Carbohydrate response element-binding protein (ChREBP) is a transcription factor that mediates glucose signaling in mammalian liver, leading to the expression of different glycolytic and lipogenic genes, such as pyruvate kinase (L-PK) and fatty acid synthase (FAS). The current model for ChREBP activation in response to sugar phosphates holds that glucose metabolization to xylulose 5-phosphate (X-5-P) triggers the activation of protein phosphatase 2A, which dephosphorylates ChREBP and leads to its nuclear translocation and activation. However, evidence indicates that glucose 6-phosphate (G-6-P) is the most likely signal metabolite for the glucose-induced transcription of these genes. The glucose derivative that is responsible for carbohydrate-dependent gene expression remains to be identified. The difficulties in measuring G-6-P and X-5-P concentrations simultaneously and in changing them independently have hindered such identification. To discriminate between these possibilities, we adapted a liquid chromatography mass spectrometry method to identify and quantify sugar phosphates in human hepatocarcinoma cells (Hep G2) and rat hepatocytes in response to different carbon sources and in the presence/absence of a glucose-6-phosphate dehydrogenase inhibitor. We also used this method to demonstrate that these cells could not metabolize 2-deoxyglucose beyond 2-deoxyglucose-6-phosphate. The simultaneous quantification of sugar phosphates and FAS and L-PK expression levels demonstrated that both X-5-P and G-6-P play a role in the modulation of gene expression. In conclusion, this report presents for the first time a single mechanism that incorporates the effects of X-5-P and G-6-P on the enhancement of the expression of carbohydrate-responsive genes.