Novel Indole-based Tambjamine-Analogues Induce Apoptotic Lung Cancer Cell Death through p38 Mitogen-Activated Protein Kinase Activation

This article is featured in Highlights of This Issue, p. 1201

Lung cancer is the most common cancer that threatens human health, accounting for more than 19% of all cancer deaths in 2012 (1). Health systems have standardized clinical procedures for treating lung cancer patients being surgery, conventional platinum-based doublet chemotherapy (e.g., carboplatin/taxol or pemetrexed), and targeted therapies (e.g., erlotinib and other novel anti-EGFR-mutated receptors) the main options (2). Nevertheless, conventional chemotherapy shows low efficacy with important treatment-related toxicity. Therefore, development of novel efficient drugs is urgently needed. In this regard, the diversity of marine environment provides an extraordinary array of biologically active metabolites for the development of new anticancer therapeutics. In late 2007, trabectedin (Yondelis, PharmaMar) became the first marine anticancer drug to be approved in the European Union (3). Over the past decade, there has been an increase in the number of new anticancer lead compounds from marine life that have entered in human clinical trials (4). Nowadays, more than 592 marine compounds are included in the pipeline of modern pharmaceuticals discovery programs, showing promising antitumor and cytotoxic activities (5). The natural alkaloids tambjamines were originally isolated from marine invertebrates including bryozoans, nudibranchs, and ascidians (6). Tambjamines possess a wide spectrum of pharmacological properties and seem to be involved in the chemical defense mechanisms of the organisms from which they derive (6). Structurally, these alkaloids are characterized by a 4-methoxybipyrrole moiety shared by other families of natural products such as the prodiginines (7). It is already well known that prodiginines have proapoptotic activity against several cancer cells including lung, breast, and hematopoietic, showing the ability to overcome the multidrug resistance phenotype (8–10). Tambjamines, like prodiginines, are very efficient anion exchangers (anionophores) in model liposomes promoting both chloride and bicarbonate transport, which are the most abundant anions in biological environments (7, 11, 12). Their anionophoric activity has an impact on cellular ion homeostasis, intracellular pH levels, and cell survival (11, 12). Recently, we have demonstrated that synthetic tambjamine analogues induce intracellular pH acidification and cytotoxicity in lung cancer cells and human-derived cancer stem cells (CSC), and how this is related to their anion transport abilities (13). Cancer stem cells are defined as immortal cells within a tumor that have the ability to perpetuate themselves through self-renewal, to spawn differentiated progeny (non-CSCs) and contribute to acquired chemotherapy resistance in cancer (14). Moreover, their facilitated transport activity triggers hyperpolarization of plasma membrane in lung cancer cells and differentiation and cell death in CSCs, leading to an effective elimination of this tumor subpopulation (13).

Although substantial work has already been done concerning chemical characterization of tambjamines and the impact of their anionophoric properties in cell survival and differentiation, their molecular mechanism of action still remains unclear. Here, we report a detailed study of how cancer cells respond to several highly active indole-based tambjamine analogues at the transcript and protein levels. Moreover, we have elucidated the molecular cell death mechanism induced by these compounds as well as assessed their therapeutic effect in vivo in several lung cancer mouse models.

Compounds 1 to 11 were synthesized by condensation of 5-(1H-indole-2-yl)-3-methoxy-1H-pyrrole-2-carbaldehyde with the appropriate amine using acetic acid as catalyst. Detailed synthetic procedures and characterization data are provided in the Supplementary Information (S1–S36).

The human lung cancer cell lines, adenocarcinoma (A549, CCL185), squamous carcinoma (SW900, HTB59), small-cell carcinoma (DMS53, CRL2062), and large-cell carcinoma (H460, HTB177) were obtained from ATCC. All cell lines were tested and authenticated by ATCC using short tandem repeat analysis. All cell lines were cultured (passage number 10–25) following ATCC recommended media (Biological Industries) supplemented with 10% heat-inactivated FBS (Gibco, Thermo Fisher Scientific Inc.), 100 U/mL penicillin, 100 μg/mL streptomycin, and 2 mmol/L l-glutamine, all from Biological Industries. Cells were grown at 37°C in a humidified incubator (Thermo Fisher Scientific Inc.) with 5% CO₂ atmosphere. The cells were mycoplasma tested using a standard PCR technique after thawing.

This protocol was approved by the local Ethics Committee (PR003/13) and the studies were conducted in accordance with the Declaration of Helsinki ethical guidelines, upon signed informed consent. Fresh human lung cancer tissues were obtained from 20 patients with primary lung cancer (age range 55–81 years) undergoing lobectomy or pneumonectomy at Bellvitge University Hospital from April 2013 to March 2014. Histologic diagnosis was determined based on microscopic features of carcinoma cells. Freshly obtained tumor tissue (within 1–2 hours after surgical removal) was washed in RPMI1640 medium (Biological Industries) containing 100 U/mL penicillin and 100 μg/mL streptomycin. Blood vessels and connective tissue were carefully removed and the cancerous area was then minced into small pieces less than 1 mm³ using a scalpel. Chopped tissue was resuspended in RPMI1640 medium containing collagenase II (Sigma-Aldrich) at a concentration of 200 U/mL and digested for 2 to 4 hours at 37°C in a humidified incubator. The enzymatic digestion was stopped when most of the cells were in a single cell suspension. Following two washes in RPMI1640, cells were transferred into standard tissue culture coated flasks (TPP, Trasadingen) and cultured in the defined keratinocyte-serum free medium (DK-SFM, Gibco, Thermo Fisher Scientific Inc.) supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L l-glutamine, 20 ng/mL EGF (Sigma-Aldrich, Merck KGaA), 10 ng/mL basic-FGF (Sigma-Aldrich, Merck KGaA), 2% B27 (Gibco, Thermo Fisher Scientific Inc.), and 0.25 mg/mL amphotericin B (Sigma-Aldrich, Merck KGaA). All lung cancer patient derived cultures were maintained at 37°C in a humidified incubator (Thermo Fisher Scientific Inc.) with 5% CO₂. Culture medium was changed every 2–3 days. Cells were passaged after detachment with TrypLE Express (Invitrogen, Thermo Fisher Scientific Inc.), when the cells reached 80%–90% confluence. We were able to establish two lung cancer primary cultures (PC), one derived from an adenocarcinoma tumor (PC#8) and the other derived from a squamous cell carcinoma (PC#13), which were used for subsequent viability assays. Lung cancer patient–derived PCs were characterized by immunofluorescence, according to their epithelial or mesenchymal biomarker expression, using cytokeratin-8 (catalog no. IF13, Oncogene, Merck KGaA) for the epithelial phenotype and vimentin (catalog no. 3932, Cell Signaling Technology Inc.) for the mesenchymal phenotype.

Cell viability was evaluated using the methylthiazoletetrazolium (MTT, Sigma-Aldrich, Merck KGaA) colorimetric assay. Cells were harvested (10⁴ cells/well) in 96-well plates in a final volume of 100 μL and allowed to grow overnight. At the following day, vehicle solution (DMSO, Sigma-Aldrich, Merck KGaA) or experimental compounds were added at a single point (10 μmol/L) or at different ranging concentrations (0.8–100 μmol/L for compounds 1 and 2, 1.6–200 μmol/L for cisplatin) to the assay plate. Cells were incubated for 24 hours and after the treatment period, 10 μL of MTT (5 mg/mL) were added and the plates were incubated for 4 hours at 37°C. Crystals were dissolved in 100 μL of DMSO, after which the reading was taken spectrophotometrically at 570 nm using a multiwell plate reader (Multiskan FC, Thermo Fisher Scientific Inc.). Cell viability and inhibitory concentration (IC) values were obtained using GraphPad Prism V5.0 for Windows (GraphPad Software). All data are shown as the mean value ± SD of three independent experiments. Statistical analysis (one-way ANOVA) was carried out with the Statgraphics plus 5.1. Statistical package (Manugistics). Cisplatin was bought from Alfa Aesar (Thermo Fisher Scientific Inc.).

The A549 cell line was seeded at a density of 2 × 10⁵ cells in 60-mm plates and allowed to grow for 24 hours. Subsequently, cells were exposed to the IC₅₀ of the tambjamine analogues 1 or 2 for 6 or 16 hours. Total RNA was extracted and purified using the column-based RNeasy Mini Kit (Qiagen), according to the manufacturer's protocol. Total RNA concentration and purity was checked in a nano spectrophotometer (Implen GmbH) and integrity was analyzed using an Agilent 2100 Bioanalyzer (Agilent Technologies). For the reverse transcription, 1 μg of total RNA was used for cDNA synthesis using a mixture of random hexamers and oligo-dT primers and following the RT² First Strand Kit protocol (Qiagen). The Human RT² RNA Quality Control PCR Array (PAHS-999ZA format A, Qiagen) was used to assess the cDNA quality and to check for genomic DNA contamination in an ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Thermo Fisher Scientific Inc.). For gene expression analysis, cDNAs of control and treated cells were mixed with RT² SYBR ROX mastermix (Qiagen) and dispensed in the RT² Profiler PCR Array of Human Cancer Drug Targets (PAHS-507A format E, Qiagen). After 2-minute centrifugation at 300 × g, the array was placed in an ABI PRISM 7900HT real-time PCR system (Applied Biosystems, Thermo Fisher Scientific Inc.) and the analysis was carried out according to manufacturer's instructions. Dissociation curves (melting curves) were carefully analyzed using SDS software v2.3 and RQ Manager v1.2 (both from Applied Biosystems, Thermo Fisher Scientific Inc.) to choose only the highly specific reaction products in the downstream analysis. The threshold cycle (C_T) values obtained were analyzed in Qiagen Data Analysis Center (Qiagen) to retrieve the fold-regulation values for each gene. The network of most altered genes upon compounds treatment was produced in Cytoscape open-source software (Cytoscape Consortium; ref. 15) with data extracted from Gene Network Central (GNC-Pro, Qiagen).

A549 cells were seeded in 100-mm culture plates (10⁶ cells) and allowed to grow for 24 hours. Afterward, they were exposed to compounds 1, 2 (IC₂₅, IC₅₀, and IC₇₅ values), or 3 (10 μmol/L) for 24 hours. Total protein extracts were obtained from cells by the addition of lysis buffer containing 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate, 50 mmol/L sodium fluoride, 40 mmol/L β-glycerophosphate, 200 μmol/L sodium orthovanadate, 1 mmol/L phenylmethylsulfonyl fluoride, and protease inhibitor cocktail (Roche Diagnostics) in 1× PBS. Protein concentration was determined by BCA protein assay (Pierce, Thermo Fisher Scientific Inc.) using BSA (Pierce, Thermo Fisher Scientific Inc.) as a standard. For Western blot analysis, 30 μg of protein extracts were first separated by SDS-PAGE and transferred to Immobilon-P polyvinylidene difluoride (PVDF) membranes (EMD Millipore, Merck KGaG). Membranes were blocked in either 5% dry milk or BSA, both diluted in TBS–Tween (50 mmol/L Tris–HCl pH 7.5, 150 mmol/L NaCl, 0.1% Tween-20) for 1 hour and then incubated overnight with primary antibodies, according to the manufacturer's instructions. Antibodies were obtained from the following sources: anti-caspase-3 (catalog no. 9662), anti-cleaved-caspase-3 (catalog no. 9661), anti-caspase-9 (catalog no. 9502), anti-PARP1 (catalog no. 9542), anti-Bak (catalog no. 6947), anti-Bax (catalog no. 2772), anti-Bcl-2 (catalog no. 4223), anti-Mcl-1 (catalog no. 4572), anti-phospho-p38 MAPK (Thr180/Tyr182, catalog no. 4511), anti-p38 MAPK (catalog no. 9212), and anti-Survivin (catalog no. 2808) were all obtained from Cell Signaling Technology Inc.; anti-Actin (I-19, catalog no. sc-1616) from Santa Cruz Biotechnology Inc and anti-Vinculin (catalog no. V-4505) from Sigma-Aldrich, Merck KGaA. Antibody binding was detected with goat anti-mouse IgG-HRP (catalog no. sc-2005), goat anti-rabbit IgG-HRP (catalog no. sc-2004), or donkey anti-goat IgG-HRP (catalog no. sc-2020), all from Santa Cruz Biotechnology Inc. and the ECL Detection Kit (Amersham, GE Healthcare). Actin or vinculin were used as gel loading controls. The results shown are representative of Western blot analysis data obtained from at least three independent experiments. Images were captured on an Image Quant LAS 500 (GE Healthcare) and band densitometries were retrieved using the Image Studio Lite software (v5.0, LI-COR Biosciences).

Determination of viable cells (A549 cell line) after different inhibitors treatment was performed using the dual DNA intercalating Fluorescent Dyes Kit MUSE Cell Count & Viability Assay (EMD Millipore, Merck KGaG). In brief, 2 × 10⁵ cells/well in DMEM medium were seeded in a six-well plate. After 24 hours, tambjamine analogues 1 and 2 at higher IC₇₅ value were added, and incubated for 6 hours in a 5% CO₂ cell culture incubator (Thermo Fisher Scientific Inc.). A pan-caspases inhibitor (Z-VAD-FMK; BD Bioscience) or a p38 MAPK inhibitor (SB202190; Cell Signaling Technology Inc.) was added at 20 and 30 μmol/L respectively, 2 hours before compound 1 or 2 exposure. Afterward, cells were collected and were incubated with the Cell Count & Viability reagent for 5 minutes. Viability of treated cells was analyzed using the flow cytometry based MUSE Cell Analyzer (EMD Millipore, Merck KGaA) according to the manufacturer's protocol. Viability was calculated as a percentage related to control cells and results shown were obtained from at least three independent experiments. Statistical analysis (one-way ANOVA) was carried out with the Statgraphics plus 5.1. Statistical package (Manugistics).

Quantitative measurements of reactive oxygen species (ROS) in A549 cells were performed using the MUSE Oxidative Stress Kit (EMD Millipore, Merck KGaG), based on dihydroethidium (DHE), a well-characterized reagent that has extensively been used to detect reactive oxidative species in cellular populations. In brief, 2 × 10⁵ cells/well in DMEM medium were seeded in a six-well plate. Twenty-four hours after seeding, tambjamine analogues 1 and 2 at higher IC₇₅ value were added, and incubated for 6 hours in a 5% CO₂ cell culture incubator (Thermo Fisher Scientific Inc.). A positive control, named tert-butyl hydroperoxide (TBHP, Sigma-Aldrich, Merck KGaA) was used at 2.5 mmol/L for 2 hours, and the nontransporter tambjamine analogue (compound 3) was used as a negative control at 10 μmol/L for 6 hours. Afterward, cells were harvested and processed with the Oxidative Stress Kit according to the manufacturer's protocol. The percentage of ROS-positive and -negative cells was calculated and results shown were obtained from at least three independent experiments. Statistical analysis (one-way ANOVA) was carried out with the Statgraphics plus 5.1. Statistical package (Manugistics).

For the purpose of this article, two murine models have been designed: the subcutaneous and the orthotopic model systems. Five-week-old female Crl:NU-Foxn1nu mice strain (Envigo) were used in this study. All animal studies were approved by the Autonomic Ethic Committee (Generalitat de Catalunya) under the protocol 9111. To generate the subcutaneous xenograft model, mice were subcutaneously implanted with 4.5 × 10⁶ DMS53 cells suspended in a 1:1 solution of RPMI1460:Matrigel (BD Bioscience). For the orthotopic model, subcutaneous xenografts of DMS53 in exponential growth from three different animals were aseptically isolated and placed at room temperature in DMEM supplemented with 10% FBS plus 50 U/mL penicillin and 50 mg/mL streptomycin and the surgical resection tumors were implanted in Crl:NU-Foxn1nu mice following previously reported procedures (16, 17). Briefly, mice were anesthetized with a continuous flow of 1% to 3% isoflurane/oxygen mixture (2 L/minutes) and subjected to right thoracotomy. Mice were situated in left lateral decubitus position and a small transverse skin incision (around 5–8 mm) was made in the right chest wall. Chest muscles were separated by a sharp dissection and costal and intercostal muscles were exposed. An intercostal incision of 2–4 mm on the third or fourth rib on the chest wall was made and a small tumor piece of 2 to 4 mm³ was introduced into the chest cavity. The tumor specimen was deposited between the second and the third lung lobule. Next, the chest wall incision was closed with surgery staples, and finally chest muscles and skin were closed. Mice were inspected twice a week, and monitored for the presence of breathing problems.

Mice bearing homogeneous subcutaneous tumors (approximately 150–200 mm³) were randomly allocated to three treatment groups (n = 7/treatment). For DMS53 orthotopic-derived model, mice were randomly allocated (n = 6/treatment) and the treatment started 30 days after tumor implantation. Compounds 1 and 2 were diluted in 7.5% DMSO/0.8% Tween-80. All treatments were intraperitoneally administrated at a dose of 6 mg/kg in alternating days during 20 days. Tumor growth was recorded two to three times per week starting from the first day of treatment (day 0) and tumor volume (in mm³), estimated according to the formula V × 0.25(ab²)/2, where a is the length or biggest diameter and b is the width or smallest diameter). After the final dose of the treatment, animals were sacrificed and tumors were dissected out and weighed.

Liver and kidney samples were fixed in 4% paraformaldehyde overnight, embedded in paraffin, and cut into sections (5 μm). Paraffin-embedded sections were deparaffinized in xylene and gradually rehydrated for hematoxylin and eosin staining. Stained liver and kidney sections were gradually dehydrated and mounted for hepatotoxicity and nephrotoxicity examination, respectively.

To measure the inhibitory effect of novel tambjamine analogues (1–11; Supplementary Fig. S1) on tumor cell proliferation, we have first performed a single-point MTT assay at 10 μmol/L in four commonly used lung cancer cell lines (A549, DMS53, SW900, and H460). After 24 hours of treatment, most of the compounds showed significant cell viability decrease, being compounds 1 and 2 the most promising anticancer agents (Fig. 1A). Then, IC₅₀ values of the selected compounds were calculated in the same cell lines as well as in both adenocarcinoma and squamous patient-derived PCs (Fig. 1B and Supplementary Figs. S37 and S38). The two selected compounds had similar potency against all the cell lines in the study, showing IC₅₀ values below 10 μmol/L. When compared with cis-diamminedichloroplatinum (II) (CDDP), commonly known as cisplatin, the standard chemotherapeutic agent used for lung cancer treatment in the clinic, the IC₅₀ values of this compound were all much higher than the ones of compounds 1 and 2. Interestingly, lung cancer patient– derived PCs were more sensitive to our compounds, showing IC₅₀ values below 5 μmol/L whereas for cisplatin they were higher than 45 μmol/L (Fig. 1B). This IC₅₀ value for cisplatin is in accordance with published data (18, 19). Therefore, the selected tambjamine analogues displayed better cytotoxic activity against a panel of lung cancer cell lines and PCs than the current clinical standard treatment.

To identify gene expression changes at the cellular level, we have used a profiler array consisting of 84 genes that are known cancer drug targets. The selected compounds 1 and 2 were tested in the adenocarcinoma cell line A549 at their respective IC₅₀ values during two incubation periods (6 and 16 hours, for raw data please check Supplementary File S1). Both compounds have produced changes in several genes from diverse functional groups (Table 1). In general, the observed alterations (higher than 1.5-fold regulation in at least two conditions) were more evident at 16 hours with a clear tendency to genes downregulation. In total, 26 genes were modulated, in which 20 were downregulated by compounds 1 or 2 exposure. Downregulated genes included those related to: apoptosis, cell cycle, growth factors and receptors, protein kinases, topoisomerases, to name a few. Especially relevant for cell viability were genes related to cell death like the antiapoptotic B-cell lymphoma 2 (BLC2) and the inhibitor of apoptosis (IAP) family member BIRC5/Survivin, as well as those related to cell cycle (e.g., CDK1 and CDK2). The remaining six genes were upregulated and related to drug metabolism, protein kinases, transcription factors, cathepsins, and heat shock proteins (Table 1). Concerning genes that are altered more than three-fold regulation in at least two conditions, we have obtained CDC25A, FIGF/VEGFD, IGF1, NFKB1, and PTGS2/Cox-2. Six randomly selected genes were reconfirmed by another methodology (Supplementary Table S1; Supplementary Fig. S39). Similar results were obtained using another lung cancer cell line from squamous carcinoma SW900 (77% gene match; Supplementary Table S2; raw data are present in Supplementary File S1), suggesting that they may be implicated in tambjamine mechanism of action. Interestingly, three genes that were previously altered more than three-fold in A549 cell line (FIGF, IGF, and NFKB1) were not altered, IGF and NFKB1 or inversely regulated, FIGF, in the SW900 cell line. We further analyzed the altered genes using protein–protein interaction data and functional data extracted from Gene Network Central (GNC-Pro, Qiagen). The results showed that compound-modulated genes formed a highly connected network (Fig. 2), suggesting that these genes work as a functional module at the molecular level. The network also showed that NFKB1 and TP53 are hubs (densely connected nodes), which integrate signals from the other nodes. Although these genes are hubs in our system, we did not observe any consistent transcriptional change in both cell line models.

Because we have obtained alterations in genes related to apoptosis and cell cycle and these processes are deeply involved in cell viability, we checked if these mechanisms are activated in our cellular model. We were unable to detect any cell-cycle arrest for either compounds tested, whereas cisplatin produced the well-documented G₂–M phase arrest (Supplementary Fig. S40; ref. 20). To elucidate if apoptosis is the molecular mechanism of cell death induced by these compounds, the expression levels of apoptosis-related proteins were analyzed by Western blot analysis because a decrease in the anti-apoptotic proteins Bcl-2 and survivin was detected in the gene expression array. The adenocarcinoma cell line A549 was treated with compounds 1 or 2 at the IC₂₅, IC₅₀, and IC₇₅ concentrations for 24 hours. The results showed that, at higher concentrations, compounds activated the proteolysis of both initiator pro-caspase-9 (CASP9 gene) and executioner pro-caspase-3 (CASP3 gene; Fig. 3A). Moreover, activation of the final apoptotic executioner PARP was also induced. Similar results were also obtained in the small cell carcinoma cell line DMS53 (Supplementary Fig. S41). In addition, we checked the levels of the apoptotic complexes Bak/Mcl-1 (BAK/MCL1 genes) and Bax/Bcl-2 (BAX/BCL2 genes). The pro-apoptotic effectors of the mitochondrial pathway (Bak and Bax) showed an increase in their protein levels after 24 hours treatment. On the contrary, Mcl-1 and Bcl-2 that are prosurvival members of the Bcl-2 family of proteins presented a decrease on their protein levels after treatment (Fig. 3A–C). The results for Bcl-2 protein expression corroborated the gene expression decrease previously observed in the profiler array (Table 1). Also in accordance with those results, downregulation of the key oncogene BIRC5/Survivin was observed at the protein level by Western blot analysis. A549 cells treated with compounds 1 or 2 (IC₂₅, IC₅₀, and IC₇₅) during 24 hours showed a dose-dependent decrease of survivin, indicating that this protein might play an important role on the cell death fate after compounds exposure (Fig. 3A).

To further investigate the involvement of caspases activation on compounds-mediated apoptosis, A549 cells were treated 2 hours with a broad-spectrum caspase inhibitor, Z-VAD-FMK, and then with compounds 1 or 2 for 6 hours. Cell viability was then assessed by flow cytometry and results showed that the decrease in cell viability induced by compounds treatment was significantly reversed by the pan-caspase inhibitor Z-VAD-FMK (Fig. 3D), corroborating that apoptosis is involved in compounds-induced cell death.

The unbalance of ionic homeostasis produced by tambjamine analogues (13) could be initiating ROS production. Therefore, we tested whether compounds 1 and 2 could promote ROS accumulation in cells. Interestingly, after 6 hours treatment with compounds 1 or 2 (at higher IC₇₅ value), there is a significant ROS formation compared with the nontreated condition or when the nontransporter tambjamine analogue (compound 3) is used (Fig. 4A). The effect of compounds is similar to the commonly used ROS stressor TBHP. The p38 MAPK signaling is one of the stress sensor pathways downstream of ROS formation that plays an essential role in inflammation, cell differentiation, growth, and death. Thus, we have investigated by Western blot analysis, the phosphorylation of p38 MAP kinase after treatment with indole-based tambjamine analogues 1, 2, or 3 in A549 cells. After 24 hours treatment with compounds 1 or 2 (IC₂₅, IC₅₀, and IC₇₅), p38 MAPK (MAPK14 gene) phosphorylation increased in a dose–response manner without any alteration in the total protein expression levels (Fig. 4B). Conversely, compound 3 (nontransporter) had no effect on the p38 MAPK activation. The ROS inducer TBHP was shown before to also activate the p38 MAPK pathway (21). To further investigate whether activation of p38 MAPK was related to cell death, A549 cells were pretreated with the p38 MAPK inhibitor SB202190 (30 μmol/L) for 2 hours before compounds treatment. A significant reversion of cell death induced by compounds 1 and 2 was observed upon inhibitor pretreatment, which indicates that compounds-mediated cell death is upstream regulated by the p38 MAPK signaling pathway (Fig. 4C).

To corroborate the anticancer activity of the indole-based tambjamine analogues seen in several lung cancer cell lines and patient-derived PCs, we evaluated the therapeutic effect of compounds 1 and 2 in established DMS53 human small cell lung carcinoma growing subcutaneous and orthotopically in nude mice models. The DMS53 cell line was selected for their ability to grow well in the flank (for subcutaneous model) and in the lung (for orthotopic model) after cell inoculation. Mice bearing DMS53 xenografts were treated with compounds 1 or 2 (6 mg/kg in saline with 7.5% of DMSO/0.8% Tween-80) or vehicle control (7.5% of DMSO/0.8% Tween-80) in alternating days during 20 days. As shown in Fig. 5A and B, compound 2 produced a significant decrease in tumor growth in subcutaneous models. Furthermore, treatment with compound 2 significantly (P < 0.05) reduced lung cancer tumor progression in the orthotopic lung model (Fig. 5C). Importantly, as a single agent the treatment with compounds 1 or 2 did not produce any obvious toxicity (hepatotoxicity and nephrotoxicity; Fig. 5D), diarrhea, or significant body weight loss (Supplementary Fig. S42).

Chemotherapeutic agents based on natural compounds are a potent source of anticancer drugs. Tambjamine analogues of natural compounds tambjamines have already demonstrated a good pharmacological activity and to have remarkable anticancer effects (6, 7, 11–13). In this study, we analyzed the cytotoxic effect and consequent induced cell death and elucidated the molecular mechanism of action of novel synthesized indole-based tambjamine analogues in lung cancer cells. We have observed a significant cytotoxic effect (less than 10 μmol/L at 24 hours) of the selected compounds in several lung cancer cell lines. Moreover, the selected compounds showed IC₅₀ values remarkably lower than cisplatin, the main chemotherapeutic agent used in the treatment of lung cancer patients.

We have recently demonstrated how tambjamine analogues are able to alter the plasma membrane potential and decrease the intracellular pH in the A549 lung adenocarcinoma cell line (13). Changes in the ion homeostasis of the plasma membrane potential have already been associated with apoptosis-induced cell death in several cell types (S49, Jurkat, HL60, and thymocytes; refs. 22, 23). These observations, in addition to our transcriptomic results and the failure to arrest cell cycle, led us to further investigate the molecular events related to the apoptotic process after cell treatment with these compounds.

The mitochondrial apoptotic pathway is closely regulated by the dynamic equilibrium of a group of proteins belonging to the Bcl-2 family with pro-apoptotic (e.g., Bax and Bak) and anti-apoptotic (e.g., Bcl-2 and Mcl-1) functions (24–26). The disruption of this equilibrium leads to the mitochondrial outer membrane permeabilization (MOMP) with the consequent release of cytochrome-c from the mitochondria and activation of initiator caspase-9. The initiator caspase-9 then activates the effector caspase-3 that cleaves regulatory and structural molecules (e.g., PARP), culminating in the death of the cell (24, 26). In this study, the activation of the intrinsic mitochondrial pathway and the cleavage of pro-caspase-9 and pro-caspase-3, as well as their substrate PARP were observed in lung cancer cells. In addition, the unbalance of the apoptotic pairs (Bcl-2/Bax and Mcl-1/Bak) towards the pro-apoptotic proteins unveils the final fate of these cells (27). In our system, the expression levels of prosurvival members of the Bcl-2 family (Bcl-2 and Mcl-1) decreased in a dose-dependent manner and the pro-apoptotic proteins (Bax and Bak) increased after the treatment. In terms of Bcl-2, this decrease further corroborates our transcriptomic results and shows that this alteration is occurring at the gene level. The IAP proteins (IAPs, e.g., XIAP and survivin) are proteins that maintain the small leakages of cytochrome-c or small activations of death receptors controlled by inhibiting the initiator and effector caspases. Survivin is a member of the IAP family that prevents mitochondrial-dependent apoptosis through the inhibition of caspase-9 and caspase-3 by direct or indirect binding (24, 28). Interestingly, after compounds treatment we have seen a downregulation of survivin, at both the transcript and protein levels, which also contributes to the triggering of the apoptotic cell death. Furthermore, using a pan-caspase inhibitor, we were able to significantly reverse the cytotoxic effect of our compounds. These results clearly indicate that compounds induce cell death through apoptosis by activating apoptosis-related proteins (caspase-9 and caspase-3) and decreasing the levels of anti-apoptotic proteins (Bcl-2, Mcl-1, and survivin).

The ionophoric activity of tambjamine analogues has an impact on cellular ion homeostasis and intracellular pH levels (13), thus, it is feasible that their actions will produce a cellular stress. Several recent works have shown that unbalanced ionic homeostasis is deeply involved in ROS generation (29–31). In our hands, treatment with the selected compounds induced a clear ROS production and this could be mediated by the anion transport because our previous work has shown that these compounds also induce cellular acidification (13). Conversely, the nontransporter compound neither produced pH changes nor generated ROS.

The MAPKs are crucial signaling players in the integration of stress signals and their conversion to cellular responses (32). One major MAPK signaling pathway that is important for detecting cellular stress, including intracellular acidosis and ROS, is the p38 MAPK cascade (32–34). The p38 MAPK-mediated apoptosis leads to caspase activation (35, 36), and several chemotherapeutic drugs have been shown to promote this cascade in order to produce apoptotic cancer cell death (36–42). Moreover, reversion of the apoptotic-mediated cell death produced by chemotherapeutic drugs that de-polymerize (nocodazole, vincristine, and vinblastine) or stabilize (taxol) microtubules was achieved by using p38 MAPK-specific inhibitors (e.g., SB203580 and SB202190). Likewise, genotoxic agents (cisplatin and oxaliplatin) and topoisomerase II inhibitors (doxorubicin) were also able to activate p38 MAPK route (43, 44). Nowadays, a dual-role of this route is known with the action as tumor suppressor or tumor promoter largely depending on the type of cancer and tumor stage (45). Interestingly, our results have shown a robust p38 MAPK activation and using the specific p38 MAPK inhibitor, SB202190, we were able to significantly restore cell viability. Reversion of cell death using the pan-caspase inhibitor or the stress kinase inhibitor is very similar, which foresight that our compounds mediate cell death by apoptosis mainly through the p38 MAPK route.

Moreover, several transcription factors have been implicated in the regulation of both anti-apoptotic proteins, survivin, and Bcl-2: Sp1, p53, NF-κB, and STAT3 (38, 46–48). Although Sp1, NF-κB, and STAT3 are generally implicated in transactivation of Bcl-2 and survivin genes, p53 was shown to be responsible for transcriptional repression of these genes. In addition, these transcription factors are all substrates of the p38 MAPK signaling pathway (49). In fact, several articles connect p38 MAPK activation with a downregulation of survivin and Bcl-2 expressions (39, 40, 50–53). Nevertheless, caution should be taken because other pathways are also involved in the activation/repression of these transcription factors and so, crosstalk with other routes is possible. Further experiments are underway to understand the regulation of these genes, which have a profound impact in the fate of the cells by apoptosis. Our study has also shown the therapeutic effect of our compounds in a preclinical setting using subcutaneous and orthotopic mouse models for lung cancer, which indicates a good potential for pharmaceutical development.

In conclusion, we have shown the cytotoxic effect of novel indole-based tambjamine analogues towards lung cancer cells in vitro and in vivo. We have also identified several gene-expression profile alterations produced by our compounds in lung cancer cells and have shown that the main molecular route of induced cell death is apoptosis, which might be activated by the p38 MAPK through ROS cellular stress induction. Ultimately, by understanding the mechanism of action through which these natural-based small molecules mediate their effect in cancer cells will provide a way to improve future studies of drug efficacy and pharmacodynamics, as well as establish drug–response biomarkers and synergistic drug combinations.

No potential conflicts of interest were disclosed.

Conception and design: L. Korrodi-Gregório, R. Quesada, V. Soto-Cerrato, R. Pérez-Tomás

Development of methodology: P. Manuel-Manresa, L. Korrodi-Gregório, E. Hernando, A. Villanueva, A.M. Rodilla, R. Ramos, V. Soto-Cerrato

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): P. Manuel-Manresa, L. Korrodi-Gregório, E. Hernando, A. Villanueva, D. Martínez-García, A.M. Rodilla, R. Ramos, J. Moya

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P. Manuel-Manresa, L. Korrodi-Gregório, D. Martínez-García, A.M. Rodilla, R. Ramos, R. Quesada, V. Soto-Cerrato, R. Pérez-Tomás

Writing, review, and/or revision of the manuscript: P. Manuel-Manresa, L. Korrodi-Gregório, D. Martínez-García, R. Ramos, M. Fardilha, R. Quesada, V. Soto-Cerrato, R. Pérez-Tomás

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Korrodi-Gregório, R. Quesada, R. Pérez-Tomás

Study supervision: L. Korrodi-Gregório, M. Fardilha, R. Quesada, V. Soto-Cerrato, R. Pérez-Tomás

We thank Beatriz Barroso from CCiTUB (Centres Científics i Tecnoloògics Universitat de Barcelona, Barcelona, Spain) for technical assistance.

This work was supported by grants from the Spanish Government and EU funds through the Fondo de Investigaciones Sanitarias (FIS, project PI13/00089) and from La Marató de TV3 Foundation (project 20132730) to R. Pérez-Tomás. R. Ramos was supported by the Sociedad Española de Neumología y Cirugía Torácica (SEPAR, Project 017/2013), R. Quesada by the Consejeri´a de Educacio´n de la Junta de Castilla y Leo´n (project BU340U13) and by the La Marató de TV3 Foundation (project 20132732) and A. Villanueva by the FIS (project PI13/01339). This work was also supported by an individual grant from FCT ((Fundação para a Ciência e a Tecnologia; SFRH/BPD/91766/2012) to L. Korrodi-Gregório) and a predoctoral fellowship awarded from the Government of Catalonia through L'Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR; FI-DGR 2016) to D. Martínez-García.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.