Tendon Development, Healing, and Adhesion Formation

In the United States, nearly 1.5 million visits to the emergency room are a result of flexor tendon injuries. Even with intricate repair of these injuries, adhesion formation remains a common complication. Significant progress has been made to better understand the mechanisms of tendon healing and adhesion formation, however there has been slow progress in clinical prevention or reversal of flexor tendon adhesions. The goal of this article is to discuss the recent literature relating to tendon development, tendon healing, and adhesion formation to identify areas in need of further research.

Introduction

Tendon injuries to the hand and wrist constitute one of the most common disorders of the human body, affecting 1 in 2,700 people each year[1, 2]. These tendon injuries can result from trauma, overuse, age-related degeneration from work, daily life, and sports activities. Injuries to tendons, tendon-bone-junctions, and related tissues (such as ligaments) can occur in numerous areas of the body, however flexor tendons covered by an intrasynovial sheath have limited vascular supply, lack sufficient cellularity, and have low growth factor activity. Due to these qualities, more than 30% of these injuries result in adhesion formation resulting in disability[3-5].

Non-operatively and operatively managed flexor tendon injuries can both be complicated by fibrotic adhesions which severely impair the function of the hand by disrupting the gliding mechanism[5, 6]. Tendon adhesions to the fibro-osseous canal and surrounding tissues have been associated with a myriad of pathological factors[5]. Many pharmacological agents (such as hyaluronic acid, 5-Fluorouracil, lubricin, and a variety of growth factors) and mechanical barriers have been investigated in reduction of adhesion formation, but none of them have been proven useful in clinical settings[7-10].

Our understanding of the formation of flexor tendon adhesions remains limited.[11]. We will discuss what is currently known about intrasynovial tendon development, tendon healing, growth factors involved in tendon healing compared those in tendon development, and the role they play in both repair and adhesion formation.

Intra-synovial Tendon Development

Limb tendons arise from the lateral plate mesoderm, which forms secondary to BMP-4 secretion provided by the ectoderm [12]. These same cells give rise to endoskeletal cartilage. Tenocytes themselves are distinct from other fibroblast-like cell types [13]. Mature tenocytes are spindle-shaped and can be identified in mouse embryos as early as embryonic day 13.5. Although tenocytes are noted to be sparse in mature tendon tissue – generally anchored to the collagen fibers they produce, changes in their structure and activity has been specifically linked with a variety of tendonopathies [14].

Tendons are composed primarily of Collagen type I with the fibrils organized along the axis of the tendon. Type I collagen is made up of two a1 molecule chains (encoded by the gene Col1a1) and one a2 molecule chain (encoded by the gene Col1a2), which form a triple helix [15]. Although much remains to be fully understood, it is thought that most of the fibril assembly takes place during the pre-natal period, while the tissue grows and matures postnatally [15]. This maturation process includes a dramatic increase in the elastic modulus [16]. In addition to collagens, small leucine-rich proteoglycans are important for tendon development and growth, particularly in terms of regulating the growth of collagens [15].

The only known marker for developing tendon is the transcription factor, scleraxis (Scx). This regulates Col1a1 in mice and is known to be important in tendon development in chick and zebrafish as well. Tenocyte overexpression of Scx causes upregulation of the gene tenomodulin (Tnmd), the protein product of which is specific to tendons and ligaments and is understood to be a marker of tendon formation [17]. Postnatally, Scx expression is largely restricted to the epitenon [18]. Two other transcription factors are known to guide tendon development in vertebrates: Mohawk (Mkx) and Early growth response 1 (Egr1). Mkx-/- mice show smaller tendons with defective collagen [19]. Egr1-/- mice also show collagen fibril defects; tendons from these animals are weaker than wildtype and have healing deficiencies after injury [20].

A variety of other factors are known to be involved in intra-synovial (flexor)tendon development. These include cytokines, chemokines, and signaling molecules. Mechanical forces also play a role. Limb tendons initiate their development independently of muscles; however, muscles are required for subsequent tendon differentiation [21]. Fibroblast growth factors (FGFs) and the transforming growth factor beta (TGFβs) family are known to promote tendon commitment of limb mesodermal cells and act downstream of mechanical forces to regulate tendon differentiation during chick limb development. TGFβ2 was noted to be tenogenic for TPCs (tendon progenitor cells) at all stages of dev in vitro [22], while FGF4 lacked tenogenicity for TPCs in vitro. However, FGF4 is believed to induce and maintain Scx expression during tendon development [23]. BMP12 signaling, via Smad 1/5/8, guides expression of Scx, Tnmd, Col1and tenascin-C (Tn-C) in TPCs in vitro. This effect was found to be positively regulated by connective tissue growth factor (CTGF) [24].

While vascularity is limited in mature tendons, vascular endothelial growth factor (VEGF) signaling is important during tendon development in human tissue – specifically within developing tendons under traction – while gliding tendons maintain an avascular zone even from the fetal period [25]. Nearly all of the 23 known matrix metalloproteinases (MMPs) as well as the 19 disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) proteins can be identified in adult tendon specimens. These are involved in regulation of the tendon extracellular matrix (ECM) as well as establishment of the muscle-tendon junction [26]. Some of the specific ECM proteins, such as fibronectin and laminin-a1, are also known to be involved in interactions at the muscle-tendon junction. Other factors known to be involved in tendon healing, which are also believed to play a role in tendon development, include insulin growth factor 1 (IGF-1), platelet-derived growth factors (PDGFs), and interleukins such as IL-6 and IL1B, and their receptors[27].

Intra-synovial Tendon Healing Process

Successful flexor tendon healing requires restoration of the baseline collagen fibers in the tendon and re-establishment of the tendon gliding within the sheath. Tendon healing is believed to involve both the extrinsic and intrinsic pathways and is comprised of three phases: inflammatory (days 1-7), fibroblastic (days 3-14), and remodeling (beyond day 10) [6, 28-31]. The extrinsic mechanism proposes that cells not resident to the local injury niche, such as immune cells and fibroblasts, are directly involved in repair[32]. The intrinsic mechanism suggests that the cells involved in tendon repair are from within the tendon[33]. Following an injury until approximately 3-7 days post injury, an acute inflammatory response is initiated. Immediately following an injury, an acute inflammatory response is initiated, with both resident intrinsic cells from the epitenon and endotenon, as well as extrinsic cells from the surrounding peritendinous are recruited to and proliferate at the injury site[6]. The strength of the tendon during this phase is almost entirely reliant on the blood clot. If a surgical repair is performed, then the surgical suture provides the majority of the mechanical strength of the tendon[34, 35].  The strength of the tendon does not begin to increase until the fibroblastic phase is initiated at day 3. During the fibroblastic phase, the injury site becomes hypercellular as components of the extracellular matrix are deposited. Initial deposition of Collage type III occurs in a disorganized fashion, which is reorganized into longitudinal structures. The Collagen type III is subsequently replaced with Collagen type I during the remodeling phase. Over the span of the ensuing 2 months, the tendon tissue matures, and the prevailing tension forces cause the fibers to reorient longitudinally. Unfortunately, the repaired tendon will never achieve its full uninjured strength[36, 37].

Factors involved in Flexor Tendon Adhesions

A great deal of research has been devoted to understanding the formation and prevention of tendon adhesion after injury and/or surgical repair[6].  Through the use of animal studies, we have identified some of the critical aspects of tendon healing and adhesion formation.

In addition to the extent of initial injury and quality of subsequent surgical repair, mechanical loading is critical to the reduction of adhesions[38-45]. Mechanical loading upregulates the expression of Collagen type III mRNA expression in tenocytes and increases the concentration of growth factors, resulting in cell proliferation, differentiation, and matrix formation at the injury site[38]. However, tendon repair rehabilitation must balance between protection of the repair from excessive forces and prevention of adhesions with some loading. Excessive loading may not only rupture the repair, but may impair healing[46].

Prolonged immobilization will also result in adhesion formation[47, 48]. Wong et al. developed a mouse model in which the flexor tendons are immobilized via the creation of a proximal tenotomy to the injury site, which greatly increased the likelihood of adhesion formation[28]. They also found that adhesions are a result not only from resident local cells but also from cells in the surrounding tissue that had trauma. These cells appear to develop an excessive amount of collagen.

Aging has also been found to be associated with impaired healing in flexor tendons, as well as patella and rotator cuff tendons[49-51]. Ackerman et al. found that at baseline tendons in older mice have a decrease in cell density. It is unknown what proportion of adult cells within the tendon are tendon progenitor cells or stem cells. However, it is likely that the number of tendon progenitor cells decreases in proportion with the age-related decrease in cell density. Additionally, as cells age within a tendon, they lose their rounded morphology taking on a more elongated shape. The number of organelles within the cells also decreases[49, 52, 53]. Collagen synthesis and collagenolytic activity diminish with age[54]. This results in an altered composition and alignment of collagen fibrils in the aging population, while collagen fibrils of the young tendon are largely homogenous and arranged in parallel[49]. Interestingly, after flexor tendon injury, there is a decrease in the amount of extracellular matrix (ECM) deposition in older mouse tendons, and the mechanical strength is diminished[49, 55]. Despite significant research on tendon healing strength, there has been minimal investigation into flexor tendon adhesions with aging.

Recapitulation of Tendon Development Processes in Tendon Healing

Tendon healing is a complex process controlled by a variety of regulatory growth factors. Many of the same processes and regulators involved in tendon development are involved in tendon healing[27, 56]. Growth factors including TGFβ, bFGF, VEGF, and PDGF have been extensively studied in a variety of tendon healing models both in vivo and in vitro (see Table 1).

TGF-B

Transforming growth factor-beta (TGF-b) has 3 main isoforms and is involved in a myriad of cellularly pathways[57]. Within tendon healing, it is known to be involved in the  initial inflammatory response, collagen synthesis, angiogenesis, and fibrosis/excessive scar formation[58-65]. TGF-β1 is expressed by tenocytes, infiltrating fibroblasts, and inflammatory cells[64, 65] and is thought to be associated with the pathogenesis of excessive scar tissue formation. Interestingly, when TGF-β1 signaling is disrupted either via antibody or miRNA after flexor tendon injury, range motion of the digit improves, however the mechanical strength of the tendon decreases[48, 65, 66]. TGFβ2 and TGFβ3 are thought to be essential for tendon formation and are a potent inducer of the tendon progenitors[22, 67]. When TGFβ signaling is disrupted during chick development, almost all tendons and ligaments are absent[67, 68].  Mechanical cues are important in the initiation of TGFβ and FGF signaling in utero. Both TGFβ/SMAD2/3 and FGF/ERK MAPK signaling pathways are decreased in tendons under immobilization conditions in developing chicks. The application of FGF4 or TGFβ2 ligands prevents SCX downregulation in immobilized developing chick limbs[21].

Exogenous delivery of TGF-β has been long studied as a treatment, both in vivo and in vitro. TGF-β1 has been thought to lead to excessive scar formation. The treatment of tenocytes in vitro with TGF-β1 promotes ECM synthesis (upregulation seen in biglycan, collagen V, collagen XII, Plasminogen activator inhibitor-1, Scleraxis, and Mohaw) and downregulate matrix remodeling MMPs[69], which is suggestive of how it may facilitate adhesion formation. The mechanical strength of injured rabbits’ Achilles tendons that received Bone Marrow-Derived Mesenchymal Stem Cells transfected with TGF-β1 cDNA was significantly increased[70]. Despite evidence that disrupting TGF-β1 reduces extent of scarring, mechanical strength of the tendon and repair site decrease[48, 65, 66], suggesting that complete blockade of TGF-β1 is not optimally therapeutic.

Unlike with TGF-β1, ectopic delivery of TGF-β3 has demonstrated some promising results. TGF-β3 promoted the tenogenic differentiation of stem cells in co-culture[71]. Jiang et al. found that the addition to TGF‑β3 to tenocytes can significantly downregulate the expression of Smad3 and upregulate the expression of Smad7 at the gene and protein levels, which may minimize scarring [72]. Exogenous delivery of TGF-β3 after Achilles tendon injuries in rats has improved the structural and mechanical properties of the tendon[73]. Further evaluation into the specific isoforms’ role in tendon healing is required to evaluate their possible therapeutic application in flexor tendon injuries.

FGF2

FGF2 is a single chain polypeptide belonging to the heparin-binding growth factor family and facilitates numerous mitogenic and angiogenic activities[74, 75]. Within tendon healing, FGF2 has been found to be associated with inflammation, neovascularization/angiogenesis, cellular proliferation, and collagen synthesis[31, 58, 76-79]. Despite FGF2 not being directly investigated in tendon formation, several other factors within the FGF family have been investigated with regards to their effects on tendon development. FGF4 and FGF8 are both expressed on muscle and tendon boundary regions during limb development, suggesting a potential role for FGF signaling pathway in muscle and tendon interactions[80]. Brent et al. demonstrated that FGF signaling may induce the formation of a tendon progenitor population that expressed Scx during somite development[23]. However, Brown et al. reported that FGF-4 did not increase Scx expression in mouse limbs in both early and late developmental stages and that it showed negative effects on Scx and Col1a1 gene expression in vitro[22, 81].

The effects of exogenous FGF delivery after tendon injury is controversial. Ectopic FGF2 has been shown to increase cell proliferation and promote neovascularization within tendon repairs, however, improvements in mechanical strength remain equivocal[78, 82]. Tang et al. demonstrated improvements in tensile strength in injured chick flexor tendons delivered with FGF2[83]. However, Thomopoulos et al. did not find improvements in mechanical or functional properties in exogenous delivery of FGF2 via a Fibrin-Heparin-Based Delivery System to dog flexor tendon injuries [77].

VEGF

The VEGF family consists of several isoforms that bind to three tyrosine kinase receptors, but their bioavailability for each receptor depends on the isoform dependent[84]. VEGF levels are elevated during tendon development. The VEGF present in human fetal tendons is thought to be responsible for the differentiation of vascular and avascular zones within tendons[25].  VEGF levels then decrease to low concentrations within healthy (homeostatic) adult tendons[85]. The presence of minimally elevated VEGF in adults is suggestive of a chronic overuse tendon injury[86]. Within tendon healing, it has been well established that VEGF is upregulated very early in the healing process and is involved in angiogenesis[35, 87, 88]. VEGF promotes neovascularization via the stimulation of MMPs to possibly degrade connective tissues to facilitate angiogenesis[85].

Ectopic VEGF delivery improves tensile strength of injured Achilles tendons[89]. However, it has also been found by Wang et al. that VEGF does not significantly upregulate collagen gene expression[90]. Therefore, it may not necessarily be the most important factor in collagen synthesis in intrasynovial tendon healing, however it clearly plays an important role in angiogenesis in tendon healing as well as in the formation of tendon.

PDGF

PDGF is a 30 kDa dimer, and its family comprises of four different polypeptide chains[91]. PDGF plays a role in the migration and proliferation of the tenocytes, fibroblasts, and mesenchymal stem cells responsible for tissue homeostasis[92]. PDGF expression is upregulated shortly after tendon injury and helps to stimulate the production of other growth factors[74].  PDGF signaling may be essential to tendon homeostasis. Sugg et al. demonstrated that inhibition of PDGF signaling prevented the normal growth response in tendon tissue to a mechanical stimulus in adult mice [93].  Little is known regarding its role in tendon development.  Exogenous delivery of PDGF improves both morphological and biomechanical properties in numerous animal models, suggesting the PDGF may help augment tendon healing[94-98].

There appear to be common growth factors and gene expression patterns between tendon development and repair. Further investigation is required to better understand the roles that growth factors, cytokines, chemokines, and/or other signaling molecules play in both tendon development and healing. Despite a wealth of knowledge regarding what factors play a role in tendon healing, a better understanding of how tendons develop would likely provide additional insights towards improving tendon repair after injury (Table 1).

Conclusion

Further exploring the similarity and differences in gene expression between tendon morphogenesis and repair may elucidate novel strategies to improve perioperative and post-operative flexor tendon injury management. Additionally, understanding the molecular mechanisms dependent on mechanical loading involved in flexor tendon healing without adhesion formation is also critical to learn how to best improve repair outcomes.  Fundamental as well as translational studies will help us decipher which growth factors, cytokines, chemokines, and/or other signaling molecules are most crucial in the prevention of adhesion formations.

Table 1: Growth factors, cytokines, chemokines, and/or other signaling molecules involved in tendon development and healing

TIMP = tissue inhibitor of metalloproteinase; EGF =  epidermal growth factor; TNFa = Tissue Necrosis Factor Alpha; BMP-2 = Bone morphogenetic protein 2