Resolution of the facilitated transport of dehydroascorbic acid from its intracellular accumulation as ascorbic acid.

We performed a detailed kinetic analysis of the uptake of dehydroascorbic acid by HL-60 cells under experimental conditions that enabled the differentiation of dehydroascorbic acid transport from the intracellular reduction/accumulation of ascorbic acid. Immunoblotting and immunolocalization experiments identified GLUT1 as the main glucose transporter expressed in the HL-60 cells. Kinetic analysis allowed the identification of a single functional activity involved in the transport of dehydroascorbic acid in the HL-60 cells. Transport was inhibited in a competitive manner by both 3-O-methyl-D-glucose and 2-deoxy-D-glucose. In turn, dehydroascorbic acid competitively inhibited the transport of both sugars. A second functional component identified in experiments measuring the accumulation of ascorbic acid appears to be associated with the intracellular reduction of dehydroascorbic acid to ascorbic acid and is not directly involved in the transport of dehydroascorbic acid via GLUT1. Transport of dehydroascorbic acid by HL-60 cells was independent of the presence of external Na, whereas the intracellular accumulation of ascorbic acid was found to be a Na-sensitive process. Thus, the transport of dehydroascorbic acid via glucose transporters is a Na-independent process which is kinetically and biologically separable from the reduction of dehydroascorbic acid to ascorbic acid and its subsequent intracellular accumulation.

Vitamin C is fundamental to human physiology (1)(2)(3). Since humans cannot synthesize vitamin C, it must be provided exogenously in the diet and transported intracellularly (4,5). Vitamin C is present in human blood at an average concentration of about 50 -100 M, and at least 95% is in the reduced form (ascorbic acid) with the remaining 5% in the oxidized form (dehydroascorbic acid) (5). The observation that cells and tissues accumulate characteristic intracellular concentrations of the reduced form of vitamin C, both in vivo and in vitro, suggests that the transport and cellular accumulation of vitamin C is a highly regulated process (5).
Data accumulated over a number of years pointed to the participation of glucose transporters in the cellular transport of vitamin C (6 -12), although functional evidence for the existence of additional transporters is also available (13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24). We recently demonstrated directly, by expression in Xenopus laevis oocytes, that the glucose transporters GLUT1, GLUT2, and GLUT4 are efficient transporters of dehydroascorbic acid (25). We extended these studies to show that glucose transporters are also the main pathway mediating the transport of dehydroascorbic acid in normal human neutrophils and HL-60 myeloid leukemia cells (25,26). There has been controversy regarding the issue of the identity of the form of vitamin C, reduced or oxidized, transported by human neutrophils. Freshly isolated human neutrophils contain millimolar concentrations of reduced ascorbic acid, and in vitro they accumulate high concentrations of ascorbic acid when incubated with ascorbic or dehydroascorbic acid (5,27). Current evidence indicates, however, that dehydroascorbic acid, and not ascorbic acid, is preferentially transported into cells (9, 18, 25, 26, 28 -36). We, and others, have shown that under the incubation conditions used by most laboratories, ascorbic acid undergoes oxidation to dehydroascorbic acid which is then transported intracellularly (25,26,37,38). The transported dehydroascorbic acid is reduced back to ascorbic acid, providing the mechanism for its cellular accumulation. The oxidation of ascorbic acid to dehydroascorbic acid appears to be part of the mechanism by which cells of the host defense system, such as neutrophils, increase their uptake of dehydroascorbic acid when activated by physiological stimuli.
Kinetic analysis of the uptake of ascorbic acid in human neutrophils has revealed the presence of two functional activities with different affinities for ascorbate, an observation that was interpreted as suggesting the existence of at least two separate transport systems involved in the cellular uptake of ascorbic acid (27). We detected two functional activities, one with high affinity and one with low affinity, involved in the uptake of dehydroascorbic acid in X. laevis oocytes expressing GLUT1 (25) and in human neutrophils and HL-60 cells (25,26). The X. laevis expression experiments suggested the association of both functional activities with the expression of a single transport protein.
The studies mentioned above were performed under conditions leading to the accumulation of ascorbic acid in cells incubated with dehydroascorbic acid. Uptake under these conditions is a function of both transport and intracellular trapping. No clear distinction between transport of dehydroascorbic acid and accumulation of ascorbic acid has been presented, and, therefore, it has been impossible to assign the functional activities detected to the steps of transport or accumulation. An understanding of the mechanisms that regulate the cellular content of ascorbic acid requires the identification of the events involved in the transport of dehydroascorbic acid as well as in the cellular accumulation of ascorbic acid. We studied this issue by analyzing the accumulation of ascorbic acid in HL-60 cells under experimental conditions that allowed us to dissociate the transport of dehydroascorbic acid from its intracellular accumulation as ascorbic acid. We identified GLUT1 as the glucose transporter expressed in HL-60 cells, characterized it functionally, established the conditions to measure transport as distinct from accumulation, and applied these conditions to measure the transport of dehydroascorbic acid. We now provide evidence indicating that only one functional system, corresponding to the glucose transporter GLUT1, is directly involved in the facilitated transport of dehydroascorbic acid by HL-60 cells. A second functional activity apparently associated with the trapping and accumulation of the reduced form of ascorbic acid is not directly involved in the transport of dehydroascorbic acid via the glucose transporters.

EXPERIMENTAL PROCEDURES
Cells-Human myeloid HL-60 cells were cultured in Iscove's modified Dulbecco's medium supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 2 mM L-glutamine. Cell viability was greater than 95% as assessed by trypan blue exclusion.
Uptake Studies-For uptake assays, the cells were suspended in incubation buffer (15 mM Hepes pH 7.6, 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl 2 , 0.8 mM MgCl 2 ), washed by centrifugation in the same buffer and resuspended at 0.5-2 ϫ 10 7 cells/ml. Uptake assays were performed in a final volume of 0.2 ml of incubation buffer containing 0.5-2 ϫ 10 6 cells, 0.1-0.4 Ci of L-[ 14 C]ascorbic acid (specific activity 8.2 mCi/mmol, DuPont NEN), a final concentration of 0.05-15 mM ascorbic acid, and 1-100 units of ascorbic acid oxidase (50 units/mg protein, Sigma). The mixture was incubated for 30 s or 10 min at room temperature, and uptake was stopped by adding 10 volumes of cold phosphate-buffered saline (4°C) containing 100 M phloretin (stopping solution). The cells were collected and washed twice by centrifugation in cold stopping solution. Samples were processed for scintillation spectrometry or high performance liquid chromatography as described (26). Hexose uptake assays were similarly performed using 1 Ci of D-3-O-[methyl-3 H]glucose (specific activity 10 Ci/mmol, DuPont NEN) (methylglucose) 1 and 0.3-20 mM methylglucose or 1 Ci of 2-[1,2-3 H]deoxy-D-glucose (specific activity 26.2 Ci/mmol, DuPont NEN) (deoxyglucose) and 0.3-20 mM deoxyglucose. When appropriate, competitors and inhibitors were added to the uptake assays, or the cells were preincubated in their presence.
Western Blot Analysis-HL-60 cells were homogenized in buffered Tris-HCl (pH 7.4) containing 0.5% deoxycholic acid, 0.5% Triton X-100, 1 mM EDTA, 1 mM EGTA, 5 g/ml each of aprotinin, leupeptin, and soybean trypsin inhibitor, and 1 mM phenylmethylsulfonyl fluoride. Cell debris was removed by centrifugation at 10,000 ϫ g for 10 min at 4°C, membranes were collected on a 45% sucrose cushion by centrifugation at 105,000 ϫ g for 2 h, and 100 g of protein were loaded on each lane of a sodium dodecyl sulfate-containing 10% polyacrylamide gel and electrophoresed at 20 mA for 6 h. After transfer to a polyvinylidene difluoride membrane (Immobilion, Millipore), blots were stained with anti-GLUT or with preimmune serum followed by incubation with horseradish peroxidase-anti-rabbit antibody. Proteins were visualized by enhanced chemiluminescence (Amersham).
Immunolocalization-For immunocytochemistry, the cells were collected by centrifugation (Cytospin, Shandon) onto microscope slides, fixed in buffered formaldehyde-acetone for 30 s, and washed with phosphate-buffered saline (pH 7.4). Fixed cells were incubated in a humid chamber for 60 min in phosphate-buffered saline containing 5% bovine serum albumin, followed by incubation for 2 h at room temperature in the same buffer containing 1% bovine serum albumin, 0.3% Triton X-100 and the respective anti-GLUT antiserum, or rabbit preimmune serum, diluted 1:100. The cells were then washed and incubated with anti-rabbit IgG-fluorescein (Life Technologies, Inc., 1:50) for 1 h, washed, and mounted.

RESULTS
Expression of GLUT1 in HL-60 Cells-Six facilitative glucose transporter isoforms have been described in mammalian cells of which five, GLUT1-GLUT5, appear to be expressed on the cell membrane, and one, GLUT7, is restricted in expression to the internal membranes of the endoplasmic reticulum (39). We used a panel of antibodies specific for each of five members of the family of glucose transporters (GLUT1-GLUT5) to identify the isoforms expressed in the HL-60 cells and responsible for the uptake of hexoses in these cells. By immunoblotting, the anti-GLUT1 antibodies reacted with a protein of M r ϳ 50,000 present in a plasma membrane-enriched fraction prepared from HL-60 cells (Fig. 1A) or in a preparation of total cellular proteins (data not shown). No reactivity was observed with anti-GLUT2, -GLUT3, -GLUT4, or -GLUT5 antibodies (Fig.  1A), although longer exposures of the blot revealed a weak reactivity with anti-GLUT5 antibodies. Control experiments using membranes or total cellular proteins prepared from cells expressing the different glucose transporter isoforms indicated that the antibodies recognized their respective antigens under identical experimental conditions (data not shown). Parallel immunolocalization experiments using fluorescein-labeled secondary antibodies indicated abundant expression of GLUT1 in the HL-60 cells, but failed to detect the presence of the other transporter isoforms (Fig. 1, B-C, and data not shown). Similar results were obtained using silver enhanced colloidal gold or alkaline phosphatase (data not shown). The specificity and the reactivity of each antibody were tested, with positive results, in cells expressing the different transporter isoforms.
Facilitated Transport of Methylglucose-Having established the presence of GLUT1 in the HL-60 cells, we carried out a detailed characterization of the uptake of hexoses in these cells. The aim of these studies was to develop experimental conditions that would allow differentiation between the transport of dehydroascorbic acid by the glucose transporters and the intracellular accumulation of ascorbic acid. In initial experiments, we analyzed the transport of the nonmetabolizable glucose analog methylglucose by HL-60 cells. Time course experiments revealed that transport of methylglucose by HL-60 cells occurred rapidly, with half the equilibrium concentration reached in about 5 min ( Fig. 2A). The rate of transport was linear for the first 30 s, and equilibrium was reached in about 20 min. The intracellular concentration of methylglucose never exceeded the respective extracellular concentrations even in samples incubated with the hexose for several hours. Similar results were obtained using concentrations of methylglucose from 1 to 20 mM.
Dose-response experiments examining uptake at 30 s indicated that transport of methylglucose approached saturation at millimolar concentrations of methylglucose (Fig. 2B), with an FIG. 1. Expression of GLUT1 in HL-60 cells. A, for immunoblotting, equal amounts of protein were loaded on each lane of a sodium dodecyl sulfate-containing polyacrylamide gel, electrophoresed, transferred to Immobilion, reacted with the different anti-GLUT antibodies, and visualized using horseradish peroxidase anti-rabbit IgG and enhanced chemiluminescence. B and C, for immunofluorescence, cytospin preparations were reacted with preimmune serum (panel B) or anti-GLUT1 antibodies (panel C), and visualized using fluorescein-coupled secondary antibodies. No reactivity was observed with anti-GLUT2, -3, -4, or -5 antibodies. apparent K m for transport of 8.5 mM (Fig. 2C). Uptake was decreased in the presence of deoxyglucose, a typical substrate of the facilitative glucose transporters (40). Deoxyglucose, at 3 mM, inhibited by 50% (IC 50 ϭ 3 mM) the transport of methylglucose by the HL-60 cells (Fig. 2D). Maltose, a disaccharide that binds to the glucose transporters but is not transported (40), inhibited the uptake of methylglucose with an IC 50 of approximately 25 mM, but no effect on transport was observed with fructose, sucrose, ␣-methyl-D-glucoside, or L-glucose (Fig.  2D). ␣-Methyl-D-glucoside is a substrate of the sodium-dependent glucose cotransporter (41) that is not transported by the facilitative glucose transporters, and sucrose and L-glucose do not interact with the glucose transporters (40). Cytochalasin B, but not cytochalasin E, inhibited the transport of methylglucose with an IC 50 of 200 nM (data not shown).
Facilitated Transport of Deoxyglucose-Deoxyglucose is a substrate of facilitative glucose transporters, and the transported deoxyglucose accumulates intracellularly in cells as deoxyglucose-6-phosphate (40). Discrimination between transport and accumulation is usually accomplished experimentally by carrying out very short uptake assays or by measuring the uptake of low micromolar concentrations of deoxyglucose. Two components were observed when the time course of the uptake of 5 mM deoxyglucose was measured in the HL-60 cells (Fig. 3,  A and B). The amount of cell-associated radioactivity increased rapidly at short incubation times, and the rate of increase was linear for the first minute of incubation. Thereafter, the rate of incorporation decreased and a second component was observed that was linear for at least 30 min. Only one component, that was linear for at least 30 min, was observed in experiments measuring the time course of uptake of 2 mM deoxyglucose (Fig.  3A). We interpreted these results as indicating that the initial linear phase of incorporation of radioactivity by the HL-60 cells represented the transport of deoxyglucose, with the second linear component representing the phosphorylation and intracellular accumulation of phosphorylated deoxyglucose. Doseresponse experiments examining uptake at 30 s indicated that transport of deoxyglucose reached saturation at millimolar concentrations of deoxyglucose ( Fig. 3C) with an apparent K m for transport of 2.5 mM (Fig. 3D). Uptake was competed by methylglucose and maltose, with IC 50 of 10 and 30 mM, respectively, but not by fructose, sucrose, ␣-methyl-D-glucoside, or L-glucose ( Fig. 3E). No measurable uptake of fructose was observed in cells incubated in the presence of radiolabeled fructose (data not shown). Cytochalasin B, but not cytochalasin E, inhibited the transport of deoxyglucose with an IC 50 of 250 nM (data not shown).
Facilitated Transport of Dehydroascorbic Acid-We analyzed the characteristics of facilitated transport of dehydroascorbic acid in HL-60 cells distinct from the accumulation of ascorbic acid. Uptake of dehydroascorbic acid proceeded in an apparently linear fashion for at least 10 min, with no evidence of saturation at 60 min (Fig. 4A). Two kinetic components were observed, however, when the time course of the uptake of dehydroascorbic acid by the HL-60 cells was measured at short incubation periods (Fig. 4B). Initially, the amount of cell-associated radioactivity increased very rapidly, and the rate of increase was linear for the first minute of incubation. Afterwards, the rate of incorporation decreased, and a second kinetic component that was linear for at least 10 min was observed. The amount of cell-associated radioactivity after 10 min of incubation with radioactive dehydroascorbic acid greatly exceeded the amount expected at equilibrium after facilitated transport. The two kinetic components were observed in uptake experiments using concentrations of dehydroascorbic acid ranging from 20 M to 4 mM, but only one component was observed at 10 M dehydroascorbic acid (data not shown). We interpreted these results as indicating that the initial linear phase of incorporation of radioactivity by the HL-60 cells represented the transport of dehydroascorbic acid, with the second, slower linear component representing the reduction of the recently transported dehydroascorbic acid to ascorbic acid leading to the intracellular accumulation of reduced ascorbic acid. Thus, the reduction of dehydroascorbic acid to ascorbic acid is the rate-limiting step when measuring uptake at concentrations of ascorbic acid equal to or greater than 20 M and when using uptake assays lasting longer than 1 min. Alternatively, measuring uptake at incubation times of less than 30 s gives information on transport at any concentration of dehydroascorbic acid.
Dose-response studies using 30-s uptake assays revealed that the transport of dehydroascorbic acid by the HL-60 cells saturated at about 4 mM (Fig. 4C). These studies also revealed  the presence of a single component with an apparent K m of 0.85 Ϯ 0.12 mM (n ϭ 8), and a V max of 4 nmol/min/10 6 cells for the transport of dehydroascorbic acid by the HL-60 cells (Fig. 4D). Parallel studies measuring uptake of dehydroascorbic acid using 10-min assays revealed the presence of one component saturating at about 15 mM of dehydroascorbic acid. A more detailed examination of the dose-response curve, however, revealed the presence of two components, one saturating at about 4 mM dehydroascorbic acid, and a second one saturating at about 15 mM dehydroascorbic acid (Fig. 4C). The Lineweaver-Burk analysis revealed the presence of two components, with apparent uncorrected K m values of 0.9 and 3.5 mM and V max of 1.6 and 2 nmol/min/10 6 cells, respectively (Fig. 4D). HL-60 cells incubated for 10 min with dehydroascorbic acid showed the presence of an additional high affinity component involved in uptake, but the kinetic analysis failed to reveal the presence of the high affinity component when measuring transport of dehydroascorbic acid using 30-s assays (Fig. 4, E and F).
Competition experiments using 30-s transport assays indicated that deoxyglucose, methylglucose, and maltose inhibited the transport of dehydroascorbic acid by the HL-60 cells with IC 50 of 2, 8, and 15 mM, respectively, but no effect of L-glucose, fructose, sucrose, or ␣-methyl-D-glucoside was observed (Fig.  4G). Cytochalasin B, but not cytochalasin E, inhibited transport with an IC 50 of approximately 200 nM (data not shown). Dehydroascorbic acid, but not ascorbic acid, inhibited the transport of methylglucose by the HL-60 cells in a dose-dependent manner, with an IC 50 of about 1 mM (Fig. 5A). The inhibition of the transport of methylglucose by dehydroascorbic acid was competitive, with a K i of 0.8 mM (Fig. 5, B and C). We also determined the effect of deoxyglucose on the cellular uptake of concentrations of dehydroascorbic acid, from 20 M to 5 mM, using incubation periods from 30 s to 10 min. Deoxyglucose completely blocked the uptake of dehydroascorbic acid under every condition tested, confirming that GLUT1 is the only pathway for the transport of dehydroascorbic acid by HL-60 cells (data not shown).
Na ϩ Independent Transport of Dehydroascorbic Acid-We explored the potential existence of other transporters involved in the transport of dehydroascorbic acid or reduced ascorbic acid in the HL-60 cells. It has been reported that the transport of ascorbate into several tissues and cells is a Na ϩ -sensitive process (5, 13-17, 19 -24, 42), but no evidence for the presence of a Na ϩ -dependent ascorbate transporter in human blood cells has been provided. We studied this issue in the HL-60 cells by measuring the uptake of dehydroascorbic acid in the absence of Na ϩ . We used choline chloride, LiCl, and sucrose to replace NaCl in the incubation medium. A substantial decrease in the uptake of dehydroascorbic acid by the HL-60 cells was observed when uptake was measured in the absence of Na ϩ and in the presence of choline ϩ (Fig. 6A). The uptake of dehydroascorbic acid in the presence of Na ϩ (controls) increased almost linearly for the first 30 min of incubation, and approximated a plateau at 120 min. In the presence of choline ϩ , however, uptake reached a plateau much earlier, and total uptake at 120 min was only 30% of the uptake observed in the control cells (Fig.  6A). The effect of Na ϩ was most evident after the first 15 min of incubation, suggesting that it is related to the accumulation of ascorbic acid rather than to the transport of dehydroascorbic acid. Consistent with the previous interpretation, short uptake experiments indicated that the transport of dehydroascorbic acid, as defined by the first 30 s of the uptake curve, was not affected by the absence of Na ϩ in the incubation medium, and decreased uptake was noticeable after the first 2 min of incubation (Fig. 6B). Similar results were obtained in experiments in which sucrose (Fig. 6C) or LiCl (Fig. 6D) were used to replace NaCl in the incubation medium, confirming that the transport of dehydroascorbic acid is Na ϩ -independent.
We further analyzed this issue by measuring uptake in HL-60 cells incubated with ascorbic acid in the absence of Na ϩ . HL-60 cells incubated with ascorbic acid accumulate lower levels of ascorbate (as compared to cells incubated with dehydroascorbic acid) due to the oxidation of ascorbic acid to the transported form, dehydroascorbic acid (26). Uptake in the presence of Na ϩ proceeded in a linear fashion for the length of the incubation period (120 min), but accounted for less than 5-7% of the uptake of dehydroascorbic acid under similar experimental conditions (Fig. 6, E and A). A decrease of approximately 40% in uptake was observed in the absence of Na ϩ at 120 min, but no effect was evident during the first 45 min of incubation (Fig. 6E). In the presence of choline ϩ , there was a small but reproducible stimulation of uptake during the first 30 min of incubation, as compared to control cells (Fig. 6F). Short uptake experiments indicated that, in the presence of Na ϩ , the cells failed to accumulate any radioactivity during the first 30 s of incubation, but uptake increased rapidly afterwards, compatible with the time-dependent generation of the transported substrate dehydroascorbic acid (Fig. 6G). Similar uptake kinetics were observed in cells incubated in the presence of choline ϩ , but uptake was always greater than in the control cells (Fig.  6G). Again, uptake under these conditions (cells incubated with ascorbic acid) was only a fraction of the uptake observed in cells incubated with dehydroascorbic acid (Fig. 6H). Similar results were obtained when sucrose or LiCl were used to replace NaCl in the incubation buffer (data not shown). Overall, these data are consistent with the concept that GLUT1, a transporter whose functional activity is Na ϩ -independent, is the only pathway involved in the cellular uptake of dehydroascorbic acid by the HL-60 cells. DISCUSSION Our data indicate that the accumulation of ascorbic acid in HL-60 cells is a process that can be dissociated into at least two components with characteristic kinetics. The initial step consists of the facilitated transport of dehydroascobate, followed by its reduction and intracellular accumulation as ascorbic acid. The transport component of uptake was very rapid and could be detected only by using very short uptake assays. At extended incubation periods, the rate-limiting step of uptake was the cellular trapping/reduction and accumulation of ascorbic acid, and it was no longer possible to measure transport separately and as distinct from accumulation. Transport was mediated by GLUT1, the member of the family of facilitative glucose transporters expressed in the HL-60 cells.
Although six different facilitative glucose transporters have been identified in mammalian cells (39), we identified GLUT1 as the main facilitative glucose transporter expressed by HL-60 cells. The HL-60 cells were able to transport deoxyglucose, a substrate specific for the facilitated glucose transporters that is not transported by the sodium-glucose cotransporters (40,41). In addition, ␣-methyl-D-glucoside, the specific substrate of the sodium-glucose cotransporter (41), did not inhibit the transport of deoxyglucose or methylglucose by the HL-60 cells. The specificity of the transporter was further confirmed by the lack of effect of L-glucose and sucrose and the inhibitory effect of maltose on the transport of deoxyglucose and methylglucose. The HL-60 cells were unable to transport fructose, and transport of methylglucose and deoxyglucose was not affected by fructose. These findings are consistent with the immunological evidence indicating that the HL-60 cells do not express GLUT2 and GLUT5. The inhibitory effect of nanomolar concentrations of cytochalasin B on the transport of deoxyglucose and methylglucose confirmed the lack of expression of GLUT2 and GLUT5 by HL-60 cells. GLUT5 transports fructose but not deoxyglucose, and its functional activity is not affected by cytochalasin B (43). GLUT2 transports fructose and deoxyglucose, and its functional activity is affected by micromolar but not by nanomolar concentrations of cytochalasin B (44). Furthermore, the K m for the transport of deoxyglucose or methylglucose by GLUT2 is in the range of 15 to 30 mM, as opposed to 3-8 mM for the transporter expressed by the HL-60 cells. Direct evidence for the presence of GLUT1 in HL-60 cells was provided by the results of immunoblotting and immunolocalization experiments with specific antibodies that indicated a clear reactivity with anti-GLUT1 and an absence of reactivity with anti-GLUT2, -GLUT3, -GLUT4, and -GLUT5 antibodies.
Our transport data are consistent with the models presented in Fig. 7. Short uptake assays enabled us to carry out a detailed kinetic analysis of the transport of dehydroascorbic acid as distinct from the accumulation of ascorbic acid. Only one functional component, with an apparent K m of 0.75 mM, was involved in the transport of dehydroascorbic acid by HL-60 cells. The competition studies indicated that dehydroascorbic acid was transported by a facilitative mechanism which, together with the identification of GLUT1 as the main glucose transporter present in the HL-60 cells, strongly supports the concept that GLUT1 mediates the transport of dehydroascorbic acid in FIG. 6. Sodium-independence of the uptake of dehydroascorbic acid by HL-60 cells. A, time course of the uptake of dehydroascorbic acid in the presence of NaCl (E) or choline chloride (q). B, time course of the uptake of dehydroascorbic acid in the presence of NaCl (E) or choline chloride (q). C, time course of the uptake of dehydroascorbic acid in the presence of NaCl (E) or sucrose (q). D, time course of the uptake of dehydroascorbic acid in the presence of NaCl (E) or LiCl (q). E, time course of the uptake of ascorbic acid in the presence of Na ϩ (E) or choline ϩ (q). F, time course of the uptake of ascorbic acid in the presence of Na ϩ (E) or choline ϩ (q). G, time course of the uptake of ascorbic acid in the presence of Na ϩ (E) or choline ϩ (q). H, time course of the uptake of ascorbic acid (AA, E) or dehydroascorbic acid (DHA, q). these cells. The competition studies, especially the absence of effect of ␣-methylglucoside on the transport of dehydroascorbic acid, indicated the lack of participation of a transporter with the functional characteristics of the sodium-glucose cotransporter in the transport of dehydroascorbic acid by HL-60 cells.
The existence of several components involved in the uptake of dehydroascorbic acid by HL-60 cells was evident only in long uptake experiments that are a complex function of the transport of dehydroascorbic acid and the intracellular trapping of ascorbic acid. We failed to detect the presence of a second, high-affinity component involved in the transport of dehydroascorbic acid by HL-60 cells. We observed, however, a highaffinity component when measuring the accumulation of ascorbic acid in cells incubated with dehydroascorbic acid for periods of time equal to or longer than 10 min. Under these conditions, the intracellular trapping of ascorbic acid rather than transport is the rate-limiting step for uptake. An additional lowaffinity component was also observed in these studies, adding another level of complexity to the overall process of accumulation of ascorbic acid. Competition experiments showed that the transport of dehydroascorbic acid was inhibited by methylglucose and deoxyglucose, and the specificity of this inhibition was demonstrated by the results indicating that other sugars un-able to interact with the glucose transporters had no effect on the transport of dehydroascorbic acid. Competition experiments, using 10-min incubation assays to measure accumulation, indicated that hexoses completely blocked the accumulation of ascorbic acid in HL-60 cells by a noncompetitive mechanism. We interpreted these results as indicating that glucose acts directly on the primary event responsible for the entry of dehydroascorbic acid, that is GLUT1-mediated transport, not on the functional component(s) involved in the intracellular accumulation of ascorbic acid. Therefore, the high-and low-affinity components detected in the 10-min uptakes are likely associated with the reduction of dehydroascorbic acid permitting the intracellular accumulation of ascorbic acid. Support for the interpretation that the high-affinity component is not related to the glucose transporter is provided by the observation that low concentrations of dehydroascorbic acid (Ͻ50 M) did not affect the transport of methylglucose or deoxyglucose. A high-affinity component has been identified in experiments measuring the accumulation of ascorbic acid in different cellular systems and the hypothesis has been advanced that it corresponds to a high-affinity transporter of reduced ascorbic acid (27). These studies were performed, however, under experimental conditions that led to the oxidation of ascorbic acid with generation of the transported form, dehydroascorbic acid. Of major concern is the fact that in such studies the kinetic constants were derived from long uptake experiments (90 min), raising the question of the identity of the rate-limiting step under those conditions (27). Our data indicating that deoxyglucose was able to completely inhibit the cellular uptake of dehydroascorbic acid in a range of experimental conditions, including different concentrations of dehydroascorbic acid and different times of incubation, are consistent with the concept that GLUT1 is the only means by which the HL-60 cells transport dehydroascorbic acid.
Our results also indicate that HL-60 cells do not express a Na ϩ -sensitive ascorbate transporter. Although we observed a major decrease in uptake of dehydroascorbic acid by HL-60 cells in the absence of Na ϩ , the kinetic data clearly showed that the step inhibited in the absence of Na ϩ was the accumulation of ascorbic acid, and not the transport of dehydroascorbic acid via GLUT1. Na ϩ -sensitive ascorbate transporters have been proposed to exist in several tissues and cells such as adrenomedullary chromaffin cells, retinal epithelial cells, osteoblasts, and the brush-border membrane of reabsorbing renal epithelia (5, 13-17, 19 -23, 42). In these studies, incubating the cells in the absence of external Na ϩ produced a marked decrease in the uptake of ascorbic acid. Evidence for Na ϩ -sensitive uptake of ascorbate was recently provided by expression studies in X. laevis oocytes injected with poly(A) ϩ RNA extracted from rabbit kidney cortex (24).
Taken together, our data indicate that the transport of dehydroascorbic acid via the glucose transporters is kinetically and biologically separable from the reduction of dehydroascorbic acid to ascorbic acid and its subsequent intracellular accumulation. These findings have major significance for our understanding of the biological control mechanisms involved in the homeostasis of the cellular content of ascorbic acid in human cells. We can envision the existence of separate mechanisms that regulate the ability of a cell to transport dehydroascorbic acid and its capacity to reduce the transported substrate intracellularly and accumulate ascorbic acid. The issue of the mechanisms involved in the intracellular reduction of dehydroascorbic acid and the concomitant accumulation of ascorbic acid has been controversial. No dehydroascorbic acid reductase activity has been consistently detected in cells or human tissues that accumulate intracellularly millimolar con- FIG. 7. Models for differentiating transport from accumulation during uptake experiments. A, the initial rate of transport of a substrate, such as methylglucose, that is not metabolized in the interior of the cell, corresponds to the early linear phase of the uptake curve. As the intracellular concentration of the recently transported substrate increases, efflux becomes significant until its rate equals the rate of influx. Under steady state conditions, the intracellular and extracellular concentrations of the transported substrate are identical, and the net rate of transport in any direction is zero. The kinetic analysis is done using data obtained at short uptake times in which the ratelimiting step of uptake is transport. OMG, methylglucose. B, the situation is more complex for a substrate such as dehydroascorbic acid that is transported intracellularly and then trapped by reduction to the non-transported species ascorbic acid. The early linear phase of the uptake curve represents the initial rate of transport, but a second linear phase that represents the trapping component and may last for a long period of time is also evident. Trapping is the rate-limiting step when using long uptake assays because the rate of transport exceeds the rate of trapping, and part of the transported dehydroascorbic acid will be transported back out of the cell. At steady state, the intracellular concentration of the trapped, non-transported species, may exceed by severalfold the extracellular concentration of the transported substrate. In a typical uptake experiment using radioactive dehydroascorbic acid, this result may give the false impression that the transported substrate, dehydroascorbic acid, is accumulating against a concentration gradient. Thus, long uptake experiments measure the accumulation of the non-transported species and its respective associated kinetic constants, rather than the transport of the substrate handled by the transporter. Similar considerations apply to the uptake of deoxyglucose. AA, ascorbic acid; DHA, dehydroascorbic acid. centrations of ascorbic acid when incubated in the presence of dehydroascorbic acid (45,46). The existence of a dehydroascorbate reductase has been clearly demonstrated in plants that use reduced glutathione to catalyze the reduction of dehydroascorbate to ascorbate, and a similar activity has been partially characterized in bovine iris-ciliary body (47). In humans, it has been proposed that the tripeptide glutathione, which is present at millimolar concentrations in cells, may play a central role in maintaining extracellular and intracellular ascorbic acid in its reduced state (48). Experimental observations in animal models have revealed a close interrelationship between the cellular content of ascorbate and glutathione. Guinea pigs and newborn rats cannot synthesize ascorbic acid and it must be provided in the diet (4). Newborn rats, made glutathione deficient by treatment with L-buthionine-(S,R)-sulfoximine, showed a decreased content of ascorbic acid in tissues, and supplementation with ascorbic acid in the diet induced increased tissue levels of ascorbic acid and glutathione (49). In guinea pigs fed an ascorbic acid-free diet, supplementation with glutathione monoester, a compound that increases the cellular levels of glutathione, delayed the onset and the development of the pathologic complications of scurvy (50). These observations established a close functional correlation between the respective in vivo levels of glutathione and ascorbic acid, but did not provide information about the mechanisms involved. On the other hand, in vitro observations have established that glutathione is able to reduce dehydroascorbic acid to ascorbic acid in the absence of any enzymatic activity through a direct chemical reaction (37,51). No direct evidence is available, however, indicating that glutathione is the physiological reducer of dehydroascorbic acid in cells or that it is directly involved in the ability of cells to take up dehydroascorbic acid and accumulate ascorbic acid (52). Recently, it was described that two well known and widely expressed enzymes, protein disulfide isomerase and glutaredoxin, possess dehydroascorbate reductase activity (53), raising the possibility that they could be involved in the reduction-dependent intracellular accumulation of ascorbic acid in human cells. In fact, the characterization and purification of two different dehydroascorbate reductase activities present in rat liver led to the identification of glutaredoxin as one of the proteins (54). The ability to differentiate the transport of dehydroascorbic acid from its accumulation as ascorbic acid offers an opportunity to address these and related issues in a controlled cellular system amenable to experimentation.