Failure of insulin secretion is now recognized as the major culprit in Type 2 Diabetes. Secretion of the hormone is mainly controlled by metabolism of glucose in the pancreatic beta-cells. However, full understanding of underlying metabolic processes has not yet been reached. One explanation is that metabolic changes upon variations in ambient glucose are complex. Recently, metabolomic analyses have emerged as a possible way to analyze the complexities of metabolism in an unbiased, and to some extent, global fashion. We have developed an analytical platform, based on gas chromatography (GC) for separation of metabolites, and mass spectrometry (MS) for identification of metabolites. We have applied this technique to extracts from clonal beta-cells to resolve the dynamic changes occurring in beta-cell metabolism after both acute and chronic changes in glucose. Upon culture at 16.7 mM glucose for 48 h, alterations in the levels of 7 metabolites derived from glycolysis, the TCA cycle and pentose phosphate shunt, and 4 amino acids were found. Sixteen metabolites deriving from glycolysis and the TCA cycle and amino acids were identified; two were only detectable in the extracts of clonal beta-cells cultured at 16.7 mM glucose: pyruvate and glucose-6-phospate. The TCA-cycle intermediates, citrate, malate and fumarate, were all increased at high glucose as was ribose-5-phosphate from the pentose phosphate shunt. Alanine and hydroxyproline were increased at high glucose, whereas glutamine and serine were decreased. Next, clonal beta-cells were starved for two hours at 2.8 mM glucose, followed by stimulation with 16.7 mM glucose. Metabolites were extracted, derivatised, and subsequently analysed by GC/MS. Orthogonal projection to latent structures (OPLS) was used to correlate the metabolic profile with the time of quenching. Proteolytic amino acids, exemplified by hydroxyproline, were continuously down-regulated, whereas serine, being an important precursor of many amino acids, purines, and pyrimidine were up-regulated. Several amino acids also showed an initial decrease followed after 15 min by a clear stagnation, which may reflect a switch from protein catabolism to anabolism. Also ribose-5-phosphate was up-regulated during the last 15 min investigated, which indicates increased biosynthesis. Succinate exhibited a late rise, which could reflect a higher level during starvation due to the entrance of amino acids proximal to succinate in the TCA cycle. All other glycolytic and TCA cycle intermediates were strongly up-regulated during the whole interval. The models were confirmed by raw data analysis and t-tests. Our results show that it is feasible to profile and quantify simultaneously multiple changes in metabolite levels in beta-cells under different conditions. Using this approach, it may be possible to reveal pathogenetic pathways in the beta-cell.