Phosphatidylinositol (PtdIns) synthesis and polyphosphoinositide (PPI) formation were measured as the incorporation of [32P]orthophosphate ([32P]Pi) or [3H]inositol into non-stimulated intact human neutrophil membrane phospholipids. The rate of PtdIns "de novo" synthesis appeared to be a slow mechanism when compared to the rapid incorporation of [32P]Pi into PPIs. Of the "de novo" synthesized [3H]PtdIns, 70% was further phosphorylated to PPI. Nevertheless, this PPI pool represented less than 0.01% of the total nmols of PPIs formed evaluated as [32P]Pi labeling, indicating that PPI formation mainly involves a no "de novo" synthesized phosphatidylinositol pool. When evaluated at short incubation times, oscillations in the formation of PPIs were detected. A rapid phase was characterized after 30 s of incubation with [32P]Pi Phosphorylation levels returned to an equilibrium state within a minute, and the second phase peaked at 5 min., returning to equilibrium at 15 min. The fluctuant kinetics though not the equilibrium level of PPI formation, could be abolished by neomycin. On the other hand, a selective inhibition of the rapid phase of PPI synthesis occurred in the presence of the tyrosine kinase inhibitor genistein. When the incorporations of [gamma-32P]-adenosine triphosphate (ATP) or [32P]Pi into human neutrophil particulate fraction membranes were evaluated, PPIs synthesis showed fluctuations independently of the precursor used. Noticeably, [32P]from [32P]Pi was incorporated more efficiently into PPIs than that from [gamma-32P]ATP, when evaluated in parallel using equal specific activities for both radiolabeled precursors and under non-ATP synthesizing conditions. Moreover, the incorporation of [32P]Pi into particulate fraction PPIs was not abolished by high concentrations of non-radiolabeled ATP, and metabolically inhibited PMNs showed high rates of PPI synthesis. These data suggest that PPI formation is not necessarily a futile cycle in PMNs.