International Journal of

Molecular Sciences Article

Anti-Oncogenic gem-Dihydroperoxides Induce Apoptosis in Cancer Cells by Trapping Reactive Oxygen Species Yuki Kuranaga 1,2 , Nami Yamada 1 , Maiko Kashiwaya 3 , Moeko Nakamura 3 , Lei Cui 3 , Minami Kumazaki 1 , Haruka Shinohara 1 , Nobuhiko Sugito 1,2 , Kohei Taniguchi 1 , Yuko Ito 4 , Tatsushi Nakayama 3 , Bunji Uno 3 , Akichika Itoh 3 and Yukihiro Akao 1, * Received: 5 November 2015; Accepted: 31 December 2015; Published: 8 January 2016 Academic Editor: William Chi-shing Cho 1

2 3

4

*

United Graduate School of Drug Discovery and Information Science, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan; [email protected] (Y.K.); [email protected] (N.Y.); [email protected] (M.K.); [email protected] (H.S.); [email protected] (N.S.); [email protected] (K.T.) Graduate School of Engineering, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan Department of Pharmacy, Gifu Pharmaceutical University, 1-25-4, Daigaku-nishi, Gifu 501-1196, Japan; [email protected] (M.K.); [email protected] (M.N.); [email protected] (L.C.); [email protected] (T.N.); [email protected] (B.U.); [email protected] (A.I.) Department of Anatomy and Cell Biology, Division of Life Sciences, Osaka Medical College, 2-7 Daigaku-machi, Takatsuki, Osaka 569-8686, Japan; [email protected] Correspondence: [email protected]; Tel.: +81-58-230-7607; Fax: +81-58-230-7604

Abstract: Organic gem-dihydroperoxides (DHPs) and their derived peroxides have attracted a great deal of attention as potential anti-cancer agents. However, the precise mechanism of their inhibitory effect on tumors is unknown. To determine the mechanism of the inhibitory effects of DHPs, we examined the effects of DHPs on leukemia K562 cells. As a result, certain DHPs used in this study exhibited growth-inhibitory activity according to a clear structure-activity relationship. The most potent DHP, 12AC3O, induced apoptosis in K562 cells, but not in peripheral blood monocytes (PBMCs) or fibroblast cells. 12AC3O induced apoptosis through the intrinsic mitochondrial pathway and thereafter through the extrinsic pathway. The activity of the former pathway was partly attenuated by a JNK inhibitor. Interestingly, 12AC3O induced apoptosis by trapping a large amount of ROS, leading to an extremely lower intracellular ROS level compared with that in the cells in the steady-state condition. These results suggest that an appropriate level of intracellular ROS was necessary for the maintenance of cancer cell growth. DHPs may have a potential to be a novel anti-cancer agent with minimum adverse effects on normal cells. Keywords: apoptosis; ROS; dihydroperoxide; JNK/MAPK; cancer cell

1. Introduction Reactive oxygen species (ROS) are always generated in the human body as a consequence of respiration and oxygen metabolism, but their accumulation can cause oncogenic damage to cells and tissues [1]. ROS are the most important redox signaling molecules, and are involved in the regulation of the MAPK signaling pathway, which induces the death of cancer cells through accumulating ROS. These pleiotropic effects of ROS are largely accounted for by changes in the thiol/disulfide status of the cell, an important determinant of the cell’s redox status [1]. Hydrogen peroxide, a ROS, also induces apoptotic and necrotic cell death through the induction of oxidative stress in cells [2,3]. In this context, we focused our attention on the effect of the peroxy group; and so we synthesized Int. J. Mol. Sci. 2016, 17, 71; doi:10.3390/ijms17010071

www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2016, 17, 71

2 of 13

Int. J. Mol. Sci. 2016, 17, 71

2 of 13

novel gem-dihydroperoxides (DHPs), which can be easily prepared from commercially available so we synthesized novel gem-dihydroperoxides (DHPs), which can be easily prepared from compounds (Figure 1) [4,5]. DHPs and their derived peroxides are known to have anti-malarial commercially available compounds (Figure 1) [4,5]. DHPs and their derived peroxides are known to and -cancer effects [6–8]. At first, we assumed that the DHPs may induce cancer cell death more have anti-malarial and -cancer effects [6–8]. At first, we assumed that the DHPs may induce cancer effectively peroxide compounds by causing the accumulation ROS in the cells dueintothe their cell deaththan more effectively than peroxide compounds by causing theofaccumulation of ROS double peroxy groups. In the current study, a structure-activity relationship of the cell growth cells due to their double peroxy groups. In the current study, a structure-activity relationship of the inhibitory effect on leukemia was found. Among the DHPs examined, 12AC3O most12AC3O efficiently cell growth inhibitory effectcells on leukemia cells was found. Among the DHPs examined, induced apoptosis. Unexpectedly, the mechanism of apoptotic cell death was due to deprivation most efficiently induced apoptosis. Unexpectedly, the mechanism of apoptotic cell death was due to of ROS in cancerofcells; the apoptotic occurred at least in part through thethrough activation deprivation ROSand in cancer cells; andcell the death apoptotic cell death occurred at least in part the of JNK/MAPK, protein kinase protein [9]. 12AC3O-induced apoptosis wasapoptosis mainly due to a activation ofa stress-activated JNK/MAPK, a stress-activated kinase [9]. 12AC3O-induced was decreased ROStolevel, but the above-mentioned signaling cascade was also induced mainly due a decreased ROS level, but thestress-inducing above-mentioned stress-inducing signaling cascade as also induced as inROS. the case of excessive ROS. These results suggest theROS essential of ROS in of in was the case of excessive These results suggest the essential roles of in theroles maintenance the maintenance of growth-related In addition, 12AC3O andeffect otheronDHPs had growth-related signaling pathways. Insignaling addition,pathways. 12AC3O and other DHPs had no cell growth effect on cellstimulated growth inperipheral concanavalin-A stimulated (PBMCs). peripheral12AC3O blood monocytes (PBMCs). in no concanavalin-A blood monocytes may thus be useful not 12AC3O may thus beagent usefulbut notalso onlyasasananagent anti-cancer agent but also as an agent for and discrimination only as an anti-cancer for discrimination between cancer normal cells, between cancer and normal cells, and for the scavenging of free radicals. and for the scavenging of free radicals.

Figure (DHP) agents agentsused usedininthis thisstudy. study. Figure1.1.Dihydroperoxide Dihydroperoxide (DHP)

Results 2. 2.Results Effects DHPsononCell CellGrowth Growthin inHuman Human Leukemia Leukemia K562 2.1.2.1. Effects of of DHPs K562 Cells Cells Firstly, examined whether DHPs exhibited a growth inhibitory effect human leukemia Firstly, wewe examined whether DHPs exhibited a growth inhibitory effect on on human leukemia K562 K562 As in shown in 2A, Figure 2A, various kinds of DHPs showed anti-cancer activity toward cells. Ascells. shown Figure various kinds of DHPs showed anti-cancer activity toward K562K562 cells at at 24 afterofthe start of treatment. in series, the C-12 thewas activity wastoshown to be 24 cells h after thehstart treatment. Especially,Especially, in the C-12 theseries, activity shown be dependent

Int. J.J. Mol. Mol. Sci. Sci. 2016, 17, 71 Int.

13 33ofof13

dependent on the position of the diperoxide group in the structure. Interestingly, a structure-activity relationship found: inhibition was greater when the number of C on the position of the diperoxidewas group in thegrowth structure. Interestingly, a structure-activity relationship atoms wasgrowth approximately 11 to when 15 and the position of the group was found: inhibition from was greater the number of C atoms wasdiperoxide approximately fromwas 11 approximately from 2of tothe 4. 12AC3O showed greatest anti-cancer activity theshowed DHPs, the but to 15 and the position diperoxide groupthe was approximately from 2 to 4.among 12AC3O it did not show any effect on the theDHPs, growth human concanavalin-A-stimulated greatest anti-cancer activity among butof it did not show any effect on the growth peripheral of human lymphocytes, ASF-4-1 human skin fibroblasts or MCF-10A human normal cells concanavalin-A-stimulated peripheral lymphocytes, ASF-4-1 human skinmammary fibroblastsepithelial or MCF-10A at 24h after themammary start of treatment Thethe same relationship wassame also human normal epithelial(Figure cells at2B–D). 24 h after startstructure-activity of treatment (Figure 2B–D). The observed for cellsrelationship of the human colorectal cancerfor cellcells lineofDLD-1 (data not shown). Thesecell results structure-activity was also observed the human colorectal cancer line indicated that DHPs exhibited growth-inhibiting effect specifically cancer cells. Based on DLD-1 (data not shown). These their results indicated that DHPs exhibited theiron growth-inhibiting effect the IC50 of on thecancer DHPscells. tested and their fortheir use stability, in subsequent experiments. specifically Based on thestability, IC50 of we the chose DHPs 12AC3O tested and we chose 12AC3O Interestingly, its IC50 value was lower Interestingly, than that of leukemia agents than such that as Etoposide and for use in subsequent experiments. its IC50 therapeutic value was lower of leukemia Cytarabine (Table [10].as Etoposide and Cytarabine (Table 1) [10]. therapeutic agents1)such

Figure Figure 2. 2. DHPs DHPs inhibited inhibited cell cell proliferation proliferation of of K562 K562 human human leukemia leukemia cells cells but but had had no no effect effect on on human human peripheral peripheral lymphocytes lymphocytes or or other other normal normal human human cells. cells. (A) (A) Human Human leukemia leukemia K562 K562 cells cells were were treated treated with for 24 24 h. h.Viable Viablecells cellswere were evaluated using trypan-blue staining. DMSO with DHPs DHPs (10 (10 µµM) M) for evaluated byby using trypan-blue staining. DMSO was was used as a control. t-test * p < 0.05, *** p < 0.001 versus the control; (B–D) Human peripheral used as a control. t-test * p < 0.05, *** p < 0.001 versus the control; (B–D) Human peripheral blood blood monocytes (PBMCs, B), human normal diploid fibroblast ASF-4-1cells cells(C), (C),and and human human normal monocytes (PBMCs, B), human normal diploid fibroblast ASF-4-1 normal mammary epithelial cells, MCF10A (D), were treated for 24 h with 12AC3O (5, 10 µM). 12AC3O mammary epithelial cells, MCF10A (D), were treated for 24 h with 12AC3O (5, DHP 10 µ M). DHP had the most growth inhibitory activity. activity. 12AC3O had potent the most potent growth inhibitory Table Table1.1.IC IC5050 values values for for the the C-12 C-12 series series of of DHPs DHPs (using (using K562 K562 cells). cells).

Agent ICIC 50 Value (µM) Agent 50 Value (µM) 12AC2O 1.65 12AC2O 1.65 12AC3O 0.81 12AC3O 0.81 12AC4O 1.69 12AC4O 1.69 12AC5O 2.14 12AC5O 2.14 12AC6O 3.83 12AC6O 3.83 Etoposide 5.30 Etoposide 5.30 Cytarabine 2.10 Cytarabine 2.10 Viable cell ratio of C-12 series of DHP or leukemia therapeutic agent-treated K562 cells at 24. Viable cell ratio of C-12 series of DHP or leukemia therapeutic agent-treated K562 cells at 24.

Int. J. Mol. Sci. 2016, 17, 71 Int. J. Mol. Sci. 2016, 17, 71

4 of 13 4 of 13

2.2. 12AC3O 2.2. 12AC3OInduced InducedApoptotic ApoptoticCell CellDeath Death 12AC3O cells time-dependently time-dependently(Figure (Figure3A). 3A).InInorder ordertoto 12AC3Oinduced inducedgrowth growth inhibition inhibition of of K562 K562 cells examine how the growth of 12AC3O-treated cells was inhibited, we stained the treated cells with examine how the growth of 12AC3O-treated cells was inhibited, we stained the treated cells with Hoechst characteristic chromatin chromatincondensation condensationand and Hoechst33342 33342and andfound foundsome some apoptotic apoptotic cells cells with characteristic fragmentation atat 4h after fragmentation 4h afterthe thestart startofoftreatment treatment(Figure (Figure3B). 3B).The Thenumber numberofofapoptotic apoptoticcells cellsincreased increased in a time-dependent manner (Figure 3C). In3C). order confirm whetherwhether 12AC3O12AC3O could induce in a time-dependent manner (Figure In to order to confirm could apoptotic induce cell death incell other cancer lines, we cell treated Jurkat T-cell leukemia cells leukemia and DLD-1 colorectal cancer apoptotic death in cell other cancer lines, we treated Jurkat T-cell cells and DLD-1 colorectal cancer and cellsthen withperformed 12AC3O aand then a viable cell assay bydye-exclusion using the cells with 12AC3O viable cellperformed assay by using the Trypan-blue Trypan-blue dye-exclusion test As anda result, Hoechst 33342 staining. As agrowth result, inhibition 12AC3O also test and Hoechst 33342 staining. 12AC3O also induced and induced apoptotic growth inhibition and apoptotic chromatin condensation and fragmentation in these cells chromatin condensation and fragmentation in these cells (Figure S1A–C). Morphologically,(Figure the EM S1A–C). the EM study also indicated apoptotic fragmentation of cells the nucleus study alsoMorphologically, indicated the apoptotic fragmentation of thethe nucleus in the treated-K562 (Figurein3E, the arrows). treated-K562 cells of(Figure 3E,autophagosomes black arrows). Inwere some of the (blue cells, arrows). autophagosomes were black In some the cells, observed Thus, 12AC3O observed (blue arrows). Thus, 12AC3O induced mainly apoptosis, and in part autophagy, in induced mainly apoptosis, and in part autophagy, in the early phase; however, later apoptoticthe cell earlydominated. phase; however, later apoptotic cell death dominated. death

Figure3.3. 12AC3O 12AC3O induced cells as estimated by morphological examination. (A) Figure inducedapoptosis apoptosisininK562 K562 cells as estimated by morphological examination. Viable cellcell ratio of 12AC3O (10 (10 µ M)-treated K562K562 cells cells up toup 24 to h after treatment. The (A) Viable ratio of 12AC3O µM)-treated 24 h the afterstart theofstart of treatment. viable cells were counted overover timetime by using trypan-blue staining. t-test *t-test p < 0.05, p < 0.001 The viable cells were counted by using trypan-blue staining. * p *** < 0.05, *** p versus < 0.001 the control; (B,C)(B,C) The The morphological characteristics of apoptosis in K562 cells were observed byby versus the control; morphological characteristics of apoptosis in K562 cells were observed < 0.01, *** p < 0.001 fluorescence microscopy using Hoechst 33342 (5 µ g/µ L). Bar is 50 µ m. t-test ** p fluorescence microscopy using Hoechst 33342 (5 µg/µL). Bar is 50 µm. t-test ** p < 0.01, *** p < 0.001 versus thecontrol; control;(D) (D)Morphological Morphological characteristics characteristics of versus the of apoptosis apoptosisin inK562 K562cells cellsobserved observedbybyelectron electron microscopy; (E) The black arrows indicate fragmented nuclei; and the blue arrows, microscopy; (E) The black arrows indicate fragmented nuclei; and the blue arrows,autophagosomes. autophagosomes.

Next, we surveyed the expression of signaling proteins associated with apoptosis by Next, we surveyed the expression of signaling proteins associated with apoptosis by performing performing Western blotting. As a result, the activation of caspase-8 continued from 4 to 24 h after Western blotting. As a result, the activationofofPARP caspase-8 start the start of treatment. The cleaved-forms werecontinued observed from from 42 to h 24 up htoafter 12 hthe after oftreatment; treatment.and The cleaved-forms of PARP were observed from 2 h up to 12 h after treatment; caspase-9 was also activated at the time corresponding to the cleavage of PARP and caspase-9 also activated at the time to the of remained PARP (Figure 4A). (Figure 4A). was However, the expression levelscorresponding of pro-apoptotic Baxcleavage and Bim almost However, the expression levels pro-apoptotic Bax and cells Bim remained unchanged. At 24 h unchanged. At 24 h after the of start of treatment, some that were almost not sensitive to 12AC3O after the start of treatment, some cells that were not sensitive to 12AC3O continued to proliferate. continued to proliferate. The expression level of Bcl-2 became elevated at 24 h. The expression level The expression level of Bcl-2 and became elevatedform at 24couldn’t h. Thebeexpression level of BID remained of BID remained unchanged, its truncated detected (Figure 4A). In order to unchanged, and that its truncated form couldn’t be detected 4A). In order to further certify further certify 12AC3O induced apoptosis, we used(Figure the caspase inhibitor Z-VAD (MBL).

Int. J. Mol. Sci. 2016, 17, 71

5 of 13

Int. J. Mol. Sci. 2016, 17, 71

5 of 13

that 12AC3O induced apoptosis, we used the caspase inhibitor Z-VAD (MBL). Pre-incubation with Pre-incubation Z-VAD inhibited the increase in the number of apoptotic K562 cells after Z-VAD clearly with inhibited theclearly increase in the number of apoptotic K562 cells after treatment of them treatment of them with (Figure the 4B).bands Biochemically, the bands cleaved-form by PARP with 12AC3O (Figure 4B).12AC3O Biochemically, of cleaved-form PARPofdemonstrated the demonstrated the treatment with 12AC3O were significantly attenuated byZ-VAD the pre-treatment treatment withby 12AC3O were significantly attenuated by the pre-treatment with (Figure 4C). with (Figure the 4C).mitochondrial Next, we examined the mitochondrial membrane byMito-Tracker. staining the Next,Z-VAD we examined membrane potential by staining thepotential cells with cells with4D, Mito-Tracker. In Figure 4D, the photo shows that potential the mitochondrial In Figure the bottom photo shows that thebottom mitochondrial membrane was clearlymembrane decreased potential was clearly decreased the cellsatafter treatment with 12AC3O 4 h compared with that in the cells after treatment within 12AC3O 4 h compared that afterattreatment with DMSO as after treatment with DMSO as the control. These findings taken together indicate that 12AC3O the control. These findings taken together indicate that 12AC3O induced apoptotic cell death mainly induced apoptotic cell death mainly signal pathway and that was the through the intrinsic apoptotic signal through pathwaythe andintrinsic that the apoptotic extrinsic apoptotic signal pathway extrinsic apoptotic signal was subsequently activated to execute apoptosis completely. subsequently activated topathway execute apoptosis completely.

Profile of ofintracellular intracellularsignaling signalingpathways pathwaysinin12AC3O-treated 12AC3O-treated K562 cells. Expression Figure 4. Profile K562 cells. (A)(A) Expression of of apoptosis-related proteins after treatment with 12AC3O µM) 8, or 12 24 or h, 24as h, assessed as assessed apoptosis-related proteins after treatment with 12AC3O (10(10 µ M) for for 2, 4,2,8,4,12 by by Western analysis; Viable ratios of 12AC3O-treated 4, 8, Western blotblot analysis; (B) (B) Viable cell cell ratios of 12AC3O-treated (10 µ(10 M)µM) K562K562 cellscells at 2, at 4, 2, 8, 12 or12 24or h 24 h after pre-treatment Z-VAD, a caspase inhibitor, (C)Change Changein in protein protein expression after pre-treatment withwith Z-VAD, a caspase inhibitor, forfor 2424h;h;(C) cleaved form. t-test *** p***< p0.001 12AC3O-treated versusversus 12AC3O-treated K562 < 0.001 12AC3O-treated 12AC3O-treated profiles of ofPARP PARPand anditsits cleaved form. t-test cells in theinpresence of Z-VAD; (D) Measurement of mitochondrial membrane potential in K562 cells K562 cells the presence of Z-VAD; (D) Measurement of mitochondrial membrane potential in K562 after treatment with 12AC3O (10 µM) at 4 h by using Mito-tracker. Blue fluorescence indicates positive cells after treatment with 12AC3O (10 µ M) at 4 h by using Mito-tracker. Blue fluorescence indicates Hoechst Hoechst 33342 nuclear positive 33342staining. nuclear staining.

2.3. The SAPK/JNK Was Up-Regulated in the Early Phase of Treatment Treatment with with 12AC3O 12AC3O effect of 12AC3O on MAP kinaseskinases and theand growth-related PI3K/Akt PI3K/Akt signaling To understand understandthe the effect of 12AC3O on MAP the growth-related pathway in the treated K562 cells, we performed Western blotting analysis. As to MAP kinases, signaling pathway in the treated K562 cells, we performed Western blotting analysis. As to MAP pErk/Erk was weakly activated; however, pp38/p38 and pJNK/JNK werewere activated untiluntil 4 h and kinases, pErk/Erk was weakly activated; however, pp38/p38 and pJNK/JNK activated 4h then then became gradually inactive from 8from h after start the of treatment (Figure 5A). Next, 5A). we pre-treated and became gradually inactive 8 the h after start of treatment (Figure Next, we cells with the JNK-IN-8 inhibitor Chemicals) in order to the role the of JNK in pre-treated cells with the JNK JNK-IN-8 JNK(EMD inhibitor (EMD Chemicals) in validate order to validate role of 12AC3O Interestingly, the apoptotic death was significantly by JNK in induced-apoptosis. 12AC3O induced-apoptosis. Interestingly, thecell apoptotic cell death wassuppressed significantly the treatment JNK-IN-8 at 1 µM (Figureat 5B), reflected the decreased level cleaved-form suppressed bywith the treatment with JNK-IN-8 1 µwhich M (Figure 5B), which reflected the of decreased level PARP. In contrast,PARP. the level PARP was 5C). increased These findings indicate JNK played of cleaved-form In of contrast, theincreased level of (Figure PARP was (Figure 5C). that These findings a key role in JNK the apoptosis induced bythe 12AC3O. Theinduced PI3K/Akt pathway was signaling activated indicate that played a key role in apoptosis by signaling 12AC3O. The PI3K/Akt until 8 h and on 8 h8 up 24 inactivated h, which could survival pathway wasinactivated activated until h to and on reflect 8 h upa compensatory to 24 h, which could signaling reflect a against 12AC3O. compensatory survival signaling against 12AC3O.

Int. J. Mol. Sci. 2016, 17, 71 Int. J. Mol. Sci. 2016, 17, 71

6 of 13 6 of 13

Figure Figure5.5. Growth-related Growth-related signaling signaling pathways pathwaysinin12AC3O-treated 12AC3O-treatedK562 K562cells. cells. (A) (A)Time-dependent Time-dependent protein proteinexpression expressionprofiles profilesofofgrowth-related growth-relatedsignaling signalingofofMAPK MAPKand andPI3K/Akt PI3K/Aktinin12AC3O-treated 12AC3O-treated K562 DMSOwas wasused usedasas a control; Viable of 12AC3O-treated cells 4, K562 cells. cells. DMSO a control; (B)(B) Viable cellcell ratioratio of 12AC3O-treated K562K562 cells at 4, 8,atand 8,24and 24 h after pre-treatment with JNK-IN-8, a JNK inhibitor, for 24 h; (C) Change in the protein h after pre-treatment with JNK-IN-8, a JNK inhibitor, for 24 h; (C) Change the protein expression expressionprofiles profilesof ofPARP PARPand andits itscleaved cleavedform. form.t-test t-test* *p p

Anti-Oncogenic gem-Dihydroperoxides Induce Apoptosis in Cancer Cells by Trapping Reactive Oxygen Species.

Organic gem-dihydroperoxides (DHPs) and their derived peroxides have attracted a great deal of attention as potential anti-cancer agents. However, the...
NAN Sizes 0 Downloads 10 Views