Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo Original paper

What was studied?

This consensus statement presents reactive oxygen species measurement guidelines for accurately defining, detecting, and interpreting reactive oxygen species (ROS), antioxidants, and oxidative damage in cells and in vivo. The authors argue that “ROS” is not a single entity but a family of chemically distinct species (for example, superoxide, hydrogen peroxide, hydroxyl radical, hypochlorous acid, and peroxynitrite) with different reactivities, lifetimes, diffusion ranges, and biological consequences, so experiments must name and target specific ROS wherever possible. They focus on best-practice methodology: how to select assays, how to avoid artefacts from commonly used commercial kits and fluorescent probes, and how to corroborate findings using orthogonal (independent) measurement approaches. Central recommendations include avoiding vague “antioxidant” interpretations unless the chemistry is plausible and oxidative damage biomarkers decrease accordingly, and prioritising validated approaches (for example, LC–MS/MS for specific probe products or lipid peroxidation biomarkers) over nonspecific fluorescence readouts.

Who was studied?

No human participants, patient cohorts, or animal populations were studied directly. Instead, an international multidisciplinary panel of redox biology experts synthesised evidence and practical experience to create methodological recommendations applicable across experimental contexts: test-tube systems, cultured cells, isolated organs, animal models, human tissues, and clinical trials. The “who” in this work is therefore the intended user community—researchers and clinicians who measure ROS or oxidative damage but may not have deep training in redox chemistry, especially those at risk of misinterpreting kit-based or nonspecific probe outputs as definitive measures of “ROS” or “oxidative stress.”

Most important findings

The statement’s core finding is that many widely used ROS assays are intrinsically prone to artefact and overinterpretation, particularly when nonspecific probes are treated as reporting a single ROS species. For example, DCFH-DA is frequently misused as a “hydrogen peroxide” detector despite requiring conversion of H₂O₂ into more reactive intermediates and being influenced by local oxygen and pH; it should be limited to preliminary “redox state change” screening and followed by more specific methods. Similarly, fluorescence from dihydroethidium or MitoSOX cannot be assumed to represent superoxide unless the superoxide-specific product (2-hydroxyethidium) is independently validated—ideally quantified by LC–MS. The authors emphasise chemically credible perturbations (for example, paraquat/quinones for superoxide generation; chemogenetic d-amino acid oxidase systems for controlled intracellular H₂O₂) and discourage reliance on poorly specific inhibitors like apocynin or diphenyleneiodonium as sole evidence for NADPH oxidase involvement. They also warn against measuring ROS in homogenates or cryosections because reactive species will be gone and sample disruption can create artificial ROS. Importantly for biomarker work, they recommend MS-based quantification for lipid peroxidation (for example, F₂-isoprostanes) and advise against TBARS-only readouts due to low specificity.

Key implications

Clinically translational ROS research, including studies linking microbiome features to “oxidative stress,” should move away from single-probe, kit-driven conclusions and toward ROS-specific, mechanism-consistent measurements supported by orthogonal validation. For intervention trials, the statement argues that antioxidant therapies are uninterpretable unless validated biomarkers demonstrate reduced oxidative damage in vivo; otherwise, negative outcomes may simply reflect failure to modify the intended biology. Adopting these guidelines improves reproducibility, reduces false biomarker associations, and strengthens any downstream attempt to catalogue meaningful, comparable signatures (including microbiome-associated host redox phenotypes) across studies.

Citation

Murphy MP, Bayir H, Belousov V, et al. Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo. Nat Metab. 2022;4(6):651-662. doi:10.1038/s42255-022-00591-z