The Pathogenesis of Tendon Microdamage in Athletes: the ... · The equine superﬁcial digital...

J. Comp. Path. 2012, Vol. 147, 227e247 Available online at www.sciencedirect.com

www.elsevier.com/locate/jcpa

SPONTANEOUSLY ARISING DISEASE: REVIEW ARTICLE

The Pathogenesis of Tendon Microdamage inAthletes: the Horse as a Natural Model for Basic

Cellular Research

Cor

002

doi

J. C. Patterson-Kane*, D. L. Becker† and T. Rich*

*Institute of Infection, Immunity and Inflammation, College of Medical, Veterinary and Life Sciences, University of Glasgow,

Bearsden Road, Glasgow G61 1QH and †Department of Cell and Developmental Biology, University College London, Gower

Street, London WC1E 6BT, UK

resp

1-99

:10.1

Summary

The equine superficial digital flexor tendon (SDFT) is a frequently injured structure that is functionally andclinically equivalent to the human Achilles tendon (AT). Both act as critical energy-storage systems duringhigh-speed locomotion and can accumulate exercise- and age-related microdamage that predisposes to ruptureduring normal activity. Significant advances in understanding of the biology and pathology of exercise-induced tendon injury have occurred through comparative studies of equine digital tendons with varying func-tions and injury susceptibilities. Due to the limitations of in-vivo work, determination of the mechanisms bywhich tendon cells contribute to and/or actively participate in the pathogenesis of microdamage requires de-tailed cell culture modelling. The phenotypes induced must ultimately be mapped back to the tendon tissueenvironment. The biology of tendon cells and their matrix, and the pathological changes occurring in the con-text of early injury in both horses and people are reviewed, with a particular focus on the use of various tendoncell and tissue culture systems to model these events.

� 2012 Elsevier Ltd. Open access under CC BY license.

Keywords: athlete; equine; pathology; tendon

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

The Importance of Injuries to Energy-Storing Tendons in Large Athletic Animals . . . . . . . . . . . . . . . . . . . . . . . . . 228

The Human Element: Tendon Equivalence in Function and Injury Susceptibility . . . . . . . . . . . . . . . . . . . . 228

Tendon Cell Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Tenocyte Networks within the Hierarchical Structure of Tendons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Modelling Tenocyte Networks In Vitro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

Tendon Cell Culture Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

Stem Cells/Progenitor Cells in Tendon Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

Tenocyte Mechanotransduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234

Coordinated Signal Transmission: the Gap Junction Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

Connexin Isoforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

Comparative Biology of Injury-Prone and Non-Injury-Prone Tendons in Horses . . . . . . . . . . . . . . . . . . . . . . . . . . 236

Functional Determination of Matrix Composition: Energy-Storing Versus Positional Tendons . . . . . . . . . . . . 236

Tendon-Specific Differences in Tenocyte Populations of Immature and Adult Horses: Cellular ‘Switch-Off’ . . 237

The Pathogenesis of Exercise-Induced Tendon Microdamage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

Evidence of Matrical and Cellular Microdamage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

ondence to: J. C. Patterson-Kane (e-mail: [email protected]).

75 � 2012 Elsevier Ltd.

016/j.jcpa.2012.05.010

Open access under CC BY license.

mailto:[email protected]

http://dx.doi.org/10.1016/j.jcpa.2012.05.010



www.sciencedirect.com/science/journal/00219975

http://www.elsevier.com/locate/jcpa

http://creativecommons.org/licenses/by/3.0/

http://creativecommons.org/licenses/by/3.0/

228 J.C. Patterson-Kane et al.

The ‘Tendonosis Cycle’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

Studying the Effects of Cell Stress and Damage in Cell Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

Contribution of Tenocyte Death to Microdamage: Primary or Secondary? . . . . . . . . . . . . . . . . . . . . . . . . . . 240

Is the Initial Lesion Inflammatory or Degenerative? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Introduction

Tendon injuries associated with athletic activity andageing are most frequent and important in horsesand people. Healing is prolonged, with failure to re-constitute normal structure contributing to high in-jury recurrence rates in both species (Crevier-Denoixet al., 1997; Ely et al., 2009; Molloy and Wood,2009). Failure of regeneration suggests that preven-tion would have a significantly greater clinical impactthan therapy. It is generally agreed that tendons sus-tain cumulative subclinical damage (i.e. microdam-age) for undefined periods of time prior to rupture,implying that there is a ‘windowof opportunity’ for in-tervention. Although our understanding is increasing,knowledge is still limited in terms of: exactly how andwhy certain tendons accumulate cellular and matrixmicrodamage; the effects of microenvironments expe-rienced by tendon cells during and after exercise ontheir viability and function including injury detectionand repair capacity; andmany aspects of fundamentaltendon cell and matrix biology, particularly in com-parison with other musculoskeletal tissues. Due to itsclinical importance and an extent of access to tissuespecimens not possible in the medical field, the mostcompletely understood tendons are those of the distallimb of the horse, in particular the injury-prone super-ficial digital flexor tendon (SDFT). Citations ofequine literature are frequent in descriptions of hu-man tendon biology and pathology (Magnussonet al., 2002; Riley, 2005; Lui et al., 2011). The special-ized nature of equine digital tendons and certain hu-man tendons essential for efficiency of high-speedlocomotion, and the large size of both species, bringthe clinical relevance of laboratory rodent and rabbitmodels of exercise-induced tendon damage into ques-tion. In addition to concerns with the principles of re-placement, reduction and refinement (the ‘three Rs’),in-vivo investigations of any tendon in any species willnot currently allow direct analysis at the microscopi-cal, ultrastructural and molecular levels required forstudy of microdamage. There has therefore been in-creasing effort to develop relevant equine cell cultureand tissue explant models. This review will cover cur-rent knowledge of equine digital tendon biology andpathology with a focus on the tenocyte (tendon fibro-

blast), discussion of the advantages and problems en-countered with in-vitro models used to study tendoncell responses to potential pathogenic factors, and in-dications of the clinical relevance of such work.

The Importance of Injuries to Energy-StoringTendons in Large Athletic Animals

Injury to the SDFT,most often involving themetacar-pal segment of the forelimb tendon, is one of the mostfrequent causes of lameness of thoroughbred horses in-ternationally, occurring in 0.6e16.1/1,000 race starts,with a reported prevalence of 11e30% (Goodshipet al., 1994; Williams et al., 2001; Kasashima et al.,2004; Pinchbeck et al., 2004; Ely et al., 2009). In a re-cent survey of National Hunt racehorses in the UK,the incidence of SDFT injury was 0.95/100 horsemonths (1 month equalling 30 days) during trainingand 42.5/100 horse months for those in racing (Elyet al., 2009). The risk of SDFT injury has been shownto increase with horse age in a number of surveys(Williams et al., 2001; Ely et al., 2004; Kasashimaet al., 2004; Perkins et al., 2005a). In one Japanesestudy, approximately 70% of 1,200 racehorses thatsustained tendon injuries (largely to the SDFT) failedto return to previous levels of performance in any race(Oikawa and Kasashima, 2002). The results of a UKstudy of performance of 401 horses sustaining anSDFT injury weremore equivocal, with no significantpost-injury reduction in maximum Racing PostRating (O’Meara et al., 2010). The absence of aninternationally agreed protocol for reporting ofinjuries and race performance makes it difficult tocompare these datasets. Re-injury rates of 23e67%have been reported in various equine sports disci-plines; 19e70% are ultimately retired due to the orig-inal or the subsequent re-injury (Marr et al., 1993;Dyson, 2004; Perkins et al., 2005b; Lam et al., 2007;O’Meara et al., 2010). All of these factors cause signif-icant economic loss and animal welfare concerns.

The Human Element: Tendon Equivalence in Function and

Injury Susceptibility

In the functionally-equivalent human Achilles ten-don (AT), increased sports participation in the

Tendon Microdamage in the Horse 229

Western world in particular is generating a wave ofexercise-related injuries. Of total AT injuries approx-imately 65e75% are sports related; the variety ofsports types involved differs between countries. Thereare also unexplained increases in older non-athleticpeople (M€oller et al., 1996). In Finland between1987 and 1999 there was an 80% increase in incidencein people in their 3rd to 4th decade (Nyyss€onen et al.,2008). In Scotland a 90% increase occurred between1995 and 2008, with AT injuries comprising>10% oftotal soft tissue injuries (Maffulli et al., 1999; Claytonand Court-Brown, 2008). In 1986 there was a suddenincrease in the incidence of AT rupture in men inScotland during a year when the CommonwealthGames were held in Edinburgh (Maffulli et al.,1999). The risk of AT injury is associated with eliteathletic activity and with age, as for SDFT injury inhorses; incidences of up to 29%, 18% and 52%have been reported in elite male runners, elite gym-nasts and running athletes, respectively, with the lat-ter figure correlated with hours spent training(Paavola et al., 2002; Knobloch et al., 2008;Emerson et al., 2010). In one study of 785 male formerelite athletes, their cumulative incidence of AT degen-eration or rupture before the age of 45 was five toeight times that of the control population; however,for AT rupture it was up to 15 times that of controlsin sprinters specifically (Kujala et al., 2005). Up to29% of patients with AT injuries may require surgery(Zafar et al., 2009); time for a return to activity variesfrom 6 weeks to almost 10 months depending on thecomplexity and chronicity of the injury, and re-rupture rates of 2e12% have been reported(Molloy and Wood, 2009; Saxena et al., 2011).

The functional equivalence of the SDFT and AT islargely due to their pivotal role in saving energy dur-ing high-speed locomotion, by reducing muscularwork (Biewener, 1998; Malvankar and Khan,2011). Most tendons (‘positional’ tendons) simplyconnect skeletal muscle and bone, stabilizing andmoving joints. In the horse, recovery of mechanicalwork during tendon elastic recoil (i.e. with theSDFT acting as a ‘biological spring’) has been esti-mated at 36% at a gallop; in the human AT, energyreturns contribute to over 50% of positive workdone at the ankle (Biewener, 1998; Farris et al.,2011). Energy-storing tendons must undergo highstrains (i.e. changes in tendon length as a percentageof initial length) to store sufficient amounts of energy.The SDFT and AT have conflicting requirements ofelasticity to maximize energy storage and strengthto support weight bearing. Comparison of SDFTstrains in vitro with those in galloping horses indicatedvery low mechanical safety margins, considered to betypical for such structures (Stephens et al., 1989;

Wilson and Goodship, 1991). It is generally agreedthat the levels and most likely also the patterns, typesand durations of physical stress on tendons are themost important aetiological factors in athletic injuries(Riley, 2005). In additional to the mechanical envi-ronment, the only other directly measured or calcu-lated stress factor specific to energy-storing tendonsis hyperthermia. Due to loss of some strain energy asheat (hysteresis), which is not effectively dispersedby the vascular system, temperatures of at least43e45�C have been measured in the SDFT core ofthe galloping horse (Wilson and Goodship, 1994).Conservative estimates of hyperthermia in the humanAT, calculated by mathematical modelling, pre-dicted a core temperature of at least 41�C during30 min of treadmill running, with the potential torise to up to 44�C in some individuals (Farris et al.,2011). Such temperatures would be expected to be le-thal or at the very least highly stressful to fibroblasticcells. Other damaging conditions that may be experi-enced during and/or following exercise within the ten-don tissue include hypoxia due to tendonhypoperfusion and increased levels of reactive oxygenspecies (ROS) induced by ischaemia-reperfusion and/or hyperthermia (Paavola et al., 2002; Bestwick andMaffulli, 2004; Millar et al., 2012).

Tendon Cell Biology

In comparison with other musculoskeletal tissues, thestructure and cell biology of tendon tissue is stillpoorly understood, but has been the subject of in-creasing research effort over the past few decades.Tendon structure is highly specialized due to the uni-directional forces to which these structures are sub-jected, and has a complex hierarchical arrangement.

Tenocyte Networks within the Hierarchical Structure of

Tendons

Tenocytes are the fibroblastic cells responsible for syn-thesis, maintenance and degradation of the extracel-lular matrix (ECM) of tendons. Their syntheticproducts include collagen, proteoglycans and glyco-proteins, and degradative enzymes including matrixmetalloproteinases (MMPs) and their regulators(i.e. tissue inhibitors ofMMPs; TIMPs). Despite theirpivotal role, numbers of tenocytes in tendon tissue arelow, with fewer than 300/mm2 in the adult SDFT(Stanley et al., 2007). Tenocytes are classified astype 1, 2 or 3 with spindle-shaped nuclei, cigar-shaped nuclei or a chondroid appearance, respec-tively (Smith and Webbon, 1996; Fig. 1). Type 1and 2 tenocytes have been equated to ‘tenocytes’and ‘tenoblasts’ in non-equine species; tenoblasts


appear to be more synthetically active and their pro-portion reduces with age in the SDFT and other ten-dons (Chuen et al., 2004; JPK personal observation).Type 3 cells are found in fibrocartilaginous areas oftendons where they ‘wrap around’ joints and experi-ence compressive as well as tensile forces. Tenocytesare arranged in widely-spaced rows within the matrixthat are bridged by their narrow cytoplasmic pro-cesses; these processes and the cell bodies (withinrows) are connected by both gap junctions (GJs)and adherens junctions (AJs). Tenocytes are con-nected to the ECM by adhesive junctions of whichthe major transmembrane components are integrins(i.e. heterodimeric receptors that physically link theECM to the actin cytoskeleton).

The (possibly sheet-like) cytoplasmic processes oftenocytes extend around collagen fibrils, the cylindri-cal units of tensile strength in tendon that are ar-ranged in parallel in the longitudinal axis (McNeillyet al., 1996; Ralphs et al., 2002); collagen comprises75e80% of tendon dry weight and in the adult95% of it is type I. Type III collagen is associatedwith smaller collagen fibril diameters (i.e. weaker fi-brils; Birk and Mayne, 1997) and is found in greateramounts in immature and damaged tendons(Williams et al., 1984). Low levels of non-fibrillar col-lagen types have also been identified. The collagen fi-brils follow a planar, zig-zag waveform termed the‘crimp’, which is thought to act as a mechanical safetybuffer (i.e. this waveform opens out under tension be-fore the fibrils themselves are stretched;Wilmink et al.,1991). A mixture of large- and small-diameter fibrilsin the SDFT is thought to provide the tendon withstrength and elasticity, respectively (Patterson-Kaneet al., 1997b; Fig. 2). The remainder of the ECM iscomprised of proteoglycans, glycoproteins, smallnumbers of elastic fibres, ions and water. Tenocytesand their ECM form subunits called fascicles, whichmeasure approximately 1 mm in diameter, are polyg-onal in cross-section and are surrounded by smallamounts of loose connective tissue termed the endote-non (Fig. 3). The fascicles appear to be able to moveindependently as the tendon is cyclically loaded, withlateral force transmission between them being smallor negligible (Goodship et al., 1994; Haraldssonet al., 2008). The endotenon carries blood vessels,which do not penetrate the fascicular substance, in

Fig. 1. Examples of the three morphological tenocyte typesreported in equine digital tendons: (A) type 1 cellswith long, spindle-shaped nuclei; (B) type 2 cells withplump, cigar-shaped nuclei; and (C) chondrocyte-liketype 3 cells, typically found in ‘wrap around’ regionsof tendons where the matrix is fibrocartilaginous. Hae-matoxylin and eosin. �400.

Fig. 2. Transmission electron micrograph of a SDFT froma thoroughbred horse, showing the collagen fibrils incross-section. The large and small diameter fibrils arethought to provide the tendon tissue with strengthand elasticity, respectively. Uranyl acetate and modi-fied lead citrate. �45,100.


addition to lymphatic vessels, nerves and fibroblasticcells; transforming growth factor (TGF)-b is onlyfound in the endotenon in the adult SDFT (Cauvinet al., 1998). Fibroblastic cells are also located in theepitenon and paratenon (i.e. connective tissue layerssurrounding the whole tendon), but the nature ofthese cell populations has not been explored in detail.

Fig. 3. Cross-section of an equine SDFT, showing polygonal fasci-cles that are surrounded by a thin sheath of looser connec-tive tissue termed the endotenon (arrows). Blood vessels areonly located within the endotenon in non-injured tendontissue. Cryosection, haematoxylin and eosin. �20.

Modelling Tenocyte Networks In Vitro

There is a need to use in-vitro experimental systemsfor studies of tenocyte responses to conditions occur-ring during exercise. The advantages of such systemsare that they are often easy to use, tractable, amena-ble to biochemical analyses and time-dependent mea-surements can bemade at both cellular andmolecularlevels. Technical, animal welfare and cost issues pre-vent the latter in experimental animal trials, particu-larly if horses are to be used. It is, however, criticalthat the cellular read-out recapitulates phenotypiccharacteristics known for the donor tissue; resultsfrommany models only have relevance to that partic-ular in-vitro environment. Access to tissue specimensfrom horses submitted to abattoirs or post-mortem di-agnostic services, with owner consent, can signifi-cantly facilitate tendon research relative to themedical field. There can, however, be problems in ob-taining full histories for the animals, in particular ex-ercise history. In the ideal situation, the extent of pre-existing damage to cells and tissue from each donorshould be quantified prior to any manipulation or ap-plied stress.

Tendon Cell Culture Systems. Two-dimensional (i.e.monolayer) culture is the easiest starting point forstudying tendon cells, however (1) tenocytes in tissueare not arranged in confluent or near-confluent sheets(as above); (2) the severity of methods used to extractthem from the tissue can potentially cause seriousproblems; (3) the process may select certain cellularsubpopulations; (4) the problem of phenotypic driftis well-documented; and (5) the relationship betweentenocytes and their matrix is critical and difficult toreplicate in culture. As tenocytes usually sit ina very small niche with intimate association withthe ECM, interference with these interactions by ei-ther enzymatic digestion or allowing outgrowthfrom a tissue fragment (explant) may create injury re-sponses that drive phenotypic drift. We have foundthat culture following enzymatic digestion of SDFTand deep digital flexor tendon (DDFT) specimenscan generate reasonable numbers of viable cellswithin a week, with slightly more rapid growth fromthe former tendon. Considerations for monolayer cul-ture in this respect include serum levels, glucose con-centration, antibiotic use and oxygen tension. Whileserum levels of 10% and higher are routinely used,it is clear that this is a non-physiological and poten-tially damaging mitogenic stimulus. Cell damagemay not be evident when using ‘binary’ live/dead as-says, even when considerable levels of genotoxic stressare being experienced and repaired. It is also impor-tant to avoid high-glucose media unless there is


a wish to model diabetic pathology, as the effects ofthis on tendon cells are not well defined. Antibioticsof the fluoroquinolone class have been linked withtendonopathy in man, and can triple the risk of ATrupture (Sode et al., 2007); the pharmacokinetic prop-erties of these drugs promote their concentration inconnective tissue. These effects extend to cell culture,with fluoroquinolone treatment of a spontaneouslyimmortalized rabbit tenocyte cell line resulting in cy-totoxicity and MMP activation by generated ROS(Bernard-Beaubois et al., 1998; Pouzaud et al.,2004). Enrofloxacin treatment of equine SDFT-derived cells caused altered synthesis of small proteo-glycans, changes in cell morphology and inhibition ofcellular proliferation; these effects were most pro-nounced when foal tendon cells were used (Yoonet al., 2004). With the use of ciprofloxacin in the me-dium, we have not noted any effects on equine tendoncell proliferation or DNA damage. Effects may beboth drug and concentration related and should betested in the specific cell culture system used if theiraddition is considered to be necessary, for exampleto removeMycoplasma spp. infection. Ambient oxygenlevels are not physiological for any cell type and areparticularly likely to be harmful in terms of ROS gen-eration for cells from poorly-vascular tissues such astendon. These levels of oxygen are considered to belargely responsible for the ‘culture shock’ experiencedby many rodent cell lines, which can lead to their pre-mature senescence (Sherr and DePinho, 2000). Re-searchers have reported accelerated proliferationand improved viability of neonatal tendon cellswhen using oxygen tensions of 2e5% (Zhang et al.,2010b). However, robust datasets for cells derivedfrom various tendons of adult animals under theseconditions are lacking.

For viability measurements, the colorimetric MTS(3-[4, 5-dimethylthiazol-2-yl]-5-[3-carboxymethoxyp-henyl]-2-[4-sulfophenyl]-2H-tetrazolium) assay formetabolic activity may be preferable to other methodsas reagent compatibility is not an issue. In general,working with equine tendon cells and tissues can beproblematic due to the need to determine reagent andantibody compatibility with this species. We havefound, for example that two commercial antibodies tocleaved caspase-3 documented as labelling equinetissue sections by other authors, did not convincinglylabel equine cells or histological sections. These anti-bodies were negative in duplicate western blots usingstaurosporine-treated equine fibroblasts; this occurreddespite a high sequence identity with species-matchedpre-proteins. Care should be taken in using antibodiesand reagents published as reactive for equine tissues orcells if formal validation is lacking. In terms of measur-ing apoptosis, researchers detecting DNA breaks using

the terminal transferase-mediated dUTP nick end la-belling (TUNEL) technique should also be aware ofthe caveats of that methodology. False-positive andfalse-negative results have been obtained, but combin-ing this technique withmorphological cell count analy-sis may improve data quality (Garrity et al., 2003).

Monolayer studies must also be analyzed with theunderstanding that the cells are selected on the basisof their ability to adhere to culture dishes followingextraction. The growth of cells from explants, whichinvolves a more lengthy initial recovery period, isalso selective in favouring cells that both migrateand replicate away from the injured tissue margins.Growth factors including insulin-like growth factor-1are often added to media to encourage growth, butthey are used at non-physiological levels and mayencourage outgrowth of poorly understood subpopu-lations (Costa et al., 2006). The success of growthfactors may in part be due to loss of inhibitory stim-uli usually provided by the ECM. Type I collagen,for example, can exert cell cycle inhibitory effectsvia integrin-linked regulation of mitogen-activatedprotein kinase activation (Sch€ocklmann et al.,2000). Another concern with primary cultures fromtendon tissue is that there are no specific markersto differentiate tenocytes from other fibroblastic cellsoriginating from the endotenon, epitenon or parate-non. Some researchers have suggested that sequen-tial enzymatic treatment can be used to separatefibroblasts from the epitenon and outer sheaths(Banes et al., 1988); however, there is no way to becertain without appropriate markers in cell cultureor in situ.

A further well-known problemwith tendon cell cul-ture is that within a few passages, substantial pheno-typic drift occurs; this is the case for cells from bothnormal and injured tendons (Schwarz et al., 1976;Bernard-Beaubois et al., 1997; Yao et al., 2006). Docu-mented changes have included increased ratios oftype III to type I collagen and reduced expressionof decorin; it has been proposed that only cells withinthe first three passages be used (Mazzocca et al.,2011). This is a difficult situation when trying to ob-tain sufficient cells from very small biopsy specimensobtained from human patients, but is less of a problemwhen working with equine models as large amounts oftissue can be acquired, and further specimens can al-ways be obtained.

Many of the above issues in tendon cell culture mayrelate to failure to provide appropriate extracellularmatrices. In two-dimensional culture, this involvesproviding surfaces that encourage normal cellular be-haviour. Collagen type I surfaces are known to bemore appropriate than either glass or plastic; how-ever, the random orientation of small collagen fibrils


in this situation provides only a subset of the organiza-tional cues provided in tissue. Methods for encourag-ing collagen fibril self-organization to mimic that seenin tissue include variations of systems that drain, shearand confine collagen to thin films or microfluidicchannels (Lee et al., 2006; Xu et al., 2011). Electro-spun collagen, which involves dispensing the proteinfrom a charged tip, has also been used to generatescaffolds for tissue engineering (Matthews et al.,2002). However, the tensile properties of such fibrilsand the extent to which they act as true biomimeticsof tissue ECM are controversial (Zeugolis et al.,2008; Jha et al., 2011). While an exhaustive descrip-tion of such methods is beyond the scope of this re-view, it should be noted that we cannot yetconstruct collagen fibrils of specified length or thick-ness. One practical solution in equine research hasbeen to use tendon tissue specimens as surfaces. Inone study, mesenchymal stem cells (bone marrow-derived) and SDFT tenocytes seeded onto acellularnative tendon matrices (ANTs) aligned in the longi-tudinal axis (sometimes in rows), down-regulatedtype I and III collagen gene expression and becamemore similar to each other than in standard two-dimensional culture (Richardson et al., 2007). TheANTs were created by cutting 70 mm SDFT cryosec-tions and sterilizing them in ethanol prior to rehydra-tion. Culturing tenocytes on surfaces of acellulartendon sections in our laboratory (without ethanoltreatment) resulted in similar alignment and a morespindle-shaped appearance.

Many of the above drawbacks associated with two-dimensional culture can be partially mitigated byperforming three-dimensional culture in which thetendon cells are seeded into various gels or scaffolds.High density culture in three dimensions has beenfound to support tenocyte differentiation and repair,reversing the de-differentiation seen in monolayers(Stoll et al., 2010). The emphasis has been on design-ing cellematrix combinations for transplantation,with a biodegradable matrix being required to drivecell survival, migration and repair (Cao et al., 2002;Kazimo�glu et al., 2003). Commonly used gel polymersinclude natural and synthetic hydrogels (Slaughteret al., 2009). Hydrogels are customizable networks ofhighly water-absorbent homopolymers, copolymersor macromers cross-linked to form insoluble polymermatrices. Natural hydrogel constituents include aga-rose, alginate, chitosan, hyaluronan, fibrin, collagenand Matrigel; Matrigel is the trade name for a gelati-nous protein mixture secreted by a mouse sarcomacell line. Synthetic gels including various polystyreneformats and poly[lactic-co-glycolic-acid] (PLGA)-scaffolds have also been employed to study the growthand differentiation properties of ex-vivo tenocytes or

stem-like cells. The synthetic versions are advanta-geous in terms of ease of manufacture and productconsistency; however, the chemistries and cross-linkers used in the production of these synthetic gelscan lead to concerns over toxic contaminants in theend-product. One problem when using any gel typeis that the normal tendon matrix is not as highly hy-drated. A promising methodology has involved usingplastic compression to drive water from collagen/cellscaffolds in order to generate more authentic tissue bi-omimicry; there is less than 10% cell death, and theresultant flat collagen sheet is then rolled along itsshort axis to form a tendon-like structure (Brownet al., 2005). More sophisticated cell model systemsemploy mixed cellular populations in order to gener-ate composite scaffolds. These are especially useful instudying graft attachment to bone (Spalazzi et al.,2006). Alternatively, tissue explants may be used asexperimental models (i.e. retaining the cells in situ intheir native matrix). Multiple explants of sufficientsize (2e3 mm thick) can be obtained from the largeequine digital tendons. A major disadvantage is thatexplants cannot be stored and there is therefore con-siderable planning and time-pressure involved inthese experiments. There will be a healing responsethat might complicate analysis although in one studyit was determined that fibroblastic proliferation andcollagen synthesis predominantly occurred at theend of explants and in the immediately subjacent en-dotenon (Murphy and Nixon, 1997).

Stem Cells/Progenitor Cells in Tendon Tissue

A significant question, aside from the nature of teno-cytes and other fibroblastic cells, is whether stem cellsare also present in tendon tissue and how they are af-fected by exercise and/or microdamage. It has beenhypothesized that tendon-specific stem or progenitorcells (TSPCs) are localized in the endotenon, al-though this has not been proven (Godwin et al.,2012). Progenitor cells within tissues are intermediatebetween stem cells and fully-differentiated cells; un-like stem cells they can only replicate a limited num-ber of times and tend to be more limited in terms ofthe cell types into which they will differentiate. How-ever, the terms ‘progenitor’ and ‘stem’ cell are oftenused interchangeably, reflecting a current lack of un-derstanding of these complex and most likely hetero-geneous populations in various tissues. Stem cellproperties of cells of both bone marrow and tendonorigin have been demonstrated including clonogenic-ity, self-renewal and multi-differentiation potential;some origin-specific differences in the latter havebeen noted, with, for example, more robust osteogen-esis from TSPCs in several studies (Bi et al., 2007;

Fig. 4. Image of a tendon fascicle from the digital extensor tendonof a Sprague-Dawley rat showing tenocyte nuclei (cyan),primary cilia (green) and the collagenous ECM (blue).The cilia, extending from surfaces of tenocytes, are alignedin the longitudinal axis of the tendon and are thought toplay a mechanosensory role. From: Donnelly et al. (2010)Primary cilia are highly oriented with respect to collagendirection and long axis of extensor tendon. Journal of Ortho-paedic Research, 28, 77e82. Reprinted with permission.


Lovati et al., 2011; Rui et al., 2012). Multipotent cellsisolated from tendons of rodents, rabbits and peoplehave shown (species-variable) differences in gene ex-pression profile and cell surface marker expressionto those of bone marrow stem cells, indicating thatthey may be identifiable in situ (Bi et al., 2007; Ruiet al., 2012). In mouse tendon, TSPCs were definedas being slow cycling during a rapid growth period,with retention of bromodeoxyuridine label for 8e14weeks (Bi et al., 2007). These cells were found withinthe collagenous matrix rather than the endotenon,and ultimately comprised approximately 6% of thecellular population; it was suggested that in tendonsthe stem cell ‘niche’ is comprised largely of ECMrather than being perivascular in nature (i.e. withinthe fascicle rather than the endotenon; Bi et al.,2007). However, the markers identified as being spe-cific for TSPCs in that study were not used to identifythem in the tissue sections.Many authors consider theperivascular niche to be more likely, with culturedperivascular cells from human supraspinatus tendonspecimens expressing both tendon cell markers (e.g.scleraxis, collagen types I and III) and stem cellmarkers (e.g. CD133, Musashi-1, nestin, CD44 andCD29); importantly, the stem cell markers and scler-axis were also expressed by these cells in situ (Tempferet al., 2009). These perivascular TSPCs may be peri-cytic, as they also express a-smooth muscle actin.

It is not known how TSPCs in specific tendons, in-cluding those that are injury-prone, respond to vari-ous exercise-induced mechanobiological factors.Results from work using rodent TSPCs suggest thatvarying levels of mechanical strain may drive thesecells down either undesirable or desirable differentia-tion pathways; as an example of the latter, in onestudy TSPCs from treadmill-exercised mice showedincreased proliferation and there was increased colla-gen production inmixedTSPC-tenocyte culture, par-ticularly when sourced from the AT, compared withthe anatomically opposing patellar tendon (Zhangand Wang, 2010; Zhang et al., 2010a). AlthoughTSPC subpopulations in the SDFT and AT do not ef-fect regenerative repair following rupture, their rolemay be more important in response to and repair ofmicrodamage. This could include induction of appro-priate responses from tenocytes, as the two cell typesare likely to be communicating within the tendon tis-sue (Zhang et al., 2010a). Additionally, studies of ro-dent tendons indicate that TSPC numbers anddifferentiation and self-renewal capacities reducewith age, this being a major epidemiological factorin tendon injury (Zhou et al., 2010). In terms of cellculture, physiological (i.e. lower than ambient) oxy-gen tensions and/or step changes in these may encour-age growth of stem/progenitor cells. Increasing

amounts of data indicate that stem or progenitorcell niches are maintained at lower oxygen levelswithin the tissue (Mohyeldin et al., 2010). One sugges-tion is that reduced free radical exposure limits stress-induced differentiation, although stem-like cells alsoappear to express robust ROS quenching capacities.The response to cell culture conditions, in additionto the tissue location, number and age- and tendon-specific repair capacities of TSPCs in equine digitaltendons require further investigation. This may beparticularly important given that the ECM that pu-tatively forms the stem cell microenvironment showssignificant differences between tendons, with ageand in response to exercise in horses and people.

Tenocyte Mechanotransduction

Tenocytes and, as discussed above, TSPCs, are highlyresponsive to mechanical loading. However, the pre-cise mechanisms by which they sense these signals andtransduce them into cellular responses remain a keyarea of research. Mechanical signals are most likelydetected via fluid flow-induced shear stress andmatrix-induced deformation of the cytoplasmic mem-brane, nucleus and/or cytoskeleton and integrin-containing cellematrix adhesion complexes(Arnoczky et al., 2002a; Wang, 2006). Tenocytesalso express a single primary cilium (i.e. a microtu-bule-based sensory organelle) that projects an axo-neme from the cell surface into the ECM and isaligned in the longitudinal axis of the tendon(Donnelly et al., 2010; Fig. 4). These cilia are deflectedin response to tensile loading and possibly also fluid


flow and are thought to play a significant role in me-chanotransduction (Lavagnino et al., 2011). In onestudy of cultured rat tail tendons, the length of teno-cyte cilia increased during stress deprivation, butthis was reversed by cyclical tensile loading(Gardner et al., 2011). Also, in response to mechanicalloading, tenocytes up-regulate n-cadherin (AJ com-ponent) and vinculin (linking integrin andAJ compo-nents to the cytoskeleton) and assemble more of theiractin fibres into stress fibres (i.e. with tropomyosin;Ralphs et al., 2002). Maintenance of cellecell contactduring stretching by these junctional complexes andstress fibres is obviously important for mechanotrans-duction and possibly also for active recovery duringrelaxation of the tendon. However, it also preventsmechanical uncoupling of the GJs that are necessaryfor coordination of tenocyte activity via direct inter-cellular communication (see below). During cyclicalloading, connexin 43 protein (Cx43), a major GJ pro-tein in tendon andmany other tissues, has been shownto co-localize with actin; recent work has indicatedthat the protein drebrin physically links Cx43 to theactin cytoskeleton, in theory acting to stabilize GJsat the plasmamembrane in addition to facilitating re-sponses to external signals (Ralphs et al., 2002;Butkevich et al., 2004; Wall et al., 2007a). The AJandGJ are connected functionally by feedback, in ad-dition to physical links as discussed below (reviewedby Giepmans, 2006).

It has been hypothesized that tenocytes establishinternal cytoskeletal tension with a mechanostat set-point that determines whether responses to mechani-cal loading will be anabolic or catabolic (Lavagninoand Arnoczky, 2005). How the exact tensile strain ex-perienced by tenocytes themselves relates to thestrains applied to the ECM as a whole are uncertain,as this appears to be influenced by cellular orienta-tion, cellular stiffness, cytoskeleton organization, sub-cellular organelles and the type and placement ofcellesubstrate adhesions, resulting in inter- andintracellular heterogeneity; however, estimates arethat the fraction of the applied strain transmitted tocells is at least moderate (Wall et al., 2007b).

Coordinated Signal Transmission: the Gap Junction

Connection

The GJs linking tenocytes allow metabolites, ions in-cluding Ca+2 and small molecules<1 kDa (includinginositol [1, 4, 5]-triphosphate; IP3) to pass directly be-tween the cytoplasm of connected cells, facilitating in-tegration and synchronization of cellular activities(i.e. the transmission of mechanically-activated sig-nalling; Wall and Banes, 2005). In tendons, GJsmay also serve the specific function of provision of nu-

trients, as most of the tenocytes are not located closelyto the endotenon-confined blood vessels (Tanji et al.,1995). GJs are clustered in their hundreds to thou-sands in cell membranes to form dynamic GJ plaques.Each individual GJ is comprised of two hemichannels(connexons; HCs) embedded in cytoplasmic mem-branes of apposing cells that dock to form a tightlysealed channel. The hemichannels are in turn eachcomprised of six connexin proteins, and Cx43 andCx32 have been identified as the major isoforms inequine, rat, avian and human tendons (McNeillyet al., 1996; Stanley et al., 2007). Some researchershave also demonstrated Cx26 expression, althoughin rat tail tendon this was found to be largely cytoplas-mic (Maeda et al., 2012). Significant amounts of Cx26protein have not been identified in equine tendon(JPK and DB, unpublished observation). The cyto-plasmic site of Cx43 protein is known to be linkedinto a macromolecular complex with AJ componentsthat include n-cadherin and catenins (reviewed byGiepmans, 2006). The Cx43 protein is also physicallylinked to the actin cytoskeleton (as mentioned above)and to the a- and b-tubulin proteins that form micro-tubules.

There can be rapid alterations in GJ expression andfunction, with half-lives of 1e5 h reported in culturedcells and intact tissue and other factors, including pHand intra-cellular calcium levels, determiningwhether a GJ is open or closed (Segretain and Falk,2004). The precise manner in which mechanical stim-ulation regulates the intercellular communication oftenocytes is still poorly understood. Gap junctional in-tercellular communication (GJIC) is known to benecessary for a number of responses to mechanicalloading including contribution to inositol triphos-phate (IP3)-mediated Ca+2 waves and up-regulationof DNA and collagen synthesis; the latter were pre-vented experimentally in avian tendon cell and tissueculture by chemical GJ blockade (Banes et al., 1999a,b; Wall and Banes, 2005). Conflicting in-vitro resultson effects of mechanical load on expression and func-tion of the GJs themselves have been published, whichmay reflect differences in load regimens, cells or ten-dons used and species of origin. Avian tenocytes sub-jected to 0.5% cyclical strain at 1 Hz for up to 7 daysup-regulatedCx43 gene expression, andwhen norepi-nephrine was also used they increased GJIC (Walland Banes, 2005). Reduction in Cx43 protein expres-sion was reported in avian tenocytes subjected to dis-continuous 0.5% cyclical strain (Banes et al., 1999b).In rat tail tendon fascicles subjected to 1 N static ten-sile load for 1 h, there was a significant reduction inCx43 protein expression and GJ permeability thatdid not occur following a 10 min loading period(Maeda et al., 2012). Similar studies using equine or


human tendon cells or tissue specimens and morephysiological cyclical mechanical loading regimenshave not been conducted.

Connexin Isoforms. Potential differential effects ofCx43 and Cx32 isoforms on tenocyte function also re-quire further investigation. These isoforms do not oc-cur within the same hemichannels (i.e. as heteromericconnexons) and have a different physical distribution,with both linking tenocytes within the same row, butonly Cx43 comprising GJs between cells in differentrows (i.e. they may link different local cellular net-works; Elfgang et al., 1995; McNeilly et al., 1996).There is some evidence that they may modulate teno-cyte collagen expression differentially with Cx43 be-ing inhibitory and Cx32 stimulatory (Banes et al.,1999a; Waggett et al., 2006). The latter study doesnot agree with increased collagen gene (Col1a1) andtotal collagen protein expression by dermal fibro-blasts in acutely Cx43-asODN-treated skin wounds;however, the environment in damaged connective tis-sue in vivo is more complex and included elevatedTGF-b1 activation (Mori et al., 2006). In osteoblasts,a Cx43-responsive element in the promoter region ofthe collagen type 1 a-1 chain has been identified(Stains et al., 2003). The effects of different mechani-cal loading regimens on specific connexin isoforms ex-pressed by tendon cells have not been determined.Finally, it should be noted that in ‘wrap around’ re-gions of tendons where type 3 tenocytes predominateand the expression of GJs is low to non-existent, coor-dination of their behaviour must occur by othermeans (e.g. via indirect cytokine and growth factorsignalling; Ralphs et al., 1998). This wrap around re-gion of the SDFT is not prone to injury.

Comparative Biology of Injury-Prone andNon-Injury-Prone Tendons in Horses

The comparison of injury-prone and non-injury-prone equine digital tendons has significantly in-creased understanding of their relative structure andfunction and the pathogenesis of exercise- and age-induced microdamage.

Functional Determination of Matrix Composition: Energy-

Storing Versus Positional Tendons

The common digital extensor tendon (CDET), whichanatomically opposes the SDFT, functions only to po-sition the digit correctly immediately prior to weight-bearing andmakes no significant contribution to elas-tic energy storage. In the SDFT, strains of up to16.6% as measured in a galloping horse approximatethe failure level, as mentioned above (Stephens et al.,

1989; Wilson and Goodship, 1991). In contrast, theCDET, as a non-load-bearing structure, is notthought to experience levels of strain greater than2.5%; as this tendon has a failure strain level of ap-proximately 10%, there is therefore a wide mechani-cal safety margin, and injury is rarely reported(Batson et al., 2003; Birch et al., 2008a). The DDFTsituated immediately cranial to the SDFT on the pal-mar surface of the forelimb is also not considered to bean energy-storing structure. The DDFT experiencessignificantly lower mechanical loads, strains andstresses during weight bearing at a walk, trot andcanter and is thought to function mainly to flex thedigit during the late swing phase (Platt et al., 1994;Butcher et al., 2009). The suspensory ligament onthe palmar aspect of the third metacarpal bone isenergy storing and is also frequently injured(Williams et al., 2001; Butcher et al., 2009), but asan intraosseous muscle rather than a tendon: it willnot be covered in this review. The differing functionsof the SDFT and CDET, which have similar lengths,are reflected in their structural and material proper-ties: the SDFT is a stiffer and stronger structure rela-tive to the CDET as it has a greater cross-sectionalarea; however, the SDFT matrix as a material hasa composition that is more compliant and elastic(with a lower elastic modulus) to allow the tendonto deform and function as an energy store (Batsonet al., 2003). These properties do show significant var-iation between individual horses (Birch, 2007). Simi-lar differences in strength and elastic modulus havealso been noted between the energy-storing humanAT and the anatomically opposing, positional andnon-injury-prone anterior tibialis tendon (ATT)(Birch, 2007). Corresponding to the tendon-specificdifferences in mechanical properties, as would be ex-pected, are very different matrices synthesized bythe respective tenocyte populations. The SDFT hasa higher content of sulphated glycosaminoglycans(GAGs), collagen oligomeric matrix protein(COMP; a glycoprotein) and water versus theCDET, with similar differences between the humanAT and ATT (Batson et al., 2003; Birch, 2007). Al-though the SDFT and CDET have similar collagencontents (as does the AT versus the ATT), theSDFT has a significantly lower collagen fibril mass-average diameter (i.e. the collagen is organized differ-ently such that it contains a greater proportion ofsmall diameter fibrils; Patterson-Kane et al., 1997b;Edwards et al., 2005). Smaller diameter collagen fi-brils have a larger surface area-to-volume ratio perunit cross-sectional area and therefore provide greaterelasticity via their greater potential to form interfibril-lar cross-links; this prevents fibrils ‘creeping’ past eachother permanently.


Tendon-Specific Differences in Tenocyte Populations of

Immature and Adult Horses: Cellular ‘Switch-Off’

In both the SDFT and CDET, the cellularity reducessignificantly during maturation (up to approximately2 years of age) and thenmore gradually during ageing;in all age groups theSDFT is significantlymore cellularthan the CDET (Stanley et al., 2007; Young et al.,2009). Similarly in man, the adult AT is more cellularthan the ATT (Birch, 2007). Between 50 days and1 year of age, the SDFT of a foal more than doublesin cross-sectional area, in association with rapid in-creases in bodyweight, and requiring a very active te-nocyte population (Kasashima et al., 2002).However, a number of studies have shown that inadulthood, the SDFT tenocytes become significantlyless active; in one such study the expression of Cx43and Cx32 proteins, synthesis of total, type I and typeIII collagens and GJ permeability were shown to re-duce significantly with maturation (Young et al.,2009). This was not the case in theCDET, inwhich ex-pression of Cx43 and synthesis of type I and type IIIcollagen proteins actually increased with maturationand exceeded levels in the (more cellular) SDFT; addi-tionally, GJ permeability was maintained (Younget al., 2009). Other studies have shown correspondingevidence of lower collagen turnover in the adultSDFT compared with the CDET, with accumulationof ‘old’ collagen, lower levels of MMP activity andlower levels of the cross-linked carboxyterminal telo-peptide of type I collagen (ICTP; i.e. a collagen break-down product; Batson et al., 2003; Birch et al., 2008b).In one study, analysis of aspartic acid racemization ofcollagen in a group of horses with a wide age range(4e30 years) indicated a mean collagen half-life of198 years in the SDFT versus 34 years for the CDET(Thorpe et al., 2010). There was significant variationbetween individual horses in these values. In cell cul-ture, SDFT tenocytes showed age-related reductionsin strain-induced collagen synthesis, but CDET cellsdid not (Goodman et al., 2004). It has been suggestedthat the SDFTof the adult horse ‘switches off’ tenocyteactivity to maintain the stiffness and elasticity of thematrixwithin theverynarrowoptimal limits for appro-priate strength and energy storage (Smith et al., 2002).If this is true, it has very serious implications for theability of the SDFT tenocyte network to respond tothe expected higher levels and frequencies of damagethan those sustained by the CDET.

The Pathogenesis of Exercise-InducedTendon Microdamage

If tenocytes and, possibly, other cellular populationsdid not constantly repair damage, all tendons would

rapidly weaken and rupture (Ker, 2002). However,when microdamage does accumulate there is debateas to (1) the extent to which it is actively initiatedand mediated by aberrant cellular activity, as op-posed to unrepaired matrix fatigue caused by me-chanical overuse and cellular loss, and (2) what thenature of the initial or very early pathology is. Dueto the prevailing opinion that the pathology is degen-erative, the pre-rupture condition is currentlyreferred to as tendonopathy (tendinopathy) or tendo-nosis rather than tendonitis.

Evidence of Matrical and Cellular Microdamage

Studies of racehorses and experimental horses sub-jected to specific treadmill exercise regimens have fa-cilitated demonstration of subclinical exercise- andage-induced microdamage to the matrix. Many ofthese changes have been located specifically in thecore of the cross-section of the metacarpal segmentsof the SDFT, where lesions occur most frequently.These have included: age-related increases in typeIII collagen and partially degraded collagen; in-creased total sulphatedGAGs, type III collagen levelsand collagen turnover in red core lesions; exercise-induced reductions in GAG content and collagen fi-bril mass-average diameter; denaturation and depo-lymerization of collagen fibres and increasedcathepsin B activity; and exercise- and age-related al-terations in the collagen crimp waveform (Miles et al.,1994; Patterson-Kane et al., 1997a, b; Birch et al.,2008a; Thorpe et al., 2010). Cell numbers, as mea-sured by quantification of DNA levels, were notshown conclusively to reduce in these areas of micro-damage (Birch et al., 2008a). Other measures of cellu-lar death, damage or inappropriate activity have notbeen made.

Studies using human tissue, in which microdamageis correlated with a well-defined exercise history, aredifficult to perform; however, there has been accessto ruptured tissue during surgical procedures. Thishas allowed observation of chronic degeneration(i.e. prior, subclinical pathological change) in re-cently ruptured tendons; in one large study degener-ation was found in 97% of tendon ruptures and,interestingly in 35% of 445 cadaveric ‘control’ speci-mens (Kannus and J�ozsa, 1991). The degenerativechanges were more severe in ruptured tendons as op-posed to tendonopathy without rupture, supportingthe microdamage theory (Tallon et al., 2001). As theequine SDFT is not usually treated surgically, accessto ruptured tendons in terms of tissue harvesting hasbeen very limited and similar pathological studies ofpotentially more advanced degenerative changehave not been performed. Degenerative changes

Fig. 5. A proposed model for a continuum of tendon pathology, inwhich there is a progression from reaction to dysrepair andthen irreversible degeneration if endogenous cellular popu-lations do not react appropriately. Not all aspects of thispathway have been identified morphologically in equinetendons. Cook and Purdam (2009) Is tendon pathologya continuum? A pathology model to explain the clinicalpresentation of load-induced tendinopathy. British Journalof SportsMedicine, 43, 409e416. Reprinted with permission.


described in human tendons are similar (but not iden-tical) to those found in equine exercise studies and in-clude evidence of new collagen synthesis withincreased type III collagen and reduced type I colla-gen, increases in levels of fibronectin, tenascin C,GAGs and proteoglycans and histological observa-tions of apparently irreversible matrix change includ-ing irregular arrangement, altered crimping,disruption and reductions in density of collagen(Riley et al., 1994a, b; J€arvinen et al., 1997; Cooket al., 2004). Fibronectin and tenascin C are glycopro-teins that are thought to modulate cellular activityand migration (Riley, 2005). Increased amounts oftype III collagen at AT rupture sites in one studywere not associated with evidence of recent synthesis(cleaved propeptides), supporting the theory thatthis accumulates prior to clinical injury (Eriksenet al., 2002). Histological degenerative lesions seen inthe human AT, but not noted in horses, include mu-coid degeneration, fatty degeneration (accumulationof adipocytes), calcification and neovascularization;this may to some extent relate to the fact that theyare not identical anatomical structures. In bothasymptomatic and painful degenerate human ten-dons, various histological changes have been notedin tenocyte populations including increased or de-creased cellularity, plump ovoid or rounded nucleiwith ultrastructural evidence of increased productionof collagen and proteoglycans, and chondroid changein more extreme cases; tendon cells cultured from le-sions have maintained a higher proliferation rateand produced greater quantities of type III collagen(Leadbetter, 1992; Maffulli et al., 2000; Rolf et al.,2001; Cook et al., 2004). Ultrastructural changes inthe tenocytes have been interpreted as indicative ofhypoxic degeneration, including alterations in sizesand shapes of mitochondria and nuclei, increasednumbers and sizes of lysosomal vacuoles, ‘hypoxic’and lipid vacuolation and intracytoplasmic or mito-chondrial calcification (Kannus and J�ozsa, 1991;Leadbetter, 1992). Not all authors agree with the def-inition of these lesions as degenerate and irreversible,referring to them as a failed healing response; the re-ality may be that both are occurring or that there isa continuum of pathology from reactive tendonop-athy through dysrepair (tenocyte nuclear roundingand matrix production) to degeneration (cell lossand death), at which point reversibility is unlikelyand rupture may occur during loading (Cook andPurdam, 2009; Fig. 5).

The ‘Tendonosis Cycle’

The tendonosis cycle is thought to begin when me-chanical overloading and other factors overwhelm

cellular repair mechanisms, resulting in increasedcell death and potentially inappropriate syntheticand degradative responses that further weaken thematrix (Leadbetter, 1992; Yuan et al., 2002); thisdoes not exclude the contribution of direct damageto the matrix, either in initiation or, particularly, inprogression of damage when compromised tissue con-tinues to be loaded. Reasons for inadequate cellularrepair may include insufficient time between episodesof exercise, a high frequency, duration and/or magni-tude of overuse that simply overwhelms capacity, andindividual susceptibility due to differences in matrixcomposition and/or cellular activity. Alterations ingene, protein and/or enzyme expression levels of var-ious MMPs, TIMPs and ADAM (A Disintegrin AndMetalloproteinase) family peptidases have been mea-sured in human degenerate tendon tissue in multiplestudies (i.e. there is significant evidence of remodel-ling of the matrix prior to the onset of clinical symp-toms; Jones et al., 2006; Riley et al., 2002).Elevations of MMP-13, MMP-1 and ADAM withThrombospondin Motifs (ADAMTS)-5 have been


documented in clinically injured equine SDFTs, in-terestingly with similar but less marked changes inthe contralateral, asymptomatic tendon (Clegget al., 2007).

Studying the Effects of Cell Stress and Damage in Cell Culture

Mechanistic studies are required to explain theabove observations in damaged equine and humantendon tissue. In cell culture systems, determinationof the effects of various stressors requires an initiallyhealthy (i.e. non-stressed) population. Importantly,cellular injury in culture systems may not necessar-ily alter numbers of viable cells, this being the pa-rameter typically measured. Many researchers alsouse growth rate as an indicator of the health ofthe system; however, stress can be a potent inducerof replication. It is not entirely clear how tenocytes,which are largely non-replicative in their naturalsetting, truly respond to the continuous mitogenicstimuli ordinarily provided in vitro. Researchersshould be aware of this and as best as they can, con-trol for such responses in their experiments by opti-mizing cell culture conditions. Sublethal levels ofcellular stress might for example be measured priorto manipulation, by quantifying ROS generationusing 20,70-dichlorodihydrofluorescein diacetate(H2DCFDA).

As tenocytes under normal conditions are subjectedto cyclical strain in vivo and many researchers havesuggested that mechanical overstimulation of teno-cytes initiates microdamage, it is critically importantto include this factor in culture systems. There is a sig-nificant advantage in this respect in using equine dig-ital tendon cells as there is published information onstrain levels, frequencies and patterns experienced atvarious gaits including galloping. This has been ob-tained from in-vivo strain gauge measurements, in-vitro measurements and modelling studies (Lochneret al., 1980; Stephens et al., 1989; Lawson et al.,2007). Many previous in-vitro studies have not repli-cated conditions normally experienced by tenocytesin vivo, inclusive of physiological levels and durationsof cyclical strain (Arnoczky et al., 2007a). It is impor-tant to recognize that tenocyte responses to strain canvary according to the tendon of origin, the age of theanimal and the differentiation status (Goodman et al.,2004; Zhang et al., 2010a). In one study, differentia-tion of rabbit ATTSPCs altered depending on the de-gree of strain; after 4% uniaxial strain at 0.5 Hz for12 h, the cells expressed more type I collagen consis-tent with tenocyte activity. However, when thatstrain level was doubled they became enriched formarkers of adipocytes, chondrocytes or osteocytes;lipid production, mucinous matrix formation and cal-

cification have all been documented in human ten-donopathy, as mentioned above.

The available devices with which to alter the me-chanical environment in cell culture vary in terms ofprecision, loadingmodality (e.g. uniaxial strain, biax-ial strain, compression) and methodology used (e.g.pneumatic, magnetic, electromagnetic) (reviewedby Brown, 2000). ‘Dial-in’ strain parameters can bedelivered to cells seeded on elastic membranes or ingel constructs. However, over extended periods oftime (i.e. high cycle numbers) the reliability withwhich these strains are delivered can vary. This canoccur due to cycle-induced changes in the immobiliz-ing membrane or gel, or irregularities in the gel poly-mer in which cells are seeded, if three-dimensionalculture is used (Bieler et al., 2009). Where three-di-mensional culture or tissue explants are used, subject-ing these to strain is typically even more technicallydemanding, in particular due to difficulties in grip-ping the specimens appropriately. The available ap-paratus, both commercial and custom-built, allsuffer deficiencies in terms of control and measure-ment of strain. One group has successfully developedan explant model in which SDFT specimens weresubjected to 5% strain at 1 Hz for 24 h. The explants,particularly those from older horses, showed dimin-ished mechanical strength that was associated withthe presence of viable cells and their MMP activity(Dudhia et al., 2007).

Interestingly, more recent work indicates that it isunderstimulation of tendon cells and tissue (in rodentand ovine models) that results in rapid up-regulationof expression of various MMPs and, in some models,expression of ADAMTS (Arnoczky et al., 2008;Smith et al., 2008; Thornton et al., 2010); ADAMTSpeptidases cleave various proteoglycans. This couldbe occurring where tendon cells are cultured in staticconditions. Mechanical understimulation in the ten-don is proposed to occur initially by damage to iso-lated collagen fibrils during exercise, either at loadsapproaching mechanical safety margins of the tendonmatrix or by localized abnormal loading (non-uni-form stress), resulting in disturbance of mechano-transduction mechanisms (Arnoczky et al., 2007a).This loss of cellularematrix contact would be acceler-ated by pericellular matrix breakdown resulting fromincreased tenocyte production ofMMPs. Isolated col-lagen fibril damage has been produced in the labora-tory by mechanical overloading of rat tail tendonfascicles, with MMP up-regulation only involving te-nocytes directly associated with those fibrils(Lavagnino et al., 2006). Matrix breakdown instress-deprived rat tail tendons in vitro was preventedby treatment with MMP inhibitors (Arnoczky et al.,2007b).


Contribution of Tenocyte Death to Microdamage: Primary or

Secondary?

Apoptotic cells have been noted in tendonopathy inman (including that of the AT), and in horses withSDFT lesions both with and without a clinical his-tory of injury (Yuan et al., 2002; Hosaka et al.,2005; Pearce et al., 2009). In patellar tendons of hu-man athletes this has been seen against a backgroundof hypercellularity, interpreted in the tendon pathol-ogy continuum model as progression from dysrepairto degeneration (Lian et al., 2007; Cook andPurdam, 2009) (Fig. 5). Sparsely cellular tendonssuch as the SDFT are potentially highly vulnerableto this loss. Mechanical stress-deprivation in vitro hasbeen shown to increase caspase-3 expression and ap-optotic death of tendon cells (i.e. apoptosis may oc-cur as a downstream event of collagen fibrildamage, as described above, through a process po-tentially involving alterations in integrin-basedcellematrix connections and/or cell shape;Egerbacher et al., 2008). Some researchers have sug-gested that high levels of cyclical strain cause celldeath directly; however, apoptosis in highly strainedtendons in in-vitro experiments could also have re-sulted from collagen fibril damage (Scott et al.,2005). Tenocytes in degenerate human patellar ten-don have been reported to up-regulate adrenergicreceptors, stimulation of which is another mecha-nism by which apoptosis might be induced(Danielson et al., 2007). Cyclical strain for15e120 min (3e9%) was shown to activate stress-activated protein kinases in canine tenocytes30e120 min later, in a manner that was propor-tional to the magnitude of the strain applied; theseproteins are important upstream regulators of theapoptosis cascade in some cell types, although theydid not induce it in this model (Arnoczky et al.,2002b). It is possible that cell death occurs in re-sponse to multiple interacting environmental factorsin the tendon during exercise including hyperther-mia. Canine patellar tendon cells subjected to tem-peratures up to 42�C showed cDNA fragmentationand caspase-3 activation, with significant increasesin both parameters when 9% cyclical strain wasadded; however, the cells were subjected to theseconditions for non-physiological periods of 6 and24 h (Tian et al., 2004). Our pilot studies using cul-tured equine SDFT cell monolayers showed a highlevel of apoptotic cell death from 2 h followinga physiological period of heating (i.e. up to45e47�C over a 10 min period as in the gallopinghorse) (Burrows et al., 2009). This cell death waslargely prevented by chemical blockade of GJs using18-b glycyrrhetinic acid, suggesting that GJIC may

spread death signals following exercise-induced heatstress. This concept is not unique to tendon tissue;propagation of ‘bystander damage’ also occurs in fo-cally injured spinal cord tissue and during reperfu-sion of the ischaemic myocardium; significantimprovements in functional recovery following spi-nal cord injury were noted in one study when anti-sense oligodeoxynucleotides were used to suppressCx43 expression (Garcia-Dorado et al., 2004;Cronin et al., 2008). Critically, it may not only beintact GJs that contribute to this phenomenon; un-coupled HC can open and allow exchange of ionsand small molecules between the cytoplasm andthe extracellular environment. Astrocytic ATP re-lease by HC following spinal cord trauma has re-cently been shown to significantly aggravatesecondary injury (Huang et al., 2012). There hasbeen no detailed investigation of HC function intendon tissue or cell culture, although we haveseen evidence of their presence on occasion in equineSDFT monolayers (JPK/DB personal observation).

Is the Initial Lesion Inflammatory or Degenerative?

Few researchers have identified inflammatory cells intendonopathy of human patients either histologicallyor ultrastructurally (J�ozsa et al., 1982; Leadbetter,1992; Astr€om and Rausing, 1995; J€arvinen et al.,1997). Where they have been noted, for example asinfiltrates of neutrophils, or lymphocytes and macro-phages, they have been interpreted as part of a healingprocess following matrix disruption and/or tenocytedeath that may precede rupture (Cetti et al., 2003).As a result, tendon microdamage has frequentlybeen proposed to be primarily degenerative. The ap-parent lack of inflammatory cells, however, may bedue to failure to visualize the very earliest lesions,with subsequent observations of progressive degener-ation actually being secondary to ineffective healing.In a recent study of subclinical tendonopathic lesionsin human subscapularis tendons, macrophages andmast cells were noted in significant numbers, al-though it is not certain at which stage of developmentthese lesions were (Millar et al., 2010). In people, ten-donopathy may be seen in conjunction with inflam-mation of the surrounding paratendon (J€arvinenet al., 1997).

The failure to see inflammatory cell infiltrates alsodoes not preclude inflammatory mediators being in-volved in aprimary fashion, forwhich there is nowcon-siderable evidence. In-vitro, humanand rabbit tendoncells and avian digital flexor tendon explants havebeen induced to produce various combinations of cy-clooxygenases, prostaglandin E2, phospholipase-A2,interleukin (IL)-1b, IL-6 and nitric oxide (NO) in


addition toMMPs, by application of cyclical strain orfluid-induced shear stress (Skutek et al., 2001;Archambault et al., 2002; Wang et al., 2004; Flicket al., 2006). Overexpression of nitric oxide synthase(NOS) has been demonstrated in Achilles tendonop-athy (Kane et al., 2008).Hypoxic tenocytes in earlyhu-man tendonopathic lesions (as demonstrated byincreased expression of hypoxia-inducible factor 1a)showedup-regulation of IL-6, IL-8 andmonocyte che-motactic protein (MCP)-1, type III collagen and somekeymediators of apoptosis (Millar et al., 2012). EquineSDFT cells heated to 40e45�C up-regulated tumournecrosis factor (TNF)-a expression, but only afternon-physiological periods (30e60 min; Hosaka et al.,2006). In human subjects exercising for prolonged pe-riods (30 min), accumulation of inflammatory media-tors including prostaglandin E2 and thromboxanehave been measured in peritendonous tissue(Langberg et al., 1999). It should be noted that stressdeprivation of rat patellar tendons and ATs in vivo

also resulted in increased expression of cytokines by te-nocytes (IL-1b, TNF-a and -b), in keeping with theabove theories on inappropriate cellular degradativeactivity in response to reducedmechanical stimulation(Uchida et al., 2005; Lavagnino et al., 2006;Wang et al.,2010). Pro-inflammatory cytokines such as IL-1 andTNF are able to stimulateMMP expression and activ-ity (Riley, 2005). Cytokines cannot always be pre-sumed to have a negative effect, however, withresults of one studyof human tenocytes in three-dimen-sional culture showing that IL-1b up-regulated Cx43and improved cell survival; GJsmay transmit ‘survivalsignals’ rather than death signals under certain cir-cumstances (Qi et al., 2011). Similar studies have notbeen conducted using equine tendon cells or tissue.

Conclusions

Exercise-induced injury to energy-storing tendons ofhorses and people occurs at a similarly high fre-quency, with failure of tissue regeneration contribut-ing to high re-injury rates or retirement from athleticactivity. There is now increasing research focus on therole of the tenocyte in accumulation of the matrix andcellular microdamage that predisposes to tendon rup-ture. Access to equine tissues and the ability to makemeasurements in vivo have already significantly im-proved our knowledge of the composition, structureand mechanical environment of injury-prone andnon-injury-prone digital tendons over a wide agerange. The imperative now is to develop tractablecell culture models using this species in order to studystress-induced damage and dysfunction; these modelswill facilitate a more authentic replication of the in-vivo microenvironment.

Acknowledgments

The authors would like to thank Dr. E. Donnelly andProf. J. Cook for provision of Figs. 4 and 5, respec-tively. Recent research on this topic has been sup-ported by: the Massey University Equine Trust,New Zealand; the Biotechnology and Biological Sci-ences Research Council (BBSRC), UK; and the Pet-plan Charitable Trust, UK.

References

Archambault JM, Elfervig-Wall MK, Tsuzaki M,Herzog W, Banes AJ (2002) Rabbit tendon cells pro-duce MMP-3 in response to fluid flow without signifi-cant calcium transients. Journal of Biomechanics, 35,303e309.

Arnoczky SP, Lavagnino M, Egerbacher M (2007a) Themechanobiological aetiopathogenesis of tendinopathy:is it the over-stimulation or the under-stimulation of ten-don cells? International Journal of Experimental Pathology,88, 217e226.

Arnoczky SP, Lavagnino M, Egerbacher M, Caballero O,Gardner K (2007b) Matrix metalloprotease inhibitorsprevent a decrease in the mechanical properties ofstress-deprived tendons: an in vitro experimental study.American Journal of Sports Medicine, 35, 763e769.

Arnoczky SP, Lavagnino M, Egerbacher M, Cabellero O,Gardner K et al. (2008) Loss of homeostatic strain altersmechanostat ‘set point’ of tendon cells in vitro. Clinical Or-thopaedics and Related Research, 466, 1583e1591.

Arnoczky SP, Lavagnino M, Whallon JH, Hoonjan A(2002a) In situ cell nucleus deformation in tendons undertensile load; a morphologic analysis using confocal lasermicroscopy. Journal of Orthopaedic Research, 20, 29e35.

Arnoczky SP, Tian T, LavagninoM,GardnerK, Schuler Pet al. (2002b) Activation of stress-activated protein ki-nases (SAPK) in tendon cells following cyclic strain:the effects of strain frequency, strain magnitude, and cy-tosolic calcium. Journal of Orthopaedic Research, 20,947e952.

Astr€om M, Rausing A (1995) Chronic Achilles tendinop-athy. A survey of surgical and histopathologic findings.Clinical Orthopaedics and Related Research, 316, 151e164.

Banes AJ, Donion K, Link GW, Gillespie Y, Bevin AG et al.(1988) Cell populations of tendon: a simplified methodfor isolation of synovial cells and internal fibroblasts:confirmation of origin and biologic properties. Journalof Orthopaedic Research, 6, 83e94.

Banes AJ, Horesovsky G, Tsuzaki M, Boitano S,Lawrence WT et al. (1999a) The connexin 43 gap junc-tion is a mechanosensitive gene in avian flexor tendoncells. In: The Biology of the Synovial Joint, B Caterson,C Archer, Eds., Harwood Academic Publishers, TheNetherlands, pp. 279e299.

Banes AJ, Weinhold P, Yang X, Tsuzaki M, BynumD et al.(1999b) Gap junctions regulate responses of tendon cellsex vivo to mechanical loading. Clinical Orthopaedics and Re-lated Research, 367, S356eS370.


Batson EL, Paramour RJ, Smith TJ, Birch HL, Patterson-Kane JC et al. (2003) Are the material properties andmatrix composition of equine flexor and extensor ten-dons determined by their functions? Equine Veterinary

Journal, 35, 314e318.Bernard-Beaubois K, Hecquet C, Hayem G, Rat P,

AdolpheM (1998) In vitro study of cytotoxicity of quino-lones on rabbit tenocytes. Cell Biology and Toxicology, 14,283e292.

Bernard-Beaubois K, Hecquet C, Houcine O, Hayem G,AdolpheM (1997) Culture and characterization of juve-nile rabbit tenocytes. Cell Biology and Toxicology, 13,103e113.

Bestwick CS, Maffulli N (2004) Reactive oxygen speciesand tendinopathy: do they matter? British Journal of

Sports Medicine, 38, 672e674.Bi Y, Ehirchiou D, Kilts TM, Inkson CA, EmbreeMC et al.

(2007) Identification of tendon stem/progenitor cellsand the role of the extracellular matrix in their niche.Nature Medicine, 13, 1219e1227.

Bieler FH, Ott CE, ThompsonMS, Seidel R, Ahrens S et al.(2009) Biaxial cell stimulation: a mechanical validation.Journal of Biomechanics, 42, 1692e1696.

Biewener AA (1998) Muscle-tendon stresses and elastic en-ergy storage during locomotion in the horse. ComparativeBiochemistry and Physiology. Part B, Biochemistry and Molec-

ular Biology, 120, 73e87.Birch HL (2007) Tendon matrix composition and turnover

in relation to functional requirements. International Jour-nal of Experimental Pathology, 88, 241e248.

Birch HL, Wilson AM, Goodship AE (2008a) Physical ac-tivity: does long-term, high-intensity exercise in horsesresult in tendon degeneration? Journal of Applied Physiol-ogy, 105, 1927e1933.

Birch HL,Worboys S, Eissa S, Jackson B, Strassburg S et al.(2008b) Matrix metabolism rate differs in functionallydistinct tendons. Matrix Biology, 27, 182e189.

Birk DE, Mayne R (1997) Localization of collagen types I,III and V during tendon development. Changes in col-lagen types I and III are correlated with changes in fi-bril diameter. European Journal of Cell Biology, 72,352e361.

Brown R, Wiseman M, Chuo C-B, Cheema U, Nazhat S(2005) Ultrarapid engineering of biomimetic materialsand tissues: fabrication of nano- and microstructuresby plastic compression. Advanced Functional Materials,15, 1762e1770.

Brown TD (2000) Techniques for mechanical stimulationof cells in vitro: a review. Journal of Biomechanics, 33, 3e14.

Burrows S, Becker DL, Fleck R, Patterson-Kane JC (2009)Gap junctional propagation of apoptosis in tendon fi-broblast monolayers in response to physiological hyper-thermia.Transactions of the Orthopaedic Research Society, 34,27.

Butcher MT, Hermanson JW, Ducharme NG,Mitchell LM, Soderholm LV et al. (2009) Contractilebehaviour of the forelimb digital flexors during steady-state locomotion in horses (Equus caballus): an initialtest of muscle architectural hypotheses about in vivo

function. Comparative Biochemistry and Physiology Part A,152, 100e114.

Butkevich E, H€ulsmann S, Wenzel D, Shirao T, Duden Ret al. (2004) Drebrin is a novel connexin-43 bindingpartner that links gap junctions to the submembrane cy-toskeleton. Current Biology, 14, 650e658.

Cao YL, Liu YT, Liu W, Shan QX, Buonocore SD et al.(2002) Bridging tendon defects using autologous teno-cyte engineered tendon in a hen model. Plastic and Recon-structive Surgery, 110, 1280e1289.

Cauvin ER, Smith RK, May SA, Ferguson MWJ (1998)Immunohistochemical localisation of TGF-b in equinetendons. A study in age-related changes in TGF-b ex-pression. International Journal of Experimental Pathology,79, A23.

Cetti R, Junge J, VybergM (2003) Spontaneous rupture ofthe Achilles tendon is preceded by widespread and bilat-eral tendon damage and ipsilateral inflammation. ActaOrthopaedica Scandinavica, 74, 78e84.

Chuen FS, Chuk CY, Ping WY, Nar WW, Kim HL et al.(2004) Immunohistochemical characterization of cellsin adult human patellar tendons. Journal of Histochemistry

and Cytochemistry, 52, 1151e1157.Clayton RAE, Court-Brown CM (2008) The epidemiology

of musculoskeletal tendinous and ligamentous injuries.Injury, 39, 1338e1344.

Clegg PD, Strassburg S, Smith RK (2007) Cell phenotypicvariation in normal and damaged tendons. InternationalJournal of Experimental Pathology, 88, 227e235.

Cook JL, Feller JA, Bonar SF, KhanKM (2004) Abnormaltenocyte morphology is more prevalent than collagendisruption in asymptomatic athletes’ patellar tendons.Journal of Orthopaedic Research, 22, 334e338.

Cook JL, PurdamCR (2009) Is tendon pathology a contin-uum? A pathology model to explain the clinical presen-tation of load-induced tendinopathy. British Journal of

Sports Medicine, 43, 409e416.Costa MA, Wu C, Bryant VP, Chong AKS, Pham HM

et al. (2006) Tissue engineering of flexor tendons: optimi-zation of tenocyte proliferation using growth factor sup-plementation. Tissue Engineering, 12, 1937e1943.

Crevier-Denoix N, Collobert C, Pourcelot P, Denoix JM,Sanaa M et al. (1997) Mechanical properties of patho-logical equine superficial digital flexor tendons. EquineVeterinary Journal, 23(Suppl.), 23e26.

Cronin M, Anderson PN, Cook JE, Green CR, Becker DL(2008) Blocking connexin43 expression reduces inflam-mation and improves functional recovery after spinalcord injury. Molecular and Cellular Neurosciences, 39,152e160.

Danielson P, Alfredson H, Forsgren S (2007) In-situ hy-bridisation studies confirming recent findings of the ex-istence of a local nonneuronal catecholamineproduction in human patellar tendinosis. Microscopy Re-

search and Technique, 70, 908e911.Donnelly E, Ascenzi MG, Farnum C (2010) Primary cilia

are highly oriented with respect to collagen directionand long axis of extensor tendon. Journal of OrthopaedicResearch, 28, 77e82.


Dudhia J, Scott CM, Draper ER, Heineg�ard D,Pitsillides AA et al. (2007) Aging enhances a mechani-cally-induced reduction in tendon strength by an activeprocess involving matrix metalloproteinase activity.Aging Cell, 6, 547e556.

Dyson SJ (2004)Medical management of superficial digitalflexor tendonitis: a comparative study in 219 horses(1992e2000). Equine Veterinary Journal, 36, 415e419.

Edwards LJ, Goodship AE, Birch HL, Patterson-Kane JC(2005) Effect of exercise on age-related changes in colla-gen fibril distributions in the common digital extensortendons of young horses. American Journal of Veterinary

Research, 66, 564e568.Egerbacher M, Arnoczky SP, Caballero O, Lavagnino M,

Gardner KL (2008) Loss of homeostatic tension inducesapoptosis in tendon cells. An in vitro study. Clinical Ortho-paedics and Related Research, 466, 1562e1568.

Elfgang C, Ekert R, Lichtenberg-Frat�e H, Butterweck A,Traub O et al. (1995) Specific permeability and selectiveformation of gap junction channels in connexin-transfected HeLa cells. Journal of Cell Biology, 129,805e817.

Ely ER, Avella CS, Price JS, Smith RKW,Wood JLN et al.(2009) Descriptive epidemiology of fracture, tendon andsuspensory ligament injuries in National Hunt race-horses in training. Equine Veterinary Journal, 41,372e378.

Ely ER, Verheyen KLP, Wood JLN (2004) Fractures andtendon injuries in National Hunt horses in training inthe UK: a pilot study. Equine Veterinary Journal, 36,365e367.

Emerson C, Morrissey D, Perry M, Jalan R (2010) Ultra-sonographically detected changes in Achilles tendonsand self reported symptoms in elite gymnasts comparedwith controlse an observational study.Manual Therapy,15, 37e42.

Eriksen HA, Pajala A, Leppilahti J, Risteli J (2002) In-creased content of type III collagen at the rupture siteof human Achilles tendon. Journal of Orthopaedic Research,20, 1352e1357.

Farris DJ, Trewartha G, McGuigan MP (2011) Couldintra-tendinous hyperthermia during running explainchronic injury of the human Achilles tendon? Journal

of Biomechanics, 44, 822e826.Flick J, Devkota A, Tsuzaki M, Almekinders L,

Weinhold P (2006) Cyclic loading alters biomechanicalproperties and secretion of PGE2 and NO from tendonexplants. Clinical Biomechanics (Bristol), 21, 99e106.

Garcia-Dorado D, Rodriguez-Sinovas A, Ruiz-Meana M(2004) Gap junction-mediated spread of cell injuryand death during myocardial ischaemia-reperfusion.Cardiovascular Research, 61, 386e401.

Gardner K, Arnoczky SP, Lavagnino M (2011) Effect ofin vitro stress-deprivation and cyclic loading on thelength of tendon cell cilia in situ. Journal of OrthopaedicResearch, 29, 582e587.

Garrity MM, Burgart LJ, Riehle DL, Hill EM, Sebo TJet al. (2003) Identifying and quantifying apoptosis: nav-igating technical pitfalls.Modern Pathology, 16, 389e394.

Giepmans BNG (2006) Role of connexin43-interactingproteins at gap junctions. Advances in Cardiology, 42,41e56.

Godwin EE, Young NJ, Dudhia J, Beamish IC,Smith RKW (2012) Implantation of bone marrow-derived mesenchymal stem cells demonstrates improvedoutcome in horses with overstrain injury of the superfi-cial digital flexor tendon. Equine Veterinary Journal, 44,25e32.

Goodman SA, May SA, Heinegard D, Smith RK (2004)Tenocyte response to cyclical strain and transforminggrowth factor beta is dependent upon age and site of or-igin. Biorheology, 41, 613e628.

Goodship AE, Birch HL, Wilson AM (1994) The pathobi-ology and repair of tendon and ligament injury. Veteri-nary Clinics of North America: Equine Practice, 10, 323e349.

Haraldsson BT, Aagaard P, Qvortup K, Bojsen-Moller J,Krogsgaard M et al. (2008) Lateral force transmissionbetween human tendon fascicles. Matrix Biology, 27,86e95.

Hosaka Y, Ozoe S, Kirisawa R, Ueda H, Takehana K et al.(2006) Effect of heat on synthesis of gelatinases and pro-inflammatory cytokines in equine tendinocytes. Biomed-ical Research, 27, 233e241.

Hosaka Y, Teraoka H, Yamamoto E, Ueda H,Takehana K (2005) Mechanism of cell death in in-flamed superficial digital flexor tendon in the horse.Journal of Comparative Pathology, 132, 51e58.

Huang C, Han X, Li X, Lam E, Peng W et al. (2012) Crit-ical role of connexin 43 in secondary expansion of trau-matic spinal cord injury. Journal of Neuroscience, 32,3333e3338.

J€arvinenM, J�ozsa L, Kannus P, J€arvinen TL,KvistM et al.(1997) Histopathological findings in chronic tendon dis-orders. Scandinavian Journal of Medicine and Science in

Sports, 7, 86e95.Jha BS, Ayres CE, Bowman JR, Telemeco TA, Sell SA et al.

(2011) Electrospun collagen: a tissue engineering scaf-fold with unique functional properties in a wide varietyof applications. Journal of Nanomaterials, doi:10.1155/2011/348268. Article ID 348268, 15 pages.

Jones GC, Corps AN, Pennington CJ, Clark IM,Edwards DR et al. (2006) Expression profiling of metal-loproteinases and tissue inhibitors of metalloproteinasesin normal and degenerate human Achilles tendon.Arthritis and Rheumatism, 54, 832e842.

J�ozsa L, B�alint BJ, R�effy A, Demel Z (1982) Hypoxic alter-ations of tenocytes in degenerative tendinopathy. Ar-chives of Orthopaedic and Traumatic Surgery, 99, 243e246.

Kane TP, Ismail M, Calder JD (2008) Topical glyceryl tri-nitrate and non-insertional Achilles tendinopathy:a clinical and cellular investigation. American Journal of

Sports Medicine, 36, 1160e1163.Kannus P, J�ozsa L (1991) Histopathological changes pre-

ceding spontaneous rupture of a tendon. Journal of

Bone and Joint Surgery, 73-A, 1507e1525.Kasashima Y, Smith RK, Birch HL, Takahashi T,

Kusano K et al. (2002) Exercise-induced tendon hyper-trophy: cross-sectional area changes during growth are


influenced by exercise. Equine Veterinary Journal,34(Suppl.), 264e268.

Kasashima Y, Takahashi T, Smith RK, Goodship AE,Kuwano A et al. (2004) Prevalence of superficial digitalflexor tendonitis and suspensory desmitis in Japanesethoroughbred flat racehorses in 1999. Equine VeterinaryJournal, 36, 346e350.

Kazimo�glu C, B€ol€ukbasi S, Kanatli U, Senk€oylu A,Altun NS et al. (2003) A novel biodegradable PCLfilm for tendon reconstruction: Achilles tendon defectmodel in rats. International Journal of Artificial Organs,26, 804e812.

Ker RF (2002) The implications of adaptable fatigue qual-ity of tendons for their construction, repair and function.Comparative Biochemistry and Physiology. Part A eMolecular

and Integrative Physiology, 133, 987e1000.Knobloch K, Yoon U, Vogt PM (2008) Acute and overuse

injuries correlated to hours of training in master runningathletes. Foot and Ankle International, 29, 671e676.

Kujala UM, Sarna S, Kaprio J (2005) Cumulative inci-dence of Achilles tendon rupture and tendinopathy inmale former elite athletes. Clinical Journal of Sport Medi-

cine, 15, 133e135.Lam KH, Parkin TD, Riggs CM, Morgan KL (2007) De-

scriptive analysis of retirement of thoroughbred race-horses due to tendon injuries. Equine Veterinary Journal,39, 143e148.

Langberg H, Skovgaard D, Karamouzis M, B€ulow J,Kjaer M (1999) Metabolism and inflammatory media-tors in the peritendinous space measured by microdialy-sis during intermittent isometric exercise in humans.Journal of Physiology, 515, 919e927.

LavagninoM, Arnoczky SP (2005) In vitro alterations in cy-toskeletal tensional homeostasis control gene expressionin tendon cells. Journal of Orthopaedic Research, 23,1211e1218.

Lavagnino M, Arnoczky SP, Egerbacher M, Gardner KL,Burns ME (2006) Isolated fibrillar damage in tendonsstimulates local collagenase mRNA expression and pro-tein synthesis. Journal of Biomechanics, 39, 2355e2362.

LavagninoM,Arnoczky SP,GardnerK (2011) In situ deflec-tion of tendon cellecilia in response to tensile loading: anin vitro study. Journal of Orthopaedic Research, 29, 925e930.

Lawson SE, Chateau H, Pourcelot P, Denoix JM, Crevier-Denoix N (2007) Effect of toe and heel elevation on cal-culated tendon strains in the horse and the influence ofthe proximal interphalangeal joint. Journal of Anatomy,210, 583e591.

Leadbetter WB (1992) Cell-matrix response in tendon in-jury. Clinics in Sports Medicine, 11, 533e578.

Lee P, Lin R, Moon J, Lee LP (2006) Microfluidic align-ment of collagen fibers for in vitro cell culture. BiomedicalMicrodevices, 8, 35e41.

Lian O, Scott A, Engebretsen L, Bahr R, Duronio V et al.(2007) Excessive apoptosis in patellar tendinopathy inathletes.American Journal of SportsMedicine, 35, 605e611.

Lochner FK, Milne DW, Mills EJ, Groom JJ (1980) In vivoand in vitro measurement of tendon strain in the horse.American Journal of Veterinary Research, 41, 1929e1937.

Lovati AB, Corradetti B, Lange Consiglio A, Recordati C,Bonacina E et al. (2011) Characterisation and differenti-ation of equine tendon-derived progenitor cells. Journalof Biological Regulators and Homeostatic Agents, 25(Suppl.2), S75eS84.

Lui PPY,Maffulli N, Rolf C, Smith RKW (2011)What arethe validated animal models for tendinopathy? Scandina-vian Journal of Medicine and Science in Sports, 21, 3e17.

Maeda E, Ye S, Wang W, Bader DL, Knight MM et al.(2012) Gap junction permeability between tenocyteswithin tendon fascicles is suppressed by tensile loading.Biomechanics and Modelling in Mechanobiology, 11,439e447.

Maffulli N, Ewen SW, Waterston SW, Reaper J, Barrass V(2000) Tenocytes from ruptured and tendinopathicAchilles tendons produce greater quantities of type IIIcollagen than those from normal Achilles tendons. Anin vitromodel of human tendon healing. American Journalof Sports Medicine, 28, 499e505.

Maffulli N,Waterston SW, Squair J, Reaper J, Douglas AS(1999) Changing incidence of Achilles tendon rupture inScotland: a 15-year study. Clinical Journal of Sport Medi-

cine, 9, 157e160.Magnusson SP, Qvortrup K, Larsen JO, Rosager S,

Hanson P et al. (2002) Collagen fibril size and crimpmorphology in ruptured and intact Achilles tendons.Matrix Biology, 21, 369e377.

Malvankar S, Khan WS (2011) Evolution of the Achillestendon: the athlete’s Achilles heel? Foot (Edinburgh),21, 193e197.

Marr CM, Love S, Boyd JS,McKellar Q (1993) Factors af-fecting the clinical outcome of injuries to the superficialdigital flexor tendon in National Hunt and point-to-point racehorses. Veterinary Record, 132, 476e479.

Matthews JA, Wnek GE, Simpson DG, Bowlin GL (2002)Electrospinning of collagen nanofibers. Biomacromole-

cules, 3, 232e238.Mazzocca AD, Chowaniec D, McCarthy MB, Beitzel K,

CoteMP et al. (2011) In vitro changes in human tenocytecultures obtained from proximal biceps tendon:multiplepassages result in changes in routine cell markers. KneeSurgery, Sports Traumatology and Arthroscopy. [Epublicationahead of print].

McNeilly CM, Banes AJ, Benjamin M, Ralphs JR (1996)Tendon cells in vivo form a three dimensional networkof cell processes linked by gap junctions. Journal of Anat-omy, 189, 593e600.

Miles CA, Wardale RJ, Birch HL, Bailey AJ (1994) Differ-ential scanning calorimetric studies of superficial digitalflexor tendon degeneration in the horse. Equine VeterinaryJournal, 26, 291e296.

Millar NL, Hueber AJ, Reilly JH, Xu Y, Fazzi UG et al.(2010) Inflammation is present in early human tendin-opathy. American Journal of Sports Medicine, 38,2085e2091.

Millar NL, Reilly JH, Kerr SC, Campbell AL, Little KJet al. (2012) Hypoxia: a critical regulator of early humantendinopathy. Annals of the Rheumatic Diseases, 71,302e310.


Mohyeldin A, Garz�on-Muvdi T, Qui~nones-Hinojosa A(2010)Oxygen in stem cell biology: a critical componentof the stem cell niche. Cell Stem Cell, 7, 150e161.

M€oller A, Astron M, Westlin N (1996) Increasing inci-dence of Achilles tendon rupture. Acta Orthopaedica Scan-dinavica, 67, 479e481.

Molloy A, Wood EV (2009) Complications of the treat-ment of Achilles tendon ruptures. Foot and Ankle Clinics,14, 745e759.

Mori R, Power KT, Wang CM, Martin P, Becker DL(2006) Acute downregulation of connexin43 at woundsites leads to a reduced inflammatory response, en-hanced keratinocyte proliferation and wound fibroblastmigration. Journal of Cell Science, 119, 5193e5203.

Murphy DJ, Nixon AJ (1997) Biochemical and site-specificeffects of insulin-like growth factor 1 on intrinsic teno-cyte activity in equine flexor tendons. American Journal

of Veterinary Research, 58, 103e109.Nyyss€onen T, L€uthje P, Kr€oger H (2008) The increasing

incidence and difference in sex distribution of Achillestendon rupture in Finland in 1987-1999. ScandinavianJournal of Surgery, 97, 272e275.

Oikawa M, Kasashima Y (2002) The Japanese experiencewith tendonitis in racehorses. Journal of Equine Science,13, 41e56.

O’Meara BO, Bladon B, Parkin TDH, Fraser B, Lischer CJ(2010) An investigation of the relationship between raceperformance and superficial digital flexor tendonitis inthe thoroughbred racehorse. Equine Veterinary Journal,42, 322e326.

Paavola M, Kannus P, J€arvinen TAH, Khan K, J�osza Let al. (2002) Achilles tendinopathy. Journal of Bone andJoint Surgery, 84, 2062e2076.

Patterson-Kane JC, Parry DA, Goodship AE, Firth EC(1997a) Exercise modifies the age-related change incrimp pattern in the core region of the equine superficialdigital flexor tendon. New Zealand Veterinary Journal, 45,135e139.

Patterson-Kane JC, Wilson AM, Firth EC, Parry DA,Goodship AE (1997b) Comparison of collagen fibrilpopulations in the superficial digital flexor tendons ofexercised and nonexercised thoroughbreds. Equine Veter-inary Journal, 29, 121e125.

Pearce CJ, Ismail M, Calder JD (2009) Is apoptosis thecause of noninsertional Achilles tendinopathy? AmericanJournal of Sports Medicine, 37, 2440e2444.

Perkins NR, Reid SW, Morris JS (2005a) Risk factors forinjury to the superficial digital flexor tendon and suspen-sory apparatus in thoroughbred racehorses in New Zea-land. New Zealand Veterinary Journal, 53, 184e192.

Perkins NR,Reid SW,Morris JS (2005b) Profiling theNewZealand thoroughbred racing industry. 2. Conditionsinterfering with training and racing. New Zealand Veter-

inary Journal, 53, 69e76.Pinchbeck GL, Clegg PD, Proudman CJ, Stirk A,

Morgan KL et al. (2004) Horse injuries and racing prac-tices in National Hunt racehorses in the UK: the resultsof a prospective cohort study. Veterinary Journal, 167,45e52.

Platt D, Wilson AM, Timbs A, Wright IM, Goodship AE(1994) Novel force transducer for the measurement oftendon force in vivo. Journal of Biomechanics, 27,1489e1493.

Pouzaud F, Bernard-Beaubois K, Thevenin M,Warnet JM, Hayem G et al. (2004) In vitro discrimina-tion of fluoroquinolones toxicity on tendon cells: in-volvement of oxidative stress. Journal of Pharmacology

and Experimental Therapeutics, 308, 394e402.Qi J, Chi L, Bynum D, Banes AJ (2011) Gap junctions in

IL-1b-mediated cell survival response to strain. Journalof Applied Physiology, 110, 1425e1431.

Ralphs JR, Benjamin M, Waggett AD, Russell DC,Messner K et al. (1998) Regional differences in cell shapeand gap junction expression in rat Achilles tendon: rela-tion to fibrocartilage differentiation. Journal of Anatomy,193, 215e222.

Ralphs JR, Waggett AD, Benjamin M (2002) Actin stressfibres and cellecell adhesion molecules in tendons: orga-nisation in vivo and response to mechanical loading oftendon cells in vitro. Matrix Biology, 21, 67e74.

Richardson LE, Dudhia J, Sadler CJ, Goodship AE,Smith RKW (2007) Extracellular matrix cues for mes-enchymal stem cell differentiation. Transactions of the Or-

thopaedic Research Society, 32, 472.Riley GP (2005) Gene expression and matrix turnover in

overused and damaged tendons. Scandinavian Journal of

Medicine and Science in Sports, 15, 241e251.Riley GP, Curry V, DeGroot J, van El B, Verzijl N et al.

(2002) Matrix metalloproteinase activities and their re-lationship with collagen remodelling in tendon pathol-ogy. Matrix Biology, 21, 185e195.

Riley GP, Harrall RL, Constant CR, Chard MD,Cawston TE et al. (1994a) Tendon degeneration andchronic shoulder pain: changes in the collagen composi-tion of the human rotator cuff tendons in rotator cufftendinitis. Annals of the Rheumatic Diseases, 53, 359e366.

Riley GP, Harrall RL, Constant CR, Chard MD,Cawston TE et al. (1994b) Glycosaminoglycans of hu-man rotator cuff tendons: changes with age and inchronic rotator cuff tendinitis.Annals of the Rheumatic Dis-

eases, 53, 367e376.Rolf CG, Fu BSC, Pau A, Wang W, Chan B (2001) In-

creased cell proliferation and associated expression ofPDGFRb causing hypercellularity in patellar tendino-sis. Rheumatology, 40, 256e261.

Rui YF, Lui PP, Lee YW, Chan KM (2012) Higher BMPreceptor expression and BMP-2-induced osteogenic dif-ferentiation in tendon-derived stem cells compared withbone-marrow-derived mesenchymal stem cells. Interna-tional Orthopaedics, 36, 1099e1107.

Saxena A, Ewen B, Maffulli N (2011) Rehabilitation of theoperatedAchilles tendon: parameters for predicting a re-turn to activity. Journal of Foot and Ankle Surgery, 50,37e40.

Sch€ocklmann HO, Lang S, Kralewski M, Hartner A,L€udke A et al. (2000) Distinct structural forms of typeI collagen modulate cell cycle regulatory proteins in me-sangial cells. Kidney International, 58, 1108e1120.


Schwarz R, Colarusso L, Doty P (1976) Maintenance ofdifferentiation in primary cultures of avian tendon cells.Experimental Cell Research, 102, 63e71.

Scott A, Khan KM, Heer J, Cook JL, Lian O et al. (2005)High strain mechanical loading rapidly induces tendonapoptosis: an ex vivo rat tibialis anterior model. BritishJournal of Sports Medicine, 39, e25.

Segretain D, Falk MM (2004) Regulation of connexin bio-synthesis, assembly, gap junction formation, and re-moval. Biochimica et Biophysica Acta, 1662, 3e21.

Sherr CJ, DePinho RA (2000) Cellular senescence: mitoticclock or culture shock? Cell, 102, 407e410.

SkutekM, van GriensvenM, Zeichen J, Brauer N, Bosch U(2001) Cyclic mechanical stretching enhances secretionof interleukin 6 in human tendon fibroblasts. Knee Sur-gery, Sports Traumatology and Arthroscopy, 9, 322e326.

Slaughter BV, Khurshid S, Fisher OZ, Khademhosseini A,Peppas NA (2009) Hydrogels in regenerative medicine.Advanced Materials, 21, 3307e3329.

Smith MM, Sakurai G, Smith SM, Young AA, Melrose Jet al. (2008) Modulation of aggrecan and ADAMTS ex-pression in ovine tendinopathy induced by alteredstrain. Arthritis and Rheumatism, 58, 1055e1066.

Smith RKW, Birch HL, Goodman S, Heineg�ard D,Goodship AE (2002) The influence of ageing and exer-cise on tendon growth and degeneration e hypothesesfor the initiation and prevention of strain-induced tendi-nopathies. Comparative Biochemistry and Physiology Part A,133, 1039e1050.

Smith RKW, Webbon PM (1996) The physiology of nor-mal tendon and ligament. In: Proceedings of the Dubai In-

ternational Equine Symposium ‘The Equine Athlete: Tendon,

Ligament and Soft Tissue Injuries’, NW Rantanen,ML Hauser, Eds., Veterinary Data, California, pp.55e81.

Sode J, Obel N, Hallas J, Lassen A (2007) Use of fluoroqui-nolone and risk of Achilles tendon rupture: a popula-tion-based cohort study. European Journal of Clinical

Pharmacology, 63, 499e503.Spalazzi JP, Doty SB, Moffat KL, Levine WN, Lu HH

(2006) Development of controlled matrix heterogeneityon a triphasic scaffold for orthopaedic interface tissue en-gineering. Tissue Engineering, 12, 3497e3508.

Stains JP, Lecanda F, Screen TJ, Towler DA, Civitelli R(2003) Gap junctional communication modulates genetranscription by altering the recruitment of Sp1 andSp3 to connexin-response elements in osteoblast pro-moters. Journal of Biological Chemistry, 278,24377e24387.

Stanley RL, Fleck RA, Becker DL, Ralphs JR, Patterson-Kane JC (2007) Gap junction protein expression andcellularity: comparison of immature and adult equinedigital tendons. Journal of Anatomy, 211, 325e334.

Stephens PR, Nunamaker DM, Butterweck DM (1989)Application of a Hall-effect transducer for measurementof tendon strains in horses. American Journal of VeterinaryResearch, 50, 1089e1095.

Stoll C, John T, Endres M, Rosen C, Kaps C et al. (2010)Extracellular matrix expression of human tenocytes in

three-dimensional aireliquid and PLGA cultures com-pared with tendon tissue: implications for tendon tissueengineering. Journal of Orthopaedic Research, 28,1170e1177.

Tallon C, Maffulli N, Ewen SW (2001) Ruptured Achillestendons are significantly more degenerated than tendi-nopathic tendons. Medicine and Science in Sports and Exer-

cise, 33, 1983e1990.Tanji K, Shimizu T, Satou T, Hashimoto S, Bonilla E

(1995) Gap junctions between fibroblasts in myotendon.Archives of Histology and Cytology, 58, 97e102.

Tempfer H, Wagner A, Gehwolf R, Lehner C, Tauber Met al. (2009) Perivascular cells of the supraspinatus ten-don express both tendon- and stem cell-related markers.Histochemistry and Cell Biology, 131, 733e741.

Thornton GM, Shao X, Chung M, Sciore P, Boorman RSet al. (2010) Changes in mechanical loading lead to ten-don specific alterations in MMP and TIMP expression:influence of stress deprivation and intermittent cyclic hy-drostatic compression on rat supraspinatus and Achillestendons. British Journal of Sports Medicine, 44, 698e703.

Thorpe CT, Streeter I, Pinchbeck GL, Goodship AE,Clegg PD et al. (2010) Aspartic acid racemization andcollagen degradation markers reveal an accumulationof damage in tendon collagen that is enhanced with ag-ing. Journal of Biological Chemistry, 285, 15674e15681.

Tian T, Lavagnino M, Gardner K, Arnoczky SP (2004)Hyperthermia increases the magnitude of caspase-3 ac-tivation and DNA fragmentation in tendon cells under-going cyclic strain: a potential intrinsic factor in theetiology of tendon overuse injuries.Transactions of the Or-

thopaedic Research Society, 29, 881.Uchida H, Tohyama H, Nagashima K, Ohba Y,

Matsumoto H et al. (2005) Stress deprivation simulta-neously induces over-expression of interleukin-1beta, tu-mour necrosis factor-alpha, and transforming growthfactor-beta in fibroblasts and mechanical deteriorationof the tissue in the patellar tendon. Journal of Biomechan-ics, 38, 791e798.

Waggett AD, BenjaminM, Ralphs JR (2006) Connexin 32and 43 gap junctions differentially modulate tenocyteresponse to cyclic mechanical load. European Journal of

Cell Biology, 85, 1145e1154.Wall ME, Banes AJ (2005) Early responses to mechanical

load in tendon: role for calcium signalling, gap junctionsand intercellular communication. Journal of Musculoskel-

etal and Neuronal Interactions, 5, 70e84.Wall ME, Otey C, Qi J, Banes AJ (2007a) Connexin 43 is

localized with actin in tenocytes. Cell Motility and the Cy-

toskeleton, 64, 121e130.Wall ME, Weinhold PS, Siu T, Brown TD, Banes AJ

(2007b) Comparison of cellular strain with applied sub-strate strain in vitro. Journal of Biomechanics, 40, 173e181.

Wang JHC (2006) Mechanobiology of tendon. Journal ofBiomechanics, 39, 1563e1582.

Wang JH, Li Z, Yang G, Khan M (2004) Repetitivelystretched tendon fibroblasts produce inflammatory me-diators. Clinical Orthopaedics and Related Research, 422,243e250.


Wang W, Tang X, Zhang J, Yan X, Ma Y (2010) Com-plete stress shielding of the Achilles tendon: ultrastruc-ture and level of interleukin-1 and TGF-b. Orthopedics,33, 810.

Williams IF, McCullagh KG, Silver IA (1984) The distri-bution of types I and III collagen and fibronectin in thehealing equine tendon. Connective Tissue Research, 12,211e227.

Williams RB, Harkins LS, Hammond CJ, Wood JL (2001)Racehorse injuries, clinical problems and fatalities re-corded on British racecourses from flat racing and Na-tional Hunt racing during 1996, 1997 and 1998. EquineVeterinary Journal, 33, 478e486.

Wilmink J, Wilson AM, Goodship AE (1991) Functionalsignificance of morphology and micromechanics of col-lagen fibres in relation to partial rupture of the superfi-cial digital flexor tendon in racehorses. Research in

Veterinary Science, 53, 354e359.Wilson AM, Goodship AE (1991)Mechanical properties of

the equine superficial digital flexor tendon. In: 7th Meet-

ing of the European Society of Biomechanics, Aarhus, Denmark,pp. p. B15.

Wilson AM, Goodship AE (1994) Exercise-induced hyper-thermia as a possible mechanism for tendon degenera-tion. Journal of Biomechanics, 27, 899e905.

Xu B, Chow MJ, Zhang Y (2011) Experimental and mod-eling study of collagen scaffolds with the effects of cross-linking and fiber alignment. International Journal of

Biomaterials, 2011, doi:10.1155/2011/172389. ArticleID 172389, 12 pages, 2011.

Young NJ, Becker DL, Fleck RA, Goodship AE, Patterson-Kane JC (2009)Maturational alterations in gap junctionexpression and associated collagen synthesis in responseto tendon function. Matrix Biology, 28, 311e323.

Yao L, Bestwick CS, Maffulli N, Aspden RM (2006) Phe-notypic drift in human tenocyte culture.Tissue Engineer-

ing, 12, 1843e1849.

Yuan J,Murrell GA,Wei AQ,WangMX (2002) Apoptosisin rotator cuff tendonopathy. Journal of Orthopaedic Re-search, 20, 1372e1379.

Yoon JH, Brooks RL, Khan A, Pan H, Bryan J et al. (2004)The effect of enrofloxacin on cell proliferation and pro-teoglycans in horse tendon cells. Cell Biology and Toxicol-

ogy, 20, 41e54.Zafar MS, Mahmood A, Maffulli N (2009) Basic science

and clinical aspects of Achilles tendinopathy. SportsMed-

icine and Arthroscopy Review, 17, 190e197.Zeugolis DI, Khew ST, Yew ESW, Ekaputra AK,

Tong YW et al. (2008) Electro-spinning of pure collagennanofibrese just an expensive way to make gelatin? Bio-materials, 29, 2293e2305.

Zhang J, Pan T, Liu Y, Wang JHC (2010a) Mouse tread-mill running enhances tendons by expanding the pool oftendon stem cells (TSCs) and TSC-related cellular pro-duction of collagen. Journal of Orthopaedic Research, 28,1178e1183.

ZhangY,Wang B, ZhangWJ, ZhouG,CaoY et al. (2010b)Enhanced proliferation capacity of porcine tenocytes inlow O2 tension culture. Biotechnology Letters, 32,181e187.

Zhang J, Wang JH (2010) Mechanobiological response oftendon stem cells: implications of tendon homeostasisand pathogenesis of tendinopathy. Journal of OrthopaedicResearch, 28, 639e643.

Zhou Z, Akinbiyi T, Xu L, Ramcharan M, Leong DJ et al.(2010) Tendon-derived stem/progenitor cell aging: de-fective self-renewal and altered fate. Aging Cell, 9,911e915.

½ R

A

eceived, April 26th, 2012

ccepted, May 14th, 2012
�

The Pathogenesis of Tendon Microdamage in Athletes: the ... · The equine superﬁcial digital...

Documents

Transcript of The Pathogenesis of Tendon Microdamage in Athletes: the ... · The equine superﬁcial digital...