MSCs are characterized by their ability to self-renew and to differentiate into cells of the mesenchymal lineage, including adipocytes, osteoblasts, chondrocytes, tenocytes, skeletal myocytes, and cells of the visceral mesoderm [2,10,11]. In addition, some studies suggested that the differentiation potential of MSCs extends beyond the conventional mesodermal lineage and that they can also differentiate into cells of ectodermal and endodermal origin, such as hepatocytes [12,13], neurons [14,15], and cardiomyocytes [16,17]. The multilineage differential potential of MSCs is commonly examined by in vitro functional assays using specific differentiation media, and these in vitro data encouraged further investigation of MSCs as a potential source of tissue repair. However, due to the lack of specific MSC markers, there is little information on the in vivo differentiation of MSCs, as compared to in vitro characterization. Studies have suggested MSC engraftment and transdifferentiation in vivo in various models of damaged or mutated bone, cartilage [18], myocardial [19,20], neural [21,22], and hepatic tissues [13], but whether the observed therapeutic effects are due to paracrine interactions or true differentiation capacity remains to be elucidated. In one study, MSCs labeled with green fluorescent protein (GFP) were injected intravenously and examined for engraftment and differentiation potential [23]. GFP-labeled MSCs were initially located in the lungs and, subsequently, MSCs were detected in other tissues at low frequencies, such as bronchiolar epithelial cells, hepatocytes, and renal tubular cells. Importantly, there was no evidence of clonal expansion and the mechanism of differentiation was not determined, suggesting that the observation of MSCs in various tissues could have been due to simple fusion events. Overall, the therapeutic potential of MSCs has been observed in various injury models, but in vivo data supporting the true differentiation and regenerative potential of MSCs are still lacking.

MSCs have significant clinical implications as they exert potent immunosuppressive and anti-inflammatory effects through the interactions between the lymphocytes associated with both the innate and adaptive immune systems. MSCs suppress T cell proliferation [24-26], B cell functions [25,27,28], natural killer cell proliferation and cytokine production [29], and prevent the differentiation, maturation, and activation of dendritic cells [30-37]. Importantly, MSCs can suppress cells independently of the major histocompatibility complex (MHC) identity between donor and recipient due to their low expression of MHC-II and other costimulatory molecules [38]. While MSCs can exert immunosuppressive effects by direct cell to cell contact, their primary mechanism is production of soluble factors, including transforming growth factor-ฮฒ [39], hepatocyte growth factor (HGF) [26], nitric oxide [40], and indoleamine 2,3-dioxygenase (IDO) [41]. Furthermore, through cell to cell contact and the production of soluble factors, MSCs induce an immunosuppressive environment by generating regulatory T cells (Tregs). The ability of MSCs to induce Tregs has been observed both in vitro [42,43] and in vivo in various models [44-49]. In addition, MSCs can induce plasmacytoid dendritic cells to produce interleukin (IL)-10 [50], which may also support the development of Tregs in vivo. These observations suggest that MSCs are key regulators of immune modulation by directly suppressing activated immune cells and indirectly recruiting Tregs.

However, MSCs are not constitutively inhibitory. MSCs are highly dependent on environmental inflammatory conditions. Under acute inflammatory conditions polarized by M1 macrophages and helper T lymphocyte (Th)-type-1 cytokines, especially the proinflammatory cytokine interferon (IFN)-ฮณ, the immunosuppressive capacity of MSCs is enhanced through increased production of ICAM-1, CXCL-10, CCL-8, and IDO [51-53]. On the other hand, under chronic inflammatory conditions when MSCs are polarized by M2 macrophages and Th2 cytokines, MSCs can be recruited into the fibrotic process [51]. Thus, the therapeutic effects of MSCs depend on the inflammatory microenvironment, which should be taken into consideration when used for therapy.

A number of studies have suggested that MSCs have the capacity to migrate to sites of inflammation and tumor microenvironments. Although the exact mechanisms underlying MSC migration remain to be elucidated, studies have shown that MSC migration is dependent on various chemokine and receptor interactions, such as stromal cell-derived factor 1 (SDF-1)/C-X-C chemokine receptor type 4 (CXCR4) [54,55], stem cell factor/c-kit, HGF/c-Met [56], vascular endothelial growth factor (VEGF)/VEGF receptor [57], platelet-derived growth factor (PDGF)/PDGF receptor [54,58], monocyte chemoattractant protein-1 (MCP-1)/C-C chemokine receptor type 2 [59], and high mobility group box 1/receptor for advanced glycation endproducts [60,61] as well as other cell adhesion molecules [55,62]. These cytokine and chemokine receptor pairs play important roles in leukocytes that respond to injury and inflammation or hematopoietic stem cells (HSC) and are thought to function similarly in MSCs. Furthermore, the tumor microenvironment closely resembles an unhealed wound that continuously produces inflammatory mediators, including cytokines, chemokines, and other chemoattractant molecules [63]. This constant inflammatory signaling may become a target for MSC migration. Among the chemokine receptor pairs, SDF-1 and CXCR4 are important mediators of stem cell recruitment to tumors [54]. In addition, many tumor microenvironments exhibit hypoxia that results in expression of proangiogenic molecules. The hypoxia-induced transcription factor HIF-1ฮฑ activates the transcription of genes, including VEGF, macrophage migration inhibitor factor, tumor necrosis factors, and numerous proinflammatory cytokines [64], inducing the generation of chemokines, such as MCP-1, involved in migration of MSCs toward tumors [59]. Many different chemokine factors and receptors have been implicated in the migration of MSCs and further studies that exploit additional chemokine/receptor interactions are needed to develop targeted MSC therapies to inflammatory and tumor sites.

The ability of MSCs to differentiate into osteoblasts, tenocytes, and chondrocytes has attracted interest for their use in orthopedic settings. First, MSCs have been shown to be beneficial in treating bone disorders, such as osteogenesis imperfecta (OI) and hypophosphatasia. OI is characterized by skeletal fragility and connective tissue alterations caused by alteration of type I collagen production by osteoblasts. Pediatric patients with OI underwent allogeneic hematopoietic stem cell transplantation (HSCT) and the transplanted bone marrow cells engrafted and generated functional osteoblasts leading to improvement in bone structure and function [65]. Although, only a low level of engraftment was achieved, a follow-up study demonstrated continued improvements in patients for 18 to 36 months posttransplantation [66]. It is important to note that these patients were transplanted with whole bone marrow instead of MSCs alone. In another follow-up study, patients who received HSCT were infused with the same donor MSCs [67]. The additional infusion of MSCs showed further benefit, but this was limited in duration. Furthermore, a fetus diagnosed with severe OI underwent in utero MSC transplantation [68]. After birth, psychomotor development and growth were normal. Hypophosphatasia is a genetic disorder of mesenchymal origin with mutation in tissue nonspecific alkaline phosphatase. Although the numbers of clinical studies are limited, pediatric patients who received BMT showed significant clinical improvements [69,70]. Administration of MSCs alone in hypophosphatasia has not yet been studied; some authors have suggested that cultured MSCs may fail to engraft after intravenous infusion due to loss of adhesion molecules and loss of self-renewal ability [71,72]. However, even patients receiving whole bone marrow did not reveal significant donor MSC engraftment despite clinical improvements [69].

Similar to studies on genetic bone disorders, there have been limited reports demonstrating the efficacy of MSCs in promoting cartilage repair in which MSCs embedded in collagen gel were transplanted into the knee joints of patients with articular cartilage defects [73-76]. MSC transplantation has been shown to produce significant clinical improvements with cartilage repair; however, the mechanisms underlying cartilage regeneration are still unknown. The transplanted MSCs may have differentiated into chondrocytes, but it is also possible that MSCs produce soluble factors to induce other cells of the microenvironment to differentiate into cartilage.

HSCT has been widely used over the past several decades to treat patients with various malignant and nonmalignant diseases. However, the procedure remains complicated by regimen-related toxicity, engraftment failure, and GVHD [77]. Preconditioning regimens, such as chemotherapy and/or radiotherapy, may damage the bone marrow and lead to a diminished engraftment of stem cells. MSCs are an attractive therapeutic approach during or after transplantation as their transplantation can minimize the toxicity of the conditioning regimens while inducing hematopoietic engraftment and decrease the incidence and severity of GHVD. In several studies, MSCs were cotransplanted with HSCs to facilitate engraftment [78-83] but their efficacy remains unclear. Similarly, the infusion of third-party haploidentical MSCs during pediatric umbilical cord blood transplantation was shown to induce prompt hematopoietic recovery [83]. On the other hand, some studies have suggested that cotransplantation of MSCs does not affect the kinetics of engraftment [82]. While there have been no trials of MSCs for hematopoiesis, the best studied therapeutic application of MSC is GVHD.

GVHD is a severe inflammatory condition that results from immune-mediated attack of recipient tissues by donor T cells during BMT. The clinical efficacy of MSCs in acute GVHD (aGVHD) was first observed in a 9-year-old boy with steroid-resistant grade IV aGVHD [84]. The patient, who was unresponsive to other therapies, showed a complete response after receiving haploidentical third-party MSCs. Following this pilot study, MSC treatment has been studied extensively in steroid-refractory GVHD [84-92]. In 2006, six of eight patients with steroid-resistant grade III to IV GVHD showed complete remission to MSC treatment [90]. The European Group for Blood and Marrow Transplantation then led a multicenter phase II study in which both pediatric and adult patients with steroid-resistant GVHD were treated with MSCs derived from various sources, including HLA-identical and haploidentical sibling donor bone marrow or third-party mismatched donor bone marrow [86]. Sixty-eight percent of these patients showed complete responses with a significantly reduced transplantation-related mortality rate. Not only did this multicenter study confirm that MSCs are a powerful therapeutic tool it also reduced concerns regarding HLA disparity between the MSC donor and recipient through extensive use of third-party-derived MSCs. Based on these properties, MSCs have been further developed into an FDA-approved commercialized "off-the-shelf" product known as Prochymal (Osiris Therapeutics Inc., Columbia, MD, USA), which is derived from the bone marrow of healthy adult donors [93]. Prochymal was used in a randomized prospective study to treat patients directly after diagnosis of GVHD [94]. Ninety-four percent of the patients had an initial response and showed no infusional toxicities or ectopic tissue formation. In a multicenter trial, a higher response rate was seen in children (84%), as compared to adults (60%) [86]. Therefore, Prochymal was used to specifically treat pediatric patients less than 18 years old with severe steroid-resistant grade III and IV aGVHD [89]. Overall, seven of 12 patients showed complete responses suggesting that pediatric patients may respond better to MSC treatment.

While studies on the use of MSCs for treatment of aGVHD have yielded promising results, the therapeutic efficacy of MSCs in chronic GVHD (cGVHD) is less clear because of the paucity of studies. While some studies indicated efficacy of MSCs, even in cGVHD [93], others suggested that MSCs are less effective in cGVHD than aGVHD [87,88,95]. In studies of MSC therapy in both aGVHD and cGVHD patients, the response rates were higher in aGVHD than cGVHD patients [96]. In addition, the infusion of MSCs following HSCT could prevent the development of aGVHD, while the development cGVHD remained unaffected. Thus, specific patient recruitment and study designs may allow critical analysis of the effects of MSC treatment in GVHD patients in the future.

Despite improvements in medical and surgical therapies, heart disease and heart failure continue to show high morbidity and mortality rates. MSC therapy is an attractive candidate for cardiovascular repair due to its regenerative and immunomodulatory properties. In preclinical studies, MSCs were shown to engraft and improve cardiac repair after administration [97-99]. Clinical trials using MSCs to improve cardiac function have also yielded encouraging results. In a pilot study, 69 patients with acute myocardial infarction received percutaneous coronary injection, and were randomized to receive intracoronary injection of autologous MSCs or standard saline as controls [100]. There were no serious adverse events following MSC administration and the MSC-treated group showed significant improvements in cardiac function, as compared to the control group. This was also the first study to follow and detect the viability of MSCs and cardiac function with cardiac electromechanical mapping. The results indicated that MSCs were still viable 3 months after transplantation. Following this study, MSCs have been used to treat acute and chronic myocardial infarction patients, with significant improvements in heart functions [101-105]. In addition to autologous MSCs, the efficacy of allogeneic MSCs has also been reported. The commercial bone marrow-derived MSC product Prochymal was administered to reperfused myocardial infarction patients in a double blind, placebo-controlled dose-range safety trial [106]. Allogeneic MSCs were well tolerated with a significant increase in left ventricular ejection fraction and lower incidences of arrhythmia and chest pain, as compared to the placebo group. Allogeneic MSCs derived from the placenta also resulted in significant clinical improvements [107]. Thus, the availability of an off-the-shelf MSC product shows promise with regard to the development of cardiac therapy.

Autoimmune diseases result from an inappropriate immune response of the body against normal cells and tissues. Based on their ability to modulate immune responses, MSCs have also been proposed as a treatment for autoimmune diseases. Patients suffering from severe autoimmune diseases do not respond to standard therapy and often require autologous or allogeneic HSCT [108]. However, HSCT presents many additional complications as well as risks such as toxicity and the incidence of GVHD. Autologous HSCT has often been criticized as identical autoimmune immune cells are being returned back to the patient. Thus, the administration of MSCs may be a safer and more feasible method of treatment. First, the therapeutic role of MSCs has been investigated in patients with Crohn disease. Crohn disease, also known as inflammatory bowel disease, is a chronic inflammatory disorder in which the immune system attacks the gastrointestinal tract. Five patients with Crohn disease were treated with autologous adipose tissue-derived MSCs [109]. The patients were given intralesional treatment of MSCs mixed with fibrin glue. Two patients showed normal healing of the infiltrated area and 75% of treated fistulas had closed and showed signs of significant repair 8 weeks after treatment. These promising results led to a phase II clinical trial [110]. Second, on the basis of preclinical studies, there have been clinical reports on the therapeutic role of MSCs in multiple sclerosis, a chronic inflammatory demyelinating disease of the central nervous system that leads to irreversible damage. Therapeutic approaches have aimed to control the immune response; however, there are still no effective treatments available. In a pilot study, 10 patients with multiple sclerosis received intrathecal injection of culture-expanded MSCs [111]. While administration of MSCs is feasible and safe, the clinical improvements are less clear. During functional assessments, six patients showed some degree of improvement in their sensory, pyramidal, and cerebellar functions, while others showed no improvement or deterioration. Furthermore, the majority of patients showed no differences in MRI assessments after 12 months, indicating that MSC therapy may have less efficacy in multiple sclerosis. Subsequent trials similarly showed mixed results [112-114]. Third, the role of MSCs has also been documented in systemic lupus erythematosus (SLE), an autoimmune inflammatory disease with multiorgan involvement including the kidney, brain, lung, and hematopoietic systems. The most widely used immunosuppressive therapy is corticosteroid administration; however, steroid-based therapies are associated with significant side effects. While MSCs seem to be an attractive therapeutic approach, a recent study suggested that MSCs derived from SLE patients show functional abnormalities [115] and, thus, MSC transplantation may be more effective, as compared to autologous MSCs. In a pilot study determining the safety and efficacy of MSC transplantation in refractory SLE patients, allogeneic MSC transplantation ameliorated disease activity, improved serological markers, and stabilized renal functions [116]. Umbilical cord-derived MSCs have also shown therapeutic potential in SLE patients [117,118]. On the other hand, the use of autologous MSCs was safe, but did not induce significant changes in disease activity [119].

Finally, rheumatoid arthritis (RA) is a T cell-mediated autoimmune disease characterized by cartilage and bone destruction. Their anti-inflammatory properties and regenerative potential indicate that MSCs could offer a novel therapeutic approach to treat RA. However, the role of MSCs in RA has not yet been reported in clinical trials. The therapeutic potential of MSCs is controversial in preclinical studies, which may have delayed their application in clinical trials. While some studies have suggested the efficacy of MSC therapy in collagen-induced arthritis (CIA) models [120,121], many others have suggested that MSCs alone do not suppress the development of Th17-mediated joint inflammation [122,123]. We have also observed that MSCs are ineffective for treatment of CIA [124]. Thus, MSCs have attracted attention as a therapeutic approach for rheumatic diseases, but the immunomodulatory mechanisms must be clarified to ensure further applications in autoimmune diseases.

MSCs have been used to treat cirrhosis in a limited number of trials. Cirrhosis is a chronic liver disease characterized by progressive hepatic fibrosis and loss of hepatic structure with formation of regenerative nodules. Liver transplantation is often the only option in advanced stage patients; however, it is limited by lack of donors, surgical complications, and rejection. MSCs have the potential to be used for the treatment of liver diseases due to their regenerative potential and immunomodulatory properties. Furthermore, MSC therapy could provide minimally invasive procedures with relatively few complications, as compared to liver transplantation. In a phase I trial, four patients suffering from end-stage liver cirrhosis were treated with autologous MSCs and showed improved quality of life with no side effects during follow-up [125]. In another phase I to II clinical trial, eight patients with end-stage liver diseases received autologous MSCs. MSC administration was well tolerated and improved liver functions [126]. Thus, MSC therapy is safe, feasible, and applicable in end-stage liver disease.

MSCs are emerging as vehicles for cancer gene therapy due to their inherent migratory abilities toward tumors [127]. Whether MSCs themselves have antitumor effects is still controversial as some studies have suggested that even unmodified MSCs inhibit tumor growth and angiogenesis [128-130], while others report that MSCs promote tumorigenesis and metastasis [131-133]. Nonetheless, MSCs have been genetically modified to overexpress various anticancer genes, such as ILs [134-138], IFNs [139-141], prodrugs [142,143], oncolytic viruses [144-147], antiangiogenic agents [148], proapoptotic proteins [149,150], and growth factor antagonists [151], for targeted treatment of different cancer types. While preclinical models using gene-modified MSCs for the treatment of cancer have been well studied, clinical trials utilizing engineered MSCs for cancer therapy have not yet been reported. The safety of MSC administration remains a concern even though MSC administration has not yet shown any major adverse events. Their potential to transform malignantly [152,153] and weaken graft versus leukemia effects following HSCT [80] are major issues with regard to guaranteeing the safety of MSC therapy. Engineered MSCs that overexpress potentially hostile molecules may pose serious problems in addition to these concerns. The lack of safety mechanisms following MSC administration has delayed the application of engineered MSCs in clinical settings. Recently, a safety system to allow control of the growth and survival of MSCs has been developed. The safety mechanism is a suicide system based on an inducible caspase-9 protein that is activated using a specific chemical inducer of dimerization (CID) [154]. Exposure to CID induced directed MSC killing within 24 hours. The development of such safety mechanisms and their incorporation into MSC therapy may allow extensive use of genetically engineered MSCs to treat cancer patients in clinical settings.