expansion of pre-isolated bonemarrow mesenchymal stromal cells in serum-free conditions.J Tissue Eng Regen Med 2013, http://dx.doi.org/10.1002/term.1713 [Epub ahead of print].

37. Shanjani Y, De Croos JN, Pilliar RM, Kandel RA, Toyserkani E.Solid freeform fabrication and characterization of porous cal-cium polyphosphate structures for tissue engineering pur-poses. J Biomed Mater Res B Appl Biomater 2010;93:510e9.

38. Shanjani Y, Hu Y, Toyserkani E, Grynpas M, Kandel RA,Pilliar RM. Solid freeform fabrication of porous calcium poly-phosphate structures for bone substitute applications: in vivostudies. J Biomed Mater Res B Appl Biomater 2013;101:972e80.