AKT activation has been reported in several types of cancer, including HNSCC (11
). To evaluate whether AKT activation correlates with tobacco exposure in HNSCC, we examined smoking histories of HNSCC patients and the status of AKT activation. A total of 35 pairs of HNSCC samples and case-matched adjacent mucosa samples were analyzed. Seven normal oropharyngeal samples from sleep apnea patients were used as normal controls. Among 35 pairs of tissue samples examined, 23 pairs were from HNSCC patients who are smokers and 12 pairs are from HNSCC patients who are non-smokers as defined in Materials and methods. Immunohistochemistry (IHC) using a phosphorylation-specific antibody revealed that AKT activation was detected more frequently in both the adjacent mucosa (61 vs. 17%, p<0.05) and HNSCC arising from smokers (91 vs. 58%, p<0.05) than those from HNSCC patients who are non-smokers (). None of the normal controls exhibited positive staining for activated AKT (). We cannot exclude the possibility that the lower percentage of AKT activation detected in the adjacent mucosa and HNSCC tumors of non-smokers is due to second-hand smoking or exposure to other environmental carcinogens. Baseline activation in HNSCC tumors is not surprising, since AKT activation is common in multiple cancer types (12
Figure 1 AKT activation occurs more frequently in the adjacent mucosa and HNSCC tumors from smokers than non-smokers. (A) Representative IHC sections of adjacent mucosa obtained from a smoker (left panel) and non-smoker (right panel) stained for active AKT (phosphorylated (more ...)
In order to explore whether the correlation between tobacco smoking and AKT activation represent a causal role of tobacco carcinogens in AKT activation, we exposed human normal oral epithelial cells (OKF6), normal esophageal squamous epithelial cells (EPC2) and HNSCC cell lines (SCC9, SCC4, and SCC25) to NNK. As shown in , NNK rapidly induced phosphorylation of AKT at both Ser473 and Thr308 in a dose- and time-dependent manner in all cell lines. AKT phosphorylation at both sites was evident as early as 15 min after 10−8
M NNK exposure, reached to a maximum level between the 30- and 60-min time-points (, left panel), and the increased phosphorylation status was maintained for 24 h (). To determine dose-dependent response to NNK, AKT was activated in each cell type with NNK doses as low as 10−8
M, but maximum phosphorylation was observed at 10−4
M (, right panel). These concentrations are physiologically relevant, as average steady-state serum concentration of NNK in active smokers has been reported at 2×10−7
M, and acute increases to 10−4
M in serum or to 10−3
M at the mucosal surface immediately after smoking have also been reported (13
Figure 2 AKT activation upon NNK exposure in normal head and neck epithelial cells and an HNSCC cell line. Left panels, time-dependent induction of AKT phosphorylation by NNK. Cells were plated and cultured in 6-well plates. The next day, the medium was changed (more ...)
Figure 3 Effects of AKT activation on downstream targets, cell proliferation, and apoptosis. (A) Effects of AKT activity on downstream targets. OFK6 cells were exposed to 10−8 M NNK with and without addition of 10 μM LY294002 for 24 h. Cells were (more ...)
To determine whether AKT phosphorylation triggered downstream signaling within normal head and neck epithelial cells, we assessed the phosphorylation status of several known AKT downstream signaling mediators, Bad, Mdm2, GSK3α/β and mTOR, which play a role in regulating apoptosis and/or proliferation. Bad is inactivated through phosphorylation, which in turn frees the survival factor BCL-XL
to exert its anti-apoptotic effects (16
). Phosphorylation of Mdm2 results in increased p53 degradation, which blunts the apoptotic-inducing role of p53 (17
). In addition, AKT activation also phosphorylates and inactivates GSK3β, which can prevent its degradation of cyclin D1 and c-myc by GSK3β phosphorylation. Finally, the mTOR pathway is well known for protein translational control and cell growth, and is a direct downstream target of AKT by phosphorylation. We observed increased phosphorylation of each of these downstream targets following NNK exposure (). To assess the requirement of AKT activation for NNK-induced phosphorylation of these downstream mediators, we pre-treated cells with a PI3K/AKT inhibitor, LY294002. As shown in , LY294002 pre-treatment significantly blocked NNK-induced AKT phosphorylation at Ser473 without altering total AKT levels, thus LY294002 is able to block AKT activation. Furthermore, blocking AKT activity resulted in decreased phosphorylation of these key downstream mediators, indicating NNK-induced activation of these molecules depends on activated AKT ().
To determine whether NNK-induced AKT activation and subsequent downstream molecular changes are able to affect cell growth and survival, we analyzed cell proliferation using a colorimetric MTT assay. As shown in , when OKF6 cells were exposed to NNK, cell proliferation was significantly higher after 72 h of NNK exposure in comparison with non-NNK exposed OKF6 cells. Further, this effect was completely blocked by LY294002 treatment confirming the requirement for PI3K/AKT signaling. We then assessed the effect of NNK-induced AKT activation on cell apoptosis. OKF6 cells were treated using a topoisomerase II inhibitor, etoposide, to induce cell apoptosis with or without NNK exposure. As shown in , in cells that were not exposed to NNK, etoposide increased cell apoptosis from 5.7 to 33.4%, as measured by annexin V-FITC staining. In the presence of NNK, etoposide-induced apoptosis was significantly decreased to 24.2% (p<0.05). Furthermore, this NNK-induced attenuation of apoptosis was completely abolished by LY294002 treatment (36.8%).
To investigate whether NNK exposure activates AKT in vivo, C57BL/6 mice were orally treated with 100 μM NNK dissolved in sesame oil three times per week. Four weeks of NNK exposure produced a hyperplastic and thickened epithelium. AKT activation was detected in murine buccal mucosa (), tongue and esophagus (data not shown) following exposure to NNK but not sesame oil control. To determine whether epithelial hyperplasia was, at least in part, due to cell proliferation and/or increased survival, a bromodeoxyuridine (BrdU) labeling experiment was performed. NNK exposure resulted in 5.3-fold increase in BrdU labeling index in mouse head and neck tissues (17.5±2.4 nuclei/mm basement membrane, n=5, ) in comparison with oil-exposed control mice (3.3±1.2 nuclei/mm basement membrane, n=5, ). Additionally, terminal deoxynucleotidyltransferase-dUTP nick-end labeling (TUNEL) assay revealed a nearly complete blockade of apoptosis in head and neck tissues from mice with NNK exposure as compared with control tissues ().
Figure 4 Mice exposed to NNK exhibit AKT activation, increased cell proliferation, and epithelial hyperplasia with no head and neck tumor formation. (A) NNK exposure induced AKT activation. Buccal mucosa, were harvested from 10-week old C57B6 mice exposed orally (more ...)
Given what has been previously demonstrated with regard to the ‘genotoxic’ effects of tobacco carcinogens, our data showing NNK-induced AKT activation in head and neck epithelium and HNSCC provide a novel mechanism of tobacco-induced HNSCC carcinogenesis. The decreased apoptosis and increased proliferation of cell damaged by NNK will synergistically augment the ‘genotoxic’ effects of NNK, which would greatly enhance the potency of tobacco carcinogens. Thus, our data indicate that AKT activation may resent a key molecular pathway involved in tobacco-induced carcinogenesis.
AKT activation is observed in adjacent mucosa of smokers at nearly four-times more than in non-smokers, suggesting that AKT activation by tobacco carcinogen occurs early in the development of HNSCC. Thus, AKT activation upon tobacco exposure could be a potential target for chemopreventive approach of HNSCC. In fact, there are several recent reports that deguelin, a naturally occurring rotenoid, possesses chemo-preventative role on lung cancer through inhibiting PI3K-AKT signaling (18
). In addition, HNSCC patients have a 10% risk of developing a second primary malignancy within 5 years after primary surgical treatment (20
) and these patients require close follow-up for recurrence. Tobacco use is an important risk factor for the HNSCC recurrence (20
). Thus, AKT activation upon tobacco exposure could have a dramatic impact on the HNSCC recurrence. Therefore, further investigation of various inhibitors of PI3K/AKT pathway in both in vitro
and in animal models of HNSCC will provide valuable information for developing new strategy of chemoprevention and/or treatment for HNSCC.
In conclusion, our studies show that AKT activation is an early event in human HNSCC development, and is associated with smoking history. In addition, application of physiologically relevant concentrations of NNK to normal head and neck epithelial and HNSCC cells activates AKT, with subsequently increased cell proliferation and survival. Furthermore, NNK exposure in vivo results in AKT activation in mouse head and neck epithelia. Thus, our studies here represent the first report that AKT activation upon NNK exposure is one of the molecular mechanisms of tobacco-induced HNSCC carcinogenesis.