Reactive Oxygen Species: How Cellular Energy Production Damages Your DNA

The Price of Breathing: Harman's Radical Idea

In 1956, a physician-biochemist named Denham Harman sat down and wrote a paper proposing something that, at the time, seemed almost too simple to be true. Ageing, he argued, was not caused by some mysterious biological clock, nor by the inevitable wearing down of tissues over time. It was caused by chemistry: specifically, by the accumulation of damage inflicted by highly reactive molecules derived from oxygen, molecules he called free radicals. The free radical theory of ageing was born.

Harman's insight was that the same process that allows us to extract energy from food, the controlled burning of organic molecules using oxygen, inevitably produces reactive byproducts. Some electrons escape from the machinery and react directly with oxygen molecules, creating chemical species with unpaired electrons that are desperately eager to react with anything nearby, including the DNA inside our cells. This was not a design flaw, exactly, more an unavoidable trade-off in the chemistry of aerobic life.

In the seven decades since Harman's proposal, the theory has been substantially revised and refined. We now understand that these reactive oxygen species (ROS) are not merely damaging bystanders but important signaling molecules at low concentrations, essential for triggering adaptive responses to stress. The relationship between ROS and health is one of dose, context, and duration, making it far more nuanced than the simple narrative of bad molecules that antioxidant marketing implies.

What Reactive Oxygen Species Actually Are

The ROS Family

Reactive oxygen species is an umbrella term for a family of molecules that share a common origin (oxygen) and a common property (high chemical reactivity). The primary species in biology are superoxide anion (O2-), hydrogen peroxide (H2O2), and the hydroxyl radical (OH-). Each has distinct chemistry and distinct biological effects.

Superoxide is the first product: it forms when a single electron is added to molecular oxygen. In the mitochondria, this happens primarily at Complexes I and III of the electron transport chain, where electrons occasionally escape from their intended path and react directly with oxygen rather than being passed to the next protein complex. Under normal conditions, roughly 0.1 to 1 percent of electrons flowing through the chain take this rogue path. The exact rate varies with metabolic state, mitochondrial membrane potential, and the presence of inhibitors or damage. When the chain is backed up, with high membrane potential and abundant NADH but limited ADP to convert to ATP, more electrons leak, and superoxide production increases.

Hydrogen peroxide is the second major species, formed when the enzyme superoxide dismutase (SOD) converts two superoxide molecules into one hydrogen peroxide and one oxygen molecule. There are two forms of SOD in the cell: MnSOD in the mitochondrial matrix and Cu/ZnSOD in the cytoplasm and intermembrane space. Hydrogen peroxide is more stable than superoxide and can diffuse across membranes, making it a more versatile signaling molecule. It is also the precursor to the most dangerous ROS of all.

The hydroxyl radical is produced when hydrogen peroxide reacts with iron or copper ions in what is called the Fenton reaction (discovered by Henry Haber and Fritz Weiss in 1934). The hydroxyl radical is extraordinarily reactive, with a half-life measured in nanoseconds. It cannot travel far before it reacts with whatever molecule is nearest, and it reacts indiscriminately with lipids, proteins, carbohydrates, and DNA. It is probably the single most destructive molecule produced in normal cellular metabolism.

How ROS Attack DNA: The Molecular Details

DNA damage by ROS has been studied intensively since the 1980s, and the catalog of lesions is extensive. The hydroxyl radical attacks DNA bases by adding across double bonds or abstracting hydrogen atoms, producing a diverse array of modified bases. The most studied lesion is 8-hydroxy-2-deoxyguanosine (8-OHdG), formed when the hydroxyl radical attacks the C8 position of guanine. This modified base is significant because it mispairs with adenine instead of cytosine, causing G-to-T transversion mutations when the damaged strand is replicated. 8-OHdG is used clinically as a biomarker of oxidative DNA damage: elevated urinary 8-OHdG levels have been found in patients with cancer, cardiovascular disease, diabetes, and neurodegenerative conditions, and higher levels correlate with faster biological ageing in some studies.

Beyond base modifications, ROS cause single-strand breaks, where one strand of the double helix is cut, and double-strand breaks, where both strands are severed simultaneously. Double-strand breaks are the most dangerous form of DNA damage: if not repaired correctly, they can lead to large chromosomal rearrangements. The cell has elaborate repair machinery for both types of damage. Base excision repair (BER) handles oxidized bases, using enzymes called glycosylases to cut out the damaged base and replace it with a correct one. Double-strand break repair uses either homologous recombination (which requires a sister chromatid as a template and is accurate) or non-homologous end joining (which is faster but error-prone).

Mitochondrial DNA is particularly vulnerable to ROS damage. It lacks the histone proteins that compact and protect nuclear DNA. It resides in the mitochondrial matrix, in close physical proximity to the electron transport chain complexes that generate ROS. And its repair mechanisms, while present, are less comprehensive than those protecting nuclear DNA. The result is that mtDNA accumulates oxidative lesions at an estimated 10 to 20 times the rate of nuclear DNA. Over a lifetime, this accumulating mtDNA damage progressively impairs the function of the electron transport chain complexes encoded by mtDNA, creating the vicious cycle that many researchers believe is the fundamental engine of biological ageing. For a detailed look at how this connects to the broader biology of getting old, see our article on the mitochondrial theory of ageing.

The Body's Antioxidant Arsenal

Evolution has not left cells defenseless against ROS. A multi-layered antioxidant system scavenges ROS at each stage of their production. Superoxide dismutase is the first line of defense, converting superoxide to the less reactive hydrogen peroxide. Catalase, found primarily in peroxisomes, converts hydrogen peroxide to water and oxygen. Glutathione peroxidase, which requires the trace mineral selenium as a cofactor, also reduces hydrogen peroxide and lipid peroxides, using glutathione (a tripeptide antioxidant) as the electron donor. Thioredoxin reductase provides another layer of peroxide reduction. The fat-soluble antioxidants vitamin E (alpha-tocopherol) and CoQ10 neutralize lipid peroxyl radicals in cell membranes before they can propagate chain reactions through the membrane bilayer.

The master regulator of this endogenous antioxidant system is a transcription factor called NRF2 (nuclear factor erythroid 2-related factor 2). Under basal conditions, NRF2 is held in the cytoplasm by a protein called KEAP1, which targets it for degradation. When cells encounter oxidative stress or certain plant chemicals called phytochemicals, KEAP1 is modified and releases NRF2, which travels to the nucleus and activates the expression of hundreds of antioxidant and cytoprotective genes. This NRF2 pathway is the reason why eating a diet rich in cruciferous vegetables, which contain sulforaphane (a potent NRF2 activator), may be more beneficial than taking high-dose antioxidant supplements. It stimulates the cell's own defenses rather than providing external antioxidants that may interfere with signaling.

The Antioxidant Paradox and Hormesis

One of the most important and counterintuitive findings in biology over the past two decades is that ROS are not simply bad. At low to moderate concentrations, superoxide and hydrogen peroxide function as essential signaling molecules. Hydrogen peroxide, for example, activates tyrosine kinase signaling pathways downstream of growth factor receptors by oxidizing and inactivating phosphatase enzymes that would otherwise switch the signal off. ROS produced during exercise signal mitochondria to increase biogenesis, telling the cell that more energy capacity is needed. Immune cells deliberately produce high concentrations of ROS using the NADPH oxidase enzyme to kill invading bacteria.

This biological reality explains what has become known as the antioxidant paradox. If ROS are beneficial at low concentrations, then flooding the system with exogenous antioxidants might blunt these adaptive signals. Multiple large clinical trials have found that high-dose antioxidant supplementation does not reduce disease risk and may in some cases increase it. The ATBC (Alpha-Tocopherol Beta-Carotene) trial found that beta-carotene supplementation increased lung cancer risk in smokers by 18 percent. Meta-analyses pooling data from dozens of antioxidant supplement trials have consistently found no mortality benefit and suggest a small but statistically significant increase in all-cause mortality with high-dose vitamin E, vitamin A, and beta-carotene supplementation.

The concept of hormesis, borrowed from toxicology, helps make sense of this. Hormesis refers to a biphasic dose-response relationship where a low dose of a stressor produces a beneficial adaptive response while a high dose causes harm. Exercise is the clearest example: a moderate amount produces health benefits through the ROS and inflammatory signals it generates; an extreme amount causes overtraining syndrome and tissue damage. The same principle appears to apply to oxidative stress itself. Some ROS, at the right time and place, are a stimulus that keeps the body's defenses sharp. This is one reason why exercise remains the most effective mitochondrial and longevity intervention, and why blunderbuss antioxidant therapy has largely failed to deliver on its initial promise.

ROS in Disease: From Long COVID to Cancer

When ROS production chronically exceeds antioxidant capacity, the resulting oxidative stress drives a recognizable pattern of cellular damage across multiple disease states. In cardiovascular disease, oxidized LDL cholesterol (created when LDL particles react with ROS) is taken up by macrophages in arterial walls, forming foam cells and initiating atherosclerotic plaques. In diabetes, chronic hyperglycemia stimulates excessive mitochondrial ROS production in endothelial cells, directly damaging blood vessel walls and contributing to the microvascular complications of the disease.

In neurodegenerative diseases, oxidative damage is among the earliest detectable pathological changes. In Alzheimer's disease, markers of oxidative damage including 8-OHdG and protein carbonylation are elevated in affected brain regions before significant amyloid or tau pathology develops. In Parkinson's disease, the selective vulnerability of dopamine neurons in the substantia nigra is partly explained by the fact that dopamine metabolism itself produces hydrogen peroxide as a byproduct, creating a uniquely high oxidative burden in these cells.

The connection to post-viral syndromes is also emerging as an important area. As explored in our article on mitochondria and long COVID, viral infections can trigger a sustained ROS overproduction state that outlasts the infection itself, damaging mitochondria and creating a self-perpetuating cycle of energy deficit and inflammation. The detailed mechanisms through which mitochondrial dysfunction underlies chronic disease are increasingly understood to involve ROS as a central molecular mediator.

Measuring and Managing Oxidative Stress

Because ROS themselves are too short-lived to measure directly in a clinical setting, oxidative stress is assessed through stable markers of the damage it causes. Urinary 8-OHdG reflects DNA oxidation. F2-isoprostanes in urine or plasma measure lipid peroxidation. Protein carbonyl content measures oxidative protein damage. Malondialdehyde (MDA) is another lipid peroxidation marker measurable in plasma. These markers can provide a snapshot of ongoing oxidative stress load, and elevated levels correlate with disease risk in epidemiological studies.

On the intervention side, the most evidence-supported strategies for maintaining healthy ROS balance are lifestyle-based. Regular moderate exercise is the strongest intervention: it transiently increases ROS signaling, which activates NRF2 and stimulates mitochondrial biogenesis, and the adapted cells become more resistant to subsequent oxidative insults. A diet rich in colorful plant foods provides polyphenols and glucosinolates that activate NRF2 through KEAP1 modification, upregulating endogenous antioxidant defenses. Adequate sleep is essential: sleep deprivation measurably increases oxidative stress markers. Avoiding unnecessary exposure to environmental ROS generators, including cigarette smoke, industrial chemicals, and excessive ionizing radiation, reduces the external oxidative burden.

For those interested in targeted supplementation, the most rational approach based on the current evidence is to support endogenous antioxidant systems rather than simply adding more exogenous antioxidants. This means ensuring adequate selenium (for glutathione peroxidase), zinc and copper (for SOD), and consuming NRF2-activating foods like broccoli sprouts, garlic, and green tea. The story of reactive oxygen species is ultimately a story about balance: one of biology's most elegant and most easily disrupted equilibria, with profound consequences for health across the entire lifespan.