Rapid Akt activation by nicotine and a tobacco carcinogen modulates the phenotype of normal human airway epithelial cells (original) (raw)
Nicotine and NNK induce Akt phosphorylation. To investigate whether components of tobacco could activate Akt, we chose to test nicotine and NNK, a nicotine-derived, tobacco-specific nitrosamine with high carcinogenic potential (14), from over 3,500 compounds in the particulate fraction of tobacco smoke. When added to NHBEs or SAECs in vitro, nicotine increased Akt phosphorylation at S473 and T308 in a time- and dose-dependent manner (Figure 1, a and b). Although stimulation of Akt phosphorylation by nicotine in each cell type was evident within 5 minutes, the patterns of S473 and T308 phosphorylation were different over time (Figure 1a, upper panels). Nicotine maximally increased S473 phosphorylation at 60 minutes in NHBEs and 30 minutes in SAECs, and increased S473 phosphorylation was maintained for 24 hours (see Figure 4a, inset). Phosphorylation of T308 by nicotine was more transient, reaching a maximum at 30 minutes in NHBEs and 15 minutes in SAECs, with decreases observed at subsequent time points. In experiments designed to test dose-dependent responses to nicotine, nicotine increased Akt phosphorylation in both cell types with doses as low as 10–100 nM, but maximum phosphorylation was observed at 1–10 μM (Figure 1b). These concentrations are achievable in smokers, since average steady-state serum concentrations of nicotine have been reported at 200 nM, and acute increases to 10–100 μM in serum or to 1 mM at the mucosal surface immediately after smoking have been reported (15–17).
Phosphorylation of Akt by nicotine or NNK in NHBEs and SAECs. (a) Nicotine increased Akt phosphorylation in a time-dependent manner in NHBEs (left panels) and SAECs (right panels), as assessed by immunoblotting with anti–phospho-S473, anti–phospho-T308, and anti-Akt antibodies. (b) Nicotine-induced Akt phosphorylation was also dose-dependent. (c) NNK increased Akt phosphorylation in a time-dependent manner. (d) NNK-induced Akt phosphorylation was also dose-dependent.
Nicotine-mediated Akt activation and survival of NHBEs. (a) Topoisomerase II inhibition. Nicotine (10 μM) protected against etoposide-induced apoptosis, as assessed by flow cytometry. Pretreatment with LY294002 decreased nicotine-mediated survival. Parallel samples were harvested for immunoblotting (inset; C, control; N, nicotine; LY, LY294002; LY/N, LY294002 + nicotine). (b) UV irradiation. Nicotine (10 μM) protected against UV irradiation–induced apoptosis, as measured using CellDeath ELISA kits. Pretreatment with LY294002 or DHβE attenuated nicotine-mediated survival. (c and d) H2O2 treatment. NHBEs were pretreated with nicotine (10 μM) (c) or NNK (d) as above, with or without H2O2 (200 μM). After 4 hours, cells were harvested, dead cells that exhibited cytoplasmic inclusion of 0.4% trypan blue were counted, and this number was compared with the total number of cells. At least 300 cells per sample were counted by a blinded observer.
The tobacco-specific carcinogen NNK stimulated Akt phosphorylation with different kinetics and was more potent (Figure 1, c and d). The time course of NNK-induced Akt phosphorylation was similar for NHBEs and SAECs and for both sites of phosphorylation (Figure 1c). For the first 10–15 minutes, we observed decreased S473 and T308 phosphorylation after NNK administration, followed by increased S473 and T308 phosphorylation that was maximal at 60 minutes. Unlike nicotine induction of S473 phosphorylation, which was maintained for at least 24 hours, NNK-induced S473 phosphorylation was maintained for 8 hours (data not shown). NNK dose–dependence was also similar for NHBEs and SAECs. Increased phosphorylation of both sites was observed with doses as low as 1 nM (Figure 1d), with maximal induction at 100 nM for NHBEs and 10 nM for SAECs. Although these data showed that nicotine or NNK could induce Akt activation in NHBEs and SAECs, the differences in dose- and time-dependence of nicotine- and NNK-induced Akt phosphorylation raised the possibility that that these compounds might use different mechanisms to activate Akt.
Effects of nicotine or NNK on Akt kinase activity and phosphorylation of downstream substrates. To confirm that increased Akt phosphorylation at S473 and T308 was indicative of increased kinase activity, we measured the phosphorylation of an exogenous peptide substrate, glycogen synthase kinase-3α/β (GSK-3α/β), after nicotine or NNK administration (Figure 2a). Nicotine or NNK increased Akt kinase activity in a PI3K-dependent manner, since the PI3K inhibitor LY294002 (18) inhibited induction of Akt activity by nicotine or NNK. To demonstrate that Akt activation propagated signaling cascades within NHBEs, and to identify substrates or cellular processes that might be affected by nicotine- or NNK-mediated Akt activation, we assessed phosphorylation of five proteins previously identified to be downstream of Akt at the time of maximal nicotinic Akt induction: a member of the forkhead transcription factor family (FKHR), GSK-3α and -3β, a ribosomal kinase (p70S6K), and a binding protein for eukaryotic translation initiation factor 4E (4EBP–B1) (Figure 2b). We observed different patterns of phosphorylation induced by nicotine or NNK. Nicotine increased the phosphorylation of all downstream substrates, but phosphorylation of GSK-3α and FKHR was induced most. NNK increased phosphorylation of all downstream substrates except FKHR and had the greatest effects on phosphorylation of GSK-3β and p70S6K. The different induction of phosphorylation of downstream substrates further suggested that the kinetics and/or the mechanism of activation of Akt for nicotine and NNK are different. Together with other studies that investigated the effects of Akt activation on these substrates (19–21), our results support the hypothesis that activation of Akt by nicotine or NNK promotes cell cycle progression, increases protein synthesis, and inhibits apoptosis.
Akt kinase activity and effect on downstream substrates. (a) We measured Akt kinase activity in NHBEs by immunoprecipitating active Akt and assessing phosphorylation of an exogenous peptide, GSK-3α/β, after administration of nicotine (left panels) or NNK (right panels). LY294002, DHβE (an α3/α4 nAchR antagonist), or α-BTX (an α7 nAchR antagonist) inhibited nicotinic induction of Akt kinase activity. (b) Phosphorylation of substrates downstream of Akt in NHBEs was increased after administration of nicotine (Nic; middle lane) or NNK (right lane), compared with that in untreated cells (Con), as assessed by immunoblotting with the indicated phosphospecific antibodies.
Expression of nAchR subunits in NHBEs and SAECs. Because nAchRs bind nicotine and NNK and mediate the biologic effects of these tobacco components, we characterized expression of nAchR subunits in NHBEs and SAECs. nAchRs belong to the superfamily of ligand-gated ion channels that are predominantly expressed in neural tissue, but they have recently been reported to be expressed in other tissues (22, 23). Functional nAchRs are composed of homopentamers derived from subunits α7–α10 or heteropentamers derived from six α subunits (α1–α6) and three β subunits (β2–β4). nAchRs containing α3 or α4 are most abundant in neural tissue (24), and α7-containing nAchRs have been described in human bronchial epithelial and endothelial cells (6). We performed nAchR subunit–specific RT-PCR analysis of subunits α1–α10 and β2–β4 in SAECs, NHBEs, and an NSCLC cell line (H157). Cell type–specific nAchR subunit expression was observed and is summarized in Table 1. SAECs selectively express α2 and α4 subunits, and NHBEs selectively express α3 and α5 subunits. Both cell types express α7–α10, β2, and β4 subunits. These results imply that SAECs and NHBEs express similar homopentamer nAchRs containing α7–α10 subunits but likely express a different repertoire of heteropentamers: namely α3β4, α3β4α5, or α3β2 nAchRs in NHBEs, and α2β2, α2β4, α4β2, or α4β4 nAchRs in SAECs.
RT-PCR expression of nAchR subunits
Effects of nicotinic antagonists on Akt phosphorylation. To determine which nAchRs might facilitate nicotinic activation of Akt, we treated NHBEs and SAECs with pharmacologic inhibitors directed against specific α subunit–containing nAchRs and measured Akt activation after treatment with nicotine or NNK. (LY294002 served as a positive control in these studies and inhibited both basal and nicotine-induced S473 phosphorylation.) In NHBEs and SAECs (Figure 3a, upper left and lower panels), the α7 antagonist α-BTX (25), the α7 antagonist MLA, or the nonspecific nAchR antagonist MCA did not attenuate nicotine-induced Akt phosphorylation, either at lower doses where specificity is greatest (data not shown) or at the higher doses shown. In contrast, the α3/α4 antagonist DHβE (26, 27) inhibited nicotine-induced Akt phosphorylation (Figure 3a) and activity (Figure 2a, left panels), suggesting that nAchRs containing α3 or α4 subunits (in NHBEs or SAECs, respectively) mediate Akt induction by nicotine. Increased Akt phosphorylation was occasionally observed with addition of nicotinic antagonists alone, and this may be related to the observation that nicotinic antagonists such DHβE and MLA can cause a compensatory increase in expression and activation of other, nontargeted nAchRs (28). The role of α3-containing nAchRs in nicotine-induced Akt phosphorylation in NHBEs was further supported by the observation that a specific α3 nicotinic agonist, α-ATX (29), increased Akt phosphorylation in NHBEs, and that this effect was inhibited by DHβE (Figure 3a, upper right panels).
Effect of nAchR antagonists on nicotinic activation of Akt in NHBEs and SAECs. (a) Nicotine. Only LY294002 or the α3/α4 antagonist DHβE inhibited nicotine-induced Akt phosphorylation in NHBEs (upper left panels) and SAECs (lower panels). To confirm the role of α3 nAchRs in activating Akt in NHBEs, α-ATX (an α3 agonist) was added to NHBEs with or without DHβE (upper right panels). (b) NNK. In contrast to nicotine-mediated Akt phosphorylation, NNK-induced phosphorylation in NHBEs was inhibited by LY294002, the α7 antagonists α-BTX and MLA, and the nonspecific inhibitor MCA. DHβE was ineffective.
In contrast to nicotine-mediated activation of Akt, NNK-induced Akt phosphorylation was inhibited by the α7-specific antagonists α-BTX and MLA, as well as by MCA. DHβE was ineffective, suggesting that NNK activates Akt through α7-containing nAchRs but not through α3- or α4-containing nAchRs (Figure 3b). Further support for the role of α7 nAchRs in mediating NNK-induced Akt activation is suggested by the dose and kinetic similarities of NNK-induced Akt phosphorylation in NHBEs and SAECs, the fact that α7 is a shared functional α subunit between NHBEs and SAECs, and the fact that α7 antagonists inhibited NNK-induced Akt phosphorylation and kinase activity (Figure 2a, right panels) at low doses (100 nM) that are specific for α7 inhibition. Thus, nicotine and NNK likely use separate nAchRs to activate Akt.
Nicotinic activation of Akt increases epithelial cell survival. Because tobacco carcinogens promote tumor formation through damage to DNA in normal cells by forming DNA adducts and causing oxidative damage, we tested whether nicotinic activation of Akt would protect airway epithelial cells against modalities that cause DNA damage and induce apoptosis. When NHBEs were exposed to a topoisomerase II inhibitor, etoposide, apoptosis increased from 2% to 44%, as measured by the formation of subgenomic DNA detected by flow cytometry (Figure 4a). In the presence of nicotine, etoposide-induced apoptosis was decreased by 61%. LY294002 attenuated the protective effects of nicotine, and the effects of nicotine and/or LY294002 on apoptosis correlated well with changes in Akt phosphorylation detected at 24 hours (Figure 4a, inset). NNK also attenuated etoposide-induced apoptosis, but protection was only evident at time points at which Akt activation was maintained (up to 8 hours; data not shown). To test UV radiation (Figure 4b), we used an apoptosis assay that measures histone release from apoptotic cells and is more sensitive than measures of subgenomic DNA. UV radiation alone was a poor apoptotic stimulus, as was serum starvation alone. When serum-deprived NHBEs were exposed to UV radiation, however, apoptosis increased from 1.2- to 2.4-fold. In the presence of nicotine, UV irradiation–induced apoptosis was decreased by 58%. When NHBEs were pretreated with LY294002 or DHβE, the protection conferred by nicotine was attenuated. We observed similar protective effects with administration of NNK in parallel experiments performed with α-BTX substituted for DHβE (data not shown). When NHBEs were exposed to H2O2, cell death approximately doubled, as assessed by trypan blue exclusion, but nicotine (Figure 4c) or NNK (Figure 4d) inhibited the cytotoxic effects of H2O2. Together, these data suggest that nicotinic activation of Akt promotes the survival of NHBEs that are exposed to conditions known to cause DNA damage and stimulate apoptosis.
Nicotinic activation of Akt alters epithelial cell growth characteristics. In addition to reducing apoptosis, Akt activation can contribute to cellular transformation by increasing cell growth and decreasing dependence on exogenous growth factors or attachment to ECM. We tested whether nicotinic activation of Akt would alter these cellular parameters. When NHBEs were incubated with nicotine, either as a single dose of 10 μM given on day 1 or as daily physiologic doses of 100 nM or 1 μM, differences in cell growth were apparent after approximately 7 days (Figure 5a). Untreated NHBEs became contact-inhibited, but nicotine-treated cells continued to grow until they crowded the tissue culture dishes and began to detach. These data showing dysregulated NHBE growth after nicotine administration are consistent with in vivo observations that bronchial tissues of active smokers have increased proliferative indices when compared with those of former smokers (30). With prolonged serum deprivation (Figure 5b), NHBEs underwent apoptosis that was attenuated by administration of nicotine. Likewise, nicotine (Figure 5c) or NNK (Figure 5d) inhibited death of NHBEs caused by inhibition of cellular attachment (anoikis; ref. 31). Protection conferred by nicotine was attenuated by LY294002 or by DHβE, and protection conferred by NNK was attenuated by LY294002 or by α-BTX. These studies show that, in addition to promoting cellular survival, nicotinic activation of Akt diminishes contact inhibition and cellular dependence on exogenous growth factors or ECM.
Nicotine alters NHBE phenotype. (a) Loss of contact inhibition. NHBEs were incubated with different concentrations of nicotine (filled symbols; asterisks indicate daily dosing) or complete media alone (open squares). Cell number was measured by absorbance at 540 nm using a 96-well microplate reader. (b) Serum starvation. NHBEs were grown in DMEM with 0.1% BSA or control media for 9 days with or without nicotine (10 μM) given once on day 1. Apoptosis was assessed using CellDeath ELISA kits. (c) Anoikis. Nicotine (left panel) or NNK (right panel) decreased anoikis, as measured using CellDeath ELISA kits. Pretreatment with LY294002 decreased protection conferred by either nicotine or NNK. Pretreatment with DHβE attenuated nicotine-mediated survival (left panel), and pretreatment with α-BTX attenuated NNK-mediated survival.
Akt phosphorylation in NNK-treated A/J mice and in human lung cancer. To determine whether nicotinic activation of Akt was observable in vivo, we assessed lung tissue derived from the A/J strain of mice, which is prone to develop lung tumors but which, with administration of NNK, will develop them at earlier time points with more aggressive histologies (32). We used phosphospecific antibodies against S473 in immunohistochemical experiments on formalin-fixed, paraffin-embedded tissues. A/J mice treated with PBS did not exhibit phosphorylated Akt (Figure 6a). In contrast, A/J mice treated with NNK showed phosphorylated Akt in airway epithelial cells, as well as in NNK-induced tumors (Figure 6b). When protein extracts were prepared from lung tissue derived from PBS-treated or NNK-treated mice and immunoblotting was performed for phosphorylated and total Akt levels, NNK-treated mice showed a 5.5-fold increase in the ratio of phosphorylated Akt to total Akt compared with PBS-treated mice (Figure 6c). These data show that exposure to NNK induces Akt activity in vivo.
Detection of Akt phosphorylation in vivo. (a) Lung tissue, including epithelial lining of airway lumen (Lu), from A/J mice given PBS orally was harvested and processed for immunohistochemistry with phosphospecific S473 antibodies as described. No staining is detectable. (b) Lung tumor (Tu) and epithelial lining of airway lumen (Lu) from A/J mice given NNK orally exhibit staining with phosphospecific S473 antibodies. (c) Quantification of phosphorylated Akt/total Akt in protein extracts derived from lung tissue from PBS- or NNK-injected mice. Quantification of immunoblots was performed using NIH Image software. (d) Phosphorylated Akt in a human lung adenocarcinoma derived from a smoker with a 48-pack-per-year smoking history.
We extended these studies by evaluating Akt phosphorylation in human lung cancer specimens derived from smokers. Akt phosphorylation was detected in ten of ten specimens. Staining varied from low to high levels in tumor tissues, but virtually none was detected in surrounding stroma (see representative specimen, Figure 6d). Because we were unable to obtain lung tissues from nonsmokers, we were unable to compare the levels of phosphorylated Akt in smokers versus nonsmokers. Staining with the phosphospecific S473 antibodies was completely abrogated by the inclusion of competing phosphopeptide (data not shown), indicating specificity of the phosphospecific Akt antibodies. These data are consistent with the detection of active Akt in lung tissues from NNK-treated A/J mice and show that active Akt can be detected in lung cancers derived from smokers.