The appearance of the CD4+CD8+ phenotype on activated T cells: possible role of antigen transfer

Stimulation of peripheral blood lymphocytes (PBL) with phytohemagglutinin (PHA) strikingly increased the proportion of CD4+CD8+ cells. Highly purified CD4+ and CD8+ lymphocyte populations cultured in the presence of PHA consistently failed to coexpress the CD8 and CD4 markers. Similarly, exposure of highly purified CD4+ cells to PHA and recombinant interleukin-2 resulted in augmented expression of CD25 but failed to induce the expression of CD8. When purified preparations of either CD4+ or CD8+ cells were activated separately for 3 days and incubated together for an additional 5 h, a considerable proportion of CD4+CD8+ cells was found in the mixture. Cycloheximide treatment did not prevent the appearance of the CD8 marker on CD4 cells. CD4+CD8+ cells isolated from PBL exposed for 3 days to PHA lost their CD8 antigenicity within 24-48 h in the absence of PHA. Increased levels of soluble CD4 and CD8 antigens were found in supernatant fluids of PHA-stimulated cells. T cells failed, however, to bind soluble markers even after prolonged incubation in the presence of supernatant fluids. Our studies show that activation of CD4+ cells per se does not elicit the CD4+CD8+ phenotype and that soluble T cell markers do not bind to T cells. Rather, it seems that direct cell-cell contact is required for the transfer of CD8 molecules from CD8+ cells to the membrane of CD4+ cells.