Sllmmsr~The cellular infiltrates of certain inflammatory processes found in parasitic infection or in allergic diseases consist predominantly of eosinophilic granulocytes, often in association with activated T cells. This suggests the existence of chemotactic agonists specific for eosinophils and lymphocyte subsets devoid of neutrophil-activating properties. We therefore examined four members of the intercrine/chemokine superfamily of cytokines (monocyte chemotactic peptide 1 [MCP-1], RANTES, macrophage inflammatory protein 1ix [MIP-ltx], and MIP-1/3), which do not activate neutrophils, for their ability to affect different eosinophil effector functions. R.ANTES strongly attracted normal human eosinophils by a chemotactic rather than a chemokinetic mechanism with a similar efficacy as the most potent chemotactic myeloid cell agonist, C5a. MIP-ltx also induced eosinophil migration, however, with lower efficacy. RANTES and MIP-lo~ induced eosinophil cationic protein release in cytochalasin B-treated eosinophils, but did not promote leukotriene C4 formation by eosinophils, even after preincubation with interleukin 3 (IL-3), in contrast to other chemotactic agonists such as C5a and formyl-methionyl-lencyl-phenylalanine (FMLP). RANTES, but not MIP-ltx, induced a biphasic chemiluminescence response, however, of lower magnitude than C5a. RANTES and MIP-lo~ both promoted identical transient changes in intracellular free calcium concentration ([Ca2+]i), with kinetics similar to those induced by chemotactic peptides known to interact with G protein-coupled receptors. No cross-desensitization towards other peptide agonists (e.g., C5a, IL-8, FMLP) was observed, suggesting the presence of specific receptors. Despite its weaker eosinophil-activating properties, MIP-lol was at least 10 times more potent on a molar basis than R.ANTES at inducing [Ca 2+ ]i changes. Interestingly, RANTES deactivated the MIP-lcz-induced [Ca2+]i changes, while the R.ANTES response was preserved after MIP-lcz stimulation. MCP-1, a potent monocyte chemoattractant and basophil agonist, as well as MIP-1/3, a peptide with pronounced homology to MIP-lo~, did not activate the eosinophil functions tested. Our results indicate that R.ANTES and MIP-ltx are crucial mediators of inflammatory processes in which eosinophils predominate.