Glioma is the most common primary tumor of the central nervous system in adults. In recent years, with the rapid development of diagnostic imaging technology, the continuous improvement of traditional therapies such as surgical techniques, radiotherapy and chemotherapy, and the emergence of new adjuvant therapies such as immunotherapy and anti-angiogenic therapy, the comprehensive treatment effect of glioma has been improved to some extent [1]. However, because gliomas often form satellite lesions by infiltrative growth into surrounding normal tissues at early stages of development, they are prone to recurrence and increased malignancy after surgical resection,[p1] and are resistant to radiotherapy and chemotherapy, so patients’ survival has not been effectively prolonged. Recent studies have confirmed that, in addition to neural stem cells (NSCs), mesenchymal stem cells (MSCs) also have the ability to migrate chemotactically to gliomas and inhibit tumor growth to a certain extent, which is expected to be a novel vehicle for gene therapy [2-3]. 1. selection of cell types Apart from NSCs, the most widely studied stem cell types for chemotactic migration to gliomas and gene therapy are MSCs of human or rat bone marrow origin [4-8]. It has been reported that both modified Transwell in vitro cell migration analysis and in vivo experiments transplanted in a rat brainstem glioma model showed no significant difference in the migratory capacity of human MSCs isolated from bone marrow, adipose and umbilical cord blood compared to human NSCs [9]. Some authors have also found that endometrial regenerative cells (ERCs) [10], adipose derived stem cells (ASCs) [11] and human cells, skin-derived stem cells (hSDSCs) [12] have the ability to migrate towards glioma and are expected to be alternative vector sources for gene therapy of glioma. In vivo and ex vivo tumorigenic effects A number of in vitro chemotactic migration experiments and in vivo transplantation by intra-tumoral, ipsilateral or contralateral hemispheres, internal carotid artery and tail vein have confirmed the ability of MSCs to migrate towards gliomas. The distribution characteristics of MSCs along the tumor periphery may contribute to their barrier effect, thus preventing the infiltrative growth of glioma cells into the brain parenchyma.Nakamizo et al [14] found that MSCs transplanted into [ p5] human U87 glial cells were able to migrate to the tumor bed, but not to the normal brain parenchyma. p5] of human bone marrow-derived MSCs[p6] transplanted into the ipsilateral internal carotid artery, the contralateral internal carotid artery, or the hemisphere contralateral to the tumor in the human U87 glioblastoma model mice were widely distributed within the brain tumor, suggesting that MSCs transplanted via the vascular route have a similar specific migratory capacity as those transplanted intracranially. hSDSCs injected via the contralateral hemisphere or tail vein were reported by Pisati et al[12] to Kim et al.]15] also reported that hSDSCs expressing secretable tumor necrosis factor-related apoptosis-inducing ligand (sSCIL) were able to migrate to human U87 glioblastoma, and were widely distributed in the tumor bed and reduced vascular density and vascular sprouting after injection into the tumor. -Han et al [10] reported that ERCs injected intravenously or intratumorally into a rat glioma model were able to migrate directionally into the glioma and inhibit its growth. Lamfers et al [11] found that the chemotactic migration of ASCs to human U87 glioblastoma and their distribution were similar to that of bone marrow MSCs, i.e., most of the cells were distributed in and around the tumor, while few were distributed in normal brain tissue. Although researchers have used different tissue sources and glioma models to transplant MSCs, in vitro migration experiments and in vivo transplantation studies suggest that MSCs have the ability to migrate specifically to gliomas, but the mechanism of this chemotactic migration is not well understood. Currently, it is thought to be related to the interaction between glioma cells and MSCs, and certain growth factors, angiogenic factors and chemotactic factors and their receptors in the glioma microenvironment may play an important role in this regard. For example, in vitro experiments by Schichor et al [16] showed that vascular endothelial growth factor-A (VEGF-A) is an important factor in increasing the chemotactic migration of human bone marrow MSCs to human glioma cells U87, U-373, U-251 and MZ-54, suggesting that MSCs can play a more important role in gliomas with high VEGF expression.Birnbaum et al [17 ] found that human glioma cells U373, U251 and MZ54 can recruit MSCs by secreting large amounts of pro-angiogenic factors such as interleukin-8, transforming growth factor and neurotrophic factor-3, but VEGF, platelet-derived growth factor, glial cell-derived neurotrophic factor, brain-derived neurotrophic factor and ciliary nerve growth factor are not involved in this process. This suggests that gliomas can attract mesenchymal cells by secreting a variety of pro-angiogenic factors [17]. However, in contrast to Birnbaum et al [17], Nakamizo [14] suggested that platelet-derived growth factor, VEGF and stromal cell-derived factor-1α play an important role in the migration of MSCs to human U87 glioma cell line, and Cheng et al [8] also reported that platelet-derived growth factor-BB promotes the migration of bone marrow-derived MSCs to rat Kim et al [18] found that interleukin-8 and growth factor-associated oncogene -α enhanced the migration of cord blood-derived MSCs to various types of human glioma cells and suggested that the significantly stronger migration of cord blood MSCs to gliomas than bone marrow-derived MSCs may be related to the interleukin-8 receptor (i.e. CX-8) in the former. -Ho et al [5] suggested that the difference in the migratory ability of MSCs from different sources to human glioma cells may be related to the difference in the expression level and activity of matrix metalloproteinase-1. 4. Gene targeting therapy based on MSCs Although virus-based gene therapy has achieved certain results in animal experiments, however, it has had little success in the clinical experimental treatment of human malignant glioma. The tumorigenic effect of MSCs and their ease of gene modification have made them a hot topic in the selection of gene therapy vectors for glioma. The most studied vectors are those carrying herpes simplex virus-thymidine kinase (HSVtk), tumor necrosis factor-related, apoptosis-inducing ligand, TRAIL), interleukin and interferon, and other therapeutic genes for the antitumor effects of MSCs. HSV-tk gene is the most widely studied oncologic drug-sensitive gene. HSV-tk/ganciclovir (GCV) kills tumor cells by the following mechanism: thymidine kinase (tk) catalyzes the phosphorylation of deoxythymidine (dThd) to produce deoxythymidine acid, and HSV-tk catalyzes the above process as well as As a purine nucleoside analogue, GCV is not toxic or has low toxicity when taken up by HSV-tk-negative cells, but phosphorylation by tk in HSV-tk-positive cells generates a phosphorylation product that is toxic to cells, thereby inhibiting cellular DNA polymerase activity, or acts as a competitive inhibitor of deoxyguanosine triphosphate, which is incorporated into Amano et al [4] have shown that transplantation of adenovirus-transfected rat bone marrow MSCs carrying HSV-tk in combination with GVC resulted in a significant reduction in the size of rat C6 gliomas and prolonged the survival of tumor-bearing rats. Kinoshita et al[20] also reported that human immortalized MSCs (hiMSCs) carrying HSV-tk could migrate through the corpus callosum to the tumor periphery after injection into the subcortical area contralateral to the tumor in a human HTB14 glioblastoma nude mouse model, and that GCV administration to mice resulted in a significant reduction in tumor size, suggesting that the combination of the two could exert antitumor effects through bystander effects . As a member of the tumor necrosis factor superfamily, TRAIL can selectively induce apoptosis in tumor cells, but has no killing effect on most normal cells. Menon et al [6] found that genetically modified human bone marrow-derived MSCs could express and release biologically active sTRAIL in vivo and ex vivo, and after transplantation into the skull of tumor-bearing mice with human U87 glioma, the number of apoptotic tumor cells increased 8-fold, the tumor volume decreased 81,6%, and the survival of transplanted mice was significantly prolonged. kim [15] also reported that that cord blood-derived MSCs carrying TRAIL exerted similar antitumor effects. In the study of interleukins and interferons, Nakamura et al [13] found that intratumoral injection of MSCs inhibited tumor growth and prolonged survival in tumor-bearing rats, and the antitumor effect was further enhanced by MSCs expressing interleukin-2 with genetic modifications. Nakamizo et al [14] found that MSCs expressing IFN-β could exert similar tumor-killing effects. In recent years, with the rapid development of molecular biology, cytogenetics, genetic engineering and other disciplines, people began to try to improve the efficacy and prognosis of glioma with gene therapy protocols, and the first clinical protocol using HSV-tK/GCV for glioma treatment was approved by NIH in 1992, which set off a new worldwide research on gene therapy for glioma. However, the commonly used viral vectors and somatic vectors have been used to treat glioma. However, the commonly used viral vectors and somatic cell vectors not only have limited efficiency in expressing target genes, but also have difficulty in crossing the blood-brain barrier and the blood-tumor barrier and targeting tumor cells and microsatellite foci that infiltrate into normal brain tissue, which makes the clinical application of gene therapy for glioma face a new bottleneck. As the understanding of stem cells deepened, people began to try to use stem cells as a vehicle for gene therapy. Although NSCs, which were once expected to have good tumorigenic effects and be easily genetically modified, their widespread application has been limited by many aspects such as difficulty in obtaining materials, insufficient number of cells and ethics. However, further studies have shown that MSCs of various origins have the ability to migrate chemotactically to human or murine gliomas in ex vivo experiments, and can be widely distributed at the margins of tumors and normal brain tissues as well as inside tumors, and that MSCs genetically modified to express specific therapeutic factors can exert good anti-tumor effects and improve the survival of tumor-bearing animals. The investigation of the mechanism of tumorigenic effect of MSCs helps to understand the interaction between glioma cells and MSCs, and also provides new ideas to enhance the tumorigenic effect of MSCs and select the suitable glioma types for MSCs treatment. Therefore, it is reasonable to believe that gene therapy using MSCs as vectors will have a wide application in the refractory disease of glioma.