Different Achilles Tendon Pathologies Show Distinct Histological and Molecular Characteristics

17 Pages·2017·3.75 MB·English

International Journal o f
Molecular Sciences
Article
Different Achilles Tendon Pathologies Show Distinct
Histological and Molecular Characteristics
FrankaKlatte-Schulz1,2,*,† ID,SusannMinkwitz1,2,†,AyshaSchmock1,NicoleBormann1,2,
AlperKurtoglu1,SerafeimTsitsilonis1,SebastianManegold1,‡ andBrittWildemann1,2,‡ ID
1 JuliusWolffInstitute,CenterforMusculoskeletalSurgery,Charité—UniversitätsmedizinBerlin,
CorporateMemberofFreieUniversitätBerlin,Humboldt-UniversitätzuBerlin,andBerlinInstituteof
Health,13353Berlin,Germany;[email protected](S.Mi.);[email protected](A.S.);
[email protected](N.B.);[email protected](A.K.);
[email protected](S.T.);[email protected](S.Ma.);
[email protected](B.W.)
2 Berlin-BrandenburgCenterforRegenerativeTherapies,Charité—UniversitätsmedizinBerlin,
13353Berlin,Germany
* Correspondence:[email protected];Tel.:[phone]
† Theseauthorscontributedequallytothiswork.
‡ Theseauthorscontributedequallytothiswork.
Received:10January2018;Accepted:26January2018;Published:30January2018
Abstract: Reasons for the development of chronic tendon pathologies are still under debate and
more basic knowledge is needed about the different diseases. The aim of the present study was
therefore to characterize different acute and chronic Achilles tendon disorders. Achilles tendon
samplesfrompatientswithchronictendinopathy(n=7),chronicruptures(n=6),acuteruptures
(n=13),andintacttendons(n=4)wereanalyzed. Thehistologicalscoreinvestigatingpathological
changeswassignificantlyincreasedintendinopathyandchronicrupturescomparedtoacuteruptures.
Inflammatoryinfiltrationwasdetectedbyimmunohistochemistryinalltendonpathologygroups,
butwassignificantlylowerintendinopathycomparedtochronicruptures. Quantitativereal-time
PCR(qRT-PCR)analysisrevealedsignificantlyalteredexpressionofgenesrelatedtocollagensand
matrixmodeling/remodeling(matrixmetalloproteinases,tissueinhibitorsofmetalloproteinases)in
tendinopathyandchronicrupturescomparedtointacttendonsand/oracuteruptures. Inallthree
tendonpathologygroupsmarkersofinflammation(interleukin(IL)1β,tumornecrosisfactor α,IL6,
IL10,IL33,solubleST2,transforminggrowthfactorβ1,cyclooxygenase2),inflammatorycells(clusterof
differentaition (CD) 3, CD68, CD80, CD206), fat metabolism (fatty acid binding protein 4, peroxisome
proliferator-activatedreceptorγ,CCAAT/enhancer-bindingproteinα,adiponectin),andinnervation(protein
geneproduct9.5,growthassociatedprotein43,macrophagemigrationinhibitoryfactor)weredetectable,
butonlyinacuterupturessignificantlyregulatedcomparedtointacttendons. Thestudygivesan
insightintostructuralandmolecularchangesofpathologicalprocessesintendonsandmightbeused
toidentifytargetsforfuturetherapyoftendonpathologies.
Keywords: chronic tendon pathologies; acute rupture; human; tendon structure; matrix;
modeling/remodeling;inflammation;fat;innervation
1. Introduction
Achillestendonpathologiescaneitherbeduetoanacuteinjury, mostlyoccurringinrelation
to sports, or have a chronic background and are called tendinopathies. Achilles tendinopathy is
oneofthemostfrequenttendonpathologiescausedbyoveruseoroverloadstresses,whichleadto
repetitivemicro-traumata[1]. Itischaracterizedbyswelling,painandreducedfunction[2]andwas
Int.J.Mol.Sci. 2018,19,404;doi:10.3390/ijms19020404 www.mdpi.com/journal/ijms
Int.J.Mol.Sci. 2018,19,404 2of17
supposedtorepresentafailedhealingresponseofthetendon[3]. Characteristicsoftendinopathic
tendons are excessive proliferation of tenocytes and changes in tenocyte morphology, disruption
of collagen fibers, increase in non-collagenous matrix such as glycosaminoglycans (GAG) and fat
tissue as well as neovascularization [4–9]. Furthermore, it was reported that alterations in the
expressionofmatrixmetalloproteinases(MMPs)andtheirnaturalantagonists,thetissueinhibitorsof
metalloproteinases(TIMPs),whichregulatethemodelingandremodelinginthetendons,arelinkedto
tendinopathy[10–13]. Thepresenceofimmunecellsandinflammatoryprocessesintendinopathyis
stillunderdebate[14–17]. Severalstudiesusingclassicalhistologicalevaluationsnegatethepresence
of immune cells in chronic tendinopathy [2,9,18]. On the other hand, more recent studies found
inflammatoryinfiltrationofBcells,Tcells,macrophages,mastcellsand/ornaturalkillercells[19–24].
Furthermore,itwashypothesizedthatinflammationisonlyrelevantintheinitiationoftendinopathy
ratherthantheprogressionofthediseaseprocess[25]. Itwasshownthatchronicpainfultendonsare
associatedwithanincreaseinneuromediatorsliketheneurotransmitterglutamateandglutamineric
receptors, substance P and protein gene product 9.5 (PGP9.5) [26–29], as well as an increase in
inflammatorymarkerssuchasclusterofdifferentiation(CD)206andCD45[27]. Additionally,itwas
shownthatthepro-inflammatorycytokinescyclooxygenase-2(COX2)andinterleukin-6(IL6)were
upregulatedinpainfulversusasymptomaticAchillestendons[30].
Taken together it seems obvious that tendinopathy is a multifactorial process and more
basic knowledge is still needed to better understand the development and progression of the
disease. Therefore,thepresentapproachaimedtocharacterizedifferenttendonpathologies,suchas
tendinopathyandchronicAchillestendonrupturescomparedtoacuterupturesandintacttendons
regarding molecular and histological alterations in matrix production, modeling and remodeling,
adipogenicdifferentiation,aswellasinflammationandinnervation.
2. Results
2.1. HistologicalAnalysis
Theintactcadaverictendonsshowedcompactcollagenfiberbundleswithlongflattenocytes
(Figure1)andwereallscoredwithzeropoints. Thetendinopathyandchronicrupturegroupsshowed
a significantly increased histological score compared to the acute ruptured tendons (Figure 2A).
Looking in detail, the tendon architecture was fully disturbed in all acute ruptured tendons,
which was significantly higher scored compared to the tendinopathic tendons (Figures 1 and 2B).
Theamountofalignedcollagendidnotsignificantlychangebetweenthegroups(Figure2C).Inthe
tendinopathyandchronicruptures,ahighGAGcontentwasdetected(Figure1). Thedifferencewas
significant in the tendinopathy group compared to the acute ruptures (Figure 2D). Fat infiltration
was seen in four tendinopathy tendons and in two chronic ruptures and was significantly higher
in the tendinopathic tendons compared to the acute ruptures, where no fat tissue was visible
(Figures1and2E). The tendinopathy and chronic ruptures were histologically characterized by
a tightly packed structure with significantly higher vascularity and hypercellularity. In acute
rupturedtendons,vesselformationwaslowandmostlyconcentratedatthetendonedges,whereas
inintacttendonsvesselswereonlypresentbetweenthecollagenfiberbundles(Figures1and2F,G).
Asignificantlyloweramountofinflammatorycells(CD45+)wasfoundinthetendinopathygroup
compared to the chronic ruptures. High amounts of inflammatory cells were also found in acute
rupturedtendons(Figure2H).TheCD45+cellswereprimarilyidentifiedasmonocytes/macrophages
(CD68+)(Figure3).
Int.J.Mol.Sci. 2018,19,404 3of17
Int. J. Mol. Sci. 2018, 19, x 3 of 17
Int. J. Mol. Sci. 2018, 19, x 3 of 17

Figure 1. Exemplary histologies of intact tendons, tendinopathic tendons, chronic ruptures and acute
Figure1.Exemplaryhistologiesofintacttendons,tendinopathictendons,chronicrupturesandacute
Friguuprteu r1e.s E uxseimngp lMaroyv haits tPoelongtaiecsh orof mineta (cMt tPe)n dstoaninsi, ntegn (dteinnodpoant haircc theintedcotunrse, ,c hgrlyocnoics armupintuorgelsy caannd (aGcuAtGe )
rupturesusingMovatPentachrome(MP)staining(tendonarchitecture,glycosaminoglycan(GAG)and
ruapndtu fraets t iusssiuneg ( lMipoidv avta cPueonlteasc)h),r αo msme o(oMthP )m sutasicnlein agc t(itne n(αdSoMn Aar)c shtaitiencitnugr e(v, agslcyucloasraitmy)in aongdl yHceamn a(GtoAxyGli)n
fattissue(lipidvacuoles)),αsmoothmuscleactin(αSMA)staining(vascularity)andHematoxylin
anEdos fiant t(iHssEu)e s(tlaipinidin vga c(uceollleusl)a)r, iαty s)m. Fo:o ftaht m tiusssuclee, aVct: inv e(αssSeMls,A C) :s tcaeilnli nclgu (svtearssc.u Tlahreit yM) aonvdat H Peemntaatcohxryolimn e
Eosin(HE)staining(cellularity). F:fattissue,V:vessels,C:cellclusters. TheMovatPentachrome
Esotsaiinn i(nHgE u)s setda isntainings (ccoellllauglaerni tiyn) .y eFl:l ofawt ttois bsuroew, Vn:, mveastsuerles ,c Col:l acgeelln c ilnu srteedr s[.3 1T]h, eG AMGov garto uPnendt asuchbrsotamnec e
stainingusedstainscollageninyellowtobrown,maturecollageninred[31],GAGgroundsubstance
stiani ntuinrqgu uosiesd, c setlali nnus ccloelil aing ebnlu ine tyoe lblloawck t oa nbdro cwytno,p mlaastmur set caoinllsa rgeednd iins hre. d [31], GAG ground substance
inturquois,cellnucleiinbluetoblackandcytoplasmstainsreddish.
in turquois, cell nuclei in blue to black and cytoplasm stains reddish.

Figure2.Cont.
Int.J.Mol.Sci. 2018,19,404 4of17
Int. J. Mol. Sci. 2018, 19, x 4 of 17
Int. J. Mol. Sci. 2018, 19, x 4 of 17

Figure 2. Evaluation of histological score: (A) Total histological score reaching 18 points as maximum,
Figure2.Evaluationofhistologicalscore:(A)Totalhistologicalscorereaching18pointsasmaximum,
wFihgiucrhe w2. oEuvladl uiantdioicna otef hai shtoiglohglyic adle sgceonreer: a(Ate)d T toetnadl hoins.t o(lBo)g iTceanl dscoonr ea rrecahcihteicntgu 1re8 paso iinntds iacsa mtoar xfiomr utmhe,
which would indicate a highly degenerated tendon. (B) Tendon architecture as indicator for the
dwihstiuchrb wanocuel do f inthdei ctaetned ao nh isgtrhulyc tudreeg e(0n–e3r apteodin ttes)n. d(oCn) . A(Bm)o Tuennt doofn a laigrcnheidte cctoullraeg aens iqnudaincatitfoier df ours itnhge
disturbanceofthetendonstructure(0–3points). (C)Amountofalignedcollagenquantifiedusing
IdmisatugerbJ a(0n–c1e0 o0f% t)h. e( Dte)n GdAonG sctorunctetunrte ( 0(–03– 3p poionintst,s M). P(C s)t aAinminogu)n. t( Eo)f Faalitg tnisesdu ec oqlulaagnetnif iqeuda unstiinfiged I muasignegJ
ImageJ(0–100%).(D)GAGcontent(0–3points,MPstaining).(E)FattissuequantifiedusingImageJ
(I0m–a1g0e0J% (,0 –M1P00 s%ta)i.n (iDng) )G. A(FG) Ccoelnlutelnart i(t0y– (30 –p3o ipnotsin, tMs,P H sEta sintainingi)n. g()E. )( GFa)t Vtiassscuuel aqruitayn t(i0f–ie3d p uosiinntgs, IαmSaMgeAJ
(0–100%,MPstaining). (F)Cellularity(0–3points,HEstaining). (G)Vascularity(0–3points,αSMA
s(0ta–i1n0i0n%g),. M(HP) sItnafilnaimngm).a t(oFr) yC ceellllusl a(nriotyt i(n0c–l3u dpeodin itns , hHisEto sltoagiincianlg s)c. o(rGe), CVDas4c5u lsatariitnyi n(0g–).3 D paotian tasr,e α gSiMveAn
staining).(H)Inflammatorycells(notincludedinhistologicalscore,CD45staining).Dataaregivenas
astsa iinndinivgi)d. u(Hal) dInoftl pamlomts awtoitrhy mceeldlsi a(nn oatn idn cinlutdereqdu ianr thilies troalnoggeic. al score, CD45 staining). Data are given
individualdotplotswithmedianandinterquartilerange.
as individual dot plots with median and interquartile range.

Figure 3. Exemplary images of inflammatory cell infiltration visualized by immunohistological CD45
Fig(Fuliergueuc3roe.c y3Et.x eEesxm) eampnldpa lrCayDryi6m i8ma (gmaegosensoo ofcfiy nitneflfsal/ammmammcaraottoporrhyya gcceeelslll) iisnntfafiiillnttriranatgtiio ionnn tvevinsiusduainaliolzipezdaetd bhyibc yi tmeinmmdmuonnuosnh, ocishhtrioosltnoogilcio crgauilpc CatuDlrC4e5sD 45
(leua(lcneoduc cayoctceuystte)e asrun) apdntCudr DCes6D.8 6(8m (monooncoycytetse/s/mmaaccrroopphhaaggeess)) ssttaaiinniinngg inin tetnenddinionpoaptahtihc itcentednodnosn, csh,rcohnriocn riucprtuupretus res
and acute ruptures.
andacuteruptures.
IInntt.. JJ.. MMooll.. SSccii.. 2021081,8 1,91,9 x, 404 5 5ofo 1f71 7
2.2. Gene Expression
2.2. GeneExpression
Using quantitative real-time PCR (qRT-PCR), the collagens Col1A1, Col3A1 and Col5A2 were
foundU stion gbeq usiagnntiiftiactaivnetlrye aulp-triemgeuPlaCteRd( qinR Tth-PeC teRn),dtihneocpoaltlhagieesn asnCdo lc1hAr1o,nCico l3ruAp1taunredsC coolm5Ap2awreedr etofo tuhned
tinotbacet scigonnitfirocla natnldy uthper eagcuultaet erdupintutrheed tetennddinoonps a(tFhiigeusraen 4dAc–hCr)o.n Ticheru Cpotul1rAes1/cCooml3pAa1r erdattioo twhaesi nntoatc t
creognutrloalteadn.d theacuterupturedtendons(Figure4A–C).TheCol1A1/Col3A1ratiowasnotregulated.
WWhheenn llooookkiinngg aat tmmatartirxi xdedgergardaidnign egneznyzmyems,e tsh,et hgeengee enxepreexspsrioesns oiof nthoef ctohlleagceonllaasgee MnaMsePM1 wMaPs 1
wsigansisfiicgannitfilyca enltelyvaeteledv ainte dthein atchuetea cruutpeturruepdt utreenddotenns dcoonmspcaormedp atore dallt ootahlelro gthroerupgsr,o uwphse,rweahs etrheea s
tehxepreexspsiroenss ioofn tohfe thcoellcaoglelangaesne asMeMMPM13P 1w3aws assigsnigifniciafinctalnyt liyncinrecarseeadse dini nthteh etetnenddinionpoaptahtyh yggrorouupp
ccoommppaarreedd ttoo tthhee iinnttaacctt ssaammpplleess ((FFiigguurree 44DD,,HH)).. TThhee ggeellaattiinnaassee MMMMPP22 wwaass ssigignnifiificcaanntltyly uuppreregguulalatetded
iinn tthhee tteennddiinnooppaatthhyy aanndd cchhrroonniicc rruuppttuurree ggrroouupp ccoommppaarreedd ttoo tthhee ininttaacct taanndd aaccuutete rruupptutureredd tetennddoonns s
((FFiigguurree 44EE)),, bbuutt MMMMPP99 wwasa snonto atffaefcfetecdte. dT.heT hexeperxepssrieosns ioofn thoef stthreomsterolymsienleyss iMneMsPM3 ManPd3 ManMdPM10M wPa1s 0
wsigansisfiigcannifitlcya ndtelcyredaescerdea isne dthine ttheendteinnodpinaothpya tahnyda ncdhrcohnrioc nriucprutuprteu rgeroguropus pcsocmopmapreadre dtot othteh eaacucutet e
rruuppttuurreedd tteennddoonnss ((FFiigguurree 44FF,,GG))..
TThhee eexxpprreessssiioonn ooff tthhee MMMMPP iinnhhiibbiittoorrss wwaass ddiiffffeerreennttiiaallllyy rreegguullaatteedd.. TTIMIMPP11 eexxppreresssisoionn wwasa s
ssiiggnniifificcaannttllyy hhigighheerra nanddT ITMIMP3Pa3n adnTdI MTPIM4ePx4p erexspsrioesnssioignn isfiigcnaniftilcyanlotwlye rloiwntehre ianc uthtee raucputtuer erduptetnudreodn s
cteonmdpoanrse dcotomtphearoetdh etrog rtohuep so.tThIeMr Pg2roeuxpprse. ssTiIoMnPw2a sesxipgrneisfisciaonnt lywealse vsaitgendifiincatnhtelyte nedleinvoatpeadt hying rtohuep
cteonmdpinaorepdatthoyt hgeroiunpta ccotmtenpdaroends taon tdhea cinuttaecrtu tpentudroenss( Faingdu raecu4tIe– Lru).ptures (Figure 4I–L).

Figure4.Cont.
Int.J.Mol.Sci. 2018,19,404 6of17
Int. J. Mol. Sci. 2018, 19, x 6 of 17

Figure 4. Relative gene expression of collagens (A–C), MMPs (D–H) and TIMPs (I–L) in tendinopathic
Figure4.Relativegeneexpressionofcollagens(A–C),MMPs(D–H)andTIMPs(I–L)intendinopathic
tendons, chronic ruptures and acute ruptures given as fold to intact tendon tissue (horizontal line,
tendons,chronicrupturesandacuterupturesgivenasfoldtointacttendontissue(horizontalline,
mean value = 1). qRT-PCR data were normalized to the expression of the house keeping gene 18S
meanvalue=1).qRT-PCRdatawerenormalizedtotheexpressionofthehousekeepinggene18SrRNA
rRNA with efficiency correction and presented as fold change to the mean of the intact tendons.
withefficiencycorrectionandpresentedasfoldchangetothemeanoftheintacttendons.Resultsare
Results are shown as individual dot plots with median and interquartile range. Significant differences
shownasindividualdotplotswithmedianandinterquartilerange.Significantdifferencesbetween
between the intact tendons and the three tendon pathology groups are marked with a star (*) above
theintacttendonsandthethreetendonpathologygroupsaremarkedwithastar(*)abovetheintact
the intact tendon and the respective significant group (T: tendinopathy, C: chronic ruptures, A: acute
tendonandtherespectivesignificantgroup(T:tendinopathy,C:chronicruptures,A:acuteruptures)
ruptures) and the p-value. The vertical dashed line separates the intact group from the tendon
andthep-value.Theverticaldashedlineseparatestheintactgroupfromthetendonpathologygroups.
pathology groups. Significant differences between the three tendon pathology groups are marked
Significantdifferencesbetweenthethreetendonpathologygroupsaremarkedwithaspanningline
wabiothv eat hspeagnrnouinpgs lainnde tahbeoivned itvhied ugarolup-pvsa launeds .tphe≤ i0n.d05ivwidausacl opn-sviadleureesd. aps ≤s ta0t.0is5t icwaallsy csoignnsiifidcearendt. as
statistically significant.
Thetendonrelatedtranscriptionfactorscleraxis(SCX)wasnotsignificantlyalteredbetweenthe
The tendon related transcription factor scleraxis (SCX) was not significantly altered between the
fourgroups. Thetendonrelatedproteoglycantenomodulin(TNMD)wassignificantlyupregulatedin
four groups. The tendon related proteoglycan tenomodulin (TNMD) was significantly upregulated
chronicversusacuterupturedtendons(Figure5A).
in chronic versus acute ruptured tendons (Figure 5A).
Whenlookingatinflammatorymarkers,theIL33expressionwassignificantlydecreasedinthe
When looking at inflammatory markers, the IL33 expression was significantly decreased in the
acuterupturescomparedtotheintacttendonsandchronicruptures(Figure5B).TheIL33receptor
acute ruptures compared to the intact tendons and chronic ruptures (Figure 5B). The IL33 receptor
solubleST2(sST2)wasnotregulatedbetweenthegroups. Inthegroupofpro-andanti-inflammatory
soluble ST2 (sST2) was not regulated between the groups . In the group of pro- and anti-inflammatory
cytokinesonlyIL6andIL10showedsignificantupregulationintheacuterupturescomparedtothe
cytokines only IL6 and IL10 showed significant upregulation in the acute ruptures compared to the
intactand/ortendinopathictendons(Figure5C,D),whereastheexpressionofIL1β,tumornecrosis
intact and/or tendinopathic tendons (Figure 5C,D), whereas the expression of IL1β, tumor necrosis
factorα(TNFα)andtransforminggrowthfactorβ1(TGFβ1)didnotchangebetweenthegroups.
factor α (TNFα) and transforming growth factor β1 (TGFβ1) did not change between the groups .
The expression of the T-cell marker CD3 was not regulated, but the expression of the
The expression of the T-cell marker CD3 was not regulated, but the expression of the
mmoonnooccyyttee//mmaaccrroopphhaaggee mmaarkrkeer rCCDD6688 wwaas sssigignnifiificacanntltyly uuppreregguulalatetded inin ththe eaacucutet erurupptuturerse scocommpparaerded
ttoo tthhee iinnttaacctt tteennddoonnss ((FFiigguurree 55EE)).. TThhee mmaaccrroopphhaaggee ppoollaarriizzaattioionn mmaarrkkeersrs CCDD8800 aanndd CCDD220066 ddidid nnoot t
ssiiggnniifificcaannttllyy cchhaannggee..
TThhee eexxpprreessssiioonn ooff mmaaccrroopphhaaggee mmiiggrraattioionn ininhhibibitiotoryry fafacctotor r(M(MIFIF),) ,a annerevrve erergegenenereartaiotino nmmarakrekre, r,
aanndd PPGGPP99..55,, aa ggeenneerraall nneerrvvee mmaarrkkeerr,, wwaass ssiiggnniiffiiccaannttllyy iinnccrreeaasseedd inin aaccuutete rruupptutureres sccoommppaarerded toto ththe e
iinnttaacctt aanndd cchhrroonniicc rruuppttuurreess oorr tteennddiinnooppaatthhyy ggrroouupp, ,rreessppeecctitviveelyly (F(Figiguurere 55FF,G,G).) .TThhee eexxppreresssisoionn oof fththe e
nneerrvvee ggrroowwtthh mmaarrkkeerr ggrroowwtthh aassssoocciiaatteedd pprrootteeiinn 4433 ((GGAAPP4433)) wwaass nnoott rreegguulalatetedd. .TThhee eexxppreresssisoionn oof fththe e
ppaaiinn--aassssoocciiaatteedd ffaaccttoorrC COOXX22w wasassi gsnigifincifaincatlnytliyn cirnecarseeadseind thine athcue teacruutpet urruepstcuormesp acoremdptaortehde tcoh rtohnei c
rcuhprotunrice sru(Fpitguurrees 5(FHig).ure 5H).
WWhheenn llooookkiinngg aatt tthhee ffaatt mmeettaabboolliissmm,, ssiiggnniiffiiccaannttllyy lolowweerr eexxpprreessssioionn oof ffafatttyty aacicdid bbininddiningg pprorotetienin
44 ((FFAABBPP44)),, ppeerrooxxiissoommee pprroolliiffeerraattoorr--aaccttiivvaatteedd rreecceeppttoorr γγ ((PPPPAARRγγ)),, aanndd CCCCAAAATT//eennhhaanncceer-rb-bininddiningg
pprrootteeiinn αα ((CCEEBBPPαα)) wwaass sseeeenn iinn aaccuuttee rruuppttuurreess ccoommppaarreedd ttoo tthhee iinnttaacctt tteennddoonnss. .OOnn ththe eotohthere rhahnadn,d ,
FFAABBPP44 aannddP PPAPRAγRγex pexrepsrseisosniown awssaisg nsiifigcnainfitclaynhtliyg hheirgihnecrh irno ncichrrounpitcu rreuspatnudretse nadnidn otpeantdhiincotpenatdhoinc s
wtehndenoncso mwpheanre cdotmopaacruetde trou pactuurtees r(uFpitguurrees 5(HFig,Iu).rTe h5eHe,Ix)p. rTehses ieoxnporefsasdioipno onfe acdtiinpo(AneDcItPinO (QA)DwIPaOsQno) t
swigans infiocta snitglyniafifcfeacnttelyd .affected.
Int.J.Mol.Sci. 2018,19,404 7of17
Int. J. Mol. Sci. 2018, 19, x 7 of 17

FigFuirgeu5re. R5.e lRaetliavteivgee gneeneex epxrpesrseisosnionof oTf NTMNMDD(A (A),)t,h tehein iflnfalmammmaatotoryryc cyytotokkinineessI ILL3333,, IILL66 aanndd IILL1100 ((BB––D),
D), the immune cell marker CD68 (E), the nerve markers MIF and PGP9.5 (F–G), the pain marker
theimmunecellmarkerCD68(E),thenervemarkersMIFandPGP9.5(F–G),thepainmarkerCOX2(H),
COX2 (H), and the fat markers FABP4, PPARγ and CEBPα (I–K) in tendinopathic tendons, chronic
andthefatmarkersFABP4,PPARγandCEBPα(I–K)intendinopathictendons,chronicrupturesand
ruptures and acute ruptures given as fold to intact tendon tissue (horizontal line, mean value = 1).
acuterupturesgivenasfoldtointacttendontissue(horizontalline,meanvalue=1).qRT-PCRdata
qRT-PCR data were normalized to the expression of the house keeping gene 18S rRNA with efficiency
werenormalizedtotheexpressionofthehousekeepinggene18SrRNAwithefficiencycorrection
correction and presented as fold change to the mean of the intact tendons. Results are shown as
andpresentedasfoldchangetothemeanoftheintacttendons.Resultsareshownasindividualdot
individual dot plots with median and interquartile range. Significant differences between the intact
plotswithmedianandinterquartilerange.Significantdifferencesbetweentheintacttendonsandthe
tendons and the three tendon pathology groups are marked with a star (*) above the intact tendon
threetendonpathologygroupsaremarkedwithastar(*)abovetheintacttendonandtherespective
and the respective significant group (T: tendinopathy, C: chronic ruptures, A: acute ruptures) and the
significantgroup(T:tendinopathy,C:chronicruptures,A:acuteruptures)andthep-value.Thevertical
p-value. The vertical dashed line separates the intact group from the tendon pathology groups.
dashedlineseparatestheintactgroupfromthetendonpathologygroups. Significantdifferences
Significant differences between the three tendon pathology groups are marked with a spanning line
betweenthethreetendonpathologygroupsaremarkedwithaspanninglineabovethegroupsandthe
above the groups and the individual p-values. p ≤ 0.05 was considered as statistically significant.
individualp-values.p≤0.05wasconsideredasstatisticallysignificant.
Int.J.Mol.Sci. 2018,19,404 8of17
3. Discussion
The present study aimed at investigating the structural, cellular, as well as molecular
characteristicsofdifferenttendonpathologies. Astendinopathyseemstobeamultifactorialprocess,
avarietyoffactorsandprocessesrelatedtotendonmatrix,modelingandremodeling,inflammation,
fatty infiltration and innervation were analyzed to better understand possible reasons for the
developmentandprogressionofdegenerativeAchillestendonpathologies.
The tendinopathy and chronic rupture groups showed a disturbed tendon structure and
were histologically characterized by a tightly packed extracellular matrix (ECM) with loss of
the hierarchical structure, high amounts of non-collagenous matrix such as GAG and fat, high
vascularityandhypercellularity,asitwasreportedpreviously[4–9]. Interestingly,despitethedifferent
pathomechanismsofthetwodegenerativetendonpathologies,whichisontheonehandtheoverused
tendonwithrepetitivemicro-traumata(tendinopathy)andontheotherhandthetendonswithoverseen
oroldmacro-trauma(chronicruptures),bothgroupsshowedsimilarcharacteristics. Onlythelower
amountofinflammatorycellinfiltration(CD45+)inthetendinopathygroupwassignificantlydifferent
compared to the chronic ruptures. Additionally, the tendinopathy group showed more distinct
differencesthanthechronicrupturescomparedtotheintactandacuterupturedtendons. Overall,
the acute ruptured tendons showed the strongest alterations compared to the intact tendons for
nearlyallanalyzedprocesses(modeling/remodeling,inflammation,immunecells,fattyinfiltration
andinnervation),whereasthetendinopathyandchronicrupturedtendonsonlyshowedsignificant
differencestotheintactcontrolsrelatedtocollagenandmatrixmodeling/remodeling.
The disturbed matrix metabolism as seen in the histology in the tendinopathy and chronic
ruptured tendons was displayed in the altered gene expression of the collagens as well as MMPs
andTIMPs. TheexpressionofCOL1A1,COL3A1andCOL5A2wassignificantlyelevatedcompared
to intact and acute ruptured tendons. Accordingly, this increase in collagen expression was also
reported by other authors [12,30,32]. An increase in type III and V collagen expression was often
reportedtoresultininferiorbiomechanicalproperties[33,34]. Furthermore,typeVcollagenvariants
wereassociatedwithchronicAchillestendinopathy[35,36]. Thelossofabilitytoformahierarchical
tendonstructureseemstobeafurtherimportantcharacteristicintendinopathy. TNMD—atendon
proteoglycanresponsibleforcollagenfibermaturation[37,38]—wassignificantlyupregulatedinthe
chronic rupture group compared to acute ruptures and not regulated in the tendinopathy group.
Non-regulationoffactorsregulatingfibrillogenesiswerealsoreportedpreviouslyfortendinopathy
versus healthy tendons [32], which partly fits with the present findings. IL33—a cytokine having
a pivotal role in innate and acquired immune responses—was described to be important in early
tendinopathy. HereitregulatesthetransitionofcollagentypeItotypeIIIsynthesisthroughitscognate
receptorST2,whichexistsinamembrane-boundaswellassolublecognateform(sST2)[39]. Presently,
IL33expressionwassignificantlydecreasedinacuterupturedversusintacttendonsbutsignificantly
increasedinchronicrupturescomparedtoacuteruptures. Thisissupportingtheprevioushypothesis
thatrecurrentmicro-traumataleadtoanincreaseinIL33expressionandthusmatrixdegradationand
inflammation,whichmightleadtoanimpairedhealingresponse[39]andpossiblytoalowertendon
strainandahighersusceptibilityforAchillestendonrupture.
Nexttothecollagens,matrixdegradingenzymeswerealsoregulatedinthetendinopathyand
chronic ruptured tendons. The expression of the gelatinase MMP2 was significantly increased in
bothgroups,whichisinaccordancetopreviousfindings[32]. Additionally,thecollagenaseMMP13
expressionwassignificantlyupregulatedinthetendinopathygroupcomparedtotheintacttendons.
The upregulation of the MMPs together with the mostly non-regulation of the TIMPs compared
to the intact tendons might lead to a possible MMP/TIMPs imbalance in the tendinopathy and
chronicrupturegroups. Suchanimbalancewashypothesizedtoleadtoafailedtendonremodeling
process[40–42]. Additionally,theexpressionofMMP3andMMP10,thestromelysines,whichhavean
importantregulatoryfunctiononotherMMPs[42],wassignificantlydecreasedinthetendinopathy
andchronicrupturegroupscomparedtotheacuterupturedtendons. AdecreaseofMMP-3expression
Int.J.Mol.Sci. 2018,19,404 9of17
wasalsoreportedbyothersandwashypothesizedtoleadtoanincorrectremodelingprocess[12,42].
PresentregulationsregardingMMPsandTIMPswerepreviouslyalsoshownfortenocytesfromaged
donorsandwithhigherdegenerativestatusofsupraspinatustendons[43]. Additionally,comparable
regulationswerefoundinacuterupturedtendons,whichweresurgicallytreatedatalatertime-point
(7–10days)postrupturecomparedtotendons,werethesurgerywasperformedatanearliertime-point
(2–4and5–6days)afterAchillestendonrupture[44]. Ashypothesizedpreviously,atalatertimepoint
thetendonsseemtoturnintoadegenerativedirection,whichcannowbeconfirmedwhencomparing
ittothepresenttendinopathyandchronicrupturegroups.
Interestingly,nosignificantchangesforthetendinopathyandchronicrupturegroupcomparedto
intacttendonscouldbeobservedfortheotherprocessessuchasinflammation,fattyinfiltrationand
innervation. Anexcessiveinflammatoryreactionwasnotpresentinthetendinopathygroupandwas
evenreduced,asseenbyaloweramountofinflammatorycellsinthetendinopathygroupcompared
tothechronicrupturesinthehistology. Inflammationispresentintendinopathyandchronicruptured
tendonsbutseemstobelesspronouncedcomparedtotheacuterupturedtendons,astheexpressionof
pro-andanti-inflammatorycytokinesaswellasinflammatorycellmarkersforT-cellsandmacrophages
wasnotdistinctlydifferentcomparedtotheexpressioninintacttendons.Thepresenceofinflammatory
cellinfiltrationintendinopathyandchronicrupturesconfirmsseveralpreviousstudies[19–24],butthe
role of inflammation in the disease is still unclear but might only be important in the initiation
of tendinopathies [25]. However, inflammation seems to be more important in acute ruptures,
asindicatedbytheincreasedexpressionofthecytokinesIL6andIL10andthemacrophagemarker
CD68. TheincreasedexpressionofIL6andCD68mightbelinkedtoeachother,asIL6wasreported
to promote the differentiation of monocytes into macrophages [45]. An increased IL6 expression
wasalsofoundbyothersinpainfulandrupturedversusasymptomatictendons[30]. Interestingly,
IL6−/− miceshowedinferiormechanicalproperties[46]. Duringacutetendonhealing,thetypical
processessuchasinflammation(pro-andanti-inflammatoryphase),proliferation,remodelingand
maturationproceed[47]. Thepresentdatasuggestthatintheacuterupturedtendons,whichwere
takenfourtosixdaysafterrupture,thehealingprocessisintheanti-inflammatoryphase,asboth
IL6andIL10haveanti-inflammatoryproperties[48]andthepro-inflammatorycytokinesIL1βand
TNFαwerenotregulated. Thisisfurthermoreunderlinedbyunregulatedmacrophagepolarization
(unchanged CD80, CD206 expression), as M1 macrophages have pro- and M2 macrophages have
anti-inflammatoryproperties[49]. However,macrophagepolarizationwaspreviouslydescribedto
beinvolvedintendinopathy,withM2macrophagesorCD206expressionisdominatinginchronic
tendon pathologies [24,50]. Additionally, we could show previously that tenocytes respond to an
inflammatoryenvironmentandinfluencemacrophagepolarization[51]. Presently,noregulationin
theexpressionofmacrophagepolarizationmarkerscouldbeobserved,eitherinacuterupturesnorin
tendinopathyorchronicrupturedtendons,whichmightbecausedbylimitationsintheusedmethods.
Aclearfattyinfiltrationwasdetectedinthehistologicalsectionsfortendinopathyandpartly
chronictendonruptures. However,nosignificantdifferencesbetweengeneexpressionlevelsofthe
adipogenicfactorsFABP4,PPARγandADIPOQweredetectedbetweentendinopathyandchronic
ruptures with intact tendons. The presence of adipogenic factors in intact tendons is not unusual,
asisolatedtenocytesofsupraspinatustendonsalsoexpressedadipogenicfactorswithoutadipogenic
stimulation [52–54]. Additionally, others found a reduced expression of markers for lipolysis and
ADIPOQandanincreaseinmarkersforfattyacidβ-oxidationinKargersfatpadinpatientswith
Achilles tendinopathy, which was discussed as being a paradox but might be caused by different
regulationofmRNAexpressionandactualenzymeactivity[55]. However,presentlytheincreasein
fattissueinthetendinopathyandchronicrupturedtendonscomparedtotheacuterupturedtendons
was confirmed by the gene expression data, with an increased FABP4 and PPARγ expression in
tendinopathyandchronicrupturedtendons. Itwaspreviouslyshownthatadiposetissueplaysan
importantroleininflammation[55,56]andthatADIPOQmighthavepro-inflammatoryproperties
Int.J.Mol.Sci. 2018,19,404 10of17
injointdiseases[57]. However,thepresentresultswerenotabletoconfirmarelationshipbetween
adiposetissueandinflammationintendinopathyandchronicruptures.
Innervationcouldbedemonstratedinthepresentstudy,withexpressionofthegeneralnerve
markerPGP9.5,thenervegrowthmarkerGAP43andthenerveregenerationmarkerMIF[58]. PGP9.5
was previously found in the plantaris tendon and peritendinous tissue of patients with Achilles
tendinopathy,indicatingstronginnervation[28]andwasadditionallyassociatedwithpain[27]. MIF
expression was not described in tendons before and no significant differences in expression were
observedbetweentendinopathy/chronicrupturesandintacttendons. However,innervationseemsto
playamoreimportantroleintheacutesituation,asacuterupturedtendonsexpressedhigheramounts
ofPGP9.5andMIFcomparedtotendinopathy,andchronic/intacttendons,respectively. Thefurther
pain-associatedfactorCOX2wasupregulatedinacuterupturedtendonscomparedtochronicruptured
tendons. AnincreasedCOX2expressionwaspreviouslyfoundinpainfulAchillestendonsbuteven
moreinacuterupturedAchillestendons[30],whichunderlinesthepresentfindings.
Limitations
Mostregulationswereobservedfortheacuterupturedtendons,whichmightbecausedbythe
highersamplenumberinthegroup. Thesamplenumberisrelativelylowfortheintact,tendinopathy
and chronic rupture group, which might account for the less distinct regulations. To increase the
samplenumberforintacttendonsanothertendonlocationcouldhavebeenused,asperformedin
otherstudies[24,59]. However,wedecidedtousethesametendonascontroltoavoidregulations
due to a different biomechanical stimulation of the tendons and surrounding tissues. The intact
AchillestendonsderivedfrompatientswithindicationsforfootsurgerynotrelatedtoprimaryAchilles
tendonpathology. SamplesweretakenduringAchillestendonlengtheningwithtendontransferfor
peronealparesisandimplantationoftotalanklearthroplasty. Theseindicationsmightharmthenormal
movementofthefootandthereforealsoAchillestendon,whichcouldhaveaffectedthepresentresults
oftheintactgroup.
Severalanalyzedfactorsandmechanismsarelinkedtoeachother,suchasMMPsandcytokines,
asMMPsalsohavenon-ECMsubstratessuchasIL1βandTNFα[60]. Thepresentstudyislimitedto
adescriptivehistologicalandmolecularbiologicallevelasspecialmechanismsandinterplayscannot
beclarified. Theavailabletendonmaterialwaslimitedandwedecidedtouseadescriptive(histology)
and quantitative (qRT-PCR) method to analyze the tendon samples in the most comprehensive
way. A quantitative protein analysis to validate the expression data was not possible with the
limitedmaterial.
Regardingthedonordemographics,theageofthegroupsvariedanddonorswithchronicruptures
weresignificantlyoldercomparedtodonorswithacuteruptures. Asweshowedpreviouslythatthe
ageofdonorscanaffectthebiologicalcharacteristicsoftenocytesoftherotatorcuff[43,54],weare
notabletofullyexcludethatchangesbetweenthetwogroupsarepartiallyaffectedbythevarying
donorage. However,fortheacuterupturegroupwedeliberatelychooseacohortofyoungrecreational
sportsmentoavoidanalyzingtendonswithdegenerativepreconditionstobetterborderthemfromthe
twochronictendonpathologygroups.
4. MaterialsandMethods
4.1. TendonSampling
AchillestendonbiopsiesfrompatientssufferingfrominsertionalAchillestendinopathy(n=7)or
chronicdelayeddiagnosedruptures(n=6;withatraumatosurgerydurationof≥3weeks)weretaken
atsurgery. Additionally,acuterupturedAchillestendonsfromrecreationalsportsmen,whichwere
takenduringminimalinvasiveAchillestendonreconstruction4to6days(mean4.9±0.9days)after
rupture(n=13)wereusedforcomparison. Patientswhoreceivedoralmedicationorlocalinjections,
whichmightinfluencetendonstructure(systemiccortisone,antibioticsandanabolicsteroids)were