Boosting tendon repair: interplay of cells, growth factors and scaffold-free and gel-based carriers

Abstract

Background

Tendons are dense connective tissues and critical components for the integrity and function of the musculoskeletal system. Tendons connect bone to muscle and transmit forces on which locomotion entirely depends. Due to trauma, overuse and age-related degeneration, many people suffer from acute or chronic tendon injuries. Owing to their hypovascularity and hypocellularity, tendinopathies remain a substantial challenge for both clinicians and researchers. Surgical treatment includes suture or transplantation of autograft, allograft or xenograft, and these serve as the most common technique for rescuing tendon injuries. However, the therapeutic efficacies are limited by drawbacks including inevitable donor site morbidity, poor graft integration, adhesion formations and high rates of recurrent tearing. This review summarizes the literature of the past 10 y concerning scaffold-free and gel-based approaches for treating tendon injuries, with emphasis on specific advantages of such modes of application, as well as the obtained results regarding in vitro and in vivo tenogenesis.

Results

The search was focused on publications released after 2006 and 83 articles have been analysed. The main results are summarizing and discussing the clear advantages of scaffold-free and hydrogels carriers that can be functionalized with cells alone or in combination with growth factors.

Conclusion

The improved understanding of tissue resident adult stem cells has made a significant progress in recent years as well as strategies to steer their fate toward tendon lineage, with the help of growth factors, have been identified. The field of tendon tissue engineering is exploring diverse models spanning from hard scaffolds to gel-based and scaffold-free approaches seeking easier cell delivery and integration in the site of injury. Still, the field needs to consider a multifactorial approach that is based on the combination and fine-tuning of chemical and biomechanical stimuli. Taken together, tendon tissue engineering has now excellent foundations and enters the period of precision and translation to models with clinical relevance on which better treatment options of tendon injuries can be shaped up.

Introduction

Tendons, which connect bone to muscle, are a crucial part of the locomotion system (Aslan et al. 2008). Tendons transmit forces from muscle to bone (Nourissat et al. 2015) and are able to withstand high tension and store elastic energy which in turn makes movements more ergonomic (Docheva et al. 2015). Tendons are very hierarchically organized tissues containing tendon-specific fibroblastic cells named according to their maturation state as tendon stem/progenitor cells (TSPCs), tenoblasts or tenocytes, the latter being the most terminally differentiated cells (Docheva et al. 2015). The cells are embedded in a three-dimensional network of extracellular matrix (ECM) consisting predominantly of type I collagen, other collagens (such type III and V), proteoglycans, elastin and fibronectin (Andres and Murrell 2008; Sayegh et al. 2015). The whole tendon unit and the tendon sub-units are wrapped with epitenon and endotenon loose connective sheets, respectively. It has been suggested that these sheets contain stem/progenitor-like cells (Docheva et al. 2015; Wu et al. 2017).

Tendon injuries are very common in trauma and orthopaedic surgery. Tendon injuries affect great range of patients from young to elderly patients, from workers to professional athletes (Docheva et al. 2015). Tendon healing follows a typical wound-healing course: inflammatory phase, proliferative phase, and then a remodelling phase (Voleti et al. 2012). The first short phase is characterized by the infiltration of inflammatory cells like platelets, monocytes, macrophages and neutrophils which release chemotactic agents activating and attracting tendon cells from the injured ends and tendon sheets. During the proliferative phase, the tendon fibroblasts start to proliferate and create abundant ECM. In the remodelling phase, the collagen fibres become parallel to the muscle force direction which is critical for the gain in tendon biomechanical strength (Evans 2012). In general, tendons have limited ability to repair as the initially formed scar tissue has inferior biomechanical properties compared to the original tendon tissue and if improperly replaced or remodelled it can be the foundation of increased rates of re-occurring ruptures.

Due to the highlighted above functional limitations and high rates of re-injury of once rupture tendon, various surgical techniques including surgical suture, tendon autograft and allograft transfer for repairing tendon injuries have been already been vastly described (Oryan et al. 2014). However, the long-term clinical outcomes of surgical treatments are still not satisfactory. Thus, new and novel techniques need to be developed. The growing interest in non-operative and conservative treatment options, even for total tendon ruptures calls for new ways to initiate endogenous tissue repair by regenerative mechanisms. Tissue engineering is a promising alternative treatment for achieving complete recovery of ruptured tendons and in general, is based on the combination of reparative cells, growth factors and carriers (Fig. 1).

Fig. 1

Summary of the key components of tendon tissue engineering. a Favourable cell types; b Pro-tenogenic growth factors; c Possible scaffolds and self-assembled materials. Various combinations between the components are possible

The aim of this review is to revise the recent literature about tendon tissue engineering with particular focus on two types of carriers, hydrogels and scaffold-free approaches. Since most reviews have focused on two of the three components of the tissue engineering approach, namely cell types and growth factors, the carriers (third component) have not often been the focus. Actually, the two types of carriers selected for evaluation in this review are particularly innovative and constitute a research field of increasing importance which can contribute to enhanced tendon tissue engineering.

The review constitutes of the following parts: (i) clinical relevance of tendon injury is provided; (ii) an overview of the areas involved in tendon tissue engineering (cells, growth factors, carriers) is given; (iii) the current status on cell types and growth factors is briefly summarized; and (iv) a detailed information on carriers with particular focus on hydrogels and scaffold-free options is delivered and discussed.

A computerized search of potentially eligible studies was performed in PubMed and the date of the last search was October 30, 2017. Database search and key inclusion criteria followed the terms “tendon” AND “tendon injury” AND “tendon repair” AND “tendon tissue engineering” AND “mesenchymal stem cells” AND “growth factors” AND “scaffold-free” OR “gel-based” OR “cell sheet”. The search was focused to publications (both forms none- and open-access) released after 2006. Articles available only as abstract or not in English language or not fitting the review scope were excluded. A summary of the search strategy and article selection in this review is shown in Fig. 2.

Fig. 2

Flow chart of the search strategy and article selection in this review

Review

Tendon clinical relevance

Tendon injuries are usually induced by intrinsic (age, gender, weight, metabolic diseases) and extrinsic (sport injury, overload and occupation) factors. To date, treating tendon injuries costs healthcare providers in USA 30 billion dollars, and in Europe over 115 billion euros per year (Abbah et al. 2014). It is reported that 30–50% of sport injuries include tendon component (September et al. 2007). Epidemiologic reports indicate that annually over 4.4 million patient clinical visits in USA are due to shoulder maladies associated with tendon disorders (Nixon et al. 2012). Chronic and acute injuries can occur in any tendon, although, the most frequently affected tendons are Achilles and rotator cuff tendons. Raikin et al. 2013 reviewed that regarding Achilles tendon injuries 76% of the ruptures are acute with average patient age of 46.4 years, whereas 24% are chronic. Rotator cuff tears affect up to 50% of patients older than 50 years and are a common cause of function-limiting pain and weakness of the shoulder (Isaac et al. 2012). Tendon repair after injury is extremely poor and inefficient because of the low cellularity, vascularity and metabolic activity of the tendon tissue (Liu et al. 2011; Thangarajah et al. 2015). Moreover, in most patients, especially in aged individuals, the healed tendon usually does not regain the mechanical properties of the uninjured tissue. As a consequence, the tendon thickens and stiffens to overcome the lower unit mechanical strength and, hence, the tendon quality and its functional activity are inferior to that of healthy tendon (Docheva et al. 2015). For the above reasons identifying and designing strategies to augment tendon healing are of very high relevance. Over the years, the tendon scientific community has particularly focused on establishing and improving various tendon tissue engineering models which become very promising for achieving a breakthrough in tendon injury management.

As previously mentioned, the most common strategy for restoring ruptured tendons is surgical repair which at times does not result in complete and satisfactory tendon healing. Especially in elderly individuals, surgical repair shows poor long-term outcomes due to re-rupture, restrictive adhesions and suboptimal strength and functionality linked to decreased biomechanical properties (Myer and Fowler 2016). Considerable side effects and complications are seen with operative treatment. In Achilles tendon repair, the majority of patients occur in men aged 30–49 and are reported due to wound and tendon healing problems, like re-rupture, tissue damage and necrosis and subsequently wound infection (Raikin et al. 2013; Thomopoulos et al. 2015). It has been known that several factors, including advanced patient age, large tear size, severe muscle atrophy and fatty infiltration, systemic metabolic diseases and smoking are associated with failed or poor tendon healing (Montgomery et al. 2012). Since traditional surgical options present limitations and complications, non-operative treatment has gained interest in recent years. The challenge of non-operative treatment is the diastasis of the ruptured tendon ends. Any gap and tendon defect may delay the healing process and may result in an insufficient tendinous tissue. Therefore, innovative treatment options such as tissue engineering of ruptured tendons has drawn great interest and driven multi-centre experimental and pre-clinical research to solve this issue.

Tendon tissue engineering

Tissue engineering aims to induce and support tissue self-repair or to produce a functional tissue replacement in vitro that is subsequently implanted in vivo at the site of injury (Youngstrom and Barrett 2016). Hence, tissue engineering may play a major role in improving the management of tendon injuries through grafting engineered tendon segment at the site of rupture (Hsieh et al. 2016b; Yin et al. 2016). We have previously reviewed in great detail (Docheva et al. 2015; Wu et al. 2017) on the two main components of tendon tissue engineering namely cell types and growth factors as well as given summary on the recent advancment of natural hard scaffolds. The current review focuses on the “third component”, the carrier and in particular on hydrogels and scaffold-free approaches. A hydrogel is a network of natural or synthetic polymer chains with high content of liquid. Due to their significant water content, hydrogels are softer and easier to mould and therefore suitable as filling materials. On the other hand, scaffold-free approaches allow the cells to form natural cell to cell and cell to matrix connections without artificial interference and with firing of specific molecular signalling events that can trigger lineage mantainance or even facilitate further maturation.

Cell types

In tendon tissue engineering, mainly stem or progenitor cells of mesenchymal origin are being implemented to create tendon grafts and support graft incorporation (Andres and Murrell 2008). Several cell types have been more favourable and these belong to the mesenchymal stem cells family (MSCs) found in different tissue sources. MSCs extracted from adipose tissues or the bone marrow or tendon-derived cells, including local MSC-related but distinct TSPCs, have been suggested as the most suitable cell types (Yin et al. 2016). These cell types have clear advantages like differentiation potential as well as paracrine effects, which have been reported to play a crucial role in their beneficial properties, by promoting angiogenesis, stimulating local progenitor and mature cells, or regulating inflammation and immune cell functions (da Silva et al. 2009).

MSCs derived from adipose tissue (ADSCs) are an attractive candidate cell type due to their easy isolation, multi-potentiality and high responsiveness to distinct environment cues (Zarychta-Wisniewska et al. 2017). Adipose tissue is abundant in human and animals and subcutaneous adipose aspirate can be easily harvested by a minimally invasive procedure (Deng et al. 2014). However, the main disadvantage of ADSCs is their preference towards adipogenesis in vivo (Neo et al. 2016).

Bone marrow mesenchymal stem cells (BMSCs) due to being best characterized are the most widely used stem cell type. One recent study showed that BMSCs are more responsive to bone morphogenetic protein-12 (BMP-12) stimulation and hence exhibited superior tenogenic differentiation capacity when compared to ADSCs (Dai et al. 2015). However, BMSCs also have some limitations, such as painful harvesting procedure with frequently low cell yield, reduced MSC quality with advanced donor age (Zhao et al. 2009), ectopic ossification and higher risk of adhesion formation when transplanted in vivo (Hsieh et al. 2016a).

TSPCs are a cell type that moved in the research spotlight due to their inherent pro-tenogenic abilities. They were first reported and described in 2007(Bi et al. 2007) and subsequently identified in different tendons, isolated from different species and characterized to some extent (Kohler et al. 2013). One clear advantage of TSPCs is their greater potential for tenogenesis. TSPCs express higher mRNA levels of tendon-related gene markers including the transcription factor Scleraxis (Scx) and the late differentiation factor Tenomodulin (Tnmd) than BMSCs (Ni et al. 2013). Furthermore, we have recently shown in two consecutive studies that native or genetically induced TSPCs transplanted in clinically relevant Achilles tendon defect model in Rattus norvegicus are superior to BMSCs as TSPCs grafting resulted in advanced, significantly less ossified and more mature ECM of the tendon at the remodelling phase of the healing process (Yin et al. 2013). However, TSPCs hold one main disadvantage namely their isolation that is associated with many limitations and co-morbidity. One strategy to overcome this difficulty is to use ADSCs or BMSCs that have been pre-differentiated towards the tendon lineage with the help of growth factors, a topic we will discuss in the next chapter and in Table 1.

Table 1 Pro-tenogenic growth factors

Growth factors

Growth factors play an important role in tendon tissue engineering. They are peptide signalling molecules with a dominant biological role in regulating cell proliferation and differentiation (Branford et al. 2014). Growth factors relevant to the tendon healing process and MSC tenogenesis include families such as bone morphogenetic protein (BMP) family, fibroblast growth factor (FGF), transforming growth factor beta (TGF-β), insulin-like growth factor (IGF), vascular endothelial growth factor (VEGF), connective tissue growth factor (CTGF) and platelet-derived growth factor (PDGF). Many studies have shown that MSCs are very sensitive to the above factors and they can influence their stemness and steer the rate of proliferation and extent of terminal differentiation (Barsby et al. 2014; Halper 2014; Han et al. 2017; Jiang et al. 2016; Lui et al. 2016; Park et al. 2010; Tan et al. 2012; Tokunaga et al. 2015; Violini et al. 2009; Zarychta-Wisniewska et al. 2017). However, only some of the growth factors have promising effects on MSC tenogenesis and example studies and their outcomes are described in Table 1.

Growth factor stimulation protocols of MSCs to achieve tenogenesis in vitro are advancing, However, in order to become very efficient and reproducible, further research understanding the exact molecular signalling events orchestrating and controlling the step-wise commitment process is required. In this respect, great knowledge can be obtained from tendon developmental studies based on gene knockout and reporter mouse strains (Dex et al. 2016). Another research area that can lead to accelerated tenogenesis is applying mechanical loading in combination with growth factors (James et al. 2008).

Carriers

Hard scaffolds

Scaffolds are used to deliver cells and drugs into the body (Garg et al. 2012; Turner and Badylak 2013). Classically, scaffolds are made of hard materials displaying distinct architecture, porosity, interconnectivity, large surface and biocompatibility. The current materials used in tendon tissue engineering have provided significant advances in structural integrity and biological compatibility and in many cases the results are superior to those observed in natural healing (Butler et al. 2008; Liu et al. 2008; Sahoo et al. 2010b). Several kinds of scaffolds have been widely used in tendon tissue engineering such as biologically-based or synthetic scaffolds (Youngstrom and Barrett 2016). Biological scaffolds which includes dermis, pericardium, small intestine submucosa, and tendon, are composed of natural collagen fibres that are bioactive and beneficial for cell attachment, viability and proliferation (Woon et al. 2011). The structure of biological scaffolds may better resemble the original tissue and be more feasible for incorporating and supporting cells but they could be more difficult to obtain and might require additional and more complicated surgery in order to be delivered into the site of injury. In addition, to avoid in vivo immunoreaction, biological scaffolds need to be decellularized prior to implantation. Furthermore, another widely used scaffold type is silk, which has been shown to perform well both in vitro and in vivo tendon studies (Chen et al. 2010). Some other natural materials e.g. human umbilical veins and hyaluronic acid-based scaffolds have been also explored and given promising results in tendon tissue engineering (Fan et al. 2014; Hofmann et al. 2008).

Synthetic scaffolds made from polymers, such as polylactic acid (PLA); poly-L-lactid acid (PLLA); polyglycolic acid (PGA); poly-D,L-lactic-co-glycolic acid (PLGA); polyuria (PU) and poly-caprolactone (PCL), aim at mimicking the native tissue properties and are frequently produced by electrospinning technology (Sahoo et al. 2010b; Sahoo et al. 2010c). TSPCs cultivated on electrospun nanofibers showed augmented tenogenic gene expression (Yin et al. 2010) and increased ECM production (Xu et al. 2014). Some scaffolds are comprised of orientated nano/microfibers which results in similar structure to the native tendon tissue, hence enabling the typical spindle morphology of tendon cells and providing structural cues for tenogenesis (Sahoo et al. 2010). However, traditional hard scaffolds have limitations including poor cell seeding and distribution, low cell adhesion especially onto synthetic materials, unsatisfactory cell proliferation and differentiation and in some cases poor biocompatibility, low biodegradability, and non-matching to the tendon biomechanical properties (Liu et al. 2008; Lui et al. 2014; Ricchetti et al. 2012). The main difficulty arises from the structural and biomechanical properties of the materials and the creation of continuity when implanted in vivo. Research has suggested that materials resembling the hierarchical ECM organisation and dimensions, as well as the elastic properties of tendon tissues are preferable (Bagnaninchi et al. 2007). Forthcoming studies should focus on investigating their further optimization and long-term behaviour in clinically relevant models of tendon injury.

Hydrogels

As mentioned previously hydrogels are a network of natural or synthetic polymer chains with significant water content, possessing a higher degree of moulding. They are commonly used in tissue engineering because of their easier handling and good biocompatibility (Yamada et al. 2007). Hydrogels can fill up various defect shapes and can reach deeper into tissue injury site by percutaneous injection with minimal invasion and low side effects which normally occur after more complicated surgery techniques when hard scaffolds are used. Furthermore, hydrogels can incorporate various cells, drugs, and growth factors through simple mixing. Adherent cells can deposit in their vicinities, natural ECM and organise it appropriately and create a niche that responds to chemical and biomechanical stimuli. Hydrogels composed of natural biomaterials such as collagen, fibrin, hyaluronic acid, alginate, and other ECM proteins are most frequently used. In Table 2, we have summarized studies which focused on the application of hydrogels functionalized with cells for tendon tissue engineering. Most of the studies have investigated how such hydrogels affect the biological features of MSCs and tendon-derived cells in vitro (Annabi et al. 2014; Bian et al. 2013). Only few studies have tested the repair potential of hydrogels alone or in combination with cells in tendon injury models in vivo (Liu et al. 2008; Shah and Federoff 2011). Type I collagen and fibrin gels have been extensively studied to create tissue engineering constructs in vitro (Sander et al. 2011). Breidenbach et al. showed that fibrin gels loaded with TSPCs exhibit improved biological, structural, and mechanical characteristics compared with TSPCs-collagen gels in vitro (Breidenbach et al. 2015). Degen et al., Li et al., and Lopiz et al. reported that the application of hyaluronic acid or alginate hydrogels resulted in enhanced biomechanical and histological properties of the tendon repair tissue in vivo compared with control groups (Degen et al. 2016; Li et al. 2016; Lopiz et al. 2017). Farnebo et al., 2014 developed a novel injectable thermosensitive hydrogel derived from tendon ECM and seeded it with ADSCs (Farnebo et al. 2014). It formed a solid gel at body temperature and had good compatibility and support of cell adhesion and proliferation. Chiou et al., 2015 showed that the hydrogels combined with ADSCs augmented the tendon healing process in a rat injury model (Chiou et al. 2015). The above studies indicated that hydrogels hold a great potential for tissue engineering as they can provide a three-dimensional environment and can serve as an easier to handle cell delivery vehicle for surgical implantation. Despite several advantages and some promising experimental outcomes of the gel-based tissue engineering approach, one very critical limitation especially with regards to tendon repair is that the hydrogels cannot provide in a full tear scenario, the desired biomechanical strength and tissue continuity. Hence, their application can be specialized for cell or drug delivery in partial tendon lesions, contained tendon defects or underneath tendon sheets (Garg et al. 2012).

Table 2 Hydrogel-based studies on tendon tissue engineering

Scaffold-free approaches

A scaffold-free approach means that cell form naturally the connections between each other and the matrix, thus tendon tissue engineered construct can be fabricated without the use of any carrier (Ni et al. 2013). With additional growth factors or mechanical stimuli, the cells can enhance the tenogenic differentiation process as well as the production of tendon-specific extracellular matrix (Kim et al. 2011). As human embryos can organize, proliferate and differentiate within itself without association to a scaffold so tendon tissue can be formed by a self-assembly process. Owaki et al. 2014 proved that TSPCs can produce scaffold-free tendon-like micro tissue (Owaki et al. 2014). This could be a powerful new approach for tendon tissue engineering. In chondogenesis, MSC pellet culture models are vastly used in vitro as well as implanted in vivo. MSCs are first condensed by centrifugation to form a pellet that in the presence of TGF-βand additional molecular factors undergo in the course of approximately 3 weeks into chondrogenesis, thus replicating key assets of embryonic cartilage formation (Mueller et al. 2013; Mueller et al. 2010). In tenogenesis, the so called cell sheet technology has been preferred (Markway et al. 2013). Cell sheets can eliminate the need for natural or synthetic carriers, thus avoiding the above listed disadvantages, as well as biocompatibility (Lui et al. 2014). A cartoon model of cell sheet formation is depicted in Fig. 3. Following cell seeding and upon reaching full confluency cells establish strong cell-to-cell contacts and produce large amounts of native ECM in their apical proximity. This allows an easy dissociation from the culture dish as a continuous integral cell layer. The layer can then be rolled up and subjected to static tension for a desired period of time. In Table 3, we have provided examples of studies dealing with cell sheets for tendon tissue engineering and their outcomes. The cell sheet technology can deliver cells to tendon and tendon-bone interface to accelerate tendon healing (Inagaki et al. 2013). One research group reported the use of anterior cruciate ligament (ACL)-derived CD34+ cell sheet that was wrapped around a tendon graft for ACL reconstruction in a rat model and concluded that the cell sheet augmented graft resulted in improved and developed a more mature bone-tendon healing (Mifune et al. 2013). One study transplanted TSPC sheets in a patellar tendon window injury model and reported that 2–6 weeks post-surgery the tendon healing was significantly promoted (Lui et al. 2014). Another study indicated that the use of a graft composed of multipotent stem cell sheets led to satisfactory reconstruction of complete musculotendinous junction rupture (Inagaki et al. 2013). Komatsu et al., 2016 proved that TSPC sheets significantly improved histological properties and collagen content at both 2 and 4 weeks after implantation into a rat Achilles tendon injury model, indicating that such an approach may effectively promote the early stages of tendon healing (Komatsu et al. 2016). One clear advantage of the cell sheet technology is the formation of native cell-to-cell and cell-to-matrix interactions which initiate the appropriate and inherent cell signalling cascades of the used cell types (Hashimoto et al. 2016; Mifune et al. 2013; Neo et al. 2016). Further application of growth factors, media supplements affecting the cell anabolism or mechanical stimuli can further boost the effectiveness of the cell sheet maturation towards enhanced in vitro tenogenesis (Tan et al. 2012; Violini et al. 2009). Follow up research is required to optimize the current protocols. For example as reported in Table 3, at present different protocols are used to form the cell sheets and there is not only one standard technique, thus further warranting investigations to optimize the cell sheet procedure. Another point for improvement is to find strategies to augment the mechanical properties of the cell sheet grafts prior to implantation in vitro which is mainly dependant on produced ECM amount and maturity level, as well as on the graft dimensions.

Fig. 3

Cartoon of the procedure to form tendon-like cell sheet. a MSCs are plated in cell culture dish; b Cell monolayers are formed; c and d The monolayer is scraped out from dish surface and rolled up into a cell sheet; e and f The three-dimensional cell sheet is cultivated under static tension and let to mature prior transplantation in vivo

Table 3 Examples of tendon cell sheet models

Conclusion

From hard scaffolds to gel-based and scaffold-free approaches, tendon tissue engineering has significantly progressed in recent years. The improved understanding of tissue resident adult stem cells, such as BMSCs, ADSCs and TSPCs, has been very helpful and a large number of studies have clarified the advantages and disadvantages of these cell types. Growth factors steering stem cell fate toward the tenogenic lineage have been identified and overall, the protocols for in vitro tenogenesis have been improved. Still, the field needs to consider a multifactorial approach that is based on the combination and fine-tuning of chemical and biomechanical stimuli in order to obtain optimal tenogenesis in vitro and in vivo. The field also has to move out of a ‘one size fits all’ strategy for treating tendon injuries and consider that different tendon defects can be treated by ‘custom design’ combination of cells and carriers and personalised physiotherapy. In particular, carrier-free and gel-based applications, in combination with autologous cells, can be very attractive option to enhance conservative treated tendon injuries as they can be delivered with minimal invasive operation procedure and may lead to quicker and better outcome. All in all, tendon tissue engineering has now excellent foundations and enters the period of precision and translation to models with clinical relevance and we think undoubtedly it will remain the most promising step forward for better treatment of tendon injuries.