Haematopoeitic system diseases, both acquired and inherited, can be currently cured by allogeneic haematopoietic stem cell transplantation. This treatment provides highly successful immune function recovery for patients receiving grafts of HLA-compatible donors but has still a great risk of complications and even failure if no suitable donor is available. Among the alternative therapeutic options, ex vivo retrovirus-mediated gene transfer into haematopoietic progenitor cells has been shown to be an efficient strategy for a substantial number of severe combined immunodeficiency-suffering patients. A recent gene therapy trial has been remarkably effective for the immunological reconstitution of patients suffering from X-linked severe combined immunodeficiency. This treatment was able to provide full correction of disease phenotype and thus, clinical benefit. However, the appearance of leukemia in several patients has put in question the safety of the procedure. This severe adverse event has been attributed to the integration of the therapeutic transgene-carrying viral vector into a known T-cell oncogene, LMO2, thereby contributing to the development of T-cell leukemia by causing aberrant expression of LMO2. Further studies mentioned the possible retroviral-mediated cis-activation of the LMO2 promoter underlying the potential ability of retroviral regulatory elements to influence neighboring gene transcription. This project aimed at decreasing the risk associated with the use of viral vectors for gene therapy through the identification of genetic insulator elements capable of isolating the vector regulatory elements in order to prevent the activation of chromosomal genes by the viral enhancers. We have established a standardized screening procedure whereby the potency of insulators can be assessed quantitatively on relevant vector elements. This assay system consists of a series of plasmids containing two reporter genes: one mimicking a therapeutic gene under the control of strong viral long terminal repeat (LTR) enhancer, and the other one standing for an endogenous gene close to the chromosomal vector integration site. The assay allowed the quantification of the enhancer blocking activity of the well-characterized chicken beta-globin 5'HS4 insulator (cHS4) in cultured cells. We assessed the insulator activity of novel synthetic elements, constructed from optimized binding sites for the insulator protein CTCF. In addition, we demonstrated the enhancer blocking activity of the nuclear factor 1 protein family (CTF/NF1) and showed that it also displays barrier properties, protecting transgene expression from silencing. We showed that both CTCF and CTF/NF1 binding sites act as insulators that mediate potent enhancer-blocker activity, resulting in a commensurate reduction of genotoxicity when implemented in viral gene therapy vectors. Finally, we used the same approach to analyze the enhancer blocking activity of well-known chromatin domain delimiters, matrix attachment regions (MAR), mainly characterized by their barrier properties. We could demonstrate the enhancer blocking ability of the 1-68 MAR that is mainly harbored by an A-T rich core sub-region.