UBC Faculty Research and Publications

Enhanced collagen type I synthesis by human tenocytes subjected to periodic in vitro mechanical stimulation Huisman, Elise; Lu, Alex; McCormack, Robert G; Scott, Alex Nov 21, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12891_2014_Article_2333.pdf [ 565.92kB ]
JSON: 52383-1.0223770.json
JSON-LD: 52383-1.0223770-ld.json
RDF/XML (Pretty): 52383-1.0223770-rdf.xml
RDF/JSON: 52383-1.0223770-rdf.json
Turtle: 52383-1.0223770-turtle.txt
N-Triples: 52383-1.0223770-rdf-ntriples.txt
Original Record: 52383-1.0223770-source.json
Full Text

Full Text

RESEARCH ARTICLEEnhanced collagen type Iexditiodamage or degeneration. Conversely, physiological mech- stretching which promote collagen synthesis by tenocytesHuisman et al. BMC Musculoskeletal Disorders 2014, 15:386http://www.biomedcentral.com/1471-2474/15/386collagen type I production by tenocytes which, in the longInstitute, Vancouver, CanadaFull list of author information is available at the end of the articleanical stimulation promotes ongoing collagen synthesis andrepair activity by resident tendon fibroblasts (tenocytes) [4].The physiological variables (strain level, frequency, repeti-tion number, etc.) which determine the tissue and cellularresponses of tendons to mechanical stretching have beencould help to refine exercise prescription, especially whenconsidering that mechanical stimulation of tendons throughexercise is a cornerstone of rehabilitation following acute orchronic tendon injury [6]. An exercise regimen (3 sets of 15heel drops performed slowly, twice per day, with weightgradually increased to patient tolerance) has been shown tobe effective for chronic Achilles tendinopathy in both theshort and long term [6]. One proposed mechanism ofaction of this type of exercise regimen is a stimulation of* Correspondence: ascott@interchange.ubc.ca1Department of Physical Therapy, University of British Columbia, Vancouver,Canada2Centre for Hip Health and Mobility, Vancouver Coastal Health and Researchstress deprivation [2,3] on the other, can lead to tendonstudy of mechanically stimulated human tenocytes, the influence of rest insertion and cycle number on (1) theprotein and mRNA levels of type I and III collagen; (2) the mRNA levels of transforming growth factor beta (TGFB1)and scleraxis (SCXA); and (3) tenocyte morphology, were assessed.Methods: Human hamstring tenocytes were mechanically stimulated using a Flexcell® system. The stimulationregimens were 1) continuous and 2) rest-inserted cyclic equiaxial strain at a frequency of 0.1 Hz for 100 or 1000cycles. Data were normalized to unstimulated (non-stretched) control groups for every experimental condition.qPCR was performed to determine relative mRNA levels and quantitative immunocytochemistry image analysis wasused to assess protein levels and cell morphology.Results: Collagen type I mRNA level and pro-collagen protein levels were higher in tenocytes that were subjectedto rest-inserted mechanical stimulation, compared to continuous stretching (p < 0.05). Rest insertion and increasedcycle number also had significant positive effects on the levels of mRNA for TGFB1 and SCXA (p < 0.05). There wasno direct relation between cell morphology and gene expression, however mechanical stimulation, overall, induceda metabolically active tenocyte phenotype as evidenced by cells that on average demonstrated a decreased major-minoraxis ratio (p < 0.05) with greater branching (p < 0.01).Conclusions: The incorporation of rest periods in a mechanical stretching regimen results in greater collagen type Isynthesis. This knowledge may be beneficial in refining rehabilitation protocols for tendon injury.BackgroundTendons experience varying loads in their natural environ-ment. Excessive mechanical loading [1] on the one hand, orstudied to some extent [5], but the optimal conditions toprevent injury or to promote collagen synthesis and repairare incompletely understood.Increased knowledge of the parameters of mechanicaltenocytes subjected to pmechanical stimulationElise Huisman1,2, Alex Lu2, Robert G McCormack3 and AleAbstractBackground: Mechanical stimulation (e.g. slow heavy loatendinopathy, however the optimal parameters of stimula© 2014 Huisman et al.; licensee BioMed CentrCommons Attribution License (http://creativecreproduction in any medium, provided the orDedication waiver (http://creativecommons.orunless otherwise stated.Open Accesssynthesis by humanriodic in vitroScott1,2*ng) has proven beneficial in the rehabilitation of chronicn have not been experimentally determined. In thisal Ltd. This is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,Huisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 2 of 8http://www.biomedcentral.com/1471-2474/15/386term, could lead to an increased tendon strength andmodulus [6].Experimental data have suggested that there may be awindow of appropriate mechanical stimulation definedby variables such as repetition number and strain magni-tude. A strain magnitude dependency was reported formRNA and protein levels of collagen type I [7], whereasan inverse strain magnitude dependency was found forMmp1 [8] (a metalloproteinase with collagenase activ-ity). Several studies have shown that tenocytes undergo-ing mechanical stretching increase their expression ofthe genes for collagen type I and III as well regulatorygenes and growth factors which influence collagen syn-thesis such as TGFB1 and SCXA [4,8-10].Surprisingly, an equivalent magnitude of increase incollagen type I synthesis rate in the human patellar ten-don was observed following both long distance running(36km) and strength training (10 sets of 10 maximalknee extensions) [11,12]. This finding has led to thespeculation that collagen type I production by tenocytesmight plateau after a number of loading repetitions aslow as n = 100 [11]. Desensitization to continuous loadcycles is a phenomenon known to occur in bone cells[13]. More importantly, the insertion of recovery periodsduring the mechanical stimulation restored the mechan-osensitivity in bone cells [13]. The insertion of rest pe-riods during cyclic stretching experiments performed onadolescent and aged murine tibia resulted in enhancedbone formation compared with controls. The level ofbone formation observed in response to rest-insertedloading was similar to a loading regimen that doubledthe load and cycle number [14,15]. This may indicatethat a high load and cycle number are not necessary toelicit a maximal response if rest periods are included. Astudy of osteogenesis in mice tibia showed that rest pe-riods inserted in between short bouts of stretching had alarger effect on osteogenesis than continuous stretching.While several studies have displayed the positive effectsof rest insertion on bone formation [13-17], the effect ofrest-insertion on the load-induced expression of collagengenes in tenocytes has yet to be examined.In vivo, tenocytes typically display an elongated morph-ology, oriented along the longitudinal direction of thetendon [18], while in response to increased mechanicalloading, they may adopt a more rounded morphology withmore prominent cytoplasm. In vitro, it has been shownthat gene expression and protein production are influ-enced by cell attachment and spreading [19]. Li et al.reported that more elongated human tendon fibroblastsexpressed higher collagen type I levels compared to lesselongated cells, while the cell spreading area was constant[20]. Alternately, other studies have shown that tendon fi-broblasts may adopt a more rounded shape when moremetabolically active. In the human patellar tendon, ovoidtendon cells expressed higher levels of TGFB1 and pro-collagen type I compared to elongated tenocytes [21]. Toour knowledge, the relation of tendon cell shape and colla-gen gene expression has not been directly investigated inmechanically stimulated tenocytes.Tenocytes were mechanically stimulated to test thehypothesis that rest insertion compared to continuousstretching initially up-regulates SCXA and TGFB1 mRNAlevels [22-24] and is accompanied by enhancement of geneexpression and protein levels of pro-collagen type I andIII, whereas low (100) and high (1000) cycle numberswould have equivalent effects on these same variables. Asa secondary question, the influence of mechanical stimula-tion on cell morphological features was analyzed.MethodsCell cultureHuman hamstring tendon pieces were obtained from ACLreconstruction surgery, after informed consent was ob-tained. Ethics approval was obtained from the Universityof British Columbia. Tendon tissue was trimmed to re-move fat and muscle then washed with phosphate buff-ered saline (PBS) and enzymatically digested with filtered1.5 mg/ml collagenase (Clostridopeptidase A; Sigma,Oakville, Ontario, Canada) in serum-free Dulbecco’s Modi-fied Eagle Medium (DMEM; HyClone, South Logan, Utah,USA) for 30 minutes at 37°C with shaking; for 5 additionalminutes, trypsin (1× TrypLE™ Select, Gibco, Life Tech-nologies, Burlington, ON, Canada) was added. The mix-ture was centrifuged at 1,200 rpm for 5 minutes and thecell pellet was resuspended. Tenocytes were cultured usingHyclone DMEM/high glucose media supplemented with10% fetal bovine serum and 1% penicillin/streptomycin(Thermo Scientific, Ottawa, ON, Canada).Mechanical stimulationCells were passaged at a concentration of 32,500 cells/ml.65,000 cells per well were plated onto 6-well BioFlex® -collagen type I coated culture plates (Flexcell InternationalCorp., Hillsborough, NC, USA). After 48 hours, the cellswere subjected to mechanical stimulation using a sinus-oidal waveform at a frequency of 0.1 Hz and a strain of10%. The frequency of 0.1 Hz was based on findings ofKongsgaard et al. who demonstrated that slow (e.g. 6-8second repetition duration), heavy resistance exercise wasbeneficial in the rehabilitation of patellar tendinopathyresulting in enhanced collagen synthesis [25]. The equiax-ial strain of 10% applied to the BioFlex® plates has beenpreviously reported to result in an average strain experi-enced by the cells of approximately 3-5% [26]. The studyconsisted of four experimental groups with a combinationof low or high cycle number and with or without rest. Theprecise groups were 1) 100 cycles of continuous stretch-ing, 2) 100 cycles with 10s rest after every cycle, 3) 1000cycles of continuous stretching, 4) 1000 cycles with 10 srest after each cycle. For every experiment group therewas a corresponding unstimulated control group, whosecells were otherwise treated identically and harvested atthe same time points. All experiments were performed inbiological and technical triplicates.RNA extraction and quantitative polymer chain reaction(qPCR)Huisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 3 of 8http://www.biomedcentral.com/1471-2474/15/386Cells were lysed using lysis buffer (Thermo Scientific,Ottawa, ON, Canada) with 1% β mercaptoethanol andstored at -80°C until further processing. Ribonucleic acid(RNA) was extracted according to the manufacturers’instructions using the GeneJet RNA purification kit(Thermo Scientific, Ottawa, ON, Canada) and storedat -80°C until further processing. Complementary deoxy-ribonucleic acid (cDNA) was synthesized using 100mMdNTP set (Life Technologies, Burlington, ON, USA) andstored at -20°C. qPCR was performed in triplicate usingSYBR Green (FastStart Universal SYBR Green MasterROX, Roche Diagnostics Corporation, Indianapolis, IN,US) on a 7500 Fast Real – Time PCR System (AppliedBiosystems, Life Technologies, Burlington, ON, USA). Theprimers used (Table 1) were designed for the target humangenes. Values for mRNA transcript levels were normalizedto corresponding Glyceraldehyde 3-phosphate dehydro-genase (GAPDH) values.Immunocytochemistry for pro-collagen types I and IIICells of all experimental groups and corresponding con-trols were fixed on the BioFlex plate stretchable substratein 4% paraformaldehyde (w/v) for 15 minutes at roomtemperature and stored at 4°C until further processing.Substrate sections of 1×1cm were cut from similar loca-tions of the membrane using a scalpel and mounted onglass slides, and permeabilized in 0.4% Triton X-100 for 5min after which the sections were blocked with PBS +1%BSA +0.2% Tween20 solution for 30 minutes. A mixtureof primary antibodies against both pro-collagen type I(Developmental Studies Hybridoma Bank. Iowa City, IA,US) and pro-collagen type III (Acris Antibodies, Inc. SanDiego, CA, US) was applied at a concentration of 1:500 forTable 1 RT-qPCR primersTarget Forward Primer Reverse PrimerGAPDH TCTTTTGCGTCGCCAGCCGAG TGACCAGGCGCCCAATACGACCollagentype ITGTTCAGCTTTGTGGACCTCCG CGCAGGTGATTGGTGGGATGTCTCollagentype IIIAATCAGGTAGACCCGGACGA TTCGTCCATCGAAGCCTCTGTGFB GCAACAATTCCTGGCGATACC AAAGCCCTCAATTTCCCCTCCScleraxis A AGAACACCCAGCCCAAACA TCGCGGTCCTTGCTCAACTT1 hour in the dark. This was followed by the incubationwith the secondary antibody solution (Alexa fluor 488 goatanti-rabbit and 596 goat anti-mouse, Life Technologies,Burlington, ON, US) at a concentration of 1:400 for 30min in the dark. Finally, nuclei were stained using Hoechst(33342 Thermo Scientific, Rockford, IL, USA) antibody ina dilution of 1:10,000 PBS for 2 minutes. All staining pro-cedures were performed in a humidified chamber at roomtemperature.Image acquisitionFluorescent micrographs were obtained using a ZeissAxio Observer.A1 (Zeiss, Oberkochen, Germany) with a10x objective. For each field of view, three images weretaken on each separate fluorescent channel to representnuclei and pro-collagens types I and III. Nuclei wereidentified by thresholding using CellProfiler™ software;the pro-collagen type I and III fluorescent stains associ-ated with each nuclei were then identified. Three imagesper fluorescent channel and per field of view were ob-tained, resulting in data from approximately 300 tendoncells of each experimental and control group.Measurement of pro-collagen I and III proteinProtein levels of pro-collagen type I and III were operation-ally defined as the intensities of the fluorescent channel ofpro-collagen type I and III, respectively, as determinedusing AxioVision software (Zeiss, Oberkochen, Germany).The fluorescence intensities were obtained through theidentified pro-collagen I and III labeling described in theprevious paragraph of the methodology. For each individ-ual fluorescent channel, the pixel intensities on a scalefrom 0 to 1 within the identified area were averaged to ob-tain mean intensity.Analysis of cell morphologyCell morphology values were obtained from each celldemonstrating positive pro-collagen type III fluorescentlabeling. This pro-collagen type III based method waschosen because all tendon cells examined constitutivelydemonstrated robust labeling (of varying intensity, as de-scribed above) throughout their cytoplasm, allowing theoverall cell morphology to be clearly assessed.To obtain solidity values a convex hull was drawnaround the cell surface. The solidity was derived as the ra-tio of surface area of the stain to total surface area of theconvex hull, thus representing an approximation of theextent of cell branching. A perfectly smooth object wouldtherefore be assigned a value of 1, whereas the morebranched the cell, the lower the solidity value.The ratio between major and minor axis per cell, wasalso obtained from the elliptical outline around the pro-collagen type III immunolabeling as an additional indica-tion of cell shape. An ellipse was fitted around the celland the length (in pixels) of the major and minor axis ofthe ellipse. The resultant output images were used tomanually filter the datasets by deleting erroneously iden-tified object sets caused by image artefacts or overlap-ping cells.Statisticsincreased after 24 h (p < 0.01) and had a mean of 0.205(0.203-0.207) while the control group displayed a mean of0.176(0.174-0.179). The stretched group showed a higher(p < 0.01) mean of 0.262(0.261-0.264) for pro-collagen typeIII intensity while the control group had value of 0.238(0.235-0.240).reIHuisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 4 of 8http://www.biomedcentral.com/1471-2474/15/386Linear mixed model analysis was performed using SPSS(SPSS Inc., Chicago, IL, USA) to test for statistically signi-ficant differences in mRNA levels. Repeated GeneralizedEstimating Equations within Generalized Linear Modelsusing SPSS was used to determine protein level, cell solid-ity and ratio of major/minor axis among the groups. Themodel tested for main effects of mechanical stimulation vscontrols, rest inserted vs continuous stretching, and low(100) vs high (1000) cycle number. The mRNA data wereexpressed as relative quantity (RQ, relative to unstimulatedtenocytes harvested under identical conditions at the sametime) after normalizing the raw data to the housekeepinggene GAPDH. RQ values were then log transformed beforestatistical analysis to obtain normal distributions. A p valueof ≤0.05 was considered statistically significant for all stat-istical tests. The qPCR results are reported as the meandifference between low and high cycle number and con-tinuous and rest inserted mechanical stimulation groupswith a 95% confidence interval (CI) of the difference. Theprotein and cell morphology results were reported asmean and 95% CI. For both methods CI adjustments weremade using Sidak correction. The figures depict the maineffects tested by the linear models, with each variable (con-tinuous vs rest-inserted loading, 100 vs 1000 repetitions)depicted in a separate figure.ResultsMechanical stimulation of human tendon cells comparedto un-stimulated cellsThe cyclic stretching regimens were well tolerated by thetenocytes (i.e. no evidence of cell death and no difference inRNA concentrations between stimulated and unstimulatedgroups). Overall, mechanical stimulation led to significantlygreater collagen type I and SCXA mRNA levels (p < 0.05)compared with unstimulated cultures (8 h, Table 2).The immunohistochemistry also indicated that thepro-collagen type I intensity of the stretched group wasTable 2 The effect of mechanical stimulation on mRNA expGroup Collagen type I Collagen type IIRest 1.15(1.00-1.34)* 0.81(0.59-1.11)Continuous 0.86(0.70-1.05) 0.67(0.46-0.98)*100 cycles 0.90(0.73-1.10) 0.93(0.60-1.55)1000 cycles 1.20(1.03-1.40)* 0.66(0.49-0.89)ƚThe data is displayed as the back transformed mean difference of experimental grolarger values for the experimental group compared to the unstimulated control culsignificant difference of p < 0.01. The back transformed means (CI) are not equivaleRest-inserted vs continuous stretchingRest insertion resulted in a greater protein level ofpro-collagen type I [1.421(1.399-1.442)] (24 h, p < 0.01)compared to continuous stretching [1.303(1.284-1.322)]while pro-collagen type III levels were greater with con-tinuous stretching [1.223(1.211-1.235)] compared with thecells stretched in the rest-inserted regimen [(1.142(1.130-1.153), 24 h, Figure 1, p < 0.01]. Overall, periods of rest in-sertion had a positive effect on mRNA levels for collagentype I with a mean difference (CI of difference) of 1.35(1.10-1.65) SCXA 1.30(1.02-1.65) and TGFB1 [1.70(1.10-2.65), 8 h, p < 0.05]. Collagen type I mRNA levels 1.20(1.01-1.43) were increased as early as 4 h (p < 0.05).Low vs high cycle numberThe pro-collagen type III protein level was greater (p < 0.01)in the 100 cycles stretch regimen [1.243(1.231-1.254)]compared to the 1000 cycle group [1.122(1.111-1.134),24 h, Figure 2].The cells stimulated for 1000 cycles showed higherexpression of TGFB1 1.49(1.20-1.85), 4 h, p < 0.01], andcollagen type I [1.31(1.07-1.60, 8 h, p < 0.05] comparedto the 100 cycle regimen.Cell morphologyThe solidity of tendon cells was greater when subjectedto mechanical stimulation [0.621(0.618-0.624)] comparedwith unstimulated controls [0.601(0.597-0.605)] (p < 0.01),indicating increased cell branching with mechanical sti-mulation. The major/minor axis, a measure of cell shape,in mechanically stretched tenocytes was lower [2.845(2.815-2.875)] than the non-stretched control group(p < 0.01, 3.005(2.958-3.053), Figure 3) indicating increasedcell rounding with mechanical stimulation.The major to minor axis ratio was lower (24 h, p < 0.05,Figure 3) with the 1000 cycles stretching regimen [2.159ssion compared with unstimulated controlsScleraxis-A Transforming growth factor β2.62(2.09-3.27)ƚ 1.21(0.83-1.76)1.79(1.22-2.64)ƚ 0.80(0.56-1.16)3.33(2.93-3.79)ƚ 1.20(0.89-1.63)1.43(1.03-1.99)* 0.83(0.50-1.38)up minus control group (confidence interval of difference). Values > 1 depicttures. The *indicates a significant difference of p < 0.05, the ƚ indicatesnt to means of the original variable due to mathematical twisting [27].Figure 1 Effect of rest-inserted stretching vs continuous stretching on mRNA expression of collagen type I (A) and III (B) and proteinlevels of pro-collagen I and III (C). The cells that underwent rest inserted stretching showed an increased mRNA expression of collagen type I(A) at 8 h post stretching (p < 0.05) compared to continuously loaded cells. Collagen type III mRNA expression did not significantly change (B).sers pheHuisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 5 of 8http://www.biomedcentral.com/1471-2474/15/386(1.757-2.561)] compared to the 100 cycle regimen [2.778(2.748-2.808)].DiscussionThe main finding of this study is that the insertion ofrest periods between cycles of mechanical stimulation oftenocytes resulted in greater expression of collagen typeI mRNA and protein compared with continuously stim-ulated tenocytes.Mechanical stimulation of cells results in mechanotrans-duction; the transmission and conversion of a mechanicalstimulus into a biological response, through a variety ofmechanisms. There are three main stages of mechanotrans-duction: mechanocoupling, cell-to-cell communication, andeffector cell response. Mechanocoupling is the processwhereby the applied load is transmitted through the tissuesand cells, resulting in different types of cellular deformation(strain, compression, fluid flow and shear [28]), and thisThe pro-collagen type I was significantly greater (p < 0.05) in the rest ingroup had an increased (p < 0.05) pro-collagen type III value at 24 hourthe ƚ indicates significant difference of p < 0.01, the error bars indicate tunstretched controls harvested at every time point.deformation is translated into a biochemical responsewhich can include the opening of ion channels suchas stretch-activated calcium channels, and the activa-tion of transmembrane signaling proteins such as integrinsand G-protein coupled receptors [29]. The mechanicallyFigure 2 Effect of cycle number on mRNA expression and protein levelevated Collagen type I mRNA levels (p < 0.05). Pro-collagen type III protei24 h). The *indicates a significant difference of p < 0.05, the indicates ƚ signAll data points from each group were normalized to unstretched controls hstimulated cell may also spread the signal to adjacent cells(cell-to-cell communication, involving the passage of cal-cium ions via gap junction, for instance) thereby amplify-ing the response [30]. In response to elevated intracellularcalcium and other signals, enzyme activity is initiatedwithin the cell (e.g. activation of MAPK family members)leading to gene transcription and production of proteinwhich can include newly synthesized extracellular matrix,or autocrine/paracrine substances like TGFB or IGF-Iwhich further amplify the adaptive responses [22]. Most ofthese processes have been documented to some extent intendon cells, which are known to be highly mechanore-sponsive. The adaptive response of tenocytes has beenshown, in vivo, to be related to the magnitude of appliedstrain; e.g., exercise in a shortened (low strain) positionleads to less tendon adaptation than exercise in a length-ened (high strain) position [31].LaMothe & Zernicke [16] and Hanson et al. [17] re-tion group compared to the continuously loaded, while the continuousost stretching (C). The *indicates a significant difference of p < 0.05,standard error. All data points from each group were normalized tosearched the effect of rest insertion and showed that restinsertion incorporated in loading protocols had a larger ef-fect on mineral deposition and osteogenesis than loadingonly [16,17]. Commonly, mechanical stimulation is appliedas a continuous bout of stretch at a certain frequency [32].el of collagen type I and III. The high cycle number (1000) causedn levels (C) were upregulated in the 100 cycles regimen (p < 0.05,ificant difference of p < 0.01, the error bars indicate the standard error.arvested at every time point.Huisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 6 of 8http://www.biomedcentral.com/1471-2474/15/386The current results suggest that rest incorporation stimu-lates the expression of collagen type I and collagen type IIImRNA to a greater degree than continuous stimulation.The number of repetitions of applied mechanical stretch-ing can elicit different gene expression patterns. In thisstudy, at a frequency of 0.1 Hz the high cycle number(1000) increased the TGFB1 (4 h) and collagen type I (8 h)mRNA expression, while no change in collagen type IIIwas observed. In a human in vitro patellar tendon modelBosch et al. showed an increase in amino terminal pro-collagen type III propeptide (P-III-NP) after 1800 and 3600stretch cycles at a strain of 5% and a frequency of 1 Hz.The higher cycle number (3600) resulted in a higherP-III-NP elevation (32%) compared to the lower cyclenumber (1800) which demonstrated a 21% increase.The carboxyterminal pro-collagen type I propeptidelevels demonstrated a 50% increase after 3600 cycles ofstretching, but not 1800. These findings are in linewith our results; a higher cycle number leads to a moreincreased collagen type I mRNA expression.SCXA regulates the transcription of collagen type 1a1and TGFB is known to induce collagen type I expression[22,23,33]. In the current study the cells stretched on aregimen of 1000 cycles upregulated SCXA (4 h), TGFB1(4 h) and collagen type I (8 h) mRNA expression; thisFigure 3 Immunocytochemistry images (pro-collagen type III), 10× mstretched cells, C) 1000 continuously stretched cells and D) 1000 cycles resappear more spread out, with a correspondingly reduced major/minor axispattern is in line with expectations of upregulated SCXAand TGFB1 preceding elevated levels of collagen type Iand collagen type III are present post stimulation.It has been shown that mechanical stresses applied tothe cells lead to altered forces within the cell which playimportant roles in the control of cell shape and cell func-tion [34,35]. For example, the application of shear stresse.g. caused by blood flow, to endothelial cells resulted inaltered cell shape from cobblestone to aligned in the direc-tion of the flow and the formation of actin stress fibers[36]. The results indicate that mechanical stimulation mayinduce a more metabolically active tenocyte phenotype,with tendon cells that are less elongated, and morebranched, particularly when subjected to cyclic stretchingthat incorporates rest insertion and longer durations; thesesame regimens are associated with higher expression levelsof a tenocyte transcriptional regulator (SCXA) and a keytendon extracellular matrix protein (type 1 collagen).These findings are in contrast with those of Li et al., whofound that more elongated cells had a higher collagen typeI protein expression [20], however they are in keeping withthose of Chuen et al. who found less elongated tenocytesto express higher levels of type I collagen [21]. Clearly, cellshape is dynamically regulated by mechanical stimulationand a clear relation with gene expression is not to beagnification. A) Un-stimulated control cells, B) 100 cycles continuouslyt insertion cells. Mechanically stimulated cells (especially C and D).Huisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 7 of 8http://www.biomedcentral.com/1471-2474/15/386expected. It may also be worth pointing out that older,morphological categorizations of tendon cells based ontheir shape (e.g. “tenoblasts” being a more rounded sub-population of tendon cells compared to the elongatedtenocytes) may not be valid [37], and that tenocyte round-ing, sometimes taken to be a feature of tendinosis [38],may in fact be associated with an adaptive response andshould not necessarily be interpreted as pathological.A limitation of this study was that the observed changesin mRNA expression were of small magnitude; however,they were in keeping with a significant upregulation at theprotein level and with the knowledge that tendon is a rela-tively slow-adapting tissue. Furthermore, our cell shapeobservations, although based on automated measurementsof many cells, do not account for possible changes in cellthickness. There is also an inherent limitation of in vitrostudies; the response to stretching may also have beenmore accentuated if the tendon cells were located withintheir native extracellular matrix, rather than cultured twodimensionally – a condition which likely disturbs integrin-mediated signaling which is thought to contribute tomechanoresponsiveness [39]. Nonetheless, in vitro studieswith tenocytes in two dimensional culture have replicatedmany of the mechanically-induced responses that areknown to occur in vivo and may therefore continue toserve as a useful experimental system.ConclusionIn this study, we demonstrated that periods of rest insertionare beneficial for collagen synthesis by human tenocytes.One implication of these findings is that when using exer-cise as a rehabilitative measure for people with chronictendinopathy, allowing an adequate time to recovery afterevery stretch cycle may induce a more substantial adaptiveresponse. Further studies could investigate the role ofmechanotransduction pathways in response to rest-insertedstretching (e.g. the role of calcium signaling) and to exam-ine whether a rest-inserted exercise program (e.g. 10 s restafter every repetition) results in improved tendon adapta-tion in humans. In addition, further optimization of mech-anical properties capable of stimulating collagen synthesiscould be undertaken, including strain rate and frequency.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsEH designed and carried out the experiments, performed the statisticalanalysis and wrote the manuscript. AL performed the immunocytochemistryand designed and carried out the quantitative image analysis. RM designedand administered the clinical protocol for screening patients and collectinghuman tendon tissue and clinical data. AS conceived of the study, designedexperiments and helped writing the manuscript. All authors read andapproved the final manuscript.AcknowledgementsAS received a Michael Smith Scholar award. AL was awarded anUndergraduate Student Research Award through Natural Sciences andEngineering Research Council of Canada. EH received a WorkSafeBCscholarship. The study was funded by the National Sciences and EngineeringResearch Council of Canada.Author details1Department of Physical Therapy, University of British Columbia, Vancouver,Canada. 2Centre for Hip Health and Mobility, Vancouver Coastal Health andResearch Institute, Vancouver, Canada. 3Department of Orthopaedic Surgery,University of British Columbia, Vancouver, Canada.Received: 26 May 2014 Accepted: 6 November 2014Published: 21 November 2014References1. Scott A, Khan KM, Heer J, Cook JL, Lian O, Duronio V: High strainmechanical loading rapidly induces tendon apoptosis: an ex vivo rattibialis anterior model. Br J Sports Med 2005, 39:1–4.2. Thornton GM, Shao X, Chung M, Sciore P, Boorman RS, Hart D a, Lo IKY:Changes in mechanical loading lead to tendonspecific alterations inMMP and TIMP expression: influence of stress deprivation andintermittent cyclic hydrostatic compression on rat supraspinatus andAchilles tendons. Br J Sports Med 2008, 44:698–703.3. Gardner K, Arnoczky SP, Caballero O, Lavagnino M: The effect of stress-deprivation and cyclic loading on the TIMP/MMP ratio in tendon cells:an in vitro experimental study. Disabil Rehabil 2008, 30:1523–1529.4. Heinemeier KM, Olesen JL, Haddad F, Langberg H, Kjaer M, Baldwin KM,Schjerling P: Expression of collagen and related growth factors in rattendon and skeletal muscle in response to specific contraction types.J Physiol 2007, 582(Pt 3):1303–1316.5. Lavagnino M, Arnoczky SP: In vitro alterations in cytoskeletal tensionalhomeostasis control gene expression in tendon cells. J Orthop Res 2005,23:1211–1218.6. Alfredson H, Cook J: A treatment algorithm for managing Achillestendinopathy: new treatment options. Br J Sports Med 2007, 41:211–216.7. Jiang C, Shao L, Wang Q, Dong Y: Repetitive mechanical stretchingmodulates transforming growth factor-β induced collagen synthesis andapoptosis in human patellar tendon fibroblasts. Biochem Cell Biol 2012,674:667–674.8. Lavagnino M, Arnoczky SP, Tian T, Vaupel Z: Effect of Amplitude andFrequency of Cyclic Tensile Strain on the Inhibition of MMP-1 mRNAExpression in Tendon Cells: An In Vitro Study. Connect Tissue Res 2003,44:181–187.9. Sun HB, Andarawis-Puri N, Li Y, Fung DT, Lee JY, Wang VM, Basta-Pljakic J,Leong DJ, Sereysky JB, Ros SJ, Klug R a, Braman J, Schaffler MB, Jepsen KJ,Flatow EL: Cycle-dependent matrix remodeling gene expression responsein fatigue-loaded rat patellar tendons. J Orthop Res 2010, 28:1380–1366.10. Heinemeier K, Langberg H, Olesen JL, Kjaer M: Role of TGF-beta1 in relationto exercise-induced type I collagen synthesis in human tendinous tissue.J Appl Physiol 2003, 95:2390–2397.11. Magnusson SP, Langberg H, Kjaer M: The pathogenesis of tendinopathy:balancing the response to loading. Nat Rev Rheumatol 2010, 6:262–268.12. Langberg H, Skovgaard D, Petersen LJ, Bulow J, Kjaer M: Type I collagensynthesis and degradation in peritendinous tissue after exercisedetermined by microdialysis in humans. J Physiol 1999, 521(Pt 1):299–306.13. Robling AG, Burr DB, Turner CH: Recovery periods restoremechanosensitivity to dynamically loaded bone. J Exp Biol 2001,204:3389–3399.14. Srinivasan S, Weimer D, Agans SC, Bain SD, Gross TS: Low-magnitudemechanical loading becomes osteogenic when rest is inserted betweeneach load cycle. J Bone Miner Res 2002, 17:1613–1620.15. Srinivasan S: Enabling bone formation in the aged skeleton via rest-inserted mechanical loading. Bone 2003, 33:946–955.16. LaMothe JM, Zernicke RF: Rest insertion combined with high-frequencyloading enhances osteogenesis. J Appl Physiol 2004, 96:1788–1793.17. Hanson AD, Marvel SW, Bernacki SH, Banes AJ, van Aalst J, Loboa EG:Osteogenic effects of rest inserted and continuous cyclic tensile strainon hASC lines with disparate osteodifferentiation capabilities. AnnBiomed Eng 2009, 37:955–965.18. Wang JH-C, Jia F, Gilbert TW, Woo SL-Y: Cell orientation determines thealignment of cell-produced collagenous matrix. J Biomech 2003,36:97–102.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionHuisman et al. BMC Musculoskeletal Disorders 2014, 15:386 Page 8 of 8http://www.biomedcentral.com/1471-2474/15/38619. McBeath R, Pirone DM, Nelson CM, Bhadriraju K, Chen CS: Cell shape,cytoskeletal tension, and RhoA regulate stem cell lineage commitment.Dev Cell 2004, 6:483–495.20. Li F, Li B, Wang QM, Wang JH: Cell shape regulates collagen type Iexpression in human tendon fibroblasts. Cell Motil Cytoskeleton 2008,65:332–341.21. Chuen FS, Chuk CY, Ping WY, Nar WW, Kim HL, Ming CK:Immunohistochemical characterization of cells in adult human patellartendons. J Histochem Cytochem 2004, 52:1151–1157.22. Maeda T, Sakabe T, Sunaga A, Sakai K, Rivera AL, Keene DR, Sasaki T,Stavnezer E, Iannotti J, Schweitzer R, Ilic D, Baskaran H, Sakai T: Conversionof mechanical force into TGF-β-mediated biochemical signals. Curr Biol2011, 21:933–941.23. Léjard V, Brideau G, Blais F, Salingcarnboriboon R, Wagner G, Roehrl MH a,Noda M, Duprez D, Houillier P, Rossert J: Scleraxis and NFATc regulate theexpression of the pro-alpha1(I) collagen gene in tendon fibroblasts. J BiolChem 2007, 282:17665–17675.24. Sharma P, Maffulli N: Biology of tendon injury: healing, modeling andremodeling. J Musculoskelet Neuronal Interact 2006, 6:181–190.25. Kongsgaard M, Kovanen V, Aagaard P, Doessing S, Hansen P, Laursen a H,Kaldau NC, Kjaer M, Magnusson SP: Corticosteroid injections, eccentricdecline squat training and heavy slow resistance training in patellartendinopathy. Scand J Med Sci Sports 2009, 19:790–802.26. Wall ME, Weinhold PS, Siu T, Brown TD, Banes AJ: Comparison of cellularstrain with applied substrate strain in vitro. J Biomech 2007, 40:173–181.27. Stats Design & Analysis Web Guide [http://dawg.utk.edu/diag/interpret_transformed_means.htm]28. Lavagnino M, Arnoczky SP, Kepich E, Caballero O, Haut RC: A finite elementmodel predicts the mechanotransduction response of tendon cells tocyclic tensile loading. Biomech Model Mechanobiol 2008, 7:405–416.29. Wall ME, Banes AJ: Early responses to mechanical load in tendon: role forcalcium signaling, gap junctions and intercellular communication.J Musculoskelet Neuronal Interact 2005, 5:70–84.30. Banes AJ, Weinhold P, Yang X: Gap Junctions Regulate Responses ofTendon Cells Ex Vivo to Mechanical Loading. Clin Orthop Relat Res 1999,367S:356–370.31. McMahon GE, Morse CI, Burden A, Winwood K, Onambélé-Pearson GL: Themanipulation of strain, when stress is controlled, modulates in vivotendon mechanical properties but not systemic TGF-β1 levels. Physiol Rep2013, 1:1–16.32. Screen HRC, Shelton JC, Bader DL, Lee DA: Cyclic tensile strain upregulatescollagen synthesis in isolated tendon fascicles. Biochem Biophys ResCommun 2005, 336:424–429.33. Robbins JR, Evanko SP, Vogel KG: Mechanical loading and TGF-betaregulate proteoglycan synthesis in tendon. Arch Biochem Biophys 1997,342:203–211.34. Maniotis AJ, Chen CS, Ingber DE: Demonstration of mechanicalconnections between integrins, cytoskeletal filaments, and nucleoplasmthat stabilize nuclear structure. Proc Natl Acad Sci U S A 1997, 94:849–854.35. Chicurel ME, Chen CS, Ingber DE: Cellular control lies in the balance offorces. Curr Opin Cell Biol 1998, 10:232–239.36. Malek AM, Izumo S: Mechanism of endothelial cell shape change andcytoskeletal remodeling in response to fluid shear stress. J Cell Sci 1996,109:713–726.37. Ippolito E, Natali PG, Postacchini F, Accinni L, De Martino C: Morphological,Immunochemical and Biochemical Study of Rabbit Achilles Tendon atVarious AGes. J Bone Joint Surg Am 1980, 62:583–589.38. Abraham T, Fong G, Scott A: Second harmonic generation analysis ofearly Achilles tendinosis in response to in vivo mechanical loading.BMC Musculoskelet Disord 2011, 12:1–26.39. Wang JH-C: Mechanobiology of tendon. J Biomech 2006, 39:1563–1582.doi:10.1186/1471-2474-15-386Cite this article as: Huisman et al.: Enhanced collagen type I synthesis byhuman tenocytes subjected to periodic in vitro mechanical stimulation.BMC Musculoskeletal Disorders 2014 15:386.Submit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items