Glycolysis and glucose uptake in intact outer segments isolated from bovine retinal rods

Glucose transport across the plasma membrane of isolated bovine rod outer segments (ROS) was measured by uptake of 14C-labeled 3-O-methylglucose and 2-deoxyglucose and was inferred from deenergization of ROS with 2-deoxyglucose. Glucose transport was mediated by a facilitated diffusion glucose transporter that equilibrated external and internal free hexose concentrations. Glucose transport in ROS displayed two components as judged from kinetic analysis of hexose equilibration and as judged from inhibition by cytochalasin B and phloretin. Transport under exchange conditions was considerably faster as compared with net hexose uptake, similar to that observed for the erythrocyte glucose transporter. Sensitivity to cytochalasin B and affinity to 3-O-methylglucose were similar to those observed for the hepatocyte glucose transporter. The cytochalasin-insensitive component appears unique to ROS and did not reflect leakage transport as judged from a comparison with L-glucose uptake. Glucose transport feeds glycolysis localized to ROS. We suggest that a major role for glycolysis in ROS is phosphorylation of GDP to GTP via pyruvate kinase and PEP, while phosphorylation of ADP to ATP can use the creatine kinase/phosphocreatine pathway as well.     

Dehydroascorbic acid uptake and intracellular ascorbic acid accumulation in cultured Müller glial cells (TR-MUL)

Vitamin C is mainly transported across the inner blood–retinal barrier (inner BRB) as dehydroascorbic acid (DHA) via a facilitative glucose transporter (GLUT) 1, and accumulates as ascorbic acid (AA) in the retina. Müller cells, huge glial cells, exhibit passive structural and metabolic functions for retinal neurons and the inner BRB. We characterized DHA transport and its corresponding transporter in a rat Müller cell line (TR-MUL5 cells). [¹⁴C]DHA uptake by TR-MUL5 cells took place in a time-dependent and Na⁺-independent manner. [¹⁴C]DHA uptake was inhibited by substrates and inhibitors of GLUTs, suggesting that Müller cells take up DHA via GLUTs. HPLC analysis revealed that most of the DHA taken up by TR-MUL5 cells was converted to AA and accumulated as AA in TR-MUL5 cells. [¹⁴C]DHA uptake by TR-MUL5 cells took place in a concentration-dependent manner with a Michaelis–Menten constant of 198 μM and was inhibited by cytochalasin B in a concentration-dependent manner with a 50% inhibition concentration of 0.283 μM. Although GLUT1, 3, and 4 mRNA are expressed in TR-MUL5 cells, quantitative real-time PCR revealed that GLUT1 mRNA expression was 5.85- and 116-fold greater than that of GLUT3 and 4, respectively. Western blot analysis supports the expression of GLUT1 protein with 45 kDa in TR-MUL5 cells. In conclusion, DHA is taken up by facilitative glucose transporters, most likely GLUT1, and converted to AA in TR-MUL5 cells.     

Direct inhibition of the hexose transporter GLUT1 by tyrosine kinase inhibitors

The facilitative hexose transporter GLUT1 is a multifunctional protein that transports hexoses and dehydroascorbic acid, the oxidized form of vitamin C, and interacts with several molecules structurally unrelated to the transported substrates. Here we analyzed in detail the interaction of GLUT1 with a group of tyrosine kinase inhibitors that include natural products of the family of flavones and isoflavones and synthetic compounds such as the tyrphostins. These compounds inhibited, in a dose-dependent manner, the transport of hexoses and dehydroascorbic acid in human myeloid HL-60 cells, in transfected Chinese hamster ovary cells overexpressing GLUT1, and in normal human erythrocytes, and blocked the glucose-displaceable binding of cytochalasin B to GLUT1 in erythrocyte ghosts. Kinetic analysis of transport data indicated that only tyrosine kinase inhibitors with specificity for ATP binding sites inhibited the transport activity of GLUT1 in a competitive manner. In contrast, those inhibitors that are competitive with tyrosine but not with ATP failed to inhibit hexose uptake or did so in a noncompetitive manner. These results, together with recent evidence demonstrating that GLUT1 is a nucleotide binding protein, support the concept that the inhibitory effect on transport is related to the direct interaction of the inhibitors with GLUT1. We conclude that predicted nucleotide-binding motifs present in GLUT1 are important for the interaction of the tyrosine kinase inhibitors with the transporter and may participate directly in the binding transport of substrates by GLUT1.     

Continuous stress-induced dopamine dysregulation augments PAP-I and PAP-II expression in melanotrophs of the pituitary gland

We studied the acquisition of dehydroascorbic acid by rat hepatocytes, H4IIE rat hepatoma cells and Xenopus laevis oocytes. Transport kinetics and competition and inhibition studies revealed that rat hepatocytes transport oxidized dehydroascorbic acid through a single functional component possessing the functional and kinetic properties expected for the glucose transporter GLUT2. On the other hand, rat hepatoma cells showed expression of at least two dehydroascorbic acid transporters with the expected functional and kinetic properties expected for GLUT1 and GLUT2. Expression studies of GLUT2 in X. laevis oocytes followed by transport kinetics and competition and inhibition studies revealed that GLUT2 is a low affinity dehydroascorbic transporter whose kinetic and functional properties match those observed for the endogenous GLUT2 transporter in rat hepatocytes and rat hepatoma cells. Therefore, GLUT2, a transporter known as a low affinity transporter of glucose and fructose and a high affinity transporter of glucosamine is also a low affinity dehydroascorbic acid transporter.     

Ascorbic acid-dependent GLUT3 inhibition is a critical step for switching neuronal metabolism

Intracellular ascorbic acid is able to modulate neuronal glucose utilization between resting and activity periods. We have previously demonstrated that intracellular ascorbic acid inhibits deoxyglucose transport in primary cultures of cortical and hippocampal neurons and in HEK293 cells. The same effect was not seen in astrocytes. Since this observation was valid only for cells expressing glucose transporter 3 (GLUT3), we evaluated the importance of this transporter on the inhibitory effect of ascorbic acid on glucose transport. Intracellular ascorbic acid was able to inhibit (3)H-deoxyglucose transport only in astrocytes expressing GLUT3-EGFP. In C6 glioma cells and primary cultures of cortical neurons, which natively express GLUT3, the same inhibitory effect on (3)H-deoxyglucose transport and fluorescent hexose 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose (2-NBDG) was observed. Finally, knocking down the native expression of GLUT3 in primary cultured neurons and C6 cells using shRNA was sufficient to abolish the ascorbic acid-dependent inhibitory effect on uptake of glucose analogs. Uptake assays using real-time confocal microscopy demonstrated that ascorbic acid effect abrogation on 2-NBDG uptake in cultured neurons. Therefore, ascorbic acid would seem to function as a metabolic switch inhibiting glucose transport in neurons under glutamatergic synaptic activity through direct or indirect inhibition of GLUT3.     

Specificity of ascorbate analogs for ascorbate transport. Synthesis and detection of [(125)I]6-deoxy-6-iodo-L-ascorbic acid and characterization of its ascorbate-specific transport properties.

Cellular ascorbic acid accumulation occurs in vitro by two distinct mechanisms: transport of ascorbate itself or transport and subsequent intracellular reduction of its oxidized product, dehydroascorbic acid. It is unclear which mechanism predominates in vivo. An easily detectable compound resembling ascorbate but not dehydroascorbic acid could be a powerful tool to distinguish the two transport activities. To identify compounds, 21 ascorbate analogs were tested for inhibition of ascorbate or dehydroascorbic acid transport in human fibroblasts. The most effective analogs, competitive inhibitors of ascorbate transport with K(i) values of 3 microM, were 6-deoxy-6-bromo-, 6-deoxy-6-chloro-, and 6-deoxy-6-iodo-L-ascorbate. No analog inhibited dehydroascorbic acid transport. Using substitution chemistry, [(125)I]6-deoxy-6-iodo-L-ascorbate (1.4 x 10(4) mCi/mmol) was synthesized. HPLC detection methods were developed for radiolabeled and nonradiolabeled compounds, and transport kinetics of both compounds were characterized. Transport was sodium-dependent, inhibited by excess ascorbate, and similar to that of ascorbate. Transport of oxidized ascorbate and oxidized 6-deoxy-6-iodo-L-ascorbate was investigated using Xenopus laevis oocytes expressing glucose transporter isoform GLUT1 or GLUT3. Oxidation of ascorbate or its analog in media increased uptake of ascorbate in oocytes by 6-13-fold compared with control but not that of 6-deoxy-6-iodo-L-ascorbate. Therefore, 6-deoxy-6-iodo-L-ascorbate, although an effective inhibitor of ascorbate transport, either in its reduced or oxidized form was not a substrate for dehydroascorbic acid transport. Thus, radiolabeled and nonradiolabeled 6-deoxy-6-iodo-L-ascorbate provide a new means for discriminating dehydroascorbic acid and ascorbate transport in ascorbate recycling.     

Direct interaction of phenylarsine oxide with hexose transporters in isolated rat adipocytes

It has previously been shown that phenylarsine oxide (PhAsO), an inhibitor of protein internalization, also inhibits stereospecific uptake of D-glucose and 2-deoxyglucose in both basal and insulin-stimulated rat adipocytes. This inhibition of hexose uptake was found to be dose-dependent. PhAsO rapidly inhibited sugar transport into insulin-stimulated adipocytes, but at low concentrations inhibition was transient. Low doses of PhAsO (1 microM) transiently inhibit stereospecific hexose uptake and near total (approx. 90%) recovery of transport activity occurs within 20 min. Interestingly, once recovered, the adipocytes can again undergo rapid inhibition and recovery of transport activity upon further treatment with PhAsO (1 microM). In addition, PhAsO is shown to inhibit cytochalasin B binding to plasma membranes from insulin-stimulated adipocytes in a concentration-dependent manner which parallels the dose-response inhibition of hexose transport by PhAsO. The data presented suggest a direct interaction between the D-glucose transporter and PhAsO, resulting in inhibition of transport. The results are consistent with the current recruitment hypothesis of insulin activation of sugar transport and indicate that a considerable reserve of intracellular glucose carriers exists within fat cells.     

Trinucleotide Insertions, Deletions, and Point Mutations in Glucose Transporters Confer K+ Uptake in Saccharomyces cerevisiae

Deletion of TRK1 and TRK2 abolishes high-affinity K⁺ uptake in Saccharomyces cerevisiae, resulting in the inability to grow on typical synthetic growth medium unless it is supplemented with very high concentrations of potassium. Selection for spontaneous suppressors that restored growth of trk1Δ trk2Δ cells on K⁺-limiting medium led to the isolation of cells with unusual gain-of-function mutations in the glucose transporter genes HXT1 and HXT3 and the glucose/galactose transporter gene GAL2. ⁸⁶Rb uptake assays demonstrated that the suppressor mutations conferred increased uptake of the ion. In addition to K⁺, the mutant hexose transporters also conferred permeation of other cations, including Na⁺. Because the selection strategy required such gain of function, mutations that disrupted transporter maturation or localization to the plasma membrane were avoided. Thus, the importance of specific sites in glucose transport could be independently assessed by testing for the ability of the mutant transporter to restore glucose-dependent growth to cells containing null alleles of all of the known functional glucose transporter genes. Twelve sites, most of which are conserved among eukaryotic hexose transporters, were revealed to be essential for glucose transport. Four of these have previously been shown to be essential for glucose transport by animal or plant transporters. Eight represented sites not previously known to be crucial for glucose uptake. Each suppressor mutant harbored a single mutation that altered an amino acid(s) within or immediately adjacent to a putative transmembrane domain of the transporter. Seven of 38 independent suppressor mutations consisted of in-frame insertions or deletions. The nature of the insertions and deletions revealed a striking DNA template dependency: each insertion generated a trinucleotide repeat, and each deletion involved the removal of a repeated nucleotide sequence.     

On the pull: periplasmic trapping of sugars before transport

Bacteria have evolved many routes for taking up nutrients, demonstrating great versatility in the types and mechanism of uptake used in different physiological conditions. The discovery of a single transporter in the bacterium Advenella mimigardefordensis for the uptake of five different sugars, including L‐glucose and D‐xylose, is described in this issue (Meinert et al., 2017 ), providing yet another example of the surprising adaptability of bacterial transport strategies. The transporter identified is a tripartite ATP‐independent (TRAP) transporter, not previously associated with sugar transport, and in fact does not transport the sugars directly at all, rather requiring them to be converted in the periplasm to their respective sugar acid forms before transport through what appears to be a novel general sugar acid transporter. In this commentary, I describe how this process is consistent with the known mechanisms of TRAP transporters and consider how the role of sugar oxidation, or oxidative fermentation, operates with multiple hexose and pentose sugars. Finally I suggest that the periplasmic conversion of nutrients acquired across the outer membrane, before transport across the inner membrane, could have potentially useful biological functions in Gram negative bacteria.     

Recycling of vitamin C by a bystander effect

Human cells transport dehydroascorbic acid through facilitative glucose transporters, in apparent contradiction with evidence indicating that vitamin C is present in human blood only as ascorbic acid. On the other hand, activated host defense cells undergoing the oxidative burst show increased vitamin C accumulation. We analyzed the role of the oxidative burst and the glucose transporters on vitamin C recycling in an in vitro system consisting of activated host-defense cells co-cultured with human cell lines and primary cells. We asked whether human cells can acquire vitamin C by a "bystander effect" by taking up dehydroascorbic acid generated from extracellular ascorbic acid by neighboring cells undergoing the oxidative burst. As activated cells, we used HL-60 neutrophils and normal human neutrophils activated with phorbol 12 myristate 13-acetate. As bystander cells, we used immortalized cell lines and primary cultures of human epithelial and endothelial cells. Activated cells produced superoxide anions that oxidized extracellular ascorbic acid to dehydroascorbic acid. At the same time, there was a marked increase in vitamin C uptake by the bystander cells that was blocked by superoxide dismutase but not by catalase and was inhibited by the glucose transporter inhibitor cytochalasin B. Only ascorbic acid was accumulated intracellularly by the bystander cells. Glucose partially blocked vitamin C uptake by the bystander cells, although it increased superoxide production by the activated cells. We conclude that the local production of superoxide anions by activated cells causes the oxidation of extracellular ascorbic acid to dehydroascorbic acid, which is then transported by neighboring cells through the glucose transporters and immediately reduced to ascorbic acid intracellularly. In addition to causing increased intracellular concentrations of ascorbic acid with likely associated enhanced antioxidant defense mechanisms, the bystander effect may allow the recycling of vitamin C in vivo, which may contribute to the low daily requirements of the vitamin in humans.     

n-3 Fatty acids modulate brain glucose transport in endothelial cells of the blood-brain barrier

We have previously shown that glucose utilization and glucose transport were impaired in the brain of rats made deficient in n-3 polyunsaturated fatty acids (PUFA). The present study examines whether n-3 PUFA affect the expression of glucose transporter GLUT1 and glucose transport activity in the endothelial cells of the blood-brain barrier. GLUT1 expression in the cerebral cortex microvessels of rats fed different amounts of n-3 PUFA (low vs. adequate vs. high) was studied. In parallel, the glucose uptake was measured in primary cultures of rat brain endothelial cells (RBEC) exposed to supplemental long chain n-3 PUFA, docosahexaenoic (DHA) and eicosapentaenoic (EPA) acids, or to arachidonic acid (AA). Western immunoblotting analysis showed that endothelial GLUT1 significantly decreased (-23%) in the n-3 PUFA-deficient microvessels compared to control ones, whereas it increased (+35%) in the microvessels of rats fed the high n-3 PUFA diet. In addition, binding of cytochalasin B indicated that the maximum binding to GLUT1 (Bmax) was reduced in deficient rats. Incubation of RBEC with 15 microM DHA induced the membrane DHA to increase at a level approaching that of cerebral microvessels isolated from rats fed the high n-3 diet. Supplementation of RBEC with DHA or EPA increased the [(3)H]-3-O-methylglucose uptake (reflecting the basal glucose transport) by 35% and 50%, respectively, while AA had no effect. In conclusion, we suggest that n-3 PUFA can modulate the brain glucose transport in endothelial cells of the blood-brain barrier, possibly via changes in GLUT1 protein expression and activity.     

Cerebral Astrocytes Transport Ascorbic Acid and Dehydroascorbic Acid Through Distinct Mechanisms Regulated by Cyclic AMP

Abstract: Cerebral ischemia and trauma lead to rapid increases in cerebral concentrations of cyclic AMP and dehydroascorbic acid (DHAA; oxidized vitamin C), depletion of intracellular ascorbic acid (AA; reduced vitamin C), and formation of reactive astrocytes. We investigated astrocytic transport of AA and DHAA and the effects of cyclic AMP on these transport systems. Primary cultures of astrocytes accumulated millimolar concentrations of intracellular AA when incubated in medium containing either AA or DHAA. AA uptake was Na⁺-dependent and inhibited by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), whereas DHAA uptake was Na⁺-independent and DIDS-insensitive. DHAA uptake was inhibited by cytochalasin B, d -glucose, and glucose analogues specific for facilitative hexose transporters. Once inside the cells, DHAA was reduced to AA. DHAA reduction greatly decreased astrocytic glutathione concentration. However, experiments with astrocytes that had been previously depleted of glutathione showed that DHAA reduction does not require physiological concentrations of glutathione. Astrocyte cultures were treated with a permeant analogue of cyclic AMP or forskolin, an activator of adenylyl cyclase, to induce cellular differentiation and thus provide in vitro models of reactive astrocytes. Cyclic AMP stimulated uptake of AA, DHAA, and 2-deoxyglucose. The effects of cyclic AMP required at least 12 h and were inhibited by cycloheximide, consistent with a requirement for de novo protein synthesis. Uptake and reduction of DHAA by astrocytes may be a recycling pathway that contributes to brain AA homeostasis. These results also indicate a role for cyclic AMP in accelerating the clearance and detoxification of DHAA in the brain.     

n-3 Fatty acids modulate brain glucose transport in endothelial cells of the blood–brain barrier

We have previously shown that glucose utilization and glucose transport were impaired in the brain of rats made deficient in n-3 polyunsaturated fatty acids (PUFA). The present study examines whether n-3 PUFA affect the expression of glucose transporter GLUT1 and glucose transport activity in the endothelial cells of the blood–brain barrier. GLUT1 expression in the cerebral cortex microvessels of rats fed different amounts of n-3 PUFA (low vs. adequate vs. high) was studied. In parallel, the glucose uptake was measured in primary cultures of rat brain endothelial cells (RBEC) exposed to supplemental long chain n-3 PUFA, docosahexaenoic (DHA) and eicosapentaenoic (EPA) acids, or to arachidonic acid (AA). Western immunoblotting analysis showed that endothelial GLUT1 significantly decreased (−23%) in the n-3 PUFA-deficient microvessels compared to control ones, whereas it increased (+35%) in the microvessels of rats fed the high n-3 PUFA diet. In addition, binding of cytochalasin B indicated that the maximum binding to GLUT1 (Bmax) was reduced in deficient rats. Incubation of RBEC with 15 μM DHA induced the membrane DHA to increase at a level approaching that of cerebral microvessels isolated from rats fed the high n-3 diet. Supplementation of RBEC with DHA or EPA increased the [³H]-3-O-methylglucose uptake (reflecting the basal glucose transport) by 35% and 50%, respectively, while AA had no effect. In conclusion, we suggest that n-3 PUFA can modulate the brain glucose transport in endothelial cells of the blood–brain barrier, possibly via changes in GLUT1 protein expression and activity.     

Retracted: Superoxide‐dependent uptake of vitamin C in human glioma cells

Glioblastomas are lethal brain tumors that resist current cytostatic therapies. Vitamin C may antagonize the effects of reactive oxygen species (ROS) generating therapies; however, it is often used to reduce therapy‐related side effects despite its effects on therapy or tumor growth. Because the mechanisms of vitamin C uptake in gliomas are currently unknown, we evaluated the expression of the sodium‐vitamin C cotransporter (SVCT) and facilitative hexose transporter (GLUT) families in human glioma cells. In addition, as microglial cells can greatly infiltrate high‐grade gliomas (constituting up to 45% of cells in glioblastomas), the effect of TC620 glioma cell interactions with microglial‐like HL60 cells on vitamin C uptake (Bystander effect) was determined. Although glioma cells expressed high levels of the SVCT isoform‐2 (SVCT2), low functional activity, intracellular localization and the expression of the dominant‐negative isoform (dnSVCT2) were observed. The increased glucose metabolic activity of glioma cells was evident by the high 2‐Deoxy‐d ‐glucose and dehydroascorbic acid (DHA) uptake rates through the GLUT isoform‐1 (GLUT1), the main DHA transporter in glioblastoma. Co‐culture of glioma cells and activated microglial‐like HL60 cells resulted in extracellular ascorbic acid oxidation and high DHA uptake by glioma cells. This Bystander effect may explain the high antioxidative potential observed in high‐grade gliomas.

     

Glucose as a regulator of insulin-sensitive hexose uptake in 3T3 adipocytes

In the present study we examined the role of glucose in the regulation of its own transport activity in the cultured 3T3 fat cell. A regulatory control of glucose became apparent after these cells were cultured in the absence of glucose. Glucose deprivation of the cells was accompanied by a specific time and protein synthesis-dependent increase in dGlc (2-deoxyglucose) uptake (up to 5-fold), which was due to an increase in the apparent Vmax of the transport system. Concomitantly, the stimulatory effect of insulin on hexose uptake almost completely disappeared. Addition of glucose to the glucose-deprived cells rapidly reversed the deprivation effects. Cycloheximide experiments revealed that the glucose deprivation-induced increase in hexose uptake required protein synthesis as well as a protein synthesis-independent response to glucose deprivation that retarded the turnover of hexose transport activity. Taken together, these data indicate that glucose deprivation is accompanied by retardation of the rate of degradation, internalization, or inactivation of hexose transporters while the increase in dGlc uptake requires at least the continuation of protein synthesis-dependent de novo synthesis, insertion, or activation of hexose transporters. Hexose competitively taken up with dGlc, including the nonmetabolizable glucose analogue 3-O-methylglucose, could replace glucose in the process of prevention and reversal of the deprivation effects, indicating that competitive transport but not the metabolism of hexose is a prerequisite for the regulatory effect of glucose on the activity of its own transport system. In conclusion, our results indicate that in cultured 3T3 fat cells glucose itself is involved in the regulation of the activity of its own transport system by influencing the rate of degradation, internalization, or inactivation of hexose transporters by a protein synthesis-independent mechanism.     

6-Bromo-6-deoxy-L-ascorbic acid: an ascorbate analog specific for Na+-dependent vitamin C transporter but not glucose transporter pathways

Vitamin C intracellular accumulation is mediated by Na(+)-dependent vitamin C transporters SVCT1 and -2 and dehydroascorbic acid transporters GLUT1 and -3. It is unclear which pathways dominate in vivo. As a new step to resolve this issue, we identified and tested 6-bromo-6-deoxy-L-ascorbic acid as a specific candidate for SVCTs. In high performance liquid chromatography and electron paramagnetic resonance analyses, the reduced compounds ascorbic acid and 6-bromo-6-deoxy-L-ascorbic acid were similar. The oxidized products 6-bromo-6-deoxy dehydroascorbic acid (BrDHA) and dehydroascorbic acid (DHA) had comparable stabilities, based on reduction recoveries. Upon expression of GLUT1 or GLUT3 in Xenopus oocytes, BrDHA was neither transported nor bound, in contrast to robust transport of DHA. The findings were not explained by differences in the oocyte reduction of DHA and BrDHA because lysed oocytes reduced both compounds equally. Further, there was no transport of the reduced compound, 6-bromo-6-deoxy-L-ascorbic acid, by GLUT1 or GLUT3. As a prerequisite for investigating 6-bromo-6-deoxy-L-ascorbic acid transported by SVCTs, SVCT2 transport activity in oocytes was enhanced 14-fold by construction and use of a vector that added a fixed poly(A) tail to the 3' end of cRNA. For SVCT1 and SVCT2 expressed in oocytes, similar K(m) and V(max) values were observed for ascorbic acid and 6-bromo-6-deoxy-L-ascorbic acid. In human fibroblasts, predicted to have SVCT-mediated ascorbate accumulation, K(m) and V(max) values were again comparable for ascorbic acid and 6-bromo-6-deoxy-L-ascorbic acid. Using activated human neutrophils, predicted to have ascorbate accumulation mediated predominantly by DHA and GLUT transporters, 6-bromo-6-deoxy-L-ascorbic acid accumulation was <1% of accumulation when compared with ascorbic acid. We conclude that 6-bromo-6-deoxy-L-ascorbic acid is the first transport substrate identified as completely specific for SVCTs, but not GLUTs, and provide a new strategy to determine the contribution of each pathway to ascorbate accumulation.