A comparative investigation of three methods of separation of CD4 and CD8 cells from human peripheral blood cells.

A comparison was made of the phenotypic purities and functional efficiencies of CD4 and CD8 cell subsets, separated by three methods. CD4 and CD8 monoclonal antibodies were used for positive and negative selection of T cells by (a) panning, (b) red cell rosetting, and (c) complement mediated killing techniques. The viability of the CD4 and CD8 cell subsets, separated by any of the three methods and assessed by the dye exclusion method, was greater than 95%. An inverse relationship was found between the yields and purities of the CD4 and CD8 cell subset. The most efficient separation of the T cells was achieved by positive selection using the rosetting method; 95.3 +/- 1.8% of the CD4 subset reacted with antibodies to CD4 and 1.4 +/- 0.5% with antibodies to CD8 antigens. The separated CD8 subset showed 90.7 +/- 0.4% reacting with CD8 and 3.5 +/- 1.3% with CD4 antibodies. This separation was slightly better than that achieved by the panning method. However, the positive methods of selection gave lower yields than the negative selection methods. The five CD4 and five CD8 cell subsets were assessed functionally by stimulating with a mitogen (phytohaemagglutinin), a bacterial antigen (purified protein derivative) and a viral antigen (herpes simplex virus). The results suggest that the uptake of [3H]thymidine is generally most efficient with the positively selected rosetting CD4 and CD8 cell subset.