It was previously shown that CD26 (DPP IV, EC 3.4.14.5) is a binding site for adenosine deaminase (ADA, EC 3.5.4.4) on T cells and that costimulation by some anti-CD26 monoclonal antibodies (mAbs) and anti-CD3 induces CD4(+) T cell proliferation, The CD26 epitopes involved in costimulation, the precise sequence of the events preceding proliferation, and the response of CD8(+) compared with CD4(+) T cells to CD26 were not extensively studied, We therefore compared the effects of the novel TA5.9 anti-CD26 mAb, recognizing an ADA-binding epitope, and the clearly distinct anti-Ta1 reference anti-CD26 mAb for their costimulatory properties in various T cell subsets, Both purified CD4(+) and CD8(+) T cells proliferated upon costimulation with anti-CD3 and either anti-CD26 mAb, but anti-TA5.9 mAb induced a more potent response than anti-Ta1. Either anti-CD26 mAb, together with anti-CD3, caused a similar sequential up-regulation of CD69, CD25 (IL-2R alpha), and CD71 (transferrin receptor) expression on CD4(+) and CD8(+) T cells, The activation markers appeared faster on the CD45R0(+) than on the CD45R0(-) subsets, After costimulation, CD4(+) T cell cultures contained significant amounts of the Th1 cytokines IL-2, interferon-gamma (IFN-gamma), and tumor necrosis factor-alpha (TNF-alpha). In CD8(+) T cell cultures relatively more IFN-gamma and TNF-alpha but almost no IL-2 was measured after triggering of CD3 and CD26, Our data demonstrate that the recognition of the ADA-binding epitope is not a prerequisite for the costimulatory capacity of anti-CDZ6 mAbs, Both CD4(+) and CD8(+) T cells and their CD45R0(-) and CD45R0(+) subsets are sensitive to various aspects of activation via CD26, but the magnitude and/or kinetics differ according to the anti-CD26 used and the T cell subset studied.