The primary objective of this study was to determine the influence of stretch-induced cell injury on the metabolism of cellular phosphatidylcholine (PC). Neonatal rat astrocytes were grown to confluency in Silastic-bottomed tissue culture wells in medium that was usually supplemented with 10 microM unlabeled arachidonate. Cell injury was produced by stretching (5-10 mm) the Silastic membrane with a 50-ms pulse of compressed air. Stretch-induced cell injury increased the incorporation of [3H]choline into PC in an incubation time- and stretch magnitude-dependent manner. PC biosynthesis was increased three- to fourfold between 1.5 and 4.5 h after injury and returned to control levels by 24 h postinjury. Stretch-induced cell injury also increased the activity of several enzymes involved in the hydrolysis [phospholipase A2 (EC 3.1.1.4) and C (PLC; EC 3.1.4.3)] and biosynthesis [phosphocholine cytidylyltransferase (PCT; EC 2.7.7.15)] of PC. Stretch-induced increases in PC biosynthesis and PCT activity correlated well (r = 0.983) and were significantly reduced by pretreating (1 h) the cells with an iron chelator (deferoxamine) or scavengers of reactive oxygen species such as superoxide dismutase and catalase. The stretch-dependent increase in PC biosynthesis was also reduced by antioxidants (vitamin E, vitamin E succinate, vitamin E phosphate, melatonin, and n-acetylcysteine). Arachidonate-enriched cells were more susceptible to stretch-induced injury because lactate dehydrogenase release and PC biosynthesis were significantly less in non-arachidonate-enriched cells. In summary, the data suggest that stretch-induced cell injury is (a) a result of an increase in the cellular level of hydroxyl radicals produced by an iron-catalyzed Haber-Weiss reaction, (b) due in part to the interaction of oxyradicals with the polyunsaturated fatty acids of cellular phospholipids such as PC, and (c) reversible as long as the cell's membrane repair functions (PC hydrolysis and biosynthesis) are sufficient to repair injured membranes. These results suggest that stretch-induced cell injury in vitro may mimic in part experimental traumatic brain injury in vivo because alterations in cellular PC biosynthesis and PLC activity are similar in both models. Therefore, this in vitro model of stretch-induced injury may supplement or be a reasonable alternative to some in vivo models of brain injury for determining the mechanisms by which traumatic cell injury results in cell dysfunction.