Ingrid Monden

Subcellular distribution and activity of glucose transporter isoforms GLUT1 and GLUT4 transiently expressed in COS-7 cells.

In adipose and muscle cells, the glucose transporter isoform GLUT4 is mainly located in an intracellular, vesicular compartment from which it is translocated to the plasma membrane in response to insulin. In order to test the hypothesis that this preferential targeting of a glucose transporter to an intracellular storage site is conferred only by its primary sequence, we compared the subcellular distribution of the fat/muscle glucose transporter GLUT4 with that of the erythrocyte/brain-type glucose transporter GLUT1 after transient expression in COS-7 cells. Full-length cDNA was ligated into the expression vector pCMV that is driven by the cytomegalovirus promoter, and introduced into COS cells by the DEAE-dextran method. Cells were homogenized and fractionated by differential centrifugation to yield plasma membranes and a Golgi-enriched fraction of intracellular membranes (low-density microsomes). In these membrane fractions, the abundance of glucose transporters was assessed by immunoblotting with specific antibodies against GLUT1 and GLUT4, and their transport activity was assayed after solubilization and reconstitution into lecithin liposomes. Uptake rates of 2-deoxyglucose assayed in parallel samples were higher in cells expressing GLUT1 or GLUT4 as compared with control cells (transfection of pCMV without transporter cDNA). Reconstituted glucose transport activity in plasma membranes was about 5-fold higher after expression of GLUT1 and GLUT4 as compared with control cells. The relative amount of GLUT4 in the low-density microsomes as detected by reconstitution and immunoblotting exceeded that of the GLUT1, but was much lower than that observed in typical insulin-sensitive cells, e.g., rat fat cells or 3T3-L1 adipocytes. These data indicate that COS-7 cells transfected with glucose transporter cDNA express the active transport proteins and can be used for functional studies.

Functional consequences of the autosomal dominant G272A mutation in the human GLUT1 gene

The first autosomal dominant missense mutation (G272A) reported within the human GLUT1 gene and shared by three affected family members was investigated in respect to functional consequences. Substitution of glycine‐91 by site‐directed mutagenesis with either aspartate or alanine resulted in a significant decrease in transport activity of GLUT1 expressed in Xenopus oocytes. Expression of mutant transporters was confirmed by immunoblot, 2‐deoxy‐glucose uptake and confocal laser microscopy. The data agree with 3‐O‐methyl‐glucose uptake into patient erythrocytes and indicate that the loss of glycine rather than a hydrophilic side chain (Gly91Asp) defines the functional consequences of this mutation.

From triple cysteine mutants to the cysteine-less glucose transporter GLUT1: a functional analysis

Two triple cysteine mutants containing Cys‐less N‐ or C‐terminal halves and the Cys‐less GLUT1 were generated by site‐directed mutagenesis. Following expression in Xenopus oocytes, the intrinsic transport activities of the multiple cysteine mutants were slightly decreased when either the cysteine residues of the C‐terminal half or all six residues were changed; substitution of serine for cysteine residues located at the N‐terminal half was without consequence for the catalytic activity. The exofacial ligand ethylidene glucose inhibited 2‐deoxy‐d‐glucose uptake of wild‐type and Cys‐less GLUT1‐expressing Xenopus oocytes with comparable half‐saturation constants (11.5 and 13.2 mM). However, each of the multiple cysteine mutants exhibited an increase in affinity for the endofacial inhibitor cytochalasin B, with the greatest effect being observed for the Cys‐less construct (decrease in K i by the factor 5–6).

The differential role of Cys-421 and Cys-429 of the Glut1 glucose transporter in transport inhibition by p-chloromercuribenzenesulfonic acid (pCMBS) or cytochalasin B (CB)

Cys‐421 and Cys‐429 of Glut1 were replaced by site‐directed mutagenesis in order to investigate their involvement in basal glucose transport and transport inhibition. Neither of the two cysteine residues was essential for basal 2‐deoxy‐D‐glucose uptake in Xenopus oocytes expressing the respective mutant M421 and M429. If applied from the external side, the poorly permeable sulfhydryl‐reactive agent pCMBS inhibited 2‐deoxy‐D‐glucose uptake of Glut1 ‐ and M421‐expressing Xenopus oocytes but failed to affect uptake of the Cys‐429 mutant. This is in agreement with the proposed two‐dimensional model or Glut1 confirming that Cys‐429 is the only residue exposed to the surface of the plasma membrane. The replacement of Cys‐421 at the exofacial end of helix eleven caused a partial protection of 3‐O‐methylglucose transport inhibition by CB; this residue may thus be involved in stabilizing an adjacent local tertiary structure necessary for the full activity of this inhibitor.

The large cytoplasmic loop of the glucose transporter GLUT1 is an essential structural element for function.

Abstract Alanine scanning mutagenesis and the introduction of deletions and insertions were used to address the role of the large cytoplasmic loop in 2-deoxyDglucose (2-DOG) uptake by GLUT1 expressed in Xenopus oocytes. Alanine scanning mutagenesis of 29 amino acid residues that are identical or homologous in GLUT1 to GLUT4 demonstrated that the transport activities of only a few variants were affected. Progressive truncation of the loop by six deletions leaving intact 59 (Δ236241), 49 (Δ231246), 39 (Δ226251), 28 (Δ221257), 18 (Δ216262), or 10 (Δ213267) amino acid residues resulted in a progressive decrease in 2-DOG uptake. Compared with wildtype GLUT1 the uptake rates varied between 33% for the Δ236241 mutant and 4% for the Δ213267 mutant. Insertional mutagenesis using hexaalanine or hexaglycine to fill in the deletion 236D-241L restored 2-DOG uptake to 73% of wildtype GLUT1 in the case of hexaalanine, whereas hexaglycine insertion was without effect. Confocal laser microscopy demonstrated that a deletion of six amino acid residues did not influence the expression level in the plasma membrane (Δ236241 mutant), whereas the plasma membrane fluorescence of the Δ213267 mutant was comparable with that of waterinjected Xenopus oocytes. Computeraided secondary structure prediction of the loop suggested that it consists of a long αhelix bundle interrupted or kinked by the highly conserved glycine-233.

Rapid Cellular Regulation of D‐Glucose Transport in Cultured Neural Cells

Abstract: Previous studies have revealed two different kinds of regulation of glucose utilization in cell lines derived from the nervous system (Keller et al., 1981). We found glucose metabolism of C‐6 glioma cells to be limited and regulated by membrane transport. In contrast, glucose utilization of C‐1300 neuroblastoma (N2A) cells was limited by the known regulatory enzymes of the Embden‐Meyerhof pathway. Under the given experimental conditions the “membrane‐limited” C‐6 glioma cells were characterized by periodically changing glucose transport rates and very low intracellular glucose concentrations, which remained constant in spite of widely differing transport rates. These findings suggest the close functional coupling between transport and phosphorylation required for the regulation of glucose transport by cellular metabolic needs.

Functional consequences of an in vivo mutation in exon 10 of the human GLUT1 gene

The functional consequences of an in vivo heterozygous insertion mutation in the human facilitated glucose transporter isoform 1 (GLUT1) gene were investigated. The resulting frameshift in exon 10 changed the primary structure of the C‐terminus from 42 in native GLUT1 to 61 amino acid residues in the mutant. Kinetic studies on a patients erythrocytes were substantiated by expressing the mutant cDNA in Xenopus laevis oocytes. K m and V max values were clearly decreased explaining pathogenicity. Targeting to the plasma membrane was comparable between mutant and wild‐type GLUT1. Transport inhibition by cytochalasin B was more effective in the mutant than in the wild‐type transporter. The substrate specificity of GLUT1 remained unchanged.