The present study was approved by the Institutional Review Board of the Institute of Zoology, Chinese Academy of Sciences and the Third Affiliated Hospital of Kunming Medical University; all tissue samples were obtained with written informed consent from the patients or their families. Tumor and adjacent normal lung tissues were obtained between June and November 2011 from 6 patients with previously untreated NSCLC. These included two men and four women, with a mean age of 58 years (range, 41-68 years). The samples were immediately frozen in liquid nitrogen following surgical resection. The expression of MSI2 at the mRNA level was assayed in The Cancer Genome Atlas (TCGA) transcrip-tome database containing 1,053 NSCLC tumor samples and 109 normal lung specimens. The survival curve was estimated by the Kaplan-Meier method and log-rank test using the Online Survival Analysis Software (30) (http://kmplot.com/analysis/index.php?p=service&cancer=lung).

Largazole was synthesized by our chemistry laboratory (31). The purity of largazole was 98% (determined by reverse-phase high-performance liquid chromatography). PS-341 was obtained from Millennium Pharmaceuticals and chloroquine (CQ) was purchased from Sigma-Aldrich; Merck KGaA. The antibodies used included rabbit anti-human MSI2 (cat. no. ab76148, Abcam; 1:1,000 for western blotting), rabbit anti-human S6K (cat. no. 9202, 1:1,000 for western blotting), rabbit anti-human p-S6K (Thr389) (cat. no. 9205, 1:1,000 for western blotting), rabbit anti-human p-S6 ribosomal protein (Ser240/244) (cat. no. 5364, 1:1,000 for western blotting), rabbit anti-human PARP (cat. no. 9542, 1:1,000 for western blotting), rabbit anti-human p-4E-BP1 (Ser65) (cat. no. 9451, 1:1,000 for western blotting) from Cell Signaling Technology, Inc., mouse anti-human actin (cat. no. A5441, Sigma-Aldrich; Merck KGaA; 1:5,000 for western blotting), rabbit anti-human p-Akt (Ser473) (cat. no. sc-7985-R, 1:500 for western blotting) and rabbit anti-human Akt (cat. no. sc-8312, 1:500 for western blotting) from Santa Cruz Biotechnology, Inc.

The human normal lung epithelial cell line 16HBE was obtained from Clonetics. The C57BL/6 murine Lewis lung carcinoma (LLC) cell line, the human leukemic cell line K562, and the NSCLC cell lines A549, H460, H520 and H1975 (harboring EGFR-L858R/T790M mutations) were obtained from the American Type Culture Collection. The 32Dcl3-Bcr-Abl-T315I (32D-BA-T315I) cell line was obtained by stably infecting 32Dcl3 cells with the MSCV-Bcr-Abl-T315I-IRES/GFP (Bcr-Abl-T315I/GFP) retroviral transducing vector, which was kindly provided by Dr Warren Pear (University of Pennsylvania). The pSRalpha plasmid constructs containing the wild-type and T315I mutant cDNAs of the Bcr-Abl tyrosine kinase were provided by Dr Brian Druker (Oregon Health & Science University). Cells expressing 32Dcl3-Bcr-Abl (32D-BA), 32D-BA-T315I and the imatinib-resistant cell line K562R were established and kept in our laboratory (32). All cell lines were cultured according to the recommended protocols.

Cell viability was determined by the Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc.) assay according to the manufacturer's protocol: Cell suspensions (100 µl) were added into each well of 96-well plates at a density of 5,000 cells/well, the cells were then exposed to the indicated concentrations of largazole for 24 or 48 h, 10 µl of CCK-8 reagent was added to each well, and the cells were incubated at 37˚C for 1‑2 h. The absorbance was measured at 450 nm using a microplate reader (Bio-Tek Instruments, Inc.). Cell apoptosis was analyzed using the Annexin V-FITC Apoptosis Detection kit (BD Biosciences) according to the manufacturer's instructions. The cells were treated with the indicated concentrations of largazole for 48 h and collected, resuspended in Annexin V binding buffer, incubated with Annexin V-FITC and propidium iodide (PI) for 15 min at room temperature in the dark, and tested by flow cytometry using a FACS Calibur flow cytometer (Becton, Dickinson and Company). FACS data were analyzed by BD CellQuest™ Pro 6.0 software (Becton, Dickinson and Company).

Cells were treated with the indicated concentrations of largazole for 12 h, then counted and resuspended in RPMI-1640 containing 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) and 0.3% low-melting-point agarose (Amresco), and plated on the bottom layer in a 35-mm plate containing 0.6% agarose; the experiment was performed in triplicate. After 12 days of culture, the colonies were stained with 0.005% crystal violet solution for 30 min at room temperature (Sigma-Aldrich; Merck KGaA) and counted.

For western blot analysis in cultured cells, NSCLC and CML cells were treated with the indicated protocols, then harvested and lysed in RIPA buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 0.1% sodium dodecyl sulfate, 1% sodium deoxycholate, 1% Triton X-100, 1 mM EDTA, 1 mM Na₃VO₄, 1 mM NaF, 1 mM PMSF and protease inhibitor cocktail. For western blot analysis in tissue specimens, frozen tissues were ground in liquid nitrogen-cooled mortar, tissue powder was lysed on ice in RIPA buffer. The lysates were centrifuged at 12,000 x g for 10 min at 4˚C, the supernatant was dissolved with 5X sample loading buffer containing 250 mM Tris-HCl (pH 6.8), 10% sodium dodecyl sulfate, 50% glycerol, 2.5% bromophenol blue, and 5% β-mercaptoethanol, and boiled for 5 min. Equivalent amounts of protein (30 µg/lane) were separated by 10-15% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. The membranes were blocked with 5% non-fat milk and incubated with the indicated primary and corresponding secondary antibodies, i.e., horseradish peroxidase-conjugated anti-mouse (cat. no. 115-035-003, Jackson ImmunoResearch Laboratories, Inc.; 1:10,000) or anti-rabbit (cat. no. 111-035-003, Jackson ImmunoResearch Laboratories, Inc.; 1:10,000). Immunoreactive bands were visualized by using Luminescent Image Analyzer LSA 4000 (GE Healthcare). Densitometry analysis was performed using ImageJ software (version 1.4.3.67; National Institutes of Health).

Total RNA was prepared with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. RT was carried out using a First Strand cDNA Synthesis kit (Fermentas; Thermo Fisher Scientific, Inc.). The RT steps were as follows: 25˚C for 10 min, 42˚C for 60 min and 70˚C for 10 min. qPCR was performed in CFX™96 Real Time System (Bio-Rad Laboratories, Inc.) using SYBR Premix Ex Taq™ (Takara Biotechnology, Inc.) according to the manufacturer's protocol. The thermocycling conditions of the qPCR step were as follows: 95˚C for 5 min followed by 40 cycles of 95˚C for 15 sec and 60˚C for 30 sec. The primer sequences were as follows: Human GAPDH, forward 5′-GGAGCGAGA TCCCTCCAAAA-3′ and reverse 5′-GGCTGTTGTCATACT TCTCATGG-3′; human MSI2, forward 5′-GTTATCTGCGAA CACAGTAGTG-3′ and reverse 5′-ACCCTCTGTGCCTGT TGGTAG-3′.

The structure of largazole was prepared by ChemBioDraw Ultra 14.0 (PerkinElmer, Inc.) and was converted to 3D model via the Ligprep module in Schrödinger Maestro 10.5 (Schrödinger LLC). We performed docking studies with a reported human MSI2 structure (28). The docking procedure was performed by glide docking in Maestro with a precision level of SP (Standard Precision). The docking results were analyzed by Biovia Discovery Studio 2016 client (Dassault Systèmes Biovia Cc).

The results are expressed as mean ± standard deviation. Two-tailed Student's t-test was used to determine the statistical significance between two groups. One-way ANOVA with Dunnett's or Bonferroni's post hoc test was used for the comparison of multiple groups. P<0.05 was considered to indicate statistically significant differences.

To assess the MSI2 expression in lung cancer, the expression levels of MSI2 in normal human lung epithelial 16HBE cells and four lung cancer cell lines (A549, H460, H520 and H1975) were first analyzed by western blotting and qPCR. Compared with the normal lung epithelial 16HBE cells, the lung cancer cell lines exhibited elevated protein and mRNA levels of MSI2 (Fig. 1A and B). The expression of MSI2 was further examined in lung cancer tissues from 6 patients with NSCLC. As shown in Fig. 1C and D, the protein levels of MSI2 were significantly increased in lung cancer tissues compared with those in adjacent normal lung tissues. Since the sample size of this study was small, we explored the expression levels of MSI2 in TCGA level 3 IlluminaHiseq RNAseqV2 tran-scriptome database containing 1,053 NSCLC tumor samples and 109 normal lung specimens. This analysis revealed that MSI2 expression was significantly higher in LUADs (n=515) and LUSCs (n=501) compared with that in normal lung tissues (Fig. 1E and F). To further investigate the correlation between MSI2 expression and the survival of lung cancer patients, Kaplan-Meier analysis was performed using the online cancer survival analysis database (http://kmplot.com/analysis/) (30). As shown in Fig. 1G, lung cancer patients with higher expression of MSI2 had poorer overall survival (OS) compared with those with lower expression of MSI2. The data mentioned above suggest that MSI2 is involved in the pathogenesis of lung cancer and may be a useful therapeutic target.

Largazole is a natural macrocyclic depsipeptide isolated from the marine cyanobacterium Symploca sp. (33). Our team has successfully synthesized this compound, as well as some analogues, and demonstrated its potent antitumor activity in certain cancers (including lung cancer) at 0.01-1 µM (31,34). By molecular docking analysis, we herein demonstrated that largazole was able to bind to MSI2. A crystal structure of the apo human MSI2 RNA-recognition motif 1 (RRM1) at 1.7 Å resolution was recently obtained (28). Docking studies with this structure were performed and it was reported that largazole can be docked into the dimer structure of MSI2 with most of the compound interacting with one of the MSI2 chains in 6DBP (chain A, red), while the thioester residue falls into the cave formed by residues in chain B (blue; Fig. 2A). The backbone part of largazole mainly interacts with the β-sheets (Arg28-Met31, Phe46-Phe49) of MSI2 (Fig. 2B). As shown in Fig. 2B and C, the valine residue forms an alkyl-alkyl bond with the side chain of Glu29. The methyl group attached at 4,5-dihydrothia-zole forms an alkyl bond with Arg28. A hydrogen bond forms between Lys2 and nitrogen in 4,5-dihydrothiazole. As regards the thiazole ring, a pi-cation interaction forms with the side chain of Arg80. Furthermore, the guanidine group of Arg80 forms a hydrogen bond with the carbonyl group of the HoSer residue. The predicted binding energy of this docking model is -30.2752 kcal/mol.

The data presented above prompted us to determine whether largazole inhibits the expression of MSI2. Western blot analysis demonstrated that treatment with indicated concentrations of largazole for 24 h drastically inhibited the expression of MSI2 in human lung cancer H460, H1975 and A549 cells (Fig. 3A). MSI2 plays a critical role in CML progression (13,14). The effects of this compound were tested on CML cells, and it was found that largazole also repressed the protein expression of MSI2 in K562 CML cells in a concentration- and time-dependent manner (Fig. 3B). Largazole decreased the expression of MSI2 in the K562R cell line, which is resistant to the Bcr-Abl tyrosine kinase inhibitor imatinib (Fig. 3C). In a pair of 32Dcl3 murine cell lines stably expressing wild-type or T315I mutant Bcr-Abl, namely 32D-BA and 32D-BA-T315I cells, treatment with largazole at 0.05-0.2 µM for 24 h inhibited MSI2 expression (Fig. 3D).

To further understand the mechanism underlying largazole-induced downregulation of MSI2, the proteasome inhibitor PS-341 and lysosome inhibitor CQ were used. As shown in Fig. 4A-D, pretreatment with PS-341 or CQ could not block largazole-induced downregulation of MSI2 in H460 and K562 cells. Furthermore, treatment of H460 and H1975 cells with largazole for 24 h also decreased the mRNA levels of MSI2 in a concentration-dependent manner (Fig. 4E and F). Treatment of K562 cells with 0.5 µM largazole for 12-24 h markedly reduced the mRNA levels of MSI2 (Fig. 4G).

Our previous study has demonstrated that largazole markedly inhibits the proliferation of multiple lung cancer cell lines and induces apoptosis of lung cancer A549 cells (34). The present study demonstrated that treatment with largazole significantly repressed the proliferation of H460, LLC, K562, K562R, 32D-BA and 32D-BA-T315I cells in a dose‑dependent manner (Fig. 5A‑C). Largazole significantly suppressed the colony-forming ability of K562 and K562R cells, as assessed by a soft agar colony formation assay (Fig. 5D). Of note, largazole markedly inhibited the proliferation of mononuclear cells that were isolated from the bone marrow of 5 patients with CML (Fig. 5E).

Largazole-induced apoptosis was investigated by Annexin V-FITC/PI double staining, and the results demonstrated that treatment of K562 and K562R cells with largazole at 1 and 2 µM for 48 h significantly increased the percentage of Annexin V⁺ apoptotic cells (Fig. 6A and B). Western blot analysis revealed that largazole caused the cleavage of poly(ADP-ribose) polymerase (PARP) in H460, H1975 (Fig. 6C), K562, K562R (Fig. 6D) and 32D-BA-T315I cells (Fig. 6E), indicating activation of the apoptosis effector caspase-3.

One of the main oncogenic pathways downstream of MSI2 is mTOR complex 1 (mTORC1), which becomes activated upon MSI2 binding to the tumor suppressor phosphatase and tensin homolog (15,35). Accumulating evidence indicates that the mTOR signaling pathway plays a key role in regulating cell metabolism, growth, proliferation and survival (36,37). The effects of largazole on the mTOR signaling pathway were examined. Western blot analysis demonstrated that largazole repressed the phosphorylation of S6K and 4E-BP1, the two best characterized downstream effector molecules of mTORC1 (38), in a concentration-dependent manner in H460, H1975 and A549 cells (Fig. 7A-C). Largazole also inhibited the phosphorylation of S6, a substrate of S6K (Fig. 7B and C). These results indicated that largazole suppresses mTORC1 signaling. mTOR forms two structurally and functionally distinct multi-protein complexes, mTORC1 and mTORC2 (39). mTORC2 phosphorylates Akt on S473 (40). It was then investigated whether largazole exerted an inhibitory effect on the mTORC2 signaling pathway. In all tested cells, largazole markedly suppressed the phosphorylation of Akt (S473), indicating that largazole inhibits mTORC2 signaling (Fig. 7B and C). The aforementioned results suggest that largazole represses cell proliferation, at least in part, by targeting MSI2 to suppress mTOR signaling pathway.