Regulation of glucose 6-phosphate dehydrogenase expression in CHO-human fibroblast somatic cell hybrids

Human--hamster somatic cell hybrids have been obtained by fusion of a CHO line (NA31) doubly deficient in hypoxanthine guanine phosphoribosyltransferase and glucose 6-phosphate dehydrogenase (G6PD) with normal G6PD(+) human fibroblasts. Analysis of NA31 extracts has revealed that, although G6PD activity is nearly absent, significant activity can be detected with 2-deoxyglucose 6-phosphate as substrate, so that the mutant and normal forms of the enzyme can both be easily detected. The cell hybrids obtained express human G6PD. The human G6PD subunits are distributed in homodimeric molecules as well as in human--hamster heterodimeric molecules. However, whereas the amount of hamster G6PD subunits present in the hybrid is similar to that in the hamster parental cells, the amount of human G6PD subunits is decreased by 3- to 10-fold when compared to the human parental cell. These results indicate that either the expression of the G6PD gene or the stability of the gene product is altered in the hybrid. By mutagenesis and selection in diamide (a substance that oxidizes intracellular glutathione), we have isolated a clone with a 3- to 5-fold increase in human G6PD activity. This derivative may have an increased rate of expression of the human G6PD structural gene.     

Association of glyceraldehyde-3-phosphate dehydrogenase with the plasma membrane of the intact human red blood cell

Glycolytic enzymes have been observed to associate in vitro with membranes and cytoplasmic filaments in a variety of systems, but their distribution in vivo is contested. We have therefore examined the distribution of glyceraldehyde-3-phosphate dehydrogenase (G3PD) in the intact human erythrocyte using indirect immunofluorescence and affinity-purified rabbit antibodies to G3PD. Antibody specificity was demonstrated by immunoblotting as well as immunofluorescence experiments with ghosts specifically depleted of and reconstituted with G3PD. Anti-G3PD immunolabeling experiments utilized both fixed whole cells and fixed cell suspensions infused with 2.3 M sucrose, frozen and thick-sectioned. In all experiments a two-step fixation protocol was employed which ensured that cytoplasmic hemoglobin was retained when cells were subjected to Triton X-100 permeabilization, the anti-genicity of G3PD was preserved, and antibody penetration was complete. We used mixtures of biotinylated affinity-purified antibodies to G3PD and dichlorotriazinylaminofluorescein-labeled, affinity-purified antibodies to hemoglobin, followed by rhodamine-streptavidin, in double-label experiments. In both whole and sectioned human erythrocytes, G3PD staining was predominantly membrane associated while hemoglobin staining was diffusely distributed throughout the cytoplasm. In isolated ghosts, some G3PD was tightly bound to the membrane and was resistant to elution with phosphate-buffered saline and NAD+/arsenate. However, in immunolabeled rat reticulocytes and erythrocytes G3PD was cytoplasmic. Nucleated human blood cells and platelets also exhibited cytoplasmic G3PD. In approximately 10% of the human erythrocyte population G3PD was also cytoplasmic. These cells were flatter in shape and exhibited strong cytoplasmic immunolabeling for hemoglobin which was sometimes concentrated along the cell membrane; possibly, these cells were late reticulocytes or early erythrocytes. We conclude that G3PD is preferentially associated with the plasma membrane of human erythrocytes in a specific fashion.     

Membrane-associated purine metabolizing enzyme activities of human peripheral blood cells

Sealed and unsealed plasma membrane vesicles were prepared from human erythrocytes and lymphocytes. Phosphoribosylpyrophosphate synthetase (PRibPP synthetase), hypoxanthine phosphoribosyltransferase (HPRTase), and adenine phosphoribosyltransferase (APRTase) activities are detectable on both inside-out and right-side-out sealed vesicles. Ghost preparations were about 0.2%, 1%, and 1.2% of the total erythrocyte and 0.5%, 5.3%, and 9.7% of the lymphocyte APRTase, HPRTase, and PRibPP synthetase activities. The rapid decrease in these enzyme activities, upon further purification of the membranes, seemed to suggest that they might be loosely bound extrinsic proteins. Evidence confirming the localization of these enzymes on the cell surface was obtained by measuring production of [14C]AMP by intact cells in medium containing [14C]adenine, ribose 5-phosphate, and Mg2+ATP. The formation of AMP was linear with time and number of cells present. Magnesium and phosphate exerted different effects on the production of extracellular AMP than on intracellular, which involves transport as well as phosphoribosylation. Cytosoluble and membrane-bound APRTase and PRibPP synthetase exhibited different catalytic properties and sensitivities to effectors. Membranes of erythrocytes of HPRTase-deficient patients contain little or no HPRTase activity when assayed in the absence of Triton. Reisolation of these membranes from admixture with normal hemolysates did not result in any bound activity; thus, the membrane-bound activity is not an artifact of the isolation procedure. Lysis with Triton released activity equal to about half that of control membranes. This is further evidence that the enzyme is firmly bound to the membrane.     

The incorporation of a purified, membrane-bound form of guanylate cyclase into phospholipid vesicles and erythrocytes

The purified membrane-bound form of guanylate cyclase was incorporated into artificial unilamellar phospholipid vesicles. The rate and extent of enzyme incorporation into the vesicles was dependent upon the phospholipid concentration and the time period of incubation. The enzyme was incorporated at a significantly faster rate after removal of carbohydrate with endoglycosidase H. The incorporation of the enzyme led to a 10-fold decrease in the apparent maximal velocity and a 2-fold increase in the apparent Michaelis constant for MnGTP. Extraction of liposomes containing guanylate cyclase with 0.2% Lubrol PX resulted in the recovery of 85% of the original amount of added activity, suggesting that the decrease in maximal velocity was not due to enzyme denaturation. Phosphatidylcholine liposomes differentially effected the activity of the membrane-form of guanylate cyclase, dependent on the nature of the fatty acid present on the phospholipid. Specific activities ranged between 458 nmol/min per mg and 2.6 mumol/min per mg, dependent upon the fatty acids present. Liposomes containing the membrane-bound form of guanylate cyclase were subsequently fused with erythrocytes using poly(ethylene glycol) 4000 in attempts to introduce the enzyme into intact cells. The enzyme was successfully introduced into the erythrocytes; greater than 90% of the enzyme activity was subsequently shown to be associated with erythrocyte membranes. Cyclic GMP concentrations of erythrocytes increased from essentially nondetectable to 4 pmol/10(9) cells after introduction of the enzyme. These results demonstrate that guanylate cyclase can be incorporated into liposomes in an active state and that such liposomes can be used to introduce the enzyme into cells where it can subsequently function to generate cyclic GMP.     

Glucose 6-phosphate dehydrogenase: isoenzymatic pattern in Oesophagostomum venulosum, Trichuris ovis and T. suis.

In the present communication we have studied the isoenzymatic pattern activity of the glucose 6-phosphate dehydrogenase (G6PD) in Oesophagostomum venulosum, Trichuris ovis and T. suis, parasites of Capra hircus (goat), Ovis aries (sheep) and Sus scrofa domestica (pig) respectively, by polyacrylamide gel electrophoresis. Different phenotypes have been observed in the G6PD isoenzymatic pattern activity in males and females of Oesophagostomum venulosum. Furthermore, G6PD activity has been assayed in Trichuris ovis collected from Ovis aries and Capra hircus. No differences have been observed in the isoenzymatic patterns attending to the different hosts. All the individuals exhibited one single band or two bands; this suggests a monomeric condition for G6PD in T. ovis. In T. suis the enzyme G6PD appeared as a single electrophoretic band in about 85.7% of the individuals.     

Structural modification of erythrocyte membranes during oxidative stress and activity of membrane bound NADH-methemoglobin reductase

The activity of NADH-methemoglobin reductase (metHb-reductase) in membranes isolated from human erythrocytes treated with phenylhydrazine at its sublytic concentration was studied. A decrease in the activity of membrane-bound metHb-reductase was shown to depend on the concentration of phenylhydrazine. Simultaneously, an increase in the level of membrane-bound methemoglobin and a change in the fluorescence parameters of membrane-bound 4,4'-diisothiocy-anatostilbene-2,2'-disulfonic acid were registered. In the case when Hb-free erythrocyte ghosts were treated with 0.2-2.0 mM phenylhydrazine, the activity of metHb-reductase did not change. The obtained results indicate that the inhibition of the activity of membrane-bound metHb-reductase by phenylhydrazine-induced oxidative stress in human erythrocytes is not caused by the direct action of the oxidant on the enzyme. The reason for this is the interaction of the products of hemoglobin oxidation with erythrocyte membrane (protein band 3) and structural changes in membrane proteins.     

Disappearance of glyceraldehyde 3-phosphate dehydrogenase from erythrocyte membrane by hemolysis with thermostable direct hemolysin of Vibrio parahaemolyticus or Vibrio cholerae El Tor hemolysin.

It is believed that the thermostable direct hemolysin (TDH) of Vibrio parahaemolyticus and El Tor hemolysin (ETH) of Vibrio cholerae damage erythrocytes and other cells by acting as pore-forming toxins. In this study, we found that a protein band with a molecular weight of 37,000 daltons specifically disappeared from erythrocyte membrane after hemolysis by TDH and ETH, but not after hypotonic hemolysis. The 37 kDa band was identified as glyceraldehyde 3-phosphate dehydrogenase (G3PD), a glycolytic enzyme, based on N-terminal 14 amino acid sequencing. The role of G3PD in hemolysis is discussed.     

Regulation of G6PD acetylation by SIRT2 and KAT9 modulates NADPH homeostasis and cell survival during oxidative stress

Glucose‐6‐phosphate dehydrogenase (G6PD) is a key enzyme in the pentose phosphate pathway (PPP) and plays an essential role in the oxidative stress response by producing NADPH, the main intracellular reductant. G6PD deficiency is the most common human enzyme defect, affecting more than 400 million people worldwide. Here, we show that G6PD is negatively regulated by acetylation on lysine 403 (K403), an evolutionarily conserved residue. The K403 acetylated G6PD is incapable of forming active dimers and displays a complete loss of activity. Knockdown of G6PD sensitizes cells to oxidative stress, and re‐expression of wild‐type G6PD, but not the K403 acetylation mimetic mutant, rescues cells from oxidative injury. Moreover, we show that cells sense extracellular oxidative stimuli to decrease G6PD acetylation in a SIRT2‐dependent manner. The SIRT2‐mediated deacetylation and activation of G6PD stimulates PPP to supply cytosolic NADPH to counteract oxidative damage and protect mouse erythrocytes. We also identified KAT9/ELP3 as a potential acetyltransferase of G6PD. Our study uncovers a previously unknown mechanism by which acetylation negatively regulates G6PD activity to maintain cellular NADPH homeostasis during oxidative stress.     

Glyceraldehyde-3-phosphate dehydrogenase in the isolated human erthrocyte membrane: selective displacement by bilirubin.

When unsealed erythrocyte ghosts in 6 mm phosphate buffer (pH 8.0, 4 °C) were incubated with bilirubin in excess of 0.1 mm and washed with buffer, a single polypeptide component (band 6 in sodium dodecyl sulfate-polyacrylamide-gel electrophoresis) diminished and was recovered in the supernatant fraction. Release of this component was virtually complete at 1 mm initial bile pigment. Since band 6 was believed to be the protomer of membrane-bound glyceraldehyde-3-phosphate dehydrogenase (G3PD), assays for this enzyme in bilirubin-treated ghosts were carried out. These revealed that enzymatic activity decreased concurrently with the disappearance of band 6. The molecular weight of the eluted polypeptide was found to be 36,000, in agreement with the known value for the G3PD protomer. When Mg²⁺-resealed ghosts were washed after exposure to 1 mm bilirubin, less than 20% of the G3PD was eluted, which is consistent with the fact that the enzyme is attached to the cytoplasmic face of the membrane. NAD⁺ in concentrations up to 2 mm displaced no more than 15% of the G3PD from unsealed ghosts. However, after preincubation with NAD⁺ (1 mm) followed by bilirubin (1 Mm) and washing, loss of G3PD was similar to that observed in the absence of cofactor. Since NAD⁺ did not attenuate release of the enzyme, it appears unlikely that such release is attributable to binding of bilirubin at the active site. Protoporphyrin acted similarly to bilirubin on unsealed ghosts, whereas rose bengal had a more pronounced effect, removing all enzymatic activity when the dye concentration reached 0.2 mm. Electrophoretic analysis of ghosts after rose bengal treatment, however, revealed a diminution not only of band 6 but bands 1, 2, and 5 as well.     

Stimulatory effect of 1,25-dihydroxyvitamin D3 on the glucose-6-phosphate dehydrogenase activity in the MCF-7 human breast cancer cell line

Human breast cancer cell lines have been shown to possess high affinity receptors for 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) and their growth is inhibited by this steroid. The present study examines the effect of 1,25(OH)2D3 on the activity of glucose-6-phosphate dehydrogenase (G6PD) in cells of a human breast cancer cell line MCF-7. G6PD, an enzyme which controls the hexose monophosphate shunt, is elevated and sensitive to 17 beta-estradiol in breast tumors. G6PD activity was stimulated by 1,25(OH)2D3 in a dose-dependent manner at very low concentrations of steroid (10(-10)-10(-12) M). 1,25(OH)2D3 increased maximum velocity without modifying the affinity constant of the enzyme for glucose-6-phosphate.     

Flow cytofluorometric analysis of enzyme reactions based on quenching of fluorescence by the final reaction product: detection of glucose-6-phosphate dehydrogenase deficiency in human erythrocytes

We developed a method for accurate cytofluorometric analysis of the final reaction product of enzyme reactions in individual cells. Glucose-6-phosphate dehydrogenase (G6PD) activity in human erythrocytes was demonstrated cytochemically, and the amount of final reaction product (formazan) per cell was detected indirectly by quenching of autofluorescence generated by glutaraldehyde fixation. Formazan quenches fluorescence in a dose-dependent manner. The method has been used for detection of G6PD deficiency. Heterozygous and homo(hemi)zygous deficiency could easily be established, even in cases of extreme "Lyonization" where microscopic inspection failed to discriminate between either normal individuals and heterozygously deficient patients or heterozygously and homozygously deficient patients. The principle of quenching of fluorescence by final reaction products of enzymes can be applied to flow cytofluorometric analysis of enzyme activity in individual cells in general.     

In vivo gene marking of rhesus macaque long-term repopulating hematopoietic cells using a VSV-G pseudotyped versus amphotropic oncoretroviral vector

BACKGROUND: Gene transfer efficiency into primitive hematopoietic cells may be limited by their expression of surface receptors allowing vector entry. Vectors pseudotyped with the vesicular stomatitis virus (VSV-G) envelope do not need receptors to enter cells, and therefore may provide superior transduction efficiency. METHODS: Using a competitive repopulation model in the rhesus macaque, we examined in vivo gene marking levels of blood cells transduced with two vectors: (i) a VSV-G pseudotyped retrovirus and (ii) a conventional amphotropic retrovirus. The VSV-G vector, containing the human glucose-6-phosphate dehydrogenase (G6PD) gene, was constructed for treatment of severe hemolytic anemia caused by G6PD deficiency. Three myeloablated animals were transplanted with peripheral blood CD34+ cells, half of which were transduced with the VSV-G vector and the other half with the amphotropic vector. RESULTS: In all animals post-transplantation, levels of in vivo marking in circulating granulocytes and mononuclear cells were similar: 1% or less with both vectors. In one animal, the human G6PD enzyme transferred by the VSV-G vector was expressed in erythrocytes, early after transplantation, at a level of 45% of the endogenous rhesus G6PD protein. CONCLUSIONS: In a clinically relevant animal model, we found similar in vivo marking with a VSV-G pseudotyped and a standard amphotropic oncoretroviral vector. Amphotropic receptor expression may not be a limiting factor in transduction efficiency, but VSV-G pseudotypes possess other practical advantages that may make them advantageous for clinical use.     

Glucose-6-phosphate dehydrogenase from Plasmodium berghei: kinetic and electrophoretic characterization

Evidence is given for the existence of a parasite-specific glucose-6-phosphate dehydrogenase (G6PD) in Plasmodium berghei by characterization of its kinetic and electrophoretic properties. From infected rat erythrocytes the parasites were isolated, washed, and lysed. G6PD was purified by affinity chromatography with 2'5'-ADP-Sepharose 4B, although the separation of the malaria-specific enzyme from that of the host cell was not complete. Malarial G6PD significantly differed from the red cell enzyme with respect to its electrophoretic properties. In cellulose acetate electrophoresis, a band with catodic mobility was observed in addition to the anodically mobile host cell enzyme at pH 7.0. The subunits of the parasite-specific G6PD have a molecular weight of 55 kDa in contrast to 59 kDa of red cell G6PD subunits. The enzyme from P. berghei shows no cross-reactivity with polyclonal antibodies against G6PD from rat erythrocytes. Thus, a close evolutionary relationship between both proteins and the presence of proteolytic modifications could be excluded. The Km value for G6P of malarial G6PD is increased by one order of magnitude compared with the host cell enzyme.     

Resealing to small solutes of white erythrocyte membranes after incubation with EDTA, Ca2+, salt, sucrose, phospholipase C

White, stable erythrocyte ghosts can recover their impermeability to small solutes after storage for several days in low-ionic-strength phosphate buffers at 0 °C. The accessibility, to their substrates, of the inner surface enzymes, glyceraldehyde-3-phosphate dehydrogenase, (G3PD), and NADH cytochrome c oxidoreductase, was used to assess resealing. The data from the two enzymes were confirmatory. None of the conditions used to investigate resealing altered the activity of the outer surface enzyme, acetylcholinesterase. Using G3PD activity, ghosts (freshly prepared by gentle stepwise hemolysis in hypotonic phosphate buffers and stored in 11 mm phosphate buffer, pH 7.4) were shown to be slightly sealed (33%). Incubation at 37 °C in the storage buffer with or without EDTA did not alter their permeability. Ionic strength rather than osmotic pressure appears to influence the sealing process since salt (286 mosm) elicited 91% sealing whereas sucrose (278 mosm) had little effect. Calcium in trace amounts caused resealing to 80%. Phospholipase C (C. welchii) completely abolished Ca²⁺-induced resealing. The data were highly reproducible although these ghosts were found to contain only 10 to 20% of the G3PD activity of the leaky ghosts prepared by shock hemolysis in 5 mm phosphate buffer, pH 8.0. The response to the resealing agents was similar regardless of the level of G3PD present. Neither calcium nor ETDA altered the chemical composition (sialic acid, cholesterol, phospholipid) of the membranes. The small amount (5%) of nonspecific loosely bound protein lost during incubation, could not be attributed to any of the test agents. The results suggest that calcium induced the recovery of impermeability by altering the association, distribution, and/or conformation of the proteins and phospholipids within the membrane.     

Chronic nonspherocytic hemolytic anemia (CNSHA) and glucose 6 phosphate dehydrogenase (G6PD) deficiency in a patient with familial amyloidotic polyneuropathy (FAP)

Summary A new glucose-6-phosphate dehydrogenase (G6PD) variant with severe erythrocytic G6PD deficiency and a unique pH optimum is described in a young patient with chronic nonspherocytic hemolytic anemia (CNSHA) and familial amyloidotic polyneuropathy (FAP). Chronic hemolysis was present in the absence of infections, oxidant drugs or ingestion of faba beans. Residual enzyme activity was about 2.6% and 63% of normal activity in erythrocytes and leucocytes, respectively. A molecular study using standard methods showed G6PD in the patient to have normal electrophoretic mobility (at pH 7.0, 8.0 and 8.8), normal apparent affinity for substrates (K_m, G6P and NADP) and a slightly abnormal utilization of substrate analogues (decreased deamino-NADP and increased 2-deoxyglucose-6-phosphate utilization). Heat stability was found to be markedly decreased (8% of residual activity after 20 min of incubation at 46°C) and a particular characteristic of this enzyme was a biphasic pH curve with a greatly increased activity at low pH. Although molecular characteristics of this variant closely resemble those of G6PD Bangkok and G6PD Duarte, it can be distinguished from these and all other previously reported variants by virtue of its unusual pH curve. Therefore the present variant has been designated G6PD Clinic to distinguish it from other G6PD variants previously described.     

Determination of glucose-6-phosphate dehydrogenase isoenzymes of human erythrocytes by a thin-layer PAG-IEF procedure

A procedure of thin-layer PAG-IEF was employed to separate glucose-6-phosphate dehydrogenase (G6PD) isoenzymes of human erythrocytes into 15-16 bands in a pH range between 4.7 and 6.5. The G6PD isoenzymes were scanned with a dual-wavelength thin-layer chromatoscanner and the relative enzyme content of each band was calculated. The method possesses the merits of simplicity, high sensitivity, rapidity and high reproducibility.     

Glucose-6-phosphate dehydrogenase – from oxidative stress to cellular functions and degenerative diseases

Abstract

Glucose-6-phosphate dehydrogenase (G6PD), the first and rate-limiting enzyme of the pentose phosphate pathway, is indispensable to maintenance of the cytosolic pool of NADPH and thus the cellular redox balance. The role of G6PD as an antioxidant enzyme has been recognized in erythrocytes for a long time, as its deficiency is associated with neonatal jaundice, drug- or infection-mediated hemolytic crisis, favism and, less commonly, chronic non-spherocytic hemolytic anemia. To a large extent, advances in the field were made on the pathophysiology of G6PD-deficient erythrocytes, and the molecular characterization of different G6PD variants. Not until recently did numerous studies cast light on the importance of G6PD in other aspects of the physiology of both cells and organisms. Deficiency in G6PD activity, and hence a disturbance in redox homeostasis, can lead to dysregulation of cell growth and signaling, anomalous embryonic development, altered susceptibility to viral infection as well as increased susceptibility to degenerative diseases. The present review covers recent developments in this field. Additionally, molecular characterization of G6PD variants, especially those frequently found in Taiwan and Southern China, is also addressed.     

Glucose-6-phosphate dehydrogenase--from oxidative stress to cellular functions and degenerative diseases

Glucose-6-phosphate dehydrogenase (G6PD), the first and rate-limiting enzyme of the pentose phosphate pathway, is indispensable to maintenance of the cytosolic pool of NADPH and thus the cellular redox balance. The role of G6PD as an antioxidant enzyme has been recognized in erythrocytes for a long time, as its deficiency is associated with neonatal jaundice, drug- or infection-mediated hemolytic crisis, favism and, less commonly, chronic non-spherocytic hemolytic anemia. To a large extent, advances in the field were made on the pathophysiology of G6PD-deficient erythrocytes, and the molecular characterization of different G6PD variants. Not until recently did numerous studies cast light on the importance of G6PD in other aspects of the physiology of both cells and organisms. Deficiency in G6PD activity, and hence a disturbance in redox homeostasis, can lead to dysregulation of cell growth and signaling, anomalous embryonic development, altered susceptibility to viral infection as well as increased susceptibility to degenerative diseases. The present review covers recent developments in this field. Additionally, molecular characterization of G6PD variants, especially those frequently found in Taiwan and Southern China, is also addressed.     

Genetic regulation of glucose 6-phosphate dehydrogenase activity in the inbred mouse

The erythrocyte glucose 6-phosphate dehydrogenase activity characteristic of each of 16 inbred mouse strains falls into one of three distinct classes. Strains C57L/J and C57BR/cdJ represent the low activity class: strains A/J and A/HeJ represent the high activity class; other strains have intermediate activities. There is no evidence that structural variation is responsible for the variation in G6PD activity, since partially purified enzyme from each class has the same thermal stability, pH-activity profile, Michaelis constants for G6P and NADP, electrophoretic mobility, and activity using 2-deoxy d-glucose 6-phosphate as substrate. The activities of 6-phosphogluconate dehydrogenase and glucose phosphate isomerase do not differ in erythrocytes of the three G6PD activity classes. Young red cells have higher G6PD activities than old red cells and there is evidence that the intracellular stability of the enzyme is reduced in red cells of strain C57L/J. G6PD activities in kidney and skeletal and cardiac muscle from animals with low red cell G6PD are slightly lower than the activities in kidney and muscle from animals with high red cell G6PD activity. The quantitative differences in red cell G6PD activity are not regulated by X-linked genes, but by alleles at two or more autosomal loci. A simple genetic model is proposed in which alleles at two unlinked, autosomal loci, called Gdr-1 and Gdr-2 regulate G6PD activity in the mouse erythrocyte.     

Targeted disruption of the housekeeping gene encoding glucose 6-phosphate dehydrogenase (G6PD): G6PD is dispensable for pentose synthesis but essential for defense against oxidative stress.

Glucose 6-phosphate dehydrogenase (G6PD) is a housekeeping enzyme encoded in mammals by an X-linked gene. It has important functions in intermediary metabolism because it catalyzes the first step in the pentose phosphate pathway and provides reductive potential in the form of NADPH. In human populations, many mutant G6PD alleles (some present at polymorphic frequencies) cause a partial loss of G6PD activity and a variety of hemolytic anemias, which vary from mild to severe. All these mutants have some residual enzyme activity, and no large deletions in the G6PD gene have ever been found. To test which, if any, function of G6PD is essential, we have disrupted the G6PD gene in male mouse embryonic stem cells by targeted homologous recombination. We have isolated numerous clones, shown to be recombinant by Southern blot analysis, in which G6PD activity is undetectable. We have extensively characterized individual clones and found that they are extremely sensitive to H2O2 and to the sulfydryl group oxidizing agent, diamide. Their markedly impaired cloning efficiency is restored by reducing the oxygen tension. We conclude that G6PD activity is dispensable for pentose synthesis, but is essential to protect cells against even mild oxidative stress.