In vitro analysis of the effect of Flightless I on murine tenocyte cellular functions.

BACKGROUND
Healing of tendons after injury involves the proliferation of tenocytes and the production of extracellular matrix; however, their capacity to heal is limited by poor cell density and limited growth factor activity. Flightless I (Flii) has previously been identified as an important regulator of cellular proliferation and migration, and the purpose of this study was to evaluate the effect of differential Flii gene expression on tenocyte function in vitro.


METHODS
The role of Flii on tenocyte proliferation, migration, and contraction was assessed using established assays. Tenocytes from Flii+/-, wild-type, and Flii overexpressing mice were obtained and the effect of differential Flii expression on migration, proliferation, contraction, and collagen synthesis determined in vitro. Statistical differences were determined using unpaired Student's t test and statistical outliers were identified using the Grubbs' test.


RESULTS
Flii overexpressing tenocytes showed significantly improved migration and proliferation as well as increased collagen I secretion. Explanted tendons from Flii overexpressing mice also showed significantly elevated tenocyte outgrowth compared to Flii+/- mice. In contrast to its role in dermal wound repair, Flii positively affects cellular processes in tendons.


CONCLUSIONS
These findings suggest that Flii could be a novel target for modulating tenocyte activity and improving tendon repair. This could have significant clinical implications as novel therapeutic targets for improved healing of tendon injuries are urgently needed.


Background
Tendons inherently have minimal capacity for healing owing to poor cell density and growth factor activity [1]. Healing of tendons involves the proliferation of spindleshaped fibroblast-like cells known as tenocytes, which comprise the main cellular content of mature tendons. Tenocytes have long cellular processes which facilitate cell-extracellular matrix (ECM) and cell-cell communication through desmosomes and tight junctions [2]. They are highly metabolically active and are responsible for synthesizing and secreting ECM. Tendon repair is characterized by the influx of intrinsic and extrinsic tenocytes which migrate to the injured area and synthesize and secrete ECM to begin the repair process [3]. Tendon healing can occur solely through epitenonbased intrinsic tenocytes that secrete larger, more mature collagen fibers compared with extrinsic tenocytes, resulting in improved physiological outcomes [4]. Tenocyte characterization during injury repair is poorly understood; however, it is well established that tenocyte function is to secrete ECM components and growth factors during healing [5]. Of particular importance is their secretion of transforming growth factor beta (TGFβ) ligands as this family of growth factors is active throughout all phases of tendon repair [5][6][7].
Tenocyte migration, proliferation, and contraction are essential processes that allow tendon healing to occur. The actin cytoskeleton is the driving force behind these processes and it also has essential roles in transmembrane signaling and healing [8,9]. In response to injury, the actin cytoskeleton is remodeled to facilitate the permeation of inflammatory cells and promote tenocyte migration [4,10]. Flightless (Flii) is a highly conserved actin remodeling protein that contains both a gelsolin and leucine-rich repeat (LRR) domain enabling it to modulate the cytoskeleton by regulating actin polymerization and depolymerization [11][12][13][14]. Flii is a known regulator of cellular migration, proliferation, contraction, and adhesion; however, this has been mainly characterized in cutaneous healing responses [8,[15][16][17]. Flii is expressed ubiquitously throughout cells and tissues of the body but its specific expression in cells and structures with migratory capabilities is of particular interest due to their involvement with wound repair [18,19]. Flii has previously been determined to be an important negative regulator of cutaneous wound healing and skin blistering [8,14,[16][17][18]. There is also strong evidence to suggest that Flii modulates TGFβ expression, with decreased levels of TGFβ1 detected in cutaneous healing studies in mice with low levels of Flii (Flii +/− ) [20]. In addition, reducing the levels of Flii decreases pro-scarring TGF-β1 and Smad 3, while causing an elevation of anti-scarring inhibitory Smad 7, suggesting a mechanistic role for Flii in TGF-β signaling [16,20].
In this study, the effect of altering Flii expression on tenocyte activity was investigated. Primary tenocytes were extracted from digital tendons of Flii +/− , wild-type (WT), and Flii Tg/Tg mice and the effect of differential Flii expression on their migration, proliferation, contraction, and ECM production assessed.

Flii mouse generation
All mouse strains were congenic on the Balb/c background and Balb/c littermates were used as wild-type (WT) control animals. Flii homozygous knockout mice are embryonic lethal due to defects with gastrulation during early embryogenesis [21]. Flii heterozygous mice (Flii +/− ) are phenotypically normal, fertile with normal life expectancy. Mice carrying the complete human Flii gene on a cosmid transgene were maintained as described previously [8,21]. Heterozygous transgenic mice Flii Tg/+ were made by crossing Flii +/+ with cosmid transgene Flii +/− . These transgenic mice were inter-crossed to obtain animals homozygous for the transgene Flii Tg/Tg and carry two copies of the mouse Flii gene and two additional copies of the human Flii gene (Flii Tg/Tg ) resulting in elevated levels of Flii protein in various tissues [8,19].

Isolation of mouse tenocytes
Murine tenocyte isolation can generally be achieved from adult mice (12-16 weeks) following established protocols [22]. Therefore, digital tendons were removed intact from the hind feet of 4 euthanized female 12-week-old Flii +/− , WT, and Flii Tg/Tg mice using micro dissection (Fig. 1a). The tendon sheath was removed, and the tip and base of the tendons discarded. The remaining tendons were chopped into 1 mm 3 sections and digested overnight in 4 mg/mL dispase (Worthington Biochem, NJ, USA), and 0.5 mg/mL collagenase (Worthington Biochem, NJ, USA) in Dulbecco's modified Eagle's media (DMEM) without serum at 37°C. The following day, the tendon mix was put through a 70-μM cell strainer (In Vitro Technologies, Victoria, Australia) and the remaining tissue discarded. The tube and strainer were washed with sterile media (DMEM + 10% fetal bovine serum [FBS] + penicillin/ streptomycin [Pen/Strep] + fungizone) and excess media added to stop the action of the dispase/collagenase mix. The suspension was spun at 1200 rpm for 5 min and the supernatant discarded. The pellet was resuspended in 4 mL sterile media and plated on collagen-coated T25 flask. The T25 was left in an incubator (37°C, 5% CO 2 ) for approximately 7 days with media changes every 2-3 days until the cells were a minimum of 90% confluent [23,24].

Migration assay
Tenocytes isolated from Flii +/− , WT, and Flii Tg/Tg mice were plated into 96-well plates at 5 × 10 5 cells/mL and left overnight in an incubator at 37°C at 5% CO 2 to reach confluence. A Woundmaker™ (Essen Bioscience, Michigan, USA) was used to create uniform wounds of 7-800 μM in each well of the 96-well plate which was subsequently placed into an Incucyte (Essen Bioscience, Michigan, USA) at 37°C and 5% CO 2 where images were automatically taken every 3 h for 24 h. The resulting images were analyzed using Image Pro Plus 7.1 as previously described [25].

Tendon outgrowth assay
Whole tendons were removed in sterile conditions from the hind feet of 6 Flii +/− , WT, and Flii Tg/Tg mice. Five millimeters was removed from the ends of each tendon and the remaining tendon was cut into 3 mm 3 sections, transferred into 12-well plates (1 section per well), and cultured in DMEM + 20% FBS + Pen/Strep + Fungizone. Cultures were maintained at 37°C at 5% CO 2 for 12 days [26]. Images were taken at days 0, 4, 8, and 12. Migration distances were measured every 90°using Image Pro Plus 7.1 (MediaCybernetics Inc., Maryland, USA) and the average distance calculated.
Following serum starvation, media were replaced with serum-free media plus Pen/Strep and Fungizone and left for 60 h to allow collagen accumulation. The cells were subsequently washed 3 times in phosphate-buffered saline (PBS), then fixed and permeabilized in ice-cold methanol for 15 min at − 20°C. The cells were aspirated and briefly air-dried before being washed 3 times in PBS. Cells were treated with 0.5% Tween 20 in PBS for 10 min and blocked in PBS + 3% normal goat serum for 30 min. The cells were incubated in 50 μL per well collagen I (#600-401-103; Rockland, PA; 2 μg/mL) in PBS and 3% normal goat serum for 2 h at room temperature. The cells were washed and incubated in 50 μL per well secondary antibody in PBS for 1 h in the dark at room temperature. Once the secondary staining was complete, the cells were washed in PBS, stained with 4′,6-diamindino-2-phenylindole (DAPI) at 1:5000 for 1 min and analyzed immediately on an Olympus Epifluorescent microscope as previously described [13].

Collagen contraction assay
Primary tenocytes and fibroblasts isolated from Flii +/− , WT, and Flii Tg/Tg mice tendons or skin, respectively, were trypsinized in 10 × trypsin/EDTA (Sigma-Aldrich, Sydney, Australia) for 5 min at 37°C and quenched using DMEM + 20% FBS. The cells were spun for 5 min at 1200 rpm and resuspended in ice-cold DMEM at 1 × 10 6 cells/mL. 3D collagen gels were prepared by mixing 8 parts of chilled collagen solution (2 mg/mL) with 1 part 10 × DMEM containing 10% FBS as previously described [16]. The pH was adjusted to 7.4 using 0.1 M NaOH. The tenocytes and fibroblasts were added to the collagen gel mixture at 1 × 10 5 cells/mL. Five hundred microliters of this mixture was added per well to a flatbottomed 48-well plate and allowed to set for 120 min at 37°C 5% CO 2 . Following this, the gel was carefully dislodged using a 200-μL pipette and the gels floated by the addition of 1 mL DMEM (+ 20% FCS, Pen/Strep, and Fungizone). Images were taken at 24, 48, and 72 h post media addition using Olympus Epifluorescent microscope and the images analyzed for degree of contraction using Image Pro-Plus 7.1 as previously described [16].

WST-1 proliferation assay
Methods were as previously described [25]. Briefly, cells were serum-starved for 6 h to synchronize the cell cycle. The media was then replaced with DMEM + 20% FBS and incubated at 37°C, 5% CO 2 for 24, 48, or 72 h. Ten microliters WST-1 was added to each well and the absorbance read on a microplate reader at 450 nm and 600 nm. Cell proliferation was normalized relative to cells with no serum added as control.

Statistical analysis
Statistical differences were determined following unblinding of the data using GraphPad software (USA) using the unpaired Student's t test. A p value of less than 0.05 was considered statistically significant. A p value of less than 0.01 was considered highly significant. Statistical outliers were determined using the Grubbs' test with a significance level of 0.05.

Tenocytes express Tenascin-C and Scleraxis
Digital tendons were removed from Flii +/− , WT, and Flii Tg/Tg mice, the tendon sheath was removed by microscopic dissection, and intrinsic tenocytes were isolated by outgrowth or enzyme digestion methods ( Fig. 1). Specific care was taken to remove the ends of the tendons to avoid phenotypically different tenocytes at the myotendinous junction (MTJ) and osteotendinous junction (OTJ). Dermal fibroblasts were also isolated from Flii +/− , WT, and Flii Tg/Tg by outgrowth or enzyme digestion methods. Tenocytes are structurally similar to fibroblasts and it is impossible to visually distinguish the two celltypes. Tendon-specific markers Tenascin-C and Scleraxis were used to identify tenocytes ( Fig. 1b- Tenascin-C is a glycoprotein expressed specifically in tenocytes [28]. Scleraxis is uniquely expressed in tissues that form tendons and ligaments [29]. Tenascin-C and Scleraxis were expressed in Flii +/− , WT, and Flii Tg/Tg tenocytes with no expression seen in fibroblasts from the same mice, (Fig. 1b-g). Tenascin-C staining was observed to be peri-nuclear whereas Scleraxis staining was observed in the cytoplasm as well as nuclear regions ( Fig. 1b-g, h-m). Tenascin-C was significantly higher in Flii +/− tenocytes when compared to WT and Flii Tg/Tg (p = 0.042 and 0.008, respectively) but no significant difference was noted in Scleraxis expression in tenocytes across the three Flii genotypes (Fig. 1n, o).

Tenocyte migration is regulated by Flii
Previous studies have established the role for Flii in regulating cell migration [9,15]. Here we examined the effect of Flii on tenocyte migration and compared it to effects seen in fibroblasts. Isolated primary tenocytes and fibroblasts from Flii +/− , WT, and Flii Tg/Tg mice were Fig. 1 Tenascin-C and Scleraxis are expressed specifically in tenocytes. a Intact digital tendons were removed from the right and left hind paws of mice for tenocyte isolation. The tendon sheath is indicated by arrows and was removed before digestion to ensure a pure population of intrinsic tenocytes. Scale bar = 1 cm. Representative images of Tenascin C (b-g) and Scleraxis (h-m) staining in tenocyte and fibroblast cells isolated from Flii +/− , WT and Flii Tg/Tg mice. Positive staining was detected in tenocyte cells for both Tenascin-C and Scleraxis, and no expression was noted in fibroblasts. Magnification × 20, scale bar = 100 μM. Graphical representation of mean fluorescence intensity of Tenascin C (n) and Scleraxis (o) staining. Tenascin-C staining was significantly higher in Flii +/− cells compared to WT and Flii Tg/Tg cells. Data is represented as mean ± SEM. *p ≤ 0.05, **p ≤ 0.01, n = 6

Tenocyte cellular outgrowth is regulated by Flii
Flii has previously been shown to regulate cellular outgrowth including both fibroblasts and keratinocytes [9,15].
To investigate the effect of Flii on tenocyte cellular outgrowth, 2 mm 3 sections of digital tendons, cleaned of their sheaths, from Flii +/− , WT, and Flii Tg/Tg mice were placed onto collagen type I coated plates in culture, and the average outgrowth measured over 12 days (Fig. 4a-d). Flii Tg/Tg tenocytes had significantly increased cellular outgrowth across all 12 days when compared to WT and Flii +/− tenocytes (p = 0.007 and 0.0005, respectively, at day 12), with Flii Tg/Tg cells showing a 2-fold increase in cellular migration by day 12 compared to Flii +/− cells (Fig. 4a-d). A dense collection of cells migrated out of Flii Tg/Tg and WT explants by day 3, whereas Flii +/− showed a less dense, slower initial migration. These cells stained positive for Scleraxis confirming the tenocyte phenotype (data not shown).

Flii has minimal effect on tenocyte contraction
Flii +/− , WT, and Flii Tg/Tg tenocytes and fibroblasts were incorporated into a 3D floating collagen gel matrix and left for 72 h to assess the rate of contraction. Minimal contraction was seen in the tenocyte gels across the 72 h with no significant difference seen between the genotypes (Fig. 6a-c, g). In contrast, significant contraction was seen in the fibroblast gels across the 72 h, with Flii +/− gels showing significantly increased contraction than WT and Flii Tg/Tg gels at 48 h (p = 0.004 and 0.0001, respectively) and Flii Tg/Tg at 72 h (p = 0.00007) ( Fig. 6d-f, h).

Flii affects collagen type I and collagen type III production in unwounded fibroblasts but not tenocytes
Collagen type I is the main component of tendons, providing stability and strength. Here the effect of Flii on collagen type I and III production in tenocytes and fibroblasts from Flii +/− , WT, and Flii Tg/Tg mice was investigated. No significant difference was observed in collagen type I production in tenocytes with different levels of Flii; however, fibroblasts were observed to have secreted significantly more collagen type I than the tenocytes (Fig. 6i-o). Overexpression of Flii in Flii Tg/Tg fibroblasts significantly decreased collagen type I production when compared with fibroblasts from WT and Flii +/− mice (p = 0.0002 and 0.03, respectively) ( Fig. 6l-o). Similarly, no significant difference in collagen type III production was observed in the tenocytes across all three genotypes ( Fig. 6p-r, v); however, fibroblasts showed an overall significant increase in collagen type III production ( Fig. 6s-v) Furthermore, Flii Tg/Tg fibroblasts showed significantly increased collagen type III secretion when compared to fibroblasts from WT and Flii +/− mice (p = 0.0002 and 0.003, respectively) (Fig. 6v).

Collagen type I production is delayed in scratch-wounded Flii Tg/Tg tenocytes
Previous studies have shown that Flii can regulate collagen I production during cutaneous wound healing [8]. Here we examined the effect of differential Flii expression on collagen type I production by wounded tenocytes using established collagen I assay [27]. Tenocytes from Flii +/− , WT, and Flii Tg/Tg mice were scratch-wounded and immunofluorescently stained for collagen type I expression at 3, 6, 12, and 24 h postwounding ( Fig. 7a-l). Flii Tg/Tg tenocytes had a significant delay in peak collagen type I expression at 12 h post-wounding compared to Flii +/− collagen type I expression at 3 h postwounding; however, the peak level of collagen type I expression was not significantly different ( Fig. 7a-m). Collagen type I expression had returned to basal levels in the Flii +/− tenocytes by 12 h and remained low for the rest of the experiment whereas there was still elevated expression in the Flii Tg/Tg tenocytes by 24 h (Fig. 7m).

Flii Tg/Tg tenocytes express significantly less TGFβ1 during healing
TGFβ1 is often referred to as the pro-scarring growth factor in the TGFβ family, and increased levels of this growth factor are generally associated with poorer healing outcomes and more significant scarring and fibrosis [30]. Previous studies have described the effect of differential Flii levels on TGF-β expression during wound healing [31]. Here we examined the effect if differential Flii levels on tenocyte TGF-β1 expression. Tenocytes from Flii +/− , WT, and Flii Tg/Tg mice were scratch-wounded and immunofluorescently stained for TGFβ1 expression at 3, 6, 12, and 24 h post-wounding ( Fig. 8a-l). TGFβ1 expression was significantly decreased across all time points in Flii Tg/Tg tenocytes compared to WT (p = 0.03, 0.0005, 0.003, and 0.01, respectively) and Flii +/− tenocytes (p = 0.0003, 0.00003, 0.00007, and 0.00004, respectively) ( Fig. 8a-m). Tenocytes showed a significant increase in expression between 3-6 h, suggesting secretion of TGFβ1 occurs early in the healing process, with a gradual decline in expression over the remainder of the experiment (Fig. 8m).

Discussion
Tendon injuries represent a multifaceted model of cellular change, requiring a complex re-organization of the actin cytoskeleton in order to facilitate essential processes for healing including migration, proliferation, contraction, and secretion. Tenocytes make up the majority of the cellular content in tendons and are integral in coordinating the healing process following injury [32]. Tenocytes are responsible for releasing signaling molecules to recruit a variety of growth factors, cytokines, and inflammatory cells to the injury site in order to facilitate successful healing [32]. It is well established that both tenocytes and dermal fibroblasts are of mesodermal origin, and both cell types exhibit similar spindle-shaped morphology when grown in in vitro culture. Both produce major extracellular matrix proteins and growth factors in response to injury. For these reasons, tenocyte function is often compared to dermal fibroblasts; however, it is important to note that they are distinct cell types. Tenocytes are mechanosensitive cells with unique properties that detect and respond to changes in mechanical load via deformation of their cell membrane and actin cytoskeleton [33]. Our findings that tenocytes and fibroblasts exhibit different cellular responses to changes in Flii gene expression aligns with previous studies that have shown tenocytes behave differently to dermal fibroblasts in vitro [34] as well as previous studies that show Flii positively affects digit regeneration and rodent vibrissa healing [35,36].
Previous studies have shown that the upregulation of Flii is associated with impaired migration, proliferation, and adhesion of both fibroblasts and keratinocytes [8,17]. Flii has previously been reported to act as an intracellular protein that affects actin remodeling and cytoskeleton organization through association with different signaling proteins [8,11,13]. Importantly, Flii translocates from the cytoplasm to the nucleus in response to wounding in fibroblasts but not keratinocytes. This suggests that the cell-specific nuclear translocation of Flii may directly regulate gene expression in fibroblasts but not keratinocytes, providing a potential mechanism of action for Flii in wound healing [31]. Using an inducible ROSA26 system to conditionally over-express Flii in fibroblasts resulted in impaired healing responses suggesting that fibroblastspecific production of Flii contributes to wound impaired healing [37]. In this study, increased Flii expression in tenocytes led to an improvement in tenocyte migration, proliferation, and outgrowth, providing more support for the important cell-specific and cell-type-dependent functions of Flii.
Cellular migration is an essential part of the healing process, allowing cells to move across a space created by an injury, to re-establish continuity and strength to an injured area. In order for cells to migrate, a complex cascade of events occurs, with actin assembly driving lamellipodia and filopodia forward, and adhesions forming to anchor the cell body to the surface [9]. Rear adhesions disassemble to allow the cell to retract its rear edge and the cell to move forward. Actin reassembly is integral to this process and actin remodeling proteins such as Flii are essential to ensure this progression occurs correctly [9]. Flii Tg/Tg tenocytes showed significantly improved migration and proliferation over 24 h compared with WT and Flii +/− counterparts, indicating the important role of Flii in regulating tenocyte adhesion, migration, and proliferation. These findings are in agreement with previous studies showing Flii association with different signaling proteins involved in actin rearrangement including Cdc42 and the Ras-dependent adenylyl cyclase pathway [15,38]. Additionally, studies have shown that Flii regulates focal adhesion turn-over and cell migration by inhibition of paxillin phosphorylation via a Rac-1-dependent pathway [13,39] and its ability to cap actin filaments [40]. Experiments presented in this manuscript show the effects of Flii on tenocyte cell function, further supporting the role of Flii in regulating cellular responses during homeostasis and in response to wounding.
Fibroblasts and tenocytes are both of mesodermal origin and both display similar morphology in vitro as well as being producers of major extracellular proteins such as collagen [41]. Collagen type I is the main component of tendons and tendon adhesions and is an essential protein synthesized during the healing process, providing structure and tensile strength. No significant difference in collagen type I production was observed in the tenocytes while fibroblasts secreted significantly more collagen type I. Overexpression of Flii in Flii Tg/Tg fibroblasts significantly decreased collagen type I production when compared to fibroblasts from WT and Flii +/− mice. Again, there was no significant difference in collagen type III secretion seen between tenocytes across all three genotypes while fibroblasts showed an overall significant increase in collagen type III production. Flii does have an effect on both collagen type I and collagen type III production as seen in fibroblasts, both in this study and previously [8]. It is, however, the lack of collagen production seen in tenocytes suggest that these cells are less metabolically active than fibroblasts during the "stationary phase" of cell mitosis; this is supported by Stoll et al. that showed slow proliferation and unstable tenocyte phenotype in monolayer cultures [42]. In response to wounding, collagen production in tenocytes is increased and the ratio of collagen type I and collagen type III is also increased [43]. One limitation of this study is the lack of in vivo studies and future in vivo studies in Flii genetic mice will provide more information about the potential effect of differential Flii levels on collagen type I and collagen type III production and organization during the healing of tendon injuries. ECM disorganization is known as a major hallmark in tendinopathy when a significant decrease in type I collagen is prevalent [43]. No significant difference was observed between Flii +/− , WT, and Flii Tg/Tg tenocytes in a 3D collagen gel contraction model suggesting that Flii does not affect tenocyte contraction. Interestingly, this was opposite to previous observations on the effect of Flii on fibroblast collagen gel contraction [16]. Embryonic tenocytes have actinrich fibripositors which act via contraction in order to form the "crimp" structure seen in established tendons. This structure is vital for the mechanical shockabsorbing behavior of mature tendons [44,45]. Adult tendons show minimal contraction during healing and the "crimp" structure tends to be lost in the reestablished tendon after healing is complete. So, although Flii appears to have minimal effect on adult tenocyte contraction in vitro, Flii is an important developmental regulator [8,21] and has been shown to be involved in fibroblast contraction [16], and so further investigations into its role in embryonic tenocyte cell contraction should be examined.
Many studies have shown that TGFβ1 is a regulator of fibrosis, and excess TGFβ1 levels have been associated with poorer healing outcomes and excessive scarring [46,47]. This is associated with the inflammatory phase of healing and prolonged expression of TGFβ1 can lead to an extension of the inflammatory phase which could contribute towards poorer healing outcomes. In a cutaneous burn injury model, elevated Flii leads to increased TGFβ gene expression and reduced Flii levels are linked with decreased TGFβ1 expression leading to better healing outcomes [48]. To investigate the possible mechanisms behind which overexpression of Flii in tenocytes may affect collagen deposition, the expression of TGFβ1 in response to in vitro scratch wounding of tenocytes was investigated. Flii Tg/Tg tenocytes expressed significantly less TGFβ1 compared with their WT and Flii +/− counterparts. TGFβ1 has been shown to increase collagen type I production resulting in excessive and disordered organization, leading to increased tendon adhesion size and severity [7]. Flii Tg/Tg tenocytes expressed significantly lower TGFβ1 levels during migration, and at 12 h post-scratch wounding, these cells had significantly lower collagen type I levels than their WT and Flii +/− counterparts. Decreased levels of pro-scarring TGFβ1 coupled with decreased and more controlled production of type I collagen, by Flii Tg/Tg tenocytes, suggest a potential positive role for Flii on healing of tendon injuries.

Conclusions
Fibroblast behavior has been extensively studied in cutaneous healing experiments, and a reduction in Flii results in improved fibroblast migration, proliferation, adhesion, and contraction [8,16]. The opposite is observed in tenocytes in this study suggesting cell-specific effects of Flii may occur in response to different stimuli. This study clearly suggests a positive role for Flii in regulating tenocyte proliferation and migration. In conjunction with decreased collagen type I and TGF-β production post wounding, this identifies Flii as a novel target for modulating tenocyte activity and for improving tendon repair. While the exact mechanism of Flii effects on tendon repair is still to be determined, the outcome of this study may help to advance new therapeutic approaches aimed at improving the healing of tendon injuries.

Funding
This work was supported by an Australian Postgraduate Award scholarship to JEJ. ZK is supported by a Foundation Fellowship from University of South Australia. AJC is supported by NHMRC Senior Research Fellowship GNT#1102617.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.