Lipopolysaccharide (LPS) in solution primes neutrophils for enhanced release of superoxide in response to N-formyl-methionyl-leucyl-phenylalanine. We show that LPS immobilized on polystyrene or polypropylene acted on neutrophils by a mechanism different from that of LPS in solution. Coating the surface with 1% plasma, either before coating with LPS (plasma/LPS) or after coating with LPS (LPS/plasma), was essential to induce the LPS response in neutrophils. However, plasma could be replaced by fibrinogen, type I collagen or type IV collagen, or, to a lesser extent, by fibronectin or vitronectin, which was not true for LPS in solution. About 20% of the LPS added was immobilized on the plastic surfaces, based on its ability to adsorb anti-LPS antibody after extensive washing. The amount of soluble LPS that might have been released from surfaces during the incubation with neutrophils was too low to account for the priming by immobilized LPS. About 13-20 min was needed for neutrophils to become primed after incubation with immobilized LPS. Immobilized LPS induced up-regulation of CD11b/CD18 and latent alkaline phosphatase and also enhanced the adhesive response of neutrophils. Priming by immobilized LPS was inhibited by anti-CD14 antibody or by treatment of neutrophils with the LPS antagonist LA-14-PP. When immobilized LPS was treated with anti-LPS-binding protein (LBP) antibody, the response of neutrophils to LPS/plasma was inhibited but the response to plasma/LPS or fibrinogen/LPS was not. Thus, the LPS in plasma/LPS or fibrinogen/LPS acted on neutrophils in an LBP-independent manner. We conclude that the CD14-dependent LPS receptor system of neutrophils was capable of working in the absence of LBP, but only when LPS was immobilized on a surface coated with protein.