Bacterial Enzymes Reveal a Hidden NADPH Weakness in Aging Cells

When mitochondria falter, scientists have long focused on a single molecular culprit: a buildup of NADH, the reduced form of nicotinamide adenine dinucleotide that normally shuttles electrons into the respiratory chain. This state, known as NADH-reductive stress, is a hallmark of primary mitochondrial disorders and has been implicated in diabetes, cancer and aging. But a new study published in Nature Metabolism suggests that fixing the NADH problem alone is not enough. A team led by Valentin Cracan at the Scintillon Institute in San Diego shows that when the respiratory chain stalls, cells suffer a second, previously underappreciated crisis: a profound depletion of NADPH, the closely related phosphorylated cofactor that powers antioxidant defense, lipid synthesis and nucleotide production. The finding reframes how researchers think about mitochondrial dysfunction and points to a surprising therapeutic strategy borrowed from bacteria.

The two cofactors, NAD(H) and NADP(H), differ by nothing more than a phosphate group, yet they occupy nearly separate biochemical worlds. NAD(H) is the currency of energy metabolism, cycling between its oxidized and reduced forms as cells extract energy from sugars and fats. NADP(H), by contrast, is the cell’s reducing reservoir, donating electrons to build fatty acids, maintain glutathione in its antioxidant state and drive the biosynthesis of DNA precursors. Because the two pools are chemically similar, enzymes that interconvert them, called transhydrogenases, can shift reducing power between them. The Cracan team exploited this property by installing a soluble transhydrogenase from Escherichia coli, dubbed EcSTH, into human cells, creating a genetically encoded dial that deliberately pushes electrons from NADPH into NADH, raising the NADH/NAD+ ratio while draining the NADPH pool.

When the researchers expressed EcSTH in primary human fibroblasts, the IMR-90 cell line widely used in aging research, the consequences were dramatic. The cells accumulated NADH and simultaneously lost NADPH, and this double redox insult triggered sweeping metabolomic and transcriptomic reprogramming. Proliferation ground to a halt, and the fibroblasts acquired features of senescence, the irreversible growth arrest that accumulates in aging tissues. The team observed disruptions across central carbon metabolism, including the tricarboxylic acid cycle, glycolysis, the pentose phosphate pathway and pyrimidine biosynthesis. Mitochondrial membrane potential became hyperpolarized, and the coenzyme Q pool, which ferries electrons within the respiratory chain, became over-reduced, a signature of electron traffic backing up behind a blocked chain.

The critical experiment came next. If NADH accumulation were the sole driver of the damage, then normalizing the NADH/NAD+ ratio should rescue the cells. The researchers achieved this in two ways: by supplementing the medium with pyruvate, which consumes NADH as it is converted to lactate, and by co-expressing LbNOX, a water-forming NADH oxidase from Lactobacillus brevis that directly burns NADH. In transformed HeLa cancer cells, these interventions worked as expected, restoring proliferation. But in primary fibroblasts they failed. Even with NADH levels brought back to normal, the cells remained arrested, still senescent, still unable to divide. Something else was missing.

Stable isotope tracing with carbon-13-labeled glucose and glutamine revealed the answer. The fibroblasts, unlike HeLa cells, rely heavily on NADPH-driven reductive carboxylation, a reversal of the tricarboxylic acid cycle in which the enzyme isocitrate dehydrogenase consumes NADPH to convert glutamine-derived alpha-ketoglutarate into citrate. This NADPH-dependent route supplies the citrate and, downstream, the aspartate needed to build pyrimidine nucleotides. When EcSTH drained NADPH, this supply line collapsed. De novo pyrimidine biosynthesis stalled, aspartate accumulated in a metabolic dead end alongside beta-alanine, and the cells could no longer manufacture the DNA building blocks required for division. Cancer cells, which lean more on oxidative metabolism, simply did not share this vulnerability.

Having identified NADPH depletion as the true bottleneck, the team engineered a tool to fix it. They turned to NAD kinase, the enzyme that phosphorylates NAD+ to produce NADP+, the immediate precursor of NADPH. Using a NAD kinase from Staphylococcus aureus as a scaffold, they developed a constitutively active variant, SaNADK, that no longer requires the calmodulin-dependent regulation typical of mammalian versions. When expressed in fibroblasts, SaNADK expanded the cellular NADP(H) pool, restoring reductive carboxylation and reviving de novo pyrimidine synthesis. Remarkably, co-expression of SaNADK with EcSTH reversed the proliferation arrest and erased the senescence features, even though the NADH/NAD+ ratio remained perturbed. The cells’ problem had never been too much NADH alone; it had been too little NADPH.

The mechanistic picture that emerges is one of a metabolic safety valve. In healthy proliferating cells, a high NADPH/NADP+ ratio drives reductive carboxylation, which partially uncouples the tricarboxylic acid cycle from the electron transport chain. This limits NADH production and keeps the coenzyme Q pool sufficiently oxidized to accept electrons from dihydroorotate dehydrogenase, the mitochondrial enzyme that couples pyrimidine synthesis to respiration. When NADPH runs low, the valve fails: carbon flows through oxidative decarboxylation instead, generating excess NADH, hyperpolarizing the mitochondrial membrane and over-reducing coenzyme Q until dihydroorotate dehydrogenase can no longer function. Pyrimidine synthesis collapses, and the cell drifts into senescence.

The implications extend beyond cell culture. Because NADH-reductive stress and NADPH depletion co-occur in mitochondrial disease and, the authors argue, plausibly in aging, the team tested SaNADK in living animals. They delivered the enzyme to the livers of aged male mice using an adeno-associated vector and compared hepatic transcriptomes against those of young controls and aged mice receiving a fluorescent marker protein instead. Hepatic SaNADK expression partially alleviated age-associated transcriptional signatures of inflammation and metabolic dysregulation, though metabolomic profiling showed only modest shifts in NAD(P)(H) pools, long-chain acylcarnitines and sphingomyelins. The effect is partial and was observed only in male mice, so the authors are careful to frame it as a proof of concept rather than an anti-aging therapy.

Methodologically, the study adds two powerful instruments to the growing toolbox of xenotopic synthetic biology, the practice of installing bacterial enzymes into mammalian cells to probe and rewire metabolism. EcSTH provides a controllable way to impose NADH-reductive stress in cells with intact mitochondria, complementing earlier tools such as LbNOX and the mitochondrial transhydrogenase-based systems developed by the Mootha laboratory. SaNADK, in turn, offers direct genetic control over the size of the NADP(H) pool, something pharmacological NAD+ precursors like nicotinamide riboside and NMN cannot do, since they feed into NAD(H) and only indirectly affect NADPH. Together they allow researchers to perturb each redox couple independently and observe which metabolic vulnerabilities emerge.

The broader lesson is that redox couples cannot be treated in isolation. Much of the excitement around NAD+ biology in recent years has centered on the age-related decline in NAD+ and the promise of replenishing it. This study suggests that in contexts of mitochondrial dysfunction, the NADPH side of the ledger may be equally important, and that interventions normalizing NADH/NAD+ while leaving NADPH depleted will fail in primary cells, even if they succeed in transformed cell lines that dominate the literature. For patients with mitochondrial disorders, and perhaps for aging tissues burdened by senescent cells, the path forward may require bolstering NADPH-dependent biosynthesis rather than simply recycling NAD+. The bacterial enzymes that made this discovery possible may, if the patent applications filed by the team bear fruit, eventually become therapeutic agents in their own right.

Subject of Research: NAD(P)H redox homeostasis, mitochondrial dysfunction and NADPH-dependent metabolic vulnerabilities in aging

Article Title: Genetic modulation of NAD(P)H homeostasis reveals distinct redox-coupled metabolic vulnerabilities

Article References: Genetic modulation of NAD(P)H homeostasis reveals distinct redox-coupled metabolic vulnerabilities. (n.d.). https://doi.org/10.1038/s42255-026-01640-7

Image Credits: AI Generated

DOI: 10.1038/s42255-026-01640-7

Keywords: NADH reductive stress, NADPH, mitochondrial dysfunction, redox metabolism, senescence, pyrimidine biosynthesis, reductive carboxylation, NAD kinase, transhydrogenase, aging, metabolomics, Nature Metabolism