Optimizing mesenchymal stem cell therapy for tendon-bone healing: Multifaceted approaches and future directions

Abstract

Tendon-bone healing remains a significant clinical challenge due to the high risk of re-rupture following injury. While mesenchymal stem cells (MSCs) show great potential in enhancing tendon-bone healing, their clinical application is limited by issues such as low delivery efficiency, restricted differentiation potential, and potential immunogenicity. Recently, various strategies combining MSCs with other approaches, such as preconditioning, biomaterial integration, gene modification, and exosome application, have been developed, resulting in improved therapeutic outcomes. This review explored the current methods used to optimize MSC therapy for tendon-bone healing, examining the advantages, disadvantages, and underlying mechanisms of each approach, providing a foundation for future research and clinical applications.

Key Words: Tendon-bone healing; Mesenchymal stem cells; Exosomes; Biomaterials; Preconditioning; Gene modification

Core Tip: This review synthesized advances optimizing mesenchymal stem cell-based therapy for tendon-bone healing, spanning metabolic or mechanical preconditioning, instructive biomaterials (aligned fibers, gradient mineralization, controlled release), gene/cargo engineering, and exosome-centered paracrine modulation. These strategies target persistent hurdles (poor homing/engraftment, lineage commitment at the fibrocartilaginous interface, hostile inflammatory milieu, and immunogenicity) while improving zonal enthesis regeneration and mechanical integration. We also highlighted scalable manufacturing and safety/readout standardization as key enablers to translate robust efficacy from preclinical models to rigorous clinical trials.

INTRODUCTION

Tendon-bone healing is a complex process that transitions tendon tissue to bone and is crucial for joint stability and mobility. However, after tendon-bone injuries, natural healing often fails to restore the original structure, leading to fibrosis, functional impairment, and high re-rupture rates[1]. Tendon-bone injuries typically require surgery, but postoperative outcomes often do not fully restore function, leaving patients at risk of poor healing and re-injury. For example, failure rates for rotator cuff tear (RCT) repair range from 20% to 94%, and anterior cruciate ligament reconstruction (ACLR) failure rates range from 10% to 25%[2]. Thus, effective strategies for promoting tendon-bone healing remain a significant clinical challenge.

Stem cell therapy, especially with mesenchymal stem cells (MSCs), has gained attention for tissue repair and regeneration[3]. MSCs are widely accessible, have immunomodulatory properties, and can differentiate into osteoblasts, chondrocytes, fibroblasts, and tenocytes, all essential for tendon-bone healing[4,5]. Moreover, MSCs secrete bioactive factors that accelerate tissue regeneration. However, challenges like low engraftment efficiency, limited differentiation potential, and possible immunogenicity hinder the broader application of MSCs[2,6,7].

To address these limitations several strategies have been developed to enhance MSC efficacy. These include physical and pharmacological preconditioning, MSC-biomaterial combinations for improved delivery, gene modifications, and novel cell-free therapies using MSC derivatives. These approaches show promise in enhancing MSC effectiveness and provide new insights for tendon-bone healing research and clinical applications (Figure 1).

Figure 1 Multiple approaches to improve mesenchymal stem cell-mediated tendon-bone healing. Current strategies to enhance mesenchymal stem cell (MSC)-based treatment for tendon-bone healing can be broadly categorized into four approaches: Preconditioning of MSCs; combination with biomaterials; genetic modification; and the use of MSC-derived exosomes. MSC: Mesenchymal stem cell; LIPUS: Low-intensity pulsed ultrasound; ESW: Extracorporeal shock wave.

STRUCTURE AND HEALING PROCESS OF TENDON-BONE INSERTION

The tendon-bone insertion (TBI) is a specialized tissue that connects tendons or ligaments to bone, providing crucial support and strength for joint movement. TBIs are classified into direct and indirect insertions. Examples of direct insertions include the rotator cuff and anterior cruciate ligament[5]. This structure consists of four zones: Tendon/Ligament (I); unmineralized fibrocartilage (II); mineralized fibrocartilage (III); and bone (IV)[2]. The transitional nature of this structure ensures a gradual shift in mechanical properties from tendon to bone, preventing stress concentration during load transmission[8]. The healing process of TBI occurs in four stages: Inflammation; proliferation; remodeling; and maturation[1]. While TBI has some capacity for self-healing, restoring its natural gradient structure is challenging. This often results in the formation of fibrous scar tissue and a high risk of re-rupture[1].

MAJOR TYPES AND CHARACTERISTICS OF MSCS

MSCs are classified into different types based on their source. Bone marrow-derived stem cells (BMSCs) are the most extensively studied and widely used, typically obtained through invasive procedures[2]. Adipose-derived stem cells (ADSCs) are easier to collect in larger quantities from subcutaneous adipose tissue. They have strong adipogenic potential, but their osteogenic and chondrogenic capabilities are comparatively weaker[9]. Synovium-derived stem cells exhibit strong chondrogenic potential but are challenging to extract and expand[2]. Tendon-derived stem cells (TDSCs)/tendon stem progenitor cells (TSPCs) and periosteum-derived stem cells show superior osteogenic, chondrogenic, and adipogenic potential compared with BMSCs although many of their properties are still being explored and validated[10,11].

MECHANISMS OF MSCS IN TENDON-BONE HEALING

MSCs promote tendon-bone healing mainly through two mechanisms: Direct differentiation and paracrine effects. In direct differentiation MSCs transform into target cells, strengthening the tendon-bone interface. The paracrine effect involves the secretion of bioactive factors and exosomes that indirectly regulate the healing process[12]. Through exosomes and growth factor release, MSCs stimulate the proliferation and differentiation of endogenous cells, thereby promoting tissue repair[13,14]. Moreover, MSCs modulate inflammation at the TBI site, reducing excessive inflammation and scar formation. They also enhance angiogenesis by releasing exosomes or growth factors, improving blood supply and supporting tissue regeneration[5]. However, their self-renewal ability carries an inherent risk of tumor formation. MSCs may contribute to tumor progression and metastasis by altering cancer cell behavior, regulating immune responses, and promoting angiogenesis[15] (Figure 2).

Figure 2 Gradient structure of tendon-bone insertion and the mechanism of mesenchymal stem cells promoting tendon-bone healing. Mesenchymal stem cells derived from various tissues promote the regeneration of the four-layered structure of the anterior cruciate ligament and rotator cuff through direct differentiation and paracrine signaling mechanisms. MSC: Mesenchymal stem cell.

LIMITATIONS OF MSCS AS A STANDALONE TREATMENT

Despite their therapeutic potential MSCs have several limitations. The most common delivery method, local injection, faces challenges such as cell leakage and low survival rates[2]. MSCs also have a short retention time at the injury site, often requiring multiple injections for optimal results, and there is currently no standardized dosage or administration protocol[7]. Additionally, stem cell therapies carry risks, including carcinogenesis, immune rejection, and ethical concerns regarding the use of embryonic stem cells[6]. Moreover, some stem cell types are difficult to obtain, and the donor’s age can significantly affect MSC functionality[7]. These challenges limit the broader application of MSC therapies and have spurred ongoing efforts to improve their effectiveness.

MULTIFACETED STRATEGIES TO OPTIMIZE MSCS

Preconditioning

Mechanical stimulation: MSCs can sense mechanical stimuli through integrin-focal adhesion complexes, mechanosensitive ion channels (such as Piezo1, transient receptor potential channels), and primary cilia. These stimuli activate pathways like Yes-associated protein/transcriptional co-activator with PDZ-binding motif, Wnt/β-catenin, and transforming growth factor-β (TGF-β)/Smad, which regulate cytoskeletal remodeling, gene expression, and differentiation[16]. Applying appropriate mechanical forces promotes MSC differentiation and enhances their immunoregulatory functions, thereby supporting tendon-bone healing.

Cyclic mechanical stretching (CMS) primarily affects MSCs by promoting osteogenesis, a process mediated through the activation of the Smad signaling pathway[17]. In mouse BMSCs CMS induces DNA demethylation by downregulating DNA methyltransferase 3 beta and upregulating Sonic hedgehog, activating the hedgehog pathway. In human BMSCs CMS activates the sirtuin 1-AMP-activated protein kinase pathway[18,19]. Animal models have also confirmed the positive effects of mechanical stimulation on MSCs in promoting tendon-bone healing. In a mouse RCT model, mechanical stretching promotes macrophage polarization toward the M2 phenotype and enhances transforming growth factor-β1 secretion, improving BMSCs cartilage differentiation[20]. In a rabbit ACLR model, mechanical stretching boosts BMSCs and tendon cell proliferation and matrix formation[21]. However, precise control of amplitude and duration is crucial. Insufficient mechanical stimulation can increase inflammatory cytokine secretion, promoting osteoclast differentiation and bone resorption, while continuous stretching suppresses runt-related transcription factor 2 (Runx2) expression and osteogenesis[22,23]. Studies suggest that 5% cyclic stretching is the most effective for MSCs[19,20].

Low-intensity pulsed ultrasound (LIPUS) is a promising preconditioning strategy for MSCs, generating cavitation, and mechanical and thermal effects. As a simple, noninvasive, and safe therapy, LIPUS has been widely used in orthopedic treatments, offering significant benefits for tendon-bone healing[24,25]. Research shows that LIPUS stimulates MSC proliferation, inhibits adipogenic differentiation, and enhances osteogenic and chondrogenic differentiation[24]. LIPUS-enhanced MSC proliferation is mediated through the phosphatidylinositol 3-kinase/protein kinase B and mitogen-activated protein kinases signaling pathways[24]. In terms of osteogenic differentiation, LIPUS activates the Rho-associated kinase-cancer Osaka thyroid oncogene/tumor progression locus 2 kinase-mitogen-activated protein kinase kinase 1-extracellular signal-regulated kinases (ERKs) pathway[26]. Additionally, LIPUS enhances chondrogenic differentiation by inhibiting the tumor necrosis factor pathway and promoting autophagy[27,28].

Extracorporeal shock wave (ESW) therapy applies compressive, tensile, and shear forces through transient pressure waves[29]. ESW has long been shown to promote tendon-bone healing with recent studies uncovering its underlying mechanisms[30]. ESW enhances BMSCs proliferation and osteogenic differentiation as well as osteogenic differentiation in TDSCs and ADSCs although their sensitivity varies: TDSCs > BMSCs > ADSCs[31,32]. Mechanistic studies indicate that ESW primarily influences the miR-138-focal adhesion kinase-ERK1/2 and reactive oxygen species-ERK1/2-bone morphogenetic protein 2 (BMP2)-Smad pathways, ultimately activating RUNX2[32,33].

Compression and fluid shear stress (FSS) can promote MSC osteogenic or chondrogenic differentiation with their effects depending on both intensity and duration. For example, MSC differentiation varies with compression levels. Human BMSCs undergo osteogenesis under 10% dynamic compression but switch to chondrogenesis at 15%[34]. Rapid FSS favors chondrogenesis while slow FSS promotes osteogenesis[35]. Additionally, intermittent FSS is more effective than continuous FSS in promoting osteogenesis, likely due to a resting period that facilitates cytoskeletal remodeling[36]. While these mechanical stimuli can theoretically enhance MSC function, their specific effects on tendon-bone healing require further investigation.

Pharmacological stimulation: Various drugs have been shown to promote BMSCs osteogenic differentiation, presenting significant potential for tendon-bone healing. Secretory leukocyte protease inhibitor, total flavonoids of Rhizoma Drynariae, polyvinylpyrrolidone iodine, baicalein, and icariin all promote MSC osteogenic differentiation, thereby enhancing tendon-bone healing[37-41]. Additionally, andrographolide preconditioning increases the resistance of BMSCs to stress and improves MSC transplant survival by inducing heme oxygenase-1 expression[42].

Other stimuli: Other stimuli also enhance MSC functionality. Dedifferentiation treatment promotes osteogenic differentiation through the Nanog/nuclear factor of activated T-cells 1/osterix pathway, improving healing[43]. Both Mg²⁺ and static magnetic fields stimulate MSC proliferation and osteogenic differentiation[44,45]. Hypoxia promotes MSC proliferation and chondrogenic differentiation although its effects on osteogenesis and adipogenesis remain debated[46]. These findings suggest additional strategies for MSC preconditioning (Table 1).

Table 1 Summary of the effects of preconditioning methods on stem cells.

Ref.	Pretreatment method	Cell type	Effects on stem cells	Mechanism	Animal model
Song et al[21], 2017	Mechanical stimulation	Rabbit BMSCs	Promote proliferation and differentiation	Increase collagen I, collagen III, ALP, OPN, tenascin C, and tenomodulin expression	Rabbit ACLR model
Wang et al[20], 2023	Mechanical stimulation	Mouse BMSCs	Promote chondrogenic differentiation	Stimulate macrophage polarization towards the M2 phenotype and secretion of TGF-β1	Mouse ACLR model
Li et al[17], 2015	Mechanical stimulation	Rat BMSCs	Inhibit adipogenic differentiation	Activate the TGFβ1/Smad2 signaling pathway	N/A
Kusuyama et al[26], 2014	LIPUS	Mouse MSCs line	Inhibit adipogenic differentiation, promote osteogenic differentiation	Regulate the ROCK-Cot/Tpl2-MEK-ERK signaling pathway and PPARγ2 activity	N/A
Chen et al[27], 2023	LIPUS	hUC-MSCs	Promote chondrogenic differentiation	Inhibit the TNF signaling pathway	Rat cartilage defect model
Wang et al[28], 2019	LIPUS	Rat BMSCs	Promote chondrogenic differentiation	Inhibit autophagy	N/A
Zhao et al[29], 2021	ESW	Human SCB-SPCs	Promote self-renewal	Activate the YAP/TAZ signaling pathway	Rabbit osteochondral defect model
Chen et al[31], 2017	ESW	Rat BMSCs	Promote proliferation and osteogenic differentiation	Increase Col1, OSX, Runx2, and ALP expression	Rat femoral shaft bone defect model
Hu et al[32], 2016	ESW	Human BMSCs, TDSCs, ADSCs	Promote osteogenic differentiation	Inhibit miR-138 to activate the FAK-ERK1/2-RUNX2 signaling pathway	Nude mouse bone induction model
Catalano et al[33], 2017	ESW	Human ADSCs	Promote osteogenic differentiation	Activate the ROS-ERK1/2-BMP2-Smad-RUNX2 signaling pathway	N/A
Wu et al[37], 2022	SLPI	Rat BMSCs	Promote migration and osteogenic differentiation	Upregulate Runx2, ALP, OCN, and OPN gene expression	Rat ACLR model
Han et al[38], 2024	TFRD	Mouse BMSCs	Promote vitality and osteogenic differentiation	Activate ERR1/2-Gga1-TGF-β/MAPK pathway	Rat ACLR model
Zhang et al[39], 2017	PVP-I	Rabbit BMSCs	Promote osteogenic differentiation	Increase the expression of BMP-2 and OPN	Rabbit ACLR model
Tian et al[40], 2018	Baicalein	Rat TDSCs	Promote osteogenic differentiation	Activate the Wnt/β-catenin signaling pathway	Rat calcaneus-Achilles tendon injury model
Wang et al[41], 2016	Icariin	Mouse MSCs	Promote osteogenic differentiation	Activate the Wnt/β-catenin signaling pathway	Mouse calvarial osteolysis model
Alipanah-Moghadam et al[42], 2023	Andrographolide	Rat BMSCs	Increase cell resistance to environmental stress	Induce the expression of HO-1	N/A
Tie et al[43], 2021	Dedifferentiated	Rabbit BMSCs	Promote osteogenic differentiation	Activate the Nanog/NFATc1/OSX signaling pathway	Rabbit ACLR model
Díaz-Tocados et al[44], 2017	Mg2+	Rat BMSCs	Promote proliferation and osteogenic differentiation	Activate the Notch1 signaling pathway	Rat femur decellularized scaffold
Kim et al[45], 2015	Static magnetic fields	Human BMSCs	Promote proliferation and osteogenic differentiation	Upregulate ALP, BSP-2, COL1A1, OCN, ON, OPN, OSX, and RUNX2 gene expression	N/A

BMSC: Bone marrow-derived stem cell; ALP: Alkaline phosphatase; OPN: Osteopontin; ACLR: Anterior cruciate ligament reconstruction; TGF: Transforming growth factor; LIPUS: Low-intensity pulsed ultrasound; MSC: Mesenchymal stem cell; ROCK: Rho-associated kinase; Cot/Tpl2: Cancer Osaka thyroid oncogene/tumor progression locus 2 kinase; MEK: Mitogen-activated protein kinase kinase 1; FAK: Focal adhesion kinase; ERK: Extracellular signal-regulated kinase; PPARγ2: Peroxisome proliferator-activated receptor gamma 2; hUC-MSC: Human umbilical cord-derived mesenchymal stem cell; TNF: Tumor necrosis factor; ESW: Extracorporeal shock wave; SCB-SPCs: Subchondral bone stem/progenitor cells; YAP: Yes-associated protein; TAZ: Transcriptional co-activator with PDZ-binding motif; Col1: Type I collagen; NFATc1: Nuclear factor of activated T-cells 1; OSX: Osterix; Runx2: Runt-related transcription factor 2; TDSC: Tendon-derived stem cell; ADSC: Adipose-derived stem cell; ROS: Reactive oxygen species; BMP2: Bone morphogenetic protein 2; OCN: Osteocalcin; SLPI: Secretory leukocyte protease inhibitor; TFRD: Total flavonoids of Rhizoma Drynariae; ERR1/2: Estrogen-related receptor 1 and 2; Gga1: Golgi-localized γ-ear containing ADP ribosylation factor-binding protein 1; MAPK: Mitogen-activated protein kinases; PVP-I: Polyvinylpyrrolidone iodine; HO-1: Heme oxygenase-1; BSP-2: Bone sialoprotein-2; COL1A1: Collagen type 1 alpha 1; ON: Osteonectin; N/A: Not available.

Combination with biomaterials

Conventional biological scaffolds: Biomaterials can serve as delivery vehicles for MSCs, enhancing their functionality to accelerate tendon-bone healing. For example, 3D-printed poly lactic-co-glycolic acid scaffolds loaded with BMSCs and biomimetic hydroxyapatite gradient scaffolds with human umbilical cord-derived MSCs have shown positive results[47,48]. Calcium silicate nanoparticles-modified natural fish scale biomaterials promote BMSCs and TSPC differentiation by activating the BMP-2/Smad/Runx2 pathway[49]. Additionally, superparamagnetic iron oxide-labeled BMSCs seeded in a biphasic scaffold under a magnetic field improve cell distribution, seeding efficiency, and chondrogenesis through the CDR1as/miR-7/fibroblast growth factor 2 pathway[50].

Hydrogel-based materials: Hydrogel materials are excellent drug carriers due to their prolonged drug retention and high drug-loading efficiency[51]. Hydrogels encapsulating stem cells are widely used in regenerative medicine, reducing cell membrane damage during injection and extending in vivo retention time[52]. A four-layered hydrogel made of UV-crosslinked gelatin/hyaluronic acid, nanoclay, and BMSCs can mimic the natural entheses structure, promoting fibrocartilage regeneration and inhibiting fat infiltration in a rat RCT model[53]. Combining ADSCs with hydrogels, such as fibrin or methacrylated gelatin, or loading ADSCs with platelet-rich plasma in extracellular matrix (ECM) hydrogels has also shown promising repair results[54,55].

Natural biomaterials: Natural biomaterials, such as ECM, cell sheet technology, demineralized bone matrix, and fibrin glue, are increasingly being studied for tendon-bone healing due to their advantages over synthetic materials. ECM provides excellent biocompatibility, bioactivity, and biosafety[56], enhancing MSC proliferation, osteogenic differentiation, and promoting osteoinductive factors in macrophages[57-59]. Cell sheet technology retains cell-to-cell connections and ECM structure, making it highly applicable in regenerative medicine[60]. Cell sheets derived from various sources, including CD34+ cells from the anterior cruciate ligament, periosteal progenitor cells, BMSCs, urine-derived stem cells, ADSCs, TDSCs, and ligament-derived stem cells, have shown promise in tendon-bone healing[61-67]. However, the combination of MSCs with demineralized bone matrix or fibrin glue still requires further validation as some studies in rat RCT models and patients with RCT did not show significant therapeutic benefits[68,69] (Table 2).

Table 2 Summary of biomaterial-enhanced stem cell therapy for tendon-bone healing.

Ref.	Biomaterial	Material type	Cell type	Functions	Model
Chen et al[47], 2020	3D-printed PLGA scaffolds	Conventional biological scaffolds	Rabbit BMSCs	Support cell growth and differentiation	Rabbit RCT model
Yea et al[48], 2020	Hydroxyapatite-gradient scaffold	Conventional biological scaffolds	hUC-MSCs	Support cell adhesion, migration, and proliferation, promoting osteogenic and chondrogenic differentiation	Rat RCT model
Han et al[49], 2023	CS-FS	Conventional biological scaffolds	Rabbit BMSCs and TSPCs	Enhance cell differentiation and activity, maintaining the phenotype	Rat and rabbit RCT models
Zhang et al[50], 2024	Magnetically seeded biphasic scaffold	Conventional biological scaffolds	SPIO-BMSCs	Increase cell seeding efficiency, promote cell distribution and concentration, and enhance chondrogenic differentiation	Rat RCT model
Ji et al[53], 2023	Cocktail-like gradient gelatin/hyaluronic acid	Hydrogel-based materials	Rat BMSCs	Simulate natural gradient structure, support long-term cell culture and embedding, promote cell growth and differentiation	Rat RCT model
Rothrauff et al[54], 2019	Fibrin or GelMA	Hydrogel-based materials	Rat ADSC	Promote chondrogenic differentiation	Rat RCT model
McGoldrick et al[55], 2017	ECM hydrogel	Hydrogel-based materials	Rat ADSC	Better biocompatibility, enhance repair efficacy	Rat calcaneus-Achilles tendon injury model
Shekaran et al[58], 2016	ECM	Natural biomaterials	Human BMSCs	Promote cell proliferation and osteogenic differentiation	Mouse ectopic mineralization model
Deng et al[59], 2021	ECM	Natural biomaterials	Human BMSCs	Promote macrophage secretion of osteoinductive factors, enhance osteogenic differentiation	Mouse femoral defect model
Mifune et al[61], 2013	Cell sheets	Natural biomaterials	Human ACL-derived CD34+ cell	Increase proprioceptive recovery, graft maturation, and biomechanical strength	Rat ACLR model
Chang et al[62], 2012	Cell sheets	Natural biomaterials	Rabbit PPCs	Maintain cell differentiation capacity, promote fibrocartilage formation	Rabbit ACLR model
Tang et al[63], 2020	Cell sheets combined with acellular scaffolds	Natural biomaterials	Rabbit BMSCs	Promote cell differentiation, enhance new bone and fibrocartilage formation	Rabbit patella-patellar tendon injury model
Chen et al[64], 2020	Cell sheets	Natural biomaterials	Canine USCs	Promote fibrocartilage formation, increase trabecular thickness and biomechanical strength	Canine RCT model
Matsumoto et al[65], 2021	Cell sheets	Natural biomaterials	Human ADSCs	Bone tunnel narrowing, increased biomechanical strength	Rabbit ACLR model
Yao et al[66], 2023	Cell sheets	Natural biomaterials	Rat TDSCs	Enhance bone formation and angiogenesis, regulate macrophage polarization and MMP/TIMP expression	Rat ACLR model
Wei et al[67], 2023	Cell sheets	Natural biomaterials	LDSCs with BMP-2/TGF-β1 gene insertion	Promote osteogenic and tenogenic differentiation, improve biomechanical strength, enhance tissue maturation, inhibit bone tunnel widening	Mouse ACLR model

PLGA: Poly lactic-co-glycolic acid; BMSC: Bone marrow-derived stem cell; RCT: Rotator cuff tear; hUC-MSC: Human umbilical cord-derived mesenchymal stem cell; CS-FS: Chitosan-silk fibroin scaffold; TSPC: Tendon stem progenitor cell; SPIO: Superparamagnetic iron oxide; GelMA: Methacrylated gelatin; ADSC: Adipose-derived stem cell; ECM: Extracellular matrix; ACL: Anterior cruciate ligament; ACLR: Anterior cruciate ligament reconstruction; PPC: Periosteal progenitor cell; USC: Urine-derived stem cell; TDSC: Tendon-derived stem cell; MMP: Matrix metalloproteinase; TIMP: Tissue inhibitors of metalloproteinase; LDSC: Ligament-derived stem cell; BMP-2: Bone morphogenetic protein 2; TGF: Transforming growth factor.

Overall, most biomaterials provide stable support for MSCs, enhancing their survival, preventing leakage, and promoting adhesion, proliferation, and differentiation, thereby improving tendon-bone healing quality[70]. Synthetic scaffolds offer high mechanical strength, aiding early tendon-bone repair, but their degradation rate may not align with tissue regeneration, and their degradation products can affect the microenvironment[71]. Hydrogels with their flexibility are ideal for stem cell and drug delivery, allowing control over mechanical properties and drug release with emerging bone-targeted applications[72]. While natural materials offer excellent biocompatibility, they suffer from rapid degradation, poor processability, and lower mechanical strength compared to synthetic scaffolds[71].

Gene modification

Gene modification techniques enhance tendon-bone healing by improving stem cell functions, providing an innovative approach to tissue regeneration with significant clinical potential. For example, lentiviral vectors overexpressing Runx1 and vascular endothelial growth factor (VEGF) A promote osteogenic and chondrogenic differentiation of BMSCs, enhance proliferation, and inhibit miR-205-5p expression, improving healing in rat ACLR and RCT models[73,74]. Adenoviral vectors overexpressing BMP-12, BMP-2, and basic fibroblast growth factor promote tenogenic differentiation, BMSC proliferation, and osteogenic differentiation, enhancing healing in rabbit RCT and ACLR models[75,76]. Additionally, adenoviral overexpression of calcitonin gene-related peptide activates the cAMP/PKA/cAMP response element-binding protein/JUNB pathway, promoting osteogenesis and upregulating Sonic hedgehog expression, improving healing in a mouse ACLR model[77].

With the rapid development of CRISPR-Cas9 gene editing, stem cell research has advanced significantly due to its precision and efficiency[78]. For instance, CRISPR interference to knock down the BMP-2 antagonist Noggin or upregulate Wnt10b and forkhead box c2 enhances osteogenic differentiation of ADSCs and BMSCs, promoting bone regeneration in rat cranial defect models[79,80]. CRISPR-Cas9 overexpression of BMP-2 and VEGF in tonsil-derived MSCs, combined with vitamin D-incorporated poly lactic-co-glycolic acid scaffolds, promotes osteogenesis, angiogenesis, and macrophage M2 polarization, aiding bone regeneration in rat models[81]. Recent studies have shown that injecting CRISPR-engineered BMSCs with SRY-box transcription factor 9 activation and RelA (also known as p65) suppression into the joint cavity promotes cartilage formation and reduces osteoarthritis progression[82]. Furthermore, combining bioinformatics and artificial intelligence can help identify key gene targets to enhance MSC functionality, improving gene editing efficiency and effectiveness. However, using CRISPR/Cas9 to enhance MSC function for tendon-bone healing may pose risks such as off-target effects and immune reactions. These risks can be minimized by designing precise single-guide RNAs and utilizing non-viral delivery systems, such as nanoparticles, to reduce immunogenicity[83]. Currently, CRISPR/Cas9 research in tendon-bone healing remains limited. Targeting key genes or pathways, or using MSC-derived exosomes as nanocarriers for CRISPR/Cas9 delivery, may be more effective strategies for tendon-bone regeneration.

MSCs derivatives

Natural exosomes: MSC-derived exosomes have garnered attention as a promising cell-free therapeutic strategy, with their bioactive substances considered key mediators of stem cell efficacy in various diseases and tendon-bone healing[14]. BMSC-derived exosomes (BMSCs-Exos) and ADSC-derived exosomes promote BMSC proliferation, migration, osteogenic, and chondrogenic differentiation, demonstrating comparable therapeutic effects[84]. BMSCs-Exos also enhance angiogenesis and promote M1 to M2 macrophage polarization via miR-23a-3p targeting interferon regulatory factor 1[85,86]. ADSC-derived exosomes improve the histological characteristics of torn human supraspinatus tendons by enhancing AMP-activated protein kinase signaling and inhibiting Wnt/β-catenin activity[87]. Furthermore, infrapatellar fat pad-derived MSC exosomes show promising healing effects in rat ACLR models[88]. Recent studies also suggest that TSPC-derived exosomes are highly effective in promoting tendon-bone healing[89]. While exosomes have lower immunogenicity than stem cells, challenges such as low yield, poor stability, and limited targeting remain. Currently, research is focused on engineering strategies to improve their therapeutic efficacy[1] (Figure 3).

Figure 3 Methods for engineering exosomes preconditioning and endogenous loading guide parent cells to produce exosomes enriched with desired proteins and nucleic acids. Exogenous loading modifies isolated exosomes to carry target molecules or to recognize specific cells through peptide-linked surface sites. VEGF: Vascular endothelial growth factor; BMP-2: Bone morphogenetic protein 2; MSC: Mesenchymal stem cell; CAP: Cartilage affinity peptide; LIPUS: Low-intensity pulsed ultrasound; miRNA: MicroRNA.

Parent cell preconditioning: Preconditioning methods not only enhance MSCs but also improve the bioactivity of their derived exosomes. For instance, hypoxia-preconditioned BMSCs-Exos accelerate healing in rat ACLR models by stimulating bone formation and angiogenesis while kartogenin-preconditioned BMSCs-Exos promote collagen maturation and cartilage formation, improving recovery in rat RCT models[90,91]. Magnetically preconditioned BMSCs-Exos using iron oxide nanoparticles and a magnetic field enhance healing in rat ACLR models by boosting miR-21-5p secretion and activating the Smad pathway[92]. LIPUS preconditioning increases miR-140 levels in BMSCs-Exos, promoting chondrogenic differentiation and suppressing adipogenic differentiation, thus improving healing in mouse RCT models[93]. Additionally, lyophilized human umbilical cord stem cell exosomes improve storage and usability while maintaining efficacy[94].

Exosomes delivery platforms: Exosomes with their inherent therapeutic effects can also be engineered as multifunctional drug delivery nanocarriers. Due to their high biocompatibility, exosomes are promising tools for precisely delivering bioactive substances to injured sites, extending therapeutic duration, and supporting mechanistic studies with broad application potential[1,95].

Endogenous loading involves introducing therapeutic molecules into parent cells before exosome generation[1]. For example, scleraxis-overexpressing platelet-derived growth factor receptor alpha (+) BMSCs-Exos secrete more miR-6924-5p, which inhibits osteoclast formation and promotes Achilles tendon healing in mice[96]. Exogenous loading on the other hand involves directly introducing therapeutic molecules into isolated exosomes, a simpler and commonly used method for developing delivery systems[1]. For instance, loading exosomes with BMP-2 or VEGF plasmids promotes bone regeneration with VEGF-loaded exosomes also enhancing angiogenesis[97,98].

A key research focus is engineering exosomes to target specific cells as targeted delivery increases therapeutic concentration and reduces side effects[99]. Lysosome-associated membrane protein 2b, one of the isoforms of lysosome-associated membrane protein 2, is widely used in exosome engineering[99,100]. By fusing cell-targeting peptides, such as human papillomavirus E7 peptide or cartilage affinity peptide, to lysosome-associated membrane protein 2b, exosomes can efficiently deliver cargo to target sites[99]. For example, exosomes targeting synovial fluid-derived MSCs or chondrocytes combined with kartogenin or matrix metalloproteinases 13 small interfering RNA can help alleviate osteoarthritis progression[101,102]. While research on targeted exosomes in tendon-bone healing is still limited, these studies offer valuable insights for future tendon-bone repair strategies, such as developing pH-targeted exosomes to deliver cytokines or microRNA to enhance healing at inflammation sites.

CONCLUSION

Currently, MSC therapy for tendon-bone healing focuses on improving delivery efficiency, enhancing therapeutic efficacy, and reducing immunogenicity. To improve delivery, combining MSCs with biomaterials such as biological scaffolds and hydrogels helps regulate their release and degradation, ensuring sustained action at the injury site. For enhanced efficacy, strategies like biomaterial support, mechanical stimulation, drug treatments, and gene editing promote MSC functions like proliferation, differentiation, migration, and anti-inflammatory properties, leading to improved healing outcomes. To reduce immunogenicity, MSC-derived exosomes help minimize the immunogenic risks associated with allogeneic MSCs, increasing safety (Table 3).

Table 3 Limitations of mesenchymal stem cell therapy and main optimization strategies.

Limitations	Strategies	Functions
Low delivery efficiency	Biomaterials	Sustained release of MSCs, reduced degradation, preventing cell leakage, and prolonging retention at the injury site
Limited direct differentiation potential	Preconditioning/gene modification	Enhancing MSC differentiation towards bone, cartilage, tendon, and other tissues
Limited cell functionality	Preconditioning/gene modification	Boosting MSC proliferation, migration, angiogenesis, and immune modulation capabilities
Immunogenicity	Exosomes	Acellular therapies that eliminate cellular immunogenicity

MSC: Mesenchymal stem cell.

Despite these advances MSC therapy still faces challenges. First, during culture MSCs inevitably undergo phenotypic, functional, and genetic changes that may lead to functional decline and potential tumorigenic risks[15]. Using autologous or allogeneic cells and reducing passage numbers can help minimize these risks[103]. The donor’s age also affects therapeutic efficacy with MSCs from younger donors showing better in vitro proliferation and differentiation potential[104]. Additionally, large-scale MSC expansion for exosome extraction can cause aging-related functional decline. Recent studies suggest replacing fetal bovine serum with human platelet lysate may help address this issue[105]. Furthermore, adhering to good manufacturing practices, regulatory frameworks, and ethical considerations remains a challenge[106].

Future research should combine strategies to overcome these limitations and improve therapeutic outcomes. Investigating new preconditioning methods will provide deeper insights into MSC biology. The use of bioinformatics and artificial intelligence for targeted gene editing could further enhance MSC efficacy. Engineering MSC-derived exosomes for cell-free therapies can help avoid immunogenic risks while improving treatment effectiveness. Exploring new biomaterials could enhance the precise delivery and local retention of MSCs and exosomes. In conclusion, advancing research into MSC optimization strategies will help address the challenges of tendon-bone healing.