Dendritic cells (DC) are critical accessory cells for primary immune responses and they may be important stimulators of transplantation reactions, but little is known of their traffic into the tissues. We have studied the migration of purified splenic DC and T lymphocytes, labeled with 111Indium-tropolone, in syngeneic and allogeneic mice. First we demonstrate that DC can migrate from the blood into some lymphoid and nonlymphoid tissues. Immediately after intravenous administration, radio-labeled DC were sequestered in the lungs, but they actively migrated into the liver and spleen and reached equilibrium levels between 3 and 24 h after transfer. At least half of the radiolabel accumulated in the liver, but the spleen was the principal site of DC localization in terms of specific activity (radiolabel per weight of tissue). DC were unable to enter Peyer's patches, or mesenteric and other peripheral lymph nodes from the bloodstream. This was also true in splenectomized recipients, where the otherwise spleen-seeking DC were quantitatively diverted to the liver. In contrast, T cells homed readily to the spleen and lymph nodes of normal mice and increased numbers were present in these tissues in splenectomized mice. Thus, unlike T cells, DC cannot recirculate from blood to lymph via the nodes. We then show that migration of DC from the blood into the spleen is dependent on the presence of T cells: DC did not enter the spleens of nude mice, but when they were reconstituted with T cells the numbers entering the spleen resembled those in euthymic mice. In nude mice, as in splenectomized recipients, the DC that would normally enter the spleen were quantitatively diverted to the liver. These findings suggest that there is a spleen-liver equilibrium for DC, that may be akin to that existing between spleen and lymph node for T cells. Finally, we followed the traffic of radiolabeled DC via the afferent lymphatics after subcutaneous footpad inoculation. DC accumulated in the popliteal nodes but did not migrate further to the inguinal nodes. There was no difference between euthymic and nude mice, showing that unlike traffic to the spleen, this route probably does not require T cells. These migration patterns were not affected by major histocompatibility barriers, and were only seen with viable, but not glutaraldehyde-fixed, DC.(ABSTRACT TRUNCATED AT 400 WORDS)