The reversal of the anti-CD3 stop signal is unaffected by the presence of CD28. (A) Interaction of Cd28^+/+ and Cd28^−/−CD4-positive T cells with DCs presenting OVA peptide. Left panel: Tracing of the migration of pre-activated Cd28^+/+ and Cd28^−/− CD4-positive T cells on LN slices. T cells were seeded with DCs alone or with DCs that had been pre-incubated with OVA peptide (0.5 μg/mL; DC-OVA). T-cell tracks have been superimposed from their starting positions. Right panel: Contact times of Cd28^+/+ and Cd28^−/− CD4- positive T cells with DCs in the presence and absence of OVA peptide on LN slices. Bottom left panel: Cd28^+/+ and Cd28^−/− CD4-positive T-cell velocities with DCs in the presence and absence of OVA peptide on LN slices. Bottom right panel: Measurements of displacement of Cd28^+/+ and Cd28^−/− CD4⁺ T cells with DCs in the presence and absence of OVA peptide on LN slices. (B) Anti–CTLA-4 reverses equally the stop-signal on both the Cd28^+/+ and Cd28^−/− CD4-positive T cells. Top panel: Tracing patterns of Cd28^+/+ and Cd28^−/− CD4-positive T cells. T cells were initially activated for CTLA-4 surface expression and then rested for 24 hours before use in experiments. Cells were monitored over 20 minutes for random movement on glass slides coated with 2 μg/mL of ICAM-1-Fc in the presence of soluble anti-CD3, anti–CTLA-4, or anti-CD3/CTLA-4. Stimulation with soluble antibody isotype served as a negative control. Rabbit anti–hamster antibody was used for crosslinking (1:4 ratio to primary antibodies). Bottom left panel: measurements of velocity; right panel: measurements of displacement. Differences between means were tested using 2-tailed Student t test (GraphPad Prism 5.0). P < .05 was considered significant. Data are representative of 4 separate experiments.