Guggulsterone Targets Smokeless Tobacco Induced PI3K/Akt Pathway in Head and Neck Cancer Cells

Background Epidemiological association of head and neck cancer with smokeless tobacco (ST) emphasizes the need to unravel the molecular mechanisms implicated in cancer development, and identify pharmacologically safe agents for early intervention and prevention of disease recurrence. Guggulsterone (GS), a biosafe nutraceutical, inhibits the PI3K/Akt pathway that plays a critical role in HNSCC development. However, the potential of GS to suppress ST and nicotine (major component of ST) induced HNSCC remains unexplored. We hypothesized GS can abrogate the effects of ST and nicotine on apoptosis in HNSCC cells, in part by activation of PI3K/Akt pathway and its downstream targets, Bax and Bad. Methods and Results Our results showed ST and nicotine treatment resulted in activation of PI3K, PDK1, Akt, and its downstream proteins - Raf, GSK3β and pS6 while GS induced a time dependent decrease in activation of PI3K/Akt pathway. ST and nicotine treatment also resulted in induction of Bad and Bax phosphorylation, increased the association of Bad with 14-3-3ζresulting in its sequestration in the cytoplasm of head and neck cancer cells, thus blocking its pro-apoptotic function. Notably, GS pre-treatment inhibited ST/nicotine induced activation of PI3K/Akt pathway, and inhibited the Akt mediated phosphorylation of Bax and Bad. Conclusions In conclusion, GS treatment not only inhibited proliferation, but also induced apoptosis by abrogating the effects of ST / nicotine on PI3K/Akt pathway in head and neck cancer cells. These findings provide a rationale for designing future studies to evaluate the chemopreventive potential of GS in ST / nicotine associated head and neck cancer.


Introduction
Head and neck squamous cell carcinoma (HNSCC) remains a significant cause of morbidity and mortality worldwide with five year survival rates of about 50%, marred by frequent recurrence or formation of second primary tumors (10-25% of cases) [1,2]. On a global scale, the use of tobacco products is the major risk factor, with smokeless tobacco (ST) consumption being linked to the high incidence of HNSCC [3][4][5]. ST is used in multiple forms namely naswar, gutkha, khaini (a mixture of ST with lime) and has or with betel quid, been classified as a human carcinogen by International Agency of Research in Cancer (IARC) [6,7]. The association of smoking with HNSCC is well known, but the link between ST use and head and neck cancer is emerging. In a recent report, Stepanov et al. [8] identified 23 polycyclic aromatic hydrocarbons (PAH) in ST in addition to nitrosamines and nicotine as reported in earlier studies [7]. Nicotine enhances proliferation, accelerates tumor growth and inhibits apoptosis in certain types of human cancer cell lines and induces angiogenesis in vivo by sustained activation of the mitogenic pathways [9][10][11]. Nicotine has been reported to inhibit apoptosis induced by opioids and genotoxic stress induced by etoposide, cisplatin, or UV irradiation in lung cancer cells [12,13]. In addition, nicotine activates PI3K/Akt pathway and inhibits the pro-apoptotic functions of Bax and Bad through phosphorylation [14,15]. These findings prompted us to investigate whether ST (khaini) induces PI3K/Akt pathway activation in head and neck cancer cells.
In this study, we investigated the activation of PI3K/Akt pathway in head and neck cancer cells on treatment with ST/ nicotine and its inactivation by GS. Further we challenged activation of Akt and its downstream targets (pS6, GSK3b and Raf), by ST/nicotine in head and neck cancer cells (SCC4) by pretreatment with GS.

Effect of ST and nicotine activate the PI3K/Akt pathway in head and neck cancer cells
The dose and time kinetics of ST and nicotine (addictive component of tobacco) treatment in head and neck cancer cells (SCC4 and HSC2), were determined by MTT assay ( Figure 1A and 1B). Similar kinetics was observed in both cell lines and the results for SCC4 cells are shown in Figure 1. Exposure to ST and nicotine increased cell proliferation in both the head and neck cancer cell lines in a dose dependant manner with an optimal dose of 20 mg/ml for ST and 10 mM for nicotine. These optimal doses have been used in all the subsequent experiments. To investigate the effect of ST or nicotine on the Akt pathway in HNSCC cells, both SCC4 and HSC2 cells were treated with 20 mg/ml of ST or with nicotine (10 mM) for different time intervals or kept untreated, and Akt activation (Akt phosphorylation at Ser-473 and Ser-309) was determined in the protein extracts by western blotting. Both ST and nicotine rapidly increased Akt phosphorylation without affecting the total Akt levels in both HSC2 and SCC4 cells ( Figure 2A and 2B respectively). These studies demonstrate that ST and nicotine activate Akt at doses that could be physiologically relevant. Further, our western blot analysis showed both ST and nicotine induced a time-dependent increase in phosphorylation of both the upstream kinases, PI3K and PDK1, as well as the downstream targets, Raf, GSK3b, and pS6 in SCC4 cells (Figure 2A and 2B), that reflected the rapid onset of increased Akt phosphorylation. Similar findings were observed in both the cell lines (SCC4 and HSC2) therefore, data for only SCC4 cells have been shown here.
Effect of guggulsterone on ST and nicotine induced activation of PI3K/Akt pathway Both ST and nicotine induced stimulation of Akt pathway, therefore, we investigated the effect of GS on ST and nicotine induced activation of Akt pathway. SCC4 cells were treated with GS (50 mM) for different time intervals and their protein extracts were tested for Akt activation. As shown in Figure 3A, GS inhibited the activation of Akt, upstream kinases PI3K, PDK1, as well as the downstream proteins, Raf, GSK3b, and pS6. The dose standardization of PI3K/Akt inhibitor, LY294002, was carried out by treating the SCC4 cells with different concentrations of the inhibitor (0.5-10 mM) for 1 h. As shown in the Figure 3B, LY294002 (10 mM) completely inhibited the expression of activated Akt; while no effect was observed on the total Akt expression levels. SCC4 cells were pre-treated with GS for 4 h or LY294002 (10 mM) for 60 min, followed by ST (20 mg/ml) for 6 h or nicotine (10 mM) for 4 h. The results indicate that GS as well as LY294002 blocked ST and nicotine-induced Akt pathway activation ( Figure 3C), though no effect was observed on the expression of total Akt and GSK3b at these time points.

Guggulsterone and PI3K-specific inhibitor LY294002 block ST and nicotine induced Bad and Bax phosphorylation
Both ST and nicotine potently stimulated serine phosphorylation of Bax and Bad in a time dependent manner ( Figure 4A and 4B) resulting in abrogation of their pro-apoptotic function. To demonstrate a functional role of Akt as the physiological ST and nicotine activated Bax and Bad kinase, LY294002, a PI3K specific inhibitor that can block the PI3K/Akt signaling pathway, was used. SCC4 cells were pre-treated with LY294002 (10 mM) for 60 min or 50 mM GS for 4 h followed by ST (20 mg/ml) for 6 h or with nicotine (10 mM) for 4 h. GS as well as LY294002 blocked ST and nicotine induced Bax and Bad phosphorylation ( Figure 4A and 4B). No effect on expression of total Bad and Bax was observed at these time points. These findings suggest that inhibition of ST and nicotine induced Bax and Bad phosphorylation by GS may restore the pro-apoptotic function of these molecules.

Akt is co-localized with Bax in cytoplasm
To assess a potential direct role for Akt as a physiological Bax kinase, sub-cellular distribution of Akt and Bax were assessed by immunofluorescence staining. Bax was primarily co-localized with Akt in the cytoplasm of SCC4 cells ( Figure 5A) suggesting that both ST and nicotine activated Akt has the potential to directly phosphorylate and inactivate Bax in SCC4 cells.

Phosphorylation of Bax and Bad results in retention of these proteins in cytosol and failure to target mitochondria
Our results demonstrated that ST and nicotine can induce Bax (Ser-184) and Bad (ser-136) phosphorylation, and phosphorylation at these sites resulted in inactivation of the pro-apoptotic function of Bax and Bad. It is well known that the pro-apoptotic activity of Bax and Bad is dependent on its cytosolic or mitochondrial localization. To test whether phosphorylation of Bax and Bad effects its sub-cellular localization, SCC4 cells were kept untreated or treated with GS (50 mM) for 4 h, ST (20 mg/ml) for 6 h and nicotine (10 mM) for 4 h. Sub-cellular fractionations were performed to isolate mitochondria and cytosol as described under material and methods. In untreated SCC4 cells, majority of Bax and Bad was found in the cytosol, and only a small amount of these proteins, was localized in the mitochondria ( Figure 5B). In GS treated cells majority of these proteins were observed in the mitochondrial fraction. In contrast, in both ST and nicotine treated cells, Bax and Bad were predominantly observed in the cytosol ( Figure 5B).
Notably, pre-treatment of SCC4 cells with GS (50 mM) for 4 h, followed by ST (20 mg/ml) treatment for 6 h or nicotine (10 mM) for 4 h, showed a shift in localization of Bax and Bad and majority of these proteins were observed in the mitochondrial fraction ( Figure 5B). To confirm the purity of the subcellular fractions obtained, succinate dehydrogenase, a protein exclusive to the mitochondria, was assessed by western blotting. Succinate dehydrogenase, was detected only in the mitochondrial fraction and not in the cytosol ( Figure 5B), confirming the purity of the , for the indicated time intervals and whole-cell extracts were prepared. Whole-cell extracts (60 mg protein) were resolved on 10% SDS-PAGE, electrotransferred to a PVDF membrane and non-specific binding was blocked with 5% non-fat milk overnight. Protein expression was determined by probing with phospho-specific antibodies for pAkt (thr-309), pAkt (ser-473). pGSKb3, pRaf, pPDK1 and Akt using enhanced chemiluminescence method. Western blotting for a-tubulin was done to show equal protein loading. doi:10.1371/journal.pone.0014728.g002 mitochondrial and cytosolic fractions. b-actin was used as the control to ensure equal protein loading in all the lanes.

ST and nicotine induced Bad phosphorylation enhances interaction with 14-3-3f
Bad phosphorylation (ser-136) promotes its translocation from mitochondria into cytosol and interaction with the scaffold protein 14-3-3. To assess whether ST or nicotine affect Bad association, SCC4 cells were treated with ST (20 mg/ml) or nicotine (10 mM) for different time intervals. Co-immunoprecipitation of 14-3-3f and Bad showed a time dependent increase in bound pBad in the immunocomplexes (IP) of 14-3-3f indicating association between 14-3-3f and pBad ( Figure 6A and 6B), on treatment with ST and nicotine respectively. Similar results were observed in reverse immunoprecipitation assays, western blotting was done for 14-3-3f in immunocomplexes obtained using pBad specific antibody, confirming the interaction of 14-3-3f with pBad ( Figure 6A and 6B). These results suggest ST and nicotine induced Bad phosphorylation resulted in sequestering Bad in the cytoplasm, functionally blocking its pro-apoptotic function.

Discussion
Tobacco in various forms is one of the main etiological factors for development and progression of HNSCC. Aberrant expression of genes involved in cell growth, survival, angiogenesis, invasion and metastasis in ST associated head and neck carcinogenesis has been reported by our group and others [27][28][29][30][31]. In this study, we demonstrated activation of Akt in HNSCC cells treated with ST/ nicotine, evidenced by sustained phosphorylation of Akt at serine 473 and threonine 308, as well as its downstream substrates (pS6, GSK3b, pRaf), in a time-dependent manner. Both ST and nicotine also induced the phosphorylation of the upstream kinase PDK1 (Ser241), and p85 subunit of another upstream kinase, PI3K in SCC4 cells. These observations are supported by the earlier reports showing activation of PI3K/Akt in cultured cortical neurons, normal human bronchial and small airway epithelial cells in response to nicotine treatment [11,[32][33][34]. Further, we showed both ST and nicotine activated Akt, induced Bad (Ser136) and Bax phosphorylation (Ser184) resulting in cytoplasmic retention of these proteins. However, treatment with LY294002, a PI3K/Akt inhibitor, potently inhibited ST and nicotine induced phosphor- ylation of Bad (Ser-136) and Bax (Ser-184) resulting in mitochondrial relocalization of these proteins, further strengthening our observation that ST and nicotine induced phosphorylation of Bad (Ser-136) and Bax (Ser-184) by activation of PI3K/Akt. Interestingly, both ST and nicotine induced phosphorylation of Bad (Ser-136) in SCC4 cells increased its association with 14-3-3f as revealed by co-IP assays. This results in sequestration of Bad in the cytoplasm, thus inhibiting the intrinsic pathway of apoptosis. Earlier reports have shown phosphorylation on either Ser112 or Ser136 facilitates formation of a complex between Bad and 14-3-3s in the cytosol, blocking their interaction with Bcl2/Bcl-xl in the mitochondria [35,36].
Recently, we demonstrated overexpression of PI Synthase, PI3-K and cyclin D1 in ST treated cells from oral lesions and oral cancer [37]. Further, analyzing clinical specimens from oral leukoplakic lesions without dysplasia, with dysplasia and oral squamous cell carcinomas (OSCCs), we provided the first evidence of increased expression of PI Synthase in early stages of oral precancer and cancer and its correlation with tumor dedifferentiation and tobacco consumption [37]. Here in we extended these findings to demonstrate GS could inhibit the induction of PI3K/ Akt by ST and nicotine.
Thus, inhibiting the induction of Akt activity by ST tobacco components might be a valuable approach to mitigate the effects of tobacco in consumers at risk for the development of HNSCC, and/or in HNSCC patients who continue to smoke or use ST products. In this respect, we observed that GS suppressed the phosphorylation of Akt (Ser473 and Thr308), PDK1 (Ser241) and p85 subunit of PI3K, leading to decreased phosphorylation of GSK3b, pRaf, pS6, the downstream targets of Akt. Notably, our study revealed that GS not only inhibited the constitutive PI3K/ Akt pathway, but its pre-treatment abrogated both ST and nicotine induced activation of PI3K/Akt pathway. ST and nicotine abrogated the pro-apoptotic effects of Bax/Bad by phosphorylation and sequestering in cytoplasm, however GS pre-treatment inhibited the effect of ST and nicotine, targeting Bax and Bad to the mitochondria, inducing apoptosis. However, the effect of GS is not specific and limited to head and neck cancer cells only. We and others have shown the anti-cancer effects of GS in a variety of cancer cell types namely, head and neck; leukemia, multiple myeloma, lung, melanoma, ovarian, gut derived adenocarcinoma, colorectal, colon, skin, prostate, esophagus, and breast [19,[21][22][23][24][38][39][40][41][42][43].
In conclusion, our study demonstrated that both ST and nicotine promote survival of human head and neck cancer cells SCC4, by inactivating the pro-apoptotic functions of Bax and Bad via its phosphorylation and sequestering these proteins in cytoplasm. GS not only inhibits constitutively active PI3K/Akt pathway, but also inhibits ST and nicotine induced activation of this pathway and promotes apoptosis signals in these cells Sixty microgram proteins from whole-cell extracts were resolved on 10% SDS-PAGE, electrotransferred to a PVDF membrane followed by blocking with 5% non-fat milk overnight. Protein expression was determined using enhanced chemiluminescence method and probed by antibodies against pBad and pBax. The blots were stripped and reprobed for Bax and Bad proteins. doi:10.1371/journal.pone.0014728.g004 ( Figure 6c). Hence, GS may be explored as a chemopreventive agent for ST induced head and neck carcinogenesis.

Preparation of Smokeless Tobacco Extract (ST)
Commonly used brand of cured tobacco (Khaini) was purchased from the local market. 25 g of tobacco was finely powdered and homogenized in 225 ml of distilled water. Mixture was stirred on a magnetic stirrer for two hours and then allowed to stand for 24 hrs at 37 uC. Thereafter, supernatant was collected after centrifugation at 5000 g for 20 minutes. The extract was sterilized by passing it through a 0.22 mm filter and stored at 4 uC till use. The final concentration of tobacco in aqueous extract was estimated to be 2.7 g% [48]. SCC4 cells were pre-treated with 50 mM GS for 4 h, followed by ST for 6 h or with nicotine for 4 h. Cytoplasmic (C) and Mitochondrial (M) extracts were prepared as described in materials and methods and separated on 10% SDS-PAGE. Proteins were then electro-transferred on PVDF membrane followed by blocking with 5% non-fat milk overnight. Blots were incubated with specific antibodies against Bax and Bad. Protein expression was determined using enhanced chemiluminescence method. Purity of the subcellular fractions obtained was determined using western blot for mitochondrial protein, succinate dehydrogenase. b-actin was used as a loading control. doi:10.1371/journal.pone.0014728.g005

Cytotoxicity Assay
Head and neck cancer cells (5610 3 /well) were plated in a 96well plate for 24 hrs. Cells were incubated in triplicates in the presence of medium containing ST/nicotine or 0.02% of DMSO which served as a vehicle control in a final volume of 100 ml for 24-96 hrs at 37 uC. Cell death was measured by adding 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) at 37 uC for 3-4 h. The formazan crystals were dissolved in 100 ml of dimethylsulphoxide (DMSO) and optical density (OD) was measured at wavelength of 570 nm. The percentage cell death was calculated individually for each dose as follows: (OD control 2OD treated /OD control )6100, as described earlier [49].

Western blotting and Co-immunoprecipitation (Co-IP) assay
Co-immunoprecipitation (Co-IP) assays and WB were carried out as described by us previously [39]. Whole-cell lysates were prepared from GS (50 mM) or nicotine (10 mM) treated SCC4 cells and protein concentration was determined using the Bradford reagent (Sigma) and equal amounts of proteins (50 mg/lane) were resolved on 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel. The proteins were then electro-transferred onto polyvinylidenedifluoride (PVDF) membrane. After blocking with 5% non-fat milk in Tris-buffered saline (TBS, 0.1 M, pH = 7.4), blots were incubated with specific antibodies as per manufacturer's recommended protocol at 4 uC overnight. Protein abundance of atubulin (Santa Cruz Biotechnology, CA) served as a control for protein loading in each lane. Membranes were incubated with HRP-conjugated secondary antibodies, (DAKO Cytomation, Glostrup, Denmark), diluted at an appropriate dilution in 1% BSA, for 2 h at room temperature. After each step, blots were washed three times with Tween (0.1%)-Tris-buffer saline (TTBS). Protein bands were detected by the enhanced chemiluminescence method (ECL, Santa Cruz Biotechnology, CA) on XO-MAT film. Co-IP assays were carried out as described earlier. assays were carried out using whole cell lysates and analyzed by western blotting as described in Materials and Methods. 14-3-3f was immunoprecipitated using specific antibody and the bound pBad protein coprecipitated with 14-3-3f was determined by western blot analysis. Reverse immunoprecipitation assays were carried out in which pBad was immunoprecipitated, followed by western blotting analysis of 14-3-3f using a specific antibody against this protein. (C) Figure shows the hypothetical model for inhibition of ST / nicotine induced activation of PI3K/Akt pathway by GS. Our results showed treatment with ST and /or nicotine activates Akt (phosphorylated at Ser-473 and Ser-309) resulting in phosphorylation of its downstream targets Raf, GSK3b, and pS6, whereas GS treatment inhibits activation of Akt pathway. Interestingly, pre-treatment of head and neck cancer cells with GS inhibits activation of Akt and its downstream targets (Raf, GSK3b, and pS6) on exposure to ST / nicotine. GS pre-treatment also releases Bad from inhibitory action of 14-3-3f, thereby activating intrinsic pathway of apoptosis. Thus, our results demonstrated GS as a potential therapeutic agent for ST-induced head and neck carcinogenesis. doi:10.1371/journal.pone.0014728.g006 Sub-cellular fractionation: Isolation of cytoplasm and mitochondria Head and neck cancer cells (2610 7 ) were treated, harvested and washed once with cold 1X PBS and resuspended in isotonic mitochondrial buffer (210 mM mannitol, 70 mM sucrose, 1 mM EGTA, 10 mM Hepes, pH 7.5, 0.1% BSA) containing protease inhibitor cocktail. The resuspended cells were homogenized with a polytron homogenizer operating for four bursts of 10 s each at a setting of 5 and then centrifuged at 2000 g for 3 min to pellet the nuclei and unbroken cells. The supernatant was centrifuged at 13,000 g for 10 min to pellet mitochondria as described [50]. The supernatant was further centrifuged at 15,000 g to pellet light membranes. The resulting supernatant contained the cytosolic fractions. The mitochondria was washed with mitochondrial buffer twice, resuspended with 1% Nonidet P-40 lysis buffer, rocked for 60 min, and then centrifuged at 13,000 g for 10 min at 4 uC. The supernatant containing mitochondrial proteins was collected. Protein (100 mg) from each fraction was subjected to 12% SDS-PAGE and analyzed by western blot analysis using anticytochrome c antibody. The purity of the fractions was confirmed by assessing localization of fraction-specific proteins including succinate dehydrogenase for mitochondria.

Co-localization studies of Akt and Bax in oral cancer cells using confocal laser scan microscopy
Head and neck cancer cells, SCC4 and HSC2, were grown on coverslips in DMEM medium supplemented with 10% FBS at 37 uC and processed for confocal laser scan microscopy as described by us previously [51]. Cells were rinsed in Dulbecco's PBS (DPBS), fixed in methanol for 5 min at 220 uC and incubated with specific primary antibodies, mouse monoclonal anti-Bax and rabbit monoclonal anti-Akt antibody, incubated as a cocktail at 4 uC overnight. After rinsing in phosphate buffer saline-0.1% Tween (PBST, 1X) the coverslips were incubated with antimouse FITC conjugated/anti-rabbit Alexa flourH594 conjugated secondary antibody for 45 min at 37 uC in the dark. Coverslips were washed and counterstained with DAPI for 30 sec. A mouse antibody against human Bax, a rabbit antibody against human Akt, and fluorescein isothiocyanate-conjugated anti-mouse (green) or rhodamine-conjugated anti-rabbit (red) secondary antibodies were used so that cells could be stained simultaneously without cross-reaction. In negative controls, the primary antibodies were replaced by non-immune mouse IgG of the same isotype to ensure specificity (data not shown). Thereafter, the slides were rinsed and mounted in fluorescence mounting medium and examined with Lieca TCS SP2 confocal laser scanning microscopy (CLSM).