15d-PGJ₂ (87893-55-8) and 3-methyladenine (3-MA; 5142-23-4) were purchased from Cayman Chemical Company (Ann Arbor, MI, USA). J11-Cl and J19 were synthesized in-house. The chemical structures of the drugs are presented in Fig. 1A. Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS) and cell culture supplements were obtained from Gibco; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Primary antibodies against SIRT1 (cat. no. 8469; 1:1,000), SIRT2 (cat. no. 12672; 1:1,000), SIRT4 (cat. no. sc-135798; 1:500), SIRT5 (cat. no. 8779; 1:1,000), SIRT6 (cat. no. 8771; 1:1,000), B-cell lymphoma-2 (Bcl-2; cat. no. 15071; 1:500), Bcl-2-associated X protein (Bax; cat. no. 5023; 1:1,000), β-actin (cat. no. 3700; 1:1,000), light chain 3 (LC3; cat. no. 3868; 1:1,000), beclin-1 (cat. no. 4122; 1:1,000), autophagy-related 3 (Atg3; cat. no. 3415; 1:1,000), Atg5 (cat. no. 12994; 1:1,000), Atg7 (cat. no. 8558; 1:1,000), α-tubulin (cat. no. 3873; 1:1,000), cleaved caspase-3 (cat. no. 9661; 1:500), cleaved caspase-9 (cat. no. 7237; 1:1,000), poly(ADP-ribose) polymerase (PARP; cat. no. 9541; 1:1,000) and acetylated p53 (cat. no. 2570; 1:500) were purchased from Cell Signaling Technology (Beverly, MA, USA). Horseradish peroxidase-conjugated secondary antibodies [anti-mouse immunoglobulin G (IgG); cat. no. sc-516102 or anti-rabbit IgG; cat no. sc-2357] were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). All other chemicals were purchased from Sigma-Aldrich; Merck KGaA. All drugs were dissolved in dimethyl sulfoxide (DMSO) and stored at −20°C until use. Chemical agents were diluted to appropriate concentrations with culture medium supplemented with 1% FBS. The final concentration of DMSO was <0.1% (v/v). DMSO was also present in the corresponding controls.

SIRT1 enzymatic activity was assessed using commercial kits (cat. no. ab156065) from Abcam (Cambridge, UK), according to the manufacturer's protocol. First, assay buffer [50 mM Tris/HCl, pH 8.0, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl₂ and 1 mg/ml bovine serum albumin (BSA)], SIRT1 enzyme, and either the solvent dimethylformamide (DMF) or different concentrations of the drugs (15d-PGJ₂, J19 or J11-C1 dissolved in DMF) were mixed with the substrate (p53) and co-substrate (NAD⁺) for 45 min. Deacetylation reactions were conducted at 37°C for 60 min, and stopped by adding 50 µl stop solution containing the developer, followed by incubation at 37°C for 30 min. Fluorescence intensity was determined by reading fluorescence using a SpectraMax M2 microplate reader (Molecular Devices, LLC, Sunnyvale, CA, USA) with an excitation wavelength of 350 nm and an emission wavelength of 450 nm. Calculations of net fluorescence were made after subtracting values for a blank consisting of buffer without NAD.

The X-ray crystal structure of SIRT1 was selected from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (PDB; code 4I5I) and prepared using the protein preparation wizard available in the Glide tool in Maestro (version 10.2; Schrödinger, LLC, New York, NY, USA). During the process, the missing side and back chains were included (16). The protein preparation wizard facility has two components: Preparation and refinement. Following ensuring the chemical accuracy, the preparation component adds hydrogen and neutralizes a side chain that is neither close to the binding cavity nor involved in the formation of salt bridges. The all-atom-optimized potentials for liquid simulations (OPLS-AA) force field was used for this purpose, and then the active site of protein was defined. Glide uses the full OPLS-AA force field at an intermediate docking stage and is considered to be more sensitive to geometrical detail compared with other docking algorithms. The water molecule occupying the protein structure was not suitable for the docking study, therefore it was removed. Finally, the optimization and minimization processes were performed until the average root mean square deviation of the non-hydrogen atoms reached 0.3 Å (17). This was followed by the generation of energy grids using the Glide protocol, as previously described (18-20). The docked ligand was used to define the energy grid boundaries with the default options. These grids were used to score the ligand 'in place' with the XP scoring function. It should be noted that considering the computational protocol followed for the generation of the covalent complexes, the XP Glide Score values can be used only in a qualitative sense to determine the binding of various ligands. The Glide XP scoring functions include improvements to the scoring of hydrogen bonds, the detection of buried polar groups, and the detection of π-cation and π-π stacking interactions. The result of the docking calculation is a top-scored predicted complex that is evaluated using the scoring function. The compounds sirtinol, 15d-PGJ₂, J11-Cl and J19 were prepared using the LigPrep software (version 3.4; Schrödinger, LLC), which can generate a number of structures from each input structure with various ionization states, tautomers, stereochemistries and ring conformations. This process eliminates molecules using various criteria including the molecular mass, specified numbers and types of functional groups present with correct chiralities for each successfully processed input structure. The OPLS 2005 force field was used for the optimization, which produced the low-energy isomer of the ligand (21). Finally, all ligand molecules formed in the complex structure for input were docked.

Human ovarian cancer cell lines (SKOV3 and OVCAR3) and normal kidney epithelial cell lines (HK-2 and NRK-52E were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). SKOV3, OVCAR3, or NRK-52E cells were maintained in DMEM supplemented with 10% FBS, 4 mmol L-glutamine, 100 U/ml penicillin and 100 µg/ml streptomycin (Gibco; Thermo Fisher Scientific, Inc.). HK-2 cells were maintained in DMEM/Ham's F12 (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 5 µg/ml insulin (Gibco; Thermo Fisher Scientific, Inc.), 5 µg/ml transferrin (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin, 100 g/ml streptomycin, 0.1 µmol/l hydrocortisone, 2 nmol/l L-glutamine plus 10% FBS. The cell cultures were incubated at 37°C in a humidified atmosphere containing 5% CO₂.

Cell viability was determined using MTT (5 mg/ml; Sigma-Aldrich; Thermo Fisher Scientific, Inc.). The cells were seeded in 96-well plates at a density of 2×10³ cells/well. Following incubation at 37°C for 24 h, the cells were treated with 15d-PGJ₂ (0.05-20 µM), J11-C1 (1-160 µM) or J19 (1-160 µM) and cultured for a further 24, 48 or 72 h. Following incubation, 15 µl MTT reagent was added to each well and plates were incubated at 37°C for 3 h in the dark. The supernatant was aspirated and formazan crystals were dissolved in 150 µl DMSO at 37°C for 15 min with gentle agitation. The absorbance of each well was measured at 540 nm using a VersaMax microplate reader (Molecular Devices LLC, Sunnyvale, CA, USA). Three independent experiments were performed for each condition and normalized to the absorbance of wells containing medium only (0%) or untreated cells (100%). The half-maximal inhibitory concentration (IC₅₀) values were calculated from the sigmoidal concentration-response curves using SigmaPlot (version 12; Systat Software, Inc., San Jose, CA, USA).

Cells were treated with 15d-PGJ₂ (1, 5 or 10 µM), J11-Cl (1, 5 or 25 µM), or J19 (1, 5 or 25 µM) for 48 h or vincristine sulfate (BML-T117-0005; Enzo Life Sciences, Inc., Farmingdale, NY, USA) at 5 nM as a positive control. The total number of cells, including those in suspension and those adhered to the walls of the wells, were harvested separately to identify sub-G₁ or other cell cycle stages, and were washed in 1% BSA before fixing in 95% ice-cold ethanol containing 0.5% Tween-20 for 1 h at −20°C. The cells (1×10⁶) were washed in a solution containing 1% BSA, stained with ice-cold propidium iodide (PI) solution (10 µg/ml PI and 100 µg/ml RNase in PBS) and incubated in the dark for 30 min at room temperature. The data were acquired and analyzed using a flow cytometer (BD Biosciences, San Jose, CA, USA).

Cells were seeded in 6-well plates (Labtek; Nalge Nunc International, Penfield, NY, USA) at 1×10⁵ cells/well and allowed to attach overnight. After 24 h of incubation at 37°C, the cells were washed with serum-free medium and treated with different concentrations of the drugs in 200 µl medium/well for 48 h. Following induction of apoptosis, the supernatant was collected and the adherent cells (2×10⁶ cells) were trypsinized from the plates. The collected cells were washed twice with PBS and centrifuged at 600 × g for 5 min. Each pellet was resuspended in 50 µl 4-(2-hydroxyethyl)-1-piperzaine-ethanesulfonic acid (HEPES) buffer (10 mM HEPES, 135 mM NaCl and 5 mM CaCl₂). Cells (2×10⁶ cells in 100 µl buffer) were transferred to flow cytometry tubes and 2 µl each of Annexin V-fluorescein isothiocyanate (FITC) and PI (each at 1 mg/ml) were added. Following incubation for 5 min at room temperature in the dark, 400 µl binding buffer was added to each tube. Samples were analyzed using a flow cytometer (Guava^® easyCyte flow cytometer, EMD Millipore, Billerica, MA, USA).

SKOV3 cells were cultured in DMEM at 37°C in a humidified atmosphere containing 5% CO₂. Following incubation for 24 h, the cells were treated with 15d-PGJ₂ (1, 5 or 10 µM), J11-Cl (1, 5 or 25 µM) or J19 (1, 5 or 25 µM), cultured for 48 h, harvested using trypsin digestion, and then washed twice with ice-cold PBS. To isolate total proteins, the cells were first suspended in PRO-PREP™ protein extraction solution (Intron Biotechnology, Inc., Seongnam, Korea). Protein concentrations were determined using a Bicinchoninic Acid protein assay kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA), according to the manufacturer's protocol. Protein samples (20 µg) of the cell extracts were separated by SDS-PAGE (6-15% gels) and transferred onto a polyvinylidene difluoride membrane (EMD Millipore), which was incubated for 1 h in TNA buffer (25 mM Tris/HCl, pH 8.5, 192 mM glycine and 20% methanol) and blocked with 5% skimmed milk powder in PBS. Subsequently, the membrane was incubated with various primary antibodies against SIRTs, Bax, Bcl-2, β-actin, PARP, cleaved caspase-3, cleaved caspase-9, p53, acetylated p53, LC3, beclin-1, Atg3, Atg5 and Atg7 at 4°C overnight. Following washing for 1 h with TNA buffer, the membrane was incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (1:10,000) for 1 h at room temperature. The blots were developed using an Enhanced Chemiluminescence Plus kit (GE Healthcare Life Sciences, Little Chalfont, UK). The data were acquired and analyzed using an ImageSaver6 (ATTO Corp., Tokyo, Japan).

Total RNA was isolated using TRIzol^® reagent (Gibco; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Samples of 2 µg RNA were reverse-transcribed for 50 min at 42°C in a 20 µl reaction mixture containing 1 µl oligonucleotide (dT)₁₅ primer (0.5 µg), 10 mM dNTP mixture, 25 mM MgCl₂ (4 µl), 0.1 M dithiothreitol (2 µl), RNaseOUT inhibitor (1 µl; Invitrogen; Thermo Fisher Scientific, Inc.), Superscript II (50 units) and X10 reverse transcription buffer (2 µl), followed by denaturation at 68°C for 15 min. The cDNAs obtained were further amplified using PCR with the specific primers. The primers sets and PCR conditions are presented in Table I. The number of PCR cycles was estimated in a preliminary study and optimized in the exponential phase of PCR. The PCR products were subjected to electrophoresis on 2% agarose gels and visualized by ethidium bromide staining and UV transillumination (WSE-5600 CyanoView; ATTO Corp.). The molecular sizes of the amplified products were determined by comparison with a molecular mass marker (100 bp DNA ladder; Intron Biotechnology, Inc.) that was run in parallel with the RT-PCR products. Each assay was performed three times.

SKOV3 cells were seeded in T-25 flasks and treated with 15d-PGJ₂ (1, 5 or 10 µM), J11-Cl (1, 5 or 25 µM) or J19 (1, 5 or 25 µM) for 48 h when the cells reached 70% confluence. At the appropriate time points, the cells were treated with 1 µg/ml acridine orange (2.7 µM) in serum-free medium at 37°C for 15 min at room temperature. Following washing with PBS, the formation of acidic vesicular organelles (AVOs) was observed using fluorescence microscopy (FV10i; Olympus Corp., Tokyo, Japan). The cytoplasm and nuclei of the stained cells fluoresced bright green, whereas the acidic AVOs fluoresced bright red. Additionally, green (510-530 nm) and red (650 nm) fluorescence emission from 1×10⁴ cells exposed to blue (488 nm) excitation light was determined using a flow cytometer.

Autophagic vacuoles were also detected by incubation with 50 µmol/l MDC in PBS at 37°C for 10 min. Following incubation, the cells were washed four times with ice-cold PBS and were fixed with 3.75% paraformaldehyde in PBS. The cells were immediately analyzed using a confocal microscope (FV10i) equipped with a filter system (excitation wavelength, 380 nm; emission filter, 525 nm).

Statistical analyses were carried out using GraphPad Prism software (version 5.03; GraphPad Software, Inc., La Jolla, CA, USA). All numerical data are presented as the mean ± standard error of the mean. Statistical analyses were performed using Student's t-test or Mann-Whitney U test. One-way analysis of variance with Tukey's post hoc test to assess differences among specific groups was used for comparison among multiple values. P<0.05 was considered to indicate a statistically significant difference.

Different cancer cell lines were used to screen the in vitro cytotoxic activity of 15d-PGJ₂ and its derivatives. Cell viability was determined using an MTT assay. Human ovarian cancer SKOV3 cells were treated with the indicated concentrations of agents for 48 h. 15d-PGJ₂ significantly decreased cell viability in a concentration-dependent manner with an IC₅₀ of 7.58 µM after 48 h of treatment. Similarly, J11-C1 and J19 markedly affected SKOV3 cell viability in a concentration-dependent manner, but the IC₅₀ values of J11-C1 (17.6 µM) and J19 (83 µM) were much higher compared with that of 15d-PGJ₂ (Fig. 1B). During incubation with 15d-PGJ₂, the cell morphology was altered to an enlarged and elongated form. However, SKOV3 cells also appeared to be more stretched following J11-Cl treatment, with a different cell morphology phenotype from that of the control cells. Following treatment with J19, the cells were initially rounded and shrunken with larger cytoplasm and distinct cellular boundaries (Fig. 1C). The cytotoxicity of 15d-PGJ₂ and its derivatives against the other ovarian cancer cell line used, OVCAR3 cells, were compared. The results indicated that the cytotoxic effect of 15d-PGJ₂, (IC₅₀, 80.7 µM), J11-Cl (49.2 µM) or J19 (20.3 µM) was lower compared with that in SKOV3 cells (Fig. 1D), therefore, SKOV3 cells were selected for our subsequent experiments. The cytotoxic effect of 15d-PGJ₂ and its derivatives against normal cells was examined using an MTT assay. Normal kidney proximal tubule epithelial HK-2 or NRK-52E cells were treated with 15d-PGJ₂ or its derivatives for 48 h. No significant cytotoxicity was observed in HK-2 or NRK-52E cells at 80 µM 15d-PGJ₂ (Fig. 2A), which significantly affected SKOV3 cells, suggesting that the cytotoxic effect of the 15d-PGJ₂ was selective towards cancer cells.

The effect of 15d-PGJ₂ and its derivatives on SIRT protein expression was examined in SKOV3 cells using western blotting with specific antibodies against acetylated SIRT1, SIRT2, SIRT4, SIRT5 and SIRT6. As presented in Fig. 3A, 15d-PGJ₂ markedly decreased the expression of SIRT1, at 1 µM. However, a high concentration (5 µM) of J11-Cl and J19 markedly decreased SIRT1 protein levels (Fig. 3A and B). To confirm the expression of SIRTs in SKOV3 cells, the mRNA levels of SIRT1, SIRT2 and SIRT4 in SKOV3 cells treated with 15d-PGJ2 (1, 5 or 10 µM), J11-Cl (1, 5 or 25 µM), or J19 (1, 5 or 25 µM). As presented in Fig. 3C and D, mRNA levels of SIRT1 and SIRT4 were significantly decreased following 15d-PGJ₂ treatment (5 and 10 µM). In addition, the mRNA levels of SIRT2 were markedly decreased in SKOV3 cells by 25 µM J11-Cl and J19 treatment (Fig. 3C). The effects of 15d-PGJ₂, J19 and J11-C1 on total SIRT1 enzymatic activity were investigated. Sirtinol was used as a reference compound for SIRT1 inhibitor. As presented in Fig. 3E, sirtinol significantly inhibited SIRT1 enzyme activity in a concentration-dependent manner. It was identified that 15d-PGJ₂ and J11-Cl significantly decreased SIRT1 enzymatic activity at high concentrations. Similar to the protein levels, J19 only significantly inhibited SIRT1 enzyme activity at high concentrations (Fig. 3E).

The aims of the molecular docking study were to elucidate whether 15d-PGJ₂, J11-Cl or J19 modulate anticancer targets and identify the actual binding pocket against the molecular target of SIRT1. The orientations and binding affinities of the drugs to SIRT1 (PDB code 4I5I) were identified. The docking results for sirtinol revealed a high docking score of −9.41 to −8.27 and the formation of a hydrogen bond of 1.94 Å (–C=O) to the backbone of Gln³⁴⁵ and π-π stacking bond lengths of 4.7 and 5.2Å with Phe²⁷³ and Phe⁴¹⁴. In the docking pose, the chemical natures of the binding site residues within a radius of 3 Å from the bound compound are presented in Fig. 4A. Similarly, the docking results of 15d-PGJ₂ revealed a docking score of −9.43 to −6.82 and formed three hydrogen bonds of 2.37 (O=C–NH–), 2.09 (–C=O) and 1.77 Å (O=C–NH–) to Asp³⁴⁸ and Ile³⁴⁷ (Fig. 4B). Likewise, the docking results of J11-C1 revealed a docking score of −7.34 to −3.98 and formed two hydrogen bonds with lengths of 1.98 Å (–C=O) to PRo₂₇₁ and of 2.51 Å (O=C–NH–) to Ile₃₄₇ (Fig. 4C). Lastly, the docking results of J19 revealed a low docking score of −4.33 to −2.85 and formed two hydrogen bonds with lengths of 2.18 and 2.24 Å (–NH=C–NH2) to Arg²⁷⁴ (Fig. 4D). In the docking pose, the binding site residues were His³⁶³, Ile⁴¹¹, Ala²⁶², Phe⁴¹³, Phe²⁷³, Phe⁴¹⁴, Phe²⁹⁷, Phe²⁹⁷ and Asn³⁴⁶; thus, the bound drugs exhibited good binding affinity and marked hydrophobic interactions, which may lead to greater stability and activity. The overall interaction of the binding site pocket (within a radius of 3 Å) is summarized in Table II.

SIRT inhibitors exhibited moderate cytotoxicity towards various human cancer cells through the induction of cell cycle arrest at a specific phase. The effects of 15d-PGJ₂ and its derivatives on cell cycle progression were determined using flow cytometry, and 15d-PGJ₂ was identified to significantly increase the number of cells in G₂/M phase after 48 h of incubation (Fig. 5). Similarly, J11-C1 and J19 significantly increased the population of SKOV3 cells in G₂/M phase following treatment for 48 h. Furthermore, the proportion of cells in G₁ phase was decreased following treatment of SKOV3 cells with the drugs (Fig. 5).

To evaluate the apoptotic death of SKOV3 cells following treatment with 15d-PGJ₂, J19 or J11-C1, Annexin V-FITC and PI staining, and western blot analyses were performed. Concentration-dependent increases in apoptotic cell death were observed following treatment with 15d-PGJ₂ and its derivatives (Fig. 6A and B). A significant increase in Bax expression and a parallel decrease in Bcl-2 expression were also observed following treatment with these drugs (Fig 6C). In addition, the expression levels of cleaved caspase-3 and -9, cleaved PARP and acetylated p53 were markedly increased in SKOV3 cells following treatment with high concentration of 15d-PGJ₂ and its derivatives (Fig. 6D and E).

To evaluate autophagic cell death induced by 15d-PGJ₂, J19 and J11-C1, western blot analysis and acridine orange staining were performed. The conversion of the soluble form of LC3-I into the autophagic vesicle-associated form LC3-II is considered to indicate autophagosome formation (22). As presented in Fig. 7A and B, a high concentration of 15d-PGJ₂, J11-Cl and J19 significantly increased the level of LC3-II, whereas unconjugated LC3-I levels were decreased. Similar to LC3-II, beclin-1 levels increased following treatment with 15d-PGJ₂, J11-Cl or J19 in a concentration-dependent manner (Fig. 7A and B). Subsequently, the induction of autophagy was confirmed by acridine orange staining. Acridine orange is a lysotropic dye that accumulates in acidic organelles in a pH-dependent manner (23-26). At neutral pH, acridine orange emits a green fluorescence, but emits bright red fluorescence within acidic vesicles when protonated and becomes trapped within the organelle (2). As presented in Fig. 7C and D, the control cells exhibited primarily green fluorescence and minimal red fluorescence, which indicated a lack of acidic vesicular organelles. However, drug-treated cells exhibited a more marked red fluorescence 48 h post-treatment compared with the control cells. The increased red fluorescence intensity observed in cultured SKOV3 cells following drug treatment indicated enhanced acidification of vesicular organelles (Fig. 7C). Fig. 7D presents the mean fluorescence intensity of the control and drug-treated cells.

To investigate the anticancer mechanism of 15d-PGJ₂, J19 or J11-C1 against autophagic cell death in SKOV3 cells, 3-MA (1 mM), an autophagy-specific inhibitor, was incubated with SKOV3 cells. This agent blocks autophagy by inhibiting phosphoinositide 3-kinase, an enzyme required for autophagy (25). As presented in Fig. 8A, the morphological changes indicated that 3-MA alone was not cytotoxic to SKOV3 cells. However, the combination of 3-MA with 15d-PGJ₂, J19 or J11-C1 decreased cell death compared with individual treatment with 15d-PGJ₂, J19 or J11-C1 (Fig. 8A). To confirm the molecular mechanism and whether autophagic cell death was affected by 3-MA treatment, the expression levels of LC3, beclin-1, Atg5 and Atg7 were determined using western blotting. The expression levels of LC3-II and beclin-1 in 15d-PGJ₂-, J19-, or J11-C1 and 3-MA-treated SKOV3 cells were markedly decreased in SKOV3 cells (Fig. 8B). Next, the induction of autophagy was confirmed using MDC staining. The vital dye, which is commonly used to study autophagy, accumulates in autophagic vacuoles following a combination of ion trapping and specific interactions with vacuole membrane lipids (26,27). As presented in Fig. 8C, the autophagic cell death induced by 15d-PGJ₂ and its derivatives was decreased by co-treatment of SKOV3 cells with 3-MA.