L-Ascorbic acid, L-sodium ascorbate, glutathione, ascorbate oxidase, 2-mercaptoethanol and dehydroisoandrosterone were purchased from Sigma Chemical (St. Louis, MO). Dehydroascorbic acid and metaphosphoric acid were from Aldrich Chemical (Milwaukee, WI), and APM was from Wako Pure Chemical (Osaka, Japan). All the chemicals were of analytical grade.

Female BALB/c mice (20~25 g) were used. The mice were given Jeil Laboratory Chow (Seoul, Korea) and water ad libitum. The animals were kept at 21~24℃ and 40~60% relative humidity with a 12-hr light/dark cycle.

A mouse plasmacytoma cell line, Adj-PC5 (5), was maintained and transplanted by the intravenous injection of 1 to 5×10⁶ cells to the BALB/c mice. The Adj.PC-5 cells mainly infiltrated into the spleen following intravenous injection. The tumor cell suspensions for the experiments in this study were prepared from the removed tumor-bearing spleens by grinding the with a syringe plunger on a fine wire mesh screen. The cells were cultured in complete RPMI 1640 medium (Gibco, Grand Island, NY) containing 5% fetal bovine serum (FBS, Gibco), 2 mM L-glutamine (Sigma), antibiotics (100 units penicillin and 100 mg streptomycin per ml) (Gibco) at 37℃ in a humidified 5% CO₂ incubator.

Colony forming assay was performed according to a previously reported method (16), with minor modifications. Briefly, the cells (3×10⁵ nucleated cells/dish) from the plasmacytoma-infiltrated spleens and normal spleens of the BALB/c mice were diluted and cultured on soft-agar (0.3%) plates. The agar culture was contained in a 35-mm tissue culture dish (Falcon, Lincoln Park, NJ) with 5 small holes made at the bottom and it was stored within a 60-mm culture dish (Falcon) for sterility. The cultures were incubated at 37℃ in a humid atmosphere that was continuously flushed with 5% CO₂ and the cultures were fed daily with 0.5 ml of fresh RPMI 1640 medium containing 5% FBS, 2 mM L-glutamine, 100 units/ml penicillin and 100 g/ml streptomycin, plus the compounds to be assessed. The fed medium percolated through the agar layer and drained out through the holes made at the bottom of the culture dish. Colonies containing 50 or more cells were counted with an inverted microscope when the cultures were 2 weeks old.

The cells were adjusted to 3×10⁵ nucleated cells per ml in RPMI 1640 medium supplemented with 5% FBS, 2 mM L-glutamine and antibiotics. Chemicals were directly added to the cell suspensions, and the suspensions were then transferred (200 µl aliquots) and cultured on 96-well microtiter plates (Costar, Cambridge, MA) at 37℃ and 5% CO₂ for 48 h. The cells were pulsed with 0.5µCi/well of [³H]-thymidine (NEN, Boston, MA) for the last 6 h of the culture period. The thymidine incorporated into DNA was measured by scintillation counting.

Ascorbate was measured by a previously described method (3), with minor modifications. The high performance liquid chromatography (HPLC) system (Shimadzu Corp., Tokyo, Japan) consisted of a CBM-10A communication module, a L-ECD-6A electrochemical detector, a LC-10AD pump system and a CTO-10A column oven. The stationary phase was a Shim-pack CLC-ODS column (packed with octadesyl coated silica particles of 5µm diameter and a 10 nm pore diameter, the column was 6 mm×15 cm) (Shimadzu) connected to a Shim-pack G-ODS guard column (4 mm×1 cm) (Shimadzu). The mobile phase consisted of 50 mM potassium phosphate and 100µM ethylenediaminetetraacetic acid and it was adjusted to pH 2.5 with phosphoric acid. The mobile phase was filtered through 0.45µm JH hydrophilic filters (Millipore, Yonezawa, Japan) prior to use. Column equilibration was performed at a flow rate of 1.0 ml/min and this flow rate was used for all the subsequent analyses. The detector was set at 0.75 V versus an Ag/AgCl reference. After the indicated incubation periods, the cultures were centrifuged to separately collect the cells and supernatants. The cells were washed with phosphate-buffered saline and extracted in 5% metaphosphoric acid solution for 10 min. In the case of the culture supernatants, an equal volume of metaphosphoric acid (10%) was added to extract the ascorbic acid. The extract was filtered (0.22µm) and then applied to the HPLC column.

G6PD activity was measured by a previously described method (17). The cells were washed twice in phosphate-buffered saline and then homogenized in ice-cold 50 mM Tris·HCl buffer (pH 7.4) that contained 10 mM MgCl₂, 0.5 mM phenylmethylsulfonyl fluoride, 1µg leupeptin per ml, and 1µg apro tinin per ml. The supernatants were collected following centrifugation (12,000 rpm) for 5 min at 4℃ and then they were used for the enzyme assay. A cuvette was loaded with 0.95 ml of the reaction mixture (50 mM Tris·HCl buffer; pH 7.4, 13.3 mM MgCl₂, 1 mM NADP⁺, 1 mM glucose 6-phosphate and 1 mM maleimide) and stored at 30℃ for 5 min. A cell lysate volume of of 50µl was added to the cuvette and mixed well. The change in absorbance at 340 nm for the first 5 min following a 5-min lag period was monitored and normalized against the protein concentration to calculate the specific enzyme activity.

Total RNA was isolated from the Adj.PC-5 cells using RN-Agents® (Promega, Madison, MI). The RNA samples (30µg) were denatured with formaldehyde, electrophoresed on a 1.2% agarose gel that contained 4% formaldehyde and then they were transferred to a nylon membrane (Amersham, Arlington Heights, IL). The transferred RNA was crosslinked to the nylon membrane by ultraviolet irradiation. Hybridization was performed overnight at 42℃ in 50% formamide, 5X SSC, 100µg yeast tRNA per ml, 7.5% dextran sulfate, 10X Denhardt's solution and 2% sodium dodecyl sulphate (SDS) with a ³²P-labeled probe that was the 0.5 kb PCR fragment of mouse G6PD cDNA. The nylon membrane was washed three times for 30 min at 40℃ each time with 2X SSPE (0.1% SDS), 1X SSPE (0.1% SDS) and 0.1X SSPE, and then it was exposed for 24 h at -70℃ for autoradiography with using an intensifying screen.

The mean±SD was determined for each treatment. The treatment groups were compared with the control groups by using Dunnett's two-tailed t-test (18), with statistical significance defined as p<0.05. All the experiments shown in this report were repeated at least three times to confirm their reproducibility.

Consistent with our previous reports (5,16), ascorbate added daily enhanced colony formation in the Adj.PC-5 cell cultures in a concentration-dependent manner at up to 100µM, but then there was a slight decrease at 200µM (Fig. 1A). However, ascorbate did not increase colony formation in the presence of ascorbate oxidase (0.5 units/ml). The oxidation and degradation of ascorbate that were catalyzed by ascorbate oxidase abrogated its stimulatory effect on colony growth, and this result suggests that the reduced form is essential for ascorbate to express its growth-stimulatory effect. Compared with colony growth, the incorporation of [³H]-thymidine into the cells growing in suspension culture represents the growth rate of the total cells. The spleen cells from the mice that were intravenously transplanted with Adj.PC-5 cells were adjusted to 5×10⁵ nucleated cells per ml and these were cultured on a 96-well plate in the presence of ascorbate derivative. Fig. 1B shows the increased growth of Adj.PC-5 cells by the addition of ascorbate. Ascorbic acid, sodium ascorbate and APM all stimulated growth at up to 100µM. APM was more efficient at enhancing cell growth than either ascorbic acid or sodium ascorbate. The addition of only 25µM APM was enough to induce a significant increase in cell proliferation. In addition, enhancement of proliferation was observed even with a quite high concentration of APM, e.g., 400µM, which is a concentration that was cytotoxic in the case of either ascorbic acid or sodium ascorbate (data not shown). Meanwhile, the cells from normal spleens showed only background levels of proliferation (<1,000 dpm) regardless of the cell type and the amounts of added ascorbate derivative (data not shown).

Although the reduced state is important for ascorbic acid to enhance cell growth, as well as to act as an antioxidant, there may also be other factors that determine the efficiency of growth stimulation, such as the cellular transport efficiency and stability in culture medium. To further characterize the factors that determine the stimulatory efficiency for cell growth, we daily fed four different derivatives of ascorbate to Adj.PC-5 cell cultures (Fig. 2). The sodium salt form of ascorbate enhanced colony formation at the same level as was observed in the ascorbic acid-treated cultures. Thus, the ascorbic acid-induced pH change does not appear to be a major factor for the modulation of cell growth by ascorbic acid. Dehydroascorbate, an oxidized form of ascorbate, was not effective at stimulating cell growth, as was anticipated from the ascorbate oxidase experiments. The highest colony forming response was observed in the cultures treated with APM, showing an approximately 10-fold increase at 100µM over the untreated control. APM is a stable derivative of ascorbic acid and slowly but steadily converted to ascorbic acid by phosphatases in both the medium and the cell cytoplasm. However, it is not yet certain which phosphatase is responsible for the dephosphorylation. This result implies that not only the redox state, but also the consistency of the ascorbic acid concentration may be important in order for it to stimulate cell growth.

APM was better than ascorbate in terms of both low cytotoxicity and growth stimulation. To elucidate that the superiority of APM on cell growth was due to its slow transformation to ascorbate as well as the resultant steady concentration of extracellular ascorbate, we examined the intracellular and extracellular ascorbate concentration kinetics following the treatment with either ascorbate or APM (200µM). As shown in Fig. 3A, the ascorbate level in the medium decreased in a time-dependent manner, leaving 38% and 4% of the initial concentration at 24 and 48 h, respectively. Meanwhile, the intracellular ascorbate increased, reaching a maximum at 12 h, and then gradually decreased. In contrast to the culture treated with ascorbate, both the extra- and intracellular ascorbate concentration in the APM-treated culture started with 0µM (Fig. 3). The ascorbate concentration in the medium increased over time until 12 h following APM treatment and then it showed a plateau for 50µM between 12~24 h. The intracellular level of ascorbate also increased to approximately 1.4 nmoles/10⁶ cells by 12 h after treatment and then it decreased thereafter. Although the level was quite low, ascorbate was detectable in both the cells and the medium of the cultures at 72 h following the start of APM treatment. However, no remaining ascorbate was detected at 72 h in either the cells or supernatants of the ascorbate-treated cultures.

We measured the effects of ascorbate on the enzyme activities involved in the regulation of cellular reduction-oxidation conditions, including superoxide dismutase, catalase, glutathione reductase and G6PD. G6PD was the only enzyme that showed a significantly increased level of activity after a 72-h incubation with ascorbate (data not shown). Since G6PD produces NADPH and ribose-5-phosphate and it is known to be a rate-limiting enzyme for cell growth, we further investigated the concentration-dependent and time-dependent effects of ascorbate on G6PD activity. The G6PD activity was increased in a concentration-dependent fashion in the Adj.PC-5 cell cultures treated with APM for 72 h (Fig. 4A). Approximately 50% enhancement was observed in the culture treated with 400µM of ascorbate. As shown in Fig. 4B, the G6PD activity was slowly increased in a time-dependent manner. This 50% increase of G6PD activity was quite consistent with the growth stimulation level by APM, suggesting the possibility that the enhanced G6PD activity is an underlying mechanism for the ascorbate-induced stimulation of cell growth. To determine whether this increase of G6PD activity was accompanied with an increase of G6PD mRNA, we performed Northern blot analysis (Fig. 4C). The level of G6PD mRNA was enhanced by the APM (200µM) treatment in a time-dependent manner and this was consistent with the results of the change of enzyme activity.

In order to determine if hydrogen peroxide produced from ascorbate has a specific role in the enhancement of growth and the G6PD activity induced by APM, catalase (10 units/ml) was exogenously added to the culture medium. While the addition of catalase to the Adj.PC-5 cell cultures increased colony growth regardless of the existence of APM, the G6PD activity in the Adj.PC-5 cells was not altered by APM (Fig. 5). These results suggest that growth enhancement by catalase was mostly an indirect effect mediated through scavenging cytotoxic hydrogen peroxide in the culture medium and that the increased G6PD by APM was not due to the hydrogen peroxide produced during ascorbate degradation.

Dehydroepiandrosterone is a well-known G6PD inhibitor (14). To determine whether the cell growth enhancement by ascorbate was associated with the increase of G6PD activity in cells, dehydroepiandrosterone was added to the Adj.PC-5 cell cultures. As anticipated, dehydroepiandrosterone was able to inhibit G6PD activity in the cytosolic fraction prepared from the Adj.PC-5 cells (Fig. 6A). Meanwhile, daily administered dehydroepiandrosterone efficiently abolished the APM-induced increase of cell growth, as measured by colony formation in the Adj.PC-5 cell culture (Fig. 6B). Thus, this data suggests that the enhancement of growth and G6PD activity in Adj.PC-5 cells that was observed following APM treatment may be closely associated.