The glutamate-glutamine cycle describes the neuronal release of glutamate into the synaptic cleft, astrocytic uptake, and conversion into glutamine, followed by release for use as a neuronal glutamate precursor. This only explains the fate of the carbon atoms, however, and not that of the ammonia. Recently, a role for alanine has been proposed in transfer of ammonia between glutamatergic neurons and astrocytes, denoted the lactate-alanine shuttle (Waagepetersen et al. [ 2000] J. Neurochem. 75:471-479). The role of alanine in this context has been studied further using cerebellar neuronal cultures and corresponding neuronal-astrocytic cocultures. A superfusion paradigm was used to induce repetitively vesicular glutamate release by N-methyl-D-aspartate (NMDA) in the neurons, allowing the relative activity dependency of the lactate-alanine shuttle to be assessed. [(15)N]Alanine (0.2 mM), [2-(15)N]/[5-(15)N]glutamine (0.25 mM), and [(15)N]ammonia (0.3 mM) were used as precursors and cell extracts were analyzed by mass spectrometry. Labeling from [(15)N]alanine in glutamine, aspartate, and glutamate in cerebellar cocultures was independent of depolarization of the neurons. Employing glutamine with the amino group labeled ([2-(15)N]glutamine) as the precursor, an activity-dependent increase in the labeling of both glutamate and aspartate (but not alanine) was observed in the cerebellar neurons. When the amide group of glutamine was labeled ([5-(15)N]glutamine), no labeling could be detected in the analyzed metabolites. Altogether, the results of this study support the existence of the lactate-alanine shuttle and the associated glutamate-glutamine cycle. No direct coupling of the two shuttles was observed, however, and only the glutamate-glutamine cycle seemed activity dependent.