Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Sep 1.
Published in final edited form as: Dev Biol. 2015 Jun 30;405(1):96–107. doi: 10.1016/j.ydbio.2015.06.020

GDF5 PROGENITORS GIVE RISE TO FIBROCARTILAGE CELLS THAT MINERALIZE VIA HEDGEHOG SIGNALING TO FORM THE ZONAL ENTHESIS

Nathaniel A Dyment 1, Andrew P Breidenbach 2, Andrea G Schwartz 3, Ryan P Russell 1, Lindsey Aschbacher-Smith 4, Han Liu 4, Yusuke Hagiwara 5, Rulang Jiang 4, Stavros Thomopoulos 3, David L Butler 2, David W Rowe 1
PMCID: PMC4529782  NIHMSID: NIHMS708924  PMID: 26141957

Abstract

The sequence of events that leads to the formation of a functionally graded enthesis is not clearly defined. The current study demonstrates that clonal expansion of Gdf5 progenitors contributes to linear growth of the enthesis. Prior to mineralization, Col1+ cells in the enthesis appose Col2+ cells of the underlying primary cartilage. At the onset of enthesis mineralization, cells at the base of the enthesis express alkaline phosphatase, Indian hedgehog, and ColX as they mineralize. The mineralization front then extends towards the tendon midsubstance as cells above the front become encapsulated in mineralized fibrocartilage over time. The hedgehog (Hh) pathway regulates this process, as Hh-responsive Gli1+ cells within the developing enthesis mature from unmineralized to mineralized fibrochondrocytes in response to activated signaling. Hh signaling is required for mineralization, as tissue-specific deletion of its obligate transducer Smoothened in the developing tendon and enthesis cells leads to significant reductions in the apposition of mineralized fibrocartilage. Together, these findings provide a spatiotemporal map of events – from expansion of the embryonic progenitor pool to synthesis of the collagen template and finally mineralization of this template – that leads to the formation of the mature zonal enthesis. These results can inform future tendon-to-bone repair strategies to create a mechanically functional enthesis in which tendon collagen fibers are anchored to bone through mineralized fibrocartilage.

Keywords: Enthesis, lineage tracing, GDF5, fibrocartilage, collagen, alkaline phosphatase, mineralization, maturation, hedgehog signaling, growth, mouse

INTRODUCTION

Muscles, tendons, and bones work in a coordinated fashion to yield efficient movement of the skeleton. These tissues possess a range of mechanical properties, structural organization, and embryological origins (Charvet, et al. 2012; Lu and Thomopoulos. 2013; Dyment, et al. 2014; Zelzer, et al. 2014). The interfaces that connect muscle to tendon (i.e., myotendinous junction) and tendon to bone (i.e., enthesis) are crucial for efficient load transfer and subsequent joint motion. However, stress is amplified at the enthesis due to the stiffness mismatch between relatively compliant tendon and stiff bone (Liu, et al. 2012; Liu, et al. 2014), which is thought to contribute to the prevalence of injuries at the enthesis (McGonagle. 2005; Benjamin and McGonagle. 2009; Lu and Thomopoulos. 2013). The enthesis is also a common site of chronic pathologies (McGonagle. 2005; Riley. 2008; Benjamin and McGonagle. 2009). In addition, inflammatory conditions such as spondyloarthropathies, including ankylosing spondylitis (Sivas, et al. 2009) and psoriatic arthritis (Mcgonagle. 2005), are localized to the enthesis (Sherlock, et al. 2012). The pathogenesis of these conditions is not fully understood, specifically regarding which cells within the enthesis are involved, but alterations in inflammation and mineralization are typical (McGonagle, et al. 2010). An improved understanding of the development and growth of the enthesis may provide insight into the etiology of these pathologies and lead to repair strategies to improve healing of these interfaces. The current paper will present an integrated spatiotemporal map of enthesis growth and maturation along with the cellular events that drive this process.

The differentiation of enthesis progenitors during normal growth and development is still not well understood and requires novel methods for elucidation. The attachment unit between tendon and the primary cartilage is defined during late fetal time points. Progenitors of the tendon midsubstance express scleraxis (Scx), while progenitors of underlying primary cartilage express the SRY-related transcription factor Sox9. At the interphase of these two populations, a unique enthesis progenitor pool has been identified that co-express Scx and Sox9 (Soeda, et al. 2010; Blitz, et al. 2013; Sugimoto, et al. 2013). Enthesis cells during this timeframe also express growth and differentiation factor (Gdf5), unlike cells in the underlying primary cartilage (Rountree, et al. 2004; Dyment, et al. 2014). At later embryonic stages, enthesis cells express Gli1 and cells within unmineralized regions of the enthesis maintain this expression during later stages of postnatal growth (Schwartz, et al. 2015). At maturity, a cell phenotype gradient forms between tendon and bone with an extracellular matrix that shares characteristics from tendon and cartilage (e.g., type I and type II collagens, aggrecan, and tenascin-C) (Galatz, et al. 2007; Liu, et al. 2011; Wang, et al. 2012). To better delineate the processes of enthesis formation, particularly in comparison to the adjacent tendon and primary cartilage, requires novel reporter mice (e.g., collagen I/II/X fluorescent reporter mice), multiple imaging modalities (e.g., two photon collagen imaging), and histomorphometry (e.g., quantification of mineral apposition via fluorescent labeling).

Better understanding of the signaling pathways that regulate enthesis development is needed to develop improved therapies for tendon-to-bone repair in adults. Recent work has established that hedgehog signaling is crucial for proper enthesis cell differentiation and maturation (Liu, et al. 2011; Liu, et al. 2013; Schwartz, et al. 2015; Breidenbach, et al. 2015). Hedgehog gain-of-function in tendon cells leads to expression of fibrocartilaginous matrix proteins within the tendon midsubstance while loss-of-function leads to reductions of these proteins within the enthesis and reduced mineralization (Liu, et al. 2013; Schwartz, et al. 2015; Breidenbach, et al. 2015). While Indian hedgehog (IHH) expression has been shown at neonatal time points, it is still unknown whether IHH is the prominent Hh ligand that is associated with mineralization, which we will demonstrate in this study. In addition, the hedgehog-responsive Gli1 population, while expressed in multiple cell layers of the enthesis during early development, becomes more restricted to regions of unmineralized fibrocartilage with age (Schwartz, et al. 2015). However, the fate of these Gli1+ unmineralized fibrochondrocytes is not clearly defined relative to the mineralization patterns in the maturing enthesis. Thus, the current study will examine the fate of enthesis cells from proliferative progenitors to mature mineralized fibrochondrocytes and demonstrate that IHH is involved in the transition from unmineralized to mineralized fibrochondrocytes.

The objective of the current study is to provide a cohesive map of the coordinated sequence of events in enthesis growth and development, including expansion of the progenitor pool, spatiotemporal changes in matrix synthesis, cell maturation from unmineralized to mineralized fibrochondrocytes, and signaling pathways that regulate these processes (all of which are summarized in figure 7). Using complementary studies as outlined in the experiment design (Fig. S1), we will establish that columnar expansion of Gdf5 progenitors contributes to enthesis growth. In addition, we will demonstrate that enthesis cells alter their collagen expression from type I to type II and finally type X as they mature into mineralized fibrochondrocytes. We will also reveal that IHH expression coincides with alkaline phosphatase activity and active mineral deposition. Finally, we will show that Gli1+ unmineralized fibrochondrocytes will mineralize in response to activated hedgehog signaling to create the mineralized fibrocartilage zone of the enthesis and disruption in Hh signaling will significantly inhibit their ability to mineralize their surrounding matrix.

Figure 7.

Figure 7

Open in a new tab

Model for cellular expansion and maturation during enthesis growth and development. Based on our current work and the work of others, we suggest five stages of maturation within the enthesis during growth and development, with cell markers for each stage. A) At condensation, there are three distinct progenitor populations at the region where tendons attach to cartilage: midsubstance tendon progenitors (TP), enthesis progenitors, and primary chondrocytes of the epiphysis (PC). The enthesis progenitors give rise to unmineralized fibrochondrocytes, mineralizing fibrochondrocytes, and mineralized fibrochondrocytes of the more mature enthesis. B) After condensation, these progenitors clonally expand and cells at the base begin to mature into unmineralized fibrochondrocytes. C) Prior to the onset of mineralization, the mineralizing fibrochondrocytes at the base express AP and ColX as they begin to mineralize. D) This leads to the active mineralization stage where mineralization can be detected. E) Mineral apposition continues and cells that are embedded within the mineralized fibrocartilage express ColX. These sequential events lead to a maturation gradient with the most mature cells at the bottom and immature cells at the top (closer to midsubstance). References: (1) (Sugimoto, et al. 2013), (2) (Blitz, et al. 2013), (3) (Soeda, et al. 2010), (4) (Rountree, et al. 2004), (5) (Dyment, et al. 2014), (6) (Schwartz, et al. 2015), (7) (Liu, et al. 2013), and (8) (current study).

MATERIALS AND METHODS

Experimental Design

The sequence of events from expansion of the enthesis progenitor pool to the development of mineralized fibrocartilage is outlined in figure S1. The patellar (PT), supraspinatus (ST), and Achilles (AT) tendons were investigated in all growing mice to monitor the temporal differences of enthesis development in these clinically relevant tendons. Each study included a time point at the onset of mineralization in the different entheses (2–2.5 weeks of age) to investigate these responses prior to and after mineralization. (1) GDF5Cre-Confetti mice were examined at 2 weeks of age (n = 4) to measure the clonal capacity of the enthesis progenitor pool. (2) Col1/2/10 mice were examined at postnatal day 1, 2 weeks, and 4 weeks of age (n = 3/group) to demonstrate the spatiotemporal expression patterns of these collagens prior to and after the onset of mineralization at 2 weeks of age. (3) Co3.6-cyan:ColX-cherry mice were injected with mineralization labels at 2.5 (demeclocycline), 4.5 (calcein), 6.5 (alizarin complexone), 8.5 (demeclocycline) weeks of age (Fig. S1) to measure the direction and mineralization apposition rate of the different entheses in this study. Mice were investigated at 4.5 and 10.5 weeks old (n=4/group). (4) To determine the fate of unmineralized fibrochondrocytes, Gli1-Ai14 mice were injected with tamoxifen and calcein at 4.5 weeks of age and sacrificed at 3 days or 4 weeks post-injection. Demeclocycline was also delivered to the latter group one day prior to sacrifice (n=3–5/group) to determine the position of the Gli1-labeled cells within respect to the advancing mineralization front. (5) ScxCre:Smof/− mice and their littermate controls were injected at 2 (demeclocycline), 4 (calcein), 6 (alizarin complexone), 8 (demeclocycline) weeks of age and sacrificed at 4 and 10 weeks of age (n=4–6/group) to determine if Smo affected the onset of mineralization and mineral apposition rate of fibrocartilage within the enthesis during growth.

Mouse Models

University animal use and care committees approved all protocols at UConn Health Center (UCHC), Washington University (WU), and Cincinnati Children’s Hospital and Medical Center (CCHMC). Five different groups of transgenic mice were used in the study across 3 institutions. (1) GDF5Cre-Confetti. In order to measure the clonal expansion of enthesis progenitors, constitutive GDF5Cre mice generously provided by Dr. David Kingsley (Rountree, et al. 2004; Dyment, et al. 2014) were crossed with R26R-Confetti mice (Gt(ROSA)26Sortm1(CAG-Brainbow2.1)Cle/J, Jackson Labs). The R26R-Confetti mice harbor a CAG promoter followed by a floxed STOP cassette and downstream brainbow 2.1 sequence all within the Rosa26 locus. Following Cre-mediated removal of the stop cassette, four potential recombination events within the brainbow 2.1 sequence will result in expression of one of four fluorescent proteins (mCFP, RFP, nGFP, or YFP) that is retained within the progeny of the cells. Therefore, the number of GDF5 clones can be determined. (2) Col1/2/10 triple (UCHC). To monitor the spatiotemporal expression of important fibrocartilage collagen types during enthesis maturation, three GFP reporter mouse strains for Col1a1 (Col3.6-Tpz) (Kalajzic, et al. 2002), Col2a1 (Col2-cyan) (Chokalingam, et al. 2009), and Col10a1 (ColX-cherry) (Maye, et al. 2011) were crossed to make Col1/2/10 triple transgenic mice. (3) Col3.6-cyan:ColX-cherry (UCHC). Double transgenic mice with Col1a1 driving CFP (Col3.6-cyan) crossed with ColX-cherry were used in study 3 (see experimental design above) involving multiple mineralization labels during growth. Col3.6-cyan was substituted for Col3.6-Tpz to reduce fluorescence interference with mineralization labels. (4) Gli1-Ai14 (WU). Tamoxifen inducible Gli1-CreERT2 mice (Gli1tm3(cre/ERT2)Alj/J, Jackson Labs) were crossed with Ai14-tdTomato Cre reporter mice (B6;129S6-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J, Jackson Labs) to trace Gli1+ cells in the enthesis during postnatal growth and mineralization. (5) ScxCre:Smof/− (SmoKO; CCHMC). Constitutive ScxCre mice generously provided by Dr. Ronen Schweitzer were crossed with Smof/f mice (Smotm2Amc/J, Jackson Labs) to target deletion of hedgehog signaling in tendons, ligaments, and their entheses (Liu, et al. 2013). Smof/+, Smof/−, or ScxCre:Smof/+ littermates were used as controls.

Mineralization Labeling

Intraperitoneal (IP) injections of demeclocycline (30μg/g), calcein (10μg/g), and alizarin complexone (30μg/g) made up in 2% NaHCO3 (pH = 7.4) were delivered to mice based on the experimental design (Fig. S1) to determine the fibrocartilage mineral apposition rate during growth.

Tamoxifen Delivery

Tamoxifen dissolved in corn oil was delivered via sub-cutaneous injection (200μg/g) to Gli1-Ai14 mice.

X-Rays

Longitudinal x-rays in sagittal plane were taken to measure bone and patellar tendon length in hindlimbs of Col3.6-cyan:ColX-cherry mice at 2.5, 4.5, 6.5, 8.5, and 10.5 weeks of age to correlate tendon growth (distance between patella and tibial tuberosity) with bone growth. Mice were anesthetized with isoflurane and imaged using the Faxitron LX 60 X-ray machine. The calcaneus, tibia, femur, patella, and PT lengths were measured using Fiji image analysis software (Schindelin, et al. 2012).

Cryohistology

Forelimbs and hindlimbs were fixed in 10% neutral buffered formalin for 1–2 days, transferred to 30% sucrose in PBS overnight, and embedded in Shandon cryomatrix (Thermo Scientific). The knee was cut in the sagittal plane to capture the PT and in the coronal plane to capture the articular cartilage in plane with the medial collateral ligament. The ankle was cut in the sagittal plane to investigate the Achilles tendon and the shoulder was sectioned in coronal plane to investigate the enthesis of the supraspinatus tendon. All sections were made from undecalcified joints using cryofilm 2C (Ushiku, et al. 2010; Dyment, et al. 2013; Dyment, et al. 2014), which maintains morphology of mineralized sections. The taped sections were glued to microscope slides using chitosan adhesive and rehydrated prior to imaging.

Immunohistochemistry

Antigen retrieval was performed using the acidic TRAP buffer for 15 minutes, which also decalcified the section. Following 3 washes in 1X PBS, the sections were blocked in Power Block (Biogenex) for 30 minutes and incubated with primary antibody against IHH (sc-1196; Santa Cruz; 1:25 dilution) in 1% BSA overnight at 4°C. Following washes, the slides were stained with fluorescent secondary antibody (Alexa Fluor, Life Technologies) for 1 hour at room temperature.

Tartrate-resistant acid phosphatase (TRAP) staining for osteoclasts

Sections were incubated in TRAP buffer (0.92% sodium acetate anhydrous, 1.14% L-(+)tartaric acid, 1% glacial acetic acid – pH 4.1 ~ 4.3) for 15 minutes then incubated with ELF97 substrate (Life Tech, E6589) in buffer for 5 minutes. The ELF97 substrate generates a yellow fluorescent signal when cleaved by TRAP.

Alkaline phosphatase staining

To measure regions of active mineralization, alkaline phosphatase (AP) staining was performed using the Vector blue alkaline phosphatase substrate kit (Vector Labs) according to manufacturer protocols. The sections were incubated in the substrate solution for 15 minutes. The AP signal was imaged using a Cy5 filter.

Imaging

Up to 6 rounds of imaging were performed on each section to correlate the spatial distribution of several response measures with respect to developmental stages prior to and after the onset of mineralization. This sequence was possible because the cryofilm tape adheres to the tissue and allows for the coverslip to be removed between imaging steps without damaging the section. The order of imaging included: (1) fluorescent reporters and fluorescent mineralization labels (tissue still mineralized), (2) second harmonic generation (SHG) signal generated from collagen on the two photon microscope, (3) antibody staining following antigen retrieval (acid), which also decalcified the tissue, (4) TRAP staining, (5) alkaline phosphatase staining with dapi counterstain, and 6) toluidine blue O (0.025% in dH2O) staining.

To demonstrate the position of the different response measures with respect to the collagen architecture in the enthesis, two photon SHG collagen imaging was performed on a Prairie Ultima IV multiphoton microscope with a Ti:Sapphire laser tuned to 890 nm and SHG signal was acquired through a 435–485 nm bandpass filter. Epifluorescent and brightfield imaging were performed on the Zeiss Axio Scan.Z1 with chroma filters for each distinct fluorophore (see supplemental table).

Mineral apposition rate measurement

Distances between mineralization labels were quantified using Fiji image analysis software to measured mineralized cartilage apposition rate (distance/time) and cumulative mineralized cartilage apposition.

Collagen fiber orientation

The collagen SHG channel from the two photon imaging was analyzed using the OrientationJ plugin in Fiji to create a color survey of fiber orientation in order to better visualize the vertical fibers of the enthesis vs. the horizontal fibers of the subchondral bone (Fig. 3). A Gaussian window of 7 pixels was applied (image resolution - 0.65μm/pixel) and a structure tensor was calculated for each window to create the color survey.

Figure 3.

Figure 3

Open in a new tab

Mineral apposition of the enthesis and subchondral bone occur in opposite directions at the attachment of the supraspinatus tendon to bone. Mice (4.5 weeks old) were injected with demeclocycline (yellow) at 2.5 weeks of age and calcein (green) one day prior to sacrifice. (A1) T Blue stained section of the supraspinatus tendon. (A2) T Blue image with mineralization labels (arrows indicate direction of mineral apposition: green – enthesis, red – subchondral bone). (A3) & (B2–3) Two photon SHG signal for collagen showing vertical alignment of fibers of the enthesis vs horizontal orientation of subchondral bone in relation to the mineral labels (yellow and green). (B3) Collagen orientation is color coordinated using OrientationJ plugin in Fiji. Mineral labels are perpendicular to collagen in the enthesis and parallel to collagen in bone. (B1–3) Higher magnification of ROI in A1. Scale bars = 200μm.

Statistics

The number of GDF5 clones, mineral apposition rate, and total mineral apposition in different tendons (PT, AT, ST) were compared via Mann Whitney U tests with significance level set to p < 0.008 to account for multiple comparisons. MAR between control and ScxCre:Smof/− mice were compared via Mann Whitney U test with significance level set to p < 0.05.

RESULTS

GDF5-labeled progenitors clonally expand to contribute to linear growth of the enthesis

Using GDF5Cre x R26R-tdTomato mice (Dyment, et al. 2014), previous work demonstrated that cells within all zones of the entheses of the patellar, Achilles, and supraspinatus tendons originated from a Gdf5 population. However, the dynamics of the enthesis progenitor population, including the clonal potential of these cells, have not been demonstrated. Therefore, we crossed GDF5Cre mice with R26R-Confetti mice to measure the number and positioning of clones within this population as they contribute to growth of the enthesis.

GDF5-labeled cells exhibited clonal expansion within the entheses of the patellar, Achilles, and supraspinatus tendons (Fig. 1). There was no difference in the relative number of clones in the different tendons (p > 0.05). Therefore, the cell counts were pooled across tendons. Over 58% of the GDF5-labeled cells were within clones of 2 or more cells. In fact, 12% of the GDF5-labeled cells were within clones of 4 or more cells, 13% were within clones of 3 cells, and 34% were within clones of 2 cells (Fig. 1B). The clones were oriented in vertical columns between collagen fiber bundles within the enthesis (brackets in Fig. 1C–D), indicating that cell proliferation was contributing to lengthening of the enthesis.

Figure 1.

Figure 1

Open in a new tab

GDF5 enthesis progenitors contribute to linear growth via clonal expansion. The entheses of GDF5Cre x R26R-Confetti mice were investigated at 2 weeks of age to assess the clonal potential of this population prior to the onset of mineralization. Following Cre recombination, the R26R-Confetti reporter results in 4 potential outcomes: nGFP (nuclear localization), YFP, RFP, and mCFP (membrane localization), which is maintained in their progeny (A). The GDF5 population expanded during growth with over 58% of the labeled cells being in clones of 2 or more cells and 12% of cells being in clones of 4 or more cells (B). The clones were oriented in a vertical fashion (cyan brackets in C–D) between collagen fiber bundles (SHG signal for collagen displayed in white in panels C2 & D2), as seen in the patellar (C1–2) and Achilles (D1–2) tendons. Panels C1 & D1 display nGFP, YFP, RFP, and mCFP reporters. Panels C2 & D2 depict fluorescent proteins with SHG collagen signal (white). PT – patellar tendon, AT – Achilles tendon, T – tibia, C – calcaneus. Scale bars = 100μm. Error bars indicate SD.

Enthesis cells express Col1a1 followed by Col2a1 and finally Col10a1 as they differentiate into mature mineralized fibrochondrocytes

Type I, type II, and type X collagens are expressed within different regions of the enthesis. However, the specific timing of their expression relative to the formation of the different zones of the enthesis is not clearly defined. Therefore, we utilized Col1/2/10 triple transgenic mice to measure expression of these critical enthesis collagens prior to, near, and after the onset of mineralization (postnatal day 1, 2 weeks of age, and 4 weeks of age, respectively). Prior to mineralization, Col1-Tpz+ cells were positioned within the tendon midsubstance and enthesis leading up to Col2-cyan+ cells of the primary cartilage (Fig. 2A1, 3). Collagen fibers, visualized via two photon second harmonic generation (SHG), ran between the Col1+ cells and anchored into the primary cartilage (Fig. 2B5; white SHG for collagen). By two weeks of age in the Achilles tendon, the cells at the base of the enthesis adjacent to the primary cartilage were beginning to mineralize as they expressed ColX-cherry (Fig. 2B3), alkaline phosphatase (Fig. 2B4; magenta) and IHH (Fig. 2B6; blue). This process yielded a maturation gradient with ColX cells at the base of the enthesis followed by Col2+ cells, then Col1/Col2+ cells, and finally Col1+ cells when moving towards the tendon midsubstance (Fig. 2B3). Mineralization of the enthesis coincided with secondary ossification of primary cartilage in the calcaneus as cells in the primary cartilage beneath the enthesis were also AP+ (Fig. 2B4). Like the AT, the ST displayed a strong band of AP activity at the base of the enthesis at two weeks of age; however, there were no detectable ColX cells (Fig. S2B2–3). In addition, there was no ColX or AP expression in the distal PT enthesis at 2-weeks of age (Fig. S2E2–3) as it mineralizes at a later age.

Figure 2.

Figure 2

Open in a new tab

Enthesis cells transition from expression of Col1 to expression of Col2 and ColX during mineralization process. Achilles tendons of Col1-Tpz/Col2-cyan/ColX-cherry triple transgenic mice were assessed at postnatal day 1 (A1–4), 2-weeks (B1–6), and 4-weeks of age (C1–4). Col1-Tpz+ cells of the enthesis interface with Col2-cyan+ cells of the primary cartilage at postnatal day 1 (A3). At 2 weeks of age, cells at the end of the collagen fibers at the base of the enthesis (B5, SHG for collagen in white) begin to mineralize indicated by expression of ColX-cherry (B1, B3), alkaline phosphatase (B2, B4; magenta), and IHH (B6; blue). By 4 weeks of age, the AP signal becomes concentrated to the tidemark (dotted line in C4) where active mineralization occurs, while ColX+ cells are located in mineralized fibrocartilage beneath the tidemark. t – tendon, e – enthesis, c – primary cartilage of calcaneus. Scale bars = 200μm.

At 4 weeks of age, AP activity was concentrated near the tidemark (dotted line in Fig. 2C4) in all tendons, which is the region of active mineralization. ColX+ cells were found in all 3 entheses studied within the mineralized fibrocartilage (Fig. 2C1, C3 and Fig. S2C, F). In contrast, Col1 expression was decreased in the midsubstance and enthesis at 4 weeks of age (Fig. 2C1 and Fig. S2C, F) and there were no detectable Col2+ cells in the enthesis, unlike the adjacent growth plate of the calcaneus (Fig. 2C1). Additionally, Col1+ cells were only found in the unmineralized regions of the enthesis at all time points in this study.

Mineral apposition of the enthesis and the subchondral bone occur in opposite directions

Next, we investigated the mineralization patterns of the collagen template during enthesis growth to measure the direction and rate of mineral apposition. Measuring dynamic cellular and extracellular matrix changes is difficult without points of reference that mark cells or tissue at different time points. To overcome this issue, multiple mineralization labels were injected to identify the mineralization front at different ages. Using these labeled references points, the direction and rate of mineral apposition can be assessed.

To determine the direction of mineralization in relation to the tissue architecture, we examined entheses from 4.5-week-old mice following a demeclocycline (yellow) injection at 2.5 weeks of age and a calcein (green) injection given the day before sacrifice. Focusing on the supraspinatus tendon insertion into the humerus, there were two parallel demeclocycline labels: one near the base of the enthesis and the other in the subchondral bone (Fig. 3A2–3, B1–2). The label in the enthesis was oriented perpendicular to collagen fibers that extend through the enthesis (Fig. 3B2–3). However, the label in the subchondral bone ran parallel to the collagen fibers (Fig. 3B2–3). The collagen fiber orientation shifted from vertical (red) in the enthesis to horizontal (teal) in the subchondral bone at the interface of these two tissues (Fig. 3B3). There were also 2 calcein labels: one at the tidemark in the enthesis and the other on the endocortical surface of the subchondral bone (Fig. 3A2–3, B1–2). Therefore, mineral apposition in the enthesis occurred from the base of the enthesis towards the tendon midsubstance while the underlying subchondral bone mineralized towards the marrow space of the epiphysis. The mineralized fibrocartilage of the enthesis also had a stronger toluidine blue signal compared to the subchondral bone (Fig. 3B1) due to increased proteoglycan content, which was another indicator (in addition to the change in collagen fiber orientation) distinguishing the enthesis from the underlying bone.

Mineralized fibrocartilage apposition rate is highest in the patellar tendon followed by the supraspinatus and Achilles tendons

After establishing the direction of mineral apposition within the enthesis, we sought to determine if the mineral apposition rate (MAR) differed among the PT, ST, and AT entheses from 2.5 to 10.5 weeks of age by measuring the distance between mineralization labels delivered at 2-week intervals (Fig. S1 & 4C). MAR was highest for the patellar enthesis of the PT at all time points (p < 0.008). The cumulative mineral apposition (Fig. 4D) of all 3 tendons displayed traditional growth curves with the ST and AT slowing dramatically over time. MAR approached zero by 6.5 weeks in the AT and 8.5 weeks of age in the ST (Fig. 4C–D). Total mineral apposition from 2.5 to 10.5 weeks was highest in the PT patellar insertion (342±32μm) followed by the PT tibial insertion (162±29μm), ST (82±12μm), and lastly the AT (58±5μm). All comparisons were significantly different among tendons for total mineral apposition (p < 0.008).

Figure 4.

Figure 4

Open in a new tab

Mineralized fibrocartilage apposition rate is highest in the patellar tendon (PT) followed by the supraspinatus (ST) and Achilles (AT) tendon entheses. Mice (10.5 week old) were given mineralization labels at 2.5 (demeclocycline, yellow), 4.5 (calcein, green), 6.5 (alizarin complexone, red), and 8.5 (demeclocycline) weeks of age. (A1–2) Radial mineral apposition was seen in fibrocartilage of the distal enthesis of PT in the anterior tibial tuberosity. (B1–2) Mineralized fibrocartilage apposition in AT enthesis, which was significantly less than the PT during this time period. The green arrows (A2, B2) indicate the measured mineral apposition in the enthesis reported in panels C–D. (C) Mineral apposition rate and (D) cumulative mineral apposition of fibrocartilage cells in the PT, ST, and AT entheses during the time periods investigated. Horizontal and vertical black bars indicate significant difference (p < 0.008). SDF – superficial digital flexor tendon. Scale bars = 200μm. Error bars indicate SD.

Following onset of mineralization, growth occurs on the anterior surface of the enthesis

Since cells within the enthesis mineralized towards the tendon midsubstance, we next sought to determine how the soft tissue above the mineralized cartilage grew. Specifically was growth concentrated to a certain region or did it occur equally across the enthesis? In order to test this we used the Col1-cyan reporter as an indicator of active collagen transcription and new matrix synthesis. In all three tendons analyzed in this study (PT, ST, and AT), there was higher Col1-cyan expression on the anterior surface compared to the posterior region of the enthesis (Fig. S3). The round cells within unmineralized fibrocartilage on the posterior half of the enthesis showed little Col1 expression (Fig. S3A3; red arrow) while more elongated cells on the anterior surface had higher Col1 expression (Fig. S3; white arrows). The increased Col1-cyan signal on the anterior surface in combination with the direction of mineralization towards this surface indicated that the enthesis thickened through collagen apposition on the anterior surface.

Limited resorption of the mineralized fibrocartilage occurs via osteoclasts from the epiphyseal bone marrow

In addition to recording mineral apposition rate in the entheses, we also monitored the organization of these labels over time to determine when or if the mineralized fibrocartilage was resorbed or remodeled during growth. We found that portions of the deepest region of the mineralized fibrocartilage were resorbed in some of the entheses, indicated by disruption of either the 2.5- (yellow) or 4.5-week (green) mineralization labels. Removal of regions of the labels was seen in the tibial tubercle where bone from the epiphysis invaded the fibrocartilage of the tuberosity as the animal aged (Fig. S4). Active remodeling occurred on these bone surfaces with both AP+ osteoblasts and TRAP+ osteoclasts (red arrows in Fig. S4A2–7). Small patches of mineralized fibrocartilage were also resorbed in the supraspinatus tendon in a similar fashion where osteoclasts from the marrow had eroded a pocket of the mineralized fibrocartilage (data not shown). However, there was no detectable remodeling by cells within the mineralized fibrocartilage of the enthesis as the distance between the 2.5-week and 4.5-week labels (i.e., yellow to green) was the same in both the 4.5- and 10.5-week-old animals investigated. Therefore, there was minimal turnover of mineralized fibrocartilage during normal stages of growth and any resorption observed in this study was caused by osteoclasts from the epiphyseal marrow. Therefore, the mineralized fibrocartilage that is produced during postnatal development is retained into adulthood.

Gli1+ enthesis cells mature from unmineralized to mineralized fibrochondrocytes during growth

Since previous studies have shown that Hh signaling is important in normal enthesis development (Liu, et al. 2013) and we found that mineral apposition occurred towards the midsubstance, we next sought to determine the location of Hh-responsive cells within the enthesis in relation to the mineralization front during growth. We conducted a pulse chase experiment where we injected calcein to mark the tidemark and tamoxifen to label Gli1-Ai14 cells within the enthesis of 4.5-week-old mice. We found that at 3 days post-labeling all of the Gli1+ cells were within the unmineralized regions of all four entheses and articular cartilage (Fig. 5A2 & S5A). Following a chase of 4 weeks, we delivered a demeclocycline label to define new mineralized fibrocartilage produced during this four-week interval. Gli1-labeled cells closest to the initial calcein line, which was deposited 4 weeks earlier, were now embedded within mineralized fibrocartilage (Fig. 5B3; arrow) or articular cartilage (Fig. S5B). These cells were between the calcein and demeclocycline labels, indicating that they matured from unmineralized to mineralized chondrocytes during this period.

Figure 5.

Figure 5

Open in a new tab

Gli1+ enthesis cells mature from unmineralized to mineralized fibrochondrocytes. Tamoxifen and calcein were delivered to Gli1-Ai14 mice at 4.5 weeks of age and sacrificed at 3 days and 4 weeks post-injection. (A1–3) Day 3 – Gli1-Ai14 cells (red) were within unmineralized regions of the proximal PT enthesis (distal to calcein label; arrow in A3). (B1–3) 4 weeks – Gli1-labeled cells adjacent to the calcein label (arrow in B3) were mineralized during this period as they were between the calcein and demeclocycline (yellow; given day before sacrifice) labels. Green arrow in B1 denotes direction of mineralization. Scale bars = 200μm.

ScxCre:Smof/− mice exhibit severely reduced mineralized fibrocartilage apposition during growth compared to littermate controls

Since Hh-responsive (Gli1+) cells were concentrated in the unmineralized fibrocartilage near the tidemark and then matured into mineralized fibrochondrocytes over time, we next sought to determine the effect of Hh signaling on the rate of mineralized fibrocartilage apposition. ScxCre:Smof/− (SmoKO) mice were used to target deletion of the downstream Hh activator Smo in tendons, ligaments, and their entheses. These mice were injected with mineralization labels from 2 to 10 weeks of age at 2-week increments to monitor the mineral apposition rate in the entheses (Fig. S1).

The mineral apposition rate of all entheses was severely impaired in the SmoKO mice compared to littermate controls (distal PT shown in figure 6 while proximal PT, ST, and AT shown in figure S6). The effect was most pronounced in the patellar insertion of the PT where total mineralized fibrocartilage apposition was only 2% of the littermate controls (p < 0.05; Fig. 6E). In addition, total mineral apposition in the tibial insertion of the PT was 20%, the AT was 22%, and the ST was 38% of their respective littermate controls (p < 0.05; Fig. 6E). Conditional deletion of Smo also affected alkaline phosphatase activity in the enthesis. The substantial band of AP activity concentrated across the tidemark in the control mice (Fig. 6A2, B2; white arrows) was disrupted and reduced in the SmoKO group, supporting the reduction in mineralized fibrocartilage apposition. However, the onset of mineralization was not delayed in the SmoKO mice as the 2-week demeclocycline label was found in both the control and SmoKO groups (Fig. S6), indicating that other factors besides Smo may activate mineralization. Finally, no significant difference was found in total mineralized cartilage apposition between these groups in the articular cartilage of the femur (Fig. 6E) because these cells were not targeted in the ScxCre mouse.

Figure 6.

Figure 6

Open in a new tab

Inactivation of Hh signaling leads to severe reductions in enthesis mineralized fibrocartilage apposition. Control (A, B) and SmoKO (C, D) mice were injected with mineralization labels at 2 (demeclocycline), 4 (calcein), 6 (alizarin complexone), and 8 (demeclocycline) weeks of age and sacrificed at either 4 (A, C) or 10 (B, D) weeks of age. Mineral apposition with separation of labels (white arrows in A1 & B1) as seen in the control mice was not found in the SmoKO group. In addition, alkaline phosphatase activity (magenta) was severely inhibited in the enthesis (arrows in A2 & B2). Total mineral apposition within this time period was measured by quantifying the distance between the labels. Total mineral apposition was 2%, 20%, 22%, and 38% of controls in the proximal PT, distal PT, AT, and ST entheses, respectively. Those tissues that show statistically significant difference between SmoKO and control are indicated by #. AC – articular cartilage. Error bars indicate SD. Scale bars = 200μm.

DISCUSSION

The current study provides a spatiotemporal map of the coordinated events (summarized in Figure 7) that lead to the formation of all zones of the enthesis from the embryo (i.e., Gdf5 labeling of progenitors) to the mature enthesis (i.e., once mineral apposition has stopped). During condensation, three distinct populations appear where tendons attach to the underlying cartilage: tendon midsubstance progenitors, enthesis progenitors, and primary cartilage progenitors (Soeda, et al. 2010; Blitz, et al. 2013; Sugimoto, et al. 2013; Schwartz, et al. 2015). The enthesis progenitors are situated between the tendon midsubstance and primary cartilage populations (Fig. 7A). They progress through several stages of differentiation during enthesis maturation. 1) They expand in columns to contribute to linear growth of the enthesis (Figs. 1 and 7B). 2) As they differentiate, these cells synthesize a collagen template that anchors into the underlying primary cartilage (Fig. 2B5) and produce proteoglycans (e.g., aggrecan and biglycan) (Galatz, et al. 2007; Liu, et al. 2011; Wang, et al. 2012) as they mature into unmineralized fibrocartilage cells (Fig. 7B–C). 3) At the onset of mineralization, the most mature cells at the base of the enthesis adjacent to the primary cartilage express AP and ColX as they mineralize the fibrocartilage (Figs. 2B and 7D). The mineralization continues in an appositional manner from the base of the enthesis towards the tendon midsubstance (Figs. 3–4 and 7E) (Tatara, et al. 2014). Hh signaling drives the mineralization process, as ablation of Smo leads to severe reductions in mineralized fibrocartilage apposition (Fig. 6) (Schwartz, et al. 2015; Breidenbach, et al. 2015). These 3 processes lead to the development of a mature zonal enthesis with a cell maturation gradient from tendon fibroblasts (Col1+) to unmineralized fibrochondrocytes (Col1+, Col2+, Gli1+) to mineralizing fibrochondrocytes (AP+, IHH+, ColX+) and finally mineralized fibrochondrocytes (ColX+) (Fig. 7).

Using the R26R-Confetti reporter mice, we demonstrated that GDF5-labeled progenitors proliferate in columns between collagen fiber bundles within the enthesis during growth (Fig. 1C–D), producing the stacked linear arrays seen in mature entheses. The clones are distributed throughout the enthesis and not concentrated to certain layers, as seen in the growth plate, suggesting that the enthesis lengthens via interstitial growth. Additionally, as demonstrated in the Col1/2/10 mice (Fig. 2), cells at the base of these stacks expressed alkaline phosphatase at two weeks of age (Fig. S7), indicating that these cells mature from the base of the enthesis towards the midsubstance. Unlike other tissues with rapid turnover, cell proliferation within the enthesis is limited, which makes identification of proliferating cells with traditional methods (e.g., BrdU, Ki67, etc.) difficult. However, the confetti mouse provided a method to identify clones and measure cell proliferation in a more sensitive manner, providing a better indicator of cell proliferation. Despite the sub-optimal recombination found in the R26R-Confetti model (Kaukua, et al. 2014), we demonstrated that GDF5-labeled progenitors have the capacity to produce clones of 4 or more cells by two weeks of age (Fig. 1B) but this capacity is still lower than tissues with higher turnover (e.g., growth plate, intestinal crypt, epidermis, etc.) (Snippert, et al. 2010; Mascre, et al. 2012; Yang, et al. 2014).

Enthesis progenitors progress through several stages of differentiation, as they become terminal mineralized fibrochondrocytes. These stages include synthesizing a collagen template and then mineralizing the template. As the cells progress through the lineage they change expression from Col1a1 to Col2a1 prior to mineralizing and then Col10a1 after mineralizing (Figs. 2 and 7). Beyond the mechanical function of these collagen types in the enthesis, they also mediate mineral deposition (Kirsch, et al. 2000; Chen, et al. 2015). Aspects of the mineralization process are regulated by Hh signaling (Liu, et al. 2013; Schwartz, et al. 2015; Breidenbach, et al. 2015) as Gli1+ unmineralized fibrochondrocytes responded to Hh signaling by mineralizing the surrounding matrix (Fig. 5). While IHH was shown to be expressed by cells of the neonatal enthesis prior to the onset of mineralization (Liu, et al. 2013), we demonstrated that IHH expression correlated with AP activity at the onset of mineralization (Fig. 2B4, 6) and is likely the prominent Hh ligand in enthesis mineralization. In fact, hedgehog signaling is a key contributor to mineralized fibrocartilage apposition in the enthesis, as inactivation of Hh signaling via conditional deletion of the activator Smo led to severe reductions in mineralized fibrocartilage apposition (Fig. 6). The largest reduction in mineral apposition was seen in the proximal enthesis of the PT. Since this ScxCre model has been shown to target the primary cartilage of the patella as well as the PT (Blitz, et al. 2013; Eyal, et al. 2015), this model may have had a combinatory effect leading to the largest reduction in this enthesis. These severe reductions in mineral apposition are consistent with prior reports examining the role of Hh signaling in enthesis development (Schwartz, et al. 2015; Breidenbach, et al. 2015).

While mineral apposition within the enthesis was severely impaired in these ScxCre-Smof/f mice, the onset of mineralization was not delayed as the KO samples contained the initial 2-week label in similar locations as the littermate controls (Fig. S6). Therefore, it is likely that Hh signaling does not regulate the initiation of mineralization, but rather controls the location of the mineralization front, the development of a mineral gradient, and/or the accumulation of mineral. Furthermore, Hh signaling may regulate proteins that participate in mineralization: Gli1 can control expression of osteopontin (Yoon, et al. 2002), a known regulator of mineral growth, and IHH can regulate Col10a1 expression via Runx2/Smads interactions (Amano, et al. 2014), which promotes mineralization through its interaction with matrix vesicles. In fact, these mice display smaller crystal size (Schwartz, et al. 2015) in the mineralized fibrocartilage, suggesting that crystal growth was impaired. These findings may have implications on repair strategies, as hedgehog signaling may not be sufficient to create mineralized fibrocartilage.

The function of Hh signaling within the enthesis may have important implications for understanding and potentially treating several pathologies of the joint. For instance, increased mineralization within the enthesis occurs in ankylosing spondylitis, psoriatic arthritis, and mineralizing enthesopathy along with increases in inflammation (Mcgonagle. 2005; Sivas, et al. 2009; Karaplis, et al. 2012). This increased mineralization could be mediated by Hh signaling among other signaling pathways and often is accompanied with increases in alkaline phosphatase activity (Karaplis, et al. 2012). In addition, IHH is increased in arthritic cartilage and conditional knockout of IHH in murine models of post-traumatic osteoarthritis yields significant reductions in cartilage degradation (Lin, et al. 2009; Zhou, et al. 2014). The likely mechanism is that matrix degradation is required in the process of cartilage mineralization and is manifested as proteoglycan loss in arthritic cartilage. In fact, IHH has been shown to regulate Runx2, which in turn can increase expression of ADAMTS5, a primary aggrecanase seen in arthritic cartilage (Lin, et al. 2009). Therefore, coordinated regulation of cartilage mineralization and maintenance of a mineralization front at its appropriate position are likely crucial to attenuating the progression of these joint pathologies. However, the relationship between inflammation and cartilage mineralization, which both occur in these pathologies, is not clear. Future therapies to regulate cartilage mineralization in addition to current therapies to attenuate inflammation may provide potent treatment to these debilitating conditions.

The three tendons investigated in this study showed significant timing differences in both the onset of mineralization and the mineral apposition rate (Figs. S2 & 4). The patellar tendon showed the highest mineralization rate followed by the supraspinatus and Achilles tendons. While the patellar tendon showed the highest mineral apposition rate during the investigated time period, the onset of mineralization in the PT tibial enthesis was also delayed compared to the Achilles tendon and supraspinatus tendons. The 2.5 week old demeclocycline label in the Col3.6-cyan:ColX-cherry mice displayed as a diffuse signal near the nucleation point of mineralization in the anterior tibial tubercle while a sharp mineralization label was found in the Achilles tendon (Fig. 4A2 vs B2). This finding indicated that the PT tibial enthesis was just starting to mineralize at 2.5 weeks of age, while the Achilles and supraspinatus began to mineralize earlier. In addition, the label given at 2 weeks of age in the control littermates in the SmoKO study showed strong signal in the Achilles and supraspinatus, but weak, if any, signal in the PT tibial enthesis. The onset of mineralization was also associated with secondary ossification of the attached bone. Secondary ossification of both the calcaneus and humerus originate near the AT and ST entheses, respectively (Schwartz, et al. 2012). However, secondary ossification of the tibia occurs within the center of the epiphysis away from the tibial tubercle. Therefore, there may be cell communication between the enthesis and primary cartilage that coordinates the mineralization process.

Matrix turnover of mineralized fibrocartilage was extremely low in the entheses investigated. This result likely contributes to the inability of enthesis to heal effectively in the adult and is consistent with a recent report showing little remodeling at entheses 5 weeks after ablating Gli1+ fibrocartilage cells (Schwartz et al., 2015). In the current study, we did find limited examples of mineralized fibrocartilage resorption. These observations all occurred within small pits at the base of the enthesis where TRAP+ osteoclasts from the underlying bone marrow resorbed segments of the mineralized fibrocartilage (Fig. S4). The mechanism of enthesis mineralized fibrocartilage resorption was also similar to that of the articular cartilage where resorption occurred at the base of mineralized cartilage by osteoclasts from the epiphyseal marrow (Shibakawa, et al. 2005). It is noteworthy that osteoclastic bone remodeling only minimally invades the quiescent mineralized fibrocartilage, suggesting that factors exist within the mineralized fibrocartilage that protect this region from osteogenic invasion.

CONCLUSIONS

Considerable research has focused on describing the organization, composition, and mechanical properties of the mature zonal enthesis (Ralphs, et al. 1998; Benjamin and McGonagle. 2009; Lu and Thomopoulos. 2013). Since the enthesis is a common site of injury that heals poorly, even after surgical repair, recapitulation of the functionally graded architecture of the native tissue is of major clinical importance. With an improved understanding of the basic molecular events that are required to create a phenotypically and functionally graded zonal enthesis, the next challenge will be to develop therapies that recapitulate these aspects in an adult repair scenario. This study suggests that successful repair may require either tissue-resident or implanted progenitors to anchor collagen to bone, synthesize a fibrocartilaginous extracellular matrix (e.g., collagen and proteoglycans), and finally mineralize the fibrocartilage to produce a solid anchor. Future studies are needed to determine the appropriate cell source within the adult that is primed to develop these attachments and then determine which signals should be delivered to these cells such that they mature properly to create the appropriate gradient needed for a functional attachment.

Supplementary Material

supplement

Highlights.

  • Enthesis progenitors expand within vertical columns to contribute to linear growth

  • Cells express type I, type II, and type X collagen sequentially during maturation

  • Indian hedgehog expression correlates with alkaline phosphatase activity

  • Gli1+ cells mineralize in response to hedgehog signaling

  • Deletion of Smoothened reduces mineral apposition but not onset of mineralization

Acknowledgments

The authors would like to thank Dr. David Kingsley and Dr. Ronen Schweitzer for graciously providing the GDF5Cre and ScxCre mice, respectively. Thanks to Dr. Kamal Khanna for providing access to the Prairie Ultima IV multiphoton microscope, and Jianping Huang for mouse breeding assistance. This work was supported by NIH grants R01-AR54713, R01-AR052374, R01-AR055580, R01-AR056943, T90-DE021989 (support for NAD) and T32-AR060719 (support for AGS).

Footnotes

COMPETING INTERESTS

The authors declare no competing financial interests.

AUTHOR CONTRIBUTIONS

N.A.D., A.P.B., A.G.S., R.P.R, R.J., S.T., D.L.B., and D.W.R. developed the concepts or approach, A.P.B., A.G.S., R.P.R., L.A., H.L., and Y.H. performed experiments or data analysis, and N.A.D., A.P.B., A.G.S., R.P.R., H.L., Y.H., R.J., S.T., D.L.B., and D.W.R. prepared or edited the manuscript prior to submission.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Amano K, Densmore M, Nishimura R, Lanske B. Indian Hedgehog Signaling Regulates Transcription and Expression of Collagen Type X Via Runx2/Smads Interactions. J Biol Chem. 2014;289:24898–910. doi: 10.1074/jbc.M114.570507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Benjamin M, McGonagle D. Entheses: Tendon and Ligament Attachment Sites. Scand J Med Sci Sports. 2009;19:520–7. doi: 10.1111/j.1600-0838.2009.00906.x. [DOI] [PubMed] [Google Scholar]
  3. Blitz E, Sharir A, Akiyama H, Zelzer E. Tendon-Bone Attachment Unit is Formed Modularly by a Distinct Pool of Scx− and Sox9-Positive Progenitors. Development. 2013;140:2680–2690. doi: 10.1242/dev.093906. [DOI] [PubMed] [Google Scholar]
  4. Breidenbach A, Aschbacher-Smith L, Lu Y, Dyment NA, Liu CF, Liu H, Wylie C, Rao M, Shearn J, Rowe DW, Kadler KE, Jiang R, Butler DL. Ablating Hedgehog Signaling in Tenocytes during Development Impairs Biomechanics and Matrix Organization of the Adult Murine Patellar Tendon Enthesis. J Orthop Res. 2015 doi: 10.1002/jor.22899. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Charvet B, Ruggiero F, Le Guellec D. The Development of the Myotendinous Junction. A Review. Muscles Ligaments Tendons J. 2012;2:53–63. [PMC free article] [PubMed] [Google Scholar]
  6. Chen L, Jacquet R, Lowder E, Jandis WJ. Refinement of collagen–mineral interaction: A possible role for osteocalcin in apatite crystal nucleation, growth and development. Bone. 2015;71:7–16. doi: 10.1016/j.bone.2014.09.021. [DOI] [PubMed] [Google Scholar]
  7. Chokalingam K, Hunter S, Gooch C, Frede C, Florer J, Wenstrup R, David Butler D. 3D-in Vitro Effects of Compression and Time in Culture on Aggregate Modulus and on Gene Expression and Protein Content of Collagen Type II in Murine Chondrocytes. Tissue Engineering Part A. 2009 doi: 10.1089/ten.tea.2008.0560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dyment N, Liu C, Kazemi N, Aschbacher-Smith L, Kenter K, Breidenbach A, Shearn J, Wylie C, Rowe D, Butler D. The Paratenon Contributes to Scleraxis-Expressing Cells during Patellar Tendon Healing. PLoS ONE. 2013;8:e59944. doi: 10.1371/journal.pone.0059944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dyment NA, Hagiwara Y, Matthews BG, Li Y, Kalajzic I, Rowe DW. Lineage Tracing of Resident Tendon Progenitor Cells during Growth and Natural Healing. PLoS ONE. 2014;9:e96113. doi: 10.1371/journal.pone.0096113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Eyal S, Blitz E, Shwartz Y, Akiyama H, Schweitzer R, Zelzer E. On the Development of the Patella. Development. 2015;142:1831–9. doi: 10.1242/dev.121970. [DOI] [PubMed] [Google Scholar]
  11. Galatz L, Rothermich S, VanderPloeg K, Petersen B, Sandell L, Thomopoulos S. Development of the Supraspinatus Tendon-to-Bone Insertion: Localized Expression of Extracellular Matrix and Growth Factor Genes. J Orthop Res. 2007;25:1621–8. doi: 10.1002/jor.20441. [DOI] [PubMed] [Google Scholar]
  12. Kalajzic I, Kalajzic Z, Kaliterna M, Gronowicz G, Clark S, Lichtler A, Rowe D. Use of Type I Collagen Green Fluorescent Protein Transgenes to Identify Subpopulations of Cells at Different Stages of the Osteoblast Lineage. Journal of Bone and Mineral Research. 2002;17:15–25. doi: 10.1359/jbmr.2002.17.1.15. [DOI] [PubMed] [Google Scholar]
  13. Karaplis A, Bai X, Falet J, Macica CM. Mineralizing Enthesopathy is a Common Feature of Renal Phosphate-Wasting Disorders Attributed to FGF23 and is Exacerbated by Standard Therapy in Hyp Mice. Endocrinology. 2012;153:5906–17. doi: 10.1210/en.2012-1551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kaukua N, Shahidi MK, Konstantinidou C, Dyachuk V, Kaucka M, Furlan A, An Z, Wang L, Hultman I, Ährlund-Richter L, Blom H, Brismar H, Lopes NA, Pachnis V, Suter U, Clevers H, Thesleff I, Sharpe P, Ernfors P, Fried K, Adameyko I. Glial Origin of Mesenchymal Stem Cells in a Tooth Model System. Nature. 2014;513:551. doi: 10.1038/nature13536. [DOI] [PubMed] [Google Scholar]
  15. Kirsch T, Harrison G, Golub EE, Nah HD. The Roles of Annexins and Types II and X Collagen in Matrix Vesicle-mediated Mineralization of Growth Plate Cartilage. J Biol Chem. 2000;275:35577–35783. doi: 10.1074/jbc.M005648200. [DOI] [PubMed] [Google Scholar]
  16. Lin AC, Seeto BL, Bartoszko JM, Khoury MA, Whetstone H, Ho L, Hsu C, Ali SA, Ali AS, Alman BA. Modulating Hedgehog Signaling can Attenuate the Severity of Osteoarthritis. Nat Med. 2009;15:1421–5. doi: 10.1038/nm.2055. [DOI] [PubMed] [Google Scholar]
  17. Liu C, Breidenbach A, Aschbacher-Smith L, Butler D, Wylie C. A Role for Hedgehog Signaling in the Differentiation of the Insertion Site of the Patellar Tendon in the Mouse. PLoS ONE. 2013;8:e65411. doi: 10.1371/journal.pone.0065411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Liu C, Aschbacher-Smith L, Barthelery N, Dyment N, Butler D, Wylie C. Spatial and Temporal Expression of Molecular Markers and Cell Signals during Normal Development of the Mouse Patellar Tendon. Tissue Engineering Part A. 2011 doi: 10.1089/ten.tea.2011.0338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Liu Y, Schwartz A, Birman V, Thomopoulos S, Genin GM. Stress Amplification during Development of the Tendon-to-Bone Attachment. Biomech Model Mechanobiol. 2014;13:973–83. doi: 10.1007/s10237-013-0548-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Liu YX, Thomopoulos S, Birman V, Li JS, Genin GM. Bi-Material Attachment through a Compliant Interfacial System at the Tendon-to-Bone Insertion Site. Mech Mater. 2012;44:83–92. doi: 10.1016/j.mechmat.2011.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lu H, Thomopoulos S. Functional Attachment of Soft Tissues to Bone: Development, Healing, and Tissue Engineering. Annu Rev Biomed Eng. 2013;15:201–26. doi: 10.1146/annurev-bioeng-071910-124656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Mascre G, Dekoninck S, Drogat B, Youssef KK, Brohee S, Sotiropoulou PA, Simons BD, Blanpain C. Distinct Contribution of Stem and Progenitor Cells to Epidermal Maintenance. Nature. 2012;489:257–62. doi: 10.1038/nature11393. [DOI] [PubMed] [Google Scholar]
  23. Maye P, Fu Y, Butler D, Chokalingam K, Liu Y, Floret J, Stover M, Wenstrup R, Jiang X, Gooch C, Rowe D. Generation and Characterization of Col10a1-Mcherry Reporter Mice. Genesis. 2011;49:410–418. doi: 10.1002/dvg.20733. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mcgonagle D, Tan A, Carey J, Benjamin M. The Anatomical Basis for a Novel Classification of Osteoarthritis and Allied Disorders. J Anat. 2010;216:279–91. doi: 10.1111/j.1469-7580.2009.01186.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mcgonagle D. Imaging the Joint and Enthesis: Insights into Pathogenesis of Psoriatic Arthritis. Annals of the Rheumatic Diseases. 2005;64(Suppl 2):ii58–60. doi: 10.1136/ard.2004.034264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ralphs J, Benjamin M, Waggett A, Russell D, Messner K, Gao J. Regional Differences in Cell Shape and Gap Junction Expression in Rat Achilles Tendon: Relation to Fibrocartilage Differentiation. J Anat. 1998;193 (Pt 2):215–22. doi: 10.1046/j.1469-7580.1998.19320215.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Riley G. Tendinopathy--from Basic Science to Treatment. Nat Clin Pract Rheumatol. 2008;4:82–9. doi: 10.1038/ncprheum0700. [DOI] [PubMed] [Google Scholar]
  28. Rountree R, Schoor M, Chen H, Marks M, Harley V, Mishina Y, Kingsley D. BMP Receptor Signaling is Required for Postnatal Maintenance of Articular Cartilage. PLoS Biol. 2004;2:e355. doi: 10.1371/journal.pbio.0020355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: An Open-Source Platform for Biological-Image Analysis. Nat Methods. 2012;9:676–82. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Schwartz A, Long F, Thomopoulos S. Enthesis Fibrocartilage Cells Originate from a Population of Hedgehog-Responsive Cells Modulated by the Loading Environment. Development. 2015;142:196–206. doi: 10.1242/dev.112714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Schwartz A, Pasteris J, Genin G, Daulton T, Thomopoulos S. Mineral Distributions at the Developing Tendon Enthesis. PLoS ONE. 2012;7:e48630. doi: 10.1371/journal.pone.0048630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Sherlock J, Joyce-Shaikh B, Turner S, Chao C, Sathe M, Grein J, Gorman D, Bowman E, Mcclanahan T, Yearley J, Eberl G, Buckley C, Kastelein R, Pierce R, Laface D, Cua D. IL-23 Induces Spondyloarthropathy by Acting on ROR-γt+ CD3+CD4-CD8− Entheseal Resident T Cells. Nat Med. 2012;18:1069–76. doi: 10.1038/nm.2817. [DOI] [PubMed] [Google Scholar]
  33. Shibakawa A, Yudoh K, Masuko-Hongo K, Kato T, Nishioka K, Nakamura H. The Role of Subchondral Bone Resorption Pits in Osteoarthritis: MMP Production by Cells Derived from Bone Marrow. Osteoarthr Cartil. 2005;13:679–87. doi: 10.1016/j.joca.2005.04.010. [DOI] [PubMed] [Google Scholar]
  34. Sivas F, Mermerci Baskan B, Erkol Inal E, Akbulut Aktekin L, Barça N, Özoran K, Bodur H. The Relationship between Enthesitis Indices and Disease Activity Parameters in Patients with Ankylosing Spondylitis. Clin Rheumatol. 2009;28:259–264. doi: 10.1007/s10067-008-1027-5. [DOI] [PubMed] [Google Scholar]
  35. Snippert H, van der Flier L, Sato T, van Es J, van den Born M, Kroon-Veenboer C, Barker N, Klein A, van Rheenen J, Simons B, Clevers H. Intestinal Crypt Homeostasis Results from Neutral Competition between Symmetrically Dividing Lgr5 Stem Cells. Cell. 2010;143:134–44. doi: 10.1016/j.cell.2010.09.016. [DOI] [PubMed] [Google Scholar]
  36. Soeda T, Deng J, De Crombrugghe B, Behringer R, Nakamura T, Akiyama H. Sox9-Expressing Precursors are the Cellular Origin of the Cruciate Ligament of the Knee Joint and the Limb Tendons. Genesis. 2010;48:635–644. doi: 10.1002/dvg.20667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sugimoto Y, Takimoto A, Akiyama H, Kist R, Scherer G, Nakamura T, Hiraki Y, Shukunami C. Scx+/Sox9+ Progenitors Contribute to the Establishment of the Junction between Cartilage and Tendon/Ligament. Development. 2013;140:2280–8. doi: 10.1242/dev.096354. [DOI] [PubMed] [Google Scholar]
  38. Tatara AM, Lipner JH, Das R, Kim HM, Patel N, Ntouvali E, Silva MJ, Thomopoulos S. The Role of Muscle Loading on Bone (Re)Modeling at the Developing Enthesis. PLoS ONE. 2014;9:e97375. doi: 10.1371/journal.pone.0097375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ushiku C, Adams D, Jiang X, Wang L, Rowe D. Long Bone Fracture Repair in Mice Harboring GFP Reporters for Cells within the Osteoblastic Lineage. J Orthop Res. 2010;28:1338–1347. doi: 10.1002/jor.21105. [DOI] [PubMed] [Google Scholar]
  40. Wang V, Bell R, Thakore R, Eyre D, Galante J, Li J, Sandy J, Plaas A. Murine Tendon Function is Adversely Affected by Aggrecan Accumulation due to the Knockout of ADAMTS5. J Orthop Res. 2012;30:620–6. doi: 10.1002/jor.21558. [DOI] [PubMed] [Google Scholar]
  41. Yang G, Zhu L, Hou N, Lan Y, Wu X, Zhou B, Teng Y, Yang X. Osteogenic Fate of Hypertrophic Chondrocytes. Cell Research. 2014;24:1266–9. doi: 10.1038/cr.2014.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Yoon JW, Kita Y, Frank DJ, Majewski RR, Konicek BA, Nobrega MA, Jacob H, Walterhouse D, Iannaccone P. Gene Expression Profiling Leads to Identification of GLI1-Binding Elements in Target Genes and a Role for Multiple Downstream Pathways in GLI1-Induced Cell Transformation. J Biol Chem. 2002;277:5548–55. doi: 10.1074/jbc.M105708200. [DOI] [PubMed] [Google Scholar]
  43. Zelzer E, Blitz E, Killian M, Thomopoulos S. Tendon-to-Bone Attachment: From Development to Maturity. Birth Defects Res C Embryo Today. 2014;102:101–12. doi: 10.1002/bdrc.21056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zhou J, Chen Q, Lanske B, Fleming BC, Terek R, Wei X, Zhang G, Wang S, Li K, Wei L. Disrupting the Indian Hedgehog Signaling Pathway in Vivo Attenuates Surgically Induced Osteoarthritis Progression in Col2a1-CreERT2; Ihhfl/Fl Mice. Arthritis Res Ther. 2014;16:R11. doi: 10.1186/ar4437. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement
NIHMS708924-supplement.pdf (2.2MB, pdf)

RESOURCES