Tendon and Ligament Healing and Current Approaches to Tendon and Ligament Regeneration

Abstract

Ligament and tendon injuries are common problems in orthopaedics. There is a need for treatments that can expedite nonoperative healing or improve the efficacy of surgical repair or reconstruction of ligaments and tendons. Successful biologically-based attempts at repair and reconstruction would require a thorough understanding of normal tendon and ligament healing. The inflammatory, proliferative, and remodeling phases, and the cells involved in tendon and ligament healing will be reviewed. Then, current research efforts focusing on biologically-based treatments of ligament and tendon injuries will be summarized, with a focus on stem cells endogenous to tendons and ligaments.

Statement of Clinical Significance:

This paper details mechanisms of ligament and tendon healing, as well as attempts to apply stem cells to ligament and tendon healing. Understanding of these topics could lead to more efficacious therapies to treat ligament and tendon injuries.

Introduction

Ligament and tendon injuries are common diagnoses, accounting for 50% of musculoskeletal injuries¹. Each year, there are approximately 17 million ligamentous injuries that require medical treatment in the United States, with an estimated economic cost of over $40 billion². In addition to the economic cost, these injuries also have significant impact on patients’ quality of life and ability to meet their occupational, recreational, and health goals. In many cases, injuries to tendon and ligaments such as common sprains and strains heal without surgical intervention. However, the process is often slow and results in the formation of inferior scar tissue, which can take years to remodel into more functional tissue. More serious injuries such as complete tear of the tendon or ligament often require surgical treatment. These surgical treatments have variable degrees of success. In cases of intra-articular ligaments, where there is limited capacity of healing, autograft tissue is often used to reconstruct the torn ligament. While these reconstruction techniques have been very successful, they are often associated with the problem of donor site morbidity^3–6. In case autograft tissue is not available, such as multi-ligamentous injuries or revision knee ligament surgeries, allografts can be used but are often inferior⁷. In cases of extra-articular tendon and ligaments that have the capacity to heal, the surgeon will reconnect or re-attach the torn tendon and ligament. These procedures usually result in the formation of inferior scar tissue which is more prone to re-rupture. For instance, 63% of large and 73% of massive rotator cuff tears fail to heal following surgical repair⁸. Despite the impact of these injuries, there is limited research on the fundamental biology that governs tendon and ligament development and healing. Thus far, the lack of biologic understanding on the mechanisms behind tendon and ligament healing has hindered the development of effective therapies for treating tendon and ligament injuries. In this article, we will discuss the current understanding of tendon and ligament healing and highlight current research efforts towards biologically-based treatments of these injuries. In particular, we will focus on the presence of stem cells found within tendons and ligaments, and the role that they may play in repair and regeneration.

Tendon and Ligament Structure and Healing

Tendon and ligaments are fibrous connective tissue that connects muscle to bone, and bone to bone, respectively. Tendons primarily function by transmitting the contraction force produced by muscle to bone, thereby enabling movement. Ligaments function to passively stabilize joints. Tendons are composed of 55–70% water, and the extracellular matrix (ECM) is primarily of aligned type I collagen fibers (65–80% dry weight) with minor extracellular matrix components such as elastin, decorin, biglycan, and fibromodulin⁹. Populations of tendon stem/progrenitor cells (TSPC) with regenerative capacity that reside in a unique niche in tendons and paratenon have been described^10,11. TSPCs have been isolated from Achilles tendon^12–14 and patellar tendon^15,16, and are thought to contribute to tendon development and regeneration both directly and through secretion of trophic factors.

Collagen in tendons are organized in hierarchical levels, beginning with tropocollagen, a triple-helix peptide chain, which coalesces into fibrils, fibers (primary bundles), fascicles (secondary bundles), tertiary bundles, and the whole tendon¹⁷. The endotenon is a reticular network of connective tissue that surrounds each fiber. The epitenon is the connective tissue sheath containing the tendon’s vascular, lymphatic, and nerve supply that surrounds the whole tendon, while the paratenon is a loose areolar connective tissue consisting of type I and III collagen fibrils, some elastic fibrils, and an inner lining of synovial cells¹⁷.

The difference in the ECM composition and organization varies with each individual tendon and ligament according to anatomic location, presumably due to variations in the mechanical loading environment. Structurally and physiologically, tendon and ligament are very similar^18,19. Additionally, the delineation between the two is not always clear. For instance, the patellar tendon can be considered a tendon if it is considered a continuation of the quadriceps tendon with the patella being as sesamoid bone (muscle to bone). But it has also been referred to as the patellar ligament since it connects the patella to the tibial tuberosity (bone to bone). Thus, for the remainder of this review, references to tendons or ligaments can be generalized to apply to both.

Acute injury to a tendon is followed by a rapid initiation of the healing process. This process is generally subdivided into three chronological stages: inflammation, proliferation, and remodeling^20–23. While these stages overlap, they are characterized by distinct cytokine profiles and cellular processes. Cytokines expressed in tendon healing vary temporally and functionally, with generally pro-inflammatory cytokines predominating early, and anti-inflammatory and restorative cytokines predominating late in the healing process²⁴.

The inflammatory stage of tendon healing begins immediately after acute injury with clot formation in damaged tissue. In this phase, clot is formed in damaged vessels, inflammatory cells are activated, and finally fibroblasts are recruited to continue the healing process²⁰. Platelets and cells within the clot release TGFβ, IGF-I, and PDGF, causing local inflammation^21,25. The clot serves as the initial scaffold for the recruited extrinsic inflammatory cells²⁶. Elaboration of these growth factors recruit neutrophils, which in turn activate macrophages to phagocytose necrotic debris^22,26. Approximately two days following injury, these cytokines released from macrophages and intrinsic cells of the endotenon and epitenon initiate the proliferative stage by recruitment of fibroblasts²¹. TGFβ is responsible for regulating proteinase activity, stimulating collagen production, and later, recruiting of fibroblasts^21,22,27. Similarly, IGF-I functions to stimulate extracellular matrix production and recruit fibroblasts to the area^28,29and, PDGF enhances DNA and protein synthesis, and therefore the expression of other growth factors²¹. These factors work synergistically to initiate the healing process.

The proliferative phase is characterized by expansion of the extracellular matrix, increased cellularity, and deposition of fibrovascular scar by fibroblasts^20,26. At the site of injury, fibroblasts migrate in and proliferate²⁰. Intrinsic cells of the endotenon and epitenon begin proliferating as well²⁰. IGF-I and TGFβ expression remains high, continuing to attract fibroblasts to the site and increase extracellular matrix production^28–30. At this time, bFGF expression from tenocytes, fibroblasts and inflammatory cells peaks, promoting angiogenesis and cellular proliferation^21,22,31,32. VEGF expression is also high, stimulating angiogenesis to provide extrinsic cells, nutrients, and additional growth factors to the area of injury^21,33. Collagen synthesis is a highly oxygen dependent process, underlying the importance of synergistic angiogenic actions of bFGF and VEGF in this stage of healing^20,34.

About two weeks following injury, remodeling of the injured area begins with reorganization of the newly deposited collagen. This process overlaps with the proliferative phase, leading to a gradual decrease in cellularity and increase in fibrous matrix²¹. Tenocytes and collagen fibers become aligned in the direction of stress, with a corresponding decrease in type III collagen, vascularity, cellularity, and water content in the forming scar²⁶. Increased action of collagenases aid in the resorption of type III collagen and replacement with type I collagen, which has more crosslinks and tensile strength^20,35. This process continues for months and years following the injury; however, the newly formed tissue lacks the native biomechanical, biochemical, and ultrastructural properties of the tendon^23,36. While most vascularized tendons and ligaments have at least some capacity to heal and the ability to form scar that gets remodeled over time, avascular tendon such as the rotator cuff and intra-articular ligaments generally do not have healing capacity.

There are multiple sub-populations of cells that migrate and contribute to tendon healing. Like most healing processes in the body, the initial pro-inflammatory phase is highlighted by the presence of M1 macrophages. As the inflammation subsides, there is a transition from M1 to M2 macrophages at the injury site. During this phase, fibroblasts or connective tissue progenitors migrate in and proliferate. The origins of these fibroblasts or fibroblast progenitor cells are still a subject of debate. A study conducted by Jones et al in 2003 labeled epitenon cells with fluorescent lipophilic membrane vital dye 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine percholate (DiI) and followed these cells in a rat flexor tendon injury. They found that these labeled epitenon cells proliferated and migrated to the lesions and synthesized new matrix³⁷. Similarly, other studies on tendon and ligament healing have also observed migration of cells from the epitenon into the injury site^12,16,38. Recent studies using murine genetic lineage tracing have identified several distinct populations of resident tendon stem/progenitor cells that are involved in tendon healing³⁹. These include cells from tendon fascicles with markers Sca-1, CD90, CD44, Scx, and Tnmd¹⁰, cells from epitendon with markers Sca-1, Laminin, αSMA and PDGFα ^40,41, and perivascular cells from epitendon with markers αSMA, Nes, and CD133^12,42,43. These tendon progenitor cells have been isolated and characterized by several different groups^14,44. They have shown the ability to self-renew and differentiate into various mesenchymal lineages in vitro. Additionally, studies have shown that they possess superior tenogenic capacity in vitro and are more adept at healing tendon injuries in vivo as comparted to other MSCs^13,42. Bhavita and Huang recently published a very thorough review on the current understanding of tendon stem/progenitor cells³⁹.

Therapeutic approaches to augment healing in tendon and ligament injuries

Biophysical Stimulation

Biophysical stimulation describes a series of conservative treatment modalities that can be used to assist in tendon and ligament healing, but is typically only applied in well-vascularized tissues that have a high likelihood of healing without surgery, or as an adjunct to surgery⁴⁵. These methods include physical therapy²⁶, cryotherapy²⁶, magnetic fields^46,47 and ultrasound⁴⁸.

Growth Factors

Many investigators have tried to enhance tendon healing by delivering various growth factors that are known to play an important role during tendon and ligament healing and development. These factors include but are not limited to: bFGF, GDF5, GDF6 (BMP13), GDF7 (BMP12), IGF1, PDGF, TGF-b1, TGF-b2, VEGF, and combinations of these growth factors⁴⁹. To deliver these growth factors, investigators have used simple local injections of recombinant protein, over-expression vectors and biomaterial carriers to achieve more sustained growth factor release. Overall, these growth factor treatments seem to provide some benefit to the healing tendon especially during early phases of tendon and ligament healing. The long-term benefit seems to be limited, resulting in varying degrees of success⁴⁹.

Biologics

In the last decade, there has been increasing interest in the utilization of biologically active adjuncts in the treatment of sports injuries. The most popular of these therapies is platelet-rich plasma (PRP), an autologous blood product that is obtained by drawing peripheral venous blood followed by centrifugation to remove red blood cells, leaving concentrated platelets and plasma⁵⁰. PRP is rich in various growth factors that include but are not limited to: PDGF, VEGF, TGF-β, EGF, FGF and IGF. Over the last 20 years, there have been numerous basic science and in vivo animal studies support the use of PRP in various orthopaedic conditions. Clinically, there also have been many isolated studies that have shown beneficial effects PRP. While the evidence regarding PRP continues to evolve, there has not been any definitive, unbiased, high-level clinical studies that have provided definitive proof of the efficacy of PRP treatment in ligament and tendon healing. There are multiple factors that have contributed to this issue. As a biologic that is directly collected from the patient there will always be significant patient to patient variability in the growth factors that are present in PRP. To compound that variability, different vendors tend to have slightly difference in how PRP is isolated. Furthermore, there is no standard protocol for the delivery of PRP in terms of composition, quantity, and frequency. Finally, there are also different PRP formulations used by physicians such as leukocyte-rich versus leukocyte-poor PRP. In general, leukocyte-poor PRP is used for osteoarthritic treatment, while leukocyte-rich PRP is used for tendonitis and similar conditions. Currently, PRP has been used for various tendon and ligament injuries, such as rotator cuff tendinopathy, patellar tendinopathy, Achilles tendinopathy, lateral epicondylitis, partial tears in ulnar collateral ligament, and partial tears of ACLs. Walters et al. recently conducted a randomized control trial on the use of PRP to augment patellar tendon autograft ACL reconstruction with minimal success. While the lack of high-level clinical evidence remains a concern, PRP remains a viable conservative treatment with a low risk of complications or adverse reactions⁵¹. The primary drawback for clinical use is the out-of-pocket costs to patients, as insurance seldom covers the cost of PRP treatment.

Stem Cells

Mesenchymal stem cell based in vivo restoration of tissue have been proposed and used since the early 1990s. Similarly, MSC-based approaches to enhance tendon healing have been investigated since 1993⁵². The main advantages of MSCs over differentiated fibroblasts are that they have the ability of self-renew, the capacity to differentiate into various mesenchymal linages, and the ability to secrete trophic factors that modulate local inflammatory response. The ability to self-renew allows for in vitro expansion to obtain adequate number of cells for in vivo intervention. The ability to differentiate into tenocytes, chondrocytes, osteoblast and myoblasts would theoretically allow for the regeneration of tendon-to-bone and tendon-to-muscle interfaces. More recently, studies have shown that MSCs rarely make direct contributions to tissue regeneration, but indirectly stimulate tissue repair by secreting trophic factors which activate residual recipient cells and/or modulate local immune response⁵³.

From our literature search, there have been over 100 studies using MSCs for tendon healing, and the majority of these studies has been published in the last 5 years. These studies have used the traditional bone marrow derived stromal cells (BMSCs), adipose derived stem cells (ASCs), endogenous ligament derived stem cells (LDSCs) or tendon derived stem cells (TDSCs), and MSCs from other sources, such as synovial fluid. MSC-based therapies have been applied to augment tendon and ligament healing in several different ways.

The most straightforward approach is the injection of MSC suspension into the injury site or with a carrier, which is usually collagen of fibrin gel. Then there are more tissue-engineered approaches in which MSCs are seeded onto a scaffold prior to implantation or MSCs are cultured on a scaffold in vitro to produce neo-tendon prior to transplantation. The benefit of the tissue-engineered approach is the initial mechanical support of the injury. To further augment tendon and ligament healing, studies have combined cells and scaffold with growth factors and/or as PRP mentioned above. A detailed list of these studies can be found in supplementary Table S1. As these studies show, there are many parameters one has to consider to optimize a given stem cell protocol for tendon and ligament healing. These include the choice of cells, quantity of cell delivered, type of growth factor, drug delivery mechanism, and type of biomaterial scaffold. All of these factors can greatly affect the outcome of any cell based therapy.

Authors’ Perspective

Given that the detailed biological processes governing tendon and ligament formation and maintenance remain largely unknown^26,54,55, it is difficult to devise treatment modalities that can model the biological processes by which tendon and ligament develop and heal. While BMSCs and ASCs have been shown to have some beneficial effects on tendon healing, the results have been inconsistent. Given recent identification and isolation of TDSCs, the most obvious biomimetic choice of stem cells for tendon healing is the endogenous tendon stem cells. They have been shown to have superior tenogenic capacity in vitro and superior tendon healing as compared to other MSCs. However, the main drawback of TDSCs is their limited supply and the need for extensive in vitro expansion.

While intrinsic progenitor cells are known to contribute to tendon healing, we and others have also hypothesized that progenitor cells from the surrounding tissue and/or vasculature also play a significant role in tendon healing. This hypothesis is partially based on the empirical observation that extra-articular tendon and ligaments with access to blood supply such as the medial collateral ligament (MCL) heal, while intra-articular ligaments which have limited access to blood supply such as the anterior cruciate ligament (ACL) do not. The importance of vascular supply to tendon healing is further highlighted by the finding that rotator cuff repairs are more likely to heal when exposed to blood supply through microfracture of the bone at its insertion on the humeral head^56,57. There have been many studies showing that a subset of TDSCs are of perivascular origin (CD146+) and can directly contribute to tendon healing in vivo^{10–12,16,38,58}. The identity and origin of these perivascular cells remain unclear as they can come from a number of different places: epitendon^12,42,43, surrounding tissue, or circulating cells. While it is known that stem cells of perivascular origin are heterogeneous in terms of phenotype, distribution, and origin⁵⁹, they do share many similarities among different organs. Thus, we have started to ask the question: do perivascular stem cells isolated from other tissues such as fat have similar tenogenic potential as that of native tendon perivascular-associated stem cells?

Recent research has identified mesenchymal stem cells (MSCs) in various organs residing in a perivascular location and has named these cells perivascular stem cells (PSCs). PSCs are comprised of two distinct cell populations, pericytes (CD31-, CD34-, CD45-, CD146+) and adventitial perivascular cells (CD31-, CD34+, CD45-, CD146-). PSCs have demonstrated the ability to self-renew and differentiate into various mesenchymal lineages in vitro⁶⁰. We are currently using a murine genetic lineage tracing method to identify the contribution of PSCs in tendon healing. Additionally, we are also conducting studies comparing the characteristics and tenogenic potential of PSCs derived from adipose tissue versus TDSCs with similar cell surface markers. If these two cell types function similarly, then PSCs isolated from fat would present an ideal cell source for tendon regeneration. Autologous PSCs obtained via liposuction are the only MSC type that can be obtained in sufficient therapeutic quantities without in vitro expansion. Multiple laboratories are currently developing PSC-based therapies for tissue regeneration of tissues including bone, tendon, and heart^61–64. Additionally, we have started in vivo studies on examining the suitability of adipose derived PSCs as a mechanistically-based therapy for ligament and tendon repair.

Summary

Treatment of tendon and ligament injuries is a significant clinical challenge. Over the last several decades, significant progress has been made in understanding of biological processes that govern tendon and ligament injury and healing. Despite this progress, the only accepted treatments remain surgery and physical therapy. In this review, we have highlighted many new strategies for tendon and ligament healing, including biophysical stimulation, bioactive agents, stem cell therapy and tissue engineering approaches. While these advances are encouraging, there remains a significant gap in knowledge that makes it difficult to design treatment strategies that model the biological processes by which tendons and ligaments develop and heal. Thus, more research on the mechanisms that govern ligament and tendon healing is needed.