A retroviral vector has been constructed that contains the human CD20 cDNA under the control of the Moloney murine leukemia virus (Mo-MuLV) LTR. Freshly isolated mononuclear cells are infected for three consecutive days in the presence of PHA and hrlL-2, and a mean 15.9% of the cells (range, 6.5 to 31.7%) acquire a CD3+CD20+ phenotype. Transduced T lymphocytes grow and expand in vitro for up to 3 weeks like mock-infected cells and, as observed for the T lymphoblastoid CEM cell line, CD20 expression is maintained for several months with no change in the growth curve of the cells. CD20-expressing CEM and fresh T lymphocytes can be positively immunoselected on columns using different anti-CD20 antibodies. Exposure to monoclonal chimeric anti-CD20 IgG1(kappa) Rituximab antibody (Roche), in the presence of complement, results in effective and rapid killing of the transduced CD3+CD20+ human T cells in vitro. This approach represents a new and alternative method to gene manipulation with "suicide" genes for the production of drug-responsive T cell populations, a crucial step for the future management of graft-versus-host disease in bone marrow transplant patients.

We have investigated the capacity of two human immunodeficiency virus type 1-derived lentivectors, differing in the presence of a 118-bp pol fragment containing the cPPT/CTS element, to transduce human normal primary cells of different hematopoietic lineages. Infection of resting monocytes with a high multiplicity of infection (MOI > 10) revealed that the lentivirus carrying the pol fragment (cPPT) is effective, transducing 75% of cells compared with 36% for the no-cPPT vector. Even at low MOIs (< or =1) the cPPT vector still shows a better transduction efficiency than the no-cPPT vector. Moreover, transduction does not require dendritic cell differentiation. In contrast, infection of nonactivated T lymphocytes showed that both vectors, tested at high MOIs, can transduce a small, although measurable, percentage of cells (up to 10%), which may correspond to G(1a) "activated" cells as detected by simultaneous staining of DNA and RNA, in our cultures in the presence of medium alone. Furthermore, we show that the sole addition of interleukin 2 or interleukin 15 represents a full proliferative signal under our conditions and permits high transduction efficiency (up to 30% with the cPPT vector and 15% with the no-cPPT vector). Still higher transduction of T lymphocytes can be achieved after stimulation with phytohemagglutinin and interleukin 2 (up to 78% with the cPPT vector vs. 42% with the no-cPPT vector). Finally, both viruses do not transduce either resting or proliferating tonsillar B lymphocytes.

Human papillomavirus (HPV) genotypes and HPV DNA load were analysed in cervical smears from 76 human immunodeficiency virus (HIV)-positive and 54 HIV-negative women. The prevalence of genotypes was similar for all women, with the exception of HPV62, which was over-represented in HIV-positive samples. HIV-positive women showed a higher prevalence of multiple genotypes that correlated neither with CD4(+) T-cell counts nor with cervical dysplasia. No significant differences were observed in terms of total or single-type HPV DNA load. The HPV DNA load in both HIV-positive and HIV-negative women was significantly higher in squamous intra-epithelial lesions than in negative Pap smears.

Summary.In order to develop a clinically feasible gene marking approach, we have used the recently described PINCO retroviral expression system, composed of the enhanced green fluorescence protein (EGFP) cDNA driven by Moloney MLV LTR and packaged in the Phoenix amphotropic cell line. Two T, five B, one erythromyeloid and three myeloid cell lines were successfully infected with % GFP þ cells ranging from 4% to 79%, showing a lineagedependent difference in infection susceptibility, with the myeloid cells being the least efficiently infected. We also infected normal mononuclear peripheral cells cultured in PHA and rhIL-2 for 2 d, and obtained an average of 30% GFP þ cells, all present within the CD3 þ population, with CD4 þ and CD8 þ cells being equally infected. Finally, the tonsillar purified B population showed lower levels of infectivity (6%) whereas high susceptibility was shown by normal human umbilical vein endothelial cells (57%). Highly purified CD34þ cells were also susceptible, varying from 6% to 10% GFP þ cells. Immature myeloid/erythroid progenitors have been infected which stably expressed the GFP protein during further differentiation in culture. The GFP þ T cells were FACS-sorted rapidly upon infection, subsequently cultured and the fluorescence intensity monitored. In all cases the difference in percentage of GFP þ cells did not correlate with the percentage of S/G 2 /M cycling cells as determined at the moment of infection or with the expression levels of Ram-1 amphotrophic receptor. The improved safety of this retroviral system, the rapidity of the technique, the high efficiency of infection with respect to normal T lymphocytes (in this last case higher than previously reported) and the lack of need for in vitro selection make this system favourable for clinical development.

No abstract