The results concerning LPS priming and inactivation are summarized in table 4. This table clearly shows that priming by and inactivation of LPS are mediated via different pathways, because there is no correlation whatsoever between priming and inactivation. LPS priming nicely correlates with CD14 expression. Monocytes express high levels of CD14, and are highly sensitive to LPS. Neutrophils express lower levels of CD14, and also require a higher concentration of LPS to obtain a primed state. Eosinophils express even lower levels of CD14 (if any), and these cells are not primed by LPS up to 150 ng/ml in the presence of serum (data not shown). [table: see text] The inactivation experiments were performed with LPS, but unpublished results from M. Pabst et al. show the same phenomenon with synthetic lipid A. This indicates that the inactivation of LPS is caused by a modification of the lipid A moiety of LPS. As a first step, we tested the hypothesis that the inactivation of LPS is a dephosphorylation of lipid A by alkaline phosphatase (AP). However, AP activity in isolated neutrophils of one of the PNH patients was completely undetectable. Nevertheless, these neutrophils were able to inactivate LPS. This result clearly shows that AP is not the enzyme that inactivates LPS. Secondly, we measured neutral pH 7 phosphatase both in whole cells, and in cell lysate. All cell types tested (neutrophils, eosinophils and monocytes) contained comparable levels of pH 7 phosphatase, but only neutrophils were able to inactivate LPS.(ABSTRACT TRUNCATED AT 250 WORDS)