Akt/mTOR Targeting Activity of Resveratrol Derivatives in Non-Small Lung Cancer

The Akt-mTOR signal is important for the survival and proliferation of cancer cells and has become an interesting drug target. In this study, five resveratrol derivatives were evaluated for anticancer activity and Akt/mTOR targeting activity in non-small lung cancer cell lines. The effects of resveratrol derivatives on cell proliferation were assessed by 2,5-diphenyl-2H-tetrazolium bromide (MTT) assay, nucleus staining, and colony formation assay. Furthermore, the effect of resveratrol derivatives on proliferation-related protein expression was analyzed by immunofluorescence and Western blotting. For the structure–activity relationship (SAR), results reveal that two derivatives of resveratrol which are 4,4′-(ethane-1,2-diyl) bis(2-methoxyphenol) (RD2) and the 4-(3-hydroxy-4-methoxyphenethyl)-2-methoxyphenol (RD3) had very similar structures but exerted different cytotoxicity. The IC50 of RD2 and RD3 were 108.6 ± 10.82 and more than 200 µM in the A549 cell line and 103.5 ± 6.08 and more than 200 µM in H23 cells, respectively. RD2 inhibited cell proliferation and induced apoptosis when compared with the control, while RD3 caused minimal effects. Cells treated with RD2 exhibited apoptotic nuclei in a concomitant with the reduction of cellular p-Akt and p-mTOR. RD3 had minimal effects on such proteins. According to these results, molecular docking analysis revealed a high-affinity interaction between RD2 and an Akt molecule at the ATP-binding and the allosteric sites, indicating this RD2 as a potential Akt inhibitor. This study provides useful information of resveratrol derivatives RD2 for treating lung cancer via Akt/mTOR inhibition.


Introduction
Lung cancer is considered one of the important cancers according to its high rate of mortality. In addition, it was found that new cases of lung cancer are relatively high in comparison to other human cancers [1]. Non-small Lung Cancer (NSCLC) is the most common type of lung cancer, and this disease is frequently found at an advanced stage (III-IV) at the time of first detection [2]. The treatment for NSCLC is surgery, radiotherapy, chemotherapy, immunotherapy, and molecularly targeted therapy [3]. High incidence of chemotherapeutic resistance in NSCLC is one factor causing failure of the treatment

Cytotoxicity and Inhibition of Colony Forming Capacity of Resveratrol Derivatives
To investigate the anti-cancer potentials of resveratrol derivatives RD1, RD2, RD3, RD4, and RD5 on lung cancer cells, we determined the cytotoxic profile of resveratrol derivatives (RD1-RD5) and resveratrol in several lines of NSCLC cells namely A549, H23, and H460 cells.
NSCLC cells were cultured overnight and treated with various concentrations (0-200 µM) of resveratrol derivatives (RD1-RD5) and resveratrol for 24 h, and then cell viability was analyzed by the 3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide (MTT) assay to evaluate the percentage of cell viability. All of the data were calculated on the basis of the results of three replicated samples. The results revealed that RD1-RD5 compounds shown potential cytotoxic effects. Among them, we found that RD2, which had a similar structure to RD3 (Figure 1a), exerted a dramatically different cytotoxic effect on lung cancer cells. The difference of the structure at C-4 was that the hydroxyl group of RD2 was modified to the methoxy group of RD3 and at C-3, the methoxy group of RD2 was changed to the hydroxyl group of RD3. RD2 significantly reduced the viability of NSCLC (A549, H23, and H460) cells with half maximal inhibitory concentrations (IC 50 ) at 108.6 ± 10.82, 103.5 ± 6.08, and 138.3 ± 25.63 µM, respectively (Figure 2a-d), while RD3 reduced cell viability in NSCLC cells (A549, H23, andH460) with the IC 50 as more than 200 µM. Furthermore, resveratrol decreased cell viability in NSCLC cells (A549, H23, and H460) with the IC 50 at more than 200 µM which was the same RD3 (Figure 2a-d). This different action on cancer cells as a result of chemical structure modification has provided an insight for the structure-activity relationship that may facilitate the further development of the potent compound from the lead structure of resveratrol derivatives.
To confirm the selective cytotoxicity of resveratrol derivatives (RD2 and RD3) in cancer cells, resveratrol derivatives (RD2 and RD3) and resveratrol were determined for the cytotoxic profile in normal human cells, dermal papilla (DP). It was found that resveratrol derivatives (RD2 and RD3) and resveratrol did not significantly affect the cell viability percentages in DP cells, as shown in Figure 2e,f.
Colony formation assay was used to confirm the anti-cancer action of resveratrol derivatives, as it is a cell proliferation assay which can detect the ability of a single cell to grow into a colony. NSCLC cells were seeded and treated with RD2 and RD3 (1-200 µM) for 24 h. The treated cells were subjected to colony formation assay for 7 days. The crystal violet-staining colony represented the colony suppression effect of RD2 and RD3 compared to the untreated control (Figure 3a,b). The RD2 compound dramatically abolished the forming colony. Colony number was absent in response to RD2 at the concentrations of 100 and 200 µM, whereas there was no significant response to RD3 (Figure 3a,b).
RD2 (10-200 µM) significantly inhibited the number of colony growth compared with the control cells in a dose-dependent manner. The results showed that the number of the forming colony was significantly reduced by 31.04%, 98.18%, and 99.40% in response to the treatment of 10, 100, and 200 µM, respectively, in A549 cells and 25.43%, 94.16%, and 98.23% in response to the treatment of 10, 100, and 200 µM, respectively, in H23 cells. Moreover, RD2 (100-200 µM) significantly inhibited the colony size compared with the control cells. The results showed that the colony size was significantly decreased by 71.66% and 85.42% in response to the treatment of 100 and 200 µM, respectively, in A549 cells and 64.42% and 65.98% in response to the treatment of 100 and 200 µM, respectively, in H23 cells (Figure 3a,b).
To confirm the selective cytotoxicity of resveratrol derivatives (RD2 and RD3) in ca cer cells, resveratrol derivatives (RD2 and RD3) and resveratrol were determined for t cytotoxic profile in normal human cells, dermal papilla (DP). It was found that resverat derivatives (RD2 and RD3) and resveratrol did not significantly affect the cell viabil percentages in DP cells, as shown in Figure 2e,f. Resveratrol derivatives (RD1-RD5) and resveratrol decreases cell viability in NSCLC cells (A549, H23, and H460) and the normal cell line (DP cells). (a-c) MTT assay was used to evaluate cell viability after treatment with various concentrations of RD1-RD5 and resveratrol (0-200 mM) for 24 h in A549, H23, and H460 cells, respectively. (d) The IC 50 of RD1-5 and resveratrol against A549, H23, and H460 cells was calculated from MTT assay by comparison with an untreated control. (e) MTT assay was used to evaluate cell viability after treatment with various concentrations of RD2, RD3, and resveratrol (0-200 mM) for 24 h in DP cells, respectively. (f) The IC 50 of RD2, RD3, and resveratrol against DP cells was calculated from MTT assay by comparison with an untreated control. Data are represented as the mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001 when compared with the control group. RD3 (10-200 µM) significantly inhibited the number of colony growth compared with the control cells, in a dose-dependent manner. The results show that the number of the forming colony was significantly reduced by 20.44%, 21.94%, and 38.91% in response to the treatment of 10, 100, and 200 µM, respectively, in A549 cells and 24.83%, 27.70% and 59.74% in response to the treatment of 10, 100, and 200 µM, respectively, in H23 cells. However, for RD3, there was no significant effect in the changing colony size in both A549 and H23 cells (Figure 3a,b).
In summary, these results suggest the promising anti-cancer effects of RD2 and RD3, a new resveratrol derivatives.  . After 1 week, colonies were stained with 1.25% crystal violet and quantified by extraction with 10% acetic acid and the analysis are determined by number and size of colony cancer cell. Data are represented as the mean ± SEM (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001 when compared with the control group. # p < 0.05, ## p < 0.01, ### p < 0.001 when compared between RD2 and RD3.

Apoptosis Induction and Apoptotic-Related Protein Alteration in Response to RD2 and RD3 Treatments
Apoptosis is considered as an important mechanism of action of anti-cancer drugs [16]. The important characteristics of apoptotic cells include the nuclear morphological changes , chromatin condensation, and DNA fragmentation. Hoechst 33342 and propidium iodide (PI) double staining was utilized to determine the mode of cell death by monitoring the nucleus morphology of apoptotic cells as well as their membrane integrity. Cells were seeded in a 96-well plate at the density of 1 × 10 4 cells/well and treated with various concentrations of RD2 and RD3 (10-200 μM) for 24 h. The cells were co-stained with Hoechst 33342 and PI. Bright blue fluorescence of Hoechst 33342 stained cells was observed and the apoptotic cells exhibiting condensed and fragmented chromatin were monitored. The red fluorescence of PI indicated a late apoptosis cell and necrotic cells. Figure 4a,b show that RD2 and RD3 significantly caused apoptosis indicated by the increase in nucleus chromatin condensed cells and apoptotic nuclei.
RD2 significantly increased the apoptotic rates compared with the non-treated control cells, in a dose-dependent manner. The apoptotic ratios were 23.25%, 61.26%, and 86.30% in the treatment of 10, 100, and 200 μM, respectively, in A549 cells. In addition, RD2 significantly also increased the apoptotic rates compared with the control cells in a dose-dependent manner. The apoptotic ratios were 36.63% and 75.79% in the treatment of 100 and 200 μM, respectively, in H23 cells (Figure 3c,d). RD3 only significantly increased apoptotic ratios by 13.25% at 200 μM in A549 cells and 18.57% at 200 μM in (Figure 4a,b). According to the previous experiment showing that RD2 had a potent cytotoxicity effect, RD2 had also caused more dead cells than RD3 at the same concentrations (10-200 μM); this was observed from the blue fluorescence of condensed nuclei indicating the early Figure 3. Resveratrol derivatives inhibit colony formation activity and induce apoptosis in NSCLC cells (A549 and H23). (a,b) Cells were treated with RD2 and RD3 (0-200 µM for 24 h). After 1 week, colonies were stained with 1.25% crystal violet and quantified by extraction with 10% acetic acid and the analysis are determined by number and size of colony cancer cell. Data are represented as the mean ± SEM (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001 when compared with the control group. # p < 0.05, ## p < 0.01, ### p < 0.001 when compared between RD2 and RD3.

Apoptosis Induction and Apoptotic-Related Protein Alteration in Response to RD2 and RD3 Treatments
Apoptosis is considered as an important mechanism of action of anti-cancer drugs [16]. The important characteristics of apoptotic cells include the nuclear morphological changes, chromatin condensation, and DNA fragmentation. Hoechst 33342 and propidium iodide (PI) double staining was utilized to determine the mode of cell death by monitoring the nucleus morphology of apoptotic cells as well as their membrane integrity. Cells were seeded in a 96-well plate at the density of 1 × 10 4 cells/well and treated with various concentrations of RD2 and RD3 (10-200 µM) for 24 h. The cells were co-stained with Hoechst 33342 and PI. Bright blue fluorescence of Hoechst 33342 stained cells was observed and the apoptotic cells exhibiting condensed and fragmented chromatin were monitored. The red fluorescence of PI indicated a late apoptosis cell and necrotic cells. Figure 4a,b show that RD2 and RD3 significantly caused apoptosis indicated by the increase in nucleus chromatin condensed cells and apoptotic nuclei.
RD2 significantly increased the apoptotic rates compared with the non-treated control cells, in a dose-dependent manner. The apoptotic ratios were 23.25%, 61.26%, and 86.30% in the treatment of 10, 100, and 200 µM, respectively, in A549 cells. In addition, RD2 significantly also increased the apoptotic rates compared with the control cells in a dosedependent manner. The apoptotic ratios were 36.63% and 75.79% in the treatment of 100 and 200 µM, respectively, in H23 cells (Figure 3c,d). RD3 only significantly increased apoptotic ratios by 13.25% at 200 µM in A549 cells and 18.57% at 200 µM in (Figure 4a,b). According to the previous experiment showing that RD2 had a potent cytotoxicity effect, RD2 had also caused more dead cells than RD3 at the same concentrations (10-200 µM); this was observed from the blue fluorescence of condensed nuclei indicating the early stage of cell death.

RD2 Inhibits Akt/mTOR Singnaling Pathways
Having shown that the resveratrol derivatives decrease cell viability, induce apoptosis, and inhibit the growth of the cells, we next investigated the potential mechanistic effect of the compounds by focusing on the proliferative and survival signal Akt/mTOR. NSCLC cells (A549 and H23) were treated with RD2 and RD3 (10-100 μM) for 12 h. After which, all cells were collected and the expression levels of Akt, p-Akt, mTOR, and p-mTOR were investigated by Western blot analysis (Figures 5a-8a). The results reveal a significant decrease of p-mTOR/mTOR protein expression levels at 100 μM and a significant decrease of p-Akt/Akt protein expression levels at 50 and 100 μM of RD2 treatment in NSCLC cells (A549 and H23) compared with the untreated control. Interestingly, RD3 increased p-Akt/Akt protein expression levels at 10-200 μM in A549 cell and at 100-100 μM in H23 cells. Interestingly, RD3 had no effect on p-mTOR/mTOR protein expression levels. These results suggest that the anticancer activity of RD2 might act, at least in part, via Akt/mTOR inhibition. Immunofluorescence staining of Akt, p-Akt, mTOR, and p-mTOR was performed in RD2 and RD3 treated cells. The results further confirm the effect of RD2 and RD3 on the Akt/mTOR pathway. While Akt and mTOR fluorescence intensity were not notably changed in RD2 and RD3 in NSCLC (A549 and H23), RD2 induced a dramatic decrease of the p-mTOR signal in NSCLC cells (A549 and H23) at 50-100 μM, while RD3 only minimally affected the intensity of p-mTOR signal (Figures 5c-8c). RD3 represents the fragmented chromatin in apoptotic cells as a percentage compared with untreated control cells and another group. Data are represented as the mean ± SEM (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001 when compared with the control group. # p < 0.05, ## p < 0.01, ### p < 0.001 when compared between RD2 and RD3.

RD2 Inhibits Akt/mTOR Singnaling Pathways
Having shown that the resveratrol derivatives decrease cell viability, induce apoptosis, and inhibit the growth of the cells, we next investigated the potential mechanistic effect of the compounds by focusing on the proliferative and survival signal Akt/mTOR. NSCLC cells (A549 and H23) were treated with RD2 and RD3 (10-100 µM) for 12 h. After which, all cells were collected and the expression levels of Akt, p-Akt, mTOR, and p-mTOR were investigated by Western blot analysis (Figures 5a-8a). The results reveal a significant decrease of p-mTOR/mTOR protein expression levels at 100 µM and a significant decrease of p-Akt/Akt protein expression levels at 50 and 100 µM of RD2 treatment in NSCLC cells (A549 and H23) compared with the untreated control. Interestingly, RD3 increased p-Akt/Akt protein expression levels at 10-200 µM in A549 cell and at 100-100 µM in H23 cells. Interestingly, RD3 had no effect on p-mTOR/mTOR protein expression levels. These results suggest that the anticancer activity of RD2 might act, at least in part, via Akt/mTOR inhibition. Immunofluorescence staining of Akt, p-Akt, mTOR, and p-mTOR was performed in RD2 and RD3 treated cells. The results further confirm the effect of RD2 and RD3 on the Akt/mTOR pathway. While Akt and mTOR fluorescence intensity were not notably changed in RD2 and RD3 in NSCLC (A549 and H23), RD2 induced a dramatic decrease of the p-mTOR signal in NSCLC cells (A549 and H23) at 50-100 µM, while RD3 only minimally affected the intensity of p-mTOR signal (Figures 5c-8c). RD3 treatment significantly increased p-Akt intensity at 10-100 µM in A549 cells ( Figure 6b) and at 100-100 µM in H23 cells ( Figure 8b) compared with untreated control cells. mTOR were investigated by Western blot analysis. The results reveal a significant decrease of p-mTOR/mTOR protein expression levels at 100 μM and a significant decrease of p-Akt/Akt protein expression levels at 100 μM of resveratrol treatment compared with the untreated control (Figure 9a-d), while RD2 could reduce p-Akt/Akt protein expression levels at a lower concentration (50-100 μM). The cells were treated with RD2 and RD3 (0-100 μM) for 24 h. (a) Western blot analysis was performed to measure the p-mTOR/mTORrelated proteins. β-Actin protein was evaluated to confirm the equal loading of each protein sample. Densitometry of each protein level was calculated and the results are presented as a relative protein level. (b) A549 cells were stained with mTOR (red fluorescence) and Hoechst 33342 (blue fluorescence). The expression of mTOR was determined by immunofluorescence. (c) A549 cells were stained with p-mTOR (green fluorescence) and Hoechst 33342 (blue fluorescence). The expression of p-mTOR was examined using immunofluorescence. Relative protein levels were quantified by densitometry. Data are represented as the mean ± SEM. * p < 0.05, when compared with the control group. Figure 5. RD2 downregulates p-mTOR in NSCLC cells (A549). The cells were treated with RD2 and RD3 (0-100 µM) for 24 h. (a) Western blot analysis was performed to measure the p-mTOR/mTORrelated proteins. β-Actin protein was evaluated to confirm the equal loading of each protein sample. Densitometry of each protein level was calculated and the results are presented as a relative protein level. (b) A549 cells were stained with mTOR (red fluorescence) and Hoechst 33342 (blue fluorescence). The expression of mTOR was determined by immunofluorescence. (c) A549 cells were stained with p-mTOR (green fluorescence) and Hoechst 33342 (blue fluorescence). The expression of p-mTOR was examined using immunofluorescence. Relative protein levels were quantified by densitometry. Data are represented as the mean ± SEM. * p < 0.05, when compared with the control group.
To investigate whether resveratrol inhibits the Akt/mTOR signaling pathway, NSCLC cells (A549 and H23) were treated with resveratrol (10-100 µM) for 12 h. After which, all cells were collected, and the expression levels of Akt, p-Akt, mTOR, and p-mTOR were investigated by Western blot analysis. The results reveal a significant decrease of p-mTOR/mTOR protein expression levels at 100 µM and a significant decrease of p-Akt/Akt protein expression levels at 100 µM of resveratrol treatment compared with the untreated control ( Figure 9a-d), while RD2 could reduce p-Akt/Akt protein expression levels at a lower concentration (50-100 µM).  The cells were treated with RD2 and RD3 (0-100 µM) for 24 h. (a) Western blot analysis was performed to measure the p-Akt/Aktrelated proteins. β-Actin protein was evaluated to confirm the equal loading of each protein sample. Densitometry of each protein level was calculated and the results are presented as a relative protein level. (b) A549 cells were stained with Akt (red fluorescence) and Hoechst 33342 (blue fluorescence). The expression of Akt was determined by immunofluorescence. (c) A549 cells were stained with p-Akt (green fluorescence) and Hoechst 33342 (blue fluorescence). The expression of p-Akt was examined using immunofluorescence. Relative protein levels were quantified by densitometry. Data are represented as the mean ± SEM. * p < 0.05, when compared with the control group.

Molecular Docking Simulations Reveals the Reveratrol Derivatives Interactions with Akt-1 Protein
We further investigated whether Akt inhibition of RD2 might be a result of direct interaction between RD2 and Akt protein. To determine the plausible modes of Akt-1 interaction, molecular docking analyses were performed using co-crystal structures of ATP-competitive inhibitor (PDB ID 3CQW) and an allosteric inhibitor (PDB ID 5KCV).
The initial validation of the docking protocol was performed by redocking. The redocking result showed that co-crystal ligands (reference compounds) exhibited similar poses to the original crystal structures (Figures 10b and 11b), which was further confirmed by the root mean square deviation (RMSD) value of 0.451 Å and 0.957 Å, respectively.  The cells were treated with RD2 and RD3 (0-100 µM) for 24 h. (a) Western blot analysis was performed to measure the p-mTOR/mTORrelated proteins. β-Actin protein was evaluated to confirm the equal loading of each protein sample. Densitometry of each protein level was calculated and the results are presented as a relative protein level. (b) H23 cells were stained with mTOR (red fluorescence) and Hoechst 33342 (blue fluorescence). The expression of mTOR was determined by immunofluorescence. (c) H23 cells were stained with p-mTOR (green fluorescence) and Hoechst 33342 (blue fluorescence). The expression of p-mTOR was examined using immunofluorescence. Relative protein levels were quantified by densitometry. Data are represented as the mean ± SEM. * p < 0.05, when compared with the control group.
In the ATP-binding site of Akt-1, the binding affinity for resveratrol and RD2 were −8.054 and −8.041 kcal/mol, respectively (Table 1). The binding modes of all ligands revealed that they occupied the same binding site of the co-crystal ligand inhibitor (Figure 10c). The results reveal that all ligands formed a hydrogen bond with key residue Ala230, which is a similar binding pattern to the reference compound (Figure 10d-f). Resveratrol formed only one hydrogen bond interaction with Ala230 in the hinge region of the ATP-binding site. Moreover, previous studies have reported that resveratrol has inhibited Akt activity via interaction with the ATP-binding site of Akt-1 [11,17]. RD2 formed three hydrogen bonds with Glu228, Ala230 in the hinge region, and Asp292 in the DFG motif.  The cells were treated with RD2 and RD3 (0-100 µM) for 24 h. (a) Western blot analysis was performed to measure the p-Akt/Aktrelated proteins. β-Actin protein was evaluated to confirm the equal loading of each protein sample. Densitometry of each protein level was calculated and the results are presented as a relative protein level. (b) A549 cells were stained with Akt (red fluorescence) and Hoechst 33342 (blue fluorescence). The expression of Akt was determined by immunofluorescence. (c) H23 cells were stained with p-Akt (green fluorescence) and Hoechst 33342 (blue fluorescence). The expression of p-Akt was examined using immunofluorescence. Relative protein levels were quantified by densitometry. Data are represented as the mean ± SEM. * p < 0.05, ** p < 0.01, when compared with the control group.
In the allosteric binding site of Akt-1, the binding affinity for resveratrol and RD2 were −8.446, and −8.546 (Table 2). The allosteric binding site of Akt-1 mainly interacts with Trp80, which plays an essential role in the binding affinity. Furthermore, three ligands were formed by hydrophobic interaction with key residue Trp80 of the PH domain. All ligands adopted a similar binding mode of miransertib (reference compound) with an interface between the PH and the kinase domain (Figure 11c-f). Resveratrol formed four hydrogen bonds with Asn54 and Thr82 in the PH domain and Gln203 and Ser205 in the kinase domain. Likewise, a previous study reported that resveratrol could bind to Akt-1 with an allosteric site [18]. RD2 formed four hydrogen bonds with Asn54, Gln79 (2 bonds) in the PH domain, and Ser205 in the kinase domain. These results illustrate that the Akt signaling pathway is involved in the inhibitory effect of resveratrol and RD2.

Molecular Docking Simulations Reveals the Reveratrol Derivatives Interactions with Akt-1 Protein
We further investigated whether Akt inhibition of RD2 might be a result of direct interaction between RD2 and Akt protein. To determine the plausible modes of Akt-1 interaction, molecular docking analyses were performed using co-crystal structures of ATPcompetitive inhibitor (PDB ID 3CQW) and an allosteric inhibitor (PDB ID 5KCV).
The initial validation of the docking protocol was performed by redocking. The redocking result showed that co-crystal ligands (reference compounds) exhibited similar poses to the original crystal structures (Figures 10b and 11b), which was further confirmed by the root mean square deviation (RMSD) value of 0.451 Å and 0.957 Å , respectively.
In the ATP-binding site of Akt-1, the binding affinity for resveratrol and RD2 were −8.054 and −8.041 kcal/mol, respectively (Table 1). The binding modes of all ligands revealed that they occupied the same binding site of the co-crystal ligand inhibitor ( Figure  10c). The results reveal that all ligands formed a hydrogen bond with key residue Ala230, which is a similar binding pattern to the reference compound (Figure 10d-f). Resveratrol formed only one hydrogen bond interaction with Ala230 in the hinge region of the ATPbinding site. Moreover, previous studies have reported that resveratrol has inhibited Akt activity via interaction with the ATP-binding site of Akt-1 [11,17]. RD2 formed three hydrogen bonds with Glu228, Ala230 in the hinge region, and Asp292 in the DFG motif.
In the allosteric binding site of Akt-1, the binding affinity for resveratrol and RD2 were −8.446, and −8.546 (Table 2). The allosteric binding site of Akt-1 mainly interacts with Trp80, which plays an essential role in the binding affinity. Furthermore, three ligands were formed by hydrophobic interaction with key residue Trp80 of the PH domain. All ligands adopted a similar binding mode of miransertib (reference compound) with an interface between the PH and the kinase domain (Figure 11c-f). Resveratrol formed four hydrogen bonds with Asn54 and Thr82 in the PH domain and Gln203 and Ser205 in the kinase domain. Likewise, a previous study reported that resveratrol could bind to Akt-1 with an allosteric site [18]. RD2 formed four hydrogen bonds with Asn54, Gln79 (2 bonds) in the PH domain, and Ser205 in the kinase domain. These results illustrate that the Akt signaling pathway is involved in the inhibitory effect of resveratrol and RD2. To test the potential binding of RD2 and RD3 on the activation site of Akt, we dock RD2, RD3, and Akt activator SC79 as ligands into the PH domain of Akt-1 (PDB ID 1UN using AutoDock Vina. A previous study reported that SC79 directly binds to the PH main and enhances Akt phosphorylation and activation [19]. In the PH domain of Akt-1, the binding affinity for RD2, RD3, and SC79 were −5.2 −6.072, and −5.62 kcal/mol, respectively ( Table 3). The hydrogen-bond interactions of PH domain, Lys14, Arg25, and Arg86, were documented to be essential for Akt activat [20]. As shown in Figure 12, RD3 formed hydrogen bonds with Ile19, Arg23, and Arg and the activator SC79 formed hydrogen bonds with the Lys14, Tyr18, Ile19, and Ar side-chain residues. The molecular docking result clarified the mechanism of Akt acti tion by RD3, which might present a reasonable interaction pose in the PH domain of A 1, similarly to Akt activator SC79 and native ligand IP4.        To test the potential binding of RD2 and RD3 on the activation site of Akt, we docked RD2, RD3, and Akt activator SC79 as ligands into the PH domain of Akt-1 (PDB ID 1UNQ) using AutoDock Vina. A previous study reported that SC79 directly binds to the PH domain and enhances Akt phosphorylation and activation [19].
In the PH domain of Akt-1, the binding affinity for RD2, RD3, and SC79 were −5.297, −6.072, and −5.62 kcal/mol, respectively ( Table 3). The hydrogen-bond interactions of the PH domain, Lys14, Arg25, and Arg86, were documented to be essential for Akt activation [20]. As shown in Figure 12, RD3 formed hydrogen bonds with Ile19, Arg23, and Arg25, and the activator SC79 formed hydrogen bonds with the Lys14, Tyr18, Ile19, and Arg23 side-chain residues. The molecular docking result clarified the mechanism of Akt activation by RD3, which might present a reasonable interaction pose in the PH domain of Akt-1, similarly to Akt activator SC79 and native ligand IP4.

Discussion
Lung cancer is one of the most important human cancers and the disease causes high mortality with poor overall prognosis. Conventional drugs are frequently not effective at an advanced stage as well as for metastatic lung cancers [21]. New treatment options including targeted therapy are found to potentially improve survival compared to those treated without targeted therapies [2]. It was found that plant-derived bioactive compounds and their synthetic analogs show promising antitumor activity as a targeted therapeutic drug [22]. One such bioactive agent is resveratrol which is a polyphenol exhibiting health beneficial effects including antioxidant, anti-inflammation, and antitumor effects. In addition, in lung cancer cells, resveratrol shows anti-cancer activities. Resveratrol was demonstrated to inhibit cancer cell growth, mediate cell cycle arrest, and induce apoptosis [23]. For lung cancer, it was demonstrated that resveratrol caused the induction of p53-dependent apoptosis in A549 cells [24] Resveratrol has been recognized as an interesting lead compound for chemical structure modifications to gain more potency and altered water solubility [25]. In addition, the anti-cancer activity of the derivatives as well as the presence of each modified moiety may lead to new knowledge of the structure-activity relationship that is useful for further development of the compound. Therefore, we modified the molecular structure of resveratrol into five derivatives (RD1-RD5) (Figure 1a). Among these compounds, RD2 was shown to have the most potent cytotoxic effect on NSCLC cells (Figure 2a-d). It is interesting to us that the RD2 and RD3 had a very similar structure. however, RD2 had more anti-cancer effect than that of RD3. RD2 exerted more effects on cancer cell viability (Figure 2d), on proliferating tumor cells (Figure 3a,b), and apoptotic cell death (Figure 4a,b). The anti-proliferation effect of resveratrol derivatives was consistent with the previous study. Regarding their parental compound, resveratrol was previously reported to inhibit cell proliferation in a concentration-and time-dependent manner in lung cancer A549 cells via the induction of cell cycle arrest at the G0/G1 phase [26]. Moreover, pterostilbene, an analog of resveratrol, was shown to significantly decrease cell viability and induced S phase arrest and the activation of caspase −3, −8, and −9 in human lung squamous cell carcinoma (H520) cells [27].
For cancer aggressiveness and treatment failure, it is known that the cellular survival signals can potentiate cell growth, immune escape, and chemotherapeutic drug resistance. The phosphoinositide 3-kinase (PI3K)/Akt signaling pathway is an important survival and proliferating signaling pathway that was shown to be up-regulated and overactivated in various types of cancer. The PI3K/Akt signal controls the hallmarks of cancer, including cell survival, proliferation, motility and metastasis, angiogenesis, and inflammatory factor recruitment [28]. Deregulation of the PI3K/Akt/mTOR pathway was shown to be critical for lung tumorigenesis and advanced disease [29]. Taken together, the inhibition of the PI3K/Akt/mTOR pathway may offer a possible option for the improvement of lung cancer therapy. In the present study, it was revealed that RD2 significantly decreased the level of active Akt (phosphorylated Akt) and active mTOR (phosphorylated mTOR) in A549 and H23 cells. The inhibition of Akt/mTOR signaling was consistent with the inhibitory effects of the compound on cell survival and the expression of related proteins (Figures 5a-8a). Consistently, several studies reported that resveratrol has an anti-tumor effect via Akt and mTOR. Resveratrol downregulated the PI3K/Akt/mTOR signaling pathway, which resulted in apoptosis in glioma cells [9]. Resveratrol also inhibited proliferation and migration through sirtuin 1-mediated post-translational modification of the PI3K/Akt pathway in hepatocellular carcinoma cells [30]. Treatment with resveratrol alone or in combination with rapamycin efficiently inhibited cell growth by inducing growth stop by targeting Akt activation and preventing mTORC1 signaling cascade phosphorylationactivation [31] Consistently, resveratrol was shown to decrease cellular levels of PI3K and p-Akt in NSCLC [12]. The combination of resveratrol and erlotinib could decrease the p-Akt, p-mTOR, and p-S6K relative to either agent alone, indicating the synergistic suppression of the Akt/mTOR pathway in NSCLC [32].
In this study, the binding affinity and the number of hydrogen bonds were suggested to play a role on the inhibition mechanism of Akt-1 by the RD2 compound. Docking analysis of RD2 with the ATP binding site and the halosteric site of Akt-1 shows the ability to interact with both binding sites; however, the Vina scoring function and hydrogen bonds of the two ligands complex with the allosteric binding site suggest them as the preferred ligand for an allosteric binding site than the ATP-binding site of Akt-1. Similarly, it has been reported that resveratrol has a dual action inhibitor of Akt. It is a potential ATP-competitive inhibitor and an allosteric inhibitor of Akt [33].

Preparation of the Resveratrol Derivatives Stock Solution
Resveratrol derivatives (RD1-RD5) were prepared as a 40 mM master stock solution by dissolving in DMSO solution and stored at −20 • C. The compounds were freshly diluted with medium to the desired concentrations before using. The final concentration of DMSO in solution was less than 0.5%, which caused no signs of cytotoxicity.

Cell Viability
Cell viability was investigated by MTT colorimetric assay. All cell lines were cultured at the density of 1 × 10 4 cells per well in a 96-well tissue culture plate and incubated in an incubator. After that, cells were treated with various concentrations of RD1-RD5 for 24 h at 37 • C and analyzed by the MTT assay according to the manufacturer's protocol (Sigma Chemical, St. Louis, MO, USA). In calculating the cell viability by a microplate reader (Anthros, Durham, NC, USA), the measured absorbance of treated cells were divided by the value of untreated cells and are reported as a percentage.

Nuclear Staining Assay
This method was applied to define apoptotic and necrotic cell death by using nuclear staining with Hoechst 33342. The cells were seeded on 96-well plates at the density 1 × 10 4 cells per well in 96-well plates and incubated overnight. The cells were treated with various concentrations of RD1-RD5 and incubated for 24 h at 37 • C. Afterward, the cells were incubated with Hoechst 33342 (10 µg/mL) / PI (5 µg/mL) solution for 30 min at 37 • C. Then, they were visualized and imaged under a fluorescence microscope (Olympus DP70, Melville, NY, USA). Results are reported as a percentage of apoptotic cells.

Colony Formation Assay
The measurement of clonogenicity was demonstrated by colony formation assay. Cells were seeded at a density of 300 cells/well into 6-well plate and then further treated with various concentrations resveratrol derivatives for 24 h after that the medium was replaced and the cell were incubated for colony number and colony size were investigated after fixing with methanol and acetic acid (3:1) solution and staining with 0.05% w/v crystal violet in 4% formaldehyde.

Western Blot Analysis
Western blotting was used to detect the presence of specific protein. Cells were seed at a density of 4 × 10 5 cells/well in 6 well plates overnight. Cells were treated with various concentrations RD1-RD5 were treated for 12 h. After treatment, the apoptosis cells were collected by centrifuging media with 1500 rpm for 5 min and aspirating supernatants. Then, cells were washed with cold 1× PBS and incubated with 1X RIPPA 40 µL containing 10× RIPA buffer 100 µL, protease inhibitors (PI) 100 µL, PMSF 10 µL, and Triton × 10 µL for 30 min on ice. BCA protein assay kit from Pierce Biotechnology (Rockford, IL, USA) analyzed protein concentrations. The extracted proteins were separated by 7.5-10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membrane (PVDF). Then, the membranes were blocked with 5% non-fat milk powder for 2 h. After that, they were incubated with primary antibody at 4 • C Akt (#9272), phosphorylated Akt (#4060), mTOR (#2983S), and p-mTOR (#5536S) overnight. The membranes were washed with Tris-buffered saline/Tween 20 3 times and incubated with secondary antibodies for 2 h. Finally, the protein bands were detected using chemiluminescence substrate and exposed by Chemiluminescent ImageQuant LAS4000. Protein bands were analyzed using Image J software. tential therapy for NSCLC through the suppression of Akt/mTOR-mediated survival. In addition, RD2 could exert the inhibition of the ATP-binding site and allosteric binding site with the Akt protein. As Akt is critical for cell survival, our results might be used in demonstrating that RD2 may represent a potential therapeutic candidate for the treatment of lung cancer.