The argument rests on a distinctive feature of red blood cell biochemistry. Mature erythrocytes lack mitochondria and nuclei, which means the pentose phosphate pathway — the metabolic route in which G6PD catalyzes the rate-limiting step — is their only source of NADPH. G6PD converts glucose-6-phosphate to 6-phosphoglucono-δ-lactone while reducing NADP+ to NADPH. That coenzyme fuels the reduction of glutathione and thioredoxin, the cell’s primary antioxidants. Without adequate NADPH, glutathione reductase cannot regenerate reduced glutathione (GSH) from its oxidized form (GSSG), and the entire antioxidant cascade that depends on GSH begins to falter. In healthy red blood cells, the NADPH/NADP+ ratio is approximately 100:1, and the GSH/GSSG ratio is around 500:1, values that must be maintained for signaling, biosynthesis, and antioxidant defense.

The clinical stakes are considerable. G6PD deficiency is the most common X-linked enzymatic disorder worldwide, and more than 400 variants have been described. The World Health Organization classifies these variants by residual enzymatic activity and clinical presentation, which helps explain why the onset and severity of hemolysis differ so widely between patients. But the perspective also emphasizes that G6PD impairment is not exclusively inherited. In diabetes, prolonged hyperglycemia activates the cAMP/PKA pathway, leading to inhibitory phosphorylation of G6PD that dramatically reduces its enzymatic activity even though protein synthesis remains normal. Acquired deficiency has also been documented in a hematological malignancy evolving into chronic myelomonocytic leukemia, where genetic analysis showed the defect arose from clonal expansion of precursor cells carrying a G6PD null variant. Both inherited and acquired forms converge on the same biochemical endpoint: reduced NADPH generation, weakened antioxidant defense, and increased vulnerability to regulated cell death.

Central to the article’s argument is a distinction between two broad categories of cell death. Accidental cell death is an uncontrollable process triggered by sudden physical or environmental damage that disrupts membrane integrity; it does not rely on intracellular signaling and cannot be targeted pharmacologically. Regulated cell death, by contrast, develops gradually, giving the cell time to mount protective responses — and offering drug discovery a foothold. The classical view of red cell loss in G6PD deficiency attributed it to osmotic fragility or mechanical breakage, particularly in the spleen, where oxidatively damaged cells stiffen as hemoglobin precipitates into Heinz bodies and lose their deformability. Ulusu contends that this passive-rupture model misses the point: red blood cells can undergo regulated death or programmed clearance before complete hemolysis ever occurs.

Among the regulated pathways, eryptosis has the strongest direct evidence in G6PD-deficient human red blood cells. Eryptosis is the predominant form of regulated cell death in mature erythrocytes, conventionally characterized by oxidative stress, cytosolic calcium influx, and membrane remodeling. In its canonical form, calcium entry activates calpain proteases that degrade cytoskeletal proteins such as Band 3 and spectrin, driving membrane blebbing. In G6PD deficiency, however, the picture is more complicated: phosphatidylserine externalization is consistently observed, but intracellular calcium levels are not always elevated at baseline. This suggests that eryptosis in G6PD-deficient cells may proceed through calcium-independent routes, driven by oxidative cross-linking of membrane proteins and severe energy depletion. Either way, the stressed cells shrink, expose phosphatidylserine on their surface, and are recognized and engulfed by macrophages — a controlled clearance that, under mild stress, may allow defective cells to escape intravascular rupture altogether.

The ferroptosis axis is more tentative. Ferroptosis is a form of regulated necrosis driven by iron-dependent lipid peroxidation, and G6PD has been highlighted in several studies as a key guardian against it. The proposed mechanism runs through the G6PD–NADPH–GSH–GPX4 chain: NADPH fuels glutathione reductase, which regenerates GSH, and the selenoprotein glutathione peroxidase 4 (GPX4) strictly requires GSH as its obligate electron donor to clear lipid peroxides. When NADPH supply is limited, the GSH pool is exhausted and GPX4-dependent lipid peroxide clearance is secondarily impaired — not because the enzyme itself is directly inactivated, but because its cofactor has run out. Mature erythrocytes are especially susceptible to this kind of damage: their plasma membranes are rich in polyunsaturated fatty acids, and free iron released from labile hemoglobin pools can catalyze lethal peroxidation of the lipid bilayer. Notably, the author argues that even mitochondria-lacking red blood cells can be pushed toward ferroptosis-like injury by upstream NADPH restriction. Yet the article is careful to state that higher lipid peroxidation does not prove ferroptosis occurs in mature erythrocytes; GPX4-dependent lipid peroxidation and ferroptotic signatures in stored red blood cells provide supporting relevance, but direct proof is still lacking.

The most speculative pathway is erythronecroptosis, a newly recognized form of regulated necrosis in mature erythrocytes driven by the classical RIPK1, RIPK3, and MLKL signaling cascade rather than physical or osmotic trauma. When activated, phosphorylated MLKL forms oligomeric pores in the plasma membrane, causing cell swelling, rupture, and the systemic release of pro-inflammatory damage-associated molecular patterns, or DAMPs. This would convert erythrocyte death from a physiologically silent event into a potent driver of intravascular inflammation and vascular pathology. Crucially, however, there is currently no direct experimental data showing that G6PD deficiency or NADPH depletion activates the RIPK1/RIPK3/MLKL pathway in mature red blood cells. The article insists that erythronecroptosis be treated strictly as a hypothesis requiring experimental proof, not as an established feature of G6PD deficiency.

The perspective also stresses that these death signatures are not equivalent outcomes and should not be conflated. Eryptosis promotes phosphatidylserine-mediated clearance of intact cells by macrophages, which can prevent intravascular rupture. Erythronecroptosis is intrinsically lytic and inflammatory. Classic hemolysis involves membrane rupture from physical or osmotic stress or toxins, against which antioxidant defenses are ineffective once damage is sudden. Ferroptosis-like lipid peroxidation involves collapse of the lipid bilayer, regulated by GPX4, though its link to membrane rupture in mature red blood cells remains incompletely defined. Rather than three independent death programs being switched on simultaneously, the author suggests G6PD deficiency may create a common redox environment in which different cell-fate programs become accessible depending on the severity of stress and the residual activity of the enzyme.

Supporting biochemical data lend the model plausibility. Tandem mass spectrometry analysis in G6PD-deficient newborns demonstrated a significant drop in GSH levels, from 370 ± 53.2 μmol/L in healthy controls to 242.9 ± 15.9 μmol/L, with a GSH/GSSG ratio of 14.9 ± 7.2 — far below the roughly 500:1 expected in healthy erythrocytes. Adult human plasma typically contains 2 to 5.17 µM GSH and around 0.14 µM GSSG. Under severe oxidative load, catalase, peroxiredoxin-2, and glutathione peroxidase enzymes work as a coupled system linked to GSH and NADPH to detoxify endogenous and exogenous oxidants. However, the article cautions that no validated threshold for NADPH/NADP+ or GSH/GSSG ratios has been established to predict when redox imbalance tips into eryptosis or hemolysis. Reported values vary with sample type, measurement method, and experimental conditions, so these indicators should be read as measures of antioxidant status rather than as predictors of a specific cellular fate. Standardized assays and well-characterized erythrocyte populations will be needed to determine whether reproducible thresholds exist.

The practical implications point upstream. If hemolysis is a terminal outcome of redox failure rather than the primary mechanism, then protecting red blood cells may depend on preserving NADPH rather than repairing secondary membrane damage. Early upstream regulators of G6PD activity — growth factors such as PDGF, EGF, and VEGF, and signaling proteins including PI3K, mTOR, and TIGAR — offer one route, but targeting them carries systemic risks, including oncogenic potential in nucleated cells. The author instead prioritizes the discovery of direct, selective small-molecule G6PD activators that could safely restore enzymatic function in both inherited and acquired deficiency without off-target tumorigenesis. As a perspective article, the work contains no primary experimental data; it evaluates current literature to advance a testable framework. Within that framework, eryptosis stands on the firmest ground, ferroptosis-like lipid injury is plausible but unproven in mature red cells, and erythronecroptosis remains an open hypothesis — a research agenda for determining whether one enzyme’s output truly decides how a red blood cell dies.

Subject of Research: The role of glucose-6-phosphate dehydrogenase and NADPH supply in redox homeostasis and regulated cell death of erythrocytes

Article Title: Glucose-6-phosphate dehydrogenase at the crossroads of redox homeostasis and regulated cell death in erythrocytes

Article References: Ulusu, N. N. (2026). Glucose-6-phosphate dehydrogenase at the crossroads of redox homeostasis and regulated cell death in erythrocytes. Frontiers in Pharmacology, 17, Article 1958205. https://doi.org/10.3389/fphar.2026.1958205

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1958205

Keywords: G6PD deficiency, NADPH, eryptosis, ferroptosis, erythronecroptosis, red blood cells, oxidative stress, glutathione, GPX4, pentose phosphate pathway, hemolytic anemia, regulated cell death