Studies on the structure and function of a mammalian sugar transport protein

The GLUT1 isoform of the mammalian passive glucose transporter family is of paramount importance in satisfying the intracellular demand for sugar by many tissues. In the absence of a three-dimensional structure at atomic resolution, the central aim of this project was to utilise a combination of methodologies to examine the structure/function relationship of GLUT1. From an amino acid sequence alignment of the sugar transporter family, the two-dimensional model of GLUT1 was refined using several predictive algorithms. Furthermore, a three-dimensional model of the protein was constructed in accord with the hypothesis that the current transporter phenotype arose from the gene duplication of an ancestral six-helix protein. Experimental evidence in favour of this arrangement was obtained by the co-expression of the N- and C-terminal halves of GLUT1 in Xenopus oocytes. Additional evidence in support of the predicted model for GLUT1 was obtained using a membrane-impermeant biotin derivative, which identified an exofacial lysine residue within the C-terminal half of GLUT1. In order to examine further the topography of the transporter, mutants of GLUT1 possessing additional sites for biotinylation were prepared for expression in CHO cells. However, analysis of a CHO cell line over-expressing wild-type GLUT1 revealed that, although high expression of functionally active GLUT1 could be achieved, the majority of the expressed protein was not targeted to the plasma membrane. These findings therefore precluded its use in topographical analyses. Nevertheless, the mutagenesis strategy also enabled the modification of proteolytic cleavage sites to aid the identification of the binding sites for transport inhibitors, such as ATB-BMPA. These mutants were shown to be functionally active in Xenopus oocytes, and one of them was characterised further by expression in the Sf9/baculovirus system. This mutant retained the ability to be labelled with ATB-BMPA and, furthermore, a labelled fragment was successfully immunopurified providing definitive evidence for the C-terminal location of the ATB-BMPA-binding site. It should now be possible to identify precisely the site of labelling by N-terminal sequencing