To date, a selective advantage of cells expressing anti-HIV ribozymes has not been shown. This study was undertaken to determine whether such a selective advantage can be demonstrated in vitro. A retroviral vector coding for a hairpin ribozyme targeting the HIV 5'LTR and for the low affinity nerve growth factor receptor (LNGF-RDelta) was designed. Since we demonstrated by RT-PCR that the amount of ribozyme transcripts was highly correlated with the level of surface LNGF-RDelta expression, the vector was utilized to assess ribozyme expression by flow cytometry. Transduced Hut78 and primary CD4+ T cells were purified and subsequently mixed with unmodified cells. After HIV challenge the percentage of ribozyme expressing cells in the cell mixture was monitored by flow cytometry. Twenty-one days after HIV infection the proportion of ribozyme expressing CD4+ T cells was 2.6 times higher in comparison to cells with the control vector. CD4+ T cells with a strong ribozyme expression conferred a 7.4-fold selective advantage at day 21 and a 11.7-fold at day 28. For Hut78 cells a selective advantage was detected exclusively for strongly ribozyme expressing cells. As a mechanism underlying the selective advantage an inhibition of HIV induced apoptosis was shown. These results demonstrate that anti-HIV ribozymes are able to confer a selective survival advantage and indicate that the protective effect is dependent on the amount of ribozyme expression. Gene Therapy (2000) 7, 408-416.