Nicotine inhalation and metabolism triggers AOX-mediated superoxide generation with oxidative lung injury

Abstract

With the increasing use of vaping devices that deliver high levels of nicotine (NIC) to the lungs, sporadic lung injury has been observed. Commercial vaping solutions can contain high NIC concentrations of 150 mM or more. With high NIC levels, its metabolic products may induce toxicity. NIC is primarily metabolized to form NIC iminium (NICI) which is further metabolized by aldehyde oxidase (AOX) to cotinine. We determine that NICI in the presence of AOX is a potent trigger of superoxide generation. NICI stimulated superoxide generation from AOX with K_m = 2.7 μM and V_max = 794 nmol/min/mg measured by cytochrome-c reduction. EPR spin-trapping confirmed that NICI in the presence of AOX is a potent source of superoxide. AOX is expressed in the lungs and chronic e-cigarette exposure in mice greatly increased AOX expression. NICI or NIC stimulated superoxide production in the lungs of control mice with an even greater increase after chronic e-cigarette exposure. This superoxide production was quenched by AOX inhibition. Furthermore, e-cigarette-mediated NIC delivery triggered oxidative lung damage that was blocked by AOX inhibition. Thus, NIC metabolism triggers AOX-mediated superoxide generation that can cause lung injury. Therefore, high uncontrolled levels of NIC inhalation, as occur with e-cigarette use, can induce oxidative lung damage.

Keywords: EPR spin-trapping; aldehyde oxidase; electronic cigarettes; free radicals; superoxide.

Figure 1

Nicotine (NIC) metabolism forms NIC iminium (NICI) that in the presence of human aldehyde oxidase-1 (hAOX1) is a potent source of O₂^•−.A, in the major pathway of NIC metabolism, NIC is first oxidized by Cytochrome P450 2A6 (CYP2A6)-catalyzed 5′-oxidation to nicotine Δ^1′(5′)-iminium ion, followed by the hAOX1 catalyzed oxidation of nicotine Δ^1′(5′)-iminium ion (NICI) to cotinine with the concomitant production of superoxide (O₂^•−) and hydrogen peroxide (H₂O₂). B, SDS-PAGE of purified recombinant hAOX1: Lane 1, molecular weight (MW) markers; lane 2, purified hAOX1, showing a single major band at ∼148 kDa. C, measurement of O₂^•− generation by cytochrome-c (Cyt-c) reduction from recombinant hAOX1 and NICI in air-saturated 50 mM phosphate buffer, pH 7.8, containing 0.1 mM EDTA, 100 μM Cyt-c, 100 U/ml catalase, and 40 nM hAOX1. The reaction was initiated by adding NICI in varying concentrations as labeled for each time course curve. D, kinetic analysis of the O₂^•− generation rate for a given NICI concentration. Data were fit to the Michaelis-Menten equation by non-linear regression with K_m and V_max values determined as shown. E, effects of the hAOX1 inhibitor raloxifene (RLX) and of SOD1. Values shown are mean ± SD, n = 4. F, EPR spin-trapping of O₂^•− generation from hAOX1 and NICI performed with 25 nM hAOX1, 10 μM NICI, 100 U/ml catalase, and 10 mM DIPPMPO in chelexed phosphate-buffered saline with 0.1 mM DTPA, pH 7.4. No O₂^•−–derived signal was detected with 0 μM NICI, but with 10 μM NICI, a large signal was seen, best simulated as a combination of DIPPMPO-OOH (84%) and DIPPMPO-OH (16%) adducts (Sim), with EPR parameters: DIPPMPO-OOH a_P ∼ 49.83 G, a_N ∼ 13.14 G, a_Hβ ∼ 11.12 G, a_HƔ ∼ 0.90 G, and DIPPMPO-OH a_P ∼ 47.92 G, a_N ∼ 12.90 G, a_Hβ ∼ 13.64 G. G, time course of the increase in the EPR signal of the O₂^•−-derived DIPPMPO–OOH adduct, associated with the spectrum in F with the spectrum shown after the signal plateaued.

Figure 2

AOX1 is expressed in the lungs with a marked increase following EC exposure leading to enhanced NICI or NIC–derived O₂^•−generation. Lung sections and lung tissue homogenates from mice exposed for 16 weeks to air (Air) or EC aerosol generated from EC liquid containing 24 mg/ml NIC (ECN) were studied. A, immunoblotting (IB) of lung homogenates for AOX1 and for GAPDH loading control. B, quantitation of AOX1 expression from IB. C, immunofluorescence (IF) of lung sections incubated with primary antibody (Ab) against AOX1 followed by fluorescent-labeled secondary Ab (green). DAPI was used as a nuclear stain (blue). D, quantitative analysis of AOX1 from IF images. E, fluorescence detection of O₂^•− in lung sections incubated with dihydroethidium (DHE; 10 μM) for 30 min alone (E1) or together with: 10 μM NICI (E2); 10 μM NICI + 100 μM SOD mimetic (SODm) MnTBAP (E3); 10 μM NICI + 100 μM AOX inhibitor raloxifene (RLX) (E4); or 100 μM NIC (E5). Red fluorescence arises from the O₂^•−-mediated oxidation of DHE. Images in C and E were obtained by confocal microscopy with the white scale bar corresponding to 50 μm. Bar graphs in E show quantitative analysis of O₂^•− generation in arbitrary units (AU). Graphs show means ± SD of data from 3 mice. Statistical analysis was done by unpaired t test in panels B & D and two-way ANOVA followed by Tukey's multiple comparison test in E. For B, D & E, p values are marked for comparison of ECN to Air. In E, # significant from DHE for matched Air or ECN treatments. For comparison across Air and ECN groups 1 to 5, p values are: for Air- 2 versus 1 (<0.0001), 3 versus 1 (0.3700), 4 versus 1 (0.9865), 5 versus 1 (<0.0001): for ECN- 2 versus 1 (p < 0.0001), 3 versus 1 (<0.0006), 3 versus 1 (0.0006), 4 versus 1 (0.0023), 5 versus 1 (p < 0.0001); $ significant from DHE/NICI for matched Air or ECN treatment. For comparison across Air and ECN groups 1 to 5, p values are: for Air- 3 versus 2 (<0.0001), 4 versus 2 (<0.0001), 5 versus 2 (0.0936); for ECN- 3 versus 2 (<0.0001), 4 versus 2 (<0.0001), 5 versus 2 (0.0724); for ECN 3 versus 2 (<0.0001), 4 versus 2 (<0.001), 5 versus 2 (0.0724).

Figure 3

Protein carbonylation, tyrosine nitration, and DNA oxidative damage. Measurements were performed in lung homogenates and sections from mice exposed for 16 weeks to air (Air) or electronic cigarette (EC) aerosol generated from EC liquid containing 0 mg/ml NIC (ECV) or 24 mg/ml NIC (ECN). A, ELISA quantitation of protein carbonyls. B, ELISA for oxidized guanine species. C, immunofluorescence (IF) of lung sections incubated with primary antibody (Ab) to nitrotyrosine (NT) followed by corresponding fluorescent-labeled secondary Ab (red). DAPI (blue) was used as a nuclear stain. D, quantitation of the NT-derived IF in C, expressed as arbitrary units (AU). E, IF of lung sections incubated with primary Ab to 8-hydroxy-2′-deoxyguanosine (8-OHdG) followed by corresponding fluorescent-labeled secondary Ab (green). F, quantitative analysis of 8-OHdG-derived IF. Images in (C) & (E) were obtained by confocal microscopy with the white bar corresponding to 50 μm. A, B, D, and F show means ± SD of data from 5 mice. Statistical analysis was done by one-way ANOVA with Tukey's multiple comparison test. p values for intergroup comparisons are as shown. EC exposure with inhalation of aerosol containing NIC triggered much greater protein and DNA oxidative damage than matched exposure without NIC.

Figure 4

AOX inhibition prevents EC-induced oxidative lung injury. Measurements were performed in lung homogenates and sections from mice exposed for 14 days to either fresh air (Air) or EC aerosol generated from EC liquid containing 24 mg/ml NIC (ECN) with daily intraperitoneal injection of vehicle or raloxifene (RLX) (25 mg/kg) (ECN + RLX). A, ELISA quantitation of protein carbonyls. B, ELISA for malondialdehyde (MDA)-protein adducts. C, immunofluorescence (IF) of lung sections incubated with primary antibody (Ab) against nitrotyrosine (NT) followed by corresponding fluorescent-labeled secondary Ab (red). DAPI (blue) was used as a nuclear stain. D, quantitation of the NT-derived IF in (C), expressed as arbitrary units (AU). E, IF of lung sections incubated with primary Ab against 8-hydroxy-2′-deoxyguanosine (8-OHdG) followed by corresponding fluorescent-labeled secondary Ab (green). F, quantitative analysis of 8-OHdG-derived IF. Images in C & E were obtained by confocal microscopy with the white bar corresponding to 50 μm. Data in (A), (B), (D), and (F) are means ± SD from 5 mice. Statistical analysis was done as in Figure 3. p values for intergroup comparisons are as shown. AOX inhibition prevented EC-induced oxidative lung damage with a decrease in protein carbonylation, lipid peroxidation, tyrosine nitration and DNA oxidative damage.