Intrafascicular chondroid‐like bodies in the ageing equine superficial digital flexor tendon comprise glycosaminoglycans and type II collagen

The superficial digital flexor tendon (SDFT) is considered functionally equivalent to the human Achilles tendon. Circular chondroid depositions scattered amongst the fascicles of the equine SDFT are rarely reported. The purpose of this study was the detailed characterization of intrafascicular chondroid‐like bodies (ICBs) in the equine SDFT, and the assessment of the effect of ageing on the presence and distribution of these structures. Ultrahigh field magnetic resonance imaging (9.4T) series of SDFT samples of young (1–9 years) and aged (17–25 years) horses were obtained, and three‐dimensional reconstruction of ICBs was performed. Morphological evaluation of the ICBs included histology, immunohistochemistry and transmission electron microscopy. The number, size, and position of ICBs was determined and compared between age groups. There was a significant difference (p = .008) in the ICB count between young and old horses with ICBs present in varying number (13–467; median = 47, mean = 132.6), size and distribution in the SDFT of aged horses only. There were significantly more ICBs in the tendon periphery when compared with the tendon core region (p = .010). Histological characterization identified distinctive cells associated with increased glycosaminoglycan and type II collagen extracellular matrix content. Ageing and repetitive strain frequently cause tendon micro‐damage before the development of clinical tendinopathy. Documentation of the presence and distribution of ICBs is a first step towards improving our understanding of the impact of these structures on the viscoelastic properties, and ultimately their effect on the risk of age‐related tendinopathy in energy‐storing tendons.


| INTRODUCTION
The equine superficial digital flexor tendon (SDFT) provides positional stability during the stance and swing phase of the stride and acts as the main energy-storing structure of the equine distal limb during locomotion. 1 SDFT injury is one of the most common causes of early retirement and wastage in performance horses and has a significant impact on equine welfare and the horse racing industry. 2 As the equine SDFT is considered functionally equivalent to the human Achilles tendon it serves as an established model for human tendinopathy. 3 The incidence of human Achilles tendon injury continues to increase in the ageing population of developed countries and is associated with chronic pain and restricted mobility. 4,5 The SDFT and the human Achilles tendon are both prone to injury as they work close to their functional limit under physiological loading conditions. 6,7 These tendons are energy-storing structures that act like springs and are considerably more elastic when compared witho positional tendons such as the equine common digital extensor or the human anterior tibialis tendon. 8,9 Characteristic features involved in the distinct biomechanics of energy-storing tendons are the interfascicular sliding capacity, as well as the crimp pattern and helical recoil mechanism of the tendon fascicles. 8,[10][11][12] Major risk factors for the development of tendinopathy are increasing age and exercise induced tendon overload. 13,14 The specific influence of repetitive load and ageing on the tendon ultrastructure and the mechanisms of energy storage and return have been studied in detail in the equine SDFT model. [15][16][17] Ageing results in the alteration of the tendon matrix composition and homeostasis and cell phenotypic variations have been recognized. [18][19][20] It has additionally been demonstrated that the tendon core region is particularly affected by age-related micro-damage. [21][22][23][24] The development of chondroid cell differentiation is commonly attributed to chronic overload or injury and has been described in multiple tendons and ligaments including the SDFT and the Achilles tendon. [25][26][27][28] However, the presence of isolated circular chondroid depositions scattered amongst the tendon fascicular structure as an effect of ageing to date has been rarely reported. 29,30 Based on this observation, the aim of this study was to further characterize the location, frequency and age-related appearance of intrafascicular chondroid-like bodies (ICBs) as detected histologically and with ultrahigh field magnetic resonance imaging (MRI) in the equine SDFT.
The distribution of chondroid depositions was assessed in horses of different age groups with the hypothesis that ICBs are predominantly found in the tendon core region of aged individuals.

| Samples
The study was conducted using SDFT samples harvested from equine cadaver limbs. Randomly selected left or right equine distal forelimbs were collected from a commercial equine abattoir or at a University teaching hospital from horses that were euthanized for reasons other than orthopedic disease or injury. Informed owner consent for tissue retention was obtained and ethical approval for the study was given by the local institutional veterinary research ethics committee (VREC214).
The collected limbs included those from young (1-9 years) and old (17-25 years) Thoroughbred, Thoroughbred Cross and Irish Sport Horses (Supporting information, S1). Horses with evidence of flexor tendon pathology were excluded from the study. Post mortem, the anonymised SDFT samples were dissected free from the surrounding soft tissues with the epitenon left intact. The tendons were excised at the level of the carpometacarpal-and the metacarpophalangeal joint, wrapped in tissue paper dampened with phosphate-buffered saline and then tin foil. Samples were subsequently stored at −20°C.

| Histological study
For histological examination, 1 cm tissue specimens were harvested from the proximal, middle, and distal metacarpal region of the SDFT.
Specimens were transected in the sagittal plane at the widest part of the tendon to collect samples containing peripheral and core tendon tissue including the dorsal and palmar epitenon. Samples were fixed in 4% phosphate-buffered formaldehyde solution for 48 h, paraffinembedded longitudinally, and 5 μm sections were mounted on poly-Llysine coated slides (VWR International Ltd.).
Histological staining included haematoxylin and eosin (H&E) for all tissue sections and the specific stains Toluidine blue, Safranin-O, Alcian blue-Periodic Acid Schiff (PAS), and the modified Von Kossa's stain for samples containing ICBs (TCS Bioscience Ltd.; Supporting information, S2). [31][32][33] Sections were viewed and imaged (×100 and ×400) using an Eclipse 80i microscope equipped with a Nikon DS 5mc digital camera 1600 × 1200 pixels (Nikon). In the H&E (×400) sections the total width and length (μm) of each ICB, length of the ICB nuclei, width and length of ICB clusters, and number of ICBs in each cluster were determined with the ImageJ Analyze tool (version 1.49, Rasband W.S., National Institute of HealthD) and recorded. 34 Immunohistological staining for aggrecan, biglycan, decorin, chondroitin-4-and chondroitin-6 sulfate, and collagen II was performed on deparaffinised tissue sections with ICBs. 31,35 The immunohistochemistry procedure and antibody details are provided in the Supporting information (S2).

| Transmission electron microscopy
A large ICB was localized on a fixed, unstained longitudinal section using light microscopy and prepared for transmission electron microscopy. The section was deparaffinised and rehydrated before being fixed in 2.5%  A total of seven overlapping transverse 3D-FISP scans were obtained for each tendon sample, to gather consecutive image data over a minimum tendon length of 14.4 cm (360 slices). 36 Following data acquisition overlapping slices were removed and DICOM images were concatenated and registered using the ImageJ StackReg tool (version 1.52a, Rasband W.S., National Institute of Health). 34 For segmentation and 3D reconstruction, files were converted into mrc Z-stacks using the IMOD "tif2mrc" command and the Z-scale was set based on image magnification and slice thickness. The 3dmod sculpt, warp and meshing tools (version 4.9, University of Colorado) were utilized to create a 3D model of the ICBs and the SDFT outline. ICBs were reconstructed in different colors to be readily distinguished.
The number, size and position of ICBs was recorded for the proximal, middle and distal one-third of the metacarpal length of the tendon (120 slices per area). ICBs were categorized as small (0.1-1.4 mm), medium (1.5-2.5 mm), or large (>2.5 mm) in size, depending on their width. The width was determined by using the 3dmod measuring tool that was calibrated based on the set size of the triangular sample rig (1 × 1 × 1 cm). In addition, ICB outlines located within 2 mm distance of the SDFT epitenon were categorized as peripheral, whilst the remaining ICBs were classed as being located in the tendon core region. 22  U test was used to analyse the difference in total ICB number between specimens from young and old horses, as detected on Ultrahigh field MRI. The difference in the ICB count between the proximal, middle, and distal metacarpal tendon was assessed using the Kruskal-Wallis test and the difference between the core and peripheral tendon regions was determined with the Wilcoxon signed-rank test. The distribution of small, medium and large ICBs was assessed with Friedman's Two-Way Analysis of Variance by Ranks. p values less than .05 were considered to be statistically significant. However, a number of tenocytes with irregular nuclei of varying size were present at the most proximal and distal extent of the larger ICBs. In addition, the tenocyte nuclei in the collagen fascicles adjacent to the ICBs were round rather than fusiform in appearance.

| Immunohistochemistry and transmission electron microscopy
Immunohistochemistry was performed in sections from aged horse SDFTs that contained ICBs (n = 7; 17-20 years, median = 18 years) to further specify the proteins associated with the ICB structure

| Ultrahigh field MRI study
Ultrahigh field MRI of the SDFT was performed in specimens obtained from young (n = 5; 1-3 years, median = 3 years) and aged horses (n = 5; 18-25 years, median = 23 years). ICBs were identified as pale circular structures on gross tendon sections and appeared as circular areas of high signal intensity on the MR images ( Figure 4).
There was a significant difference (p = .008) in the ICB count between the groups of young and old horses with no ICBs or other structures suggestive of chondroid deposition detected in any of the SDFT samples of the young horses examined. In the aged horse group, all SDFT specimens contained multiple distinct circular areas of high signal intensity with the number of ICBs ranging from 13 to 467 (median = 47, mean = 132.6) per tendon ( Table 1).
The 3D reconstruction of the ICBs confirmed the spherical-to elliptical shape of the ICBs and allowed for accurate recognition of the ICB position within the tendon ( Figure 5).
There was no significant difference in the number and size the of ICBs between the proximal, middle and distal aspect of the tendon (p = .50; Figure 6A) but there were significantly more ICBs in the periphery (within 2 mm distance to the epitenon) when compared with the core tendon region (p = .010; Figure 6B). The majority of ICBs were small in diameter and large ICBs were significantly less prevalent (p = .007; Table 1 and Figure 6).

| DISCUSSION
In this study intra-fascicular chondroid-like bodies (ICBs) were de- ICBs were not identified in young horses aged 1-9 years in this study. Earlier reports have described the replacement of the IFM with focal chondroid metaplasia in the equine SDFT starting at the age of 5 years. 29,30 Chondroid metaplasia in tendons is commonly described to be caused by injury or compressive forces leading to tissue hypoxia and subsequent chondrogenic cell differentiation. 25,26 It was therefore hypothesized that there would be more ICBs in the tendon core region where SDFT injuries most commonly occur, or at the level of the fetlock canal where compressive forces act upon the tendon tissue during weight bearing leading to a tendon matrix that is more chondrogenic in nature. 23,30 ICBs however appeared to be randomly dispersed throughout the length of the SDFT in the current study with more ICBs present in the proximal one-third in only one tendon ( Figure 5D). In addition, a significantly larger number of ICBs were detected in the tendon periphery when compared to the tendon core region. A possible explanation could be that ICBs are a response to continuous inter-fascicular sliding with thinning of the IFM and resulting increase of shear forces rather than chondrogenesis due to compressive forces alone. 15,17,24,33,37 As the the tendon's enthesis or in areas that are exposed to compressive forces. 19 and subsequent ossification has been described as an incidental finding as well as a result of tendon injury. 28