We present a custom experimental platform and tissue culture protocol that recreates fibrocartilaginous change driven by impingement of the Achilles tendon insertion in murine hind limb explants with sustained cell viability, providing a model suitable for exploring the mechanobiology of tendon impingement.
Tendon impingement upon bone generates a multiaxial mechanical strain environment with markedly elevated transverse compressive strain, which elicits a localized fibrocartilage phenotype characterized by accumulation of glycosaminoglycan (GAG)-rich matrix and remodeling of the collagen network. While fibrocartilage is a normal feature in impinged regions of healthy tendons, excess GAG deposition and disorganization of the collagen network are hallmark features of tendinopathy. Accordingly, impingement is clinically recognized as an important extrinsic factor in the initiation and progression of tendinopathy. Nevertheless, the mechanobiology underlying tendon impingement remains understudied. Prior efforts to elucidate the cellular response to tendon impingement have applied uniaxial compression to cells and excised tendon explants in vitro. However, isolated cells lack a three-dimensional extracellular environment crucial to mechanoresponse, and both in vitro and excised explant studies fail to recapitulate the multiaxial strain environment generated by tendon impingement in vivo, which depends on anatomical features of the impinged region. Moreover, in vivo models of tendon impingement lack control over the mechanical strain environment. To overcome these limitations, we present a novel murine hind limb explant model suitable for studying the mechanobiology of Achilles tendon impingement. This model maintains the Achilles tendon in situ to preserve local anatomy and reproduces the multiaxial strain environment generated by impingement of the Achilles tendon insertion upon the calcaneus during passively applied ankle dorsiflexion while retaining cells within their native environment. We describe a tissue culture protocol integral to this model and present data establishing sustained explant viability over 7 days. The representative results demonstrate enhanced histological GAG staining and decreased collagen fiber alignment secondary to impingement, suggesting elevated fibrocartilage formation. This model can easily be adapted to investigate different mechanical loading regimens and allows for the manipulation of molecular pathways of interest to identify mechanisms mediating phenotypic change in the Achilles tendon in response to impingement.
A multitude of tendons, including the Achilles tendon and rotator cuff tendons, experience bony impingement due to normal anatomical positioning1,2,3,4. Tendon impingement generates compressive strain directed transversely to the longitudinal fiber axis5,6,7. Regions of tendon impingement demonstrate a unique fibrocartilage phenotype in which shrunken, round cells (fibrochondrocytes) are embedded within a disorganized collagen network with markedly increased glycosaminoglycan (GAG) content2,3,4,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24. Prior studies suggest the disparate mechanical environment produced by tendon impingement sustains this GAG-rich matrix by driving the deposition of large aggregating proteoglycans, most notably aggrecan, although the underlying mechanisms are unclear1,3,12,13,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39. While fibrocartilage is a normal feature in impinged regions of healthy tendons, aberrant proteoglycan metabolism associated with excessive fibrocartilage formation is a hallmark feature of tendinopathy, a common and debilitating disease that disproportionately emerges in chronically impinged tendons1,40,41,42,43,44,45,46,47,48,49. Accordingly, tendon impingement is clinically recognized as an important extrinsic factor driving several of the most common tendinopathies, including rotator cuff disease and insertional Achilles tendinopathy (IAT)50,51,52. Currently, treatment of tendinopathy is inefficient. For example, approximately 47% of patients with IAT require surgical intervention after failed conservative management, with variable postoperative outcomes53,54,55,56. Despite the apparent relationship between impingement and tendinopathy, the mechanobiological mechanisms by which cells in impinged tendon sense and respond to their mechanical environment are poorly described, which obscures understanding of tendinopathy pathogenesis and results in inadequate treatment.
Explant models are useful tools in the study of tendon mechanobiology57,58. As a first step towards understanding the mechanobiology of tendon impingement, several prior studies have explored cellular response following the application of simple uniaxial compression to cells or excised tendon explants27,29,30,31,32,33,34,39. However, cells in vitro lack extracellular and pericellular matrices that facilitate strain transfer, sequester important growth factors and cytokines released by mechanical deformation, and provide substrate for focal adhesion complexes that play a role in mechanotransduction57,59. Additionally, both in vitro and excised explant studies fail to recapitulate the multiaxial mechanical strain environment generated by tendon impingement in vivo, which depends on anatomical features of the impinged region5,6. In the context of the impinged Achilles tendon insertion, this includes surrounding tissues such as the retrocalcaneal bursa and Kager's fat pad60,61,62,63. Conversely, in vivo models of tendon impingement25,28,36,37,38,64,65,66 allow minimal control over the magnitude and frequency of load applied directly to the tendon, which is a well-recognized limitation of in vivo models for studying tendon mechanobiology57,58,67,68,69,70. Given challenges in measuring tendon strain in vivo, the internal strain environment generated within these models is often poorly characterized.
In this manuscript, we present a custom experimental platform that recreates impingement of the Achilles tendon insertion upon the calcaneus within whole murine hind limb explants that, when paired with this tissue culture protocol, maintains viability over 7 days in explant culture and allows for study of the biologic sequelae of tendon impingement. The platform is built upon a 3D printed polylactic acid (PLA) base that provides the foundation for the attachment of the grips and 3D printed PLA volume reduction insert. The grips are used to clamp the upper leg and knee proximal to the Achilles myotendinous junction with the caudal aspect of the hind limb facing upward, allowing the Achilles tendon to be imaged from above using an ultrasound probe or inverted microscope (Figure 1A). The volume reduction insert slides along a track on the base and reduces the required volume of tissue culture media. A braided line wrapped around the hind paw is routed out of the platform utilizing the base design and a 3D printed PLA clip. By pulling on the string, the hind paw is dorsiflexed, and the Achilles tendon insertion is impinged against the calcaneus, resulting in elevated transverse compressive strain5,6 (Figure 1A). The platform is contained within an acrylic bath that maintains the hind limb explants submerged in tissue culture media. Securing the taut string to the outside of the bath with adhesive tape maintains ankle dorsiflexion to produce static impingement of the Achilles tendon insertion. CAD files for 3D printed components are provided in multiple formats (Supplementary File 1), allowing import into a range of commercial and free, open-source CAD software for modification to suit experimental needs. If access to 3D printers is not available for fabrication, CAD files can be provided to online 3D printing services that will print and ship the parts at low cost.
Importantly, the triceps surae-Achilles musculotendinous complex spans both the knee and ankle joints71,72,73. Consequently, tensile strain in the Achilles tendon is influenced by knee flexion. Knee extension places the Achilles tendon under tension, whereas knee flexion reduces tension. By first extending the knee and then passively dorsiflexing the ankle, compressive strains at the impinged insertion can be superimposed upon tensile strains. Conversely, by passively dorsiflexing the ankle with the knee flexed, tensile strain is reduced, and compressive strain remains. The current protocol explores three such conditions. 1) For static impingement, the foot is dorsiflexed to < 110° with respect to the tibia to impinge the insertion, with the knee flexed to reduce tension. 2) For the baseline tension group, the ankle is extended above 145° of dorsiflexion with the knee extended, generating predominately tensile strain at the insertion. 3) For the unloaded group, explants are cultured in a Petri dish with the knee and ankle in neutral positions in the absence of externally applied load. The angles referred to above are photographically measured relative to a coordinate system where the foot and tibia are parallel at an angle of 180° and perpendicular at an angle of 90°.
Key steps of the protocol include 1) dissection of hind limb explants and careful removal of the skin and plantaris tendon; 2) explant culture following a 48 h dexamethasone pretreatment; 3) tissue sectioning and histological staining; and 4) color image analysis to assess fibrocartilage formation. Following dissection, each hind limb explant is pretreated for 48 h in culture media supplemented with dexamethasone74. Contralateral limbs from each mouse are assigned to separate experimental groups for pairwise comparison, which helps control biological variability. After pretreatment, explants are positioned into platforms as described above and cultured for 7 more days (Figure 1B). Additional comparisons are made to a pretreated (day 0) group in which explants are removed immediately following the 48 h pretreatment.
After explant culture, hind limbs are trimmed down, formalin fixed, decalcified and embedded in paraffin. Serial sectioning in sagittal orientation provides visualization of the Achilles tendon from the myotendinous junction to the calcaneal insertion while allowing section depth to be tracked through the entire tendon. Terminal deoxynucleotidyl transferase (TdT)-mediated dUTP X-nick labeling (TUNEL) is used to visualize DNA damage secondary to apoptosis and assess viability. Toluidine blue histology and custom color image analysis are performed to quantify changes in GAG staining. Toluidine blue stained tissue sections are then used for SHG imaging to characterize alterations in collage fiber organization (Figure 1B).
The provided representative results suggest altered histological staining of the GAG-rich matrix and disorganization of the extracellular collagen network generated by 7 days of static impingement within the model. This model can be utilized to explore molecular mechanisms underlying impingement-driven fibrocartilaginous change.
All animal work was approved by the University of Rochester Committee on Animal Resources.
1. Preparation of tissue culture media
2. Explant dissection and dexamethasone pretreatment
3. Explant culture and loading platforms
4. Fixation, decalcification and paraffin embedding
5. Tissue sectioning
6. Deparaffinization/rehydration and slide selection
7. TUNEL to assess Achilles tendon viability
8. Toluidine blue histology to characterize fibrocartilage formation
9. SHG imaging to investigate change in collagen network organization
Representative images of TUNEL stained tissue sections demonstrate minimal apoptotic nuclei within the body of the Achilles tendon after 7 days of explant culture across experimental groups (Figure 2A). Quantification of these images provides evidence that the tissue culture protocol maintains up to 78% viability on average within the Achilles tendon after 7 days of explant culture across loading conditions (Figure 2B).
Qualitatively, enhanced Toluidine blue staining is appreciated after 7 days of static impingement compared to unloaded controls (Figure 3A,B), particularly in deep quadrants adjacent to the calcaneus where we have previously measured maximum transverse compressive strain5,6. Quantification of these Toluidine blue stained tissue sections indicates significant change in hue (Figure 3D), saturation (Figure 3E), and fibrocartilage score (Figure 3F) after 7 days of static impingement compared to contralateral unloaded explants. This altered GAG staining likely represents fibrocartilage formation secondary to static impingement within this model. Statistically significant differences in saturation and fibrocartilage score were only detected in quadrant 1 after 7 days of baseline tension (Figure 3E,F). Moreover, saturation and fibrocartilage score decreased relative to contralateral unloaded tendons, an opposite trend compared to static impingement. This data may support the hypothesis that tensile loading can attenuate or reverse impingement-driven fibrocartilage formation and warrants additional investigation in the future.
Quantification of SHG imaging suggests subtle but significant change in collagen fiber orientation after 7 days of static impingement, again in the deep and distal region adjacent to the calcaneus, as indicated by significant differences in dispersion (Figure 4C).
Figure 1: Explant platform and experimental design. (A) Photographic and schematic representation of the explant platform. (Top) Photograph of a whole murine hind limb explant loaded into the platform, with critical components labeled. (Bottom) Schematic depicting insertional Achilles tendon impingement elicited by passively applied ankle dorsiflexion in this explant model. (B) Schematic of study design and explant culture. Hind limb explants are pretreated in 100 nM dexamethasone for 48 h74. From each mouse, one hind limb explant is then cultured in a dish unloaded for 7 days, while the contralateral limb is loaded into the experimental platform to place the Achilles tendon under either baseline tension or static impingement for 7 days. Serial tissue sections from contralateral pairs of limbs are either stained with Toluidine blue to visualize GAG-rich tissue or used for TUNEL staining. Toluidine blue stained sections are then used for SHG imaging to assess collagen organization. Please click here to view a larger version of this figure.
Figure 2: Explant viability. (A) Representative images of TUNEL stained sections from pretreated (day 0) explants (top left), unloaded tendons (top right), tendons maintained at baseline tension (bottom left), and statically impinged tendons (bottom right). The Achilles tendon is outlined in a dashed white line and labeled AT, while the calcaneus is outlined in a solid yellow line and labeled C. Apoptotic nuclei appear red (TUNEL), all nuclei appear blue (DAPI). (B) Quantification of dead cell fraction was on average less than 22% for the unloaded, baseline tension, and static impingement groups. Data represents mean ± standard deviation. Outliers are present in these data sets with greater than 50% cell death, which may be attributed to tendon damage induced by poor dissection technique. The average dead cell fraction with statistical outliers removed is less than 15% for all groups. No statistically significant difference in cell death was found between unloaded, baseline tension and static impingement groups (p < 0.05). Day 0 control: n=8. Unloaded: n=18. Baseline Tension: n=8. Static Impingement: n=12. Please click here to view a larger version of this figure.
Figure 3: Toluidine blue histology. (A) Representative Toluidine blue stain after 7 days of unloaded culture depicting the compressive tendon fibrocartilage (CTF), attachment zone fibrocartilage (AZF), and periosteal fibrocartilage (PF) at the Achilles insertion16. Calcaneus labeled C. ROI capturing the CTF is outlined in red and is further subdivided into 4 quadrants to provide additional spatial information. (B) Representative Toluidine blue stain after 7 days of static impingement, with enhanced staining of the CTF, especially in quadrant 1 where we have previously measured maximum transverse compressive strains5,6. (C) Schematic of the hue-saturation color space used for color image analysis. Color is described by a combination of hue (0°-360°) and saturation (0-255)76. The average color in each ROI is normalized to the average color of the PF by calculating Euclidean distance separating colors in each ROI from the color in the PF, which provides a normalized fibrocartilage score. (D-F) Changes in Toluidine blue stain after 7 days of static impingement (top) and baseline tension (bottom) compared to contralateral unloaded tendons. Data represents mean ± standard deviation. (D) Quantification of hue indicates significant differences in average hue across all regions after 7 days of static impingement compared to contralateral unloaded tendons. No similar trend was detected after 7 days of baseline tension compared to contralateral unloaded tendons. (E) Quantification of saturation, with significant changes in average saturation across the entire CTF and within quadrants 1-3 after 7 days of static impingement compared to contralateral unloaded tendons. Conversely, significant differences were only detected in quadrant 1 after 7 days of baseline tension, where saturation decreased compared to contralateral unloaded tendons. (F) Quantification of fibrocartilage score, with significant changes in fibrocartilage score across all regions after 7 days of static impingement compared to contralateral unloaded tendons. Conversely, significant differences were only detected in quadrant 1 after 7 days of baseline tension, where fibrocartilage score decreased compared to contralateral unloaded tendons. The data suggest enhanced GAG stain brought about by static impingement indicative of increased fibrocartilage formation. Data describing the total CTF were compared using Wilcoxon matched-pairs signed rank tests. Quadrant data were compared using repeated measures two-way ANOVA with Šídák's multiple comparisons test. *p < 0.05. **p < 0.01. ***p < 0.005. ****p < 0.001. Baseline Tension vs. Unloaded: n = 6 pairs. Static Impingement vs. Unloaded: n = 8 pairs. Please click here to view a larger version of this figure.
Figure 4: SHG imaging. (A) Representative SHG image after 7 days of unloaded culture compared to (B) 7 days of static impingement. CTF is outlined in red and divided into 4 quadrants as labeled, consistent with ROIs in Toluidine blue analysis above. (C) Within each quadrant, Fourier spectrum analysis was performed in a moving sub-window using the Directionality plugin in FIJI. The spread of the distribution of fiber orientations within each sub-window is termed dispersion (°) and is inversely related to fiber alignment. Dispersion = 0° represents perfect parallel alignment of fibers. Increasing dispersion represents decreasing collagen fiber alignment. Data represent mean ± standard deviation. Statistically significant differences in dispersion were detected in quadrant 1 following 7 days of static impingement compared to contralateral unloaded tendons, suggesting increased collagen disorganization. Quadrant data were compared using repeated measures two-way ANOVA with Šídák's multiple comparisons test. * p < 0.05. Static Impingement vs Unloaded: n = 5 pairs. Please click here to view a larger version of this figure.
Supplementary File 1: CAD files. These CAD files, compiled in multiple file formats, can be used to 3D print the platform base, volume reduction insert, and clip. Please click here to download this File.
Supplementary File 2: MATLAB files. The compiled MATLAB files allow for quantification of Toluidine blue histology as described in section 8 of the protocol to assess fibrocartilage formation through spatial change in GAG staining at the Achilles tendon insertion. Please click here to download this File.
Supplementary File 3: Guidelines for implementing MATLAB code. This document describes a pipeline for quantifying changes in GAG staining at the Achilles tendon insertion via image analysis of Toluidine blue histology using the provided MATLAB code. Please click here to download this File.
Supplementary File 4: Sample image of Toluidine blue histology. This RGB color image of Toluidine blue histology can be used to execute the provided MATLAB code. Please click here to download this File.
The experimental murine hind limb explant platform paired with the tissue culture protocol described in this study provide a suitable model for studying the mechanobiology of impingement-driven fibrocartilage formation at the Achilles tendon insertion. The utility of this explant model is demonstrated by the representative results, which indicate maintenance of cell viability concomitant with significant and spatially heterogeneous change in Toluidine blue staining after 7 days of static impingement. These findings suggest altered metabolism of GAG-rich matrix molecules secondary to mechanical impingement, consistent with other models and clinical studies3,4,9,13,23,25,28,29,30,31,32,33,34,35,36,37,38,44,49. Additionally, subtle yet significant changes were detected in collagen organization via SHG imaging after 7 days of static impingement, consistent with clinical features of impingement-associated tendinopathy and in vivo models of tendon impingement28,37,38,43,44,49,64.
Prior efforts to explore the biologic sequelae of tendon impingement have applied simple compression to isolated tendon cells27,29. Once a mechanobiological mechanism of interest has been identified, these models provide superior therapeutic screening platforms compared to our explant model. However, the findings of these studies must be interpreted with caution, as isolated cells cultured in vitro lack a three-dimensional extracellular environment that influences mechanoresponse57,59. Additionally, cells isolated from other collagenous tissues (e.g., cartilage) demonstrate altered mechanosensitive ion channel activity when mechanically challenged in vitro in the absence of an extracellular environment78,79. The model presented addresses these limitations by providing a platform to subject in situ tendon cells within viable explants to physiologically relevant loading conditions.
Explant models are popular experimental systems for probing tendon biology and physiology57,58. In the specific context of tendon impingement, several prior studies have employed partial- and whole-tendon explants to explore tendon cell response to artificial uniaxial compression30,31,32,33,34,39. While these studies demonstrated a direct link between uniaxial compression and fibrocartilage formation, the in vivo mechanical strain environment generated by tendon impingement is multiaxial and depends on anatomical features of the impinged region5,6. For example, the strain environment created by impingement of the calcaneus upon the Achilles tendon during ankle dorsiflexion is influenced by insertion angle, area of attachment, and the presence of surrounding soft tissues60,61,62,63,80,81. Our explant platform preserves these external structures and reproduces the multiaxial strain environment generated by impingement while maintaining cells within their native environment6.
Animal models have been fundamental in substantiating an immediate relationship between mechanical impingement and the quintessential features of tendon fibrocartilage and pathology, which share analogous phenotypes1,25,28,37,64,65,66. In recent decades, the majority of in vivo impingement research has utilized rodent models to study rotator cuff disease through surgical reduction of the subacromial space to artificially impinge the rotator cuff tendons. These animal models have been used to describe the cellular and molecular pathogenesis of impingement tendinopathy in vivo and can be extended to evaluate novel therapeutic strategies to promote clinical translation of impingement research36,37,38,66,82,83,84,85.
Established in vivo models of tendon impingement, however, have several important limitations in studying mechanobiology57,58. These models surgically introduce or remove an external source of tendon impingement and resume physical activity (free cage activity, forced treadmill running, etc.) This approach lacks control over the magnitude, orientation, and frequency of force applied directly to the tendon throughout the duration of the experiment. Given the difficulty in measuring tissue strain in vivo, these models offer limited control over, or characterization of, the mechanical environment generated within the tendon in response to impingement. This limitation is well-recognized within the field of tendon mechanobiology research57,58,67,68,69,70. Elsewhere, sophisticated experimental platforms have been designed to apply controlled mechanical overload (fatigue) directly to tendons in vivo under anesthesia67,68,69. These models control the application of mechanical stimuli for short durations under anesthesia and offer no control over loading history throughout the remainder of the experiment as the animals resume normal cage activity for days to weeks following mechanical overload. The explant model described in this protocol offers controlled prescription of Achilles tendon impingement to generate measurable and well-described patterns of tissue strain6, allowing for rigorous study of impingement mechanobiology. It should be noted, however, that this explant model does not obviate the need for in vivo investigation of tendon impingement, but rather, it serves as a supplementary tool to enrich data derived from these indispensable animal models. Cellular and molecular pathways defined in relation to impingement mechanics using this explant model can be characterized in vivo and evaluated as therapeutic targets using animal models in a complementary approach to enhance clinical translation of mechanobiology research. The power of this integrated strategy is being exploited in other realms of tendon mechanobiology research58,86,87,88, and this explant model bestows its application to tendon impingement.
The explant model is not without limitations. As with all explant models, mechanobiology can only be studied over relatively short time scales while the tissue remains viable57,89. While the model is a powerful tool for studying the immediate cellular response to mechanical impingement, it is not suited for studying chronic tendon disease associated with impingement. Additionally, the model is unable to account for systemic response and tissue crosstalk since blood supply and neural communication are not maintained. Nevertheless, the model provides a platform to delineate the intrinsic response of endogenous tendon cells to mechanical impingement. A further limitation of the model is that according to prior research32,90,91,92, we expect many large GAG-rich proteoglycans to diffuse out of the tendon during pretreatment and early time points of explant culture, particularly fragments of aggregating proteoglycans not anchored to the ground matrix. Therefore, changes in GAG stain secondary to impingement in the model are likely to reflect changes in newly synthesized GAG-rich macromolecules that have not yet been catabolized and remain anchored within the extracellular matrix. Given that these GAG-rich proteoglycans localize to the pericellular space22,30,93,94,95, the explant model may facilitate investigation of PCM remodeling secondary to tendon impingement.
The dissection of murine hind limb explants described in this protocol requires practice, but proficient technique can be achieved quickly. Complete degloving (i.e., skin removal) can prove challenging, particularly around the digits of the hind paw. While the skin around the toes does not need to be entirely removed, it does reinforce sterility throughout culture as bacteria is present on the skin and fur.
The most difficult aspect of the dissection is removal of the plantaris tendon medial to the Achilles tendon proximally and passing superficial to the Achilles tendon distally. The plantaris muscle-tendon unit spans the knee and ankle joints in close association with the triceps surae and Achilles tendon81,96,97,98,99. Mechanical loads transferred across the posterior aspect of the ankle during passive ankle dorsiflexion are distributed between the plantaris and Achilles tendons, and the plantaris tendon is known to bear high stress in vivo100,101. Additionally, anatomical variability in the human plantaris is well described96,102,103 and while this variability is less prevalent in rodents97, it has the propensity to affect the mechanical environment within the Achilles tendon in situ. Consistent with prior techniques used to measure strain within the Achilles tendon in situ6,104, the plantaris tendon should be removed to control the internal strain environment generated by externally applied tendon impingement.
Careful removal of the plantaris is necessary to avoid dissection-induced cell death in the Achilles tendon, and the variability in cell apoptosis present in the data (Figure 2B) may reflect dissection-induced damage due to poor dissection technique. Removal of the plantaris tendon benefits from fine tip forceps to carefully separate the plantaris from the Achilles and free the plantaris muscle-tendon complex proximally. Serrated forceps help grip the proximal free end of the plantaris tendon to pull distally, around the calcaneus and towards the digits of the paw for complete removal. Lee and Elliott provide a helpful anatomical description of the rat plantaris and Achilles tendons and associated musculature97.
While we found the dissection technique described in this study to be sufficient for ensuring sterility, the protocol can be adjusted or modified as needed in settings where maintaining sterility proves challenging. Surgical tools can be autoclaved, betadine solution can be applied topically to the skin, and dissection can be carried out in a biological safety cabinet. We have found that swift and efficient benchtop dissection is critical and allows for dissected limbs to be quickly transferred into sterile media in a biological safety cabinet after the sterile boundary provided by the skin is breached.
Another prime target for troubleshooting contamination is the technique used for sterilizing platform components. In our hands, we have had consistent success with simple soaks in ≥ 10% bleach solution with adequate rinsing in tap water without autoclaving. Additional sterilization using 70% ethanol can be applied as needed but be aware that any component 3D printed with PLA is incompatible with EtOH and will warp. Additionally, all parts of the platform aside from the acrylic bath and 3D printed PLA components can be autoclaved, and components soaked in ≥ 10% bleach solution can be rinsed with autoclaved or purified water. To avoid rusting, it is encouraged to wash, and hand dry all metal components immediately following each experiment.
The methodology for quantifying Toluidine blue staining to characterize fibrocartilage formation described in this protocol employs an innovative custom color image analysis that offers broad application to tendon research. Significant advantages are provided by converting pixel data from the RGB to the HSV color space76. Here, hue encodes a dominant wavelength as a characteristic color from 0°-360°, while saturation defines the purity of each hue from 0-255. Value, conversely, describes brightness of color and as such, the HSV color scheme separates brightness (value) from color (hue, saturation). In the context of histological analyses, HSV pixel data is more robust to changes in light transmission during microscopy and variations in tissue section thickness76. Additionally, working in the HSV color space allows for delineation of specific wavelengths of interest (via hue) depending on the histological stain76, e.g., hues around 240° for Toluidine blue staining.
An innovative aspect of the Toluidine blue image analysis presented in this study is the fibrocartilage score, a metric that is related to the Euclidean distance separating the color of each ROI to the color of the periosteal fibrocartilage. The periosteal fibrocartilage emerges as a secondary cartilage immediately after birth, whereas the compressive tendon fibrocartilage is absent postnatally and, like other sesamoid fibrocartilages, appears to be regulated by mechanical demand in regions of elevated tendon compression16,17,28,77. By comparing fibrocartilage scores between experimental groups, we can determine not only if the Toluidine blue stain color is changing secondary to impingement, but whether this change results in an appearance closer to fibrocartilage. This analysis can be extended to other popular GAG stains such as Alcian blue and Safranin O, as well as other regions of tendon fibrocartilage and fibrocartilaginous tendon entheses, as long as a control tissue is present akin to the periosteal fibrocartilage. For example, changes in GAG staining in the tendons and ligaments of the knee could be normalized via the fibrocartilage score to staining in the fibrocartilaginous menisci.
A crucial aspect of this protocol is serial tissue sectioning and tracking section level through the Achilles tendon, which requires repeatable paraffin embedding and scrutinous identification of entry into the Achilles tendon insertion for tracking of section depth. This sectioning and staining protocol undoubtedly requires practice and refinement but can be extended to cryosectioning.
The histological approach described in this manuscript can easily be extended to immunohistochemistry or immunofluorescence. In this regard, an important aspect of the above experimental design is the pairwise comparison of contralateral limbs from each mouse. Here, all statistical comparisons are drawn between tissue sections at comparable section level through the Achilles tendon from contralateral limbs from the same mouse assigned to different experimental groups. This strategy helps control biological variability. Additionally, level-matched pairs of contralateral tissue sections are stained on the same slide rack simultaneously (histology) or on-slide simultaneously using identical working solutions for all reagents, buffers, and antibody solutions (immunostaining) to help control for stain-to-stain variability inherent in histology and immunolabeling, as well as anatomical heterogeneity dependent upon section level, location, and orientation.
Several other modifications to the protocol described in this manuscript may be implemented for further study of additional research questions. For instance, while the representative data provided here is focused on 7 days of static impingement, this approach can be adapted to investigate the biologic sequelae of intermittent or cyclic impingement by coupling the string wrapped around the hind paw through the clip to a mechanical actuator such as a syringe pump. In addition, the platform provides the opportunity to interrogate molecular mechanisms of interest by supplementing the culture media with small molecule agonists/antagonists, or by using hind limb explants from transgenic mouse strains.
In conclusion, we have presented a novel murine hind limb explant model for studying mechanobiology underlying Achilles tendon impingement. This is endowed by a tissue culture protocol that maintains cell viability across loading conditions over 7 days. This model recapitulates impingement-driven fibrocartilage formation and collagenous change as described in other model systems as well as in degenerative tendon disease. The experimental platform allows controlled application and quantification of mechanical strain patterns, and therefore provides an excellent model for rigorous study of impingement mechanobiology. This model can be applied to interrogate molecular mechanisms of interest to identify potential therapeutic targets in the treatment of insertional tendinopathies.
The authors have nothing to disclose.
The authors are grateful for support and assistance provided by Jeff Fox and Vidya Venkatramani of the University of Rochester Center for Musculoskeletal Research's Histology, Biochemistry, and Molecular Imaging (HBMI) Core, funded in part by P30AR06965. Additionally, the authors would like to thank the Center for Light Microscopy and Nanoscopy (CALMN) at the University of Rochester Medical Center for assistance with multiphoton microscopy. This study was funded by R01 AR070765 and R01 AR070765-04S1, as well as 1R35GM147054 and 1R01AR082349.
Absorbent underpads | VWR | 82020-845 | For benchtop dissection |
Acrylic bath | Source One | X001G46CB1 | Contains the explant platform submerged in culture media |
Autoclave bin | Thermo Scientific | 13-361-20 | Used as secondary containment, holds two platforms |
Base | – | – | 3D printed from CAD files provided as Supplementary Files |
Braided line | KastKing | 30lb test | Used to wrap around paw and apply ankle dorsiflexion |
Clip | – | – | 3D printed from CAD files provided as Supplementary Files |
Cover glass | Fisherbrand | 12-541-034 | Rectangular, No. 2, 50 mm x 24 mm |
Cytoseal XYL | VWR | 8312-4 | Xylene-based mounting media for coverslipping Toluidine blue stained tissue sections |
Dexamethasone | MP Biomedical LLC | 194561 | CAS#50-02-2 |
Dimethyl sulfoxide (DMSO), anhydrous | Invitrogen by ThermoFisher | D12345 | CAS#67-68-5, use to solubilize dexamethasone into concentrated stock solutions |
Double-sided tape | Scotch Brand | 34-8724-5195-9 | To attach sandpaper to Grip platens |
Dulbecco's Modified Eagle Medium (1X DMEM) | Gibco by ThermoFisher | 11965092 | high glucose, (-) pyruvate, (+) glutamine |
EDTA tetrasodium salt dihydrate | Thermo Scientific Chemicals | J15700.A1 | CAS#10378-23-1, used to make 14% EDTA solution for sample decalcifcation |
Ethanol, 200 proof | Thermo Scientific | T038181000 | CAS#64-17-5, 1 L supply |
Foam biopsy pads | Leica | 3801000 | Used with processing cassettes, help hold ankle joints in desired position during fixation and decalcification |
Forceps, #SS Standard Inox | Dumont | 11203-23 | Straight, smooth, fine tips |
Forceps, Micro-Adson 4.75" | Fisherbrand | 13-820-073 | Straight, fine tips with serrated teeth |
Garnet Sandpaper, 50-D Grit | Norton | M600060 01518 | Or other coarse grit sandpaper |
Glacial acetic acid | Fisher Chemical | A38S-500 | CAS#64-19-7, for adjusting pH of sodium acetate buffer used for Toluidine blue histology, as well as 14% EDTA decalcification solution |
Grips | ADMET | GV-100NT-A4 | Stainless steel vice grips, screws and springs described in the protocol are included |
Histobond Adhesive Microscope Slides | VWR | 16005-108 | Sagittal sections of hind limbs explants reliably adhere to these slides through all staining protocols |
In situ Cell Death Detection Kit, TMR Red | Roche | 12156792910 | TUNEL assay |
Labeling tape | Fisherbrand | 15-959 | Or any other labeling tape of preference |
L-ascorbic acid | Sigma-Aldrich | A4544-100G | CAS#50-81-7, for culture media formulation |
Neutral buffered formalin, 10% | Leica | 3800600 | For sample fixation, 5 gallon supply |
Nunc petri dishes | Sigma-Aldrich | P7741-1CS | 100 mm diameter x 25 mm height, maintain explants submerged in 70 mL of culture media as described in protocol |
Penicillin-streptomycin (100X) | Gibco by ThermoFisher | 15140122 | Add 5 mL to 500 mL 1X DMEM for 1% v/v (1X) working concentration |
Polylactic acid (PLA) 1.75 mm filament | Hatchbox | – | Choose filament diameter compatible with your 3D printer extruder, in color of choice. |
Processing cassettes | Leica | 3802631 | For fixation, decalcification and paraffin embedding |
Prolong Gold Antifade Reagent with DAPI | Invitrogen by ThermoFisher | P36931 | Mounting media for coverslipping tissue sections after TUNEL |
Proteinase K | Fisher BioReagents | BP1700-50 | CAS#39450-01-6, used for antigen retrieval in TUNEL protocol |
Scissors, Fine | FST | 14094-11 | Straight, sharp |
Slide Staining Set, 12-place | Mercedes Scientific | MER 1011 | Rack with 12 stain dishes and slide dippers for Toluidine blue histology |
Sodium acetate, anhydrous | Thermo Scientific Chemicals | A1318430 | CAS#127-09-3, used to make buffer for Toluidine blue histology |
Tissue-Tek Accu-Edge Low Profile Microtome Blades | VWR | 25608-964 | For paraffin sectioning |
Toluidine Blue O | Thermo Scientific Chemicals | 348601000 | CAS#92-31-9 |
Volume Reduction Insert | – | – | 3D printed from CAD files provided as Supplementary Files |
Xylenes | Leica | 3803665 | 4 gallon supply for histological staining |