Ligament and tendon injuries are common clinical problems. Healing of these tissues occurs, but their properties do not return to normal. This predisposes to recurrent injuries, instability and arthritis, loss of motion and weakness. Gene therapy offers a novel approach to the repair of ligaments and tendons. Introduction of genes into ligaments and tendons using vectors has been successful. Marker genes and therapeutic genes have been introduced into both tissues with evidence of corresponding functional alterations. In addition, gene transfer has been used to manipulate the healing environment, opening the possibility of gene transfer to investigate ligament and tendon development and homeostasis, in addition to using this technology therapeutically. Several factors modulate the ‘success’ of gene transfer in these tissues.
Gene transfer has been accomplished in ligaments and tendons with marker and therapeutic constructs, but the number of reports indicating successful transfer of therapeutic genes is limited. Genetic investigation to determine the roles of certain molecules in tendon healing has also been reported. Therefore, while gene intervention (therapy or investigation) is possible, the current state of the art is in its infancy. Work on several fronts is ongoing to address the many challenges faced due to the properties of the tissues and the accessibility of cells in these tissues. The work can be grouped into efforts common to gene intervention and efforts unique to musculoskeletal tissues such as ligaments and tendons.
In this review, we will begin by defining the clinical problems that are being, or may be, addressed with gene intervention. These clinical problems will be put in the context of normal ligament and tendon properties. Gene therapy and investigation will then be considered for ligaments and tendons separately. We will end with a view to future directions regarding gene intervention for ligaments and tendon, highlighting issues unique to these tissues.
The major clinical problems addressed with gene intervention have focused on improving the healing of complete disruption of ligaments and tendons following injury. In both cases, the injury heals with formation of scar tissue, a material inferior to the normal structure. An additional problem for healing of some tendon injuries is the formation of adhesions. Adhesions connect the healing tendon to surrounding tissues, interfering with the gliding motion of tendons that is required for joint motion, and thus, compromise function. While not the focus of gene intervention studies to date, ‘adhesions’ or loss of normal gliding between joint capsular ligaments interferes with extremity function by limiting joint motion as well.
For both tissues, the events of healing are qualitatively similar. There are three major phases to healing; an inflammatory stage, a proliferative stage and a remodeling stage.1,2 The inflammatory stage occurs over the first several days following the injury. It begins with the formation of a fibrin clot, which is then invaded by polymorphonuclear leukocytes and other inflammatory cells. These cells remove debris and damage. They also release growth factors promoting angiogenesis and proliferation of fibroblasts. The second, or proliferative phase, takes place over the first couple of weeks to months following the injury. Blood vessels develop and fibroblasts proliferate to produce matrix that fills the gap in the ligament or tendon caused by the injury. The proliferative phase gives way to the remodeling phase. The remodeling phase occurs from months to years after the injury and is characterized by decreasing cellularity and realigning of the matrix to respond better to the forces applied to the tissues.
Ligament healing has been characterized biomechanically and biochemically. From a biomechanics perspective, rabbit medial collateral ligament scars fail at measures between 30 and 50% of normal values.2,3,4 Even at subfailure, or low loads, scar has more creep (stretches out more) than normal ligament.5 Biochemically, water content, collagen amounts and types, and other matrix components are altered.2,3,4 The inferior properties of ligament scars lead to joint instability and predisposes to osteoarthritis.
Tendon healing also shows altered properties. Biochemically, water content increases, collagen amounts decrease and collagen crosslinks are altered with increases in reducible crosslinks and decreases in mature nonreducible pyridinoline crosslinks.6 Clinically, inferior properties of tendon scars can predispose to rerupture.7,8,9 For intrasynovial tendons (digits), gap formation affects tendon excursion and the resultant active joint motion. Adhesions are a problem, particularly for flexor tendon injuries of the hand, limiting both active and passive joint motion.
Properties of ligaments and tendons
Normal ligaments and tendons have similar compositions with some minor variations.1 Both tissues are hypocellular with <5% of the total volume occupied by cells. However, the cells do have connections with gap junctions.2,10 Ligament and tendon are relatively hypovascular.9,11 Specialized nerve receptors exist that participate in joint proprioception and recent work has shown the importance of innervation in the healing of ligaments.12,13,14
The major research efforts have focused on the description of ligament and tendon matrix as well as their biomechanical properties.1,4,15 The matrix is predominantly composed of water (65–80%). The solid components of the matrix are primarily collagen, with a preponderance of type I collagen. Other solid components include other collagens (types III, V), proteoglycans including decorin, elastin, fibronectin and other molecules that represent small portions of the solid matrix. Type I collagen is viewed as the major load-bearing component of the solid matrix. The other molecules of the solid matrix have less-defined roles. Some of the molecules such as collagen type V and decorin have been implicated in collagen fibrillogenesis.2,16,17,18 Electron microscopic analysis of ligaments has shown a bimodal distribution of collagen fibril diameters in normal rabbit medial collateral ligaments.19 Histologically, collagen in ligament and tendon displays a crimp pattern, which refers to a regular, wavy pattern of the matrix when viewed in unloaded conditions.2,5,15,20
Biomechanically, ligaments and tendons are viscoelastic materials.2 Viscoelasticity refers to the time-dependant responses of a materials' force and length properties. For example, a ligament or tendon subjected to a constant force will get longer with time, or ligament and tendon subjected to a constant length will see a dissipation of force with time. These lower load (subfailure) properties have higher load (failure) correlates; the magnitude of the failure force of ligament and tendon varies with the rate at which the force is applied. Traditionally, higher load (failure) properties have been reported, but more recently, lower load (sub failure) properties have been extensively characterized.5 Viscoelastic properties in general are a phenomenon of the water/solid matrix composition of ligaments and tendons. The exact role of each component remains to be determined, and the roles may be more evident or important for particular components at different loading conditions (low versus high; subfailure versus failure loads).
While biomechanical properties in and of themselves are important parameters to replicate, biomechanical influences undoubtedly affect molecular and histological end points. Examples include altered gene expression of cultured cells on membranes that can modify strain environments, and altered gene expression of hydraulically loaded ligament tissues.21,22 Realignment of collagen fibrils in ligaments is altered with immobilization versus mobilization of joints in vivo.19 Also equally important is to consider biomechanical influences of the joint in vivo. Studies on blood flow to rabbit knee joints have shown that joint structures not injured with the experimental intervention nevertheless have altered blood flow following ligament injuries.23 The implication is that the ligament injury with subsequent joint instability has altered the biomechanical loading of the other joint structures in association with the blood flow changes.
Gene intervention – ligaments
Studies on gene intervention to ligaments have used rabbit knee models.16,17,24,25,26,27,28 The medial collateral ligament (MCL) and anterior cruciate ligament (ACL) have been evaluated. The MCL heals but has inferior properties compared to normal MCLs, while the ACL does not heal in an appreciable manner to recover biomechanical function.1,2,3 Ligament studies have focused on the feasibility of gene transfer using gene markers and the introduction of therapeutic genes. Issues such as the effect of ligament (MCL versus ACL), method of viral gene delivery, and effect of therapeutic genes on ligament properties have been addressed.
Several authors laid the ground work for gene transfer to ligaments by looking at expression of β-galactosidase with a Lac Z gene transfection (Table 1).24,25,26,27 The initial work began with transfections of ligament cells in vitro. Gerich et al,26 obtained cells from the MCL, ACL and posterior cruciate ligament (PCL) of 3-month-old rabbits. They found the adenovirus – Lac Z construct transfected 100% of cells without cytotoxicity at a multiplicity of infection (MOI) of 10. A BAG retrovirus that expressed Lac Z and neor genes led to nearly complete transduction of ligament cells from all three tissues with the use of the synthetic antibiotic G418 that selected for cells expressing neor.
In vivo experiments also used the Lac Z gene transfer system to show cell transduction through β-galactosidase activity. Hildebrand et al,24 used skeletally mature female rabbits and studied the MCL and ACL. The expression of β-galactosidase was evaluated at 10 days, 3 and 6 weeks after the intervention. Two different viral vectors were used; a direct injection of adenovirus and injection of MCL and ACL cells (from skeletally mature female rabbits) that were transduced with a BAG retrovirus. The effect of injury on β-galactosidase gene expression was evaluated in the MCL only. β-Galactosidase gene expression was detected in both the MCL and ACL (Figure 1).24 Expression was dependent on the vector used. The length of transgene expression lasted at least 6 weeks in the ACL and between 3 and 6 weeks for the MCL, with the adenovirus vector. Expression using the retrovirus (ex vivo, with allograft ligament cells) occurred over a shorter time frame, between 10 days and 3 weeks in both the MCL and ACL. When considering the possible effect of injury of the MCL on expression of β-galactosidase, there was no difference in the time frame of expression. Again, expression was longer with the adenovirus, lasting between 3 and 6 weeks compared to 10 days and 3 weeks with the retrovirus. The encouraging observation was that marker gene expression occurred in the setting of an injury, where a nonspecific inflammatory response is incited due to the injury. This is in distinction to a normal ligament where the response is lacking. The use of viral vectors can elicit an inflammatory response which is associated with a diminution of gene expression.29,30 Using viral vectors in a setting where the inflammatory response is heightened could limit the time of transgene expression; however, this was not the case in the work published by Hildebrand et al.24
Menetrey et al,25 also studied marker gene expression (β-galactosidase) in skeletally mature rabbit ACLs. They looked at the length of transgene expression using direct injection of a replication-deficient adenovirus with the Lac Z gene or the introduction of this vector construct ex vivo via injection of previously transduced rabbit pup primary ACL cells or myoblasts. Similar to Hildebrand et al, Menetrey et al, found that direct injection of the viral vector led to longer gene expression in the ACL. Direct injection of adenovirus vectors led to transgene expression for at least 6 weeks, whereas ex vivo gene transfer led to transgene expression of between 3 and 6 weeks. The mechanisms leading to the longer gene expression with direct injection of the viral vector are unclear, but studies of retrovirally transfected skin fibroblasts showed that the transplanted cells could be detected up to 8.5 months after implantation, while gene expression decreased >1500-fold after 1 month in a rat wound-healing model.31 This implies that, at least in skin wound healing, regulation of the transferred gene was altered in the skin fibroblasts. The authors studying gene transfer with transduced cells in MCLs and ACLs did not comment on whether the transplanted cells were present even though they were not expressing β-galactosidase.24,25
Methods of vector delivery, either viral or liposomal, to ligaments have also been studied.27,28 The methods include direct injection into the ligament, intra-articular injection or intra-arterial injection of the vectors. Injection of viral vectors, or transduced allografted cells, directly into ligaments was used by Hildebrand et al,24 and Menetrey et al.25 With the injection into one area of the ACL, or the ruptured ends of the MCL, most of the β-galactosidase activity was found near the injection sites (Figure 1).24 Menetrey et al,25 injected proximal, middle and distal aspects of the ACL and reported β-galactosidase activity throughout the ACL. Intra-articular injections of myofibroblasts transfected with adenovirus containing Lac Z showed β-galactosidase-positive myoblasts adhering to the ACL in immature rabbits knees.27 Myoblasts went on to fuse into stable myotubes with continued expression of β-galactosidase in the ACL of adult SCID mice for up to 35 days.27 Distribution of β-galactosidase activity was not specified in this study.
The final study on delivery methods of vectors involved a comparison of two different direct injection techniques with an intra-arterial injection.28 These authors used adult rabbit MCL scars as the ligament model. The vector used was a hemagglutinating virus of Japan (HVJ)-conjugated liposome construct that delivered a fluorescent marker. Two weeks after a standard MCL injury, a second surgery was performed. Three groups were compared. Group 1 had a direct, freehand injection of HVJ-liposomes into the MCL scar. Group 2 had a systematic direct injection of the HVJ-liposomes, where 10 μl was dispensed through square holes of mesh grid system placed over the MCL. The mesh allowed injection of the scar every 0.5 mm. Group 3 had an injection of the HVJ-liposomes into the femoral artery, which feeds the MCL. The femoral vein was clamped for 10 min to allow pooling of the HJV – liposomes in the MCL. The rabbits were killed 1 day later and the percentage of cells taking up the fluorescent markers tagged on oligodeoxynucleotides (ODN) were estimated. The direct injection freehand group had ≈10% of the cells showing fluorescence, the systematic injection through the mesh group had approximately 60% of the cells labeled while the intra-articular injection group had only 0.2% of the cells showing fluorescence, mostly around blood vessels. Thus, a systematic injection of the whole scar led to the greatest number of transfected cells.32 It was determined that 25% of cells still showed fluorescence with injection of the vector using the mesh grid system 7 days after injection.32
While the previous studies had shown gene transfer was possible to ligaments, another series of studies have determined that MCL healing can be manipulated with positive functional effects.16,17,28 Using a well-established model of MCL healing in an adult female rabbit knee where a 4 mm gap is created and marked, the scar that forms to fill the gap was injected with HVJ-liposomes containing antisense ODNs to decorin 2 weeks after the first surgery.16,17,28,33 Decorin is a small leucine-rich proteoglycan that is implicated in collagen fibrillogenesis, where increased levels of decorin are associated with smaller diameter collagen fibrils.17 Decorin is a component of normal and healing rabbit MCLs.16,17 MCL healing is characterized by a uniform distribution of smaller collagen fibril diameters in distinction to the bimodal distribution of smaller and larger fibril diameter of normal MCLs.4,19 These alterations in fibril diameters are associated with inferior biomechanical measures of ligament strength in MCL scars.4,19 These investigators tested the hypothesis that downregulating decorin expression would be associated with a return of the fibril diameters to a more normal bimodal distribution and this in turn would improve biomechanical measures of the MCL scar.
The optimal antisense ODN sequence for decorin was determined using skeletally mature rabbit MCL scar cells. A panel of five candidate antisense ODN (∼20-mers), which span the translational start site of the message were assessed in vitro and the most effective candidate was chosen for further in vivo investigation based on decreased decorin mRNA levels in cultures of MCL cells (DA Hart et al, unpublished data). The in vivo experiments compared injection of antisense decorin ODNs, sense decorin ODNs and a third group that was injected with HVJ-liposomes alone. A further group was ‘poked’ with an empty needle the same number of times as the ODN-treated animals to mimic trauma to the scar tissue. The ODNs were delivered in the HVJ-liposome vector system via injections into the MCL scars 2 weeks after gap creation using the mesh grid. Semiquantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis showed a significant decrease in decorin mRNA levels in the antisense group when compared to the sense group 2 days and 3 weeks following injection.28 Western blot analysis of the MCL scars showed a corresponding significant decrease in the protein levels of decorin in the antisense group when compared to the sense group 4 weeks after injection.28 Electron microscopic evaluation of the scars showed both smaller and larger fibril diameters in the antisense group, similar to normal MCLs while the sense group and the HVJ-liposomes or ‘poke’ control groups had only smaller fibril diameters as seen in previous studies of MCL scars (Figure 2).4,16,19 There was a significant inverse correlation between fibril diameters and decorin mRNA levels when considering all MCLs in this study.16 Standard hematoxylin and eosin (H&E) sections showed a more normal appearance of the antisense scars, with some return of the normal crimp pattern seen on polarized light microscopy.16 However, this was not uniform throughout the scar tissue, a finding which may reflect the nonuniform delivery of the HVJ-liposomes to the tissue. Finally, biomechanical evaluation of the healing MCL scars 4 weeks after injection (6 weeks after injury) also showed improvements in the antisense group when compared to the sense and liposome groups or the ‘poke’ control group.16 The antisense scars had peak failure stresses around 85% greater than the other groups, while for lower load properties, the antisense scars had about 20% less creep, or stretching out, when compared to the other groups.
The results of antisense decorin gene therapy on the healing rabbit MCL are encouraging. However, while an effect was shown, caution is required. First, the failure stresses of the antisense group 4 weeks after injection (6 weeks after the original injury) were still only ≈20% of normal MCL failure stresses.16 Second, the presence of larger fibril diameters in the antisense scars was patchy in its appearance, likely a function of the difficulty in delivering the vector through the ligament tissue.16 Finally, the linear hypothesis of downregulating decorin expression with antisense ODN leading to a more normal fibril diameter distribution in the ligament scar which in turn improves biomechanical measures may not be due to such a simplistic cause and effect relationship in vivo.17 While decorin antisense ODNs may be specific in down-regulating decorin mRNA in cell culture systems, the alteration of mRNA synthesis in whole ligament specimens either in vitro or in vivo with decorin antisense ODNs is not as specific.17,28 Looking at MCL scars 4 weeks after antisense decorin ODNs were injected using HVJ-liposomes, mRNA levels of biglycan, lumican and versican (other proteoglycans), collagen types I and III, and transforming growth factor-beta 1 (TGF-β1) were also decreased.17 Total RNA levels were not altered by antisense decorin ODN therapy, indicating that the antisense decorin ODN therapy did not inhibit cell metabolism in general, and there were other molecules whose mRNA levels were unaffected.17 Thus, in the in vivo setting, the milieu is very complex and pertubations of one molecule can alter other molecule systems. For example, decorin and other small leucine-rich proteoglycans bind TGF-β1 in tissues rendering it inactive.17 Furthermore, decorin binds to the epidermal growth factor (EGF) receptor and can modify cell behavior via this mechanism also.34,35,36
In summary, several studies have shown the feasibility of gene intervention in ligaments. Variables include the ligament in question, the environment (normal versus healing), the vector system and the delivery method of the vector. Ligament healing has been successfully manipulated with gene therapy, but the in vivo consequences of manipulating one factor may have simultaneous effects on several biological processes which may have additive, synergistic or even antagonistic impacts on the final (clinical) result.
Gene intervention – tendons
Several investigators have studied gene transfer to tendons. Three model systems have been used; the patellar tendon of the knee,28,37 the digital flexor tendon,38,39,40,41 and the Achilles tendon.42 In these model systems, repair strength and gap formation of the tendon scar is important while with digital flexor tendons a second, equally important issue is adhesion formation between the healing tendon and the surrounding tendon pulley sheaths. Studies reported to date have considered a number of variables including different vectors, delivery methods and manipulation of the healing environment.
Four investigations have considered the feasibility of gene transfer in digital flexor tendons of chickens and dogs, and patellar or Achilles tendons of rats.37,39,41,42 Different vectors were used to deliver the β-galactosidase marker gene (Table 2). Nakamura et al,37 used the HVJ-liposome construct to deliver β-galactosidase to a rat patellar tendon. The medial half of the patellar tendon was transected in the mid portion and 3 days later the HVJ-liposomes were injected directly into the wound site. Approximately 7% of the cells in the wound were β-galactosidase positive 7 days after injection and this decreased to 0.2% of cells 56 days after injection. Interestingly, the authors observed β-galactosidase-positive cells throughout the length of the patellar tendon at later time points, suggesting that the cells in the wound that were transfected infiltrated to noninjured areas.37
Two studies considered the digital flexor tendons in dogs and chickens (Table 2).39,41 Using modified cationic liposomes, and a permeability step with a mild detergent (lysolecithin) to enhance transfection, an expression vector containing β-galactosidase was introduced into digital flexor tendons of adult dogs.41 The digital flexor tendons were cut in the first and fifth forepaw toes, and the liposomes were injected 2 min after the permeability step was begun. The dogs were killed 6 days later, and β-galactosidase staining was detected in the tendon, the tendon healing site, tendon sheath and skin. Transfection efficiencies were reported as 100% in the flexor tendon, including deep to the surface layer.41 The authors reported no evidence of immune response. The second digital flexor tendon model was that of chicken flexor tendons.39 An adenoviral-Lac Z construct was used. The vector was injected into the digital flexor tendon sheath percutaneously. β-Galactosidase expression was detected in the tendon and tendon sheath at 3, 30 and 75 days after injection. An estimated 2–5% of cells were β-galactosidase positive and most of these cells were on the surfaces of the tendon and tendon sheath.39
The fourth study used a transected rat Achilles tendon model without suturing to evaluate β-galactosidase expression.42 At 4 days after transection, recombinant adenovirus carrying LacZ was placed in the transection site. The authors detected β-galactosidase expression up to 17 days after injection (21 days after transection). Transduction rates increased with higher doses of injected virus, with 3% of cells positive with a dose of 109 PFU. Using a gelatin sponge to introduce the viral vector into the transection site of the Achilles tendon increased the number of β-galactosidase expressing cells to 6.7% and decreased the β-galactosidase activity in the adjacent muscles when compared to introducing the viral vector without the gelatin sponge.42
The previous four studies with tendons used direct injection of the vectors in or around the tendons in question. Özkan et al,43 investigated the feasibility of intra-arterial injection of HVJ-liposomes for transfecting cells of the patellar tendon. The patellar tendons of rats were completely transected. After 3 days, the femoral artery was injected with HVJ-liposomes carrying the β-galactosidase gene vector. The femoral artery and vein were clamped for 5 min following the injection to allow blood pooling around the patellar tendon. The authors found that between 8 and 12% of the cells in the healing wound of the patellar tendon were β-galactosidase positive in the first 2 weeks after injection of the HVJ-liposomes. By 56 days after the injection, approximately 0.7% of the cells were β-galactosidase positive.43
Gene therapy and investigation
Three studies have tried to manipulate the healing environment of tendons (Table 3).32,38,40 Two studies evaluated digital flexor tendons, studying adhesion formation and the tendon scar strength.38,40 Adhesion formation between the tendon and the surrounding tendon pulley limits the excursion of the tendon with muscle contraction and hence joint motion under the influence of the muscle–tendon unit. Lou et al,40 used gene intervention to manipulate the healing environment of chicken digital flexor tendons to investigate whether focal adhesion kinase (pp125 FAK) plays a role in tendon adhesions. FAK is an intracellular kinase involved with integrin-initiated cytoplasmic signal transduction and has been suggested to play a role in cell spreading/motility and wound healing.40 The authors made an adenoviral construct with sense FAK and used as controls adenoviral constructs with β-galactosidase or antisense FAK. In vitro studies showed that cultured chicken tendon fibroblasts overexpressed FAK after transection and transduction with adenovirus – sense FAK constructs when compared to cultured chicken tendon fibroblasts transduced with the other two constructs. Three groups of adult chickens had the right long toe injected with adenoviral constructs containing sense FAK, antisense FAK or β-galactosidase. After 2 days, the flexor profundus tendon of each toe was transfected to 50% of its width and not repaired. The feet were immobilized in casts and 4 weeks later the chickens were killed. Immunohistochemical studies showed overexpression of FAK in the sense FAK group when compared to the other two groups. Biomechanical measures of the work required to flex the chicken toes showed an almost 2 × greater work of flexion value in the sense FAK group when compared to the other two groups. The differences between the sense FAK groups and the other two groups were statistically significant.40 While the precise mechanisms through which FAK overexpression increased the work of flexion were not determined, this study did show that gene transfer may be used as a method to manipulate the healing environment.
The second chicken digital flexor tendon study examined the effect of a growth factor, bone morphogenetic protein-12 (BMP-12), on tendon healing.38 Unlike other members of the BMP family, injection of BMP-12 was found to induce transformation of soft tissues (muscle, subcutaneous fat) into ligament- or tendon-like material instead of bone. Lou et al,38 used gene transfer to transduce tendon cells and tendons in vivo with BMP-12 to investigate its effects on the tendon substance. They used an adenoviral construct with BMP-12 to transduce primary chicken tendon cells in vitro. Collagen type I synthesis was increased 30% by cells transduced with the adenoviral-BMP-12 construct when compared to control cells or cells transduced with an adenovirus- β-galactosidase construct. However, alkaline phosphatase activity, a marker of active bone formation, was not different between these three groups. The in vivo model used was a complete laceration of the middle toe flexor digitorum profundus bilaterally with repair of the tendon. Immediately after wound closure, one toe received adenovirus-BMP-12 injections while the contralateral toe received adenovirus- β-galactosidase injections. The chickens had both feet immobilized in casts and were killed 2 or 4 weeks later. While there were no significant differences in the ultimate failure force or the stiffness at 2 weeks, by 4 weeks these values for the adenoviral-BMP-12 group were approximately 2 × greater than the values of the adenoviral- β-galactosidase group. The differences between the two groups were significant at 4 weeks.38 The authors did not evaluate adhesion formation in this study, as this model of complete transection requires immobilization. Immobilization is known to enhance adhesion formation.
The third study on manipulating tendon healing involved an immunohistologic evaluation of rat patellar tendon healing following implantation of platelet-derived growth factor-B (PDGF-B) complementary deoxyribonucleic acid (cDNA).32 The mid-portion of the medial half of the patella tendon in Wistar male rats was transected. HVJ-liposomes containing the PDGF-B cDNA were injected into the wound site and in the fascial pocket over the patella tendon at the injury site. Empty HVJ-liposomes served as controls. Morphologic analysis was performed after rats were killed 1, 4 and 8 weeks following the surgery. PDGF-B distribution was detected with polyclonal antibodies to human PDGF-BB. Diffuse PDGF-B staining was present in both groups at the tendon injury site and the adjacent tendon at 1 week. At 4 weeks, PDGF-B was found in the wound site and the old tendon in the PDGF-B group, but at the wound site only in the control group. However, by 8 weeks, PDGF-B was expressed only at the wound site in both groups. Angiogenesis was evaluated with indirect fluorescence immunohistochemistry using polyclonal antibodies to rat laminin. There was a significant increase in the vascularity of the PDGF-B patellar tendon wounds when compared to control tissues 1 week after surgery, but there were no significant differences between the two groups at 4 and 8 weeks postsurgery.32 Collagen deposition was evaluated with polyclonal antibodies to rat collagen type I and matrix deposition was visualized histologically with Masson's Trichrome staining. In general, collagen deposition increased with time in both groups. Tendons from the PDGF-B group contained significantly more collagen at 4 weeks, but there were no differences between the groups 1 and 8 weeks postsurgery. Masson's Trichrome staining was greater in the PDGF-B group at 1 and 4 weeks postsurgery. Biomechanical evaluation was not reported.32 Thus, Nakamura et al, showed that with introduction of PDGF-B cDNA using HVJ-liposomes, rat patellar tendon healing had an early increase in angiogenesis in the first week, associated with increased collagen deposition at 4 weeks and increased matrix deposition 1 and 4 weeks after surgery. By 8 weeks, there were no differences between the PDGF-B group and control group with respect to angiogenesis, collagen deposition and matrix staining with Masson's Trichrome.
In summary, several studies have demonstrated gene intervention with tendons. Transfer of genes to tendons is feasible, and the healing environment can be manipulated, in the short term, at least. For tendons, the ‘therapeutic’ examples have been overexpression of genes such as PDGF-B to improve healing, while gene intervention by upregulation of a gene such as FAK was performed to investigate its role in tendon adhesions.
Gene intervention offers the opportunity to deliver ‘therapeutic’ agents or to manipulate the healing environment to examine roles of various molecules of interest. Examples for ligaments and tendons have been illustrated.16,17,32,38,40 While these reports are promising, there is still a paucity of reports demonstrating functional improvement in healing using such interventions. This is likely due to a variety of issues including identification of appropriate targets, tissue and host heterogeneity, vectors, delivery systems and access to the tissue, optimal timing of delivery, and insufficient molecular and cell biology knowledge of how the tissues mature and are maintained (Table 4). If one reorients thinking from a focus on improving healing to using gene transfer approaches to understand better maturation and homeostasis in specific tissues, the research findings may serve a dual purpose of identifying appropriate targets and elucidating the molecular and cellular mechanisms regulating uninjured tissues. This concept, initially raised during a previous workshop,44 will be developed further below.
Identification of appropriate targets for gene therapy
Tissue healing is a complex process with many factors that change over time. The identification of appropriate targets to improve the healing of ligaments and tendons is central to any rational gene therapy strategy. It is also quite clear that at the present state of knowledge, the choice of targets has been based on the literature, and not necessarily the ligament and tendon literature. While overexpression of PDGF-B showed some early improvements in tendon healing during the inflammatory phase, this early improvement was not evident in later phases.32 Another example is the antisense decorin approach which was effective in modifying the phenotype of the MCL scar tissue, although the improvement cannot be attributed solely to an effect on collagen assembly as decorin can modulate cell activity via interactions with the EGF receptor 34,35,36,45 and via modulating the availability of TGF-β.17 These examples illustrate some aspects of the complexity in the form of a time effect for PDGF-B and the multiple effects a ‘single’ intervention may have.
It is apparent that effective intervention will be a compromise between focusing on identifying the single target that is the mythical ‘magic bullet’ versus developing a gene therapy ‘cocktail’ containing a multitude of molecular specificities. However, the best approach may be to identify molecular targets that are associated with events occurring early after injury, and events that set the stage for the subsequent processes. In other systems, based in part on differences between the so-called scarless fetal healing compared to adult healing, some potential targets that could be applied to ligaments and tendons have been identified.46,47
Tissue and host heterogeneity
Tissue heterogeneity likely will play a role in both the choice of potential gene therapy targets, and their effective application to different tendons and ligaments throughout the body. Tissue heterogeneity is obvious comparing ligaments and tendons, but heterogeneity occurs within a tendon or ligament as well. For example, some tendons have fibrocartilagenous areas where they undergo compressive loading, while healing of the rabbit MCL varies with the site of injury within the MCL (midsubstance versus ligament insertion).48,49
Some factors that may contribute to this tissue heterogeneity can be considered mechanical or biological. Some structures are in the so-called ‘high-load environments’ while others likely function in ‘low-load environments’. As there are detectable differences in gene expression during maturation of the patellar tendon (high-load environment) versus the medial collateral ligament (low-load environment) of the rabbit (Lo et al, submitted), it is not unreasonable to assume that the choice of target and the effectiveness of a target may also vary between these tissues later during the healing process as the tissues start to be loaded again.50 An example of biologic influences on healing relates to the presence or absence of a synovial environment. A recent preliminary report from this laboratory has indicated that healing of injuries to the sheep MCL (extrasynovial) proceed very differently from those to the ACL (intrasynovial), particularly as the healing process proceeds to later stages.51 Gelberman's group has provided evidence that the healing of intrasynovial and extrasynovial tendon grafts are different in digital flexor tendon reconstructions.7
Another issue which is critical, but has not attracted much discussion is host heterogeneity. Age is an obvious host factor and the basis for age-related alterations in the healing process are fairly well described in the literature.52,53 In part, this may involve an age-related decline in inflammatory aspects of healing, particularly at the level of macrophage function which is critical early in the healing response of tissues such as skin.54,55 In the case of females, this age-related decline in healing may also be related to a loss of estrogen in the postmenopausal state.56 Thus, heterogeneity with regard to the host at different stages of the lifespan (ie prepuberty, sexual maturation, and in the case of females, the postmenopausal state) may also influence the choice of target to improve healing via gene therapy approaches.
Host heterogeneity may also be apparent at the level of genetics. Such genetic heterogeneity may contribute to phenotypic variation in the healing process and thus be the target of gene therapy interventions in their own right. Alternatively, some genetic heterogeneity may have no ‘overt’ phenotypic impact on the outcome of the healing process, but it may impact the efficacy of specific gene therapy interventions. Relevant to this discussion are recent studies from this laboratory indicating that skin wound healing in two different strains of pigs yields very different outcomes.57,58,59,60,61,62 Interestingly, gene expression patterns in ligaments, tendons and other connective tissues also differ between the two strains of pigs (Hart, Reno, Gallant, Wang and Olson, in preparation). While it remains to be determined how such differences impact the healing of tendon and ligament injuries in these porcine models, the studies do indicate that it is possible that genetic factors may impact ligament and tendon healing in different populations, and thus, would influence the choice of gene therapy target and the expected outcomes.
Another important variable in defining the identification of an appropriate target for gene therapy intervention relates to whether it will be used in a ‘normal’ host, or one in which healing is compromised in some manner. The healing process in otherwise healthy individuals is likely well optimized already and trying to improve on a functionally well-evolved process may be extremely difficult. However, one area that may lead to very fruitful research is in the area of compromised healing such as in those with diabetes. Biological interventions (ie administering preformed proteins) have been performed with some successes.63,64,65 While some of these reports have been associated with the process of skin wound healing there is no a priori reason why many of the biological events should not be similar.
Vectors and delivery systems for ligament and tendon
The design of a satisfactory vector and a delivery system to optimize efficiency of the gene transfer is a daunting task with factors unique to ligament and tendons and factors common to all tissues. In the skeletally mature individual, ligaments and tendons are dense hypocellular tissues that make effective delivery difficult. While such delivery can be more readily accomplished in less dense more hypercellular scar tissue, the efficiency of gene transfer is not 100%.16,28 Therefore, a target molecule that needs to be modulated in all cells in the tissue may not be the most appropriate, while a molecule that could exert an effect even if only a subset of cells were genetically modified (ie a growth factor that acts in a paracrine fashion) may be more appropriate. Other important properties of ligaments and tendons are that they are normally under tension, and comprise primarily water, making it difficult to administer a solution containing a vector in any quantity or uniformly to even a scar tissue, let alone a normal tissue. Some investigators have used the alternative approach of modifying cells in vitro and then transferring them to an early scar environment.24,25,27 However, in the relatively short-term (up to 6 weeks) studies published to date on gene transfer in ligaments, cell-based (ex vivo) techniques have not produced marker gene expression for as long as direct injection of viral vectors (in vivo).24,25 The mechanisms underlying these observations remain to be elucidated.
A second variable relates to choice of the vector to express the therapeutic function. While a number of viral vectors such as adenoviruses or adeno-associated viruses, as well as others, have been developed66 they are still not optimal for long-term expression of genes in many situations. Liposomes as delivery systems have some advantages, but due to the charge on the lipids usually used in such systems there can be interference with in vivo delivery due to molecules, such as proteoglycans, which are abundant in tissues such as ligaments and tendons. The use of a ‘gene gun’ approach to deliver genes to ligaments and tendons has not been reported, but the effectiveness of such an approach in hypocellular tissues would certainly be a ‘hit and miss’ proposition, and also leaves small gold particles behind in the tissues.
Another issue is the method used to modulate gene expression. Transfer of whole genes or small ODNs have been discussed in this article. The ODN approach led to some functional improvement in ligament healing, but it was not possible to determine a direct cause and effect relationship.17 Interestingly, recent reports using other systems have also indicated that the antisense approach may be more complicated than previously appreciated, even in vitro, and caution should be exercised in interpreting the results of such approaches.67 Such caution is not restricted to antisense approaches as recent in vitro studies in our laboratories using small interfering ribonucleic acids (siRNA) directed to other targets has also revealed that this gene therapy approach also leads to multiple targets being affected even though there is no molecular ‘cross-reactivity’ between the affected molecules (Wang, Ma, Tsao, Olson and Hart in revision).
Use of the gene intervention: improve healing or a research tool?
The above discussion has focused on the use and limitations of gene therapy approaches to improve healing of ligaments and tendons. This, of course, is where there is the greatest clinical need. However, one should not overlook the use of such interventions for other purposes. Gene transfer approaches could also be applied to achieving a better understanding of normal ligament and tendon maturation and maintenance.44 This approach to gene transfer is more a fundamental research tool which is essentially a tissue-specific refinement or modification of genetic approaches successfully used in mouse models (ie conditional transgenic and knockout mice) around the world.
Recent studies have indicated that ligaments and tendons appear to follow a reproducible, somewhat tissue-specific pattern of gene expression during maturation, which is associated with collagen assembly and large fibril formation.10 Based on this information, it may be possible to modulate the process using gene transfer approaches to modify the environment in the MCL or a specific tendon in one leg of a very young animal and assess the impact on the maturation process. Such an experimental design would control for physiological and loading environment and potentially allow for more accurate assessment of the critical factors operative in vivo. While the examples provided are obvious ones, there are a number of additional possibilities to use the gene transfer strategy to improve our understanding of the important regulatory processes that are critical to the functioning of these tissues in vivo.
Potential applications of gene intervention to ligamentous problems not currently being addressed
The discussion has focused on tendons and ligaments, with particular emphasis on using such approaches to improve healing. An area related to ligaments, is that of joint capsules, which in some joints can be a thin connective tissue, while in others such as the shoulder, there is a spectrum of tissues which can resemble ligaments (broad, narrow, thin, thick) or capsules. Problems associated with these tissues have not attracted much gene therapy research attention. While the complexity of the human shoulder poses a set of real challenges with regard to such interventions, injuries to ligament, tendon and capsular structures of the shoulder are very frequent and can also lead to decreased quality of life for many young people and athletes, with increased risk for development of osteoarthritis later in life.
Another set of clinical problems related to joint capsules is joint contracture where following injury the ‘healed’ tissue has lost its normal elasticity and has become rigid with loss of joint motion and function.68,69,70,71 While the molecular basis for such complications to the healing process is now receiving more attention,69 it has not been the subject of any gene therapy reports based on our search of the literature. In part, this is due to lack of complete understanding of the mechanisms involved, but this is also an area where gene therapy approaches may lead to improvement in joint function and potentially impact a number of patients.
In summary, gene therapy applications for use in ligaments and tendons range from use in very clinically relevant situations to improve healing of injured tissues with a goal to improving function and preventing degenerative joint disease, to more applied uses focused on generating better ligament and tendon replacement tissues (which could be autologous and avoid potential infectious disease complications), to very basic applications to provide more detail of the molecular events operative in normal tissues. In many respects, this basic to clinical spectrum of applications can be viewed as a continuum, with the different application approaches interdependent. If one views gene therapy from such a perspective, there is great potential for development of successful therapeutic interventions to restore compromised ligament and tendon properties to levels that are required for effective function.
Hildebrand KA, Frank CB . Ligaments: structure, function and response to injury and repair. In: Dee R, Hurst LC, Gruber MA, Kottmeier SA (eds) Principles of Orthopaedic Practice, 2nd ed. McGraw-Hill: New York, 1997, pp 109–117.
Woo SL-Y et al. Anatomy, biology, and biomechanics of tendon and ligament. In: Buckwalter JA, Einhorn TA, Simon SR (eds). Orthopaedic Basic Science: Biology and Biomechanics of the Musculoskeletal System, 2nd ed. 2000, pp 581–616.
Hildebrand KA et al. The effects of platelet-derived growth factor-BB on healing of the rabbit medial collateral ligament. Am J Sports Med 1998; 26: 549–554.
Frank CB, Amiel D, Akeson WH . Healing of the medial collateral ligament of the knee. Acta Orthop Scand 1983; 54: 917–923.
Thornton GM, Boorman RS, Shrive NG, Frank CB . Medial collateral ligament autografts have increased creep response for at least two years and early immobilization makes this worse. J Orthop Res 2002; 20: 346–352.
Goldfarb CA et al. The effect of variations in applied rehabilitation force on collagen concentration and maturation at the intrasynovial flexor tendon repair site. J Hand Surgery 2001; 26A: 841–846.
Seiler JGI et al. Autogenous flexor tendon grafts. Clin Orthop 1997; 345: 239–247.
Gelbermann RH et al. The effect of gap formation at the repair site on the strength and excursion of intrasynovial flexor tendons: an experimental study on the early stages of tendon-healing in dogs. J Bone Joint Surg 1999; 81: 975–982.
Manske PR . Flexor tendon healing. J Hand Surgery 1988; 13-B: 237–245.
Lo IKY et al. The cellular networks of normal ovine medial collateral and anterior cruciate ligaments are not accurately recapitulated in scar tissue. J Anatomy 2002; 200: 283–296.
Bray RC, Rangayyan RM, Frank CB . Normal and healing ligament vascularity: a quantitative histological assessment in the adult rabbit medial collateral ligament. J Anatomy 1996; 188: 87–95.
Ivie TJ, Bray RC, Salo PT . Denervation impairs healing of the rabbit medial collateral ligament. J Orthop Res 2002; 20: 990–995.
Gronblad M et al. Immuno reactive neuropeptides in nerves in ligamentous tissue. An experimental neuroimmunohistochemical study. Clin Orthop 1991; 265: 291–296.
Ackerman PW et al. Anatomic innervation of tendons, ligaments and joint capsules. A morphologic and quantitative study in rats. J Orthop Res 2001; 19: 372–378.
Amiel D et al. Tendons and ligaments: a morphological and biochemical comparison. J Orthop Res 1984; 1: 257–265.
Nakamura N et al. Decorin antisense gene therapy improves functional healing of early rabbit ligament scar with enhanced collagen fibrillogenesis in vivo. J Orthop Res 2000; 18: 517–523.
Hart DA et al. Complexity of determining cause and effect in vivo after antisense gene therapy. Clin Orthop 2000; 379S: S242–S251.
Niyibizi C, Kavalkovich K, Woo SL-Y . Type V collagen is increased during rabbit medial collateral ligament healing. Knee Surgery Sports Traumatol 2000; 8: 281–285.
Frank CB et al. Collagen fibril diameters in the healing adult rabbit medial collateral ligament. Connective Tissue Res 1991; 27: 251–263.
Thornton GM, Boorman RS, Shrive NG, Frank CB . Early immobilization further impairs the inferior long-term creep behavior. 1999; 42: 391–392; The American Society of Mechanical Engineers. Bioengineering Conference 1999. Goel VK, Spilker RL, Ateshian GA, Soslowsky LJ.
Majima T et al. Compressive compared with tensile loading of medial collateral ligament scar in vitro uniquely influences mRNA levels for aggrecan, collagen type II, and collagenase. J Orthop Res 2000; 18: 524–531.
Banes AJ et al. PDGF-BB, IGF-I and mechanical load stimulate DNA synthesis in avian tendon fibroblasts in vitro. J Biomech 1995; 28: 1505–1513.
Bray RC, Butterwick DJ, Doschak MR, Tyberg JV . Coloured microsphere assesment of blood flow to knee ligaments in adult rabbits: effects of injury. J Orthop Res 1996; 14: 618–625.
Hildebrand KA et al. Early expression of marker genes in the rabbit medial collateral and anterior cruciate ligaments: the use of different viral vectors and the effects of injury. J Orthop Res 1999; 17: 37–42.
Menetrey J et al. Direct-, fibroblast- and myoblast-mediated gene transfer to the anterior cruciate ligament. Tissue Eng 1999; 5: 435–442.
Gerich TG et al. Gene transfer to the rabbit patellar tendon: potential for genetic enhancement of tendon and ligament healing. Gene Therapy 1996; 3: 1089–1093.
Day CS et al. Myoblast-mediated gene transfer to the joint. J Orthop Res 1997; 15: 894–903.
Nakamura N et al. A comparison of in vivo gene delivery methods for antisense therapy in ligament healing. Gene Therapy 1998; 5: 1455–1461.
Sawchuk SJ et al. Anti-T cell receptor monoclonal antibody prolongs transgene expression following adenovirus-mediated in vivo gene transfer to mouse synovium. Human Gene Ther 996; 7: 499–506.
Tripathy SK, Black HB, Goldwasser E, Leiden JM . Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replication-defective adenovirus vectors. Nat Med 1996; 2: 545–550.
Palmer TD, Rosman GJ, Osborne WRA, Miller AD . Genetically modified skin fibroblasts persist long after transplantation but gradually inactive introduced genes. Cell Biol 1991; 88: 1330–1334.
Nakamura N et al. Early biological effect of in vivo gene transfer of platelet-derived growth factor (PDGF)-B into healing patellar ligament. Gene Therapy 1998; 5: 1165–1170.
Chimich D et al. The effects of initial end contact on medial collateral ligament healing: a morphological and biomechanical study in a rabbit model. J Orthop Res 1991; 9: 37–47.
Santra M, Reed CC, Iozzo RV . Decorin binds to a narrow region of the epidermal growth factor (EGF) receptor, partially overlapping but distinct from the EGF-binding epitope. J Biol Chem 2002; 277: 35671–35681.
Iozzo RV, Moscatello DK, McQuillan DJ, Eichstetter I . Decorin is a biological ligand for the epidermal growth factor receptor. J Biol Chem 1999; 274: 4489–4492.
Patel S et al. Decorin activates the epidermal growth factor receptor and elevates cytosolic Ca2+ in A431 carcinoma cells. J Biol Chem 1998; 273: 3121–3124.
Nakamura N et al. Transient introduction of a foreign gene into healing rat patellar ligament. J Clin Invest 1996; 97: 226–231.
Lou J et al. BMP-12 gene transfer augmentation of lacerated tendon repair. J Orthop Res 2001; 19: 1199–1202.
Lou J, Manske PR, Aoki M, Joyce ME . Adenovirus-mediated gene transfer into tendon and tendon sheath. J Orthop Res 1996; 14: 513–517.
Lou J et al. In vivo gene transfer and overexpression of focal adhesion kinase (pp125FAK) mediated by recombinant adenovirus-induced tendon adhesion formation and epitendon cell change. J Orthop Res 1997; 15: 911–918.
Goomer RS et al. Nonviral in vivo gene therapy for tissue engineering of articular cartilage and tendon repair. Clin Orthop 2000; 379S: S189–S200.
Dai Q, Manfield L, Wang Y, Murrell GAC . Adenovirus-mediated gene transfer to healing tendon – enhanced efficiency using a gelatin sponge. J Orthop Res 2003; 21: 604–609.
Özkan I et al. Direct in vivo gene transfer to healing rat patellar ligament by intra-arterial delivery of haemagglutinating virus of Japan liposomes. Eur J Clin Invest 1999; 29: 63–67.
Hart DA, Evans CH . Breakout session 8: ligament and tendon. Clin Orthop 2000; 379S: S260–S261.
Iozzo RV . The family of the small leucine-rich proteoglycans: key regulators of matrix assembly and cellular growth. Crit Rev Biochem Mol Biol 1997; 32: 141–174.
Bullard KM, Longaker MT, Lorenz HP . Fetal wound healing: current biology. World J Surg 2003; 27: 54–61.
Peled ZM et al. Matrix metalloproteinases and the ontogeny of scarless repair: the other side of the wound healing balance. Plastic Reconstr Surg 2002; 110: 801–811.
Malaviya P et al. An in vivo model for load-modulated remodeling in the rabbit flexor tendon. J Orthop Res 2000; 18: 116–125.
Frank CB, Loitz BJ, Shrive NG . Injury location affects ligament healing. Acta Orthop Scand 1995; 66: 455–462.
Hart DA et al. Similarities and differences between in vivo and in vitro analysis at the functional and molecular levels. Recent Res Dev Biophys Biochem 2002; 2: 153–177.
Lo IKY et al. The healing response of ACL and MCL injuries is initially similar but diverges with time. Proceedings of World Congress of Biomechanics 2002, Calgary, Canada: August.
Thomas DR . Age related changes in wound healing. Drugs Aging 2001; 18: 607–620.
Ashcroft GS, Mills SJ, Ashworth JJ . Ageing and wound healing. Biogerontology 2002; 3: 337–345.
Ashcroft GS, Horan MA, Ferguson MWJ . Aging alter the inflammatory and endothelial cell adhesion molecule profiles during human cutaneous healing. Lab Invest 1998; 78: 47–58.
Hackman DJ, Ford HR . Cellular, biochemical and clinical aspects of wound healing. Surg Infect 2002; 3 (Suppl 1): 23–35.
Ashcroft GS et al. Estrogen accelerates cutaneous wound healing associated with an increase in TGF-beta 1 levels. Nat Med 1997; 3: 1209–1215.
Gallant CL, Wright JB, Olson ME, Hart DA . The red Duroc pig model of hypertrophic wound healing. Wound Repair Regeneration 2001; 9: 140.
Gallant CL, Olson ME, Hart DA . Skin wound healing in red Duroc pigs: potential role of growth factors and cytokines in scar formation. Wound Repair Regeneration 2002; 10: A19.
Gallant CL, Olson ME, Hart DA . Full- and partial-thickness skin wounds in red Duroc pigs leads to hypercontracted, hyperpigmented scars. Wound Repair Regeneration 2002; 10: A19.
Gallant CL, Olson ME, Hart DA . Molecular and histologic analysis of skin wound healing in Yorkshire X red Duroc F1 pigs: Evidence for the dominant influence of the red Duroc genetic contribution. Wound Repair Regeneration 2003; 11: A7.
Wang JF . Molecular and cell biology of skin wound healing in a pig model. Connective Tissue Res 2000; 41: 195–211.
Wang JF . The pig as a model for excisional skin wound healing: characterization of the molecular and cellular biology, and bacteriology of the healing process. Compar Med 2001; 51: 341–348.
Bennett SP et al. Growth factors in the treatment of diabetic foot ulcers. Br J Surg 2003; 90: 133–146.
Eming SA, Smola H, Krieg T . Treatment of chronic wounds: state of the art and future concepts. Cells Tissues Organs 2002; 172: 105–117.
Pierce GF et al. Tissue repair processes in healing chronic pressure ulcers treated with recombinant platelet-derived growth factor BB. Am J Pathol 1994; 145: 1399–1410.
Evans CH, Robbins PD . Current concepts review: possible orthopaedic applications of gene therapy. J Bone Joint Surg 1995; 77: 1103–1114.
Fisher AA et al. Evaluating the specificity of antisense oligonucleotide conjugates: a DNA array analysis. J Biol Chem 2002; 277: 22980–22984.
Akeson W, Amiel D, Woo SL-Y . Immobility effects on synovial joints the pathomechanics of joint contracture. Biorheology 1980; 17: 95–110.
Hildebrand KA et al. Myofibroblast numbers are elevated in human elbow joint capsules following trauma. Clin Orthop Relat Res 2004, In press.
Hildebrand KA, Holmberg M, Shrive NG . A new method to measure post-traumatic joint contractures in the rabbit knee. J Biomech Eng 2004, In press.
Hildebrand KA, Sutherland C . Post-traumatic rabbit knee joint contractures are permanent. Trans Orthop Res Soc 2003; 28: 1275.
We thank Carla Gronau for preparing the manuscript. Dr KA Hildebrand is a Clinical Investigator with the Alberta Heritage Foundation for Medical Research, Dr CB Frank is the McCaig Professor of Surgery, University of Calgary, and Dr DA Hart is the Calgary Foundation – Grace Glaum Professor in Arthritis Research, University of Calgary.
About this article
Cite this article
Hildebrand, K., Frank, C. & Hart, D. Gene intervention in ligament and tendon: current status, challenges, future directions. Gene Ther 11, 368–378 (2004) doi:10.1038/sj.gt.3302198
- gene therapy
- gene intervention
Compound phenotype of osteogenesis imperfecta and Ehlers–Danlos syndrome caused by combined mutations in COL1A1 and COL5A1
Bioscience Reports (2019)
Macromolecular Bioscience (2018)
Association of COL5A1 gene polymorphisms and risk of tendon-ligament injuries among Caucasians: a meta-analysis
Sports Medicine - Open (2018)
Effect of Uniaxial Tensile Cyclic Loading Regimes on Matrix Organization and Tenogenic Differentiation of Adipose-Derived Stem Cells Encapsulated within 3D Collagen Scaffolds
Stem Cells International (2017)
Different combinations of growth factors for the tenogenic differentiation of bone marrow mesenchymal stem cells in monolayer culture and in fibrin-based three-dimensional constructs