Reply to Queimado et al.: E-cigarettes induce DNA damage and inhibit DNA repair in mice and human cells

We believe that Queimado et al. (1) may have inadvertently missed several key details presented in our recent PNAS publication (2).

First, for the determination of nicotine effects on DNA adduct formation and DNA repair activity, the nicotine concentrations used for 1-h treatments of human lung and bladder epithelial cells were based on cytotoxicity (killing <20%). For lung cells, the concentrations used were 0, 100, and 200 μM. For urothelial cells, the concentrations used were 0, 1, and 2.5 μM (163–410 ng/mL), which are similar to the nicotine level generated by 100 puffs of an E-cigarette (2, 3).

Second, human cells were treated with pure nicotine or with the nicotine-derived nitrosamine ketone (NNK), not E-cigarette smoke (ECS) (2). Therefore, all effects on DNA adduct formation and DNA repair activity are unambiguously due to nicotine and NNK exposure, and thus are not derived from “other electronic cigarette genotoxic” components, as suggested by Queimado et al. (1).

Third, our results showed that both nicotine and NNK induced γ-hydroxy-propano-deoxyguanosine (γ-OH-PdG) as well as O⁶-methyl-dG (2). While there is a slight possibility that γ-OH-PdG can be induced by aldehydes generated by lipid peroxidation (LPO), this process cannot induce O⁶-methyl-dG. However, it is well established that NNK can induce O⁶-methyl-dG (4). Taken together, these results further support our argument that nicotine induces DNA adducts, and that this process is mediated by its nitrosation product, NNK.

Fourth, we demonstrated that nicotine and NNK reduced not only DNA repair activity but also DNA repair proteins XPC and hOGG1/2 (2). We performed both of these two experiments mindful that a reduction of repair proteins is not necessarily equivalent to a reduction of DNA repair activity.

Finally, Queimado et al. (1) argue that the DNA adducts we detected in nicotine-treated human cells or in mice exposed to ECS may not be induced by nicotine or ECS, but rather are an effect of accumulation of endogenous DNA damage because nicotine and ECS inhibit DNA repair. We did consider the possibility that the nicotine metabolism byproduct formaldehyde may trigger LPO and that the aldehyde byproducts of LPO may induce γ-OH-PdG (2). However, this still does not account for the induction of O⁶-methyl-dG. Since nicotine treatment induces both γ-OH-PdG and O⁶-methyl-dG (2), we conclude that it is most likely that nicotine metabolites induce γ-OH-PdG and O⁶-methyl-dG by direct interaction with genomic DNA.

We have previously reported that both tobacco smoke (side stream) and E-cigarette smoke induce mainly γ-OH-PdG adducts, and that both cause DNA repair inhibition (2, 5). Therefore, it is reasonable to assume that the amount of consumption is the crucial factor contributing to the harmful effects of tobacco smoke versus E-cigarette smoke.