Coronarin D induces apoptotic cell death and cell cycle arrest in human glioblastoma cell line

molecules Article Coronarin D Induces Apoptotic Cell Death and Cell Cycle Arrest in Human Glioblastoma Cell Line Yollanda E. M. Franco 1,2,*,†, Marcia Y. Okubo 3,4,†, Adriana D. Torre 3, Paula P. Paiva 3, Marcela N. Rosa 5, Viviane A. O. Silva 5, Rui M. Reis 5,6,7 , Ana L. T. G. Ruiz 4,8 , Paulo M. Imamura 9, João E. de Carvalho 4,8 and Giovanna B. Longato 1,2 1Research Laboratory in Molecular Pharmacology of Bioactive Compounds. São Francisco University, Bragança Paulista 12916–900, SP, Brazil; [email protected] 2Posgraduate program in Health Science, São Francisco University, Bragança Paulista 12916–900, SP, Brazil 3Chemical, Biological and Agricultural Pluridisciplinary Research Center (CPQBA), University of Campinas–UNICAMP, Paulínia 13148–218, SP, Brazil; [email protected] (M.Y.O.); [email protected] (A.D.T.); [email protected] (P.P.P.) 4Posgraduate program in dentistry, Piracicaba Dental School, University of Campinas, Piracicaba 13 414–903, SP, Brazil; [email protected] (A.L.T.G.R.); [email protected] (J.E.d.C.) 5Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos 14.784–400, SP, Brazil; nr[email protected] (M.N.R.); [email protected] (V.A.O.S.); ruir[email protected] (R.M.R.) 6Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, 4710–057 Braga, Portugal 7ICVS/3B’s–PT Government Associate Laboratory, 4710–057 Braga, Portugal 8Faculty of Pharmaceutical Sciences, University of Campinas, UNICAMP, Campinas 13081–970, SP, Brazil 9Institute of Chemistry, University of Campinas–UNICAMP, P.O. Box 6154, Campinas 13083–970, SP, Brazil; [email protected] *Correspondence: [email protected]; Tel.: +55-11-996672895 †These authors contributed equally to this work. Academic Editor: Roberto Fabiani Received: 25 October 2019; Accepted: 21 November 2019; Published: 9 December 2019   Abstract: Glioblastoma (GBM) is the most frequent and highest–grade brain tumor in adults. The prognosis is still poor despite the use of combined therapy involving maximal surgical resection, radiotherapy, and chemotherapy. The development of more efficient drugs without noticeable side effects is urgent. Coronarin D is a diterpene obtained from the rhizome extract of Hedychium coronarium, classified as a labdane with several biological activities, principally anticancer potential. The aim of the present study was to determine the anti–cancer properties of Coronarin D in the glioblastoma cell line and further elucidate the underlying molecular mechanisms. Coronarin D potently suppressed cell viability in glioblastoma U–251 cell line, and also induced G1 arrest by reducing p21 protein and histone H2AX phosphorylation, leading to DNA damage and apoptosis. Further studies showed that Coronarin D increased the production of reactive oxygen species, lead to mitochondrial membrane potential depolarization, and subsequently activated caspases and ERK phosphorylation, major mechanisms involved in apoptosis. To our knowledge, this is the first analysis referring to this compound on the glioma cell line. These findings highlight the antiproliferative activity of Coronarin D against glioblastoma cell line U–251 and provide a basis for further investigation on its antineoplastic activity on brain cancer. Keywords: coronarin D; glioblastoma; apoptosis; cell cycle arrest; natural products Molecules 2019,24, 4498; doi:10.3390/molecules24244498 www.mdpi.com/journal/molecules Molecules 2019,24, 4498 2 of 16 1. Introduction Comprising over 100 diseases, cancer is characterized by disordered cell growth and tissue invasion that can spread to other regions of the body, leading to metastasis. It is a multifactorial heterogeneous disease and one of the major worldwide causes of mortality [ 1 ]. Among all types of cancers, brain tumors are one of the less prevalent, accounting for about 2% of all types of malignant tumors, but considered one of the most worrying ones [ 2 , 3 ]. As the most frequent (70–75%), gliomas were originally characterized as lesions originated from glial cells, which play a role of support, protection, and nourishment for neurons in the central nervous system (CNS) [ 4 ]. However, stem–like cells within the CNS are now thought to be the cells of origin of several primary brain tumor types, including glioblastomas [ 5 , 6 ]. Gliomas are one of the most fatal tumors, presenting a high mortality rate, 29–35%, of the CNS tumors in adolescents and young adults [ 7 , 8 ]. Among gliomas, glioblastomas (GBM) are the most aggressive and frequent subtype [ 9 , 10 ]. GBM is highly invasive and presents a median survival of only 14.6 months, even after aggressive treatment with surgery, radiation, and chemotherapy. The difficulties in human GBM therapy are due to the pathological characteristic and numerous drug–resistance mechanisms. Temozolomide (TMZ) comprises the standard treatment for glioblastoma, but unlike classic chemotherapeutics, TMZ does not induce DNA damage or misalignment of segregating chromosomes directly. It is a DNA alkylating agent, which leads to base mismatches that initiate futile DNA repair cycles; eventually, DNA strands break, which in turn induces cell death. The addition of TMZ to the standard treatment protocol was hailed as a major breakthrough in GBM therapy. Despite this, patients’ prognosis remains dismal with a five–year overall survival below 10% [ 11 ]. For this reason, more efficient therapeutic approaches are required for enhancing the treatment effect [12]. One of these approaches is based on cell death induction. There are many cell death morphotypes described in the literature [ 13 ]. Searching for new anticancer agents, investigation into cell death mechanisms, such as apoptosis, necrosis, necroptosis or other under–explored forms of cell death, is a significant strategy to afford more selective and efficient drugs. Further, as many clinically–established drugs are based on natural products, there are still many researches focused on finding new compounds from natural products [14]. Native from Asia and the Pacific [ 15 ], Hedychium coronarium (Zingiberaceae family) is an invasive species in Brazil [ 16 ]. Popularly known as white garland–lily, butterfly lily, napoleon, narcissus, Olympia, white ginger or “l í rio do brejo” (in Brazil), the H. coronarium rhizomes are used as a starch source [ 17 ] and in traditional medicine for treatment of inflammation, diabetes, and rheumatic pain, among other uses [ 15 ]. Among pharmacological evaluations, the ethanolic extract of H. coronarium rhizomes induces apoptosis on HeLa cells by promoting cell cycle arresting at G1 phase, upregulating p53, p21, and Bax expression as well as downregulating cyclin D1, cyclin–dependent kinases CDK–4, CDK–6, and Bcl–2 expression [ 18 ]. Moreover, chemical evaluation of these extracts afforded the isolation of several labdane–like diterpenes with anti–inflammatory action [ 19 , 20 ], antiallergic [ 21 ], antibacterial [ 22 ], and cytotoxic effects over A–549– lung cancer, SK–N–SH– human neuroblastoma, MCF–7 breast cancer, and HeLa cervical cancer cell lines [23]. One of these diterpenes, coronarin D, has been reported as a promising antiproliferative and anti–inflammatory agent. Coronarin D inhibits the β –hexosaminidase release in RBL–2H3 cells [ 21 ] in addition to increasing the in vivo inhibition of the acetic acid–induced vascular permeability in mice [ 19 ]. Further, Coronarin D exhibits antiproliferative, pro–apoptotic, anti–invasive, antiangiogenic, antiosteoclast, and anti–inflammatory activity by suppressing NF– κ B and the gene products regulated by this pathway of osteoclastogenesis [ 24 ]. Recently, Coronarin D has been described as inducing apoptosis in human hepatocellular carcinoma (HCC) [ 25 ] and in human oral cancer (OSCC) [ 26 ] through the c–Jun N–terminal kinases (JNK) pathway while it has induced reactive oxygen species–mediated cell death in human nasopharyngeal cancer cells (NPC) through inhibition of p38 mitogen–activated protein kinase (MAPK) and activation of JNK [27]. Molecules 2019,24, 4498 3 of 16 Based on these significant activities, the present study sought to further elucidate the Coronarin D mechanism of action on cell death of the human tumor cell line U–251 (glioblastoma). As far as we know, this is the first report concerning the Coronarin D mechanism of action on glioblastoma cancer cell line. 2. Results 2.1. Isolation and Characterization of Coronarin D Coronarin D (Figure 1) was obtained from the dichloromethane crude extract of Hedychium coronarium rhizomes. The rhizomes were collected by Dr. Paulo Matsuo Imamura and identified at the herbarium of the State University of Campinas (UEC 163701). The identification of Coronarin D was done by comparison of experimental 1 H– and 13 C–NMR data (Figure S1 and Table S1) with those described by Itokawa et al. [28]. Molecules 2019, 24, x 2 of 17 know, this is the first report concerning the Coronarin D mechanism of action on glioblastoma cancer cell line. 2. Results 2.1. Isolation and Characterization of Coronarin D Coronarin D (Figure 1) was obtained from the dichloromethane crude extract of Hedychium coronarium rhizomes. The rhizomes were collected by Dr. Paulo Matsuo Imamura and identified at the herbarium of the State University of Campinas (UEC 163701). The identification of Coronarin D was done by comparison of experimental 1H– and 13C–NMR data (Figure S1 and Table S1) with those described by Itokawa et al. [28]. Figure 1. Coronarin D molecular structure, data from [29]. 2.2. In Vitro Antiproliferative Activity Assay Coronarin D presented an interesting antiproliferative activity (Figure 2, Table 1), with U–251 (glioblastoma), 786–0 (kidney), PC–3 (prostate), and OVCAR–3 (ovary) as the most sensitive ones, and total growth inhibition (TGI) values <50 µM. Figure 1. Coronarin D molecular structure, data from [29]. 2.2. In Vitro Antiproliferative Activity Assay Coronarin D presented an interesting antiproliferative activity (Figure 2, Table 1), with U–251 (glioblastoma), 786–0 (kidney), PC–3 (prostate), and OVCAR–3 (ovary) as the most sensitive ones, and total growth inhibition (TGI) values <50 µM. Molecules 2019,24, 4498 4 of 16 Molecules 2019, 24, x 2 of 17 (a) (b) Figure 2. In vitro antiproliferative activity of (a) Coronarin D and (b) doxorubicin hydrochloride (positive control) after 48 h of treatment. Concentration range: 0.785 – 785 µM for Coronarin D; 0.043 – 43.1 µM for doxorubicin hydrochloride. Human tumor cell lines: U–251 (glioblastoma), MCF7 (breast), NCI–ADR/RES (multidrug resistant ovary), 786–0 (kidney), NCI–H460 (lung, non–small cells tumor), PC–3 (prostate), OVCAR–3 (ovary), HT–29 (colon), K562 (chronic myelogenous leukemia). Human non–tumor cell line: HaCaT (keratinocyte). 10-3 10-2 10-1 100101102103 -100 -75 -50 -25 0 25 50 75 100 Cell Growth (%) Concentration (μM) U251 MCF7 NCI/ADR-RES 786-0 NCI-H460 HT29 K-562 HaCaT PC-3 OVCAR-3 10-3 10-2 10-1 100101102 -100 -75 -50 -25 0 25 50 75 100 Cell Growth (%) Concentration (μM) U251 MCF7 NCI/ADR-RES 786-0 NCI-H460 PC-3 OVCAR-3 HT29 K-562 HaCaT Figure 2. In vitro antiproliferative activity of ( a ) Coronarin D and ( b ) doxorubicin hydrochloride (positive control) after 48 h of treatment. Concentration range: 0.785–785 µ M for Coronarin D; 0.043–43.1 µ M for doxorubicin hydrochloride. Human tumor cell lines: U–251 (glioblastoma), MCF7 (breast), NCI–ADR/RES (multidrug resistant ovary), 786–0 (kidney), NCI–H460 (lung, non–small cells tumor), PC–3 (prostate), OVCAR–3 (ovary), HT–29 (colon), K562 (chronic myelogenous leukemia). Human non–tumor cell line: HaCaT (keratinocyte). The glioma cell line (U–251) was chosen to continue the in vitro experimental procedures, taking into account that the treatment for this tumor type is still scarce and requires alternative therapies, as previously mentioned. Considering the TGI value, the concentrations of 2.5, 5, and 10 µ M were chosen to proceed with the cell cycle; concentrations of 10, 20, and 40 µ M were chosen for the flow cytometry and 40 µM for the Western blot assay. Molecules 2019,24, 4498 5 of 16 Table 1. Antiproliferative effect of Coronarin D and doxorubicin hydrochloride expressed as the concentration required for total growth inhibition (TGI, µM) after 48 h of exposition. Cell Lines Coronarin D Doxorubicin Hydrochloride TGI (µM) TGI (µM) U–251 18.6 ±0.5 2.4 ±0.4 MCF7 105.0 ±5.1 11.6 ±1.8 NCI–ADR/RES 550.1 ±79.9 >43.1 * 786–0 36.4 ±4.2 17.8 ±3.4 NCI–H460 640.8 ±11.3 26.7 ±1.8 PC–3 17.1 ±0.6 21.4 ±4.2 OVCAR–3 41.6 ±2.1 19.2 ±1.9 HT–29 534.1 ±31.8 >43.1 * K562 56.6 ±2.4 10.0 ±1.2 HaCaT 12.9 ±1.7 1.1 ±0.1 TGI (concentration required for total growth inhibition of each cell line) values expressed as mean ± standard error of two independent experiments. *: TGI values higher than the highest experimental concentration. Human tumor cell lines: U–251 (glioblastoma), MCF7 (breast), NCI–ADR/RES (multidrug resistant ovary), 786–0 (kidney), NCI–H460 (lung, non–small cells tumor), PC–3 (prostate), OVCAR–3 (ovary), HT–29 (colon), K562 (chronic myelogenous leukemia). Human non–tumor cell line: HaCaT (keratinocyte). 2.3. Cell Cycle Assay Comparing to untreated U–251 cells, Coronarin D induced cell cycle arrest at G1 phase, in a concentration–dependent way and independent of time exposure (Figure 3). The increasing G1 subpopulation was concomitant with a significant reduction on cell subpopulations at S phase (Figure 3a) and G2/M phase (Figure 3a,b), proportionally to Coronarin D concentration. Figure 3c,d reveal the histogram of the most representative concentration (10 µM). Molecules 2019, 24, x 2 of 17 Figure 3. Quantification of U–251 in phases G1, S, and G2 after (a) 24 h and (b) 48 h of treatment with vehicle (DMSO) and Coronarin D at concentrations of 2.5, 5, and 10 µM. Histograms of the most representative concentration (10 µM) are presented at (c) 24 h and (d) 48 h. The values were expressed as mean ± standard deviation of two replicates of the same experiment. * p < 0.05; ** p < 0.01 and *** p < 0.001. (Two–way ANOVA: Bonferroni). 2.4. Phosphatidylserine (PS) Externalization Assay According to Figure 4, after 12 h of U–251 exposition, the concentrations 20 and 40 µM reduced cell viability and increased the number of cells labeled with annexin V–PE (17.88% and 25.88%, consecutively) and doubly labeled with annexin V–PE/7–AAD (7.30 and 13.00%, consecutively). After 24 h of treatment with Coronarin D at 10, 20, and 40 µM, the cell viability was dramatically reduced in comparison with the control (63.4%, 52.00%, 28.88%) and there was an increase of cells labeled with annexin V–PE (26.32%, 23.18%, 22.75%) and doubly labeled with annexin V–PE/7–AAD (9.50%, 19.42%, 42.00%, consecutively). Coronarin D induces cell death through a concentration–dependent effect—the higher the concentration, the more advanced cell death process. The population of non– viable cells labeled only by 7–AAD did not increase significantly, indicating that the treatments with Coronarin D induced cell death characterized by phosphatidylserine exposure, being a type of programmed cell death. Figure 3. Quantification of U–251 in phases G1, S, and G2 after ( a ) 24 h and ( b ) 48 h of treatment with vehicle (DMSO) and Coronarin D at concentrations of 2.5, 5, and 10 µ M. Histograms of the most representative concentration (10 µ M) are presented at ( c ) 24 h and ( d ) 48 h. The values were expressed as mean ± standard deviation of two replicates of the same experiment. * p<0.05; ** p<0.01 and *** p<0.001. (Two–way ANOVA: Bonferroni). Molecules 2019,24, 4498 6 of 16 2.4. Phosphatidylserine (PS) Externalization Assay According to Figure 4, after 12 h of U–251 exposition, the concentrations 20 and 40 µ M reduced cell viability and increased the number of cells labeled with annexin V–PE (17.88% and 25.88%, consecutively) and doubly labeled with annexin V–PE/7–AAD (7.30 and 13.00%, consecutively). After 24 h of treatment with Coronarin D at 10, 20, and 40 µ M, the cell viability was dramatically reduced in comparison with the control (63.4%, 52.00%, 28.88%) and there was an increase of cells labeled with annexin V–PE (26.32%, 23.18%, 22.75%) and doubly labeled with annexin V–PE/7–AAD (9.50%, 19.42%, 42.00%, consecutively). Coronarin D induces cell death through a concentration–dependent effect—the higher the concentration, the more advanced cell death process. The population of non–viable cells labeled only by 7–AAD did not increase significantly, indicating that the treatments with Coronarin D induced cell death characterized by phosphatidylserine exposure, being a type of programmed cell death. Molecules 2019, 24, x 2 of 17 Figure 4. Percentage of U–251 cells stained with annexin V–PE and 7–AAD after (a) 12 h and (b) 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM concentrations. Dotplots are presented at (c) 12 h and (d) 24 h. The values are expressed as mean ± standard deviation of two replicates of the same experiment. * p < 0.05 and *** p < 0.001. (Two–way ANOVA: Bonferroni). 2.5. Detection of Activated Caspases The results obtained for caspases corroborate with annexin assay. The highest concentrations (20 and 40 µM) led to caspases activation without cell membrane disruption in 14.3% and 12.5% of cells, respectively. The percentage of cells doubly labeled, which means, caspases activation with cell membrane disruption increased for 20 µM concentration and this percentage was even higher for 40 µM concentration (49.7%) (Figure 5). Figure 5. (a) Percentage of U–251 cells stained with SR–VAD–FMK and 7–AAD after 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM concentrations. Dotplots are presented at (b). The values are expressed as mean ± standard deviation of two replicates of the same experiment. * p < 0.05, ** p < 0.01, and *** p < 0.001. (Two–way ANOVA: Bonferroni). 2.6. Mitochondrial Membrane Potential Assay The induction of death by intrinsic apoptosis is usually triggered by some stimulus or stress that leads to a mitochondrial response and may result in the depolarization of its outer membrane. Figure 4. Percentage of U–251 cells stained with annexin V–PE and 7–AAD after ( a ) 12 h and ( b ) 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µ M concentrations. Dotplots are presented at ( c ) 12 h and ( d ) 24 h. The values are expressed as mean ± standard deviation of two replicates of the same experiment. * p<0.05 and *** p<0.001. (Two–way ANOVA: Bonferroni). 2.5. Detection of Activated Caspases The results obtained for caspases corroborate with annexin assay. The highest concentrations (20 and 40 µ M) led to caspases activation without cell membrane disruption in 14.3% and 12.5% of cells, respectively. The percentage of cells doubly labeled, which means, caspases activation with cell membrane disruption increased for 20 µ M concentration and this percentage was even higher for 40 µM concentration (49.7%) (Figure 5). 2.6. Mitochondrial Membrane Potential Assay The induction of death by intrinsic apoptosis is usually triggered by some stimulus or stress that leads to a mitochondrial response and may result in the depolarization of its outer membrane. Untreated cells showed high intracellular fluorescence intensity indicating that mitochondria were able to sequester a greater amount of rhodamine 123, whereas in cells treated with Coronarin D at 20 µ M and 40 µ M for 6, 9, and 12 h, there was an intracellular fluorescence signal reduction (Figure 6a–c), being more Molecules 2019,24, 4498 7 of 16 intense at 12 h of treatment. This result suggests that Coronarin D induces loss of mitochondrial membrane preceding or concomitant with caspase activation and phosphatidylserine externalization. Molecules 2019, 24, x 2 of 17 Figure 4. Percentage of U–251 cells stained with annexin V–PE and 7–AAD after (a) 12 h and (b) 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM concentrations. Dotplots are presented at (c) 12 h and (d) 24 h. The values are expressed as mean ± standard deviation of two replicates of the same experiment. * p < 0.05 and *** p < 0.001. (Two–way ANOVA: Bonferroni). 2.5. Detection of Activated Caspases The results obtained for caspases corroborate with annexin assay. The highest concentrations (20 and 40 µM) led to caspases activation without cell membrane disruption in 14.3% and 12.5% of cells, respectively. The percentage of cells doubly labeled, which means, caspases activation with cell membrane disruption increased for 20 µM concentration and this percentage was even higher for 40 µM concentration (49.7%) (Figure 5). Figure 5. (a) Percentage of U–251 cells stained with SR–VAD–FMK and 7–AAD after 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM concentrations. Dotplots are presented at (b). The values are expressed as mean ± standard deviation of two replicates of the same experiment. * p < 0.05, ** p < 0.01, and *** p < 0.001. (Two–way ANOVA: Bonferroni). 2.6. Mitochondrial Membrane Potential Assay The induction of death by intrinsic apoptosis is usually triggered by some stimulus or stress that leads to a mitochondrial response and may result in the depolarization of its outer membrane. Figure 5. ( a ) Percentage of U–251 cells stained with SR–VAD–FMK and 7–AAD after 24 h of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM concentrations. Dotplots are presented at ( b ). The values are expressed as mean ± standard deviation of two replicates of the same experiment. *p<0.05, ** p<0.01, and *** p<0.001. (Two–way ANOVA: Bonferroni). Molecules 2019, 24, x 2 of 17 Untreated cells showed high intracellular fluorescence intensity indicating that mitochondria were able to sequester a greater amount of rhodamine 123, whereas in cells treated with Coronarin D at 20 µM and 40 µM for 6, 9, and 12 h, there was an intracellular fluorescence signal reduction (Figures 6a– c), being more intense at 12 h of treatment. This result suggests that Coronarin D induces loss of mitochondrial membrane preceding or concomitant with caspase activation and phosphatidylserine externalization. Figure 6. Percentage of cells with high (rhodamine +) and low (rhodamine –) intracellular fluorescence intensity after (a) 6 h, (b) 9 h, and (c) 12 h of treatment with vehicle (DMSO) and Coronarin D at 20 µM and 40 µM. The values were expressed as mean ± standard deviation of two replicates of the same experiment. ** p < 0.01 and *** p < 0.001. (ANOVA Two–way: Bonferroni). 2.7. Measurement of Hydrogen Peroxide (H2O2) Generation The induction of death by intrinsic apoptosis is usually triggered by a stress that leads to depolarization of the outer membrane of mitochondria and the release of reactive oxygen species (ROS), more specifically hydrogen peroxide. This was measured by examining the fluorescence intensity of DCF. The intensity of fluorescence is proportional to intracellular hydrogen peroxide levels [30]. Data suggest that over 80% of the cell population presented high fluorescence intensity (DCF +) after 90 min of treatment with Coronarin D, even at the lowest concentration (10 µM), indicating the presence of H2O2 on these cells (Figure 7). Figure 6. Percentage of cells with high (rhodamine +) and low (rhodamine –) intracellular fluorescence intensity after ( a ) 6 h, ( b ) 9 h, and ( c ) 12 h of treatment with vehicle (DMSO) and Coronarin D at 20 µ M and 40 µ M. The values were expressed as mean ± standard deviation of two replicates of the same experiment. ** p<0.01 and *** p<0.001. (ANOVA Two–way: Bonferroni). Molecules 2019,24, 4498 8 of 16 2.7. Measurement of Hydrogen Peroxide (H2O2) Generation The induction of death by intrinsic apoptosis is usually triggered by a stress that leads to depolarization of the outer membrane of mitochondria and the release of reactive oxygen species (ROS), more specifically hydrogen peroxide. This was measured by examining the fluorescence intensity of DCF. The intensity of fluorescence is proportional to intracellular hydrogen peroxide levels [ 30 ]. Data suggest that over 80% of the cell population presented high fluorescence intensity (DCF +) after 90 min of treatment with Coronarin D, even at the lowest concentration (10 µ M), indicating the presence of H2O2on these cells (Figure 7). Molecules 2019, 24, x 2 of 17 Figure 7. (a) Percentage of cells with high (DCF +) and low (DCF –) 2,7–dichlorofluorescein intracellular fluorescence intensity after 90 min of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM. Histograms are presented at (b). The values are expressed as mean ± standard deviation of two replicates of the same experiment. *** p < 0.001. (ANOVA Two–way: Bonferroni). 2.8. Western Blotting Assay In order to confirm the cell signaling pathway of Coronarin D, some proteins involved with proliferation, cell death and cell cycle were evaluated after 24 h of treatment. Figure 8 revealed a decrease of total ERK protein, cleavage of poly (ADP–ribose) polymerases (PARP) and cleavage of caspases 3, 7, and 9 as well as increase of phosphorylation of ERK and p–H2AX histone. The p21 protein, related with cell cycle arrest, was also overexpressed. Of note, Coronarin D activated caspases 7 and 9, as well as PARP and p21 in a more intense way compared with the chemotherapeutic drug TMZ. Figure 8. Effects of Coronarin D and temozolomide (TMZ) on expression of caspase 9, caspase 7, cleaved–caspase 3, poly (ADP–ribose) polymerase (PARP), p–H2AX, ERK, and p–ERK proteins in glioblastoma cell line (U–251). The α–tubulin was used as a positive control. The values for caspase 7, caspase 9, PARP, p–H2AX, p–ERK, and p21 were expressed as mean ± standard deviation of three replicates of the same experiment. Values for cleaved–caspase 3 and ERK were obtained from one experiment. 3. Discussion Since Coronarin D has shown several biological activities as mentioned before and presents a potential clinical application in cancer therapy, it is important to have a clear understanding of its mechanism of action. In this study, we showed that most of the tumor cell lines evaluated were Figure 7. ( a ) Percentage of cells with high (DCF +) and low (DCF –) 2,7–dichlorofluorescein intracellular fluorescence intensity after 90 min of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µ M. Histograms are presented at ( b ). The values are expressed as mean ± standard deviation of two replicates of the same experiment. *** p<0.001. (ANOVA Two–way: Bonferroni). 2.8. Western Blotting Assay In order to confirm the cell signaling pathway of Coronarin D, some proteins involved with proliferation, cell death and cell cycle were evaluated after 24 h of treatment. Figure 8revealed a decrease of total ERK protein, cleavage of poly (ADP–ribose) polymerases (PARP) and cleavage of caspases 3, 7, and 9 as well as increase of phosphorylation of ERK and p–H2AX histone. The p21 protein, related with cell cycle arrest, was also overexpressed. Of note, Coronarin D activated caspases 7 and 9, as well as PARP and p21 in a more intense way compared with the chemotherapeutic drug TMZ. Molecules 2019, 24, x 2 of 17 Figure 7. (a) Percentage of cells with high (DCF +) and low (DCF –) 2,7–dichlorofluorescein intracellular fluorescence intensity after 90 min of treatment with vehicle (DMSO) and Coronarin D at 10, 20, and 40 µM. Histograms are presented at (b). The values are expressed as mean ± standard deviation of two replicates of the same experiment. *** p < 0.001. (ANOVA Two–way: Bonferroni). 2.8. Western Blotting Assay In order to confirm the cell signaling pathway of Coronarin D, some proteins involved with proliferation, cell death and cell cycle were evaluated after 24 h of treatment. Figure 8 revealed a decrease of total ERK protein, cleavage of poly (ADP–ribose) polymerases (PARP) and cleavage of caspases 3, 7, and 9 as well as increase of phosphorylation of ERK and p–H2AX histone. The p21 protein, related with cell cycle arrest, was also overexpressed. Of note, Coronarin D activated caspases 7 and 9, as well as PARP and p21 in a more intense way compared with the chemotherapeutic drug TMZ. Figure 8. Effects of Coronarin D and temozolomide (TMZ) on expression of caspase 9, caspase 7, cleaved–caspase 3, poly (ADP–ribose) polymerase (PARP), p–H2AX, ERK, and p–ERK proteins in glioblastoma cell line (U–251). The α–tubulin was used as a positive control. The values for caspase 7, caspase 9, PARP, p–H2AX, p–ERK, and p21 were expressed as mean ± standard deviation of three replicates of the same experiment. Values for cleaved–caspase 3 and ERK were obtained from one experiment. 3. Discussion Since Coronarin D has shown several biological activities as mentioned before and presents a potential clinical application in cancer therapy, it is important to have a clear understanding of its mechanism of action. In this study, we showed that most of the tumor cell lines evaluated were Figure 8. Effects of Coronarin D and temozolomide (TMZ) on expression of caspase 9, caspase 7, cleaved–caspase 3, poly (ADP–ribose) polymerase (PARP), p–H2AX, ERK, and p–ERK proteins in glioblastoma cell line (U–251). The α –tubulin was used as a positive control. The values for caspase 7, caspase 9, PARP, p–H2AX, p–ERK, and p21 were expressed as mean ± standard deviation of three replicates of the same experiment. Values for cleaved–caspase 3 and ERK were obtained from one experiment. Molecules 2019,24, 4498 9 of 16 3. Discussion Since Coronarin D has shown several biological activities as mentioned before and presents a potential clinical application in cancer therapy, it is important to have a clear understanding of its mechanism of action. In this study, we showed that most of the tumor cell lines evaluated were sensitive to the treatment with Coronarin D and, among them, U–251 (glioblastoma) was the cell line chosen to continue the evaluation of the mechanism of action of this compound. We demonstrated that Coronarin D induces cell cycle arrest at G1 phase and apoptosis of glioblastoma cells in a concentration–time dependent manner. Many natural products can suppress proliferation by arresting cells at phases in the cell cycle [ 25 ]. The p21 is a small protein with 165 amino acids and belongs to the CIP/Kip family of CDK inhibitors. The p21 can arrest the cell cycle progression in G1/S and G2/M transitions by inhibiting CDK4,6/cyclin–D and CDK2/cyclin–E, respectively [ 31 , 32 ]. In addition, some studies have shown that H2AX is required for p21–induced cell cycle arrest after replication stalling [ 33 ]. The results herein presented indicate that Coronarin D inhibits glioblastoma (U–251) cell growth by inducing cell cycle arrest at G1 phase after increasing expression of p21, likely mediated by the phosphorylation of H2AX. Coronarin D was also able to trigger cell death with the activation of caspases 9, 3, and 7 and phosphatidylserine exposure, characteristics of apoptosis, and with further rupture of the cell membrane [ 13 ]. There was a gradual and time–dependent reduction of the mitochondrial membrane potential (MMP) in U–251 cells treated with Coronarin D, which is a feature of the intrinsic apoptotic pathway that occurs in response to various intracellular stress conditions centered on mitochondria [ 34 ]. The ROS production after the cell treatment with Coronarin D suggests that it can act as a second messenger, signaling to the activation of the apoptotic process, since this can activate effector caspases. In addition, ROS can lead to DNA damage that, in turn, activates the p21 pathway and results in cell cycle arrest [35]. Coronarin D led to an increase in the expression of protein kinase ERK, as well as PARP cleavage. ERK is part of the MAPK family and, when activated, can mediate mechanisms of cell proliferation and apoptosis [ 36 , 37 ]. Some studies have reported that the activation of ERK could be a result of DNA damage that subsequently leads to cell cycle arrest and apoptosis [ 38 , 39 ]. In addition, it is known that intracellular ROS lead to the activation of ERK and subsequent apoptosis [40–43] . Poly (ADP–ribose) polymerases (PARPs) are a family of enzymes involved in cellular homeostasis, including DNA transcription, cell–cycle regulation, and DNA repair [ 44 ]. This protein is really relevant in the apoptosis pathway, because it has a positive regulation in tumors and when it is inactivated leads to the cleavage of caspase 3 that is involved with the apoptosis process. Studies describe that some natural products are responsible for cleaving PARP as well as activating the caspase cascade as a mechanism of action on the induction of apoptosis [ 45 , 46 ]. Relating all these data reported we can suggest that Coronarin D could induce apoptosis involving ROS generation and ERK activation in U–251 cell line through an intrinsic and caspase–dependent pathway. It has been reported in the literature that Coronarin D has pro–apoptotic potential, including potentiation of PARP cleavage and a reduction in the expression of anti–apoptotic gene products, such as apoptosis protein–1, TRAF–2 cellular inhibitory proteins, surviving, and Bcl–2 [ 24 ]. Consistent with our findings it has been shown that Coronarin D triggers apoptosis by activating caspase–dependent proteins and altering the expression of Bcl–2, Bcl–xL, and Bak in human hepatoma cell lines [ 25 ]. Moreover, Dimas et al. and Mahaira et al. demonstrated that compounds that contain labdane–type diterpenes triggered apoptosis in human colon cancer cells and myeloid leukemia cells [47,48]. The results obtained propose that Coronarin D elevates the generation of ROS (H 2 O 2 ), which promotes phosphorylation of H2AX and consequently damage to the DNA. In addition, an increase in the expression of p21 leads cell cycle arrest at the check point between G1 and S. ROS generation also increases ERK phosphorylation and loss of mitochondrial membrane potential that allows the release of cytochrome c (not evaluated in this work), and consequently the cleavage of caspases (9, 3, and 7) and PARP protein. The caspase activation, in turn, leads cell to death through the Molecules 2019,24, 4498 16 of 16 40. Tan, B.-J.; Chiu, G.N.C. Role of oxidative stress, endoplasmic reticulum stress and ERK activation in triptolide–induced apoptosis. Int. J. Oncol. 2013,42, 1605–1612. [CrossRef] 41. Ramachandiran, S.; Huang, Q.; Dong, J.; Lau, S.S.; Monks, T.J. Mitogen–Activated Protein Kinases Contribute to Reactive Oxygen Species–Induced Cell Death in Renal Proximal Tubule Epithelial Cells. Chem. Res. Toxicol. 2002,15, 1635–1642. [CrossRef] 42. Zhuang, S.; Yan, Y.; Daubert, R.A.; Han, J.; Schnellmann, R.G. ERK promotes hydrogen peroxide–induced apoptosis through caspase–3 activation and inhibition of Akt in renal epithelial cells. Am. J. Physiol. 2007 , 292, 440–447. [CrossRef] [PubMed] 43. Wang, X.; Martindale, J.L.; Holbrook, N.J. Requirement for ERK Activation in Cisplatin–induced Apoptosis. J. Biol. Chem. 2000,15, 39435–39443. [CrossRef] [PubMed] 44. Ricks, T.K.; Chiu, H.-J.; Ison, G.; Kim, G.; McKee, A.E.; Kluetz, P. Successes and Challenges of PARP Inhibitors in Cancer Therapy. Front. Oncol. 2015,5, 222. [CrossRef] [PubMed] 45. Wang, S.; Huang, M.; Li, J.; Lai, F.; Lee, H.; Hsu, Y. Punicalagin induces apoptotic and autophagic cell death in human U87MG glioma cells. Acta. Pharmacol. Sin. 2013,30, 1411–1419. [CrossRef] [PubMed] 46. Lv, L.; Zheng, L.; Dong, D.; Xu, L.; Yin, L.; Xu, Y. Dioscin, a natural steroid saponin, induces apoptosis and DNA damage through reactive oxygen species: A potential new drug for treatment of glioblastoma multiforme. Food Chem. Toxicol. 2013,59, 657–669. [CrossRef] [PubMed] 47. Dimas, K.; Demetzos, C.; Vaos, V.; Ioannidis, P.; Trangas, T. Labdane type diterpenes down–regulate the expression of c–myc protein, but not of bcl–2, in human leukemia T–cells undergoing apoptosis. Leuk Res. 2001,25, 449–454. [CrossRef] 48. Mahaira, L.G.; Tsimplouli, C.; Sakellaridis, N.; Alevizopoulos, K.; Demetzos, C.; Han, Z. The labdane diterpene sclareol (labd–14–ene–8, 13–diol) induces apoptosis in human tumor cell lines and suppression of tumor growth in vivo via a p53–independent mechanism of action. Eur. J. Pharmacol. 2011 ,666, 173–182. [CrossRef] 49. Yang, S.H.; Li, S.; Lu, G.; Xue, H.; Kim, D.H.; Zhu, J.-J. Metformin treatment reduces temozolomide resistance of glioblastoma cells. Oncotarget 2016,7, 48. [CrossRef] 50. Silva–Oliveira, R.J.; Melendez, M.; Martinho, O.; Zanon, M.F.; de Souza Viana, L.; Carvalho, A.L. AKT can modulate the in vitro response of HNSCC cells to irreversible EGFR inhibitors. Oncotarget 2017 , 8, 53288–53301. Sample Availability: Samples of the compound Coronarin D are not available from the authors. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).