Seven clones from the Caco-2 cell line, three isolated from passage 29 (PD7, PD10, PF11) and four from passage 198 (TB10, TC7, TF3, TG6), all of them selected on the basis of differences in the levels of expression of sucrase-isomaltase and rates of glucose consumption, were analysed for the expression of hexose-transporter mRNAs (SGLT1, GLUT1-GLUT5) in relation to the phases of cell growth and the associated variations of the rates of glucose consumption. All clones showed a similar pattern of evolution of the rates of glucose consumption, which decreased from the exponential to the late-stationary phase, but differed, in a 1-40-fold range, in the values observed at late postconfluency. According to these values, clones could be divided into high- (PD10, PF11) and low-glucose-consuming cells (PD7, TB10, TC7, TF3 and TG6). GLUT1 and GLUT3 mRNAs were expressed in all clones and showed a similar pattern of evolution: their level decreased, from the exponential to the stationary phase, in close correlation with the decrease in rates of glucose consumption, with only high-glucose-consuming clones maintaining high levels in the stationary phase. In contrast, SGLT1, GLUT2 and GLUT5 mRNAs were only expressed, like sucrase-isomaltase mRNA, in the low-glucose-consuming clones, and their level increased from the exponential to the stationary phase, in parallel with the differentiation of the cells. GLUT4 was undetectable in all the clones. Glucose deprivation generally resulted in a discrete decrease in the levels of all transporter mRNAs in all clones, one exception being GLUT2, which in the high-glucose-consuming clones is only detectable when the cells are grown in low glucose. These clones should be ideal tools with which to study in vitro, at the single-cell level, how these transporters concur to the utilization and transport of hexoses and how their exclusive or co-ordinated expression is regulated.

No abstract

The human colon carcinoma cell line Caco-2 was used as an enterocyte model to study the expression of the facilitative glucose transporters GLUT-1 and GLUT-2, and of the putative hexose transporter GLUT-5, which are expressed specifically in the gut. Northern blots indicate that Caco-2 cells express GLUT-1 and GLUT-5 mRNAs but not the mRNA coding for the basolateral glucose transporter GLUT-2. The level of GLUT-5 mRNA is growth dependent, being detectable only in postconfluent differentiated cells. In addition, the expression of GLUT-5 increases with the number of cell passages and is approximately 10 times higher in later passages (passage 184) than in early ones (passage 26). With the use of polyclonal antibodies directed against the COOH-terminus of GLUT-5, indirect immunofluorescence and Western blotting indicate that GLUT-5 is mainly localized to the brush border of Caco-2 cells. GLUT-5 is also found to be associated with the brush border of epithelial cells from fetal and normal adult human small intestine, but is absent from the colon.

The effect of cyclic AMP on the expression of the fructose transporter, GLUT5, was studied in Caco-2 cells, a human colon cancer cell line that differentiates spontaneously in culture into cells with the properties of small intestine enterocytes. Treatment of differentiated Caco-2 cells with 50 microM forskolin, which stimulates adenylate cyclase and raises intracellular cyclic AMP levels, increased fructose uptake 2-fold and raised GLUT5 protein and mRNA levels 5- and 7-fold respectively. The increased GLUT5 mRNA levels in forskolin-treated cells are a result of stabilization of GLUT5 mRNA in these cells and increased transcription. The effect of cyclic AMP on GLUT5 transcription was assessed by measuring the activity of human GLUT5 promoter-reporter gene constructs in forskolin-treated differentiated Caco-2 cells. The results showed that forskolin stimulated the activity of the GLUT5-reporter gene constructs and this stimulatory effect was mediated by cis-acting regulatory sequences.