Therapeutic Applications of Genes and Gene-Engineered Mesenchymal Stem Cells for Femoral Head Necrosis

Abstract

Osteonecrosis of the femoral head (ONFH) is a common and disabling joint disease. Although there is no clear consensus on the complex pathogenic mechanism of ONFH, trauma, abuse of glucocorticoids, and alcoholism are implicated in its etiology. The therapeutic strategies are still limited, and the clinical outcomes are not satisfactory. Mesenchymal stem cells (MSCs) have been shown to exert a positive impact on ONFH in preclinical experiments and clinical trials. The beneficial properties of MSCs are due, at least in part, to their ability to home to the injured tissue, secretion of paracrine signaling molecules, and multipotentiality. Nevertheless, the regenerative capacity of transplanted cells is impaired by the hostile environment of necrotic tissue in vivo, limiting their clinical efficacy. Recently, genetic engineering has been introduced as an attractive strategy to improve the regenerative properties of MSCs in the treatment of early-stage ONFH. This review summarizes the function of several genes used in the engineering of MSCs for the treatment of ONFH. Further, current challenges and future perspectives of genetic manipulation of MSCs are discussed. The notion of genetically engineered MSCs functioning as a “factory” that can produce a significant amount of multipotent and patient-specific therapeutic product is emphasized.

Introduction

Osteonecrosis of the femoral head (ONFH) is a refractory disabling hip disease caused by the decrease in blood supply to the femur head and the impairment of bone marrow cell activity.^1,2 According to its etiology, ONFH is categorized as traumatic or non-traumatic. The traumatic ONFH is the consequence of physical injuries, such as femoral neck fracture (FNF) or hip dislocation, directly affecting blood flow to the femoral head. The non-traumatic ONFH is often associated with long-term or high-dose treatment with glucocorticoids (GCs), alcoholism, and sickle cell disease.

The traumatic ONFH is typically secondary to FNF in older patients. However, the prevalence of ONFH in children and adolescents with FNF remains high, regardless of the treatment.³ On the contrary, non-traumatic ONFH, in particular, the GC-induced ONFH (GIONFH), commonly affects young adults and physically active patients.⁴ ONFH results in significant psychological stress for patients and their families and places a sizable burden on the health care system.⁵ Unfortunately, the pathogenesis of ONFH has not yet been fully elucidated, and the treatment options remain limited.

Clinically, the therapy of early-stage ONFH is mainly restricted to physical interventions, pharmacotherapy, surgical core decompression (CD), osteotomy, and vascularized bone grafting.^6

–10 However, the outcomes of these joint-preserving techniques are not consistently satisfactory.¹¹ Once the collapse of the femoral head has occurred, reconstruction of the joint remains the last resort. Although this type of surgery can significantly restore mobility, reduce pain, and improve patients' quality of life, the associated complications and high cost constitute a considerable barrier in its application.^12,13 In addition, particularly in the case of young patients, the limited lifespan of implants increases the risk that revision surgery will be required.¹⁴ Therefore, further research into potential novel treatments of early ONFH is necessary.

Mesenchymal stem cells (MSCs) have been the subject of extensive research related to bone regeneration. Interest in MSCs results from their unique characteristics of self-renewal, immune privilege, osteogenic properties, secretion of regeneration-promoting factors, and easiness of genetic modification.^15,16 Nevertheless, the functional benefits of stem cell transplantation are limited in the inhospitable environment present in the necrotic area, such as inflammatory reaction, immune rejection, hypoxia, and oxidative stress.^17,18

To overcome this obstacle, different methods have been recently explored to increase the survival of grafted stem cells and enhance their therapeutic impact. Stem cells engineered to deliver therapeutic proteins have been successfully tested experimentally and even applied in clinical trials. However, this strategy has some disadvantages, such as the relatively short half-life of the delivered protein under physiologic conditions, presence of adverse side effects, difficulty to achieve sustained, controllable delivery, and problems with the delivery of intracellular proteins.¹⁹ In addition, the delivery of a gene vector together with stem cells to the site of injury is often imprecise and short-lasting due to the vulnerability of pure DNA to nucleases in vivo.²⁰

Genetic modification of MSCs is emerging as a promising strategy to enhance intrinsic mechanisms of bone repair and to protect against harsh conditions by increasing the autocrine activity of the cells and regulating their function. Herein, we will discuss the therapeutic significance of various genes utilized in the genetic engineering of MSCs for ONFH treatment. Further, the current challenges and the future of MSC-based genetic manipulation strategy for ONFH will be analyzed. We envision that genetic modifications of MSCs can enhance their regenerative function that can be harnessed for the femoral head-preserving therapy in ONFH patients. The preparation methods and application of genetically engineered MSCs for ONFH are listed in Table 1.

Table 1.

Preparation methods and application of genetically engineered mesenchymal stem cells for osteonecrosis of the femoral head

Model	Animal	Cells	Gene Modification	Method of Transplant	Longest Follow-Up Period	Adverse Effects	Efficacy	Ref.
GIONFH	Rats	BMSCs	LV-MDR1	MSCs; 1 × 10⁷; i.v.	10 weeks	None	Higher osteogenic variables	³³
							Lower adipocytic variables
							Reduced the incidence of GIONFH
	Rats	BMSCs	LV-DKK-1 siRNA	MSCs; 1 × 10⁶; i.m; Femur	8 weeks	None	Prevented the progression of necrosis	⁴²
	Rabbits	BMSCs	LV-HIF-1α	CD + MSCs; 5 × 10⁶	4 weeks	None	Promoted angiogenesis and bone formation	⁴⁹
	Rabbits	BMSCs	LV-bFGF	CD + MSCs + XACB; cell density ≥5 × 10⁶/mL	12 weeks	None	Inhibited TNF-α expression	^57 –59
							Improved OPG expression
							Promoted angiogenesis and bone formation
	Rabbits	BMSCs	AV-hHGF	CD + MSCs; 1 × 10⁶	4 weeks	None	Obvious osteogenesis and angiogenesis	⁶⁵
	Rabbits	BMSCs	AV-hHGF	CD + MSCs; 1 × 10⁶	4 weeks	None	Rich bone marrow in hematopoietic tissue	⁶⁵
	Rabbits	BMSCs	AV-hHGF	CD + MSCs + FG; 1 × 10⁷	8 weeks	None	Improved the long-term efficacy compared with uncompounded MSCs	⁶⁶
	Rats	hBMSCs	LV-SDF-1α	MSCs; 1 × 10⁷; i.m; Tibia	6 weeks	None	Promoted angiogenesis and bone regeneration	⁷⁴
Traumatic ONFH	Goats	BMSCs	AV-hBMP2	CD + MSCs + β-TCP; 1 × 10⁷	16 weeks	None	Higher volume of new bone	⁸¹
	Goats	BMSCs	AV-hBMP2	CD + MSCs + β-TCP; 1 × 10⁷	16 weeks	None	Better mechanical properties	⁸¹
	Rabbits	BMSCs	Non-viral BMP2	MSCs + Mg rod	12 weeks	None	Recovery with normal running ability	⁸²
	Rabbits	BMSCs	Non-viral BMP2	MSCs + Mg rod	12 weeks	None	Prevented the occurrence of ONFH	⁸²
	Rabbits	BMSCs	AAV-hVEGF-165	CD + MSCs	8 weeks	None	Promoted angiogenesis and bone formation	⁸⁶
	Rabbits	BMSCs	AAV-hVEGF-165	CD + MSCs	8 weeks	None	Recovery with compressive strength	⁸⁶
	Dogs	BMSCs	Non-viral hVEGF-165	CD + MSCs; 2 × 10⁷	16 weeks	None	Promoted angiogenesis and bone formation	⁸⁷
	Rabbits	BMSCs	AV-hHGF	MSCs; 1 × 10⁶	4 weeks	None	Recovery with decreased empty lacunae and hematopoietic tissue	⁸⁹
	Rabbits	BMSCs	AV-hHGF	MSCs; 1 × 10⁶	4 weeks	None	Promoted angiogenesis and bone repair	⁸⁹
	Rabbits	BMSCs	AV-hHGF (Tet-on)	CD + MSCs; 1 × 10⁷; orally with 5 mg/kg Dox	Not given	None	Reduced the incidence of ONFH	⁹²
	Rabbits	BMSCs	Co-transfection; rAAV-hVEGF-165 and rAAV-hBMP2	CD + MSCs; 1 × 10⁹	8 weeks	None	Large new blood vessels and bone formation	⁹⁴
	Dogs	BMSCs	AV-BMP2-bFGF	DBM + MSCs; 4 × 10⁹	12 weeks	None	Promoted angiogenesis and bone formation	⁹⁷
	Dogs	BMSCs	AV-BMP2-bFGF	DBM + MSCs; 4 × 10⁹	12 weeks	None	Enhanced the compression and bending strength	⁹⁷

β-TCP, beta-tricalcium phosphate; AAV, adenovirus-associated virus; AV, adenovirus; bFGF, basic fibroblast growth factor; BMP2, bone morphogenetic protein 2; BMSCs, bone marrow mesenchymal stem cells; CD, core decompression; DBM, demineralized bone matrix; DKK-1, Dickkopf-1; Dox, doxycycline; FG, fibrin glue; GIONFH, glucocorticoid-induced osteonecrosis of the femoral head; HGF, hepatocyte growth factor; HIF-1α, hypoxia-inducible factor 1α; i.m, intramedullary injection; i.v, intravenous injection; LV, lentivirus; MDR1, multidrug resistance gene 1; ONFH, osteonecrosis of the femoral head; OPG, osteoprotegerin; SDF-1α, stromal cell-derived factor 1α; siRNA, small interfering RNA; TNF-α, tumor necrosis factor-alpha; VEGF, vascular endothelial growth factor; XACB, xenogeneic antigen-cancellous bone.

Non-Traumatic ONFH

To date, several hypotheses explaining the pathogenesis of non-traumatic ONFH have been proposed, including abnormal fat metabolism,^21,22 inflammation,²³ circulatory impairment,²⁴ and cell dysfunction.^25

–28 The cell dysfunction hypothesis proposes that the crucial pathological mechanism of the onset and development of non-traumatic ONFH involves the decrease in proliferation ability, increase in cell apoptosis, and disruption of differentiation pathways in bone marrow MSCs (BMSCs) in the microenvironment of the femoral head.^22,29 A switch in the differentiation of BMSCs into adipocytes results not only in the reduction in the pool of stem cells available for the generation of osteoblasts but also in vascular compression by excessive fat, ultimately leading to insufficient repair and remodeling of the necrotic bone.

Therefore, the restoration of normal cell differentiation by genetic modification appears as an optimal choice to promote bone regeneration and prevent the progress of necrosis in ONFH. Given that the research on the application of genetically engineered MSCs in the treatment of non-traumatic ONFH is focused on the GIONFH model, this review will concentrate on GC-dependent osteonecrosis. Typically, the research effort was concentrated on the modification of MSCs with genes regulating the side effects of GCs on stem cells, with the objective of improving their function.

Drug transporters

P-glycoprotein (P-gp) is a plasma membrane protein encoded by multidrug resistance gene 1 (MDR1); it regulates drug absorption and distribution in the human body. Active P-gp extrudes its substrates, including GCs, from the cell, reducing their intracellular availability and concentration.³⁰ Several lines of evidence point to the close association of P-gp with the development of GIONFH. The MDR1 gene polymorphism implicates the host susceptibility to GIONFH.³¹ The aberrant hypermethylation modification of the MDR1 promoter in MSCs results in lower cell viability and is more frequent in patients with GIONFH than in patients with FNF. Further, the decrease in P-gp activity increases the level of GCs in BMSCs, which, in turn, leads to an elevation in intracellular oxidative stress and adipocyte conversion.²⁸

An early study by Han and coworkers³² found that rifampicin-induced activation of P-gp significantly reduced the risk of GIONFH in rats, from 80% in the absence of rifampicin to 50% in its presence. This reduction was most likely due to the inhibition of adipogenesis and cell apoptosis in the femoral head.

Recently, the same laboratory successfully transduced BMSCs with lentiviral vectors carrying the green fluorescent protein (GFP) and MDR1 genes. Overexpression of MDR1 in vitro promoted the intracellular level and activity of P-gp, reduced GC accumulation, upregulated the expression of osteogenesis-related markers (Runx2), and enhanced osteogenic differentiation, while significantly reducing adipogenesis. In vivo, intravenous injection of 1 × 10⁷ BMSCs carrying the GFP and MDR1 genes promoted P-gp and Runx2 expression in the femur head in rats, increasing the rate of mineral apposition. Importantly, the incidence of GIONFH was 10% by post-injection week 10; whereas in the untreated group the incidence was 90%, and in the group treated with BMSCs carrying only GFP it was 55%.

Moreover, GFP-positive cells with an osteoblast-like morphology attached to trabecular surfaces in the femoral head were observed.³³ Thus, increasing the expression of MDR1 in MSCs appears to prevent the incidence of GIONFH.

Transcriptional regulators and transcription factors

Dickkopf-1

The main components of the canonical Wnt/β-catenin signaling pathway are Wnt ligands, the seven-transmembrane receptor Frizzled, its co-receptor, lipoprotein-related protein five and six (LRP5/6), and downstream key molecule β-catenin. The Wnt/β-catenin signaling pathway constitutes a critical mechanism regulating bone metabolism and homeostasis, and its inhibition negatively affects cellular functions in the early stage of GIONFH.³⁴ For example, aberrant CpG island hypermethylation in the Frizzled 1 gene was observed in BMSCs from GIONFH patients, and it was associated with a decreased cell viability and osteogenesis, and enhanced adipogenesis. Pretreatment of BMSCs with 5′-Aza-dC, an inhibitor of DNA methyltransferase, reversed these effects.³⁵ Therefore, the Wnt/β-catenin signaling pathway could be used as a therapeutic target in the treatment of GIONFH.

Dickkopf-1 (DKK-1) is an antagonist of LRP5/6 and is implicated in various bone diseases.³⁶ A high level of DKK-1 inhibits bone morphogenetic protein 2 (BMP2)-induced osteoblast differentiation.³⁷ Mice overexpressing DKK-1 develop severe osteopenia.³⁸ Conversely, osteocyte-specific DKK-1 deletion protects against GC-induced bone loss.³⁹ In ONFH patients, the upregulation of the expression of DKK-1 and proapoptotic protein Bad in bone tissue and higher serum level of DKK-1 correlated with the GC-induced apoptosis of BMSCs and the progression of ONFH.⁴⁰ Downregulation of DKK-1 in BMSCs extracted from GIONFH patients improved the osteogenic differentiation and proliferative activity of the cells in vitro.⁴¹ Therefore, the inhibition of DKK-1 in BMSCs may arrest the progression of necrosis in GIONFH, most likely by the inhibition of apoptosis and enhancement of bone formation.

Zhun and coworkers⁴² examined the possibility of therapeutic application of lentivirus-infected BMSCs stably overexpressing small interfering RNA (siRNA) against DKK-1 in an animal model of GIONFH. They have found that allotransplantation of 1 × 10⁶ siRNA-overexpressing BMSCs into the femoral medullary cavity of rats with GIONFH effectively prevented the progression of the disease at a histological level at 8 weeks after injection, but the effects were not significantly improved in comparison with therapy using unmodified BMSCs. These results may be related to the multifactorial characteristics of GIONFH, and further studies are needed to identify the mechanism involved.

Hypoxia-inducible factor 1α

Hypoxia-inducible factor 1α (HIF-1α), a key nuclear factor activated in response to hypoxia, is easily degraded under normoxic conditions. However, under low oxygen concentration, HIF-1α forms a stable heterodimer with HIF-1β and increases the transcription of numerous hypoxia-responsive genes that affect cell survival and metabolism.⁴³ Importantly, HIF-1α plays a central role in angiogenic-osteogenic coupling during bone regeneration induced by MSCs. It also upregulates the expression of multiple angiogenic factors, such as vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), angiopoietin, stromal cell-derived factor 1 (SDF-1), placenta growth factor, and stem cell factor.^44
–46

Therefore, the activation of HIF-1α might trigger the synergistic effects of multiple growth factors, resulting in more effective osteogenesis and angiogenesis than when induced by a single factor.

The expression of HIF-1α is markedly increased in the femoral head in response to hypoxia at the early stage of ischemic injury, thus constituting a compensatory regulatory mechanism. However, this response decreases at later stages of necrosis and restoration.⁴⁷ The increased expression of HIF-1α in the ischemic femoral head of animals was accompanied by the upregulation of SOX9 expression, indicating a chondroprotective role of HIF-1α.⁴⁸

Ding et al. ⁴⁹ documented that BMSCs stably overexpressing HIF-1α introduced by lentiviral vector were characterized by an enhanced VEGF secretion, expression of osteogenic genes, and osteogenic differentiation. These cells were used in a rabbit model of GIONFH; 5 × 10⁶ of engineered BMSCs were injected into the femur head through the bone tunnel made by a drill. After 4 weeks, intact and well-distributed subchondral trabeculae accompanied by a small amount of granulation tissue, significant new bone mass, and dense vasculature were formed. In addition, exosomes derived from BMSCs overexpressing HIF-1α facilitated the repair of GIONFH.⁵⁰ These studies provide support for the application of HIF-1α-expressing MSCs in the treatment of ONFH.

Growth factors

Basic fibroblast growth factor

bFGF plays a pivotal role in bone metabolism. The disruption of the bFGF gene results in a dramatic decrease in bone formation and bone mass in mice.⁵¹ Moreover, bFGF also exhibits many desirable functions, including upregulation of VEGF expression in osteoblasts, and activation of proliferation, migration, and osteogenic differentiation of MSCs.^52
–54

Recently, extensive research has focused on the application of MSCs genetically engineered to express bFGF. The level of bFGF in the culture supernatant of adipose tissue-derived MSCs (ADSCs) transduced with the GFP-Luc-bFGF vector increased two-fold for up to 14 days. In vivo, GFP-positive cells could be detected for at least 21 days after injection into the muscle adjacent to the fracture site in mice, promoting active intramembrane bone formation, augmenting angiogenesis, and accelerating the restoration of bone strength.⁵⁵ In a segmental bone defect model, the implantation of a biodegradable porous beta-tricalcium phosphate (β-TCP) ceramics containing bFGF-BMSCs promoted significantly new bone formation accompanied by abundant regeneration of functional capillaries.⁵⁶

Osteoprotegerin (OPG) is a pseudo-receptor of nuclear factor-κB ligand (RANKL). OPG prevents RANKL from binding to RANK expressed on the surface of osteoclast precursor cells, thereby inhibiting the maturation and differentiation of osteoclasts.

Peng and coworkers^57
–59 implanted a xenogeneic antigen-cancellous bone (XACB) seeded with rabbit bFGF-BMSCs into the necrotic region after CD in the rabbit model of early GIONFH. The expression of OPG in the femoral head significantly increased at 3, 6, and 12 weeks after implantation. In contrast, the expression of tumor necrosis factor-alpha (TNF-α), a proinflammatory cytokine, remained decreased for up to 12 weeks post-transplant. Histological analysis confirmed that bFGF-MSCs/XACB tissue-engineered bone dramatically promoted the formation of new bone and prevented the collapse of the femoral head. Under hypoxic conditions, MSCs transfected with the bFGF gene exhibited an improved survival both in vitro and in vivo.⁶⁰

These data indicate that MSCs transfected with the bFGF gene implanted together with a carrier represent a powerful gene therapy tool for the repair of the necrotic segment of the bone.

Hepatocyte growth factor

Hepatocyte growth factor (HGF) is a multifunctional factor secreted by MSCs that can nourish and promote repair of various tissues, including the liver, heart, nerves, and bone.⁶¹ HGF exerts its cellular effects by binding to its receptor c-Met. Depending on cell type, HGF induces different biological responses through three distinct transduction pathways: Ras-ERK1/2, p38 MAPK, and PI3K/Akt.⁶² In MSCs, which express c-Met, HGF elicits divergent biological responses in a concentration-dependent manner; a low concentration of HGF (20 ng/mL) preferentially promotes osteogenic differentiation and cell migration but attenuates proliferation, whereas a high concentration of HGF (100 ng/mL) strongly activates cell proliferation.^63,64

Wen et al. ⁶⁵ used adenoviral vectors to express HGF in human BMSCs and found that the concentration of HGF in the culture supernatant reaches 133 ng/mL at 48 h after the infection and then tapers off to 19 ng/mL approximately 2 weeks later. Thus, HGF gene-modified BMSCs could meet the requirement of concentration-dependent effects for the proliferation of MSCs shortly after injury and bone differentiation at a later time interval. To further test the in vivo efficacy of these engineered BMSCs, the cells were transplanted into the CD track of the rabbit models of GIONFH. After 4 weeks, the transplantation of engineered cells generated regularly arranged trabeculae, massively reconstructed blood vessels under the cartilage, and stimulated robust osteogenesis.

In addition, transplantation of the engineered BMSCs compounded within a medical fibrin glue functioning as a supporting biomaterial improved the maintenance of local HGF concentration and favored cell adhesion in the necrotic field, thus improving the microenvironment and achieving significant long-term enhancement of the recovery of ONFH for up to 8 weeks.⁶⁶ In another study performed in mice with ovariectomy-induced osteoporosis, the injection of HGF-modified dental pulp-derived MSCs through the tail vein effectively prevented bone loss.⁶⁷ These results indicate that the HGF gene might represent an ideal target for the engineering of stem cells used for disease-associated bone loss, enhancing the efficacy of cell transplants.

Chemokines

SDF-1, a member of the CXC motif chemokine subfamily, exerts important biological effects by binding to a variety of cell surface CXC receptors (CXCR, e.g., CXCR4 and CXCR7) involved in cellular proliferation, survival, migration, and differentiation.⁶⁸ SDF-1 is a potent stem cell homing factor; the expression of SDF-1 increases in damaged areas, inducing the CXCR4-positive MSCs to migrate and contribute to tissue repair.^69
–71 In a mouse model of femoral bone damage, the administration of CXCR4 antagonists or local intervention with an anti-SDF-1 antibody prevented the accumulation of MSCs in the injured area.⁷² Thus, SDF-1 functions as an essential regulator in endogenous repair.

However, the expression of SDF-1 in ADSCs can be inhibited in vitro by GCs in a concentration-dependent manner, and the concentration of SDF-1 is lower in the plasma and ADSCs isolated from GIONFH patients.⁷³ These results suggest that the decreased SDF-1 expression may be associated with the reduced regenerative ability in GIONFH. Yang and coworkers⁷⁴ enhanced the expression of SDF-1α in human BMSCs by lentiviral gene transfer and found that overexpression of SDF-1α ameliorated the damage to MSCs induced by high-dose GCs; the beneficial effects included tube formation, migration, and osteogenic differentiation ability in vitro.

In the rat model of GIONFH, the injection of 1 × 10⁷ SDF-1α-BMSCs into the bone marrow of tibia markedly enhanced the reconstruction of the necrotic region, resulting in a significant improvement of the trabecular structure and formation of new blood vessels. These findings indicate that MSCs engineered to secrete SDF-1 could provide a new approach to the treatment of ONFH; the therapeutic effect could be achieved by either increasing their own activity or attracting endogenous cells.

Traumatic ONFH

Because of the intrinsically limited blood supply, the femur head is prone to ischemic necrosis. Histologically, osteonecrosis can be detected within 3 days after trauma. The pathologic progression of traumatic ONFH involves four main stages.¹⁰

The first phase is tissue necrosis involving the death of bone marrow cells, capillary endothelial cells, adipocytes, and osteocytes. The second stage is tissue repair, during which endogenous MSCs and endothelial cells migrate to the site of the fracture and contribute to osteogenesis and angiogenesis. The next stage is bone remodeling, which is achieved by the osteoclast-mediated bone resorption and is accompanied by insufficient bone reconstruction; this phase repeatedly produces trabecular microfractures resulting from the mechanical stress exerted on the femur head. The last stage is the development of the impairment of the joint.

Therefore, it is critical to enhance bone repair in the necrotic area before the collapse of the head. At present, the prevailing strategy for genetic engineering of MSCs to be used in the treatment of traumatic ONFH involves their modification with genes coding osteogenic or angiogenic growth factors, alone or in combination.

Bone morphogenetic proteins

BMPs, members of the transforming growth factor–β superfamily, are pleiotropic molecules involved in the formation of bone and cartilage, and cell growth and differentiation in embryonic and adult tissues.⁷⁵ BMP2 exhibits excellent osteogenic potential and has been granted the Food and Drug Administration approval for clinical application in fractures, non-unions, and spinal fusion.^76,77 However, a high dose of BMP2 is necessary when administered systemically, which results in adverse complications, such as ectopic bone formation and inflammatory response.⁷⁸

BMPs-transfected MSCs possess better osteogenic potential than primary MSCs. The expression of BMPs and osteogenic and vascular trophic factors is significantly upregulated in BMPs gene-modified MSCs.^79,80 Tang and coworkers⁸¹ investigated the effectiveness of adenovirus-hBMP2-transduced BMSCs in goats with liquid nitrogen-induced ONFH. Three weeks after the induction of ONFH, the hBMP2 gene-transfected BMSCs and β-TCP were implanted into the tunnel after CD. In comparison with the control group, at 16 weeks after the treatment, this protocol resulted in a higher volume of new bone generated in the macropores of the scaffold and better mechanical properties of the regenerated tissue, including the maximum compressive strength and Young's modulus. Importantly, no adverse effects have been identified throughout the course of the experiment.

In a related study, Katiella and collaborators⁸² found that plasmid-transfected BMSCs overexpressing BMP2 delivered together with magnesium alloy rod effectively prevented experimentally induced rabbit ONFH. Twelve weeks after its insertion, the magnesium alloy was successfully absorbed without apparent visceral damage and all animals in the magnesium rod/BMSCs group were able to run normally. The treatment group maintained the normal contour of the femur head, improved bone mineral density with a higher number of endocytic cells, better arranged trabecular structures, and near-normal cancellous bone at the implantation site. These data suggest that the MSCs transiently expressing BMP2 can improve the repair of traumatic ONFH.

Vascular endothelial growth factor

Circulation impairment in the femur head constitutes direct evidence of the etiology and pathogenesis of traumatic ONFH. VEGF is one of the most important proangiogenic peptides and in the bone environment is secreted mainly by osteoblasts. It enhances vascularization by promoting endothelial cell proliferation and migration, and it affects skeletal development and osteogenesis.⁸³ Of note, implantation of MSCs enhances bone formation but does not significantly increase vascularization in the model of traumatic ONFH.⁸⁴ Thus, to improve the survival of implanted cells and subsequent bone repair, it is critical to develop strategies to facilitate revascularization.

Different isoforms of VEGF resulting from the alternative splicing of VEGF pre-mRNA have been identified in humans. Among them, VEGF-165 is one of the most abundant isoforms exhibiting strong angiogenic potential.⁸⁵ To date, two studies have investigated the therapeutic benefits of VEGF-expressing MSCs for ONFH repair.

In a rabbit model of traumatic ONFH induced by liquid nitrogen, Liu and Zhao⁸⁶ employed recombinant adeno-associated virus (AAV) to induce hVEGF-165 overexpression in MSCs and found that the expression of the exogenous hVEGF-165 gene could be detected 48 h after the transfection and peaked at 10 days. With the help of an arthroscope, the cells were accurately implanted into the bone track where the necrotic bone was emptied. After 8 weeks, hVEGF-165 could be still detected by immunohistochemistry, and more capillaries and new bone were formed in the filling area. The compressive strength of the bone in the hVEGF-165-treated group approached the level measured in normal controls, indicating high biological activity and the bone-regenerating potential of these engineered cells.

Similar conclusions have been reached by Hang and coworkers⁸⁷ in a canine model of traumatic ONFH. The transplantation of autologous hVEGF-165-expressing BMSCs (2 × 10⁷) transfected with hVEGF plasmid using lipofectamine markedly enhanced the isotope uptake measured by three-phase 99mTc-MDP bone scintigraphy and induced the formation of new trabecular bone as early as 2 weeks after implantation. Quantitation of immunofluorescent staining of von Willebrand factor, a specific marker of endothelial cells, showed that the vessel density was significantly increased in the treated group in comparison with animals treated with non-engineered BMSCs and the CD-alone group. Similarly, there was no indication of increased angiogenesis in the non-transfected BMSCs group.

Previous studies demonstrated that GCs downregulate VEGF expression in primary osteoblasts and the loss of VEGF might contribute to the initial stage of GIONFH,⁸⁸ further emphasizing the critical function of VEGF in ONFH treatment and highlighting the potential of this growth factor as a key therapeutic target.

Hepatocyte growth factor

Through a mechanism similar to that functioning in the treatment of GIONFH. HGF-overexpressing BMSCs also contribute to bone regeneration at the early stages of a rabbit model of traumatic ONFH when transplanted 1 week after the trauma. This conclusion is based on the reduction in collagen I expression in the trabeculae, increase in expression of VEGF, and the presence of better organized trabecular bone and new capillaries.⁸⁹

A major safety concern related to the use of MSCs genetically engineered to express HGF is the possibility of the development of sarcomas due to the sustained high expression of HGF in vivo.^90,91 To eliminate the problem of tumorigenesis, Pan and coworkers⁹² successfully developed a gene switch (Tet-on system) allowing the activation of HGF expression in BMSCs by the administration of tetracycline or its derivatives, for example, doxycycline (Dox). The transplantation of autologous BMSCs with HGF expression controlled by Dox resulted in a significant reversal of ONFH. Thus, to avoid the adverse effects of high concentrations of HGF, the ability to control the expression of this growth factor in MSCs must be achieved before its therapeutic use in ONFH patients.

Gene co-modification

As discussed earlier, osteogenesis and angiogenesis are tightly coupled during bone repair and remodeling. Therefore, efforts have been made to functionally co-modify MSCs with genes providing synergic effects.⁹³ At present, two major methods of cell-based gene co-delivery are available. The first is the transplantation of mixed cell populations, each of them transduced with a different gene. The second approach is to transplant cells co-transfected with different genes.

Ma and collaborators⁹⁴ innovatively explored possible synergic effects of VEGF and BMP2 in traumatic ONFH induced by the method of Liu and Zhao.⁸⁶ In these experiments, rabbit BMSCs were co-transfected with hVEGF165 and hBMP2 by AAV, and they were transplanted to the necrotic zone of the femoral head after removing the necrotic tissue under arthroscopy. This protocol resulted in the formation of a large number of new blood vessels and bone in the damaged region. A smooth femoral head surface was obtained, and cystic lesions were not identified by an X-ray examination at 8 weeks. Of note, apparently inconsistent outcomes of in vivo bone formation have been reported with different VEGF-to-BMP ratios.^95,96 Therefore, the optimal BMP2-to-VEGF ratio should be determined in ONFH models.

In addition to its pro-osteogenic role, bFGF acts synergistically with VEGF to stimulate angiogenesis. Based on this notion, Peng and Wang⁹⁷ investigated whether a combination of demineralized bone matrix (DBM) and BMSCs transduced with an adenoviral vector carrying the BMP2 and bFGF genes can promote the repair of ONFH in a canine model. Twelve weeks after the implantation of the DBM seeded with engineered BMSCs into the necrotic femoral head, the newly generated bone area, neovascularization density, and compression and bending strength parameters in the treated animals were superior to those in the control group.

Current Challenges and the Future of Genetically Engineered MSCs for ONFH

Challenges

Early treatment is critical for ONFH patients, and stem cell-based therapies have high application potential. As one of the important approaches to cell therapy, genetic modifications can enhance the benefits of the treatment with MSCs. The preparation of genetically engineered stem cells for therapeutic use includes three main steps: (1) acquisition and propagation of stem cells from various sources; (2) manipulation of the expression of the gene of interest by several available methods; and (3) use of the engineered stem cell for disease treatment (Fig. 1). Although an increasing number of studies have revealed a tremendous regenerative potential of MSC-based gene therapies in several diseases associated with ischemia and tissue damage,^98,99 the clinical application of gene-engineered MSCs for ONFH continues to face many challenges.

Figure 1.

Schematic diagram of the procedure for applying genetically engineered MSCs in the treatment of early-stage ONFH. AT, adipose tissue; BM, bone marrow; CM, conditioned medium; iMSCs, induced pluripotent stem cell-derived mesenchymal stem cells; iPSCs, induced pluripotent stem cells; MSCs, mesenchymal stem cells; ONFH, osteonecrosis of the femoral head; Exos, exosomes. Color images are available online.

The first challenge is the choice of vectors for gene delivery. The viral vectors, for example, retrovirus or lentivirus, enable long-term and consistent transgene expression by integrating the transgene into the host genome but carry the risk of insertional mutagenesis and tumor formation.^100,101 To enhance the safety of this approach, significant effort is focused on the improvement of virus-packaging systems, development of the suicide gene strategy, and pre-detection of insertion mutations before transplantation.^{98,100
–102} Incorporation of a suicide gene to viral vector, which allows destroying the residual delivering cells after the therapeutic factor has been delivered, is highly attractive in cell-based gene therapy.

Non-viral vectors have great potential for gene delivery, particularly due to their safety. However, their low delivery efficiency remains the major problem. Recent advances in material sciences have led to the development of a series of new synthetic polymer- and lipid-based non-viral delivery systems with improved gene transfer efficiency, such as polymers, cationic lipids, and virus-like particles.^102,103 Nevertheless, the non-integration vectors, including non-viral vectors, adenovirus, AAV, baculovirus, and others, provide only transient gene expression. It is also necessary to explore safer and more efficient techniques of genetic manipulation.

The second challenge is the proper selection of the therapeutic gene. Since non-traumatic ONFH and traumatic ONFH have distinct pathology, the therapeutic effects of the transgene should be fully tailored to different types of necrosis.

For non-traumatic ONFH, cell apoptosis and the imbalance between osteogenesis and adipogenesis should be fully considered when selecting potential candidate genes. For example, GC-induced leucine zipper (GILZ), a member of the leucine-zipper family of transcription factors, inhibits GC-induced adipocyte differentiation and its overexpression shifts the balance of osteoblastogenesis and adipogenesis of MSCs toward the osteogenic pathway.^104,105 Oxidative stress is a crucial mediator of cell apoptosis and the induction of adipogenesis in MSCs. NADPH oxidase Nox2 serves as the major source of intracellular reactive oxygen species (ROS); inhibition of Nox2 in stem cells decreases ROS formation and promotes survival in ischemic environment.¹⁰⁶

However, as discussed earlier, for traumatic ONFH, an acceleration of early osteogenesis and angiogenesis is the key to its prevention. Numerous other growth factors or transcription factors have shown this activity, including BMP6/7/9,^107
–109 human platelet-derived growth factor,¹¹⁰ calcitonin gene-related peptide,¹¹¹ Runx2,¹¹² and the Forkhead/Fox transcription factor.¹¹³ It is noteworthy that some adverse effects may result from the uncontrolled expression of the transgenes. For instance, overexpression of HGF may lead to the development of sarcoma and negatively affect osteogenesis and bone formation.^83,90 Uncontrolled synthesis of VEGF might induce angiomas and, in turn, impair osteogenesis.¹¹⁴ Thus, more precise and controllable gene switches must be developed in the future.

The third challenge is the establishment of ideal preclinical animal models. Multiple distinct methods have been developed for creating experimental ONFH models, but the efficacy of an intervention in animals cannot be directly compared with that in humans because the models do not mimic full-range ONFH.¹¹⁵ Last but not least, although gene-engineered MSCs were shown to be safe and effective, the optimal time of cell delivery, the dose of cells, implantation method, distribution in the patient's body, and long-term safety and clinical outcomes remain to be investigated.

Perspectives

Although the problems of gene therapy for ONFH outlined earlier continue to be actively investigated, several strategies to improve the clinical outcomes and the safety of genetically modified stem cells are being explored as well.

Non-coding RNA

In addition to the engineering of MSCs with growth factor-coding genes, non-coding RNA species can be potentially utilized in ONFH therapy. Non-coding RNAs, such as microRNA and long non-coding RNA, are involved in the pathogenesis of ONFH.^41,116,117 These molecules modulate cellular functions by transcriptional regulation of specific genes, and, therefore, can be used in the future as a form of gene therapy for ONFH.

Induced pluripotent stem cell-derived MSCs

MSCs derived from different sources, such as adipose tissue, synovium, placenta, umbilical cord, as well as induced pluripotent stem cells (iPSCs), could become alternative host cells for the transgene. Our laboratory is particularly focused on the application of iPSCs and iPSC-derived MSCs (iMSCs) in stem cell-based bone reconstruction, due to their increasingly recognized strong potential for bone regeneration. By using the classical “OSKM” system (Oct4, Sox2, Klf4, and c-Myc), our team has successfully reprogramed the BMSCs from non-traumatic ONFH patients into iPSCs and then differentiated them into MSCs. The cells generated with this protocol exhibited increased proliferation and osteogenic differentiation potential both in vitro and in vivo (Reprogrammed mesenchymal stem cell-based therapy for the treatment of steroid-associated osteonecrosis of the femoral head in Sprague-Dawley rats. Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, 2016-2019; Unpublished data).

Tissue engineering

Bone tissue engineering provides a suitable extracellular environment for stem cell proliferation, survival, retention, and differentiation. In addition, some synthetic scaffolds also provide mechanical support, preventing the collapse of the weight-bearing area in the femur head. The implantation of MSCs into the necrotic region provides better therapeutic benefit if the cells are loaded on carriers, such as DBM, XACB, fibrin glue, bone-marrow buffy coat, and small intestine submucosa matrix, or on scaffolds, such as poly(lactic-co-glycolic) acid, biphasic calcium phosphate, tantalum rods, and β-TCP.¹¹⁸

Exosomes

Recent evidence documented that the regenerative effect of MSCs may be mediated predominantly by the secretion of bioactive factors, including exosomes, cytokines, chemokines, and growth factors, rather than by the engraftment per se.^119,120 The use of conditioned medium or exosomes produced by gene-engineered MSCs provides an alternative to address the problems of safety.^121

–125 The novel viewpoint that the genetically engineered MSCs can function as a “factory” that can produce a significant amount of multipotent and patient-specific therapeutic product is particularly attractive.

Conclusions

It is our strong conviction that discoveries of efficient new genes, the development of optimal vectors, utilization of gene-editing technology such as CRISPR-Cas systems, the application of novel biomaterials, and in-depth characterization of the biology of MSCs will have an enormous impact on ONFH treatment.

Footnotes

Authors' Contributions

P.S., J.Y.X., and S.N.Y.: conception design, article writing and revision; D.L.S. and Z.X.S.: literature search; S.T.: structure and figure design; Q.P. and C.J.W.: language revision. All authors have given approval to the final version of the article.

Author Disclosure

No competing financial interests exist.

Funding Information

This work was supported by the National Natural and Scientific Foundation of China (no. 81472083); the Independent Innovation Research Fund of Huazhong University of Science and Technology (no. 2017KFYXJJ058).

References

Mont

, Cherian

, Sierra

, et al. Nontraumatic osteonecrosis of the femoral head: where do we stand today? A ten-year update. J Bone Joint Surg Am, 2015; 97:1604–1627.

Zhang

, Chen

, Ai

, et al. Osteonecrosis of the femoral head, nonunion and potential risk factors in Pauwels grade-3 femoral neck fractures: a retrospective cohort study. Medicine (Baltimore), 2016; 95:e3706.

Spence

, Dimauro

, Miller

, et al. Osteonecrosis after femoral neck fractures in children and adolescents: analysis of risk factors. J Pediatr Orthop, 2016; 36:111–116.

Bassett

, Schink-Ascani

, Lewis

. Effects of pulsed electromagnetic fields on Steinberg ratings of femoral head osteonecrosis. Clin Orthop Relat Res, 1989; 246:172–185.

Zalavras

, Lieberman

. Osteonecrosis of the femoral head: evaluation and treatment. J Am Acad Orthop Surg, 2014; 22:455–464.

, Cheng

, Wang

, et al. Herbal Fufang Xian Ling Gu Bao prevents corticosteroid-induced osteonecrosis of the femoral head-A first multicentre, randomised, double-blind, placebo-controlled clinical trial. J Orthop Translat, 2018; 12:36–44.

Yuan

, Guo

, Yan

. The use of bisphosphonate in the treatment of osteonecrosis of the femoral head: a meta-analysis of randomized control trials. Osteoporos Int, 2016; 27:295–299.

Saggini

, Di Stefano

, Saggini

, et al. Clinical application of shock wave therapy in musculoskeletal disorders: part II related to myofascial and nerve apparatus. J Biol Regul Homeost Agents, 2015; 29:771–785.

, Zhang

, Chen

, et al. Effectiveness of various hip preservation treatments for non-traumatic osteonecrosis of the femoral head: a network meta-analysis of randomized controlled trials. J Orthop Sci, 2018; 23:356–364.

10.

Bachiller

, Caballer

, Portal

. Avascular necrosis of the femoral head after femoral neck fracture. Clin Orthop Relat Res, 2002; 399:87–109.

11.

Sadile

, Bernasconi

, Russo

, et al. Core decompression versus other joint preserving treatments for osteonecrosis of the femoral head: a meta-analysis. Br Med Bull, 2016; 118:33–49.

12.

Ravi

, Pincus

, Wasserstein

, et al. Association of overlapping surgery with increased risk for complications following hip surgery: a population-based, matched cohort study. JAMA Intern Med, 2017; 178:75–83.

13.

Smith

, Dieppe

, Howard

, et al. Failure rates of metal-on-metal hip resurfacings: analysis of data from the National Joint Registry for England and Wales. Lancet, 2012; 380:1759–1766.

14.

Bayliss

, Culliford

, Monk

, et al. The effect of patient age at intervention on risk of implant revision after total replacement of the hip or knee: a population-based cohort study. Lancet, 2017; 389:1424–1430.

15.

Salem

, Thiemermann

. Mesenchymal stromal cells: current understanding and clinical status. Stem Cells, 2010; 28:585–596.

16.

Gimble

, Katz

, Bunnell

. Adipose-derived stem cells for regenerative medicine. Circ Res, 2007; 100:1249–1260.

17.

Balsam

, Wagers

, Christensen

, et al. Haematopoietic stem cells adopt mature haematopoietic fates in ischaemic myocardium. Nature, 2004; 428:668–673.

18.

Zhang

, Methot

, Poppa

, et al. Cardiomyocyte grafting for cardiac repair: graft cell death and anti-death strategies. J Mol Cell Cardiol, 2001; 33:907–921.

19.

Evans

. Gene delivery to bone. Adv Drug Deliv Rev, 2012; 64:1331–1340.

20.

Raisin

, Belamie

, Morille

. Non-viral gene activated matrices for mesenchymal stem cells based tissue engineering of bone and cartilage. Biomaterials, 2016; 104:223–237.

21.

Murata

, Kumagai

, Miyata

, et al. Osteonecrosis in stroke-prone spontaneously hypertensive rats: effect of glucocorticoid. J Orthop Sci, 2007; 12:289–295.

22.

Wang

, Li

, Mao

, et al. Alcohol-induced adipogenesis in bone and marrow: a possible mechanism for osteonecrosis. Clin Orthop Relat Res, 2003; 410:213–224.

23.

, Feng

, He

, et al. IL-4 administration exerts preventive effects via suppression of underlying inflammation and TNF-alpha-induced apoptosis in steroid-induced osteonecrosis. Osteoporos Int, 2016; 27:1827–1837.

24.

Fan

, Li

, Yu

, et al. Hypoxia-inducible factor prolyl hydroxylase inhibitor prevents steroid-associated osteonecrosis of the femoral head in rabbits by promoting angiogenesis and inhibiting apoptosis. PLoS One, 2014; 9:e107774.

25.

Chen

, Yang

, Feng

, et al. Impairment of two types of circulating endothelial progenitor cells in patients with glucocorticoid-induced avascular osteonecrosis of the femoral head. Joint Bone Spine, 2013; 80:70–76.

26.

Calder

, Buttery

, Revell

, et al. Apoptosis—a significant cause of bone cell death in osteonecrosis of the femoral head. J Bone Joint Surg Br, 2004; 86:1209–1213.

27.

Hao

, Yang

, Xu

, et al. MiR-708 promotes steroid-induced osteonecrosis of femoral head, suppresses osteogenic differentiation by targeting SMAD3. Sci Rep, 2016; 6:22599.

28.

Sun

, Yang

, Ye

, et al. Aberrant CpG islands' hypermethylation of ABCB1 in mesenchymal stem cells of patients with steroid-associated osteonecrosis. J Rheumatol, 2013; 40:1913–1920.

29.

Wang

, Cui

, Balian

. The Nicolas Andry award. The pathogenesis and prevention of steroid-induced osteonecrosis. Clin Orthop Relat Res, 2000; 370:295–310.

30.

Pariante

. The role of multi-drug resistance p-glycoprotein in glucocorticoid function: studies in animals and relevance in humans. Eur J Pharmacol, 2008; 583:263–271.

31.

, Zhao

, Wang

. ABCB1 gene polymorphisms and glucocorticoid-induced avascular necrosis of the femoral head susceptibility: a meta-analysis. Med Sci Monit, 2014; 20:2811–2816.

32.

Han

, Yan

, Guo

, et al. Effects of p-glycoprotein on steroid-induced osteonecrosis of the femoral head. Calcif Tissue Int, 2010; 87:246–253.

33.

Han

, Li

, Cai

, et al. P-glycoprotein overexpression in bone marrow-derived multipotent stromal cells decreases the risk of steroid-induced osteonecrosis in the femoral head. J Cell Mol Med, 2016; 20:2173–2182.

34.

Baron

, Kneissel

. WNT signaling in bone homeostasis and disease: from human mutations to treatments. Nat Med, 2013; 19:179–192.

35.

, Jiao

, Liu

, et al. Hypermethylation of Frizzled1 is associated with Wnt/beta-catenin signaling inactivation in mesenchymal stem cells of patients with steroid-associated osteonecrosis. Exp Mol Med, 2019; 51:23.

36.

Pinzone

, Hall

, Thudi

, et al. The role of Dickkopf-1 in bone development, homeostasis, and disease. Blood, 2009; 113:517–525.

37.

Qiang

, Barlogie

, Rudikoff

, et al. Dkk1-induced inhibition of Wnt signaling in osteoblast differentiation is an underlying mechanism of bone loss in multiple myeloma. Bone, 2008; 42:669–680.

38.

, Sarosi

, Cattley

, et al. Dkk1-mediated inhibition of Wnt signaling in bone results in osteopenia. Bone, 2006; 39:754–766.

39.

Colditz

, Thiele

, Baschant

, et al. Osteogenic Dkk1 mediates glucocorticoid-induced but not arthritis-induced bone loss. J Bone Miner Res, 2019; 34:1314–1323.

40.

, Wang

, et al. Increased Dickkopf-1 expression accelerates bone cell apoptosis in femoral head osteonecrosis. Bone, 2010; 46:584–591.

41.

Dai

, Jin

, Zheng

, et al. MiR-217 promotes cell proliferation and osteogenic differentiation of BMSCs by targeting DKK1 in steroid-associated osteonecrosis. Biomed Pharmacother, 2019; 109:1112–1119.

42.

Zhun

, Donghai

, Zhouyuan

, et al. Efficiency of cell therapy to GC-induced ONFH: BMSCs with Dkk-1 interference is not superior to unmodified BMSCs. Stem Cells Int, 2018; 2018:1340252.

43.

Dai

, Xu

, Wang

, et al. HIF-1alpha induced-VEGF overexpression in bone marrow stem cells protects cardiomyocytes against ischemia. J Mol Cell Cardiol, 2007; 42:1036–1044.

44.

Riddle

, Khatri

, Schipani

, et al. Role of hypoxia-inducible factor-1alpha in angiogenic-osteogenic coupling. J Mol Med (Berl), 2009; 87:583–590.

45.

Ceradini

, Kulkarni

, Callaghan

, et al. Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med, 2004; 10:858–864.

46.

Kelly

, Hackett

, Hirota

, et al. Cell type-specific regulation of angiogenic growth factor gene expression and induction of angiogenesis in nonischemic tissue by a constitutively active form of hypoxia-inducible factor 1. Circ Res, 2003; 93:1074–1081.

47.

Zhang

, Yuan

, Pei

, et al. In vivo and in vitro characteristic of HIF-1alpha and relative genes in ischemic femoral head necrosis. Int J Clin Exp Pathol, 2015; 8:7210–7216.

48.

Zhang

, Yang

, Cornelia

, et al. Hypoxia-inducible factor-1 is a positive regulator of Sox9 activity in femoral head osteonecrosis. Bone, 2011; 48:507–513.

49.

Ding

, Gao

, Hu

, et al. HIF-1alpha transgenic bone marrow cells can promote tissue repair in cases of corticosteroid-induced osteonecrosis of the femoral head in rabbits. PLoS One, 2013; 8:e63628.

50.

, Liu

, Li

, et al. Exosomes secreted from mutant-HIF-1alpha-modified bone-marrow-derived mesenchymal stem cells attenuate early steroid-induced avascular necrosis of femoral head in rabbit. Cell Biol Int, 2017; 41:1379–1390.

51.

Montero

, Okada

, Tomita

, et al. Disruption of the fibroblast growth factor-2 gene results in decreased bone mass and bone formation. J Clin Invest, 2000; 105:1085–1093.

52.

D'mello

, Elangovan

, Salem

. FGF2 gene activated matrices promote proliferation of bone marrow stromal cells. Arch Oral Biol, 2015; 60:1742–1749.

53.

Xiao

, Sobue

, Esliger

, et al. Disruption of the Fgf2 gene activates the adipogenic and suppresses the osteogenic program in mesenchymal marrow stromal stem cells. Bone, 2010; 47:360–370.

54.

Saadeh

, Mehrara

, Steinbrech

, et al. Mechanisms of fibroblast growth factor-2 modulation of vascular endothelial growth factor expression by osteoblastic cells. Endocrinology, 2000; 141:2075–2083.

55.

Zhang

, Kot

, Lay

, et al. Acceleration of fracture healing by overexpression of basic fibroblast growth factor in the mesenchymal stromal cells. Stem Cells Transl Med, 2017; 6:1880–1893.

56.

Guo

, Zheng

, Kulbatski

, et al. Bone regeneration with active angiogenesis by basic fibroblast growth factor gene transfected mesenchymal stem cells seeded on porous beta-TCP ceramic scaffolds. Biomed Mater, 2006; 1:93–99.

57.

Peng

, Zhang

, et al. Expression of osteoprotegerin and receptor activator for the nuclear factor-κB ligand in XACB/LV-bFGF/MSCs transplantation for repair of rabbit femoral head defect necrosis. J Cell Biochem, 2019; 120:1427–1434.

58.

Peng

, Dong

, Zhang

, et al. Effects of transplantation of FGF-2-transfected MSCs and XACB on TNF-alpha expression with avascular necrosis of the femoral head in rabbits. Biosci Rep, 2019; 39:BSR20180765.

59.

Zhang

, Peng

, Wang

, et al. Role of FGF-2 transfected bone marrow mesenchymal stem cells in engineered bone tissue for repair of avascular necrosis of femoral head in rabbits. Cell Physiol Biochem, 2018; 48:773–784.

60.

Song

, Kwon

, Lim

, et al. Transfection of mesenchymal stem cells with the FGF-2 gene improves their survival under hypoxic conditions. Mol Cells, 2005; 19:402–407.

61.

Wang

, Wang

, Zhang

, et al. Hepatocyte growth factor gene therapy for ischemic diseases. Hum Gene Ther, 2018; 29:413–423.

62.

Aenlle

, Curtis

, Roos

, et al. Hepatocyte growth factor and p38 promote osteogenic differentiation of human mesenchymal stem cells. Mol Endocrinol, 2014; 28:722–730.

63.

Forte

, Minieri

, Cossa

, et al. Hepatocyte growth factor effects on mesenchymal stem cells: proliferation, migration, and differentiation. Stem Cells, 2006; 24:23–33.

64.

Wen

, Zhou

, et al. Change in hepatocyte growth factor concentration promote mesenchymal stem cell-mediated osteogenic regeneration. J Cell Mol Med, 2012; 16:1260–1273.

65.

Wen

, Ma

, Chen

, et al. Treatment of avascular necrosis of the femoral head by hepatocyte growth factor-transgenic bone marrow stromal stem cells. Gene Ther, 2008; 15:1523–1535.

66.

Wen

, Zhou

, Luo

, et al. Pro-osteogenic effects of fibrin glue in treatment of avascular necrosis of the femoral head in vivo by hepatocyte growth factor-transgenic mesenchymal stem cells. J Transl Med, 2014; 12:114.

67.

Kong

, Shi

, Xiao

, et al. Transplantation of hepatocyte growth factor-modified dental pulp stem cells prevents bone loss in the early phase of ovariectomy-induced osteoporosis. Hum Gene Ther, 2018; 29:271–282.

68.

Kortesidis

, Zannettino

, Isenmann

, et al. Stromal-derived factor-1 promotes the growth, survival, and development of human bone marrow stromal stem cells. Blood, 2005; 105:3793–3801.

69.

Chin

, Furukawa

, Kurotaki

, et al. Facilitation of chemotaxis activity of mesenchymal stem cells via stromal cell-derived factor-1 and its receptor may promote ectopic ossification of human spinal ligaments. J Pharmacol Exp Ther, 2019; 369:1–8.

70.

Wang

, Sun

, Lv

, et al. Stromal cell-derived factor-1 accelerates cartilage defect repairing by recruiting bone marrow mesenchymal stem cells and promoting chondrogenic differentiation. Tissue Eng Part A, 2017; 23:1160–1168.

71.

Xiao Ling

, Peng

, Jian Feng

, et al. Stromal derived factor-1/CXCR4 axis involved in bone marrow mesenchymal stem cells recruitment to injured liver. Stem Cells Int, 2016; 2016:8906945.

72.

Okada

, Kawao

, Yano

, et al. Stromal cell-derived factor-1 mediates changes of bone marrow stem cells during the bone repair process. Am J Physiol Endocrinol Metab, 2016; 310:E15–E23.

73.

Kato

, Khanh

, Sato

, et al. SDF-1 improves wound healing ability of glucocorticoid-treated adipose tissue-derived mesenchymal stem cells. Biochem Biophys Res Commun, 2017; 493:1010–1017.

74.

Yang

, Xue

, Guan

, et al. Stromal-cell-derived factor (SDF) 1-alpha overexpression promotes bone regeneration by osteogenesis and angiogenesis in osteonecrosis of the femoral head. Cell Physiol Biochem, 2018; 46:2561–2575.

75.

Salazar

, Gamer

, Rosen

. BMP signalling in skeletal development, disease and repair. Nat Rev Endocrinol, 2016; 12:203–221.

76.

Ronga

, Fagetti

, Canton

, et al. Clinical applications of growth factors in bone injuries: experience with BMPs. Injury, 2013; 44 Suppl 1:S34–S39.

77.

Krishnakumar

, Roffi

, Reale

, et al. Clinical application of bone morphogenetic proteins for bone healing: a systematic review. Int Orthop, 2017; 41:1073–1083.

78.

Shields

, Raque

, Glassman

, et al. Adverse effects associated with high-dose recombinant human bone morphogenetic protein-2 use in anterior cervical spine fusion. Spine (Phila Pa 1976), 2006; 31:542–547.

79.

Hsieh

, Wu

, Chen

, et al. BMP-2 gene transfection of bone marrow stromal cells to induce osteoblastic differentiation in a rat calvarial defect model. Mater Sci Eng C Mater Biol Appl, 2018; 91:806–816.

80.

Kim

, Kang

, Kim

, et al. Evaluation of mesenchymal stem cell sheets overexpressing BMP-7 in canine critical-sized bone defects. Int J Mol Sci, 2018; 19:2073.

81.

Tang

, Lu

, Yue

, et al. Treatment of osteonecrosis of the femoral head with hBMP-2-gene-modified tissue-engineered bone in goats. J Bone Joint Surg Br, 2007; 89:127–129.

82.

Katiella

, Yanru

, Hui

. Magnesium alloy transfected BMSCs-BMP-2 composite in repair of femoral head necrosis with assessment of visceral organs. Springerplus, 2016; 5:1857.

83.

, Olsen

. The roles of vascular endothelial growth factor in bone repair and regeneration. Bone, 2016; 91:30–38.

84.

Yan

, Hang

, Guo

, et al. Fate of mesenchymal stem cells transplanted to osteonecrosis of femoral head. J Orthop Res, 2009; 27:442–446.

85.

Phillips

, Stone

, Jones

, et al. Vascular endothelial growth factor (rhVEGF165) stimulates direct angiogenesis in the rabbit cornea. In Vivo, 1994; 8:961–965.

86.

Liu

, Zhao

. Treatment for osteonecrosis of femoral head by hVEGF-165 gene modified marrow stromal stem cells under arthroscope. Zhonghua Yi Xue Za Zhi, 2009; 89:2629–2633.

87.

Hang

, Wang

, Guo

, et al. Treatment of osteonecrosis of the femoral head with VEGF165 transgenic bone marrow mesenchymal stem cells in mongrel dogs. Cells Tissues Organs, 2012; 195:495–506.

88.

Varoga

, Drescher

, Pufe

, et al. Differential expression of vascular endothelial growth factor in glucocorticoid-related osteonecrosis of the femoral head. Clin Orthop Relat Res, 2009; 467:3273–3282.

89.

Wen

, Jin

, Zhou

, et al. HGF-transgenic MSCs can improve the effects of tissue self-repair in a rabbit model of traumatic osteonecrosis of the femoral head. PLoS One, 2012; 7:e37503.

90.

Maier

, Mariotti

, Albini

, et al. Over-expression of hepatocyte growth factor in human Kaposi's sarcoma. Int J Cancer, 1996; 65:168–172.

91.

Cortner

, Vande Woude

, Rong

. The Met-HGF/SF autocrine signaling mechanism is involved in sarcomagenesis. EXS, 1995; 74:89–121.

92.

Pan

, Zhang

, Cheng

, et al. Treatment of femoral head necrosis with bone marrow mesenchymal stem cells expressing inducible hepatocyte growth factor. Am J Ther, 2016; 23:e1602–e1611.

93.

, Wang

, Qiu

, et al. Synergistic effects of vascular endothelial growth factor on bone morphogenetic proteins induced bone formation in vivo: influencing factors and future research directions. Biomed Res Int, 2016; 2016:2869572.

94.

, Cui

, Zhao

. Vascular endothelial growth factor/bone morphogenetic protein-2 bone marrow combined modification of the mesenchymal stem cells to repair the avascular necrosis of the femoral head. Int J Clin Exp Med, 2015; 8:15528–15534.

95.

Peng

, Usas

, Olshanski

, et al. VEGF improves, whereas sFlt1 inhibits, BMP2-induced bone formation and bone healing through modulation of angiogenesis. J Bone Miner Res, 2005; 20:2017–2027.

96.

Lee

, Ko

, Kim

, et al. Osteogenesis and angiogenesis are simultaneously enhanced in BMP2-/VEGF-transfected adipose stem cells through activation of the YAP/TAZ signaling pathway. Biomater Sci, 2019; 7:4588–4602.

97.

Peng

, Wang

. Adenovirus-mediated expression of BMP-2 and BFGF in bone marrow mesenchymal stem cells combined with demineralized bone matrix for repair of femoral head osteonecrosis in beagle dogs. Cell Physiol Biochem, 2017; 43:1648–1662.

98.

Wei

, Huang

, Li

, et al. Improved therapeutic potential of MSCs by genetic modification. Gene Ther, 2018; 25:538–547.

99.

Shapiro

, Lieber

, Gazit

, et al. Recent advances and future of gene therapy for bone regeneration. Curr Osteoporos Rep, 2018; 16:504–511.

100.

Oggu

, Sasikumar

, Reddy

, et al. Gene delivery approaches for mesenchymal stem cell therapy: strategies to increase efficiency and specificity. Stem Cell Rev, 2017; 13:725–740.

101.

Hacein-Bey-Abina

, Von Kalle

, Schmidt

, et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science, 2003; 302:415–419.

102.

Salameh

, Zhou

, Ward

, et al. Polymer-mediated gene therapy: recent advances and merging of delivery techniques. Wiley Interdiscip Rev Nanomed Nanobiotechnol, 2019:e1598.

103.

Patil

, Gao

, Lin

, et al. The development of functional non-viral vectors for gene delivery. Int J Mol Sci, 2019; 20:5491.

104.

Pan

, Cao

, Yang

, et al. Role of glucocorticoid-induced leucine zipper (GILZ) in bone acquisition. J Biol Chem, 2014; 289:19373–19382.

105.

Shi

, Shi

, Li

, et al. A glucocorticoid-induced leucine-zipper protein, GILZ, inhibits adipogenesis of mesenchymal cells. EMBO Rep, 2003; 4:374–380.

106.

Feng

, Zhang

, Ding

, et al. Blocking Nox2 improves mesenchymal stem cells therapy in myocardial infarction via antagonizing oxidant and promoting survival. J Cell Physiol, 2018; 233:7004–7015.

107.

Liao

, Zhong

, Liu

, et al. Bone mesenchymal stem cells co-expressing VEGF and BMP-6 genes to combat avascular necrosis of the femoral head. Exp Ther Med, 2018; 15:954–962.

108.

Zhang

, Wang

, Qiang

, et al. Angiopoiesis and bone regeneration via co-expression of the hVEGF and hBMP genes from an adeno-associated viral vector in vitro and in vivo. Acta Pharmacol Sin, 2010; 31:821–830.

109.

Fujioka-Kobayashi

, Sawada

, Kobayashi

, et al. Recombinant human bone morphogenetic protein 9 (rhBMP9) induced osteoblastic behavior on a collagen membrane compared with rhBMP2. J Periodontol, 2016; 87:e101–e107.

110.

Zhang

, Yu

, Niibe

, et al. The effects of platelet-derived growth factor-BB on bone marrow stromal cell-mediated vascularized bone regeneration. Stem Cells Int, 2018; 2018:3272098.

111.

, Liu

, Chen

, et al. CGRP gene-modified rBMSCs show better osteogenic differentiation capacity in vitro. J Mol Histol, 2018; 49:357–367.

112.

Thiagarajan

, Abu-Awwad

HAM

, Dixon

. Osteogenic programming of human mesenchymal stem cells with highly efficient intracellular delivery of RUNX2. Stem Cells Transl Med, 2017; 6:2146–2159.

113.

You

, Gao

, Fan

, et al. Foxc2 regulates osteogenesis and angiogenesis of bone marrow mesenchymal stem cells. BMC Musculoskelet Disord, 2013; 14:199.

114.

Zhang

, Sheng

, He

, et al. Continuous occurrence of both insufficient neovascularization and elevated vascular permeability in rabbit proximal femur during inadequate repair of steroid-associated osteonecrotic lesions. Arthritis Rheum, 2009; 60:2966–2977.

115.

Fan

, Peng

, Qin

, et al. Experimental animal models of osteonecrosis. Rheumatol Int, 2011; 31:983–994.

116.

Zha

, Sun

, Zhang

, et al. Regulatory effect of microRNA-34a on osteogenesis and angiogenesis in glucocorticoid-induced osteonecrosis of the femoral head. J Orthop Res, 2018; 36:417–424.

117.

Yang

, Jia

, Li

, et al. Long noncoding RNAs: new players in the osteogenic differentiation of bone marrow- and adipose-derived mesenchymal stem cells. Stem Cell Rev, 2018; 14:297–308.

118.

Wang

, Ma

, Chai

, et al. Multiscale stem cell technologies for osteonecrosis of the femoral head. Stem Cells Int, 2019; 2019:8914569.

119.

Ratajczak

, Jadczyk

, Pedziwiatr

, et al. New advances in stem cell research: practical implications for regenerative medicine. Pol Arch Med Wewn, 2014; 124:417–426.

120.

Liang

, Ding

, Zhang

, et al. Paracrine mechanisms of mesenchymal stem cell-based therapy: current status and perspectives. Cell Transplant, 2014; 23:1045–1059.

121.

Liao

, Ning

, Xu

, et al. BMSC-derived exosomes carrying microRNA-122-5p promote proliferation of osteoblasts in osteonecrosis of the femoral head. Clin Sci (Lond), 2019; 133:1955–1975.

122.

Liu

, Li

, Niu

, et al. Exosomes secreted from human-induced pluripotent stem cell-derived mesenchymal stem cells prevent osteonecrosis of the femoral head by promoting angiogenesis. Int J Biol Sci, 2017; 13:232–244.

123.

Choi

, Jeong

, Han

, et al. IL-10-secreting human MSCs generated by TALEN gene editing ameliorate liver fibrosis through enhanced anti-fibrotic activity. Biomater Sci, 2019; 7:1078–1087.

124.

Gong

, Liu

, Wang

, et al. Exosomes derived from SDF1-overexpressing mesenchymal stem cells inhibit ischemic myocardial cell apoptosis and promote cardiac endothelial microvascular regeneration in mice with myocardial infarction. J Cell Physiol, 2019; 234:13878–13893.

125.

, Lv

, Chen

, et al. Vascular endothelial growth factor over-expressed mesenchymal stem cells-conditioned media ameliorate palmitate-induced diabetic endothelial dysfunction through PI-3K/AKT/m-TOR/eNOS and p38/MAPK signaling pathway. Biomed Pharmacother, 2018; 106:491–498.