Structure-Activity Relationships and Molecular Docking Analysis of Mcl-1 Targeting Renieramycin T Analogues in Patient-derived Lung Cancer Cells.

Myeloid cell leukemia 1 (Mcl-1) and B-cell lymphoma 2 (Bcl-2) proteins are promising targets for cancer therapy. Here, we investigated the structure–activity relationships (SARs) and performed molecular docking analysis of renieramycin T (RT) and its analogues and identified the critical functional groups of Mcl-1 targeting. RT have a potent anti-cancer activity against several lung cancer cells and drug-resistant primary cancer cells. RT mediated apoptosis through Mcl-1 suppression and it also reduced the level of Bcl-2 in primary cells. For SAR study, five analogues of RT were synthesized and tested for their anti-cancer and Mcl-1- and Bcl-2-targeting effects. Only two of them (TM-(–)-18 and TM-(–)-4a) exerted anti-cancer activities with the loss of Mcl-1 and partly reduced Bcl-2, while the other analogues had no such effects. Specific cyanide and benzene ring parts of RT’s structure were identified to be critical for its Mcl-1-targeting activity. Computational molecular docking indicated that RT, TM-(–)-18, and TM-(–)-4a bound to Mcl-1 with high affinity, whereas TM-(–)-45, a compound with a benzene ring but no cyanide for comparison, showed the lowest binding affinity. As Mcl-1 helps cancer cells evading apoptosis, these data encourage further development of RT compounds as well as the design of novel drugs for treating Mcl-1-driven cancers.


Introduction
Estimates of the worldwide cancer records of incidence and mortality from all types of cancers have revealed that lung cancer is the most common cause of cancer death, and is showing an increasing incidence. Furthermore, the 5-year survival rate from lung cancer is critically low due to its resistance to cancer therapy and due to disease relapse [1,2]. The improved clinical response as a result of targeted therapies has suggested that the use of more precise drugs focused on the molecular targets underlying the aggressiveness of the cancer and its drug resistance would be highly beneficial.
It has been widely shown that one of the important hallmarks of cancer is the ability of the tumor cells to evade apoptosis [3]. In general, programmed apoptosis is a well-controlled cell death mechanism for the removal of unwanted or harmful cells. However, what frequently occurs in cancer is the upregulation of the anti-apoptotic B-cell lymphoma 2 (Bcl-2) family of proteins [4], which consequently facilitates oncogenesis through cell death resistance [5]. The Bcl-2 family of proteins belong to the class of BH-domain-containing proteins [6], which can be classified into anti-apoptotic members (Bcl-2, Mcl-1, Bcl-w, Bcl-xl, and Bfl-1/A1) and pro-apoptotic members (Bax, Bak, Bad, Bid, Bak, Bim, Puma, Noxa, Hrk, and Bmf). The anti-apoptotic proteins interact and are antagonized specifically to certain BH3-only proteins and pro-apoptotic proteins [7]. The pro-apoptotic proteins Bax and Bak are the key members that trigger permeabilization of the mitochondrial membrane and the release of pro-apoptogenic proteins. Recent evidence has suggested that the ability to evade apoptosis, survival during metastasis, and resistance to therapy are likely to be dependent on the expression and function of the pro-survival Bcl-2 proteins [8]. Consequently, it is not unexpected that the amplification and overexpression of pro-survival Bcl-2 proteins, such as Bcl-2 and Mcl-1, are found in many cancer types (e.g., non-small-cell lung cancer, breast cancer, ovarian cancer, prostate cancer, and pancreatic cancer) [9][10][11]. Bcl-2 family members, like Bcl-2 and Mcl-1, have been shown to dominate negative regulation for apoptosis control and to be responsible for chemotherapeutic resistance [12]. An analysis of cancer specimens revealed that up to 50% of lung cancer specimens exhibited elevated levels of Mcl-1 expression, which is associated with a poor prognosis for lung cancer patients [13,14]. Not unpredictably, the downregulation of Mcl-1 increases cancer cell sensitivity to standard anticancer drugs, such as etoposide, doxorubicin, and ABT-737 [14]. Moreover, the reduction of Mcl-1 sensitized osteosarcoma cells to chemotherapeutic agents [15]. In lung cancer, evidence from a microarray analysis indicated that Mcl-1 is highly expressed in the cancer tissue of non-small cell lung cancer (NSCLC) patients [16,17]. In addition, it was shown that NSCLC cell lines exhibited high levels of Mcl-1, and the inhibition of such a protein by siRNA could potently mediate NSCLC cell apoptosis [18]. Additionally, evidence has suggested that, in NSCLC, Mcl-1 inhibition showed superior potential for having a cancer therapeutic effect when compared with Bcl-xl inhibition [18]. Taken together, new drugs with modes of action involving eliminating Mcl-1 in lung cancer cells are of interest as candidates for Mcl-1-targeted therapy.
Marine organisms function as a reservoir of potent marine-derived agents capable of inhibiting the growth of cancer cells, as has been demonstrated in in vitro and in vivo studies [19,20]. For instance, there are four marine agents approved for use in the treatment of cancers: cytarabine, trabectedin, eribulin mesylate, and the conjugated antibody brentuximab vedotin [21]. Moreover, it has been found that substances from marine organisms, such as renieramycins, have the potential to prevent tumor formation and induce cell death via the apoptosis pathway [22]. A previous study reported that renieramycin T (RT), a renieramycin-related compound isolated from the blue sponge Xestospongia sp., was dominantly toxic to lung cancer cells and mainly exerted this effect through apoptosis induction via the targeting of Mcl-1 for ubiquitin-proteasomal degradation [23]. As RT has a complex structure composed of several chemical moieties, understanding the structure-activity relationships (SARs) is a necessity for identification of the active moieties that are critical for drug action and that hold promise to increase drug precision and potency. Using RT as a lead compound, we aimed to establish such structure-activity relationships (SARs) and the subsequent SAR-directed optimization for treatment. The newly synthesized simplified parts of RT were developed and the active parts as well as the required moieties of the compound for the Mcl-1-targeted effect were evaluated in the present study utilizing protein analysis in combination with molecular docking simulation.

Cytotoxicity and Apoptosis-inducing Effect of RT on Patient-derived Primary Lung Cancer Cells
Chemotherapeutic drug resistance is accepted to be a major cause of therapeutic failure, tumor recurrence, and disease progression in lung cancer [24]. Mcl-1, an anti-apoptotic member of the Bcl-2 family, was demonstrated to be mainly involved in chemotherapeutic resistance as this protein is frequently found to be highly expressed in lung cancer [25] and the diminishment of Mcl-1 can lead to cancer cell death [26,27]. To characterize the potency of the anti-cancer activity of RT (Figure 1a), we determined the cytotoxic profile of RT in chemotherapeutic resistant primary lung cancer cells (ELC12, ELC16, ELC17, and ELC20) and lung cancer cell lines (H460). The basic cell morphology of the NSCLC and patient-derived primary cancer cell lines and the molecular characteristics are shown in Figure 1b. The results indicated that RT exerted a superior cytotoxic potency when compared with the commonly used chemotherapeutic drugs, including cisplatin, etoposide, and doxorubicin, at the equivalent concentrations ( Figure 1c). Figure 1c shows that nearly all of the lung cancer cells were resistant to cisplatin at 0-10 µM, as the cell viability was found to be above 90% after treatment, while doxorubicin and RT showed comparable potent cytotoxic effects and both compounds could reduce cancer cell viability by approximately 70% at the 10 µM concentration. The half maximal inhibitory concentrations (IC 50 ) values of RT and the commercial drugs were calculated and the results indicated that the IC 50 of RT was generally lower than that of the chemotherapeutic drugs. Importantly, RT showed greater potency compared to that of doxorubicin in all the cells (Figure 1d). The apoptotic cell death and necrosis were further evaluated by Hoechst33342 and propidium iodide (PI) staining, respectively. We tested the apoptosis induction effect of cisplatin, etoposide, and doxorubicin in H460 cells and found consistent results with the cytotoxicity results, showing that doxorubicin caused the highest apoptosis, as indicated by the fragmented or condensed nuclei ( Figure 1e). Then, the apoptosis induction effect of RT was evaluated in all lung cancer cells (H460, H292, H23, A549, ELC12, ELC16, ELC, 17,and ELC 20). The result revealed that RT caused an increase in apoptosis in a concentration-dependent manner, whereas it exhibited a minimal necrotic cell death effect, as shown in Figure 1e,f. We confirmed the apoptotic cell death by determination of cleaved PARP protein using Western blot analysis. The result showed an increase of cleaved PARP in response to RT treatment compared to control (Figure 1g).

Cytotoxic and Mcl-1-targeting Activities of the Simplified Right-half model of RT Compounds
The SAR of a bioactive compound is very useful information that could facilitate the development of drugs, as this information describes the critical parts to be exploited for drug action as well as helps to increase the potency of the initially detected activity. As RT has been shown to have promising anti-cancer activity against lung cancer through its Mcl-1-targeted activity, as demonstrated in our previous study [23], we designated RT as a lead compound and generated simplified right-half models in order to elucidate the SAR. The structures of RT and the simplified right-half model of  (e-f) Cells were seeded and treated with 0-10 µM of RT or chemotherapeutic drugs (cisplatin, etoposide, and doxorubicin) for 24 h before adding Hoechst 33342 and PI to stain the cell nucleuses. Images were detected by using a fluorescence microscope and the percentages of nuclear-fragmented and propidium iodide (PI)-positive cells were calculated. (g) ELC12, ELC16, ELC17, and ELC20 cells were treated with 0-1 µM of RT for 24 h. Western blot analysis was performed to detect the PARP and cleaved PARP protein levels. The blots were reprobed with β-actin to confirm an equal loading of each of the protein samples and densitometry was used to calculate the protein expression levels. Densitometric values of protein levels were presented as the fold changes relative to uncleaved form of the protein. Data represent the mean ± SEM (n = 3) (* 0.01 ≤ p < 0.05, ** p < 0.01, compared with the untreated control).   In order to verify the Mcl-1-targeted activity of the simplified right-half compounds of RT, we first confirmed the Mcl-1-targeted action of RT in primary lung cancer cells. We determined the Mcl-1 level in H460, ELC12, and ELC16 by immunofluorescence assay. The results showed that RT reduced the fluorescence intensity, reflecting the Mcl-1 protein, in H460, ELC12, and ELC16 cells (Figure 3a). Western blot analysis was used to confirm the effect of RT on Mcl-1 and its possible effect on the anti-apoptotic Bcl-2 proteins. The Mcl-1, Bcl-2, and proapoptotic Bax in the Bcl-2 family proteins were determined and the results showed that RT significantly decreased the level of Mcl-1 and Bcl-2 in primary lung cancer cells. In contrast, RT had no effect on the Bax protein ( Figure 3b). It was noteworthy that while we showed in this study that RT reduced both Mcl-1 and Bcl-2 in primary lung cancer cells, RT was previously shown to have a minimal effect on Bcl-2 in H460 cells [28].

Discussion
In cancer biology, it is well known that the evasion of apoptosis is a prominent hallmark of cancer [3]. Apoptosis is an important component of various processes as it contributes to the elimination of unwanted cells to maintain the equivalence between cell survival and cell death. Therefore, the dysregulation of apoptosis results in sustained cell proliferation and enhanced tumor development [30], and consequently, drugs or treatment strategies that can restore the apoptotic signaling pathways may benefit the management of cancer.
The essential regulators of apoptosis are the Bcl-2 family of proteins [31]. Regarding evading apoptosis, the roles of anti-apoptotic members, like Mcl-1 and Bcl-2, have been shown in many in vitro and in vivo studies [5,15,[32][33][34][35][36]. The protein levels of Bcl-2 and Mcl-1 in various types of tumor tissues were found to be frequently augmented at a higher ratio than the other anti-apoptotic proteins in the Bcl-2 family [26]. In particular, increased expressions of Bcl-2 and Mcl-1 reflect a poor prognosis

Discussion
In cancer biology, it is well known that the evasion of apoptosis is a prominent hallmark of cancer [3]. Apoptosis is an important component of various processes as it contributes to the elimination of unwanted cells to maintain the equivalence between cell survival and cell death. Therefore, the dysregulation of apoptosis results in sustained cell proliferation and enhanced tumor development [30], and consequently, drugs or treatment strategies that can restore the apoptotic signaling pathways may benefit the management of cancer.
The essential regulators of apoptosis are the Bcl-2 family of proteins [31]. Regarding evading apoptosis, the roles of anti-apoptotic members, like Mcl-1 and Bcl-2, have been shown in many in vitro and in vivo studies [5,15,[32][33][34][35][36]. The protein levels of Bcl-2 and Mcl-1 in various types of tumor tissues were found to be frequently augmented at a higher ratio than the other anti-apoptotic proteins in the Bcl-2 family [26]. In particular, increased expressions of Bcl-2 and Mcl-1 reflect a poor prognosis for many malignancies, including lung cancer [37][38][39]. Not only is their increased expression critical for oncogenesis and cancer progression, but these proteins are also involved in conferring chemotherapeutic drug resistance [35,[40][41][42][43]. Research was performed using Bcl-2 as a target for overcoming chemoresistance through BCL2 gene silencing to improve the clinical outcome in small-cell lung cancer [44]. Furthermore, in a mouse lung adenocarcinoma model, Mcl-1 overexpression was shown to help tumor progression by inhibiting Myc-induced apoptosis [40]. Taken together, compounds with potent activity for eliminating Bcl-2 or Mcl-1 in cancer cells are of great interest as good candidates for targeted therapy. In agreement with the use of Mcl-1 as a target for cancer therapy [11], our previous study highly supported this concept, where in our experiments, the treatment of RT in an NSCLC cell line (H460) resulted in apoptotic cell death through an Mcl-1 proteasomal degraded mechanism [23]. In the same way, not only in a known lung cancer cell line, but we also proved that the absence of Mcl-1 after RT treatment in primary lung cancer cell lines derived from patients (Figure 3a,b) could trigger the apoptotic pathway, which led to the death of cancer cells (Figure 1f,g).
Renieramycin T (RT) (Figure 1a) is a tetrahydroisoquinoline alkaloid compound that is part of the renieramycin family and was first isolated from the blue sponge Xestospongia sp. by pretreatment with potassium cyanide in a study in 2009 [45]. Recently, its anti-cancer activities have been reported against several types of cancer cells, such as colon (HCT116), prostate (DU145) [46], non-small cell lung (H292, H460, and QG56) [47], breast (T47D), and pancreatic (AsPC1) cancer cells [45]. Moreover, a modified form of RT, 5-O-acetyl-renieramycin T, was shown to induce the death of lung cancer stem cells and sensitize cisplatin-mediated apoptosis in lung cancer cells [48]. Our previous study revealed that the effects of RT on the apoptotic mechanism depended on the disappearance of Mcl-1 through the increase in Mcl-1 protein degradation [23]. In this study, we further confirmed that RT also had anti-cancer activities as well as Mcl-1-targeted activity in patient-derived primary lung cancer cells, with a lower IC 50 compared to other first-line chemotherapeutic drugs (Figure 1b-g, Figure 3a,b). It was newly discovered that RT could decrease the level of the Bcl-2 protein in primary lung cancer cells, while we did not observe this effect before in the lung cancer cell line H460 [23]. The explanation for this may due to the specific mutations of H460 cells, including KRAS and PI3K. The mutation of these two oncogenes may cause a high expression of Bcl-2 in H460 cells [28]. Gathering all the collected information, it is not surprising that RT could be an outstanding applicant for further clinical investigation in anti-cancer drug development.
In spite of the marvelous anti-cancer effects of RT, it is hard to deny that RT is difficult to fabricate in massive large-scale synthesis due to its large and complicated structure. Besides, its structure-activity relationships are still unrecognized. Therefore  (Figure 4c,d), but as we previously discussed in the section about the RT-treatment conditions, Mcl-1 depletion was the main mechanism by which these compounds induced apoptosis.
It has been shown that the anti-apoptotic potency of the Bcl-2 family of proteins primarily relies on their stability [49]. Mcl-1 is considered a very unstable protein compared to the other Bcl-2 family proteins and the degradation of Mcl-1 can be induced by anti-cancer agents [50][51][52][53]. The stabilization and degradation of Mcl-1 were reported to be regulated by a phosphorylation mechanism at the PEST region, which is the N-terminus of Mcl-1 enriched in proline (P), glutamate (E), serine (S), and threonine (T) residues and rich in putative phosphorylation sites [54]. Several pieces of evidence have suggested that the phosphorylation of Mcl-1 induced by extracellular signal-regulated kinases (ERK), c-Jun NH2-terminal kinase (JNK), and p38 at Threonine 92, Threonine 163, and Serine 121 can stabilize the Mcl-1 protein level [55][56][57]; whereas, the phosphorylation by glycogen synthase kinase-3 (GSK-3) at Serine 155 and Serine 159 destabilizes the Mcl-1 protein level and inhibits the interaction between Mcl-1 and the pro-apoptotic protein Bim [58][59][60][61]. From this information, we made the hypothesis that, supposing our compounds attached Mcl-1 at ERK, JNK, or p38 binding sites, they might prevent the phosphorylation mechanism of those molecules and lead to destabilization of the Mcl-1 protein level. To prove our hypothesis, we applied computational molecular docking using the mitogen-activated protein kinase (MAPK) docking motif, also known as the D-motif, found in the Mcl-1 protein sequence 137-143 as the target for the compounds and Mcl-1 interactions to characterize the behavior of these small molecule compounds at the binding site of the Mcl-1 protein [62,63]. Likewise, the circled part of the compounds in Figure 5a, which it is supposed are necessary for their mechanisms, was deployed as a docking site for examination of the binding properties and SARs. After investigation (Figure 6a-e), the results revealed that the binding affinity of RT, TM-(-)-18, and TM-(-)-4a was high and in a similar range, which indicated their ability to bind properly with Mcl-1 and induce destabilization of the protein; whereas, the binding affinity of TM-(-)-45, which has a benzene ring but no cyanide, was lower than the others, thus reflecting its incapability to induce Mcl-1 destabilization, as previously supported by the experimental data.
The appearance of both cyanide and a benzene ring (Figure 5a) in RT, TM-(-)-18, and TM-(-)-4a was found to be necessary for the induction of Mcl-1 destabilization. This study energetically supported the SAR concept, which is key to many aspects of new drug discovery [64] and provides beneficial information for further anti-cancer drug modification and drug development as SARs can then be used to predict the activities of new molecules from their molecular structure because there is a relationship between molecular structures and their biological activity. This allows the modification of the effect or the potency of a bioactive compound by changing its chemical structure. Therefore, the SAR concept is essential in drug discovery to guide the acquisition or synthesis of desirable new compounds, as well as to further characterize existing molecules. Our

Patient-derived Primary Lung Cancer Cell Line Preparation from Malignant Pleural Effusion
The patient-derived malignant cancer cells were isolated from pleural effusions of recurrent or advanced stage non-small cell lung cancer patients who had been diagnosed at the King Chulalongkorn Memorial Hospital. The protocol of conduction was approved by the Ethics Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (IRB 365/62) and was obtained informed consents from all participants. This study was carried out in accordance with the principles of World Medical Association Declaration of Helsinki. Primary cancer cells were collected from pleural effusion (500-1000 mL) through thoracentesis. The collected samples were centrifuged at 300 g for 10 min, at 4 • C and the cells were resuspended in RPMI medium with 10% FBS, 2 mM L-glutamine, and 100 units/mL of each of penicillin and streptomycin. After culturing for 10-15 passages, they were characterized as the patient-derived primary cancer cell lines (ELC12, ELC16, ELC17, and ELC20). The characteristics as well as their status of mutation were presented in the Figure 1b.

Simplified Right-half Model of RT Compounds Preparation
The compounds in solid form were dissolve in DMSO to 50 mM stock solution and stored at −20 • C. They were freshly diluted to concentrations used in the experiments with an awareness that final concentration of DMSO should be less than 0.5%.

Nuclear Staining Assay
Hoechst 33342 and PI double staining were applied to define apoptotic and necrotic cell death through nuclear co-staining. NSCLC cell lines (H460, H292, H23, and A549) and patient-derived primary lung cancer cell lines (ELC12, ELC16, ELC17, and ELC20) were seeded into 96-well plates at the density of 1 × 10 5 cells/well overnight and treated with 5 and 10 µM of chemotherapeutic agents (cisplatin, etoposide, and doxorubicin), RT, and right-half model of RT compounds (TM-(-)-18 and TM-(-)-4a) for 24 h. Afterwards, the cells were stained with 10 µg/mL Hoechst 33342 for 15 min at 37 • C and then stained with 5 µg/mL PI before immediately detecting fluorescence of nuclear-stained cells by fluorescent microscope (Nikon ECLIPSE Ts2, Tokyo, Japan). The number of nuclear condensed and DNA fragmented cells was reported as the percentage of apoptotic cells.

Immunofluorescence for Mcl-1
Immunofluorescence was introduced to evaluate whether or not cells in particular samples express Mcl-1 through antibody specification. NSCLC cell line (H460) and patient-derived primary lung cancer cell lines (ELC12 and ELC16) were seed overnight in 96-well plates at the density of 1 × 10 5 cells/well. Then, they were treated with 1 µM of RT, TM-(-)-18, and TM-(-)-4a and incubated for 24 h. After that, cells were fixed with 4% of paraformaldehyde for 30 min, permeabilized by 0.5% of Triton X-100 in PBS for 5 min, followed by blocking with 10% of FBS in 0.1% of Triton X-100 for further 1 h at room temperature (RT). Primary antibody of Mcl-1 at proportional 1:100 in 10% of FBS was applied to the cells before incubation overnight at 4 • C. After incubation time, Alexa Fluor 488 conjugated with goat anti-rabbit IgG secondary antibody was added and incubated in dark for 1 h at RT. Cell nucleuses were stained with Hoechst 33342 and then visualized under fluorescent microscope (Nikon ECLIPSE Ts2, Tokyo, Japan).

Western Blot Analysis
Western blot analysis was used to determine the amount of specific proteins in the cells. NSCLC cell line (H460) and patient-derived primary lung cancer cell lines (ELC12, ELC16, ELC17, and ELC20) were seeded overnight at the density of 4×10 5 cells/well and treated with RT, TM-(-)-45, TM-(-)-18, TM-(-)-4a, TM-(-)-52, and TM-(-)-55) for 24 h. Cells were then collected by centrifuging media with 1500 rpm for 5 min and lysed with radioimmunoprecipitation assay (RIPA) lysis buffer containing 25 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS, and a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IN, USA) for 30 min at 4 • C. The lysates were collected and their protein contents were determined by a BCA protein assay kit (Pierce Biotechnology, Rockford, IL, USA). Equivalent amount of proteins from each sample (70 µg) was separated using SDS-polyacrylamide gel electrophoresis and further transferred to 0.2 µm polyvinylidene difluoride (PVDF) membranes (Bio-Rad Laboratories, Hercules, CA, USA). 5% skim milk in TBST (Tris-buffer saline with 0.1% tween containing 25 mM Tris-HCl pH 7.5, 125 mM NaCl and 0.1% tween 20) was applied to block the separating blots for 2 h at RT. Membranes were incubated with primary antibodies specific for PARP, Mcl-1, Bcl-2, Bax, and β-Actin at 4 • C overnight, washed with TBST and then were incubated with secondary antibody for 2 h at RT. Finally, the immunoreactive proteins were detected with the enhanced chemiluminescent detection system (Supersignal West Pico, Pierce, Rockford, IL, USA) and subsequently exposed to X-ray film. The intensity of protein bands was analyzed by the ImageJ software (version 1.52, National Institutes of Health, Bethesda, MD, USA). Densitometric values of protein expression levels were calculated as the fold changes relative to β-actin. Detailed information can be found at Supplementary Figures S1-S3.

Computational Mcl-1 modelling and Molecular Docking
Molecular docking was applied to detect an interaction between RT or simplified right-half model of RT compounds and Mcl-1 protein. The target sequence of Mcl-1 (350 amino acids) was retrieved from UniProt, accession code Q07820 [68]. Since the three-dimensional (3D) structure of Mcl-1 has not been available yet, homology modelling, a promising tool to predict 3D structure of protein, was performed using the I-TASSER server [69,70]. The quality of the constructed homology model was then estimated by the Ramachandran plot using PROCHECK server [71]. The information in Figure 6b showed that the majority of amino acids were mainly found in most favored and additional allowed regions with values of 77.8% and 17.6%, respectively, which is suggestive of a reliable quality of this model. Before docking calculations, the modeled structure of Mcl-1 was relaxed by short molecular dynamics simulation for 20 ns using the AMBER16 software package according to standard procedures [72][73][74][75].
To prepare for the docking study, the chemical structures of all studied ligands were built using the Gaussian09 program [76]. Afterward, quantum chemistry calculation with the B3LYP/6-31G* level of theory was used for geometry optimization of all compounds. The docking calculations were carried out with AutoDock Vina [77]. Each ligand was docked to the expected binding site of Mcl-1 (residues 137-143). A grid box size was set to of 20 × 20 × 20 Å, whereas the grid was centered at the position of the residues 137-143, at x, y, z coordinates of 54.0, 31.0, 51.5. The UCSF Chimera package [78] was used for the graphical presentation of the data.

Statistical Analysis
The data from at least three independent replicated experiments (n = 3) was presented as the mean ± standard error of the mean (SEM). Statistical differences between multiple groups were analyzed using an analysis of variance (ANOVA) which calculated SPSS software program version 16 (SPSS Inc., Chicago, IL, USA). Statistical significance was considered at p < 0.05.

Conclusions
This study provides supporting evidence that RT and its simplified right-half model compounds TM-(-)-18 and TM-(-)-4a exert an anti-cancer action through Mcl-1 suppression and in part by the decrease in Bcl-2. Furthermore, by synthesizing structurally modified compounds as analogues of RT, and by performing in vitro protein analysis and molecular docking experiments, we were able to clarify the SAR information of RT compounds, which indicated that the cyanide and benzene ring compositions of RT play key functions in targeting the Mcl-1 protein. RT, TM-(-)-18, and TM-(-)-4a were the active compounds that were demonstrated to have potent anti-cancer activity in lung cancer cells, and information on the SARs of these compounds could encourage the development of related compounds having these groups for Mcl-1 suppression.