Downloaded 2021-09-11T20:12:34Z Some rights reserved. For ...

Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published

version when available.

Downloaded 2022-05-25T20:13:02Z

Some rights reserved. For more information, please see the item record link above.

Title Preferential tendon stem cell response to growth factorsupplementation.

Author(s) Holladay, Carolyn; Abbah, Sunny-Akogwu

PublicationDate 2014

PublicationInformation

Holladay C, Abbah SA, O'Dowd C, Pandit A, Zeugolis DI.(2014) 'Preferential tendon stem cell response to growth factorsupplementation'. Journal Of Tissue Engineering AndRegenerative Medicine, .

Publisher Wiley

Link topublisher's

versionhttp://dx.doi.org/10.1002/term.1852

Item record http://hdl.handle.net/10379/4242

https://aran.library.nuigalway.ie

http://creativecommons.org/licenses/by-nc-nd/3.0/ie/

Preferential tendon stem cell response to growthfactor supplementationCarolyn Holladay1,2, Sunny-Akogwu Abbah1, Colm O’Dowd2, Abhay Pandit1 and Dimitrios I. Zeugolis1*1Network of Excellence for Functional Biomaterials (NFB), National University of Ireland Galway (NUI Galway), Ireland2Vornia Biomaterials, Galway, Ireland

Abstract

Tendon injuries are increasingly prevalent around the world, accounting for more than 100 000 newclinical cases/year in the USA alone. Cell-based therapies have been proposed as a therapeuticstrategy, with recent data advocating the use of tendon stem cells (TSCs) as a potential cell sourcewith clinical relevance for tendon regeneration. However, their in vitro expansion is problematic,as they lose their multipotency and change their protein expression profile in culture. Herein, weventured to assess the influence of insulin-like growth factor 1 (IGF-1), growth and differentiationfactor-5 (GDF-5) and transforming growth factor-β1 (TGFβ1) supplementation in TSC culture. IGF-1preserved multipotency for up to 28days. Upregulation of decorin and scleraxis expression wasobserved as compared to freshly isolated cells. GDF-5 treated cells exhibited reduced differentiationalong adipogenic and chondrogenic pathways after 28days, and decorin, scleraxis and collagen type Iexpression was increased. After 28days, TGFβ1 supplementation led to increased scleraxis,osteonectin and collagen type II expression. The varied responses to each growth factor may reflecttheir role in tendon repair, suggesting that: GDF-5 promotes the transition of tendon stem cellstowards tenocytes; TGFβ1 induces differentiation along several pathways, including a phenotypeindicative of fibrocartilage or calcified tendon, common problems in tendon healing; and IGF-1promotes proliferation and maintenance of TSC phenotypes, thereby creating a population sufficientto have a beneficial effect. Copyright © 2013 John Wiley & Sons, Ltd.

Received 17 March 2013; Revised 30 September 2013; Accepted 6 November 2013

Additional supporting information may be found in the online version of this article at the publisher’s web-site.

Keywords tendon stem cell phenotype maintenance; tendon stem cell differentiation; growth factorsupplementation; tendon stem cell culture; tenogenesis; tendon healing

1. Introduction

In the USA, tendon and ligament injuries account forapproximately 50% of all musculoskeletal conditions,with approximately 100 000 new cases/year (Zeugoliset al., 2011a, 2011bQ2 ). Tendon autografts, the goldstandard in clinical practice, which have demonstratedlimited success in knee and anterior cruciate ligamentsurgery (Crossett et al., 2002; Krych et al., 2008), stillpresent the disadvantages of creating a secondary injury

site during harvesting to only partially repair the injuredtissue (Ikeda et al., 2011; James et al., 2011a, 2011b Q3),and there is not sufficient supply to treat degenerativeconditions (Nakamura and Katsuki, 2002; Chu et al.,2008). Allograft and xenograft alternatives are character-ized by delayed remodelling and substantial stability andstill face the risk of potential transmission of infectiousdiseases (Stone et al., 2007). To address the clinical needfor functional tendon regeneration, numerous syntheticand natural biomaterials have been developed in recentyears (Zeugolis et al., 2011a, 2011b). However, syntheticsubstitution results in a non-functional and significantlythinner and weaker neotissue and/or in a disorderedfibrous capsule (Cao et al., 2006). Natural biopolymers,such as collagen (Kato et al., 1991; Cavallaro et al.,1994; Zeugolis et al., 2009; Kew et al., 2011; Kishore

* Correspondence to: Dimitrios I. Zeugolis, Network of Excel-lence for Functional Biomaterials (NFB), National Universityof Ireland Galway (NUI Galway), Ireland. E-mail: [email protected]

Copyright © 2013 John Wiley & Sons, Ltd.

JOURNAL OF TISSUE ENGINEERING AND REGENERATIVE MEDICINE RESEARCH ARTICLEJ Tissue Eng Regen Med (2013)Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/term.1852

Journal Code Article ID Dispatch: 17.12.13 CE:T E R M 1 8 5 2 No. of Pages: 16 ME:

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

et al., 2011), chitosan (Bagnaninchi et al., 2007), fibrin(Hohendorff et al., 2008) and silk (Kardestuncer et al.,2006; Sahoo et al., 2010) have also been assessed fortendon and ligament reconstruction, with promisingin vitro and in vivo results. However, only a few productshave been clinically approved to date (Derwin et al.,2006). In fact, there is little evidence to suggest that theinvasive new therapies have improved the 4–13% failurerate observed with autografts (Longo et al., 2010; Shearnet al., 2011).

To this end, cell-mediated strategies are under develop-ment for tendon and ligament regeneration. It is believedthat the limited number of resident tendon cells mightexplain the limited inherent regeneration capacity oftendon, and therefore delivery of an appropriate cellpopulation may be able to encourage regeneration whilstpreventing the development of fibrocartilage regions(Richardson et al., 2007; Sutter 2007; Bullough et al.,2008; Yin et al., 2010). Mesenchymal stem cells (MSCs)(Hoffmann et al., 2006; Krampera et al., 2006; Krygeret al., 2007; Ben-Arav et al., 2009; Schnabel et al., 2009)and adipose-derived stem cells (Park et al., 2010; Uysaland Mizuno, 2010, 2011) have been shown to havegreater proliferation rates than tenocytes, whilst animalmodels and clinical trials have demonstrated significantlyimproved biomechanical, biological and biochemicalcharacteristics in response to stem cell injection (Caplanet al., 1998; Young et al., 1998; Awad et al., 1999; Wanget al., 2005; Juncosa-Melvin et al., 2006). However,ectopic bone formation has been observed previouslyafter MSCs implantation in tendon models (Harris et al.,2004; Bi et al., 2007; Lui et al., 2011). Moreover, the lackof tendon-specific markers hampers quantification oftenogenic differentiation of these cells (Richardson et al.,2007). The recent discovery of tendon stem cells (TSCs)and their characterization in terms of multipotency,clonogenicity and self-renewal (Bi et al., 2007; Rui et al.,2010) may provide a valuable solution towards tendonregeneration. However, TSCs lose their phenotype in vitrowith time and passageing (Tan et al., 2012a, 2012bQ4 ), whichsignificantly limits their clinical potential.

Herein, it was hypothesized that treatment of TSCswith the appropriate concentration of growth factors willfacilitate ex vivo tendon stem cell phenotype maintenancein terms of multipotency, proliferation, marker expressionand extracellular matrix (ECM) production over longerintervals than previously possible. The rationale of usinggrowth factors to achieve this relies on the fact that oneof the most potent techniques for maintaining cell pheno-type, or indeed inducing cells to differentiate towards anew phenotype, is treatment with growth factors, aloneor in combination (Mitchell et al., 1993; Tateno andYoshizato, 1999; Huang et al., 2009). Induction oftenogenic differentiation in embryonic, adipose-derivedor bone marrow-derived stem cells has been attemptedusing a variety of growth factors, including insulin-likegrowth factor-1 (IGF-1), growth and differentiationfactor-5 (GDF-5) and transforming growth factor-β1(TGFβ1) (Farng et al., 2008; Kapacee et al., 2010;

Okamoto et al., 2010; Schneider et al., 2011). These threegrowth factors were selected because they representdifferent areas of growth factor-mediated effects. Specifi-cally, IGF-1 is known to play a pivotal role in the mainte-nance of tenocyte phenotype (Costa et al., 2006; Qiuet al., 2012) and tendon repair after injury (Klein et al.,2002a, 2002b Q5; Qiu et al., 2012; Raghavan et al., 2012)and has been used in a clinical trial to enhance collagensynthesis in the patellar tendon of Ehlers–Danlos patientsand healthy controls (ClinicalTrials.Gov; Trial ID:NCT01446783). GDF-5 supplementation has been shownto maintain tenocyte phenotype (Wolfman et al., 1997;Hogan et al., 2011; Keller et al., 2011), promote tenogenicdifferentiation (Hayashi et al., 2011; James et al., 2011a,2011b; Tan et al., 2012a, 2012b) and is known to play acritical role in tendon development (Francis-West et al.,1999; Clark et al., 2001; Chhabra et al., 2003). TGFβ1 ishighly upregulated during development (Pryce et al.,2009) and after tendon injury (Molloy et al., 2003;Kashiwagi et al., 2004; James et al., 2008) and has beenused as a supplement in a variety of tenocyte culture stud-ies (Wolfman et al., 1997; Klein et al., 2002a, 2002b;Schneider et al., 2011; Chen et al., 2012; Heisterbachet al., 2012; Laumonier et al., 2012; Mendias et al.,2012). Furthermore, it is a key component of platelet-richplasma (PRP), an increasingly popular treatment fortendon strain and tendinopathy (Chen et al., 2012).

2. Materials and methods

2.1. MaterialsAll primary cells were obtained in accordance with theethical guidelines of the National University of Ireland,Galway, from male Lewis rats aged 8–9weeks (CharlesRiver, UK). Growth and differentiation factor-5 (GDF-5)and insulin-like growth factor-1 (IGF-1) were obtainedfrom ProSpec (Israel). Transforming growth factor-β1(TGFβ1) was obtained from Millipore (Ireland). Unlessotherwise specified, all other consumables and cell cul-ture media were obtained from Sigma-Aldrich (Ireland)or Fisher Scientific (Ireland).

2.2. Tendon stem cell isolation andcharacterization

Rat patellar tendon-derived stem cells (TSCs) wereharvested as described previously (Bi et al., 2007). Briefly,the tendons were explanted and cleaned aseptically witha scalpel to remove all paratendon, fat and muscle. Thetendons were then minced using a scalpel into 1mm-thickslices and placed in a 0.3% w/w collagenase solution(Clostridium histolyticum, > 125 CDU/mg; Sigma-Aldrich,Ireland). After 30min, the partially digested tendon wasplaced on a 70μm cell strainer and washed with Hank’sbalanced salt solution (HBSS). The tendon was thenplaced in fresh 0.3% collagenase solution and digested

2 C. Holladay et al.

Copyright © 2013 John Wiley & Sons, Ltd. J Tissue Eng Regen Med (2013)DOI: 10.1002/term

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

at 37°C for 2 h. The liquid portion after this digest wasfiltered through a 70μm cell strainer and the remainingtendon was washed in basal TSC medium (Vornia Bioma-terials Ltd, Ireland). The filtered fraction was centrifugedat 300 × g for 5min; the supernatant was removed, thenresuspended in basal TSC medium, placed in a tissueculture flask and allowed to attach for 5 days withoutdisturbance. After 10 days, colonies were identified andthe morphology compared to the literature.

Characterization for multipotency, clonogenicity, sur-face marker expression and extracellular matrix (ECM)production (see below) was carried out when the cellsreached 80% confluency (approximately 14–21 days).Standard chondrogenic, osteogenic and adipogenic induc-tion protocols were followed, using a Mesenchymal StemCell Functional Analysis kit according to the manufac-turer’s protocol (SC010, R&D Systems, UK). Osteogenicand adipogenic induction experiments were conductedby culturing TSC monolayers in media supplemented withthe osteogenic or adipogenic supplements provided withthe kit. Adipogenic induction could be observed as soonas 10 days after induction. Osteogenic induction required21 days. To analyse chondrogenic potential, TSC pelletswere treated with chondrogenic induction medium for28 days and then the pellet was fixed and cryosectioned.Immunohistochemical staining for fatty acid bindingprotein-4 (FABP4), osteonectin and collagen type II(R&D Systems) was used to quantify the fraction of cellsthat were successfully induced (adipogenic, osteogenicand chondrogenic induction, respectively) and oil red O,alizarin red and toluidine blue staining was used toconfirm the antibody staining qualitatively.

Standard clonogenicity assays were carried out toconfirm clonogenicity over time. Briefly, TSCs were platedat density of 10 cells/cm2 and cultured for 7 days. The cellswere then stained with 0.5% crystal violet (Sigma-Aldrich)and counted. Surface marker expression was analysed viaflow cytometry (BD FACS Canto, UK) and immunohisto-chemical staining. For flow cytometry, trypsinized cells(5 × 105) were stained with APC-Cy7 anti-CD90+,AlexaFluor 647 anti-CD44+, PerCP anti-CD34– and FITCanti-CD31–. These results were confirmed with immunohis-tochemical staining with unlabelled analogues (Santa Cruz,USA) and a FITC-labelled donkey anti-mouse secondaryantibody (Invitrogen, Ireland). Tendon-specific markersand ECMmolecules were also analysed immunohistochem-ically, using tenomodulin (TNMD) and scleraxis (SCX) ordecorin (DCN), collagen type I and tenascin-c (TEN-C),respectively. The expression of collagen type II, osteonectinand FABP4 were also analysed immunohistochemicallywithout any induction, using the same antibodies as in themultipotency analysis.

For the cell surface markers, the positive cell fraction(i.e. fraction of cells expressing the given markers) wasquantified using flow cytometry gated to threshold levelsusing positive and negative controls. These measure-ments were then confirmed using immunohistochemicalstaining and image analysis software (Image Proplus v 5,Media Cybernetics), with thresholds set using positive

and negative controls. For markers such as collagen typeII and osteonectin, which are secreted proteins, onlyimmunohistochemical analysis was used, as the secretedproteins were not associated with the cell membraneenough to be quantified using flow cytometry.

2.3. Growth factor treatment study outline

TSC populations from three donors were used in allstudies to form a biological replicate of three. The studyoutline is illustrated in Figure F11. At time 0 (approximately18 days after the cells were extracted), 200 000 cells wereseeded into 30 flasks (10 flasks/donor). At this point, asubset of these cells was set aside for multipotencyanalysis and their baseline expression of markers andECM molecules was quantified. One flask/donor wasselected for treatment with 1, 10 and 100 ng/ml of eachgrowth factor and one flask/donor was set aside andtreated with basal tendon cell medium (0ng/ml growthfactor). The cells in the treatment groups were treated withthe appropriate growth factor-supplementedmedium every2–3days for 14days. At this time point, the cellswere trypsinized and subsets of the cells from each flask(200 000/flask) were seeded into a new flask and culturedto continue the study. The remaining cells were preservedin RIPA buffer (Thermo Scientific, Germany) for ELISAanalysis, seeded for multipotency analysis, or seeded ontooptical plates for immunohistochemistry. This procedure

Figure 1. Diagram depicting study donors. These cells weretrypsinized at day 0 and either seeded into new flasks for the studyor analysed for multipotency, marker expression and ECM proteinexpression. The seeded flasks were treated with growth factors(replenished every 2–3days) for 14days. At this point, the flaskswere all trypsinized and either seeded into new flasks and culturedfor another 14days (28 total) or used for analysis. After 28days, allcells were analysed for multipotency, marker expression and ECMprotein expression

Colour

online,B&W

inprint

Growth factor supplementation of tendon stem cell culture 3


12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

was repeated after 28days without reseeding of the flasks.Thus, all of the cells had been trypsinized three times by theend of the 28days.

2.4. Multipotency studies

At each time point, cells were trypsinized and seeded formultipotency analysis. As described in Section 2.2, standardosteogenic and adipogenic induction protocols werefollowed, using the R&D Systems (UK) Functional AnalysisKit. Immunohistochemical staining for FABP4, osteonectinand collagen type II was used to quantify the fraction ofcells that were successfully induced, and oil red O, alizarinred and alcian blue staining were used to confirm theantibody staining qualitatively. The fraction of cells express-ing each marker was analysed in order to compare theefficacy of each growth factor in phenotype maintenance.

2.5. Immunohistochemistry analysis

Immunohistochemical analysis was used to determine theexpression of CD31, CD34, CD44, CD90, DCN, SCX,osteonectin, collagen types I and II, FABP4 and TEN-C. Inall cases, the cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Anti-CD44 and anti-CD90were obtained from BioLegend (UK), anti-collagen II andanti-FABP4 from R&D Systems (UK) and the rest fromSanta Cruz (USA). The four CD markers were selected inlight of the work previously conducted to demonstrate theTSC phenotype in rats (Rui et al., 2010; Zhang et al.,2011). The antibodies to these markers were raised in mice,as was the anti-FABP4; anti-TNMD was raised in rabbit;anti-collagen type II was raised in sheep; and the antibodiesto DCN, SCX, osteonectin, collagen type I and TEN-Cwere raised in goat. The secondary antibodies usedwere: AlexaFluor 488 donkey anti-mouse (Invitrogen);AlexaFluor 488 mouse anti-sheep (Invitrogen); Texas reddonkey anti-rabbit (Invitrogen, Ireland); FITC rabbit anti-goat (Vector Laboratories, UK); and AlexaFluor 594 donkeyanti-goat (Invitrogen). When dual staining was used, theantibodies were tested on control selections to ensurenegligible interactions.

2.6. Direct enzyme-linked immunosorbent assays(ELISAs)

As suitable antibody pairs for sandwich ELISA were notavailable for most of the target proteins, direct ELISAprotocols were optimized instead, using the peptidesagainst the antibody-binding sequences as a standard.These peptides are approximately 20 amino acids inlength and are matched to the antibody. It should benoted that the measured concentration refers to themass of antibody-binding sequences/volume, not themass of the full protein; in the case of large proteogly-cans, such as DCN, this is an important distinction. The

exception to this was collagen type II, as a collagentype II standard was available at a high purity andconcentration.

Briefly, antigen-containing samples were coated ontoan immunoplate (Thermo Scientific, Germany) overnightat room temperature. The following day, samples wereblocked with 5% bovine serum albumin (BSA) and 0.1%Tween in phosphate-buffered saline (PBS), treated with theappropriate primary antibody at a 1:250 dilution, treatedwith anti-goat HRP at 1:500 dilution, and developed usingOne-step™ Ultra TMB-ELISA (Thermo Scientific). The pro-tein concentration was normalized to total protein contentusing a BCA assay (Thermo Scientific).

2.7. Histological staining

Standard oil red O, alizarin red and toluidine bluestaining protocols were used to confirm the results ofthe immunohistochemical multipotency analysis andcheck for unintended differentiation of the stem cells ateach time point. Briefly, adipogenic cultures were fixed,rinsed, incubated for 5min with 60% isopropanol andincubated for 5min with oil red O stock solution. The oilred O was then removed and the cells washed thoroughlywith tap water before counterstaining with haematoxylin.Osteogenic cultures were fixed, washed with PBS,incubated for 20min in alizarin red working solution,washed with distilled water, and washed again withPBS. Sections of chondrogenic pellets were hydrated indistilled water and stained in the toluidine blue workingsolution for 3min. The sections were then washed indistilled water, dehydrated and mounted.

2.8. Statistical analysis

All statistical analyses were conducted using IBM SPSS v. 20.Normality was calculated using the Kolmogorov–Smirnovtest and/or the Shapiro–Wilk test. Non-parametric datawere compared using the independent samples Mann–Whitney U-test or the independent samples Kruskal–Wallistest. Normal data with equal variance were analysed usingone-way analysis of variance (ANOVA) to determinestatistical significance within groups, and the Bonferronipost hoc test was used to determine where the differenceslay. Outliers that lay outside of three interquartile rangeswere removed from the dataset. Significance was set atp< 0.05 and all graphs are presented as mean± standarderror of the mean (SEM).

3. Results

3.1. TSC differentiation potential and markerexpression in vitro

When extracted from tendon, TSCs rapidly lost theirdifferentiation potential. A significant loss of adipogenic



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

(p< 0.0001) and osteogenic (p< 0.0001) potential wasobserved in terms of the fraction of cells expressingFABP4 and osteonectin, respectively (FigureF2 2a). Therewas also a decrease in the fraction of cells expressingCD90 after 14 days (p< 0.001) and significant increasesin the fractions of cells expressing CD31 after 28 days(p< 0.0001), implying significant changes in thephenotype of the cells (Figure 2b). CD34 expressionsignificantly decreased from approximately 6% to negli-gible between day 0 and day 14 (p< 0.001) and CD44expression did not significantly change from 100% at

any time point. Representative images used for theseanalyses are included in Figure S1 (see supportinginformation). Protein expression analysis on day 14revealed an increase in all assayed proteins, includingDCN, SCX, TEN-C, collagen type I, osteonectin andcollagen type II (Figure 2c). Subsequent protein eval-uation on day 28 revealed that, besides SCX, theexpression level of the other proteins decreasedsignificantly when compared to freshly isolated cells,as well as to the levels on day 14 (p< 0.0001). Theexpression of SCX was significantly higher on day 28

Figure 2.Q13 TSC phenotype is altered over time. The fraction of cells expressing differentiation markers (a) decreased over time in theadipogenic and osteogenic groups, although there was no loss in chondrogenic potential over time. In terms of marker expression (b),there was no decrease in the fraction of cells expressing tenomodulin or CD44, while there was a significant decrease in the fraction ofcells expressing CD34 (from 6% to 0%) and a temporary decrease in the fraction of cells expressing CD90. There was also a significantincrease in the fraction of cells expressing CD31 over time, suggesting a change in phenotype. The protein expression (c) in theuntreated cells was variable with time, with significant increases in all of the proteins after 14days, but only SCX and collagen typeII had significant changes after 28days. Qualitatively (d), the expression of the markers was significantly reduced over time and,although collagen type II-expressing chondrogenic pellets did form, the collagen was less dense and the pellets were not as wellformed. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

abbah

Inserted Text

. Scale bar represents 100 micrometer.

abbah

Typewritten Text

abbah

Typewritten Text

compared to day 14 (p< 0.0001). Similarly, Figure 2dshows that cells in pellets were capable of expressingosteonectin, FABP4 and collagen type II after 0, 14and 28 days. However, the densities of cells express-ing these markers appeared to decrease with time inthe untreated cells (Figure 2d).

3.2. TSC multipotency in the presence of IGF-1

After treatment for 14 and 28 days with 10 ng/ml IGF-1,there was no significant loss in adipogenic (p> 0.05;FigureF3 3a) or osteogenic (p> 0.05; Figure 3b) potential.Osteogenic potential was maintained with all doses ofIGF-1 after 28 days. There was a transient decrease inthe fraction of cells expressing CD90 in the untreatedcells and in cells treated with 100 ng/ml IGF-1 after14 days (p< 0.001 and p< 0.003, respectively; Figure 3c)but this effect was not statistically significant after

28 days. Furthermore, while the fractions of untreatedcells expressing CD31 significantly increased after 14and 28 days in comparison to the day 0 cells (p< 0.0001;Figure 3d), there were no significant increases in thegroups treated with 10 or 100 ng/ml IGF-1. After 28 days,TSCs from all IGF-1-treated groups were capable offorming collagen type II-expressing pellets (see supportinginformation, Figure S2a).

3.3. TSC marker expression in the presence ofIGF-1

IGF-1 supplementation did not affect DCN, TEN-C andosteonectin levels (Figure F44a, c, f, respectively; p> 0.05).SCX expression (Figure 4b) was slightly increased at alldoses after 28days (p< 0.002), whilst collagen type Iexpression (Figure 4d) was slightly decreased atall doses after 14 days and at 10 and 100 ng/ml

Figure 3. IGF-1 supplementation reduced the loss in stem cell phenotype observed in the TSCs. There was no significant decrease inadipogenic differentiation (a) in the group treatedwith 10ng/ml at either time point and no change after 28days in the 100ng/ml group.The loss in osteogenic potential (b) was also reduced to negligible in the 10ng/ml group and was not significant after 28days at any dose.Chondrogenic potential was also evaluated and all groups were capable of forming collagen type II-expressing pellets at all time points.The temporary decrease in CD90 expression (c) observed after 14days in the untreated cells was not observed in the group treated with10ng/ml IGF-1, neither was there any significant increase in CD31 expression (d) in either the 10 or 100ng/ml groups at any time point.The 1ng/ml dose did not appear sufficient to completely prevent changes in phenotype, but it was partially effective as compared tounsupplemented medium. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colouronlin

e,B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

after28 days (p< 0.003 and p< 0.001, respectively).Markers of adipogenic differentiation were not ob-served (see supporting information, Figure S3a) and astatistically significant downregulation of collagen typeII expressionwasobservedat all doses (p< 0.006; Figure4e).No significant decrease in proliferation was observed after28days (p> 0.05; FigureF5 5a).

3.4. TSC multipotency in the presence of GDF-5

After 14 and 28 days of TSC culture in media supplemen-ted with 1, 10 or 100 ng/ml GDF-5, significant losses inTSC multipotency were observed (Figure F66). Significantdecreases in adipogenic potential were observed afteronly 14 days at all doses (p< 0.009) and after 28 days at

Figure 4. IGF-1 supplementation maintained the levels of some, but not all, tendon-, bone- and cartilage-related proteins. SCX ex-pression slightly increased over time in all groups (b), but DCN (a), TEN-C (c) and osteonectin (f) expression did not significantlychange in the groups treated with IGF-1 at any dose. Collagen type I expression decreased at all doses, although this effect was onlystatistically significant after 28days in the 10 and 100ng/ml groups. There was also decreased collagen type II expression in allgroups at both 14 and 28days. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

1 and 100ng/ml doses of GDF-5 (p< 0.008; Figure 6a).Osteogenic potential was significantly decreased after14 days in all groups (p< 0.026; Figure 6b), but this effectwas no longer significant after 28 days. CD90 expressionwas also reduced after 14 days in cells treated with0–10 ng/ml GDF-5 doses (p< 0.001Q6 ) and after 28 days ingroups treated with 10 and 100 ng/ml doses of GDF-5 (p0.022) and CD31 expression was increased (p< 0.035).Chondrogenic pellets formed from TSCs treated for28 days with 10 and 100 ng/ml GDF-5 dissolved be-fore culture was complete, suggesting a significantdecrease in chondrogenic differentiation potential atthe higher GDF-5 doses (see supporting information,Figure S2b).

3.5. TSC marker expression in the presence ofGDF-5

GDF-5 treatment of TSC cultures induced statistically sig-nificant changes in expression levels of ECM and tendon-

related proteins. DCN expression was slightly increasedafter 28 days in the 1 ng/ml group (p< 0.025; Figure F77a),while SCX expression was elevated compared to theuntreated controls after 28 days in all groups (p< 0.003;Figure 7b). TEN-C was downregulated in the 10 ng/mlgroups after 28 days (p< 0.042; Figure 6c) and collagentype I expression was significantly decreased in allgroups after 14 days (p< 0.004; Figure 6d), butunchanged after 28 days. Significant decreases wereobserved in collagen type II expression at the 1 and100 ng/ml doses (Figure 6e). After 14 days, osteonectinexpression was significantly increased in the 1 ng/mlgroup and decreased in the 100 ng/ml group (p< 0.002and p< 0.039, respectively; Figure 6f). After 28 days,however, no significant changes in osteonectinexpression were observed in TSCs treated with GDF-5(p> 0.05). FABP4 expression was negligible in allsamples (see supporting information, Figure S3b). Nosignificant change in proliferation was observed after28 days (p> 0.05; Figure 5c).

Figure 5. Proliferation as a function of time and growth factor dose. The fold change in cell numbers was used as a measure of pro-liferation. The cell proliferation rate between day 28 and day 14 did not significantly decrease as compared to the fold change betweenday 14 and day 0 in the cells treated with IGF-1 at any dose (a), although the fold change in cell numbers from day 0 to 14 was signif-icantly less in the 100ng/ml group than in the untreated cells. The proliferation rate temporarily decreased in the cells treated with1–100ng/ml GDF-5 (b; as compared to the untreated cells) but recovered after 28days to be greater than that of the untreated cells.The proliferation rate of TSCs treated with TGFβ1 (c) significantly decreased as a function of dose after 28days and was significantlyincreased in all groups after 14days. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

3.6. TSC multipotency in the presence of TGFβ1

Significant losses in adipogenic potential were observed inresponse to treatment with all doses of TGFβ1 after 14days(p< 0.014; FigureF8 8a). Osteogenic potential was alsodecreased in all groups after 14days (p< 0.048), but thiseffect was not statistically significant after 28days(Figure 8b). CD90 expression was significantly decreasedafter 14days in all treatment groups (p< 0.019; Figure 8c)but not after 28days, while CD31 expression significantlyincreased after 28 days in all groups (p< 0.004) andafter 14 days in the group treated with 100 ng/ml TGFβ1(p< 0.03; Figure 8d). All groups at all time points formedcollagen type II-expressing chondrogenic pellets (seesupporting information, Figure S2c).

3.7. TSC marker expression in the presence ofTGFβ1

After 28days, there were significant changes in the expres-sion of several markers. DCN was significantly upregulated

after 28 days in the group treated with 1 ng/ml TGFβ1, ascompared to initial DCN expression levels (p< 0.003;Figure F99a). SCX expression after 28 days was significantlyupregulated in comparison to time 0 at all doses of TGFβ1(p< 0.014; Figure 9b). A significant increase in TEN-Cexpression was observed in the 100 ng/ml group after28 days (p< 0.04; Figure 9c). Collagen type I expressionwas also increased after 28 days in the 1 and 100 ng/mlgroups (p< 0.047; Figure 9d). Collagen type II expres-sion was significantly increased (levels more thandoubled) at all doses after 28 days in comparison to TSCsat day 0 (p< 0.005; Figure 9e). Osteonectin expressionwas significantly increased when treated with1 ng/mlTGFβ1 and decreased after 14 days in response to both10 and 100 ng/ml TGFβ1 (p< 0.048; Figure 9f). Thesechanges are suggestive of a chondrocyte-like pheno-type in these cells or in a subpopulation of the cells,which was further supported by the rounded shapeof many of the cells and the staining for sulphatedglycosaminoglycans in these cells (toluidine blue)(Figure 9g). A significant decrease in cell proliferationas compared to untreated TSCs was observed at both

Figure 6. GDF-5 supplementation did not maintain the stemness of the TSCs in terms of either adipogenic (a) or osteogenic (b)differentiation. Furthermore, TSCs treated with 10 or 100ng/ml GDF-5 formed chondrogenic pellets but they dissolved before culturewas complete, making it impossible to assess their collagen type II production. Significant decreases in CD90 expression wereobserved at both 10 and 100ng/ml doses after 28days (c) and CD31 expression increased significantly over time at the 10 and100ng/ml doses as well. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

time points in the 10 and 100 ng/ml groups,suggesting the acquisition of a less proliferative phe-notype (p< 0.0001; Figure 9h). FABP4 expressionwas negligible in all groups (see supporting informa-tion, Figure S3c).

4. DiscussionAlthough cell-based therapies are becoming increasinglyimportant in the field of tissue engineering and regenera-tive medicine, widespread acceptance and clinical

Figure 7. GDF-5 supplementation increased the expression of tendon-related markers. After 28days, DCN expression (a) increased signifi-cantly at the 1ng/ml dose, while SCX expression increased significantly at all doses (b). TEN-C expression (c) decreased marginally in re-sponse to the 10ng/ml dose after 28days, while collagen type I expression was downregulated temporarily at day 14 in all groups, butrecovered after 28days (d). Collagen type II expression decreased in both the 1 and 100ng/ml groups (e), while there were temporarychanges after 14days in osteonectin expression that were not statistically significant after 28days (f). No FABP4 expression or fat dropletswereobserved in any groupat any timepoint.Dataare expressed asmean±SEM; *significant change from the values for freshly isolated cells

Colouronline,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

translation is still limited, owing to cell phenotype drift inculture and subsequent loss of therapeutic potential.Growth factor supplementation has been advocated toremedy tenocyte phenotypic loss in culture (Klein et al.,2002a, 2002b; Takahashi et al., 2002; Tang et al., 2003;Thomopoulos et al., 2005, 2010; Keller et al., 2011; Qiuet al., 2012; Caliari and Harley, 2013) and has beenextensively studied as means of modulating stem celldifferentiation (Musarò et al., 2004; Farng et al., 2008;Hayashi et al., 2011; Rose et al., 2012; Trosan et al.,2012). In the present study we evaluated the influenceof growth factor supplementation on the maintenance oftendon stem cell phenotype, a cell population known tolose its phenotype in culture with time and passageing(Tan et al., 2012a, 2012b).

A significant loss in adipogenic and osteogenic capacitywas observed, although the rate of this loss was fasterthan described elsewhere (within three passages, notfive). This could be attributable to the deliberately lowseeding numbers and, thus, the longer culture timebetween passages (i.e. phenotypic loss was likely similarwhen comparing overall culture time instead of passages).

Further changes in CD90 or CD44 expression may nothave been visible until after longer culture times. Interest-ingly, an increase in the expression of CD31, a moleculethat plays key role in angiogenesis, was observed overthe 28 day assay period in all groups, particularly withTGFβ1 treatment. CD34 expression, as expected, wasobserved to be low to negligible at all time points. Impor-tantly, we observed increased expression of SCX at 14 and28 days, as compared to freshly isolated cells. However,this finding is in contrast to a previous report showingdecreased SCX expression with time (Tan et al., 2012a,2012b). This is also in variance to the apparent biphasicexpression of the other proteins evaluated, showingsignificantly higher expression levels on day 14 comparedto day 28. The increased expression of SCX observed inthe present study could indicate a shift of TSC from theprimitive (stemness) phenotype with capacity to differen-tiate into bone- and cartilage-like tissues to a more maturetenocyte phenotype with diminished multipotency. This isconsistent with the progressive loss in multipotency, asrevealed by downregulation of markers for osteogenesisand adipogenesis.

Figure 8. TGFβ1 induced differentiation of TSCs away from their native phenotype. At all doses and time points, significant decreasesin adipogenic differentiation were observed (a). All groups were able to form collagen type II-expressing pellets. After 14days, signif-icant decreases in osteogenic potential were also observed (b), but this effect was not significant after 28days. The temporary de-crease in the fraction of cells expressing CD90 (c) observed in the untreated cells was similar to that in TGFβ1-treated cells at alltime points, and the increase in the fraction of cells expressing CD31 (d) was more pronounced after treatment with TGFβ1 in allgroups after 28days. Data are expressed as mean±SEM; *significant change from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

Figure 9. TGFβ1 induced significant changes in expression of tenogenic markers and significantly upregulated production of osteo-genic and chondrogenic genes. DCN expression (a) was significantly upregulated after 28days by treatment with 1ng/ml TGFβ1,but unaffected at all other doses. SCX expression (b) was increased at all doses after 28days. TEN-C expression (c) was increased onlyat the 100ng/ml dose after 28days, while collagen type I synthesis (d) was upregulated at both the 1 and 100ng/ml doses after28days. Collagen type II expression (e) was significantly upregulated in response to all doses after 28days, as was osteonectinexpression. Microscopic examination showed the presence of small, rounded cells with strong staining for glycosaminoglycans andproteoglycans (g), and estimation of the change in proliferation rate indicated a significant decrease in the proliferation characteris-tics of terminally differentiated, relatively quiescent cells such as chondrocytes. Data are expressed as mean±SEM; *significantchange from the values for freshly isolated cells

Colou

ron

line,

B&W

inprint



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

Considering the influence of the different growthfactors in the present study on TSC phenotype and differ-entiation, IGF-1 treatment at 10 or 100 ng/ml maintainedTSC multipotency and phenotype, evident by mainte-nance in the fraction of cells capable of differentiatingdown adipogenic, osteogenic and chondrogenic pathwaysand the expression of CD90, CD31, DCN, TEN-C andosteonectin. This observation is in accordance with previ-ous studies, where blocking the IGF-1 receptor resulted insignificantly decreased SSEA3 expression and increasedapoptosis in cultures of embryonic stem cells (Bendallet al., 2007), and in decreased Oct-4 and Nanog expres-sion in spermatogonial stem cells (Huang et al., 2009),demonstrating the crucial role of IGF-1 signalling inmaintaining stem cells in culture. In a broader context,IGF-1 overexpression attenuated the effects of ageing oncardiac stem cells in mice (Capogrossi, 2004) and mayplay a role in delaying senescence in humans (Rinconet al., 2004), suggesting a potent role in maintainingmultiple cellular functions. In the context of tendons,IGF-1 injections have been investigated as a potentialtherapeutic tool, both in animal studies (Kurtz et al.,1999) and in a human clinical trial (Trial ID:NCT01446783). The results have been encouraging,suggesting a role of IGF-1 in strengthening the tendinousECM (Kurtz et al., 1999). In tenocyte cultures, IGF-1supplementation has been employed to promote tenocytephenotype maintenance, in terms of scleraxis, decorin andtenomodulin expression (Klein et al., 2002a, 2002b; Qiuet al., 2012) and proliferation (Thomopoulos et al.,2010; Caliari and Harley, 2013). This may explain theincreased scleraxis expression observed in all groups after28 days.

Treatment with GDF-5, on the other hand, resulted inthe preservation of tendon-related phenotype. However,this treatment was unable to preserve adipogenic andchondrogenic potentials. The loss in multipotency wascoupled with a significant increase in SCX expression atall time points without significant increases in collagentype II or osteonectin expression, which would be associ-ated with chondrogenic or osteogenic differentiation,respectively. These findings are consistent with other pub-lished reports which have used GDF-5 to induce a tendon-like phenotype in mesenchymal stem cells (Farng et al.,2008; Hayashi et al., 2011) and adipose-derived stem cells(Park et al., 2010; James et al., 2011a, 2011b), or to main-tain tenocyte phenotype during in vitro culture (Hoganet al., 2011; Keller et al., 2011). In some of these studies,GDF-5 supplementation was used in combination withanother stimulus, such as topography (James et al.,2011a, 2011b) or mechanical stimulation (Farng et al.,2008). This suggests that GDF-5 may also be an idealsupplement for the culture of tendon-like constructsgrown on textured materials or cultured in bioreactors,where they can be mechanically stimulated. Our findingsare also consistent with the known importance of GDF-5in tenogenesis during development (Wolfman et al.,1997; Francis-West et al., 1999; Stone, 2000) and intendon repair after injury (Stone, 2000; Chhabra et al.,

2003). Although the mechanism of GDF-5 action intendon development and repair is not yet fully identified,these results further support the use of GDF-5 formaintaining tenocyte cultures and potentially inducingtenogenic differentiation.

Finally, although treatment with TGFβ1 did not signifi-cantly enhance the expression of tendon-related markersin the present study, it appeared to induce a chondro-cyte-like phenotype, characterized by increased collagentype II expression, reduced proliferation, increased GAGstaining and increased numbers of round-shaped cells,as has been observed previously (Kawamura et al.,2005). Treatment with TGFβ1 also increased the fractionof cells expressing CD31 and the levels of osteonectin(at the lowest dose only), suggesting that someosteogenic differentiation may have occurred. Consider-ing the role of TGFβ1 as a potent regulator of prolifera-tion, differentiation and cytokine secretion (Molloy et al.,2003; Kashiwagi et al., 2004), it was expected to inducemore than one type of differentiation. In this case, the in-creased expression of markers associated with cartilage,bone and endothelium suggests that TGFβ1 upregulationobserved after injury may actually contribute to thedevelopment of vascularized fibrocartilaginous regionsor calcified areas seen in tendinopathy (Xu and Murrell,2008). As these regions have poor mechanical propertiesand may contribute to degeneration of the surroundingtendon, this natural upregulation of TGFβ1 may be morepathogenic than reparative. In fact, if TGFβ1 upregulationcauses differentiation of TSCs, it may increase the speedof basic repair by inducing the cells to differentiate andproduce more ECM (i.e. DCN, TEN-C and collagens) butlimit the overall tendon repair capacity by reducing thenumbers of TSCs or redirecting their differentiationtowards osteogenesis and chondrogenesis, as has previ-ously been hypothesized (Rui et al., 2011). Thus, it wouldappear that TGFβ1 supplementation of TSC cultures gen-erates an in vitro cell population more representative oftendinopathy than healthy tendon, and would be morevaluable in generating disease models than in preparationof cells for therapeutic injections.

TSCs have been shown to spontaneously lose theirmultipotency and self-renewal when cultured in vitro(Tan et al., 2012a, 2012b; Zhang and Wang, 2013). Devel-opment of in vitro cell culture conditions that enhance themaintenance of their stemness and multipotency is criticalin the quest to realize their therapeutic potential. In thepresent study, treatment of TSCs with IGF-1 and, to alesser extent, with GDF-5 enabled the expression oftendon-related markers during the 28day culture period.TGFβ1, on the other hand, promoted a more chondrogenicphenotype.

5. Conclusions

TSC phenotype is lost over time in culture, evident by a re-duction in the expression of related stem cell surface



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

markers, concurrent with a significant upregulation ofcollagen type II and osteonectin expressions, depictinglineage commitment. Continuous supplementation with10 ng/ml IGF-1 significantly improved the maintenanceof the TSC phenotype for up to 28 days. Whereas supple-mentation with GDF-5 maintained and promoted markersassociated with TSC phenotype (SCX, TEN-C and DCN), itwas ineffective in maintaining their multipotency. TGFβ1,on the other hand, induced differentiation of a subset ofcells towards a chondrogenic phenotype. Therefore, wedemonstrate here that continuous supplementation ofbasic cell culture media with 10 or 100 ng/ml IGF-1maintains TSC phenotype. This may prove particularlyuseful in allowing long-term expansion of these cells,thereby greatly improving their therapeutic potential.

Conflict of interest

The authors h Q14ave declared that there is no conflict ofinterest.

Acknowledgements

The authors thank Vornia Ltd and members of the NFB for theirassistance in this project, specifically Andrew English and AyeshaAzeem. This study has received funding (to D.Z.) from the EuropeanUnion Seventh Framework Programme (Grant No. FP7/2007-2013)under the Marie Curie, Industry–Academia Partnerships andPathways (IAPP) award, part of the People programme (under GrantAgreement No. 251385, Tendon Regeneration) and ScienceFoundation Ireland (Project No. SFI_09-RFP-ENM2483).

References

Awad HA, Butler DL, Boivin GP et al. 1999;Autologous mesenchymal stem cell-mediated repair of tendon. Tissue Eng 5:267–277.

Bagnaninchi PO, Yang Y, Zghoul N et al.2007; Chitosan microchannel scaffolds fortendon tissue engineering characterizedusing optical coherence tomography.Tissue Eng 13: 323–331.

Ben-Arav A, Snedeker JG, Zilberman Y et al.2009; Smad8/BMP-2 engineered mesen-chymal stem cells induce rapid Achillestendon full defect regeneration. Mol Ther17: S267–267.

Bendall SC, Stewart MH, Menendez P et al.2007; IGF and FGF cooperatively establishthe regulatory stem cell niche ofpluripotent human cells in vitro. Nature448: 1015–1021.

Bi YM, Ehirchiou D, Kilts TM et al. 2007;Identification of tendon stem/progenitorcells and the role of the extracellular ma-trix in their niche. Nat Med 13: 1219–1227.

Bullough R, Finnigan T, Kay A et al. 2008;Tendon repair through stem cell interven-tion: cellular and molecular approaches.Disabil Rehabil 30: 1746–1751.

Caliari SR, Harley B. 2013; Compositegrowth factor supplementation strategiesto enhance tenocyte bioactivity inaligned collagen GAG scaffolds. TissueEng A 19: 1100–1112.

Cao D, Liu W, Wei X et al. 2006; In vitro ten-don engineering with avian tenocytes andpolyglycolic acids: a preliminary report.Tissue Eng 12: 1369–1377.

Caplan AI, Fink DJ, Bruder SP et al. 1998;The regeneration of skeletal tissues withmesenchymal stem cells. In Frontiers inTissue Engineering. Pergamon: Oxford;471–480.

Capogrossi MC. 2004; Cardiac stem cells failwith aging: a new mechanism for the age-dependent decline in cardiac function. CircRes 94: 411–413.

Cavallaro JF, Kemp PD, Kraus KH. 1994; Col-lagen fabrics as biomaterials. BiotechnolBioeng 43: 781–791.

Chen L, Dong SW, Tao X et al. 2012; Autolo-gous platelet-rich clot releasate stimulatesproliferation and inhibits differentiationof adult rat tendon stem cells towards

nontenocyte lineages. J Int Med Res 40:1399–1409.

Chhabra A, Tsou D, Clark RT et al. 2003;GDF-5 deficiency in mice delays Achillestendon healing. J Orthop Res 21: 826–835.

Chu PJ, Lee HM, Hou YT et al. 2008; Exten-sor-tendons reconstruction using autoge-nous palmaris longus tendon grafting forrheumatoid arthritis patients. J OrthopSurg 3: 16.

Clark RT, Johnson TL, Schalet BJ et al. 2001;GDF-5 deficiency in mice leads to disrup-tion of tail tendon form and function.Connect Tissue Res 42: 175–186.

Costa MA, Wu C, Pham BV et al. 2006; Tissueengineering of flexor tendons: optimiza-tion of tenocyte proliferation usinggrowth factor supplementation. TissueEng 12: 1937–1943.

Crossett LS, Sinha RK, Sechriest VF et al.2002; Reconstruction of a ruptured patel-lar tendon with Achilles tendon allograftfollowing total knee arthroplasty. J BoneJoint Surg Am 84A: 1354–1361.

Derwin KA, Baker AR, Spragg RK et al. 2006;Commercial extracellular matrix scaffoldsfor rotator cuff tendon repair – biomechan-ical, biochemical, and cellular properties. JBone Joint Surg Am 88A: 2665–2672.

Farng E, Urdaneta AR, Barba D et al. 2008;The effects of GDF-5 and uniaxial strainon mesenchymal stem cells in 3D culture.Clin Orthop Rel Res 466: 1930–1937.

Francis-West PH, Abdelfattah A, Chen P et al.1999; Mechanisms of GDF-5 action duringskeletal development. Development 126:1305–1315.

Harris MT, Butler DL, Boivin GP et al. 2004;Mesenchymal stem cells used for rabbittendon repair can form ectopicbone and express alkaline phosphataseactivity in constructs. J Orthop Res 22:998–1003.

Hayashi M, Zhao C, An KN et al. 2011; Theeffects of growth and differentiation factor5 on bone marrow stromal cell transplantsin an in vitro tendon healing model. JHand Surg Eur 36E: 271–279.

Heisterbach PE, Todorov A, Fluckiger R et al.2012; Effect of BMP-12, TGFβ1 and autolo-gous conditioned serum on growth factorexpression in Achilles tendon healing.

Knee Surg Sports Traumatol Arthrosc 20:1903–1910.

Hoffmann A, Pelled G, Turgeman G et al.2006; Neotendon formation induced bymanipulation of the Smad8 signallingpathway in mesenchymal stem cells. J ClinInvest 116: 940–952.

Hogan M, Girish K, James R et al. 2011;Growth differentiation factor-5 regulationof extracellular matrix gene expression inmurine tendon fibroblasts. J Tissue EngRegen Med 5: 191–200.

Hohendorff B, Siepen W, Spiering L et al.2008; Long-term results after operativelytreated Achilles tendon rupture: fibringlue versus suture. J Foot Ankle Surg 47:392–399.

Huang YH, Chin CC, Ho HN et al. 2009;Pluripotency of mouse spermatogonial stemcells maintained by IGF-1-dependent path-way. FASEB J 23: 2076–2087.

Ikeda J, Sun YL, An KN et al. 2011; Applica-tion of carbodiimide derivatized synovialfluid to enhance extrasynovial tendongliding ability. J Hand Surg Am 36A:456–463.

James R, Kesturu G, Balian G et al. 2008;Tendon: biology, biomechanics, repair,growth factors, and evolving treatment op-tions. J Hand Surg 33: 102–112.

James R, Kumbar SG, Laurencin CT et al.2011a; Tendon tissue engineering:adipose-derived stem cell and GDF-5 me-diated regeneration using electrospunmatrix systems. Biomed Mater 6: 025011.

James R, Kumbar SG, Laurencin CT et al.2011b; Tendon tissue engineering:adipose-derived stem cell and GDF-5mediated regeneration using electrospunmatrix systems. Biomed M Q7ater 6: •••–•••.

Juncosa-Melvin N, Boivin GP, Gooch C et al.2006; The effect of autologous mesenchy-mal stem cells on the biomechanics andhistology of gel–collagen sponge con-structs used for rabbit patellar tendon re-pair. Tissue Eng 12: 369–379.

Kapacee Z, Yeung CYC, Lu YH et al. 2010;Synthesis of embryonic tendon-like tissueby human marrow stromal/mesenchymalstem cells requires a three-dimensionalenvironment and transforming growthfactor-β3. Matrix Biol 29: 668–677.



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

Kardestuncer T, McCarthy M, KarageorgiouV et al. 2006; RGD-tethered silk substratestimulates the differentiation of humantendon cells. Clin Orthop Relat Res 448:234–239.

Kashiwagi K, Mochizuki Y, Yasunaga Y et al.2004; Effects of transforming growthfactor-β1 on the early stages of healing ofthe achilles tendon in a rat model. ScandJ Plastic Reconstruct Surg Hand Surg 38:193–197.

Kato YP, Dunn MG, Zawadsky JP et al. 1991;Regeneration of Achilles tendon with acollagen tendon prosthesis. Results of aone-year implantation study. J Bone JointSurg Am 73: 561–574.

Kawamura K, Chu CR, Sobajima S et al.2005; Adenoviral-mediated transfer ofTGF-β1 but not IGF-1 induceschondrogenic differentiation of humanmesenchymal stem cells in pellet cultures.Exp Hematol 33: 865–872.

Keller TC, Hogan MV, Kesturu G et al. 2011;Growth/differentiation factor-5 modulatesthe synthesis and expression of extracellu-lar matrix and cell-adhesion-related mole-cules of rat Achilles tendon fibroblasts.Connect Tissue Res 52: 353–364.

Kew SJ, Gwynne JH, Enea D et al. 2011;Regeneration and repair of tendon and lig-ament tissue using collagen fibre biomate-rials. Acta Biomater 7: 3237–3247.

Kishore V, Paderi JE, Akkus A et al. 2011;Incorporation of a decorin biomimetic en-hances the mechanical properties of elec-trochemically aligned collagen threads.Acta Biomater 7: 2428–2436.

Klein MB, Yalamanchi N, Pham H et al.2002a; Flexor tendon healing in vitro: ef-fects of TGF-β on tendon cell collagen pro-duction. J Hand Surg Am 27A: 615–620.

Klein MB, Yalamanchi N, Pham H et al.2002b; Flexor tendon healing in vitro: ef-fects of TGF-β on tendon cell collagen pro-duction. J Hand Surg 27: 615–620.

Krampera M, Pizzolo G, Aprili G et al. 2006;Mesenchymal stem cells for bone, carti-lage, tendon and skeletal muscle repair.Bone 39: 678–683.

Krych AJ, Jackson JD, Hoskin TL et al. 2008;A meta-analysis of patellar tendon auto-graft versus patellar tendon allograft inanterior cruciate ligament reconstruction.Arthroscopy 24: 292–298.

Kryger GS, Chong AKS, Costa M et al. 2007;A comparison of tenocytes and mesenchy-mal stem cells for use in flexor tendontissue engineering. J Hand Surg 32:597–605.

Kurtz CA, Loebig TG, Anderson DD et al.1999; Insulin-like growth factor I acceler-ates functional recovery from Achilles ten-don injury in a rat model. Am J Sports Med27: 363–369.

Laumonier T, Michel M, Gabbiani G et al.2012; Autologous transplantation ofculture-born myofibroblasts into intactand injured rabbit ligaments. Int Orthop36: 1733–1738.

Longo UG, Lamberti A, Maffulli N et al. 2010;Tendon augmentation grafts: a systematicreview. Br Med Bull 94: 165–188.

Lui PPY, Rui YF, Ni M et al. 2011; Tenogenicdifferentiation of stem cells for tendon re-pair – what is the current evidence? J Tis-sue Eng Regen Med 5: e144–163.

Mendias CL, Gumucio JP, Lynch EB. 2012;Mechanical loading and TGF-β change

the expression of multiple miRNAs in ten-don fibroblasts. J Appl Physiol 113: 56–62.

Mitchell JJ, Woodcockmitchell JL, Perry Let al. 1993; In vitro expression of the α-smooth muscle actin isoform by rat lungmesenchymal cells – regulation by culturecondition and transforming growthfactor-β. Am J Resp Cell Mol Biol 9: 10–18.

Molloy T, Wang Y, Murrell GAC. 2003; Theroles of growth factors in tendon and liga-ment healing. Sports Med 33: 381–394.

Musarò A, Giacinti C, Borsellino G et al.2004; Stem cell-mediated muscle regener-ation is enhanced by local isoform ofinsulin-like growth factor 1. Proc Natl AcadSci U S A 101: 1206–1210.

Nakamura S, Katsuki M. 2002; Tendongrafting for multiple extensor tendon rup-tures of fingers in rheumatoid hands. JHand Surg Br 27: 326–328.

Okamoto N, Kushida T, Oe K et al. 2010;Treating Achilles tendon rupture in ratswith bone marrow cell transplantationtherapy. J Bone Joint Surg Am 92A:2776–2784.

Park A, Hogan MV, Kesturu GS et al. 2010;Adipose-derived mesenchymal stem cellstreated with growth differentiation factor-5 express tendon-specific markers. TissueEng A 16: 2941–2951.

Pryce BA, Watson SS, Murchison ND et al.2009; Recruitment and maintenance oftendon progenitors by TGFβ signaling areessential for tendon formation. Develop-ment 136: 1351–1361.

Qiu Y, Wang X, Zhang Y et al. 2012; Develop-ment of a refined tenocyte expansionculture technique for tendon tissue engineer-ing. J Tissue Eng Regen Med ••: •••–•••.Q8Q9

Raghavan SS, Woon CYL, Kraus A et al.2012; Optimization of human tendon tis-sue engineering: synergistic effects ofgrowth factors for use in tendon scaffoldrepopulation. Plastic Reconstruct Surg129: 479–489.

Richardson LE, Dudhia J, Clegg PD et al.2007; Stem cells in veterinary medicine– attempts at regenerating equinetendon after injury. Trends Biotechnol25: 409–416.

Rincon M, Muzumdar R, Atzmon G et al.2004; The paradox of the insulin/IGF-1signaling pathway in longevity. Mech Age-ing Dev 125: 397–403.

Rose LC, Fitzsimmons R, Lee P et al. 2012; Ef-fect of basic fibroblast growth factor inmouse embryonic stem cell culture and os-teogenic differentiation. J Tissue Eng RegenMedicine ••: •••–•••.Q10

Rui YF, Lui PPY, Chan LS, et al. 2011; Doeserroneous differentiation of tendon-derived stem cells contribute to the patho-genesis of calcifying tendinopathy? ChinMed J 124: 606–610.

Rui YF, Lui PPY, Li G et al. 2010; Isolationand characterization of multipotent rattendon-derived stem cells. Tissue Eng A16: 1549–1558.

Sahoo S, Toh SL, Goh JCH. 2010; A bFGF-releasing silk/PLGA-based biohybrid scaf-fold for ligament/tendon tissue engineer-ing using mesenchymal progenitor cells.Biomaterials 31: 2990–2998.

Schnabel LV, Lynch ME, van der Meulen et al.2009; Mesenchymal stem cells and insulin-like growth factor-I gene-enhanced mesen-chymal stem cells improve structural as-pects of healing in equine flexor digitorum

superficialis tendons.0. J Orthop Res 27:1392–1398.Q11

Schneider PRA, Buhrmann C, Mobasheri Aet al. 2011; Three-dimensional high-density co-culture with primary tenocytesinduces tenogenic differentiation inmesenchymal stem cells. J Orthop Res 29:1351–1360.

Shearn JT, Kinneberg KRC, Dyment NA et al.2011; Tendon tissue engineering: prog-ress, challenges, and translation to theclinic. J Musculoskel Neuron Interact11: 163–173.

Stone CA. 2000; Unravelling the secrets offoetal wound healing: an insight into frac-ture repair in the mouse foetus and per-spectives for clinical application. Br JPlastic Surg 53: 337–341.

Stone K, Walgenbach A, Turek T et al. 2007;Anterior cruciate ligament reconstructionwith a porcine xenograft: a serologic,histologic, and biomechanical study inprimates. Arthroscopy 23: 411–419.

Sutter WW. 2007; Autologous cell-basedtherapy for tendon and ligament injuries.Clin Tech Equine Pract 6: 198–208.

Takahashi S, Nakajima M, Kobayashi M et al.2002; Effect of recombinant basic fibro-blast growth factor (bFGF) on fibroblast-like cells from human rotator cuff tendon.Tohoku J Exp Med 198: 207–214.

Tan Q, Lui PPY, Rui YF. 2012a; Effect ofin vitro passaging on the stem cell-relatedproperties of tendon-derived stem cells –

implications in tissue engineering. StemCells Dev 21: 790–800.

Tan SL, Ahmad RE, Ahmad TS et al. 2012b;Effect of growth differentiation factor 5on the proliferation and tenogenicdifferentiation potential of human mesen-chymal stem cells in vitro. Cells TissuesOrgans 196: 325–338.

Tang JB, Xu Y, Ding F et al. 2003; Tendonhealing in vitro: promotion of collagengene expression by bFGF with NF-κB geneactivation. J Hand Surg 28: 215–220.

Tateno C, Yoshizato K. 1999; Growth poten-tial and differentiation capacity of adultrat hepatocytes in vitro. Wound RepairRegen 7: 36–44.

Thomopoulos S, Das R, Sakiyama-Elbert Set al. 2010; bFGF and PDGF-BB for tendonrepair: controlled release and biologic ac-tivity by tendon fibroblasts in vitro. AnnBiomed Eng 38: 225–234.

Thomopoulos S, Harwood FL, Silva MJ et al.2005; Effect of several growth factors oncanine flexor tendon fibroblast prolifera-tion and collagen synthesis in vitro. J HandSurg 30: 441–447.

Trosan P, Svobodova E, Chudickova M et al.2012; The key role of insulin-like growthfactor I (IGF-I) in limbal stem cell differen-tiation and the corneal wound healing pro-cess. Stem Cells Dev 21: 3341–3350.

Uysal AC, Mizuno H. 2010; Tendon regenera-tion and repair with adipose derived stemcells. Curr Stem Cell Res Ther 5: 161–167.

Uysal AC, Mizuno H. 2011; Differentiationof adipose-derived stem cells for tendonrepair. In •••••••••••, Gimble JM,Bunnell BA (eds). Humana: •••••;443–451. Q12

Wang QW, Chen ZL, Piao YJ. 2005; Mesen-chymal stem cells differentiate intotenocytes by bone morphogenetic protein(BMP) 12 gene transfer. J Biosci Bioeng100: 418–422.



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

Wolfman NM, Hattersley G, Cox K et al. 1997;Ectopic induction of tendon and ligament inrats by growth and differentiation factors 5,6, and 7, members of the TGF-β gene family.J Clin Invest 100: 321–330.

Xu YH, Murrell GAC. 2008; The basic scienceof tendinopathy. Clin Orthop Rel Res 466:1528–1538.

Yin Z, Chen X, Chen JL, OuyangHW. 2010; Stemcells for tendon tissue engineering and regen-eration. Expert Opin Biol Ther 10: 689–700.

Young RG, Butler DL,WeberW et al. 1998; Useof mesenchymal stem cells in a collagen

matrix for Achilles tendon repair. J OrthopRes 16: 406–413.

Zeugolis DI, Chan JCY, Pandit A. 2011a;Tendons: Engineering of functionaltissues. In Tissue Engineering, Pallua N,Suscheck CV (eds). Springer: Berlin,Heidelberg; 537–572.

Zeugolis DI, Keeney M, Collin E et al. 2011b;Xenogenic tissues and biomaterials for theskeletal system. In Comprehensive Bioma-terials, Ducheyne P, Healy KE, HutmacherDW et al. (eds). Elsevier: Amsterdam;387–404.

Zeugolis DI, Paul GR, Attenburrow G.2009; Cross-linking of extruded collagenfibres – a biomimetic three-dimensionalscaffold for tissue engineering applications.J Biomed Mater Res A 89: 895–908.

Zhang J, Li B, Wang JHC. 2011; The role ofengineered tendon matrix in the stemnessof tendon stem cells in vitro and the pro-motion of tendon-like tissue formationin vivo. Biomaterials 32: 6972–6981.

Zhang J, Wang JH. 2013; Human tendon stemcells better maintain their stemness in hyp-oxic culture conditions. PLoS One 8: e61424.

Supporting information on the internet

Additional supporting information may be found in the online version of this article at the publisher’s web-site:

Figure S1. Images of extracted tendon stem cells, showing CD90- and TEN-C-expressing cells (a), CD44-expressingcells without collagen expression (b), DCN-expressing cells with negligible CD31 expression (c) and scleraxis-express-ing cells with negligible CD34 expression (d)Figure S2. Chondrogenic pellets prepared from (a) untreated TSCs at each time point; (b) TSCs treated with 1, 10 or 100ng/ml IGF-1; (c) TSCs treated with 1, 10 or 100 ng/ml GDF-5; (d) TSCs treatedwith 1,10 or 100 ng/ml TGF-β1, stained forcollagen type II (green) and counter-stained with DAPI (blue)Figure S3. FABP4 expression as an indicator of adipogenesis in TSCs at each time point



12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364

6566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128

Author Query Form

Journal: Journal of Tissue Engineering and Regenerative Medicine

Article: term_1852

Dear Author,

During the copyediting of your paper, the following queries arose. Please respond to these by annotating your proofs withthe necessary changes/additions.• If you intend to annotate your proof electronically, please refer to the E-annotation guidelines.• If you intend to annotate your proof by means of hard-copy mark-up, please refer to the proof mark-up symbols guide-lines. If manually writing corrections on your proof and returning it by fax, do not write too close to the edge of thepaper. Please remember that illegible mark-ups may delay publication.

Whether you opt for hard-copy or electronic annotation of your proofs, we recommend that you provide additional clar-ification of answers to queries by entering your answers on the query sheet, in addition to the text mark-up.

Query No. Query Remark

Q1 AUTHOR: Supplementary figs are not numbered in the supporting information.I’ve assumed they are placed in numerical order

Q2 AUTHOR: The citation “Zeugolis et al., 2011” (original) has been changed to“Zeugolis et al., 2011a, 2011b throughout the text”. Please check if appropriate.

Q3 AUTHOR: The citation “James et al., 2011” (original) has been changed to “Jameset al., 2011a, 2011b throughout the text”. Please check if appropriate.

Q4 AUTHOR: The citation “Tan et al., 2012” (original) has been changed to “Tanet al., 2012a, 2012b throughout the text”. Please check if appropriate.

Q5 AUTHOR: The citation “Klein et al., 2002” (original) has been changed to “Kleinet al., 2002a, 2002b Throughout the text”. Please check if appropriate.

Q6 AUTHOR: MS contains message Error! Reference source not found. Is there amissing ref here?

Q7 AUTHOR: Insert page range of ref. James R, et al. 2011.

Q8 AUTHOR: Reference Qiu Y, et al. 2012 has a duplicate in reference list, to avoidduplication one of them has been deleted in the reference, please check if this iscorrect and appropriate.

Q9 AUTHOR: Insert volume and page numbers of ref. Qiu Y, et al. 2012.

Q10 AUTHOR: Insert volume and page numbers of ref. Rose LC, et al. 2012.

Q11 AUTHOR: Please supply given name to van der Meulen, of ref. Schnabel LV,et al. 2009.

Q12 AUTHOR: Insert book title (in italics) and CITY of publisher of ref. Uysal AC,2011.

Q13 AUTHOR: Figure 2 contains small and poor quality text. Please check andresupply if necessary.

Q14 AUTHOR: Please check Conflict of Interest section if captured correctly

USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION

Required software to e-Annotate PDFs: Adobe Acrobat Professional or Adobe Reader (version 7.0 or above). (Note that this document uses screenshots from Adobe Reader X) The latest version of Acrobat Reader can be downloaded for free at: http://get.adobe.com/uk/reader/

Once you have Acrobat Reader open on your computer, click on the Comment tab at the right of the toolbar:

1. Replace (Ins) Tool – for replacing text.

Strikes a line through text and opens up a text box where replacement text can be entered.

How to use it

Highlight a word or sentence.

Click on the Replace (Ins) icon in the Annotations section.

Type the replacement text into the blue box that appears.

This will open up a panel down the right side of the document. The majority of tools you will use for annotating your proof will be in the Annotations section, pictured opposite. We’ve picked out some of these tools below:

2. Strikethrough (Del) Tool – for deleting text.

Strikes a red line through text that is to be deleted.

How to use it

Highlight a word or sentence.

Click on the Strikethrough (Del) icon in the Annotations section.

3. Add note to text Tool – for highlighting a section to be changed to bold or italic.

Highlights text in yellow and opens up a text box where comments can be entered.

How to use it

Highlight the relevant section of text.

Click on the Add note to text icon in the Annotations section.

Type instruction on what should be changed regarding the text into the yellow box that appears.

4. Add sticky note Tool – for making notes at specific points in the text.

Marks a point in the proof where a comment needs to be highlighted.

How to use it

Click on the Add sticky note icon in the Annotations section.

Click at the point in the proof where the comment should be inserted.

Type the comment into the yellow box that appears.

USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION

For further information on how to annotate proofs, click on the Help menu to reveal a list of further options:

5. Attach File Tool – for inserting large amounts of text or replacement figures.

Inserts an icon linking to the attached file in the appropriate pace in the text.

How to use it

Click on the Attach File icon in the Annotations section.

Click on the proof to where you’d like the attached file to be linked.

Select the file to be attached from your computer or network.

Select the colour and type of icon that will appear in the proof. Click OK.

6. Add stamp Tool – for approving a proof if no corrections are required.

Inserts a selected stamp onto an appropriate place in the proof.

How to use it

Click on the Add stamp icon in the Annotations section.

Select the stamp you want to use. (The Approved stamp is usually available directly in the menu that appears).

Click on the proof where you’d like the stamp to appear. (Where a proof is to be approved as it is, this would normally be on the first page).

7. Drawing Markups Tools – for drawing shapes, lines and freeform annotations on proofs and commenting on these marks.

Allows shapes, lines and freeform annotations to be drawn on proofs and for comment to be made on these marks..

How to use it

Click on one of the shapes in the Drawing Markups section.

Click on the proof at the relevant point and draw the selected shape with the cursor.

To add a comment to the drawn shape, move the cursor over the shape until an arrowhead appears.

Double click on the shape and type any text in the red box that appears.

Downloaded 2021-09-11T20:12:34Z Some rights reserved. For ...

Documents

Transcript of Downloaded 2021-09-11T20:12:34Z Some rights reserved. For ...