The delivery to joints of genes able to inhibit inflammation and bone loss represents an attractive therapeutic strategy in arthritis. However gene transfer into synoviocytes is limited by the fact that these cells are naturally nonproliferative. As lentiviral vectors are capable of transducing nondividing cells, contrary to the other retroviral vectors, they could represent an interesting solution. Moreover, the recent description of the role in viral DNA nuclear import of a central DNA flap in HIV-1 genome allows the construction of HIV vectors with an increased capacity to introduce genes into nondividing cells. Therefore, we tested such a vector ex vivo on human synoviocytes, either nontreated, or growth-arrested by the addition of aphidicolin, or treated with the reverse transcriptase inhibitor AZT. An adenoviral vector was used as positive control, and a murine retroviral vector as negative control. Table 1 shows that the HIV vector, contrary to the control murine retroviral vector, was able to transduce synoviocytes, even in absence of cell division. Lentiviral vectors, that induce no inflammation and are stably integrated, could represent an interesting tool for gene therapy of arthritis. In vivo results will be presented.