INTRODUCTION

Humans cannot synthesize vitamin C, and therefore the vitamin must be obtained from external sources and transported intracellularly (1, 2). In mammals with the capacity to synthesize vitamin C de novo, synthesis is restricted to the liver, and therefore other tissues must express membrane transport systems that mediate the cellular uptake of the vitamin. Reliable kinetic analysis of the transport and accumulation of vitamin C in human cells has proved difficult because of the chemical properties of the vitamin. In solution, ascorbic acid undergoes oxidation to dehydroascorbic acid, a process that is reversible (2). The oxidation of ascorbic acid can be catalyzed and greatly accelerated by traces of metals in the solution and prevented in the presence of metal chelators or reducing agents, such as dithiotreithol or glutathione (3-5). The oxidation of ascorbic acid in solution has not always been taken into consideration when designing transport or accumulation studies, making the interpretation of transport data difficult (6). Notwithstanding these problems, transport studies have revealed that mammalian cells possess at least two different systems involved in the cellular uptake of vitamin C. Cells from the adrenal glands and the small intestine, and also some osteoblast cell lines, express vitamin C transporters with the capacity to transport ascorbic acid, the reduced form of the vitamin (2, 7, 8). In these specialized tissues, transport of ascorbic acid appears to be an active process driven by a Na⁺ electrochemical gradient and mediated by a Na⁺-dependent cotransporter whose molecular identity is not known (7-9). On the other hand, human neutrophils, erythrocytes, and HL-60 cells express vitamin C transporters with the capacity to transport only the oxidized form of the vitamin, dehydroascorbic acid (10-14). Data accumulated over several years pointed to the participation of glucose transporters in the cellular uptake of dehydroascorbic acid (11, 15-20). We recently demonstrated that human neutrophils and HL-60 cells transport dehydroascorbic acid through facilitative glucose transporters (5, 21-23). We also showed, by expression in Xenopus laevis oocytes, that the glucose transporters GLUT1, GLUT2, and GLUT4 are efficient transporters of dehydroascorbic acid but lack the capacity to transport ascorbic acid (21). Given the presence of facilitative glucose transporters in all mammalian cells and tissues, the data point to the mammalian facilitative glucose transporters as the universal transporters of dehydroascorbic acid in human cells.

Vitamin C is present in human blood at a concentration of approximately 50 µM, with greater than 95% of it in the reduced form, ascorbic acid (2). As in blood, only ascorbic acid appears to be present intracellularly in human cells (2). The concentration of ascorbic acid in human cells and tissues can exceed that in the blood by one order of magnitude, and its steady-state levels appear to be tightly controlled in a cell-specific manner. Organs such as the adrenal gland accumulate high intracellular concentrations of ascorbic acid by a process thought to be energy-dependent, through Na⁺-ascorbate cotransporters (7). Host defense cells such as human neutrophils also accumulate intracellular concentrations of ascorbic acid in the millimolar range (2, 5, 21, 22). In vitro, human neutrophils and HL-60 cells incubated in the presence of dehydroascorbic acid accumulate high intracellular concentrations of ascorbic acid in a matter of minutes, but no dehydroascorbic acid is detected intracellularly. HL-60 cells are able to accumulate high concentrations of ascorbic acid by a mechanism involving the facilitated transport of dehydroascorbic acid down a concentration gradient through the glucose transporters, followed by the reduction of dehydroascorbic acid to ascorbic acid and the intracellular trapping of ascorbic acid that cannot be transported by the glucose transporters (22).

The observation that only ascorbic acid is present intracellularly in mammalian cells and tissues regardless of the chemical form, reduced or oxidized, that is transported suggests that mammalian cells possess very efficient cellular mechanisms for the regeneration of ascorbic acid through the reduction of dehydroascorbic acid. The molecular identity and the functional characteristics of the cellular mechanisms involved in the reduction of dehydroascorbic acid in human cells remain controversial (24-28). The tripeptide glutathione (N-(N-L--glutamyl-L-cysteinyl)glycine) is the predominant nonprotein thiol in cells (29). In vivo as well as in vitro evidence points to an important role for glutathione in the cellular recycling of ascorbic acid and the maintenance of the vitamin in its reduced state. Studies using animals, experimentally rendered scorbutic or glutathione-depleted, have revealed a close functional link between the redox cycles of glutathione and ascorbic acid (30-33). Like humans, guinea pigs and newborn rats cannot synthesize ascorbic acid (1). Newborn rats rendered glutathione-deficient by treatment with L-buthionine-(S,R)-sulfoximine (BSO)¹ showed decreased concentrations of ascorbic acid in tissues (30, 31). When glutathione-deficient newborn rats were supplemented with ascorbic acid, they were found to have increased levels of glutathione, and in adult mice capable of de novo ascorbic acid synthesis, glutathione deficiency resulted in increased hepatic synthesis of ascorbic acid (32). The ability of glutathione to maintain ascorbic acid has also been explored. In guinea pigs fed a scorbutigenic diet, the onset and pathological features of scurvy could be delayed or prevented by supplementation with glutathione monoester, a compound that raises intracellular concentrations of glutathione (33). The exact nature of the functional interaction between glutathione and ascorbic acid in vivo remains controversial. The hypothesis has been advanced that the direct nonenzymatic chemical reaction between glutathione and dehydroascorbic acid may be sufficient to explain the ability of mammalian cells to reduce dehydroascorbic acid, because high concentrations of reduced GSH can convert dehydroascorbic acid to ascorbic acid in a cell-free system (34). There is also evidence, however, for the existence of several enzymatic activities in mammalian cells with the capacity to reduce dehydroascorbic acid. For instance, the enzymes glutaredoxin and protein disulfide isomerase have glutathione-dependent dehydroascorbate reductase activity (26, 35), and several glutathione-dependent dehydroascorbate reductases have been identified in rat liver and bovine eye (28, 36). Moreover, a dehydroascorbate reductase activity that is independent of glutathione but requires NADPH has also been shown to exist in rat liver, a finding that indicates that several redundant and independent systems could be involved in the reduction of dehydroascorbic acid in mammalian cells (27).

We analyzed the possible role of glutathione in the accumulation of ascorbic acid in HL-60 cells that transport only dehydroascorbic acid. We found that cells treated with BSO and diethyl maleate under conditions that resulted in a two to three order of magnitude decrease in cellular glutathione showed no alteration in their ability to accumulate millimolar concentrations of ascorbic acid when incubated with dehydroascorbic acid. The dose and temperature dependence of the direct chemical reduction of dehydroascorbic acid by glutathione was also clearly different from the cellular reduction of dehydroascorbic acid. The data indicate that the cellular reduction of dehydroascorbic acid by HL-60 cells is enzymatic in nature and cannot be explained by a direct chemical reaction between glutathione and dehydroascorbic acid. The data are also consistent with the concept that HL-60 cells express highly efficient dehydroascorbic acid reductase activities that are not dependent on glutathione.

EXPERIMENTAL PROCEDURES

Human myeloid HL-60 cells were cultured in Iscove's modified Dulbecco's medium containing 10% fetal bovine serum, 1% penicillin/streptomycin, and 1% L-glutamine (22).

Cells were incubated with BSO (Sigma) or diethyl maleate (Sigma) for the time indicated in the respective figure legends before resuspension in incubation buffer (3 mM Hepes (pH 7.5), 27 mM NaCl, 1 mM KCl, 0.35 mM CaCl₂, 0.16 mM MgCl₂) at 1-6 × 10⁶ cells/ml for uptake essays or directly processed for glutathione determination.

For ascorbate uptake, 1-6 × 10⁶ cells were added to incubation buffer containing 50 µM ascorbic acid, 0.2 µCi of L-[1-¹⁴C]ascorbic acid (DuPont NEN). For dehydroascorbic acid uptake, 2-5 units of ascorbate oxidase (Sigma) were added to the incubation mix and incubated for 2 min before adding the cells (22). The oxidation of ascorbic acid was monitored by the decrease in absorbance at 266 nm and also by high performance liquid chromatography (HPLC; see below). For hexose uptake, the incubation medium contained 0.2 mM 2-deoxy-D-glucose (deoxyglucose) and 0.5 µCi 2-[1,2-³H]-2-deoxy-D-glucose, or 1 mM 3-O-methylglucose (methylglucose) and 0.5 µCi of ³H-labeled 3-O-methylglucose (22). Following incubation, samples were diluted in cold phosphate-buffered saline, centrifuged at 4 °C, and rapidly washed twice with cold Ca²⁺- and Mg²⁺-free phosphate-buffered saline. After lysis in 10 mM Tris-HCl (pH 8.0) containing 0.2% SDS, the incorporated radioactivity was determined by liquid scintillation spectrometry.

HL-60 cells (5 × 10⁶) were lysed in 60% methanol, 1 mM EDTA (pH 8.0). Lysates were stored at 70 °C until use. HPLC analyses were performed using a Whatman strong anion exchange reverse phase column (Partisil 10 SAX, 4.6 mm × 25 cm, 10-µm particle) with a silica preconditioning column (5). The HPLC system was equipped with a two-channel UV diode detector and a radioactivity detector arranged in series. The elution conditions were: temperature, 25 °C; flow, 1.0 ml/min from 0 to 20 min, and 2.0 ml/min from 20 to 60 min; buffers, buffer A (0.007 M KH₂PO₄, 0.007 M KCl, pH 4.0) and buffer B (0.25 M KH₂PO₄, 0.5 M KCl, pH 5.0); mobile phase, isocratic buffer A (0-5 min), linear gradient of buffer A  100% buffer B (5-20 min), isocratic buffer B (20-37 min), linear gradient of buffer B  100% buffer A (37-52 min), and isocratic buffer A (52-60 min). Under these conditions, dehydroascorbic acid eluted at 4.4 min, and ascorbic acid eluted at 10.4 min.

Reduced glutathione (Sigma) was tested for its ability to reduce dehydroascorbic acid (Aldrich) in vitro at different pH values and temperatures. Reduction of 0.01, 0.05, 0.1, 0.5, and 1 mM dehydroascorbic acid by 0.3-200 mM GSH was continuously monitored at 266 nm for 10 min at 0-37 °C using a diode array spectrophotometer (Beckman DU 7500). Stock solutions of glutathione were prepared daily in 100 mM sodium phosphate buffer, pH 7.5, containing 0.1 mM EDTA. Concentrated solutions of dehydroascorbic acid (100×) (Aldrich) were prepared in sodium phosphate, pH 5.5. Alternatively, dehydroascorbic acid was prepared by oxidation of ascorbic acid with bromine (24).

Ten million cells were washed twice with normal saline and lysed with 250 µl of 0.4% Triton X-100. One volume of 5% sulfosalicylic acid was added, the precipitated protein was removed by centrifugation for 1 min at 10,000 × g, and the supernatant was stored at 70 °C until assay. For assay, 25 µl of supernatant were added to 96-well microtiter plate wells containing 140 µl of 100 mM NaHPO₄, 5 mM NaEDTA (pH 7.5), 25 µl of 6 mM 5,5-dithio-bis-(2-nitrobenzoic acid), and 10 µl of 100 units/ml glutathione reductase (type III, Sigma). Blanks were adjusted to a final concentration of 0.2% Triton and 2.5% sulfosalicylic acid. The reaction was started by adding 50 µl of 1.1 mM NADPH, and the change in absorbance at 405 nm was monitored for 10 min. A standard curve for glutathione was run with each assay. This methodology allows for the determination of reduced as well as oxidized glutathione (37).

RESULTS

We used BSO and diethyl maleate, two compounds that have been used to deplete cells of glutathione (37), to obtain HL-60 cells containing different intracellular concentrations of glutathione. BSO is an inhibitor of the enzyme -glutamylcysteine synthetase that catalyzes the first step of the synthesis of glutathione and therefore inhibits its synthesis (38). Diethyl maleate reacts covalently and irreversibly with glutathione and rapidly depletes cellular glutathione (39), although it may also react with free cysteine or protein thiol groups. The content of glutathione in untreated HL-60 cells varied from 2.4 to 8.3 mM (4.7 ± 2.1 mM), depending on the cell sample (Fig. 1). Incubation of HL-60 cells with 200 µM BSO resulted in a continuous decline of intracellular glutathione during the first 18 h of treatment, with maximal depletion at 24 h. At 24 h, 200 µM BSO reduced the concentration of glutathione by approximately 90%, with no further depletion seen with concentrations of BSO as high as 1 mM. Cells incubated for 24 h in the presence of 200 µM BSO contained concentrations of glutathione in the range of 250 to 700 µM (0.45 ± 0.19 mM) (Fig. 1). BSO showed no cellular toxicity under the experimental conditions studied, and cell viability as assessed by trypan blue exclusion was always greater than 95%.

Diethyl maleate had an acute effect on cellular glutathione. One millimolar diethyl maleate induced maximal depletion in cellular glutathione in 60 min. The final glutathione concentration in cells incubated for 60 min with 1 mM diethyl maleate was in the range of 0.9 to 1.5 mM (1.14 ± 0.23 mM) (Fig. 1). Dietyl maleate did not show cellular toxicity under the experimental conditions used, and cell viability was greater than 95%. More efficient depletion of cellular glutathione was observed in cells incubated for 24 h with 200 µM BSO followed by 60 min with 1 mM diethyl maleate, with treated cells containing levels of glutathione in the range of 1 to 200 µM (0.09 ± 0.07 mM) (Fig. 1). Even when we observed a difference of three orders of magnitude in the intracellular concentration of glutathione between untreated and treated cells, there were no changes in cell viability (data not shown).

We determined the ability of cells containing various concentrations of glutathione to transport dehydroascorbic acid and accumulate ascorbic acid. We verified the identity of the chemical form of radiolabeled vitamin C present in both the incubation medium and cell extracts by HPLC. Dehydroascorbic acid and ascorbic acid were clearly separated by HPLC, with elution times of 4.4 and 10.4 min, respectively (data not shown). Analysis of the incubation medium containing reduced ¹⁴C-labeled ascorbic acid by HPLC showed that more than 98% of the radioactivity eluted with authentic ascorbic acid. A small amount of radioactive material eluted in several small peaks with an elution time different from that of ascorbic or dehydroascorbic acid. The nature of the contaminant material is not known. After treatment with ascorbate oxidase, only one radioactive peak containing more than 95% of the initial radioactivity applied to the column was detected that eluted with the retention time of authentic dehydroascorbic acid. No radioactive peak that eluted in the position of ascorbic acid was observed even when overloading the column with excess sample. In contrast, HPLC chromatograms of extracts prepared from HL-60 cells incubated for 1-30 min with different concentrations of ¹⁴C-labeled dehydroascorbic acid generated by treatment of ¹⁴C-labeled ascorbic acid with ascorbate oxidase showed that greater than 95% of initial radioactive material loaded into the HPLC co-eluted with authentic reduced ascorbic acid. The remaining radioactive material eluted in several small peaks that did not coincide with the elution of ascorbic or dehydroascorbic acid. Thus, the HPLC data clearly indicated that only ascorbic acid accumulated intracellularly in HL-60 cells that transported dehydroascorbic acid.

Transport studies using short uptake assays (22) showed that depleting the HL-60 cells of glutathione did not affect the transport of dehydroascorbic acid (Fig. 2A). HL-60 cells containing as low as 1 µM glutathione transported as much dehydroascorbic acid as cells containing 0.3, 4.1, or 6 mM glutathione. Similar results were obtained when the assays were extended to 30 min (Fig. 2B), in which the rate-limiting step in the uptake reaction is the reduction/trapping of dehydroascorbic acid/ascorbic acid (22). HL-60 cells treated with 200 µM BSO for 24 h followed by 1 mM diethyl maleate for 60 min contained 1 µM glutathione and accumulated an intracellular concentration of ascorbic acid of approximately 15 mM, compared with approximately 17 mM for control HL-60 cells containing 6 mM glutathione (Fig. 2B). We obtained cells covering the full spectrum of glutathione concentrations from 6 mM to 1 µM by subjecting the cells to treatment with BSO, diethyl maleate or combinations of both, and then measured the transport of dehydroascorbic acid (30 s uptake assays) and the accumulation of ascorbic acid (30-min uptake assays). A three-order of magnitude decrease in the concentration of cellular glutathione did not affect the ability of the HL-60 cells to transport dehydroascorbic acid and accumulate ascorbic acid (Fig. 2C). Similarly, no effect on the ability to transport dehydroascorbic acid and to accumulate ascorbic acid was observed in HL-60 cells whose glutathione was augmented to 20 mM by simply incubating the cells in fresh culture medium for 12 h (data not shown). Viability determinations as assessed by trypan blue exclusion showed no decrease in the viability of the treated cells depleted of glutathione compared with control untreated cells; viability was always greater than 95%.

As controls for the integrity of the cells under depletion conditions, we measured the ability of the HL-60 cells to transport methylglucose and accumulate deoxyglucose. These experiments were informative because we previously demonstrated that in HL-60 cells dehydroascorbic acid is transported by GLUT1, the main glucose transporter expressed by HL-60 cells (22). Methylglucose is a substrate of the glucose transporters that is not metabolized and, therefore, can be used to assess the functional state of the transporter (40). Glutathione depletion did not affect the ability of the HL-60 cells to transport methylglucose, confirming that the functional activity of GLUT1 was not affected by the depletion of cellular GSH. Deoxyglucose is transported by glucose transporters and is rapidly phosphorylated to deoxyglucose 6-phosphate which is not metabolized and accumulates intracellularly (40). Different cellular insults that have general negative effects on cellular physiology rapidly affect the ability of cells to accumulate sugars as deoxyglucose. Glutathione-depleted HL-60 cells accumulated deoxyglucose to the same extent as that of the control cells, indicating that the depletion of glutathione did not have a general negative effect on cell physiology.

When exposed to dehydroascorbic acid, HL-60 cells accumulate intracellular concentrations of ascorbic acid that are much higher than the extracellular concentration of dehydroascorbic acid (5, 22). Also, increasing the extracellular concentration of dehydroascorbic acid leads to increased accumulation of intracellular ascorbic acid. The ratio between the intracellular concentration of ascorbic acid relative to the extracellular concentration of dehydroascorbic acid decreases as extracellular dehydroascorbic acid increases in concentration, suggesting that the HL-60 cells have a finite capacity to reduce the transported dehydroascorbic acid (5). Therefore, if glutathione has a fundamental role in the reduction of dehydroascorbic acid, its role should became evident in cells depleted of glutathione and incubated in the presence of high concentrations of dehydroascorbic acid. Accumulation studies using 10-min uptake assays (to measure accumulation of ascorbic acid as opposed to transport of dehydroascorbic acid) revealed that HL-60 cells depleted of glutathione maintained a remarkable ability to accumulate high intracellular concentrations of ascorbic acid when incubated with concentrations of dehydroascorbic acid ranging from 10 µM to 10 mM (Fig. 3A). At 10 mM external dehydroascorbic acid, HL-60 cells containing 40 or 170 µM glutathione accumulated intracellular concentrations of ascorbic acid of approximately 130-160 mM, similar to the concentration of ascorbic acid accumulated by control HL-60 cells containing 3.5 mM glutathione. A two-order of magnitude decrease in the cellular content of glutathione, from 3.5 mM to 40 µM, did not affect the dehydroascorbate-reducing capacity of the HL-60 cells (Fig. 3B). The three cell groups accumulated similar intracellular concentrations of ascorbic acid at all of the concentrations of dehydroascorbic acid tested, and the ratio of the intracellular concentration of accumulated ascorbic acid versus the extracellular concentration of dehydroascorbic acid was not changed by depletion of glutathione in the HL-60 cells (Fig. 3B).

The data derived from the experiments using HL-60 cells depleted of glutathione suggested that the direct chemical reaction between glutathione and dehydroascorbic acid cannot explain the ability of HL-60 to accumulate high intracellular concentrations of ascorbic acid. Additional data consistent with the previous concept were obtained in cell-free experiments in which we measured the dose dependence for the reduction of dehydroascorbic acid by varied concentrations of glutathione. Samples containing dehydroascorbic acid were incubated at 37 °C for 10 min in the presence of concentrations of glutathione ranging from 100 µM to 10 mM, and the generation of ascorbic acid was monitored by following the change in absorbance at 266 nm. An incubation time of 10 min was selected based on the data from the accumulation of ascorbic acid in HL-60 cells depleted of glutathione. Because the absorbance of a solution of ascorbic acid 0.1 mM is approximately 1.6 at 266 nm, we had a lower limit of detection of about 2 µM dehydroascorbic acid in these experiments. On the other end of the concentration curve, we were limited to using concentrations of dehydroascorbic acid no higher than 250 µM. There was a marked dose dependence for the reduction of dehydroascorbic acid by glutathione under cell-free conditions (Fig. 4A). A vast molar excess of glutathione was required to completely reduce a determined concentration of dehydroascorbic acid. Thus, although 10 mM glutathione completely reduced 150 µM dehydroascorbic acid (60-fold molar excess of glutathione over dehydroascorbic acid), only 40-60% of 150 µM dehydroascorbic acid was reduced in the presence of 1-3 mM glutathione. No measurable amount of reduced ascorbic acid was generated when the glutathione concentration in the assay was 100 µM or less (Fig. 4A).

The previous data refer to the amount of ascorbic acid generated in an incubation period of 10 min. We subsequently analyzed the dose dependence of the inital rate of the reduction of dehydroascorbic to ascorbic acid in the presence of various concentrations of glutathione. The rate of generation of ascorbic acid was negligible at concentrations of glutathione lower than 100 µM (Fig. 4B). A clear dose-response curve was observed, however, at concentrations of glutathione greater than 300 µM, and the rate of ascorbic acid generation was maximal at 10 mM glutathione at all the initial concentrations of dehydroascorbic acid tested. In the cell-free system, the reduction of 50 µM dehydroascorbic acid occurred at a rate of 40 nmol/min/ml in the presence of 10 mM glutathione and decreased to approximately 30, 10, and 3 nmol/min/ml at 6, 3, and 1 mM glutathione, respectively. On the other hand, HL-60 cells containing from 1 µM to 6 mM glutathione and incubated with 50 µM dehydroascorbic acid for 10 min accumulated ascorbic acid at a constant rate of approximately 900 nmol/min/ml of cell water (Fig. 4C). Under these conditions, the limiting step in uptake is the reduction of the recently transported dehydroascorbic acid (22), and therefore, the rate of trapping/accumulation of ascorbic acid is dependent on the rate of reduction of dehydroascorbic by the HL-60 cells. Considering a content of 2.4-8.3 mM glutathione in control cells, our data indicate that the HL-60 cells reduce dehydroascorbic acid at a rate that exceeds by approximately 30- 200-fold the rate of the direct chemical reduction of dehydroascorbic acid by glutathione. Overall, the data indicate that only a minor portion of the ascorbic acid that is accumulated in the HL-60 cells is generated by the direct chemical reduction of dehydroascorbic acid and the cell's glutathione.

The temperature dependence of the rate of accumulation of ascorbic acid by the HL-60 cells was also clearly different from that of the rate of the direct chemical reduction of dehydroascorbic acid by glutathione. The rate of the chemical reduction increased linearly with temperature, and a plot of the rate of reduction versus temperature resulted in a straight line with a slope of 1.04 nmol/min/ml/ °C (Fig. 5A). There was a 14.4-fold increase in the rate of reduction, from 2.5 to 36 nmol/min/ml, when the temperature was increased from 4 to 37 °C. At 21 °C, the initial rate of reduction was approximately 50% of that observed at 37 °C. On the other hand, the rate of trapping/accumulation of ascorbic acid in the HL-60 cells showed a complex temperature dependence that resulted in a plot of rate of accumulation versus temperature with a marked upward curvature (Fig. 5B). As a result, there was a 100-fold increase in the rate of accumulation, from 10 to 1000 nmol/min/ml, when the temperature was increased from 4 to 37 °C, and the rate of accumulation at 21 °C was only 17% of that observed at 37 °C.

The temperature dependence for the trapping of deoxyglucose was very similar to that for the trapping/accumulation of ascorbic acid (Fig. 5C). The plot of rate of accumulation versus temperature showed a marked upward curvature, and as a result, there was an 87-fold increase in the rate of trapping, from 9 to 780 nmol/min/ml, when the temperature was increased from 4 to 37 °C. The rate of accumulation at 21 °C was only 24% of that observed at 37 °C. On the other hand, the temperature dependence of the transport of methylglucose by the HL-60 cells was clearly different from that of the trapping/accumulation of ascorbic acid (Fig. 5D). A plot of the rate of transport versus temperature was linear, with a slope of 406 nmol/min/ml/ °C. There was a 13-fold increase in the rate of transport, from 0.2 to 2.4 µmol/min/ml, when the temperature was increased from 4 to 37 °C, and the rate of transport at 21 °C was approximately 45% of that observed at 37 °C. The temperature dependence for the transport of dehydroascorbic acid by the HL-60 cells was also clearly different from that of the trapping/accumulation of ascorbic acid (Fig. 5E). A plot of the rate of transport versus temperature was linear, with a slope of 59 nmol/min/ml/ °C. There was a 39-fold increase in the rate of transport, from 50 to 1940 nmol/min/ml, when the temperature was increased from 4 to 37 °C, and the rate of transport at 21 °C was approximately 50% of that observed at 37 °C. Overall, the data are consistent with the concept that the direct chemical reaction between dehydroascorbic acid and cellular glutathione is not the primary mechanism by which the HL-60 cells reduce the recently transported dehydroascorbic acid and trap it intracellularly as ascorbic acid.

DISCUSSION

We directly addressed the role of glutathione in the ability of HL-60 cells to transport dehydroascorbic acid and to reduce dehydroascorbic acid and accumulate ascorbic acid. HL-60 cells depleted of glutathione by treatment with BSO and diethyl maleate under conditions that resulted in a two- to three-order of magnitude decrease in cellular glutathione, showed no alteration in their ability to accumulate millimolar concentrations of ascorbic acid when incubated with dehydroascorbic acid. Furthermore, cellular glutathione depletion did not affect the rate of transport of dehydroascorbic acid or methylglucose, or the trapping of deoxyglucose, by the HL-60 cells. These findings have direct relevance to our understanding of the control mechanisms involved in the regulation of the cellular content of ascorbic acid in human cells and the identity and functional properties of the cellular systems that participate in the reduction of dehydroascorbic acid. The data are consistent with the existence of several overlapping mechanisms, enzymatic as well as nonenzymatic, that regulate the ability of the HL-60 cells to reduce dehydroascorbic acid and accumulate ascorbic acid. Our findings suggest a secondary role for glutathione, and by extension for glutathione-dependent dehydroascorbic acid reductases, in the regulation of the cellular content of vitamin C in HL-60 cells.

The reported normal content of glutathione in human cells and tissues is in the range of 1 to 10 mM (29), the concentration range at which glutathione is able to generate ascorbic acid from dehydroascorbic acid by a direct chemical reaction. This led to the proposal that the direct chemical reaction between glutathione and dehydroascorbic acid may be responsible for the accumulation of ascorbic acid in cells (34). HL-60 cells have intracellular concentrations of glutathione in the range of 3.4 to 8.2 mM, and therefore, the ascorbic acid accumulated by the HL-60 cells incubated with dehydroascorbic acid could be the product of the direct chemical reaction between glutathione and the recently transported dehydroascorbic acid. Our data indicate, however, that the direct chemical reduction of dehydroascorbic acid by glutathione occurred only in the presence of a great molar excess of glutathione over dehydroascorbic acid, with no reduction observed at concentrations of glutathione of 300 µM or less. On the other hand, HL-60 cells, containing concentrations of glutathione that ranged from as low as 1 µM to a normal content of 3 mM and higher, were able to accumulate very high intracellular concentrations of ascorbic acid, indicating that the direct chemical reaction between glutathione and dehydroascorbic acid was not necessary for the accumulation of ascorbic acid by HL-60 cells. Consistent with this interpretation were the results showing that HL-60 cells containing as little as 1 µM intracellular glutathione accumulated ascorbic acid at the same rate as cells containing 6 mM glutathione. The rate of direct chemical reduction of dehydroascorbic acid by glutathione was dependent on the concentration of glutathione, and even at a vast excess of glutathione it was only 2-5% of the rate of trapping/accumulation of ascorbic acid by the HL-60 cells. Thus, in cells containing a normal concentration of glutathione, less than 5% of the cellular ascorbic acid present in HL-60 cells is expected to be generated from the direct chemical reaction between glutathione and dehydroascorbic acid. We conclude that the cellular reduction of dehydroascorbic acid by HL-60 cells is mostly enzymatic in nature and not due to a direct chemical reaction between glutathione and dehydroascorbic acid.

Further evidence for the existence of enzymatic processes involved in the cellular reduction of dehydroascorbic acid by HL-60 cells, as opposed to the direct chemical reaction between glutathione and dehydroascorbic acid, was provided for the difference in temperature dependence of these processes. The trapping/accumulation of ascorbic acid by HL-60 cells was highly temperature dependent and increased by 100-fold with an increase in temperature from 4 to 37 °C, compared with less than a 15-fold increase for the direct chemical reaction. On the other hand, the temperature dependence of the trapping/accumulation of ascorbic acid by HL-60 cells was similar to the temperature dependence of the trapping of deoxyglucose by the cells, a process that is known to be enzymatic (40). These experiments also confirmed our previous findings that performing uptake experiments for different lengths of time (30 s versus 10 min or longer) gives information on different aspects of the overall uptake process (22), as shown here by the difference in temperature dependence of transport of dehydroascorbic acid and methylglucose compared with the trapping of ascorbic acid and deoxyglucose.

Our data also address the issue of the participation of the enzymes glutaredoxin and protein disulfide isomerase, which have dehydroascorbic acid reductase activity (26, 35), in the cellular reduction of dehydroascorbic acid and accumulation of ascorbic acid in HL-60 cells. Glutaredoxin and protein disulfide isomerase have a K_m for glutathione of 3.4 and 2.9 mM, respectively, and they are absolutely dependent on glutathione for their activity. Thus, both enzymes theoretically could play an important role in the cellular accumulation of ascorbic acid in HL-60 cells containing normal concentrations of glutathione that range from 2.4 to 8.3 mM. If that were the case, a marked decrease of cellular glutathione would have a major effect in the ability of the HL-60 cells to accumulate ascorbic acid when incubated with dehydroascorbic acid. The data in the HL-60 cells, however, indicated that the trapping/accumulation of ascorbic acid by cells incubated with dehydroascorbic acid was not affected by a three-order of magnitude decrease in the cellular content of glutathione, indicating that these enzymes do not have a central role in mediating the accumulation of ascorbic acid by HL-60 cells.

In conclusion, our data indicate that HL-60 cells have highly efficient systems for the reduction of dehydroascorbic acid that do not require glutathione.