Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Hexose transport and phosphorylation capillaries isolated from rat brain A. LORRIS BETZ, JUDIT CSEJTEY, AND GARY Departments of Neurology and Pediatrics, University San Francisco, California 94143 by W. GOLDSTEIN of California School of Medicine, tubular segments of metabolically active microvessels (19) exhibiting energy-dependent rubidium transport (17) and Na+-dependent and Na+-independent amino acid transport (9>. Thus, this preparation is useful for the study of transport and metabolism in brain capillaries and provides a valuable tool for further characterization of the BBB. Glucose transport across the BBB has been well studied in vivo (3-8, 10-12, 20, 27-B). An earlier report from our laboratory on sugar uptake into brain capillaries in vitro (18) described an apparent transport system with much higher affinity and different stereospecificity than is reported for the BBB transport of glucose in vivo. However, this study did not adequately distinguish between transport and subsequent phosphorylation of the sugar. The present study further characterizes glucose transport into brain capillary endothelial cells and examines the relationship between sugar transport and phosphorylation. blood-brain barrier; glucose; Z-deoxyglucose; 3-O-methylglucase; endothelial cells; cytechalasin B; phloretin; phlorizin; accelerative exchange diffusion; glucose transport stereospecCity Isolation of capillaries. The method used for isolation of capillary segments from rat brain was a minor modification of the one described previously (19) Twenty to forty Sprague-Dawley rats, 30-days old and weighing loo-125 g, were killed by cervical dislocation. The brains were immediately removed and placed in a buffer consisting of oxygen-saturated Ringer solution with 1.2 mM MgCl,!, 15 mM N-2-hydroxy-ethylpiperazine-N’-2-ethanesulfonic acid (HEPES), pH 7.4, 5 mM sodium pyruvate, and 1 % fraction V bovine serum albumin. The brainstem cerebellum, and meninges were discarded and cortical shells, free of choroid plexus and ependyma, were minced in btier. A 10% (wt/vol> homogenate was made using 20 up-and-down strokes in a Teflon and glass homogenizer (0.25 mm clearance) at 390 rpm. The homogenate was centrifuged at 1,000 x g for 10 min. To remove cellular debris and myelin, the pellet was resuspended to a concentration of 16% (wt/ vol) in the same buffer now containing 25% bovine serum albumin and centrifuged at 1,000 x g for 15 min. The new pellet consisted of a mixture of free nuclei and capillary segments of various sizes. To obtain a. more uniform suspension of capillaries, the pellet was resuspended in buffer and then passed through a 118~pm nylon mesh under gentle suction. The capil laries were separated from nuclei by passing the suspension of a selective permeability barrier between blood and brain is well established. Although the cells or structures that constitute this blood-brain barrier (BBB) are not known with certainty, many investigators propose that the capillary endothelial cells limit movement of solutes into the brain This hypothesis is based on ultrastructural studies showing that brain capillaries form a continuous layer of endothelial cells joined together by tight junctions (31). Furthermore, when macromolecules such as horseradish peroxidase (31) or microperoxidase (30) are injected into the bloodstream, they penetrate up to, but not through, the tight junctional complexes. However, there is no direct evidence that the endothelial cells are responsible for the selective cerebral uptake of low-molecular-weight compounds. The recent development of methods for isolating brain capillaries (19) offers an opportunity for the in vitro study of cells that are the probable site of the BBB in vivo. The isolated capillary preparation consists of short THE EXISTENCE C96 MATERIALS 0363-6143/79/0000-0000$01.25 AND METHODS l Copyright 0 1979 the American Physiological Society Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 BETZ, A. LORRIS, JUDIT~SEJTEY, ANDGARY W. GOLDSTEIN. Hexose transport and phosphoryla tion by capillaries isolated from rat brain. Am. J. Physiol. 236(l): C96-ClO2, 1979 or Am. J. Physiol.: Cell Physiol. 5(l): C96-C102, 1979, -Hexose transport and phosphorylation were studied in capillary segments isolated from rat brain. Uptake of 3-@methyl-n-glucose (3MG) could be inhibited by cytochalasin 13, phloretin, and phlorizin, but not by 2,4-dinitrophenol or ouabain. 2-Deoxy-nglucose (2DG), D-ghCOSe, galactose, and mannose inhibited 3MG uptake, while L-glucose, fructose, and ribose did not. Accelerative exchange diffusion of 3MG was demonstrated. At equilibrium, the intracellular concentration of hexose did not exceed the external concentration, and transport was, therefore, equilibrative rather than accumulative. Transport of 2DG and n-glucose was not rate limiting for metabolism. When incubated in 5 mM n-glucose, the endothelial cells contained a large pool of free glucose. L-Glucose entered capillaries more slowly than other hexoses and served as a marker for simple diffusion of sugars into the cells. Our results suggest that sugar uptake into isolated brain capillaries occurs by a transport system similar to the one responsible for glucose transport across the blood-brain barrier in vivo. SUGAR TRANSPORT INTO ISOLATED BRAIN c97 CAPILLARIES mM 3MG, 80 mM mannitol, and 50 @/ml [:‘H]3MG. Measurement of metabolic cowersion of sugars. Radiolabeled metabolites of glucose were separated from free glucose by ion-exchange chromatography (25). Immediately after terminating glucose uptake and washing the cells, the filters were placed in 1.5 ml boiling water for 5 min. The cooled samples were centrifuged at 1,000 x g for 10 min to remove denatured proteins. We separated amino acids from the supernatant by column chromatography using Dowex AG-SOW resin. The effluent was passed through a second column containing Dowex AG-1X8 to remove carboxylic acids and sugar phosphates. When 2-deoxy-D-glucose (2DG) was used, only the Dowex AG-1X8 column was necessary because 2DG is phosphorylated, but not further metabolized (33) m Determination of the water space. A suspension of capillaries was preincubated for 15 min at 37OC and then centrifuged for 5 min at 200 x g. The pellet was quickly resuspended in iced buffer containing 2 x 10” cpm/ml [14C]L-glucose, and aliquots were added to each of four Pasteur pipettes with their narrow ends plugged with parafilm. The loaded pipettes were centrifuged for 3 min at 1,000 x g at 4°C. The pellet was separated from the supernatant by breaking the pipette through the pellet. The packed cells were added to preweighed aluminum pans, weighed, dried to constant weight in a vacuum oven at 4O”C, and reweighed. The dried pellets were dissolved in 0.1 N NaOH and the amount of 14C in a neutralized aliquot was determined. The intracellular water space was calculated as the difference between the wet and dry weights minus the L-glucose water space. It was assumed that L-glucose remained completely extracellular during this experiment. We calculated the water space of the contaminating erythrocytes using the hemoglobin content of the preparation (0.034 mg hemoglobin/mg protein) and a mean corpuscular hemoglobin concentration of 34 g/100 ml of packed erythrocytes. MuteriaZs. The following materials were obtained from New England Nuclear Corp. (Boston, MA): 23-O-[methyE-H]methyl-n-glu[“H(G)]deoxy-n-glucose, case, D- [:6-:sH]glucose, and L- [lJ*C]glucose. Phloretin was purchased from K & K Laboratories (Plainview, NY); 2,4-dinitrophenol from Mallinckrodt (St Louis, MO); D-mannose from SchwarzlMann (Orangeburg, NY); and insulin from Eli Lilly and Co. (Indianapolis, IN). All other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO). l RESULTS Properties of 3-0-methylglucose uptake. 3MG was used as model substrate for characterization of the glucose transport system. This sugar is nonmetabolizable (13) and has been shown to compete with D-glucose for BBS transport in several species (5, 10, 12, 27, 29). Figure 1 shows a time course for the uptake of 5 mM 3MG at 25°C and its inhibition by 0.05 mM cytochalasin B uptake of the nontransported sugar L-glucose was used as a measure of the rate of simple diffusion into the preparation. Cytochalasin B is an inhibitor of glu- Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 through a 1.2 x 1.5 cm column containing 0.25mm glass beads. Nuclei were not retained by the beads and could be removed by washing with buffer. The capillaries remained attached to the beads and were recovered by gentle agitation in buffer. After the beads settled, the capillaries were collected by centrifugation at 500 x g for 5 min. The quality of each preparation was judged by its appearance under phase microscopy. We determined cell protein by the method of Lowry et al. (26) after overnight solubilization in 1% sodium dodecyl sulfate, using bovine serum albumin as a standard. Measurement of sugar uptake. The capillaries were diluted with the preparative buffer to a concentration of approximately 2 mg cell protein/ml, kept on ice, and used within 2 h. Assays for sugar uptake were performed at 37°C or 25OC in 1.5-ml plastic test tubes. Cell suspension (0.06 ml) was brought to temperature and the reaction was started by the rapid addition of 0.18 ml buffer containing 3.3 or 6.7 &i of :jH-labeled sugar and various amounts of unlabeled sugar. The reaction was terminated by the rapid addition of a 0.2-ml aliquot of the incubation mixture to 5 ml iced stopping solution (see below) and immediate filtration through a Whatman GF/A glass-fiber filter. The filter and cells were washed twice with 5 ml stopping solution and then placed in liquid scintillation vials containing 0.5 ml water and 10 ml scintillation fluid. Samples for background radioactivity were obtained by adding 0.05 ml cell suspension to iced stopping solution before the addition of 0.15 ml of the isotope solution. The stopping solution, based on a similar solution used to terminate sugar transport in red blood cells (16), contained Ringer solution, 1 mM HgCly, 0.1 mM phloretin, and 1% (vol/vol) ethanol. Preliminary experiments demonstrated that use of this solution resulted in a 10% increase in the amount of radioactivity retained by the cells. Quantitative retention of capillaries by the glassfiber filters was also demonstrated. These filters have the advantage of permitting rapid filtration with low nonspecific binding of labeled sugars. The entire procedure of stopping and washing could be completed in less than 6 s. Sugar uptake in the presence of inhibitors. The effect of glucose analogues on 3-O-methyl-n-glucose (3MG) uptake was determined by incorporating inhibiting sugars into the isotope solution. When uptake was measured in the presence of cytochalasin B, phloretin, or phlorizin, 0.01 ml inhibitor in 24% (vol/vol) ethanol was added to 0.05 ml cell suspension before addition of the isotope solution. Control incubations contained the same amount of ethanol without inhibitor. The final ethanol concentration was 1% (vol/vol) in these studies. The effects of 2,4-dinitrophenol, ouabain, and insulin were examined after a 30-min preincubation at 37°C in the presence of these compounds. Accelerative exchange diffusion of 3MG was demonstrated by suspending capillaries in either 100 mM 3MG or 100 mM mannitol. The cells were preincubated at 37°C for 30 min to allow equilibration and then brought to 25°C. Uptake of [sH]3MG was initiated by adding isotope solutions containing unlabeled 3MG and mannitol at concentrations desiffned to result in final external concentrations of 20 BETZ, CSEJTEY, AND GOLDSTEIN 1 1.o 0.5 mm TIME (min) case transport in several other cell systems including Even at 25”C, 3MG uptake was nearly half-equilithe BBB (15). In isolated brain capillaries it clearly brated by 10 s. Because this was the earliest time point inhibited 3MG transport and reduced the initial uptake that could be reliably measured using the filt&ion assay, it was not possible to determine transport kinetto nearly that of the diffusion marker (Fig. 1). The final equilibrium water space for 3MG was 2.2 ics based on initial velocities. Although a kinetic anal~1 water/mg capillary protein. This is less than the ysis was not possible, the relative inhibition by sugar total intracellular water space of 4.2 $/mg protein t 0.7 transport inhibitors and glucose analogues could be (SD) that was determined as the difference between wet determined. Table 1 shows that 3MG uptake was more and dry weights with L-glucose used as a correction for effectively inhibited by phloretin than by phlorizin. extracellular water. A similar discrepancy has been Cytochalasin B was the most potent inhibitor tested. There was no inhibition of 3MG uptake in the presence reported by Kimmich (22) for isolated intestinal epithelial cells. For these cells, the equilibrium 3MG space of ouabain or 2,4-dinitrophenol ,, and insulin did not enhance uptake. Table 2 shows that 3MG uptake was was 2.3 $/mg protein while the total intracellular water space calculated by the difference between wet inhibited by high concentrations of ZDG, D-glucose, galactose, mannose, and to a lesser extent, xylose. and dry weights minus the polyethylene glycol water space was 5.0 ,ul/mg protein. The reason for a difference There was no significant inhibition by L-glucose, frucbetween the equilibrium sugar space and the total cell tose, or ribose. The data in Fig. 2 indicate that 3MG uptake could be water space is not clear. It is possible that the sugar is excluded from intracellular organelles, such as the stimulated by preloading the cells with a high concennucleus or mitochondria. On the other hand, it may be tration of 3MG. This phenomenon, accelerative exchange diffusion has been described for several other that some of the intracellular [“H]3MG is lost during filtration and washing of the cells; however, we have carrier-mediated glucose transport systems, including the BBB (6). not been able to demonstrate this effect experimentally Relationship between sugar uptake and phosphoryl(results not shown). Furthermore, using the filtration method, we have previously demonstrated that the ation. 2DG is-a glucose analogue that can be phosphoequilibrium water space for amino acids that are not rylated by mammalian cells, but not further metaboconcentrated by the capillaries is 4.2 pl/mg protein (9). lized (33). It is also an effective inhibitor of glucose transport across the BBB (4,27,29), and thus is a useful The erythrocyte water space was less than O-1 pl/mg protein in our preparation and thus red blood cells made model substrate for a study of the relationship between sugar transport and phosphorylation in isolated capila negligible contribution to the total 3MG uptake. Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 FIG. 1. Time course for uptake at 25°C of 5 mM 3-0-methylglucose (a), 5 mM 3-0-methylglucose in presence of 0.05 mM cytochalasin B (0) and 5 mM L-glucose (A). Insert: expanded view of 1st min of uptake. Data shown are averages of 3 determinations -+ SD. SUGAR TRANSPORT TABLE 1, Effect 3-0-methylglucose INTO of inhibitors BRAIN B on Uptake, nmollmg 0.05 0.50 0.50 1.00 0.50 0.10 U/ml 0.24 1.64 2.90 4.33 4.31 k * 1 f f 3.60 2 0.64 Percent Control 0.64 0.33 0.57 0.73 0.75 6 37 67 97 96 80 2. Inhibition by glucose analugues Added Sugar 2-Deoxy-n-glucose D-Glucose 3-@methyl-D-glucose n-Galactose D-Mannose D-Xylose D-Fructose D-Ribose L-Glucose Control of 3-0-methylglucose Uptake, nmollmg 0.94 1.22 1.23 1.57 1.92 2.30 2.96 3.31 3.39 3.26 k * 4 k -Ik k + 4 k 0.37 0.36 0.35 0.45 0.71 0.33 0.78 0.32 0.38 0.38 uptake Percent Control tion of 2DG-PO4 have been reported for rat kidney slices (23) Figure 3 shows the time course for 2DG uptake and phosphorylation by capillaries with 0.1 mM 2DG at 37°C. This concentration of 2DG was used to more clearly define the relationship between transport and phosphorylation as well as to permit comparison with our previous results (18). The rapid equilibrat.ion of free 2DG with an intracellular water space of 2.2 $/mg protein was followed by a slower accumulation of 2DGPO,. These data suggest that transport was not rate limiting for metabolism at a 2DG concentration of 0.1 mM. Figure 4 shows similar results with D-glucose at the physiologic concentration of 5 mM. The free sugar equilibrated with a water space of 2.1 pl/mg protein and this process was more rapid than metabolism. DISCUSSION The movement of glucose between blood and brain has been extensively studied in vivo. It occurs via a mediated transport system (3-8, 10-12, 20, 27-29) with very little free diffusion (8). The kinetic constants for transmort fall within a relatively narrow range from 29 38 38 48 59 70 91 101 104 100 Values are averages of 4 determinations * SD. We corrected for diffusion using the uptake of [:iH]L-glucose (Fig. 1). Results are expressed as percent of control containing 100 mM D-mannitol to correct for osmotic effects. Uptake of [“H]3MG was determined after a 30-s incubation in the presence of 5 mM 3MG and 100 mM inhibiting sugar. lary endothelial cells. Capillaries were incubated with [:‘H]2DG for various periods of time and then the reaction was stopped, the cells washed, and the retained radioactivity fractionated into free 2DG and 2DG-PO4 by ion-exchange chromatography. Preliminary studies showed that the method by which the radioactivity was released from the cells was very important (Table 3). When cells were incubated for 30 min at 37OC in 0.1 mM 2DG, washed, and then sonicated to release radioactivity, we recovered only 10.4% of the intracellular label as 2DGPO,. However, release of intracellular radioactivity by sonication, as had been done in our previous study (la), would not necessarily protect 2DG-PO4 from dephosphorylation by glucose-6-phosphatase and other phosphatases, e.g., alkaline phosphatase, which are known to be present in brain capillaries (19). When the washed cells were immediately placed in boiling water to inactivati enzymes and then sonicated, 94.8% of the intracellular radioactivity was recovered as 2DG-PO4 (Table 3). Thus, the use of boiling water to inactivate phosphatases avoids dephosphorylation of intracellular 2DGPO,. Similar artifacts caused by rapid dephosphoryla- 10 0.25 0.50 0.75 TIME 1.0 (min) FIG. 2. Accelerative exchange diffusion of 30methylglucose. Capillaries were preincubated with either 100 mM 3MG (a) or 100 mM mannitol (0). Uptake of 3MG was then determined in presence of 20 mM 3MG and 80 mM mannitol at 25°C. Data shown are averages of 3 determinations &SD. TABLE 3, Importance of boiling in recovery of phosphorylated 2-deoxyglucose from capillary cells Release by Sonication Boiling, sonication Percent 2DG 89.5 k 8.2 5.2 k 0.5 Values are averages of 3 determinations * SD. incubated at 37°C with 0.1 mM 2DG for 30 min. filtered, washed, and intracellular radioactivity either sonication alone or by treating with boiling sonication. 2DG and 2DG-6-Po4 were separated chromatography on Dowex AG-1X8. Total sugar nmollmg k 0.10 SD. Percent 10.5 94.8 2DG-6-PO, 2 3.0 4 3.0 Capillaries were Cells were then was released by water and then by ion-exchange uptake was 4.34 Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 Values are averages of 4 determinations -+- SD; we corrected for diffusion using uptake of [iJH]L-glucose (Fig. 1). Uptake of [:$HJ3MG was determined after a 30-s incubation at 25°C in the presence of 5 mM 3MG. Control uptake for studies with cytochalasin B, phloretin, and phlorizin was 4.34 nmollmg * 0.32 SD. There was a 30-min preincubation at 37°C at the stated concentration of inhibitor for those experiments in which ouabain, 2,4-dinitrophenol, or insulin was present. Corresponding control sample was also preincubated and had an uptake of 4.47 nmollmg * 0.81 SD. TABLE c99 CAPILLARIES uptake Concentration, mM Inhibitor Cytochalasin Phloretin Phlorizin Ouabain 2,4-Dinitrophenol Insulin ISOLATED BETZ, 0.5 0.1 DP / /D I 1 I I 2 TIME 3 I 4 1 5 (An) 3. Time course for uptake of 0. I mM 2-deoxyglucose. After incubation at 37”C, intracellular radioactivity was separated into free 2DG (a) and 2DG-6-P04 (0) by ion-exchange chromatography. Data shown are averages of 3 determinations *SD. FIG. species to species. Reported values for the apparent K,, are within a range of 6-9 mM while the apparent V,,,,, varies from 1.2 to 3.0 (pmol/g whole brain)/min (3, 8, 11, 20, 28, 29). The stereospecificity for BBB glucose transport is well described in rat (29) and dog (4). Although phlorizin and phloretin are both competitive inhibitors of glucose transport into brain, phloretin is more potent than phlorizin (4). In contrast, Na+-dependent sugar transport in intestine (14) and kidney (1) is much more sensitive to phlorizin than phloretin. Cytochalasin B, which is a potent noncompetitive inhibitor of glucose transport into dog brain (E), has no effect on Na+-dependent glucose transport in the intestine (21). Sugar transport into brain is unaffected by ouabain (29) or insulin (8) and exhibits accelerative exchange diffusion when the brain glucose concentration is increased (6). Further more, transport of glucose into brain is not rate limiting for subsequent brain metabolism (7). In general, BBB glucose transport is thought to be mediated by a Na+-independent facilitated diffusion system similar to the one in the erythrocyte (7). Our results for 3MG uptake by brain capillaries in vitro are entirely consistent with studies of BBB sugar transport in vivo. The stereospecificity data shown in Table 2 are nearly identical to the in vivo data from rat (29) and dog (4). The relative potency of cytochalasin B, phloretin, and phlorizin are also the same (4, X5), and there was no effect of ouabain, 2,4-dinitrophenol (29), or insulin (8). Both isolated capillaries (Fig. 2) and the BBB in vivo (6) show accelerative exchange diffusion AND GOLDSTEIN for sugars. Furthermore, in isolated capillaries, transport is equilibrative and is not rate limiting for metabolism. Kolber and Morel1 (24) have recently presented preliminary data that also show equilibrative 3MG uptake into isolated brain capillaries. It would be useful to compare the kinetics of glucose transport into isolated capillaries with the kinetics of transport across the BBB. This kind of data could not be obtained with our studies because the rate of D-glucose uptake by capillaries is too rapid at 37°C. Thus, at 5 mM, D-glucose, uptake was half-equilibrated by 10 s. Reducing the temperature would lower the rate of uptake, but also significantly alter both the K,, and V H-liis (32). An alternate approach utilizes integrated rate equations and time courses of uptake to describe the kinetics of transport (16). Unfortunately, the significant rate of simple diffusion into the isolated capillary preparation (Fig. 1, L-glucose) makes such equations prohibitively complex. The first study of sugar transport into isolated brain capillaries by Goldstein et al. (18) presented results inconsistent with data from in vivo studies or with the results of the present study. The time course for 2DG uptake was linear for more than 5 min and the apparent K,,, for this process was 0.093 mM. In addition, a study of stereospecificity showed inhibition by D-fructose, but not by D-galactose. These properties are opposite to those found for BBB glucose transport in vivo (4), and are, in fact, similar to the properties of mammalian hexokinase (35). Other investigators have recently reported that 2DG transport into isolated capillaries is 12 10 2 TIME (min) 4. Time course for uptake of 5 mM D-glucose. After incubation at 37”C, intracellular radioactivity was separated into D-glucose (a) and glucose metabolites (0) by ion-exchange chromatography. Data shown are averages of 3 determinations *SD. FIG. Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 0.4 .-c aI ‘0 & f 0.3 \ ;2 0 E = 0.2 CSEJTEY, SUGAR TRANSPORT INTO ISOLATED BRAIN Cl01 CAPILLARIES require little energy because it is not concentrative or Na+-dependent and does not require phosphorylation of the sugar. Our data also indicate that there is a sizable pool of free glucose in endothelial cells under physiological conditions (Fig. 4). Furthermore, sugar transport into capillaries and, presumably, also out of capillaries, occurs very rapidly. This would permit rapid response of the glucose supply to changes in cerebral metabolism. In conclusion, this study demonstrates the usefulness of isolated brain capillaries for the investigation of transport phenomena thought to occur at the BBB. Our observation that hexose transport into isolated brain capillaries is mechanistically similar to glucose transport across the BBB does not by itself prove that the BBB is located at the capillary endothelial cell. However, we anticipate that further studies with the isolated capillary preparation will more firmly establish this relationship. We thank A. Ste. Marie for her excellent technical assistance and I. Diamond and A. Gordon for their helpful review of the manuscript. This work was supported by the National Foundation of the March of Dimes, and Grants ES-01164 and HL-22088 from the National Institutes of Health. A. L. Betz is the recipient of the National Institutes of Health postdoctoral fellowship NS-05807. G. W. Goldstein is the recipient of the National Institutes of Health Research Career Development Award NS-00278. Received 30 May 1978; accepted in final form 17 August 1978. REFERENCES 1. ALVARADO, F. Hypothesis for the interaction of phlorizin and phloretin with membrane carriers for sugars. Biochim. Biophys. Acta 135: 483-495, 1967. 2. ANCHORS, J. M., D. F. HAGGERTY, AND M. L. KARNOVSKY. Cerebral glucose-6-phosphatase and the movement of 2-deoxy-nglucose across cell membranes. J. Biol. Chem. 252: 7035-7041, 1977. 3. BACHELARD, H. S., P. M. DANIEL, E. R. LOVE, AND 0. E. PRATT. The transport of glucose into the brain of the rat in vivo. Proc. R. Sot. London Ser. B 183: 71-82, 1973. 4. BETZ, A. L., L. R. DREWES, AND D. D. GILBOE. Inhibition of glucose transport into brain by phlorizin, phloretin and glucose analogues. Biochim. Biophys. Acta 406: 505-515, 1975. 5. BETZ, A. L., AND D. D. GILBOE. Kinetics of cerebral glucose transport in vivo: inhibition by 3-O-methylglucose. Bruin Res. 65: 368-372, 1974. 6. BETZ, A. L., D. D. GILBOE, AND L. R. DREWES. Accelerative exchange diffusion kinetics of glucose between blood and brain and its relation to transport during anoxia. Biochim. Biophys. Acta 401: 416-428, 1975. 7. BETZ, A. L., D. D. GILBOE, AND L. R. DREWES. The characteristics of glucose transport across the blood-brain barrier and its relation to cerebral glucose metabolism. In: Advances in Experimental Medicine and Biology, edited by G. Levi, L. Battistin, and A. Lajtha. New York: Plenum Press, 1976, p- 133-149. 8. BETZ, A. L., D. D. GILBOE, D. L. YUDILEVICH, AND L. R. DREWES. Kinetics of unidirectional glucose transport into the isolated dog brain. Am. J. Physiol. 225: 586-592, 1973. 9. BETZ, A. L., AND G. W. GOLDSTEIN. Polarity of the blood-brain barrier: neutral amino acid transport into isolated capillaries. Science 202: 225-227, 1978. 10. BIDDER, T. G. Hexose translocation across the blood-brain interface: configurational aspects. J. Neurochem. 15: 867-874, 1968. 11. BUSCHIAZZO, P. M., E. B. TERRELL, AND D. M. RECEN. Sugar transport across the blood-brain barrier. Am. J. Physiol. 219: 1505-1513, 1970. 12. CUTLER, R. W. P., AND J. C. SIPE. Mediated transport of glucose between blood and brain in the cat. Am. J. Physiol. 220: 1182- 1186,1971. 13. CZAKY, T. Z., AND J. E. WILSON. The fate of 3-U-14CH,-glucose in the rat. Biochim. Biophys. Acta 22: 185-186, 1956. 14. DIEDRICH, D. F. Competitive inhibition of intestinal glucose transport by phlorizin analogs. Arch. Biochem. Biophys. 117: 248-256, 1966. 15. DREWES, L. R., R. W. HORTON, A. L. BETZ, AND D. D. GIL~OE. Cytochalasin B inhibition of brain glucose transport and the influence of blood components on inhibitor concentration. Biochim. Biophys. Acta 471: 477-486, 1977. 16. EILAM, Y., AND W. D. STEIN. Kinetic studies of transport across red blood cell membranes. In: Methods in Membrane Biology, edited by E. D. Korn. New York: Plenum Press, 1974, vol. 2, p. 283-354. 17. GOLDSTEIN, G. W. Relation of potassium transport to oxidative metabolism in isolated brain capillaries. J. Physiol. London. In press. GOLDSTEIN, G. W., J. CSEJTEY, AND I. DIAMOND. Carrier-mediated glucose transport in capillaries isolated from rat brain. J. Neurochem. 28: 725-728, 1977. 19. GOLDSTEIN, G. W., J. S. WOLINSKY, J. CSEJTEY, AND I. DIAMOND. Isolation of metabolically active capillaries from rat brain. J. Neurochem. 25: 715-717, 1975. 20. GROWDON, W. A., T. S. BRATTON, M. C. HOUSTON, H. L. TARPLEY, AND D. M. REGEN. Brain glucose metabolism in the intact mouse. Am. J. PhysioZ. 221: 1738-1745, 1971. 21. HOPPER, U., K. &GRIST-NELSON, E. AMMANN, AND H. MURER. Differences in neutral amino acid and glucose transport between brush border and basolateral plasma membrane of intestinal epithelial cells. J. CeU. Physiol. 89: 805-810, 1976. 22. KIMMICH, G. A. Preparation and characterization of isolated intestinal epithelial cells and their use in studying intestinal transport. In: Methods in Membrane Biology, edited by E. D. Korn. New York: Plenum Press, 1974, vol. 5, p. 51-113. 23. KLEINZELLER, A., AND E. M. MCAVOY. Transport and phosphorylation of D-galactose in renal cortical cells. Biochim. Biophys. Acta 455: 109-125, 1976. 24. KOLBER, A. R., AND P. MORELL. Monosaccharide transport in Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017 affected by anoxia (34). UsiAg more refined methods we now show that these previous studies did not adequately account for phosphorylation of ZDG. Thus, they did not distinguish -between transport and metabolism. Our findings also point out that 2DG is not always a satisfactory glucose analogue for transport studies. It should be used cautiously and only after a thorough study of its phosphorylation and dephosphorylation by the tissue under investigation. Anchors et al. (2) have recently demonstrated rapid phosphorylation of 2DG in isolated brain capillaries after IO- to 30-s incubations at 37°C; however, they could not detect free 2DG in the cells. These results are in direct contrast to our data. It is likely that the centrifugation method used bY Anchors et al. to wa .sh the ce11s at the end of their tra nsport experiments resulted in loss of all intracellular 2DG, while the 2DG-PO4 was trapped. In fact, this phenomenon was the reason we found it necessary to stop the transport reaction by filtration and to wash the cells with a solution containing transport inhibitors. The data presented in this study are consistent with the hypothesis that the BBB for sugars is located at the brain capillary. Thus, glucose would be transported from plasma into the endothelial cell cytosol and then out of the endothelial cell into the brain interstitial fluid before being available for neuronal and glial metabolism. Glucose transport across capillaries would Cl02 BET& capillaries 219, 1977. 25. 26. 27. 28. 29. 30. isolated from rat brain. Sot. for Neurosci. A bstr. 3: A. V. The uptake and metabolism of n-[U-14C]glucose by the choroid plexus. Brain Res. 112: 435-441, 1976. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275, 1951. OLDENDORF, W. H, Brain uptake of radiolabeled amino acids, amines, and hexoses after arterial injection. Am. J. Physiol. 221: 1629-1639, 1971. PAPPENHEIMER, J. R., AND B. P. SETCHELL. Cerebral glucose transport and oxygen consumption in sheep and rabbits. J. Physiol. London 233: 529-551, 1973. PARDRIDGE, W. M., AND W. II. OLDENDORF. Kinetics of bloodbrain barrier transport of hexoses. Biochim. Biophys. Acta. 382: 377-392, 1975. REESE, T. S., N. FEDER, AND M. W. BRIGHTMAN. Electron microscopic study of the blood-brain and blood-cerebrospinal LORENZO, 31. 32. 33. 34. 35. CSEJTEY, AND GOLDSTEIN fluid barriers with microperoxidase. J. Neuropathol. Exp. Neurol. 30: 137-138, 1971. REESE, T, S., AND M. J. KARNOVSKY. Fine structure localization of a blood-brain barrier to exogenous peroxidase. J. CeZL. BioZ. 34: 207-217, 1967. SEN, A. K., AND W. F. WIDDAS. Determination of the temperature and pH dependence of glucose transfer across the human erythrocyte membrane measured by glucose exit. J. Physiol. London 160: 392-403, 1962. SOLS, A., AND R. K. CRANE. Substrate specificity of brain hexokinase. J. Biol. Chem. 210: 581-595, 1954. SPATZ, M., D. MICIC, AND I. KLATZO. Effect of anoxia on :$H-2deoxy-n-[“HJglucose uptake in isolated cerebral capillaries. Sot. for Neurosci. Abstr. 3: 324, 1977. WALKER, D. G. The nature and function of hexokinases in animal tissues. In: Essays in Biochemistry, edited by P. N. Campbell and G. D. Greville. New York: Academic, 1966, vol. 2, p. 33-67. Downloaded from http://ajpcell.physiology.org/ by 10.220.32.247 on June 17, 2017