Science in brief: Recent advances into understanding tendon function and injury risk

Tendon injury is one of the most common musculoskeletal injuries sustained by horses competing in all disciplines, with a particularly high prevalence reported in racing Thoroughbreds 1-4. The forelimb superficial digital flexor tendon (SDFT) is far more likely to be injured than any other tendinous structures in the equine limb, with ∼90% of tendon injuries in racehorses occurring to this tendon 1, 5. Other tendinous structures at risk of injury are the suspensory ligament (SL), which accounts for ∼7% of tendon injuries 1, 5. While tendon injuries occurring to horses competing in other disciplines are less well reported, a small number of studies have shown a significant risk of injury to the SDFT, SL and deep digital flexor tendon (DDFT), depending on the discipline undertaken (4, 6; for a more detailed review of tendon injury risk, please refer to Thorpe et al. 7 and references within). Tendon injury is thought to occur due to an accumulation of microdamage over a number of loading cycles rather than an acute injury 8. It is well established that the risk of sustaining an SDFT injury increases significantly with ageing 1, 5, 9, 10, which may be because aged tendons have undergone a larger number of loading cycles and so are more likely to have accumulated microdamage. While tendon injury risk has been well characterised, the efficacy of many treatments is still to be established, and recurrence of tendon injuries is common 11. The intrinsic healing property of tendon is low, and few horses return to previous activity levels 12, 13. While some treatments, including stem-cell therapy, show promising results in preliminary studies 14, 15, more research is required to establish the true efficacy of these treatments. In order to develop effective treatment regimens, the alterations that occur during tendon injury and healing need to be understood better; however, this is difficult when there is still much that is not known about basic structure–function relationships within tendon and how these are altered with ageing and injury. All tendons have the same basic structure, in which collagen type I molecules are aligned in a hierarchical manner, aggregating to form fibrils, fibres and fascicles and tendons 16 (Fig 1). At the larger hierarchical levels, the collagen is interspersed with a predominantly noncollagenous matrix, rich in proteoglycans, glycoproteins and other fibrous proteins, including elastin 16, 17 (Fig 1). This organisation results in a structure with high uniaxial strength, in which extension is facilitated by sliding between collagenous units at each hierarchical level 18-21. Recent research has shown that there are numerous differences in tendon structure, composition and organisation between tendons with different functions 22-24; these differences confer the distinct mechanical properties required by functionally different tendons. Schematic illustrating the structure of the equine superficial digital flexor tendon (SDFT). Fascicles, the largest subunit of the tendon, are surrounded by the interfascicular matrix (also called the endotenon). The interfascicular matrix has recently been implicated as critical for facilitating fascicle sliding and enabling high strains in energy-storing tendons 28. The composition and structure of the interfascicular matrix are poorly characterised, but it is known to be highly cellular and to contain proteoglycans, glycoproteins and elastin 35, which are likely to modulate the fascicle sliding 28. Note the presence of the helix at the fascicular level 30. Recent work has focused on characterising structure–function relationships in the equine SDFT, due to both its extreme energy-storing role and its high risk of injury. Along with the SL, this energy-storing tendon stretches and recoils up to 16% with each stride 25 and experiences stresses of up to ∼90 MPa 26. These values approach the failure properties of the SDFT measured in vitro 27, 28. It is therefore important to understand how the SDFT is specialised to be able to resist the continual application of large stresses and strains in vivo and to characterise changes that occur with ageing that predispose aged tendons to injury. Comparison of the SDFT with the positional common digital extensor tendon (CDET), which experiences much lower stresses and strains in vivo 22, has provided important insights into specialisation of tendon mechanical properties. Several studies have shown that the basic matrix composition differs between the SDFT and CDET, with greater cellularity, glycosaminoglycan content, water content and type III collagen 27, 29 measured in the SDFT. In addition, it has been shown that the profile of the collagen cross-links, which link the collagen molecules together, also differs between tendon types 23, 24. However, it is not clear how these alterations in matrix composition affect tendon mechanical properties. More recent studies have demonstrated that the rate of matrix turnover differs between the SDFT and CDET, with collagen being turned over at a slower rate in the SDFT, whereas the noncollagenous matrix is turned over more rapidly in the SDFT 24. While this may seem counterintuitive, the noncollagenous matrix is far less stiff than the collagenous components, so likely to be far more highly strained and therefore damaged when the tendon is loaded, requiring subsequent repair. The turnover rate of specific proteins within the noncollagenous matrix is yet to be determined; these data may provide important information regarding maintenance of healthy tendon function. Characterisation of the mechanical properties of healthy tendon has provided important information regarding structure–function relationships in functionally distinct tendons. It has previously been shown that the SDFT is less stiff than the CDET, such that it is able to extend further before failure (i.e. the SDFT has a higher failure strain than the CDET) 27, 28. These properties are likely to be important for efficient energy-storing function, but until recently, it has not been established how such differences are attained. Our studies have recently characterised the failure properties of the SDFT and CDET at different hierarchical levels, demonstrating that the failure strain of fascicles isolated from the SDFT is only just over half of that of the whole tendon (Fig 2a). Indeed, SDFT fascicles have a lower failure strain than those from the CDET, indicating that differences in whole-tendon mechanics cannot be due to differences in fascicle properties 28. However, the authors also showed that there was greater capacity for sliding between adjacent fascicles in samples from the SDFT, which is large enough to account for the difference in fascicle and tendon failure strain in this tendon type (Fig 2b). These findings suggest that interfascicular sliding provides the greater extensibility required in energy-storing tendons, such as the SDFT 28. These results indicate that the interfascicular matrix (IFM; also called the endotenon; see Fig 1) may be important in modulating this sliding behaviour, and variation to this matrix will therefore result in altered tendon mechanical properties. However, the composition and structure of the interfascicular matrix is poorly characterised and, as such, is an important area for further research. Mechanical behaviour of the superficial digital flexor tendon (SDFT) and common digital extensor tendon (CDET) and their constituent fascicles and interfascicular matrix. a) Typical stress-strain curves for the SDFT and CDET and their constituent fascicles. While the SDFT has a higher failure strain and lower failure stress than the CDET, this difference is reversed at the fascicle level, with SDFT fascicles failing at lower strains than their counterparts from the CDET, such that the failure strain of SDFT fascicles is approximately half of that of the whole tendon. b) Typical force-extension curves for the interfascicular matrix. c) Percentage interfascicular matrix extension plotted at different percentages of failure force in the SDFT and CDET. The failure properties of the interfascicular matrix do not differ between the SDFT and CDET, but at physiological loads there is greater interfascicular matrix extension in the SDFT than in the CDET. This extension is sufficient to account for the difference in failure strain between the SDFT as a whole and its constituent fascicles. Adapted from Thorpe et al. 28. Work has also been undertaken to characterise the microstructural strain response of fascicles from the SDFT and CDET, using confocal microscopy combined with mechanical testing to visualise the response of collagen fibres to applied strain 30. These studies demonstrate differences in micromechanics between fascicles from functionally distinct tendons, characterised by rotation in response to applied strain in the SDFT, which is absent in the CDET 30. This rotation is indicative of helical substructures within SDFT fascicles. We have used scanning electron microscopy to visualise the ultrastructure of fascicles from the SDFT and CDET. Fascicles from the SDFT had a helical appearance, which was absent in the CDET (Fig 3), supporting the micromechanical data in indicating a helical substructure. This specialisation may allow SDFT fascicles to act as springs, resulting in more efficient energy storage and return. Indeed, studies have demonstrated that SDFT fascicles are better able to recoil and recover than those from the CDET 30, further indicating that their structure is specialised for maximal energy-storing capacity. Scanning electron microsopic images of fascicles from the superficial digital flexor tendon (SDFT) (a) and common digital extensor tendon (CDET) (b). Fascicles were dissected from the SDFT and CDET from young horses (n = 3, aged 3–6 years). Fascicles were fixed in 4% glutaraldehyde, then dehydrated in ethanol and hexamethyldisilizane. Samples were dried overnight and then mounted on scanning electron microsope stubs and sputter coated with a gold layer (10 nm) before scanning at 10 kV. Note the helical appearance of the SDFT fascicle, not replicated in the CDET fascicle. Scale bar = 200 μm. In addition to considering the tendon response to quasi-static loading, it is also important to assess fatigue resistance, because this will provide important information regarding the effect of exercise on tendon properties. We have shown that fascicles from the SDFT have a much greater fatigue resistance than those from the CDET, with a greater number of loading cycles before failure measured in the SDFT (16,825 cycles in the SDFT 31 vs. 2017 cycles in the CDET). Furthermore, assessment of fascicle microstructural strain response in SDFT fascicles post fatigue loading has demonstrated that fatigue loading results in mild damage to the fascicle matrix, characterised by fibre kinking. This is accompanied by decreased rotation in response to applied strain and decreased recovery following the removal of load 31, 32. These results suggest that fatigue loading causes alterations to the helix substructure, decreasing the ability of fascicles to recoil efficiently 31, 32. A decreased ability to recoil may lead to an impaired response to further loading, potentially increasing the likelihood of injury occurring in fatigued tendons. Studies have also been undertaken to assess the effect of ageing on tendon composition and mechanical properties, in order to understand age-related alterations in tendon properties that may predispose aged tendons to injury. An early study reported that ageing had no effect on cellularity, total collagen content or glycosaminoglycan levels in the SDFT, but did result in increased levels of type III collagen in the core of the tendon 33. More recent work assessing the rate of matrix turnover in the SDFT and CDET with increasing age has shown that there appears to be an accumulation of partly degraded collagen within the aged SDFT 24. This has important implications because it suggests that the aged SDFT is more resistant to proteolysis, such that damage to the matrix cannot be repaired, potentially reducing the mechanical integrity of the tendon. When considering the effect of ageing on tendon mechanical properties, a recent study has shown that neither tendon nor fascicle failure properties are altered with increasing age 34. However, it has been shown that the interfascicular matrix becomes stiffer in aged SDFTs, but does not alter with age in the CDET 34. The reduced interfascicular matrix extensibility with increasing age in the SDFT is likely to be of great importance, because this would mean that, at any given strain, a greater proportion of the load would be transferred to the fascicles, increasing the likelihood of damage occurring to the fascicles, which may predispose the tendon to subsequent injury. The age-related alterations to interfascicular matrix composition and structure that result in the observed increased stiffness are yet to be established and so remain an important area for future study, because elucidation of alterations to specific interfascicular matrix components may provide potential targets for regenerative therapies. When considering the effect of ageing on the fascicle response to applied strain, studies have identified several ageing changes in the microstructural strain response of SDFT fascicles, characterised by reduced rotation in response to applied strain and a decreased ability to recoil and recover 30. These changes are similar to those seen as a result of fatigue loading in young tendon. No age-related changes in microstructural strain response were identified in the CDET, supporting previous studies, which suggest that ageing does not have a detrimental effect on the CDET 24, 29, 34. The fatigue response in aged SDFTs has also been studied; it has been demonstrated that the number of cycles to failure is significantly reduced in fascicles from aged SDFTs (decreasing from 16,825 to 4062 cycles) 31. This is accompanied by damage to the matrix, which is more severe than that seen in fatigue-loaded samples from young tendon, characterised by fibre kinking and widening of the interfibre space 31 (Fig 4). The microstructural strain response is also altered, with increased sliding between collagen fibres in response to applied strain 31. The age-related alterations to the helix substructure in the SDFT, indicated by decreased rotation 30, may mean that the fascicles from aged tendon are less able to withstand cyclic loading and are therefore more prone to damage and early failure. Schematic diagram showing alterations in superficial digital flexor tendon (SDFT) fascicle microstructural strain response with fatigue loading and ageing. In young, unloaded samples, rotation was observed upon the application of strain. In young, fatigue-loaded samples, the amount of rotation decreased, but was still present. In old, unloaded samples, less rotation was observed when strain was applied. In old, fatigue-loaded samples, matrix damage was observed, characterised by fibre kinking and widening of the interfibre space. Application of strain resulted in very little rotation, but fibre sliding dominated instead in these aged, fatigue-loaded samples. Adapted from Thorpe et al. 31. Recent studies investigating structure–function relationships in equine tendon have provided the first important insights into components of the tendon matrix that may be essential for efficient function. The interfascicular matrix, a component of the tendon that has previously been largely ignored, has been shown to be essential for providing the high extensibility required by the SDFT. Furthermore, age-related alterations to the interfascicular matrix have been identified, which may help to explain the increased risk of injury in aged SDFTs. In addition, it has been demonstrated that the fascicles within the SDFT also have a specialised structure, in which the formation of a helix provides a greater ability to recoil and increased fatigue resistance. This helical structure appears to be altered as a result of both fatigue loading and ageing, resulting in a decreased ability to resist further loading. This information may also help to explain why aged tendons are more prone to injury. However, more work is required to characterise fully the structural optimisations within the energy-storing SDFT and to establish the age-related alterations that result in increased risk of injury. This information will aid in the development of effective preventative strategies and treatment options for tendon injury. Specifically, there is a clinical need to develop imaging techniques to define fully the aspects of structural compromises that occur in early tendon failure. Further regenerative techniques require targeting towards the structural aspects of tendon function, such as the interfascicular matrix, which are key to normal SDFT function. Our research on tendon function and injury risk has been supported by the Horserace Betting Levy Board, PRJ 752.