Equine tendon mechanical behaviour: Prospects for repair and regeneration applications

Abstract

Tendons are dense connective tissues that play an important role in the biomechanical function of the musculoskeletal system. The mechanical forces have been implicated in every aspect of tendon biology. Tendon injuries are frequently occurring and their response to treatments is often unsatisfactory. A better understanding of tendon biomechanics and mechanobiology can help develop treatment options to improve clinical outcomes. Recently, tendon tissue engineering has gained more attention as an alternative treatment due to its potential to overcome the limitations of current treatments. This review first provides a summary of tendon mechanical properties, focusing on recent findings of tendon mechanobiological responses. In the next step, we highlight the biomechanical parameters of equine energy‐storing and positional tendons. The final section is devoted to how mechanical loading contributes to tenogenic differentiation using bioreactor systems. This study may help develop novel strategies for tendon injury prevention or accelerate and improve tendon healing.

Graphical Abstract: This review provides a summary of recent findings of tendon mechanical properties and mechanobiological responses, focusing on equine tendon. This study may help develop novel strategies for tendon injury prevention or accelerate and improve tendon healing.

1. INTRODUCTION

Tendons are fibrous connective tissues that transmit tensile forces to bones to generate movement. Tendons have additional functions in the equine superficial digital flexor tendon (SDFT) and the human Achilles tendon (AT) to allow energy storage (Ehrle et al., 2021; Zhang et al., 2022). This difference in tendon function is due to variability in the tendon structure and composition.

Tendons alter their structure, composition and mechanical properties in response to mechanical forces during development, ageing and injury in a process which is called tissue mechanical adaptation (Zamboulis et al., 2020). Mechanical responses of tendons depend on the tendon fibroblast (tenocyte). Tenocytes are aligned between fibres within fascicles with fibroblast‐like morphology. Despite tenocytes, tendons contain a pool of tendon stem/progenitor cells (TSCs), the exact location of which is unknown but may be present in endotenon, peritenon and perivascular regions with rounded morphology (Walia & Huang, 2019). TSCs may play a major role in tendon physiology as well as pathology such as tendinopathy. Overloading can induce the production of inflammatory mediators that may be attributed to the differentiation of TSCs into non‐tenocytes and also cause tendon micro‐damage leading to tendinopathy (Wang et al., 2020). There is also evidence of the presence of diverse cell populations present within tendons involved in maintaining tendon homeostasis, which are affected by ageing with the dysregulation of genes expression (De Micheli et al., 2020; Kendal et al., 2020; Zamboulis et al., 2023).

The tendon healing process is slow and is often associated with incomplete repair and often the inability to achieve the full range of motion. It is well known that repair tissues must be loaded in a controlled manner to enhance tendon healing while reducing scar tissue formation and tendon adhesions (Yang et al., 2016). Understanding the biomechanics and mechanobiology of the tendon during development, ageing, injury and healing processes is important to achieve the design of novel therapies. The horse is an accepted model for human tendon disease because SDFT and common digital extensor tendon (CDET) have an analogous function, to the human AT and anterior tibialis tendon, respectively (Lui et al., 2011). Moreover, tendon injuries in the SDFT show a very similar injury risk, aetiology and pathology to human AT injuries. There are similar differences in mechanical properties between energy‐storing and positional tendons in both species. However, there are fewer differences between tendon types in human than in horses. One reason for this may be related to differences in energy‐storing function as human AT stores less energy than equine SDFT (Patel et al., 2021).

Therefore, evaluating treatment strategies for tendon injuries in horses may be applicable to humans. Moreover, equine tendon injuries are a serious cause of loss of athletic potential, which results in considerable time and economic cost.

The traditional treatment options have unsatisfying results in achieving functional tissue. In recent years, tendon tissue engineering has gained attention as an alternative tissue source for autograft and allograft therapies due to their potential to overcome major problems, such as limited availability, donor site morbidity and immune rejection (Shojaee & Parham, 2019). Providing an appropriate biomechanical signal is a critical component for the success of functional tendon tissue engineering. However, optimal regimes of mechanical stimulation for different cells and materials are not yet clarified (Engebretson et al., 2018). Designing an in vitro loading protocol that mimics the in vivo situation is a challenging task. It is necessary to determine which parameter values lead to a tendon‐like construct.

Tendon experiences physiological mechanical stimulus, including tensile strain, compression and shear force. Mechanical stretching is the main type of mechanical stimulus. Therefore, parameters obtained from tensile tests provide very important insights into the new construction of engineered tendons to mimic the in vivo mechanical stimulation. Moreover, measuring the mechanical parameters of the tendon also is important for optimizing the rehabilitation protocol regardless of the type of treatment chosen to treat the lesion. For example, the effect of heel lift (18 mm) compared to barefoot showed that heel lift could reduce the strain and force on AT during rehabilitation training (Zhang et al., 2020). In horse case, the effect of different shoes under following conditions: barefoot, glue on heart bar and aluminium racing plates on tendon strain was calculated, and it was found that that aluminium racing plates with packing material may be most suitable for most racehorses as they do not significantly increase SDFT strain (Ault et al., 2015).

In the first part of this review, we overview tendon structure and the role of composition in the modulation of mechanical properties in every aspect of tendon biology. In part two, we refer readers to techniques used to measure tensile mechanical parameters. Following this background material, we present the results of published papers on tendon mechanical responses in current model systems. Because measuring the mechanical properties of the tendons is useful for the prevention or treatment of injuries, we summarize studies that calculate mechanical parameters in response to various factors, focusing on horses as an accepted model. In our previous review paper (Shojaee & Parham, 2019), current strategies for in vitro equine tenogenic differentiation such as the use of different scaffolds and mechanical stimulation, which are principles of tendon tissue engineering were discussed. But, due to the main function of the tendon, mechanical stimulation is a vital element in engineering load‐bearing tissues (Sheng et al., 2020); therefore, in the final section, we summarize studies using bioreactor system that is critical for tendon tissue engineering for tenogenic differentiation and highlight the mechanotransduction process and underlying signalling pathways. We hope that increased knowledge about tendon mechanical behaviour can help readers to design a suitable scaffold with tendon‐like mechanical strength and optimize the rehabilitation protocols and mechanical parameters of bioreactors inspired by the natural load.

2. TENDON STRUCTURE AND COMPOSITION

Tendons have a hierarchy of fibrillar arrangement structures with a high tensile strength (Lipman et al., 2018). The crimp or sinusoidal pattern of collagen fibrils facilitates the stretching of the tendon (1%–3%) to prevent rupture from sudden mechanical loading. The diameter of fibril varies from 10 to 500 nm depending on the species, age and location of sample (Ribitsch et al., 2020). Changes in hierarchical structure and compositions may lead to differences in mechanical properties.

The fascicles are surrounded by the interfascicular matrix (IFM), commonly termed endotenon, which is more elastic in energy‐storing tendons than in positional tendons and is critical for facilitating sliding between fascicles. The IFM niche has a more complex proteome and a more diverse range of cell populations than the fascicle matrix and, unlike the fascicle matrix, changes under loading during post‐natal development (Zhang et al., 2022). The fascicles forming the tendon are bound together by the epitenon (Lipman et al., 2018). Most tendons are surrounded by dense connective tissue, called paratenon that facilitates tendon movement, and where tendons pass over the joint, they are covered with synovial sheath to reduce sliding friction. Tendon is attached to the muscle by the myotendinous junction (MTJ) and to bone by a fibrocartilaginous tissue called enthesis (Lee, 2021; Subramanian & Schilling, 2015). The enthesis and mid‐substance of tendons have different compositions, resulting in different responses to mechanical loading. The tensile forces at enthesis may be four times more powerful than mid‐substance (Kaya, 2020). Moreover, regional differences in mechanical properties of rat AT were reported by Disser et al. (2022), which showed a 3.72‐fold increase in the Young modulus from the proximal origin (at the MTJ) to middle region and an additional 1.34‐fold increase from the middle to distal region (at the enthesis).

Tendon tissue is characterized by large amounts of extracellular matrix (ECM) and a low number of cells. About 55% of the weight of the tendon is water, which reduces friction to facilitate the gliding of fibrils in response to mechanical loading. Water has a significant relationship with the stiffness of tendon tissue (Lozano et al., 2019).

The ECM of the tendon is turned over by tenocytes, and the most abundant component in the ECM is collagen type I. The tensile strength of tendon depends on the crosslink of collagen in the tendon matrix. Collagen type III is present in the endotenon and epitenon of normal tendons, and it is also found in ageing tendons and at the insertion sites of highly stressed tendons. It is also found abundantly during the early phase of repair and reduces mechanical strength due to its ability to form smaller fibrils and less organized fibrils with rapid crosslinks. Other collagens, including III, V, VI, IX–XII and XIV, are only present in small amounts in tendons, which have different roles and localizations (Docheva et al., 2015; Nourissat et al., 2015; Schneider et al., 2018; Zabrzyński et al., 2018). Collagen XII is critical to the hierarchal structure and mechanical properties of tendon by the regulation of intercellular communication and regulation of fibrillogenesis (Izu et al., 2021). Tendons in a galloping horse experience ∼16 kN/kg of body mass, which highlights the persistence of collagen. However, the persistence of fibrils means that new collagen fibrils have different organizations compared to surrounding tissue after injury, leading to scar formation (Revell et al., 2021).

Besides collagens, tendons also contain non‐collagenous extracellular components, including proteoglycans (PGs), glycoproteins and other molecules that are responsible for the viscoelastic and time‐dependent behaviour of tendons. Most PGs are small leucine‐rich proteoglycans (SLRPs), and its content is associated with mechanical loading conditions (e.g. tension vs. compression) of the tendon site. For example, in the tensile region, SLRPs, such as decorin, biglycan, fibromodulin and lumican, are present, whereas PGs, such as aggrecan, biglycan, and lubricin, are mostly present in compressed regions (Iozzo & Schaefer, 2016; Thorpe, Birch, et al., 2013).

Decorin and biglycan are the most abundant SLRPs in tendon, but biglycan is found at a lower level than decorin and both regulate fibrillogenesis and ECM assembly. Decorin knockdown resulted in minor changes in tendon structure and no changes in mechanics, whereas biglycan knockdown resulted in expensive changes in tendon structure and mechanics (Beach et al., 2022).

Lubricin is another PG particularly important in energy‐storing tendons in which both glycoprotein and proteoglycan isoforms have been identified. The interfascicular gliding is disturbed in lubricin null mice, and fascicle viscoelastic properties are altered with lubricin depletion. Lumican and fibromodulin are other members of the SLRP family. In studies of knockout models, fibromodulin was shown to have a greater mechanistic effect than lumican (Eisner et al., 2022).

The glycoproteins that exist in the tendon ECM include tenascin‐C, elastin, fibronectin and cartilage oligomeric matrix protein (COMP) as the most abundant glycoprotein in the tendon. Elastin is a highly extensible, fatigue‐resistant fibrous protein that may influence fascicle sliding and contribute to the crimped collagen fibres recovery after stretching. Healing tendons have higher levels of elastin compared to intact tendons, possibly indicating the protective role of elastic fibres to prevent re‐injuries during early tendon healing (Svard et al., 2020).

As changes in hierarchical structure and composition may contribute to the differences in mechanical properties, here we highlight some differences between SDFT and CDET to clarify the mechanisms behind tendon disease. The equine SDFT has higher water, lower collagen content, higher levels of DNA, COMP and sulphated GAG than the CDET (Birch et al., 2013). The collagen fibril diameters of the positional tendons are larger than that of energy‐storing tendons, which increases stiffness due to the greater possibility of crosslinks (O'Brien et al., 2021). SLRPs are present within the IFM and may also contribute to sliding between fascicles. Biglycan, elastin, lubricin and lumican are often localized in IFM of both tendon types. Fibromodulin is more intense in the CDET than in the SDFT and exhibits significantly more staining in the fascicular matrix (FM). There are no significant differences in decorin staining between tendon types or regions (FM and IFM) (Thorpe, Karunaseelan, et al., 2016; Thorpe, Peffers, et al., 2016). COMP is more abundant in weight‐bearing tendons compared to unloaded tendons. The expression of COMP is very low in foal tendons and increases with maturity by mechanical loading (Posey et al., 2018). Tenascin‐C staining is restricted to the IFM and is absent from the fascicle through post‐natal development (0–2‐year old) (Zamboulis et al., 2020). The schematic diagram describes some structural properties between SDFT and CDET (Figure 1).

FIGURE 1.

Schematic diagram of superficial digital flexor tendon (SDFT) and common digital extensor tendon (CDET) structure: anatomy of equine forelimb tendons (a); comparison of structural properties between SDFT and CDET (b) (Zamboulis et al., 2020; Thorpe, Birch, et al., 2013; Thorpe, Udeze, et al., 2013; Thorpe, Klemt, et al., 2013; Birch, 2007). CSA, cross‐sectional area; IFM, interfascicular matrix. ‘++’ indicates higher values; ‘+’ indicates lower values.

3. MECHANICAL PROPERTIES OF TENDON

3.1. In vitro and in vivo mechanical testing

The most of knowledge about tendon mechanical properties is derived from an isolated tendon sample. The quasi‐static tensile tests mimic in vivo conditions of loading tendons. This is the preferred method for recording specimen deformation, where force is applied to a sample being pulled at a constant speed in order to failure (Wang et al., 2018). The load cell records the tensile deformation resulting from the displacement of the actuator. A force–deformation curve can be obtained from this test (Figure 2a). The slope of the curve describes stiffness (N/m) of sample, and its optimal level is vital for effective muscle–tendon interactions. The shape of the obtained curves is different between specimens, so force and deformation data are both normalized by the cross‐sectional area (CSA) and initial length to give stress and strain, respectively (Heinemeier & Kjaer, 2011).

FIGURE 2.

Schematic quasi‐static and time‐dependent mechanical properties of tendon: force–extension curve (a), stress–strain curve (b), creep (c), stress relaxation (d) and hysteresis (e). CSA, cross‐sectional area, ΔL; change in length; L ₀, original length; F, tendon force.

The schematic stress–strain curve for SDFT is shown in Figure 2B displaying different regions. In the toe region, the crimped collagen fibres are strained to ∼3%–4%. The difference in angle and length of the crimp pattern depends on the type of tendon and sample site. In the linear region, the tendon is stretched between ∼3.6% and 10.6%, and the slope of this linear is described as Young's module (stiffness has been normalized to tendon CSA and length). Micro‐damage has been recorded at strain levels of 12%–18% that corresponds to the strain levels found during galloping. The yield and failure regions are where partial tearing strain and complete rupture occur. The stretching ability of the tendon is different in various species and tendons. Studies of damage and failure in equine tendons suggest that positional tendons fail at lower strains (15%–22%) than energy‐storing tendons (18%–28%) (Screen et al., 2015). Typical parameters obtained from a stress–strain curve are Young's modulus (GPa), ultimate stress (i.e. stress at failure, MPa) and ultimate strain (i.e. strain at failure, %).

Tendon viscoelastic properties are directly correlated to collagenous and non‐collagenous proteins, water and the interactions between collagen and proteoglycans, as well as remarkably depending on age and activity. The viscoelastic behaviours of tendons have clinical importance to prevent fatigue failure of tendons. One of the viscoelastic characteristics of tendons is hysteresis. SDFTs exhibit low mechanical hysteresis than CDET (Thorpe, Karunaseelan, et al., 2016; Thorpe, Peffers, et al., 2016; Thorpe, Riley, et al., 2016). Another behaviour of viscoelastic tissue is time‐dependent properties defined by creep and stress relaxation. Schematic diagram describes the time‐dependent mechanical properties of tendon (Figure 2c–e).

Ex vivo or in vivo mechanical tests are extensively used to overcome the limitation of in vitro testing procedures associated with gripping samples and preserved methods for isolated tendon, which may lead to a premature fraction and misunderstanding of tendon properties, respectively. In ex vivo animal protocols, the muscle–tendon complex is intact, and loading is imposed by stimulation‐induced muscle contraction. The in vivo measurements of human tendon mechanics have been greatly advanced during the past years (Heinemeier & Kjaer, 2011). In these protocols, the limb is fixed on the load cell of a dynamometer plate to record the applied force. Moreover, isometric muscle contractions are generated by stimulation and MTJ displacement is recorded using ultrasonography.

3.2. Tendon response to mechanical loading

The role of mechanical loading in tendon has mostly been studied in the development and homeostasis, healing and differentiation. Some of the mechanobiological adaptations of tendon include changes in the diameter and crimp pattern of collagen fibrils, and the expression of tendon markers are discussed below.

The mechanical forces on tendons during normal locomotion are influenced by several factors, including tendon type, tendon regions (mid‐tendon vs. enthesis), muscle contraction level, tendon size and different daily activities. The different activities induce different levels of forces, even on the same tendon. For example, the maximal force in the SDFT is 3110 ± 1787 N at the walk, 5652 ± 2472 N at the trot, 7030 ± 2948 N in the leading forelimb, 6453 ± 2940 N in the trailing forelimb at the canter and the force in the SDFT cannot be measured at the gallop (Takahashi et al., 2010).

Therefore, measuring the load on the tendons is useful for the prevention and treatment of injuries.

In addition to changing the mechanical properties of tendon, mechanical loads also cause biochemical changes in tendons. Exercise and training lead to biochemical changes in tendons in an anabolic and catabolic way by increasing matrix proteins (e.g. collagen) production and stimulating the release of matrix‐degrading enzymes like matrix metalloproteinases (MMPs), respectively. Tissue inhibitor of MMP‐1 interacts with MMPs to prevent the excessive degradation of the ECM and tissue remodelling (Benage et al., 2022).

Moreover, the ability of the tendon ECM to loading adaptation decreases with maturation and ageing due to changes in cellular populations and reduction in cell numbers (Dowling & Dart, 2005).

A recent study showed that functional adaptation occurs post‐natally, following mechanical loading through transforming growth factor β (TGF‐β) signalling, and is related to the non‐collagenous IFM, suggesting that regenerative medicine strategies should shift focus from the fibrous to the non‐collagenous matrix (Zamboulis et al., 2020).

3.2.1. Physiological and pathological mechano‐responses

Normal physiological loads are necessary for appropriate tendon development and homeostasis, and they are also used in clinical rehabilitation protocols for tendon injuries after repair or regeneration.

The absence of muscle or the loss of continuous force transmission during development impairs tendon formation, size and biomechanical properties. It is likely because of muscle‐derived mechanical stimulation and/or secreted soluble (e.g. growth) factors (Edom‐Vovard et al., 2002; Maeda et al., 2011; Maredziak et al., 2015).

Pathological mechanical responses of tendons due to overuse, disuse or accumulation of tendon micro‐damage as an effect of ageing and repetitive strain have been reported. Measurements of mechanical property after acute exercise and long‐term training with a certain frequency serve as the basis to understand the mechanism of different modes of exercise on tendon injury to develop effective treatment intervention. High strain energy‐storing tendons, such as human AT and equine SDFT, are most commonly injured in athletes (Patel et al., 2021; Thorpe, Clegg, et al., 2010).

The effect of treadmill exercise (long and short terms) on the collagen fibrils of the SDFT in Thoroughbred (TB) horses with 24‐ and 39‐month old at the termination of the short‐term and long‐term regimens, respectively (Edwards et al., 2005). They showed that the mass average diameter (MAD) of the central region fibrils was significantly lower than that of control horses (walking exercise) and in a similar manner reduction in MAD of collagen fibrils in CDET. Exercise did not significantly affect fibril diameter while accelerating age‐related changes in SDFT compared to CEDET. Age‐related reduction in collagen fibril MAD agreed with findings (Logan & Nielsen, 2021) that showed the adaptability in mature horses (>4 years of age) is lower compared to juvenile horses (<2 years of age) in SDFT. In addition, a reduction in collagen fibril crimp has been demonstrated in response to ageing and exercise (Logan & Nielsen, 2021).

However, it is difficult to separate the effects of age from exercise in these studies. Evaluation of changes related to loading along with ageing on fibril morphology showed that training reduces fibril diameter in mature horses.

The effect of exercise on the CSA of the SDFT in mature horses demonstrated conflicting results. It is most likely that exercise cannot increase the CSA of the SDFT (Dowling & Dart, 2005). Similarly, the CSA of the SDFT increased during 5–8 months of age in the pasture and pasture along with forced exercise while there was a trend towards a larger CSA in a pasture along with forced exercise with ageing. Spontaneous pasture exercise compared to pasture along with forced exercise may be as beneficial for young horses because there is less risk of injury when starting race training (Moffat et al., 2008). Therefore, both a lack of exercise and over‐exercise in young horses can damage the tendon. Training horses before SDFT reaches maturity when the horse is of 2 years of age can produce stronger tendons at maturity (Wang & Zhao, 2022).

In vivo model showed that mechanical loading drastically alters the expression of tendon markers along with changes in mechanical parameters (Birch et al., 2008; Dudhia et al., 2007). Comparing two groups of female horses with high‐ or low‐intensity exercise training for 18 months showed that SDFT collagen fibril diameter decreased in the high‐intensity group, but no changes in collagen content and mechanical properties. In response to high‐intensity exercise, SDFT shows very little adaptation and reduction in glycosaminoglycan content, whereas CDET shows a higher elastic modulus; moreover, the values of their mechanical parameters are presented in Table 1 (Birch et al., 2008).

TABLE 1.

Mechanical properties of the energy‐storing (superficial digital flexor tendon [SDFT]) and positional tendons (CDET) in response to various factors.

				Results
Factors effecting the study outcome				Mechanical parameters values
Tendon type	Age	Breed	Model characteristics	Modulus (MPa)	Ultimate strain (%)	Ultimate stress (MPa)	Ultimate force (N/KN)	CSA (mm2)	Brief finding	References
1: SDFT 2: CDET	18 month	TB	Exercise for 18 month: (a)Low intensity (walking) (b)High intensity (gallop + trot)	1: (a) 1303 ± 110		1: (a) 128 ± 33	1: (a) 13,520 ± 4188	1: (a) 105 ± 25	High intensity: ↓ GAG content No change in collagen content; ↓ collagen fibril diameters No change of mechanical properties in SDFT Higher elastic modulus and lower water content in CDET	Birch et al. (2008)
				(b) 1245 ± 121	–	(b) 123 ± 24	(b) 13,335 ± 2923	(b) 98 ± 8
				2: (a) 1310 ± 95		2: (a) 187 ± 43	2: (a) 5420 ± 1474	2: (a) 32 ± 2
				(b) 1514 ± 55		(b) 203 ± 34	(b) 5307 ± 789	(b) 31 ± 3
SDFT	(1) 1–3 years (2) 4–10 years (3) 11–30 (4) Combined		(a) Control	–	1: (a) 22.86 ± 6.64	–	–	–	Age‐related effect of cyclical strain ↓ Ultimate strain;↑ levels and activity of MMP‐2, MMP‐9 and release of degraded COMP	Dudhia et al. (2007)
			(b) Strained (Stretch: 5%,1 Hz, cyclic strain 24 h)		(b) 17.70 ± 5.28
					2: (a) 20.83 ± 5.52
					(b) 19.42 ± 5.38
					3: (a) 16.53 ± 4.88 (b)22.80 ± 6.86
					4: (a) 19.96 ± 5.92
					(b)19.51 ± 6.62
SDFT	15–20 years	Mix breed	(a) Normal tendon	–	(a)12.4 ± 1.7	(a) 107 ± 11	(a) 12,356 ± 1333	(a) 117 ± 1.7	Greatest hypertrophy ↓ Stress and stiffness ↑ CSA in order to prevent overstraining in late stage of tendon healing	Crevier‐Denoix et al. (1997)
			(b) spontaneous injuries		(b) 10.8 ± 2	(b) 63.29 ± 29	(b) 14,498 ± 3582	(b) 282 ± 131
SDFT	5 months 1 month	WB	(a)No exercise (b)Exercise (c)Pasture	–	1: (a) 12 ± 1	1: (a) 119 ± 8	Range: 1:	1: (a) 63 ± 6	↑ CSA at young age with exercise ↑ Force rupture with aging in non‐exercise and decrease with pasture that may impact on later susceptibility injury	Cherdchutham et al. (2001)
					(b) 11 ± 4	(b) 112 ± 31	7.4 ± 0.8–9.3 ± 1.3	(b) 68 ± 9
					(c) 12 ± 1	(c) 99 ± 11		(c) 102 ± 26
					2: (a) 12 ± 2	2: (a) 106 ± 7	2: 11.5 ± 1.1	2: (a) 118 ± 20
					(b) 11 ± 2	(b) 116 ± 8	12.4 ± 1.3	(b) 103 ± 11
					(c) 11 ± 1	(c) 100 ± 10		(c) 115 ± 11
Collagenase‐induced injuries of SDFT	7–14 years	WB	Cyclically loading: (a)With cast immobilization (b)Without cast immobilization				(a) 6710 ± 2030 (b) 7450 ± 2170		Immobilization ↓ the lesion size; not affect force at rupture	Bosch et al. (2010)
1: SDFT 2: CDET	(a)3–8 years (b)9–4 years (c)15–20 years	TB	_	1: (a) 676.3 ± 142.2	1: (a) 21.1 ± 4.8	1: (a) 116.9 ± 19.4	1: (a) 1.49 ± 1.09	1: (a) 68.3 ± 20	No significant difference between age group Degraded collagen accumulates with age in SDFT	Thorpe, Birch et al. (2013), Thorpe, Udeze et al. (2013) and Thorpe, Klemt et al. (2013)
				(b) 626.8 ± 97.7	(b) 24.8 ± 6.0	(b) 130.7 ± 31.0	(b) 1.57 ± 0.78	(b) 97.5 ± 18.3
				(c) 586.2 ± 93.7	(c) 21.7 ± 3.3	(c) 104.0 ± 21.7	(c) 1.50 ± 0.93	(c) 98 ± 40.6
				2: (a) 1036.8 ± 176	2: (a) 15.3 ± 3	2: (a) 148.5 ± 33.7	2: (a) 1.38 ± 0.97	2: (a) 19.7 ± 3.5
				(b) 935.91 ± 75.6	(b) 15.8 ± 3.4	(b) 145.4 ± 33.1	(b) 1.76 ± 1.07	(b) 29.1 ± 8.7
				(c) 1024.0 ± 24.1	(c) 19.7 ± 3.7	(c) 160.3 ± 24.8	(c) 1.61 ± 1.11	(c) 27.2 ± 6.7
1: SDFT 2: CDET	3 to 12 years	(a) TB (b) WB (c) FH	–	1: (a)595.7 ± 00.1 (b)397.3 ± 87.1 (c)507.7 ± 41.4 2: (a)1012.6 ± 3.8 (b)971 ± 131.1 (c)929.6 ± 29.3	1: (a)25.3 ± 4.6 (b)32.6 ± 6.0 (c)26.4 ± 4.1 2: (a)21.5 ± 4.5 (b)20.5 ± 4.7 (c)21.4 ± 5.6	1: (a)121.4 ± 18.6 (b)116.9 ± 17.8 (c)110.4 ± 20.6 2: (a)171.9 ± 50.2 (b)153.3 ± 21.4 (c)160.5 ± 38.7	1: (a)9982 ± 3486 (b)12,606 ± 1936 (c)1010 ± 1583 2: (a)4332 ± 1229 (b)5583 ± 971 (c)4939 ± 976	1: (a)89.0 ± 26.3 (b)110.7 ± 27.5 (c)93.4 ± 16.9 2: (a)26.2 ± 7.3 (b)37.0 ± 7.7 (c)31.2 ± 2.8	SDFT from WB had significant lower modulus than TB and higher strain and load than TB and FH SDFT had larger CSA than CDET CDET had higher stress and modulus than SDFT No difference between breed in CDET	Verkade et al. (2019)
SDFT			(a) CT (b) DC	(a) 122.25 ± 0.78 (b) 140.22 ± 51.9	(a) 8.09 ± 1.54 (b) 11.86 ± 5.69	(a) 7.95 ± 3.12 (b) 16.66 ± 7.41			Decreases of ∼80%–87% residual DNA Eq‐DCT reseeding with hu‐FTP showed higher quantity of cells	Aeberhard et al. (2020)

Note: ‘↑’ means increased levels; ‘↓’ means decrease levels.

Abbreviations: BT, bone tendon – complex; CDET, common digital flexor tendon; COMP, cartilage oligomeric matrix protein; CSA, cross‐sectional area; CT, control; DC, decellular (by freeze/thaw [F/T] cycles + NaCl 1 M); FH, Friesian Horse; GAG, glycosaminoglycan; hu‐FPTs, human foetal progenitor tenocytes; MMP, matrix metalloproteinases; SDFT, superficial digital flexor tendon; TS, Tendon substance; WB, Warmblood.

Another study highlighted the role of mechanical loading during early development in the regulating tendon development with the gain and loss of function to induce hypermotility and paralysis in chicken embryos (Pan et al., 2018). Paralysis led to decreased modulus and also decreased the expression of lysyl oxidase (LOX) that plays a critical role in crosslinking collagen and elastin.

It is well known that tendon healing is promoted by adequate mechanical loading. However, the beneficial effect of mechanical loading depends on the injury site and the type of tendon. During the inflammatory phase, tendon lesions are usually enlarged that have a worse prognosis for functional repair (Bosch et al., 2010). A study on the effect of cyclically loading with and without cast immobilization on the propagation of collagenase‐induced SDFT lesions showed that the combination of enzymatic and mechanical stimulation induces lesion propagation and cast immobilization reduced the enlargement of primary tendon lesions. The study suggested that mechanical forces alone are not sufficient to cause lesion propagation, and other factors, such as the activity of proteolytic enzymes, also play a role (David et al., 2012).

In vivo model of surgically induced SDFT core lesions showed cast immobilization in the early healing phase effectively reduced lesion size compared to bandaging (David et al., 2012). Immobilization or unloading during the tendon healing process has a detrimental effect on the properties of the tendon that is time‐ and dose‐dependent and is also related to tendon type. This effect is completely reversible with remobilization among species and has faster and stronger effect in light species (David et al., 2012). It has been suggested that immobilization in human tendon may be used for about 5–7 days during the acute phase (David et al., 2012).

In addition to current model systems, computational models have been used to predict mechanical loading parameters and investigate how mechanical loading is attributed to tendon pathologies and healing at the cell and bulk tissue levels (Chen et al., 2018).

4. FACTORS INFLUENCING MECHANICAL PROPERTIES OF TENDONS

In addition to mechanical loads, other factors that affect the mechanical properties of tendons can be divided into (1) experimental factors (e.g. preserved or fresh specimen orientation) and (2) biological factors (e.g. age, diabetes and obesity subject). These factors can partially explain the different outcomes in the same tendon tissue. The specimen condition during tensile testing and the environmental factors, including temperature and hydration may affect the mechanical response of tendon tissue (Guney et al., 2015; Legerlotz et al., 2013; Oswald et al., 2017). Decellularized SDFT (150 × 10 × 1.2 mm) by freeze/thaw cycles plus NaCl (1 M) had higher values of strain, stress, modulus and energy to rupture compared to control group as presented in Table 1 (Aeberhard et al., 2020). A study measured the mechanical parameters of human AT at two strain rates, 10 and 1%/s, representing physiological activity and quasi‐static, respectively, and showed higher failure stresses and strains at faster rates with no changes in modulus (Wren et al., 2001).

Several studies have shown that ageing, as a biological factor, affects the tendon properties and may cause stiffer and stronger tendons, whereas other studies have challenged these results. This discrepancy may be partly due to the use of very young specimens and consequence changes of maturation, which may mask an actual ageing effect (Thorpe, Birch, et al., 2013; Thorpe, Udeze, et al., 2013; Thorpe, Klemt, et al., 2013; Svensson et al., 2016). In the equine SDFT, partly the accumulation of degraded collagen has been reported with ageing, but not in the CDET. Additionally, decreased ability for sliding between fascicles SDFT with ageing due to the high stiffness of IFM has been shown that may predispose aged tendons to tendinopathy (Thorpe, Birch, et al., 2013; Thorpe, Udeze, et al., 2013; Thorpe, Klemt, et al., 2013). The fascicle CSA in the core region of the SDFT increased with ageing. Other values of the mechanical parameters of two types of tendons are presented in Table 1.

The result of the tensile test performed for AT of diabetic and non‐diabetic patients showed that non‐diabetic tendons exhibit superior biomechanical parameters compared to diabetic patients (Guney et al., 2015). These changes are the result of increased collagen crosslinking caused by advanced glycation end‐products (AGEs), which is accelerated by hyperglycaemia. AGEs are sugar‐induced non‐enzymatic crosslinks that accumulate with age and/or diabetes in ECM. Thorpe, Clegg et al. (2010) and Thorpe, Streeter et al. (2010) found a difference in AGE content and collagen turnover in SDFT and CDET with ageing.

A recent in vivo study investigated the effect of obesity on tendons and found that CSA and stiffness of AT were positively correlated with body mass and body mass index (BMI) in young but not older adult men, and the negative impacts of obesity on weight‐bearing tendons are exacerbated with ageing (Tomlinson et al., 2021). Equine studies have shown a potential relationship between obesity, or body weight:height ratio, with the development of orthopaedic problems. A linear relationship between CSA of SDFT and BMI was also reported for 6‐year‐old horses (Agut et al., 2009; Gruyaert et al., 2020). Another study investigated several risk factors for SDF tendinopathy Odds ratio (OR) and showed an increased OR with increasing body weight (≥470 kg) and also reported that risks of SDF tendinopathy in males and geldings were significantly higher than in females (Ikeda et al., 2019).

The therapeutic potential of corticosteroids due to their anti‐inflammatory properties can be considered for the management of tendinopathy, but there is a risk of post‐treatment laminitis after, especially topical use, and frequent administration (Cornelisse & Robinson, 2013). Therefore, the use of steroids may predispose tendons to rupture and reduces mechanical properties such as stiffness and ultimate stress after administration, leading to impairing the tendon healing process (Dean et al., 2014).

The effects of methylprednisolone acetate (MPA) on dorsal fibrocartilaginous region of forelimb DDFTs‐derived cells showed that the higher MPA concentration significantly down‐regulated tendon ECM‐related genes after 24 h of treatment although the mechanical responses were not investigated.

5. MECHANICAL PROPERTIES OF THE ENERGY‐STORING AND POSITIONAL TENDON

Not surprisingly, each type of tendon experiences a different range of mechanical properties because it performs different functions within the body. The assessment of mechanical properties of tendons has received more attention because of its importance for normal activity and because tendon injuries are rather prevalent and difficult to treat.

For efficient force transfer, positional tendons should be stiffer under physiological loads, and strains should not exceed 3%. In contrast, energy‐storing tendons require a high degree of extensibility because they experience large strains during high‐speed exercise. For example, strains in the SDFT of galloping TB can reach up to 16%, and strain in the human AT has been measured to exceed 10% during one‐legged hopping exercise (Thorpe, Birch, et al., 2013; Thorpe, Udeze, et al., 2013; Thorpe, Klemt, et al., 2013).

Thorpe et al. (2012) demonstrated great differences in quasi‐static properties of whole tendons compared to the fascicle level. SDFT has a higher failure strain, lower modulus, lower failure stress and larger CSA than the CDET. These studies also show that SDFT fascicles have a lower failure strain and no difference in fascicle failure stress or modulus than in CDET fascicles. SDFT fascicles exhibit less hysteresis and greater recovery in response to the applied strain than CDET fascicles. Micromechanical response to loading is directly related to the cycle number of applied loading (Thorpe, Birch, et al., 2013; Thorpe, Udeze, et al., 2013; Thorpe, Klemt, et al., 2013). The nanoscale mechanical responses of SDFT and CDET were not reported. The horse with natural tendonitis of SDFT showed larger CSA, lower ultimate stress, lower ultimate strain and higher maximal loads compared to normal SDFT; the values of mechanical parameters are presented in Table 1 (Crevier‐Denoix et al., 1997).

In contrast to studies using a mix of horse breeds, one study compared the mechanical properties of the CDET and SDFT between breeds. The type of sport and the locomotor performance level are associated with differences in tendon injuries. In TB racehorses, the forelimb SDFT is often injured, which is localized in the core of the mid‐metacarpal region. The lower ultimate strains predispose to overstrain injury in the TB SDFT, which is a major cause of wastage. Suspensory ligament injuries are common in sport horses (Friesian Horse [FH], Warmblood [WB]) (Murray et al., 2006). The in vitro rupture pattern of FH tendons showed rupture in the core with intact outer fibre rather than a clean break in the TB and most of the WB tendons. The difference in the rupture pattern in FH tendons compared to other breeds may be due to different collagen distributions in FH tendons; it should be noted that the age range was relatively large, and exercise history was unknown (Verkade et al., 2019). As presented in Table 1, the mechanical properties of the CDET and CSA of the SDFT did not differ between breeds (Verkade et al., 2019, 2020). A summary of studies that have assessed quasi‐static mechanical properties of SDFT and CDET in different cases under the influence of various conditions, such as ageing, breed, orthotic shoes, sex and activity, are shown in Table 1.

The assessment of in vivo mechanical properties of human AT showed that tendinopathic AT had greater CSA, lower tendon stiffness and modulus compared to age‐ and gender‐matched controls (Arya & Kulig, 2010). Typical mechanical parameters have been reported for many tendons, including modulus in the range of 500–1850 MPa. Failure stresses are in the 50–125 MPa range, and failure strains at tendon mid‐substance are 5%–16% (Wren et al., 2001).

6. MECHANICAL LOADING IN TENDON ENGINEERING APPLICATIONS

Since Altman et al. in 2002 designed the first 3D engineered tendon/ligament bioreactor system, the effect of mechanical stimulation on the engineered tendon/ligament has received increasing attention (Wang et al., 2013; Dyment et al., 2020).

Bioreactors are used to mimic in vivo loading regimes and provide a dynamic culture system with the low risk of contamination and high reproducibility as well as generate functional tissue (Salehi‐Nik et al., 2013). The basic components of bioreactors are an actuating system designed for mechanical stimulation and the culture chamber with a controlled culture environment. The common actuators for uniaxial strain used in tendon bioreactors are pneumatic actuators, linear motors and step motor ball screws (SMBSs). The accuracy of pneumatic actuators is ∼±0.1 μm, which is not insignificant when typical tendon bioreactors require <5% strain on 1–5‐cm tissues. The linear motor is popular and the highest accuracy ∼±1 μm. In multi‐chamber systems, SMBS is widely used with an accuracy of around ±5 μm. It is suggested that the stimulation pattern should be adjusted based on the maturation of the engineered tendon/ligament, with higher loading dose applied at the later stage of tissue culture (Wang et al., 2013).

Mechanical stretching can be controlled by strain, frequency, distance and duration. The strain experienced in equine tendons at a canter (with a stride frequency of ∼2 Hz) can be applied in tendon tissue engineering, whereas most studies applied lower than 1 Hz (Atkinson et al., 2020). According to periods of around 20 min on the canter in training horses, rest intervals should be considered in tendon tissue engineering in order to restore cell mechanosensitivity to achieve the best conditions for inducing tenogenic differentiation. Stride frequency experienced in equine tendons ranged from 88.3 to 94.0 strides/min at the trot and from 117.5 to 123.5 strides/min at the canter (Groppi, 2008). In vivo strain rates at the gallop of 200% per second are in agreement with predicted in vitro strain rates of 150%–200% per second (Dowling & Dart, 2005).

There are a few studies in mouse models exist that compare in vitro and in vivo mechanical loadings with identical loading protocols to determine whether reproducible results from an in vivo study occurs (Kim et al., 2016; Fleischhacker et al., 2020). These types of research studies could help tissue engineers to determine to what extent in vitro stimulation set‐ups can mimic in vivo characteristics.

In addition, the bioreactor based on operational needs may also include a medium circulation system, monitoring system, feedback system and a medium analysis system (Wang et al., 2013). Initial bioreactors that are designed to apply uniaxial strains and commercial bioreactor systems are not able to provide the torsional loading. As tendons can experience both uniaxial and biaxial strains, custom‐built bioreactors have been developed to apply multidimensional strains, such as shear stress, tension (uniaxial and biaxial) and compression (Wang et al., 2018). The conflicting results on tenogenic differentiation under both uniaxial and biaxial stretching have been reported (Wang et al., 2018; Kayama et al., 2016). Mechanical stimulation depends on stimulation regimes (percentage of strain) that can induce the tenogenic or osteogenic differentiation of stem cells (Zhang & Wang, 2010; Burk et al., 2016; Chen et al., 2015).

It has been reported that tensile loading induces tenogenic differentiation, whereas fibrocartilage‐like phenotype enhances following compressive loading in MSCs (Schiele et al., 2013). The loading parameters derived from embryonic development may improve tenogenic differentiation towards tendon formation (Schiele et al., 2013). The proper stimulation induces the tenogenic markers including scleraxis (SCX) as tendon progenitors marker, Mohawk (MKX) and early growth response1 and 2 (Egr1/2) as well as genes encoding tendon‐specific ECM proteins such as COL (Liu et al., 2014).

Equine SDFT fascicles subjected to 2%–12% applied uniaxial strain and 1800 cycles at a frequency of 1 Hz designed to mimic high‐intensity exercise indicated that increased amounts of inflammatory mediators and collagen degradation markers were induced in response to overload (Thorpe et al., 2015). Despite the improved knowledge using 2D culture combined with mechanical stimulation, which is only suitable for loading monolayer cell cultures, 3D culture was developed to mimic in vivo cellular behaviour using hanging drop cultures (Kraus et al., 2017; Theiss et al., 2015) and self‐assembly (Bavin et al., 2015). Moreover, despite the developed 3D culture, the lack of the homogeneous distribution of nutrients is one of the main disadvantages of these techniques compared to 3D culture under static bioreactors. Because mass transfer into a 3D culture under static bioreactors (e.g. spinner flasks and stirred systems) is very limited, 3D culture combined with dynamic mechanical loading, in the form of static or cyclic stimulation, overcomes the above‐mentioned weakness. A summary of studies on tenogenic differentiation by using bioreactor systems is given in Table 2. In these studies, the effect of loading regimes (excitation duration and amplitude frequency) or the repetition pattern of the loading process was considered and discussed below.

TABLE 2.

Summarize of studies using bioreactor system for tenogenic differentiation.

						Results
Cell source	Scaffold (dimension)	Bioreactor type	Actuating system	Parameters of mechanical stimulation	Measurement method	Biochemical changes	Biomechanical parameter	References
eq‐BM‐MSCs	eq‐DCT 12.4 ± 1.1 mm 1.7 ± 0.1 mm	Custom‐made steepr motor bioreactor	Cyclic strain	0%, 3% or 5% strain at 0.33 Hz for up to 1 h daily for 11 days	elongated at 0.5% per second until failure	3% strain increased the elastic modulus, promoted tenocytic differentiation and decreased COL3	Failure stress: (CT: 14 ± 5; 3% DC: 17.7 ± 3.8); Elastic modulus: (CT: 98 ± 25; 3% DC: 119 ± 44)	Youngstrom et al. (2015)
Eq‐BM–MSCs ASCs TDSCs	eq‐DCT Length: 17.26 ± 2.73 mm, width: 2.48 ± 0.33	Stepper motor linear actuator bioreactor	Cyclic strain	1 h per day at 3% strain and 0.33 Hz	0.1% strain per second to failure	TDSCs expressed the greatest levels of SCX, COL1 and COMP	No differences in elastic modulus among groups cell‐laden constructs significantly increased in tensile strength compared to controls	Youngstrom et al. (2016)
Equine tenocyte	Self‐assembly constract		Uniaxial strain	10% strain at 0.67 Hz 20 min every 24 h/day for 14 days	1–0.1 Hz at 1% strain			Atkinson et al. (2020)
eq‐ASCs	eq‐DCT	Custom‐made uniaxial Cyclic strain bioreactor	Uniaxial Cyclic strain	2% strain 1 Hz		COL1a2 decreased and DCN, COL3a1, SPP1, COL3 increased TNC and SCX decreased initially but increased 24 h after stimulation		Burk et al. (2016)
eq‐teno‐iPSCs	poly(ɛ‐caprolactone nanofiber scaffold	Customized bioreactor	Cyclic uniaxial sinusoidal	1.0 Hz with 0%–6% sinusoidal wave of strain		Increased expression of SCX, EGR1, COL1A2, TNC		Yang et al. (2019)
R‐BM‐MSCs	eq‐DCT	Stepper Linear Actuator	Cyclic strain and perfusion 15–30–60 min on‐off two times/day for 7 days	3% strain at 0.33 Hz		Superior production and organization of newly formed collagen matrix		Talo et al. (2020)

Abbreviations: ASCs, adipose‐derived mesenchymal stem cells; COL1/III, collagen type1/3; COMP, cartilage oligomeric matrix protein; CT, control; DC, decellular; DCN, decorin; eq/R‐BM‐MSCs, equine/rabbit bone marrow‐derived mesenchymal; eq‐DCT, equine decellularized tendon; SCX, scleraxis; SPP1, osteopontin; TDSCs, tendon‐derived MSCs; teno‐iPSCs, equine tenocyte‐derived iPSCs; TNC, tenascin‐C.

Decellularized equine SDFT has been widely used for tendon tissue engineering applications (Table 2). Decellularized eq‐SDFT has been reported to be a potential source for human hand tendon grafting thanks to its large dimensions and less immunogenic effects compared to other species like bovine (Aeberhard et al., 2020). Engebretson et al. (2018) investigated the effects of varying duration and frequency on tenogenic differentiation of human BM‐MSCs cultured on decellularized human umbilical vein and showed beneficial effects after 7 days; meanwhile, they suggested further examination is needed to determine whether the regime is optimal for longer culture or a different stimulation regime is required.

Dynamic perfusion culture not only improves the cell metabolism but is also required in the case of highly oriented, dense collagen tissues like tendons compared to rotating culture. Tissue engineering system simulation is useful for predicting the effect of factors such as 3Dscaffold geometries and bioreactor geometries on engineered outputs. Talo et al. (2020) designed an oscillating stretch–perfusion bioreactor and developed a bidirectional flow without an external pump. They used computational flow dynamics analysis and achieve a uniform perfusion velocity of 100 μm/s on the decellularized eq‐SDFT. The results of bioreactor functionality are presented in Table 2.

Moreover, mechanical stimulation applied to human and equine stem cells alone or in combination with other factors, such as the scaffold material (Nirmalanandhan et al., 2007), scaffold anisotropy (Grier et al., 2017) and growth factor supplementation (Govoni et al., 2017; Raabe et al., 2013; Rinoldi et al., 2019), affects the response of tissue‐engineered tendon constructs.

However, these experiments using conventional methods require a large number of cells, samples and bulky incubators, a large volume of culture fluid as well as costly equipment. Besides large‐scale classical dynamic bioreactors, micro‐bioreactors have become valuable tools for 3D cell cultures (Altmann et al., 2021). In this way, the primary micro‐bioreactor designed for the application of ligament‐like multidimensional mechanical strains was also able to support the selective differentiation of bone marrow cells towards ligament‐like cells in the absence of specific growth/differentiation factors (Altmann et al., 2021).

In addition, magnetic force can be used as a mechanical stimulus to reduce the negative effects of strain‐based bioreactors on cultured cells and scaffold properties. Magnetic field stimulation enhanced the tenogenic differentiation of human‐ADSCs cultured with magnetic nanoparticle–decorated fibrous scaffolds and modulated inflammatory responses (Tomas et al., 2019). Maredziak et al. (2015) demonstrated that static magnetic field enhanced the proliferation and synthesis of secreted microvesicles in equine ASCs, although no studies on the tenogenic differentiation of equine stem cells have been conducted.

The long‐term goal is for these tissue‐engineered tendons to eventually be used clinically. Some in vivo studies evaluated the functionality of engineered tissue and showed promising results of musculoskeletal tissue regeneration in horses (Bolanos et al., 2017; Diloksumpan et al., 2020). The production of the tissue‐engineered tendon with the ability to withstand mechanical load is needed for clinical application in athletic horses. It should be considered that equine tendon lesions are located in the centre of SDFT, the design and development of an effective and strong surgical glue is necessary to attach to the tissue under dynamic movement for the treatment of tendon injury. The strong, stretchy hydrogel adhesive may be a promising option not only as a patch that can be cut to the desired size for use in many tissues but also as an injectable solution for deeper injuries, which can be used to fill tissue defects with different shapes and sizes (Liu et al., 2020).

7. MECHANOTRANSDUCTION MECHANISMS IN TENDON CELLS

The adaptive cellular responses mentioned above raise a question about mechanotransduction. The mechanism of converting mechanical stimuli into biochemical signals may be through direct cell‐to‐cell connections (e.g. gap junctions and cadherins) or cell‐to‐ECM adhesion molecules (CAMs, e.g. integrins) leading to the activation of signalling pathways eventually resulting in the adaptive response to mechanical loading (Arvind & Huang, 2017). The connexins (Cx) 32 and Cx43 are involved in the tendon gap junction. Cx32 connects tendon cells in the stretch direction that may stimulate collagen synthesis, whereas Cx43 links cells laterally and longitudinally (Schiele et al., 2013). Integrins are composed of α and β subunits and interact with ECM proteins, including fibronectin (ligand for α5β1) and laminin (ligand for α6β1) (Schwartz, 2010). The transmission of mechanical stimulus is mostly through cell–ECM interaction within the 3D‐tendon‐like constructs (Wang et al., 2018). ECM deformation can be transmitted to the cytoskeletal components resulting in actin cytoskeleton remodelling and signalling cascades within the cell (Schiele et al., 2013; Wang et al., 2012). Additionally, growth factors/cytokines may activate the signalling pathway with mechanical loading that can affect adaptive response (Wang et al., 2020; Shojaee et al., 2022).

Major signalling pathways involved in the mechanobiological responses in tendon development, repair and differentiation include the TGF­β and FGF–ERK/MAPK (mitogen‐activated protein kinase) signalling pathways, whereas it should be noted that the effect of FGF–ERK/MAPK is different among species (Maeda et al., 2011). The tenogenic differentiation induced by mechanical stimulation may be mediated by focal adhesion kinase, actin cytoskeleton and RhoA/Rock signalling pathways, phosphatidylinositol 3‐kinase/protein kinase B (PI3K/A KT), yes‐associated protein/transcriptional coactivator with PDZ binding (YAP/TA Z) (Wang et al., 2018; Schiele et al., 2013; Xu et al., 2012). The activation of the c‐Jun N‐terminal kinase/stress‐activated protein kinase (JNK/SAPK) in patellar tendon fibroblasts is regulated by a magnitude‐dependent mechanism (Wang et al., 2020). Effects and signalling pathways involved with the same mechanical stimulation protocol can differ between a 2D and a 3D culture systems under dynamic bioreactors (Wang et al., 2018).

Expanding knowledge in tendon mechanobiology may contribute to tendon differentiation before clinical application as well as identification of specific mechanotransductive signalling pathways and their pharmaceutical manipulation in combination with rehabilitation regimens may help to manage tissue inflammation and scar/adhesion formation during tendon healing (Wang et al., 2012; Shojaee et al., 2019).

8. CONCLUSION

Tendons are mechano‐responsive tissues and the appropriate range of mechanical forces (e.g. intensity and duration) is fundamental to their development, clinical rehabilitation protocols and successful functional tendon tissue engineering that provides the basic rationale for its medical or surgical treatments following injury.

In vitro mechanical tests show differences in the quasi‐static and time‐dependent mechanical behaviour of energy‐storing and positional tendons according to their function. Despite in vitro testing techniques being remarkably advanced, there are many challenges to measuring tendon strain during dynamic movements that require more in vivo studies on the effect of loading in intact and healing tendons. When evaluating the function of a given tendon, it is essential to investigate the function of the associated muscle.

Tendon injury occurs when the stress imposed is higher than the tissue can withstand and often responds poorly to treatments. The type and prevalence of injury depend on several factors, such as age, daily activity levels, breed and sex.

The findings discussed in the present review suggest that a deep and complete understanding of the biomechanical properties of tendons can help develop more efficient ways to prevent the injury. Furthermore, it would be useful to design treatment methodologies and optimal rehabilitation protocols, which accelerate the return of athletes to normal daily activities and sports. Due to different loading conditions, further studies are required to evaluate the effects of using different regimes to optimize specific conditions and should be adapted to the cell's response and the properties of the scaffold.

Finally, the data on the biomechanical properties of SDFT may have a great translational value for human AT injuries. However, further studies will be needed to evaluate tensile testing mechanical properties with optimal obtained parameters by considering the effect of preconditioning and strain rates as well as comparing them and their composition with those obtained from AT humans.

AUTHOR CONTRIBUTION

Writing—review and editing: Asiyeh Shojaee.

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no conflicts of interest that could be perceived as prejudicing the impartiality of the research reported.

PEER REVIEW

The peer review history for this article is available at https://publons.com/publon/10.1002/vms3.1205

Shojaee, A. (2023). Equine tendon mechanical behaviour: Prospects for repair and regeneration applications. Veterinary Medicine and Science, 9, 2053–2069. 10.1002/vms3.1205

DATA AVAILABILITY STATEMENT

The sources for the information discussed in this review can be obtained from the papers cited in the references.