We demonstrated that a primary exposure to the lymphocytosis promoting factor (LPF) of Bordetella pertussis-induced T cell colony formation. Colony formation was observed when mononuclear cells (MNC) were cultured at concentrations of more than 1 X 10(6)/ml, and reached a peak on day 8. However, the number of colonies generated with LPF was about one-third induced with phytohaemagglutinin (PHA). Removal of monocytes from MNC or T cells resulted in the failure of colony formation, but colony growth could be restored by the addition of monocytes or B enriched cells, indicating that they were required for the optimal colony growth induced by LPF. In the absence of accessory cells, optimal colony growth from monocyte depleted T cells could be obtained when an appropriate concentration of phorbol myristate acetate (PMA) or interleukin-2 (IL-2) was added in the cultures with LPF. PMA did not enhance LPF-induced colony formation in the cultures containing a sufficient amount of exogeneous IL-2. These findings suggest that IL-2 is essential to LPF-induced colony formation. Surface marker analysis showed that most of LPF-induced colony cells were T cells. The percentages of T4+ and T gamma cells of LPF-induced colony cells were more, and T8+ cells less, than those of PHA-induced colony cells. Ia1, T9 and Tac antigens were detected on many colony cells induced by LPF or PHA. These results indicate that the phenotype of LPF-induced colony cells differs from those of PHA, but the sequential antigen expression on lymphocytes triggered by IL-2 might be similar in both LPF- and PHA-induced colony formation.