Manoalide Shows Mutual Interaction between Cellular and Mitochondrial Reactive Species with Apoptosis in Oral Cancer Cells

We previously found that marine sponge-derived manoalide induced antiproliferation and apoptosis of oral cancer cells as well as reactive species generations probed by dichloro-dihydrofluorescein diacetate (DCFH-DA) and MitoSOX Red. However, the sources of cellular and mitochondrial redox stresses and the mutual interacting effects between these redox stresses and apoptosis remain unclear. To address this issue, we examined a panel of reactive species and used the inhibitors of cellular reactive species (N-acetylcysteine (NAC)), mitochondrial reactive species (MitoTEMPO), and apoptosis (Z-VAD-FMK; ZVAD) to explore their interactions in manoalide-treated oral cancer Ca9-22 and CAL 27 cells. Hydroxyl (˙OH), nitrogen dioxide (NO2˙), nitric oxide (˙NO), carbonate radical-anion (CO3˙–), peroxynitrite (ONOO–), and superoxide (O2˙–) were increased in oral cancer cells following manoalide treatments in terms of fluorescence staining and flow cytometry. Cellular reactive species (˙OH, NO2·, ˙NO, CO3˙–, and ONOO–) as well as cellular and mitochondrial reactive species (O2˙–) were induced in oral cancer cells following manoalide treatment for 6 h. NAC, MitoTEMPO, and ZVAD inhibit manoalide-induced apoptosis in terms of annexin V and pancaspase activity assays. Moreover, NAC inhibits mitochondrial reactive species and MitoTEMPO inhibits cellular reactive species, suggesting that cellular and mitochondrial reactive species can crosstalk to regulate each other. ZVAD shows suppressing effects on the generation of both cellular and mitochondrial reactive species. In conclusion, manoalide induces reciprocally activation between cellular and mitochondrial reactive species and apoptosis in oral cancer cells.


Introduction
Oral cancer is a global disease with approximately 710,000 new cases of cancers of the oral cavity and pharynx per year and over 359,000 deaths worldwide [1]. In the world, cancers of the oral cavity and pharynx rank as the 7th more prevalent cancer and rank as the 9th cause of cancer death. To date, the incidence rate of oral and pharynx cancers among Taiwanese men ranks the highest worldwide [1,2]. In Taiwan, cancers of the oral cavity and pharynx were the fourth most prevalent cancer among males [3]. At present, oral cancer is primarily treated by surgical resection, radiation therapy, chemotherapy, or a combination of the above therapies [4]. The anticancer drug development for oral cancer cells is still imperative.
Recently, we reported that manoalide exhibited antiproliferation, apoptosis, and DNA damage effects against oral cancer cells by inducing the cellular reactive species as probed by dichloro-dihydrofluorescein diacetate (DCFH-DA) [12]. However, the DCFH-DA was reported to be unreliable probe to detect H 2 O 2 and other kind of ROS [15,16]. Moreover, superoxide anion was reported to be incapable of crossing the mitochondrial membrane [17]. However, our previous study showed that MitoTEMPO (MT) [18], an mitochondrial superoxide (MitoSOX) inhibitor, suppressed manoalide-induced DNA damages (γH2AX and 8-oxodG) [12], suggesting that MitoSOX may cross the mitochondrial membrane to induce DNA damage in oral cancer cells. Accordingly, the MitoSOX traffic to exit mitochondria is controversial. Therefore, the traffic between manoalideinduced cellular and mitochondrial reactive species remains unclear. It warrants for detailed investigation for the involvement of more different cellular and mitochondrial reactive species after manoalide treatment.
In the present study, we aimed to determine the changes of several types of reactive species using several available probes [19] in oral cancer Ca9-22 and CAL 27 cells following manoalide treatment. Levels of the cellular reactive species such as nitrogen dioxide (NO 2˙) , carbonate radical-anion (CO 3˙-), hydroxyl (˙OH), peroxynitrite (ONOO -), and nitric oxide (˙NO) as well as the cellular and mitochondrial superoxide (O 2˙-) were estimated.
Using the inhibitors for cellular and mitochondrial oxidative stresses (N-acetylcysteine (NAC) and MitoTEMPO (MT)), the sources of cellular and mitochondrial reactive species and its apoptosis-modulating effect in oral cancer cells after manoalide treatment were analyzed. Using the inhibitors for apoptosis (Z-VAD-FMK; ZVAD), the cellular and mitochondrial reactive species-modulating effect of apoptosis in oral cancer cells after manoalide treatment was explored. Therefore, the possibility that manoalide induced the mutual interaction between cellular and mitochondrial reactive species and apoptosis in oral cancer cells were examined in the current study.  [20]. Cell viability for 6 h manoalide treatment (10 μM) was determined by MTS assay [12]. Apoptosis was determined by both annexin V/7-aminoactinmycin D (7AAD) (Strong Biotech Corporation, Taipei, Taiwan) and pancaspase activity (Abcam, Cambridge, UK) [21] assays as previously described.
2.2. Probes for Several Reactive Species. Measurements for several reactive species could be detected using several probes (Sigma, St Louis, MO, USA) as follows [19]. DCFH-DA is a probe for NO 2˙, CO 3˙¯, and˙OH. Hydroxyphenyl fluorescein (HPF) is a probe for˙OH and ONOO-. 4-amino-5methylamino-2 ′ ,7 ′ -difluorofluorescein (DAF-FM) is a probe for˙NO. Dihydroethidium (DHE) and MitoSOX Red are probes for cellular and mitochondria O 2˙¯, respectively [16]. These probes were dissolved in DMSO and all experiments with or without probes had the same concentration of 0.1% DMSO.  [25]. Data were analyzed by Flow Jo (FlowJo LLC, Ashland, OR, USA).

Fluorescence
To evaluate the suppression powder of inhibitors on manoalide-induced reactive species, we use the formula of suppression (fold) to calculate as follows: Suppression fold of reactive species inhibitors = (mean intensity of manoalide / mean intensity of control) / (mean intensity of inhibitors and manoalide / mean intensity of inhibitors), where 2 Oxidative Medicine and Cellular Longevity inhibitors can be NAC, MT, and ZVAD. When manoalide concentration is zero, the suppression (fold) of inhibitor is 1.   (Figure 1(b)), suggesting short-term exposure of manoalide exhibited a detectable cell killing effect to oral cancer cells.

Several Reactive Species Were Differentially Generated in
Oral Cancer Cells after Manoalide Treatment. Using flow cytometry, the levels of several reactive species were measured in manoalide-treated oral cancer cells by using available probes (Figure 2(a)). Since free radicals were short-lived intermediates [16,31], all test probes (DCFH-DA, HPF, DAF-FM, DHE, and MitoSOX Red) were detected in short time (0, 10 min, 1 h, and 6 h). These test probes showed differential increase for their corresponding reactive radicals in a time-dependent manner to oral cancer Ca9-22 and CAL 27 cells after manoalide treatment (10 μM) (Figure 2(b)). Moreover, cellular reactive species (probed by DCFH-DA, HPF, DAF-FM, and DHE) and mitochondrial reactive species (probed by MitoSOX Red) were differentially induced in manoalide-treated oral cancer cells. Since 6 h manoalide treatment (10 μM) showed the highest intensity for all reactive species ranging from 0 to 6 h, the following experiments were performed according to this condition.

Discussion
Drugs with redox-modulating ability have the potential for selective killing on cancer cells [32][33][34]. Manoalide was validated to have this redox-modulating ability for selective killing on oral cancer cells [12]; however, its redox evidence of manoalide relies on DCFH-DA and MitoSOX Red-detected reactive species. Moreover, the DCFH-DA was reported to be unreliable probe to detect H 2 O 2 (˙OH) [15,16]. DCFH-DA also crossdetected NO 2˙a nd CO 3˙¯. Accordingly, more probes detecting other reactive species as indicated in Figure 1 are necessary to clarify the redox-modulating ability of manoalide.
In the present study, we investigated the sources of cellular and mitochondrial oxidative stresses in oral cancer cells after manoalide treatment. Moreover, the interaction among these manoalide-induced reactive species and apoptosis in oral cancer cells were explored.

Cellular Reactive Species May Regulate Mitochondrial
Reactive Species. Based on the finding using the inhibitor pretreatment (NAC) of cellular reactive species, the manoalideinduced cellular reactive species as probed by DCFH-DA, HPF, DAF-FM, and DHE were suppressed (Figures 4-7). Similarly, FasL-stimulated cellular reactive species as probed by dihydrorhodamine (DHR for H 2 O 2 detection), HPF, and DHE were suppressed by NAC in Jurkat cells [35]. Thrombin-induced cellular reactive species as probed by DHE was suppressed by NAC in platelets in vitro [27]. Moreover, NAC pretreatment also suppressed the mitochondrial    Oxidative Medicine and Cellular Longevity

10
Oxidative Medicine and Cellular Longevity    reactive species (MitoSOX) (Figure 8). Similarly, antimycin A [36] and withanolide C [37] -induced MitoSOX generations were suppressed by NAC in oral and breast cancer cells, respectively. Therefore, cellular reactive species may induce mitochondrial reactive species generations in oral cancer cells.

Mitochondrial Reactive Species May Regulate Cellular
Reactive Species. Superoxide may be derived from the sources of NAPDH oxidase (NOX) [38] and mitochondria [39]. For mitochondria, complexes I and III are responsible for continuously producing reactive species during electron transfer [40]. Mitochondrial superoxide was reported to be highly membrane impermeable [41], which was supported by the finding that complex I-dependent superoxide is exclusively fluxed to matrix without escaping from mitochondria to cytoplasm [42]. However, this team also reported that that complex III can release superoxide to both the matrix and outer mitochondrial membrane [42], which may partly release to cytoplasm. Therefore, the traffic of mitochondrial reactive species to exit mitochondria is controversial.
Based on our findings using the inhibitor pretreatment (MT) of mitochondrial reactive species, the manoalideinduced mitochondrial reactive species as probed by Mito-SOX Red were suppressed ( Figure 8). Moreover, MT pretreatment also suppressed cellular reactive species as probed by DCFH-DA, HPF, DAF-FM, and DHE (Figures 4-7). Therefore, manoalide-induced mitochondrial reactive species may induce cellular reactive species generations in oral cancer cells, suggesting that mitochondrial reactive species may exit from mitochondria to cytoplasm to regulate the cellular reactive species.

Both Cellular and Mitochondrial Reactive Species May
Regulate Apoptosis. At 6 h manoalide treatment, apoptosis is triggered in oral cancer Ca9-22 and CAL 27 cells. This manoalide-induced apoptosis was differentially suppressed by NAC, MT, and ZVAD in oral cancer cells (Figure 3). In addition to radical species scavenging (Figures 4-8), both NAC and MT can inhibit apoptosis after manoalide treatment in oral cancer cells. Therefore, cellular and mitochondrial reactive species can induce manoalide-induced apoptosis.

Apoptosis May Regulate Both Cellular and Mitochondrial
Reactive Species. It is well known that reactive species can induce apoptosis. However, the role of apoptosis in the induction of reactive species is rarely investigated. ZVAD, a common pancaspase inhibitor to suppress apoptosis, was used to investigate the modulating effect of apoptosis to reactive species response. For example, ZVAD inhibits etoposideinduced caspase activation and DCFH-DA-detected reactive species generation in cervical cancer HeLa cells [43]. ZVAD inhibits cytosine analogue ferropoptoside N69-induced DCFH-DA-detected reactive species generation in melanoma cells [44]. ZVAD also suppresses oxidized black carbon-induced DCFH-DA detected reactive species generation in lung cancer cells [45]. These studies suggest that sev-eral drug-induced apoptosis can induce DCFH-DA-detected reactive species.
Similarly, we found that manoalide induced a number of cellular (Figures 4-7) and mitochondrial ( Figure 8) reactive species generations, which were suppressed by ZVAD pretreatment in oral cancer cells. It shows that manoalide induces a caspase-dependent reactive species generation in oral cancer cells. These results suggest that apoptosis may trigger both manoalide-induced cellular and mitochondrial reactive species generations. Therefore, manoalide can induce apoptosis as well as cellular and mitochondrial reactive species in oral cancer cells, and they have the reciprocal activation between each other.
Moreover, many drug-induced apoptosis in cancer cell studies [46][47][48][49] also triggered oxidative stress but these studies only relied on the cellular reactive species detection by DCFH-DA. Our finding demonstrates that more cellular and mitochondrial reactive species as probed by DCFH-DA HPF, DAF-FM, DHE, and MitoSOX red also contribute to manoalide-induced redox changes to induce apoptosis. Therefore, multiple kinds of cellular and mitochondrial reactive species are suggested to be considered in drug-induced apoptosis studies.  Figure 9: Expected mechanism of multifaceted inductions for cellular and mitochondrial reactive species to apoptosis on manoalide-treated oral cancer cells. NAC is an inhibitor for cellular reactive species (as probed by DCFH-DA, HPF, DAF-FM, and DHE), MT is an inhibitor for mitochondrial reactive species (as probed by MitoSOX Red), and ZVAD is an inhibitor for apoptosis. We proposed a possible mechanism that manoalide (10 μM, 6 h) can induce (1) cellular and (2) (6) ZVAD, suggesting that (7) cellular and (8) mitochondrial radical species can trigger apoptosis. Interestingly, (6) ZVAD also inhibits both (7) cellular and (8) mitochondrial reactive species, suggesting that apoptosis may induce manoalide-induced cellular and mitochondrial reactive species in oral cancer cells. Therefore, manoalide exhibits reciprocally activation between cellular reactive species, mitochondrial reactive species, and apoptosis in oral cancer cells. Note: arrow and T symbol indicate the activating and inhibiting effects. DCFH-DA is the probe for NO 2˙, CO 3˙¯, anḋ OH. HPF is the probe for˙OH and ONOO-. DAF-FM is the probe for˙NO. DHE and MitoSOX Red are the probes for cellular and mitochondrial O 2˙¯.