Photolabelling of the liver-type glucose-transporter isoform GLUT2 with an azitrifluoroethylbenzoyl-substituted bis-D-mannose.

The bis-D-mannose photolabel ATB-BMPA (2-N-[4-(1-azi-2,2,2- trifluoroethyl)benzoyl]-1,3-bis-(D-mannos-4-yloxy) propyl-2-amine) has been used to radiolabel the glucose transporter present in liver plasma membranes. The labelling was inhibited by 4,6-O-ethylidene-D-glucose. Approx. 7% of the liver plasma-membrane protein that was photolabelled in a 4,6-O-ethylidene-D-glucose-inhibitable manner was specifically immunoprecipitated by either an anti-(GLUT2 C-terminal peptide) antibody or by an anti-(GLUT2 exofacial-loop peptide) antibody. After correction for non-specific labelling and precipitation, the ratio of immunoprecipitable GLUT2 to GLUT1 was approximately 5:1, suggesting that GLUT1 was not a major component of liver plasma membranes. The low levels of immunoprecipitation of the photolabelled transporter may be due to low antibody affinity for GLUT2 or may indicate that the photolabelling reagent has labelled another glucose-transporter-like protein. The hexose-transport inhibitors phloretin, cytochalasin B and 4,6-O-ethylidene-D-glucose all inhibited the photolabelling by ATB-BMPA of immunoprecipitable GLUT2. D-Glucose inhibited approx. 57% of the ATB-BMPA labelling of GLUT2. D-Fructose also inhibited the GLUT2 labelling confirming that it is a substrate for GLUT2 [Gould, Thomas, Jess & Bell (1991) Biochemistry 30, 5139-5145]. From photolabel displacement by a range of concentrations of non-labelled ATB-BMPA, the affinity constant (Kd) of ATB-BMPA was found to be 250 +/- 78 microM, whereas the Bmax. (total number of binding sites) value was 2.1 +/- 0.29 pmol of GLUT2/mg of membrane protein. Since GLUT1, GLUT4 and GLUT2 have approximately equal affinities for the external ligand ATB-BMPA, but have widely varying affinities for equilibrated and transported substrates, it is suggested that the isoforms may differ in their ability to bind hexoses at the internal site.