Brain-type glucose transporter (GLUT-1) is selectively localized to the blood-brain barrier. Studies with quantitative western blotting and in situ hybridization.

W M Pardridge,R J Boado,C R Farrell

doi:10.1016/s0021-9258(18)38267-x

Abstract

The hypothesis that the GLUT-1 glucose transporter isoform is expressed selectively in brain at the capillary endothelium, i.e. the blood-brain barrier (BBB), was tested by using quantitative Western blotting, cytochalasin B binding, and in situ hybridization in bovine brain cortex. Purified human red cell glucose transporter was used as the standard for quantitative Western blots, because the mobility of the human erythrocyte and BBB glucose transporters in electrophoretic gels was identical. The concentration of immunoreactive glucose transporter in bovine BBB plasma membranes was 10.8 +/- 0.9 pmol/mgp (mean +/- S.E., n = 6). This value was not statistically different from the estimate of the maximal binding sites of D-glucose-displaceable [3H]cytochalasin B binding in the BBB membrane preparations, 11.7 +/- 3.5 pmol/mgp. In situ hybridization experiments using 35S-labeled antisense and sense riboprobes corresponding to nucleotides 385-932 of the GLUT-1 cDNA showed prominent hybridization of the antisense probe over brain microvascular endothelium, but no hybridization over neuropil greater than that found with the 35S-labeled sense probe. These studies are consistent with the following conclusion: (a) essentially 100% of the glucose transporter binding sites at the BBB can be accounted for by the GLUT-1 isoform; (b) in situ hybridization studies confirm previous Northern blot analysis and indicate the GLUT-1 gene is expressed selectively in microvascular endothelium in brain with minimal, if any, expression of this gene in neurons or glial cells in vivo.

Highlights

The hypothesis that the GLUT-l glucose transporter isoform is expressed selectively in brain at the capillary endothelium, i.e. the blood-brain barrier (BBB), was tested by using quantitative
In situ hybridization experiments using 3SS-labeled antisense and sense riboprobes corresponding to nucleotides 385-932 of the GLUT-l cDNA showed prominent hybridization of the antisense probe over brain microvascular endothelium, but no hybridization over neuropil greater than that found with the 35S-labeled sense probe
A typical standard curve of the quantitative Western blot analysis is shown in Fig. 1, and these studies show that the standard curve is linear up to 100 ng of purified human GT (hGT) applied to each well

Summary

STUDIES WITH QUANTITATIVE WESTERN BLOTTING AND IN SITU HYBRIDIZATION’

In situ hybridization experiments using 3SS-labeled antisense and sense riboprobes corresponding to nucleotides 385-932 of the GLUT-l cDNA showed prominent hybridization of the antisense probe over brain microvascular endothelium, but no hybridization over neuropil greater than that found with the 35S-labeled sense probe These studies are consistent with the following conclusion: (a) essentially 100% of the glucose transporter binding sites at the BBB can be accounted for by the GLUT-l isoform; (b) in situ hybridization studies confirm previous Northern blot analysis and indicate the GLUT-l gene is expressed selectively in microvascular endothelium in brain with minimal, if any, expression of this gene in neurons or glial cells in vivo. The performance of the quantitative Western blot entailed the isolation of: (a) purified human erythrocyte GT as a standard, [6] isolation of BBB plasma membranes, and (c) isolation of capillary-depleted synaptosomal or brain cell (BC) membranes. Comparison of these data with estimates of total GT binding sites using [3H]cytochalasin B radioreceptor assays allows for a quantitative analysis of the expression of the GLUT-l protein in the BBB membrane fraction

PROCEDURES

RESULTS

DISCUSSION