Antiproliferation Effects of Marine-Sponge-Derived Methanol Extract of Theonella swinhoei in Oral Cancer Cells In Vitro

The purpose of this study aimed to assess the antiproliferation effects of methanol extract of T. swinhoei (METS) and explore the detailed responses of oral cancer cells compared to normal cells. METS effectively inhibits the cell proliferation of oral cancer cells but does not affect normal cell viability, exhibiting preferential antiproliferation function. METS exerted more subG1 accumulation, apoptosis induction, cellular and mitochondrial oxidative stress, and DNA damage than normal cells, reverted by oxidative stress inhibitor N-acetylcysteine. This METS-caused oxidative stress was validated to attribute to the downregulation of glutathione. METS activated both extrinsic and intrinsic caspases. DNA double-strand breaks (γH2AX) and oxidative DNA damage (8-hydroxy-2-deoxyguanosine) were stimulated by METS. Therefore, for the first time, this investigation shed light on exploring the functions and responses of preferential antiproliferation of METS in oral cancer cells.


Introduction
Oral cancer patients show high morbidity and mortality [1] and show high incidence worldwide [2]. Surgery, radiation, and chemotherapy are common clinical therapies for oral cancer, but chemoradiation occasionally generates side effects [3]. It is necessary to identify new anticancer drugs for oral cancer treatment.

METS Preparation
The sponge Theonella swinhoei was collected by scuba diving in Orchid Island, Taitung County, Taiwan, in April 2011. A voucher specimen (OISP-1) was deposited at the Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung, Taiwan. The animal material was extracted by ethanol thrice to provide a crude extract (469.0 g). This extract was aliquoted into ethyl-acetate-soluble and water-soluble fractions, and the former portion was further partitioned between 75% methanol (aq) and hexanes. The 75% methanol soluble layer (43.9 g) of T. swinhoei was termed METS.

Cell Cultures and Reagents
Oral cancer cell lines such as tongue-derived CAL 27 (ATCC, Manassas, VA, USA) and Gingiva-derived Ca9-22 (HSRRB, Ibaraki, Osaka, Japan) were used. To assess the drug safety of METS, a non-malignant cell line derived from gingival epithelial Smulow-Glickman (S-G) was chosen as normal control cells [28][29][30]. The culture medium was prepared by mixing DMEM with F12 (3:2) (Gibco, Grand Island, NY, USA), 10% fetal bovine serum, and P/S antibiotics [31]. Cells were seeded at 4 × 10 4 /well for a 12-well plate for growth overnight and subjected to drug treatment for flow cytometry experiments.

Cell Viability Assay
MTS reagent (Promega Corporation, Madison, WI, USA), a tetrazolium dye, was used to detect cell viability. Ca9-22, CAL 27, and S-G cells were plated at 4, 4, and 6 × 10 3 /well for a 96-well plate. Subsequently, cells were incubated overnight. Finally, cells were then treated with drugs for 24 h. Finally, MTS reagents were added and incubated for 1 h before ELISA reader detection (490 nm) [35].

Cytometric Cell Cycle Assay
Cell cycle phases are proportional to the DNA content, which was stained using 1 µg/mL of 7-aminoactinmycin D (7AAD) (Biotium, Hayward, CA, USA) for 30 min for 75% ethanol fixed cells [36]. DNA content was assessed using a flow cytometer (Guava easyCyte, Luminex, TX, USA), and cell cycle phases were analyzed using Flow Jo 10 software (Becton-Dickinson; Franklin Lakes, NJ, USA).

Statistical Analysis
The one-way ANOVA analysis and post hoc test (JMP software, SAS Institute Inc., Cary, NC, USA) provides the connecting letters for determining the significant difference for multiple comparison. Connecting letters without an overlap indicate a significant difference, while it differs non-significantly when the connecting letters are overlapping. Examples were mentioned in the figure legend to explain significance.

HPLC Analysis of METS
HPLC-PDA fingerprint profiles of METS and the isolated major compound, theonellapeptolide 1d, were provided ( Figure 1A). The retention time of theonellapeptolide 1d was found at 14.205 min, which overlapped the major peak of METS. The linear equations (y = 10 7 x − 79228, R 2 = 0.9995) of theonellapeptolide 1d was deduced by the HPLC peak area in four different concentrations ( Figure 1B). In METS, theonellapeptolide 1d accounts for 21.0% of the whole amount.

Statistical Analysis
The one-way ANOVA analysis and post hoc test (JMP software, SAS Institute Inc., Cary, NC, USA) provides the connecting letters for determining the significant difference for multiple comparison. Connecting letters without an overlap indicate a significant difference, while it differs non-significantly when the connecting letters are overlapping. Examples were mentioned in the figure legend to explain significance.

HPLC Analysis of METS
HPLC-PDA fingerprint profiles of METS and the isolated major compound, theonellapeptolide 1d, were provided ( Figure 1A). The retention time of theonellapeptolide 1d was found at 14.205 min, which overlapped the major peak of METS. The linear equations (y = 10 7 x − 79228, R 2 = 0.9995) of theonellapeptolide 1d was deduced by the HPLC peak area in four different concentrations ( Figure 1B). In METS, theonellapeptolide 1d accounts for 21.0% of the whole amount.       Figure 1B (Ca9-22 cells), the lower-case letters for METS at 0, 1, 4, and 6 μg/mL are "a, b, cd, and e", revealing significant differences between each other. By contrast, the lower-case letters for METS at 1, 2, and 3 μg/mL are "b, bc, and bc", indicating non-significant differences since they overlap with the same letter "b".

METS Causes More subG1 Accumulation to Oral Cancer Cells Than Normal Cells
The subG1 events of histograms were used to measure primarily apoptosis. METS caused the accumulation of subG1 events in oral cancer cells (Ca9-22 and CAL 27) under dose and time course experiments ( Figure 3A,B). CAL 27 cells showed a higher extent of subG1 events than Ca9-22 cells. In contrast, METS exhibited lower subG1 events in normal S-G cells than oral cancer cells. Accordingly, METS exerts a greater extent of subG1 in oral cancer than in normal cells.
The G1 events were increased, and G2/M events were decreased at 6 μg/mL of METS for Ca9-22 cells. By contrast, the G1 events were reduced, and G2/M events were increased at 6 μg/mL of METS for CAL 27 and S-G cells. Accordingly, METS exerts differential cell cycle disturbance for various cell lines.
Additionally, NAC was used to test the contribution of ROS in cell cycle disturbance caused by METS. NAC inhibited subG1 increment of METS-treated oral cancer cells (  , the lower-case letters for METS at 0, 1, 4, and 6 µg/mL are "a, b, cd, and e", revealing significant differences between each other. By contrast, the lower-case letters for METS at 1, 2, and 3 µg/mL are "b, bc, and bc", indicating non-significant differences since they overlap with the same letter "b".

METS Causes More subG1 Accumulation to Oral Cancer Cells than Normal Cells
The subG1 events of histograms were used to measure primarily apoptosis. METS caused the accumulation of subG1 events in oral cancer cells (Ca9-22 and CAL 27) under dose and time course experiments ( Figure 3A,B). CAL 27 cells showed a higher extent of subG1 events than Ca9-22 cells. In contrast, METS exhibited lower subG1 events in normal S-G cells than oral cancer cells. Accordingly, METS exerts a greater extent of subG1 in oral cancer than in normal cells.
The G1 events were increased, and G2/M events were decreased at 6 µg/mL of METS for Ca9-22 cells. By contrast, the G1 events were reduced, and G2/M events were increased at 6 µg/mL of METS for CAL 27 and S-G cells. Accordingly, METS exerts differential cell cycle disturbance for various cell lines.
Additionally, NAC was used to test the contribution of ROS in cell cycle disturbance caused by METS. NAC inhibited subG1 increment of METS-treated oral cancer cells ( Figure 3B), revealing that oxidative stress modulated METS-triggered subG1 increment. NAC inhibited G1 increment and G2/M decrement of METS-treated Ca9-22 cells. In contrast, NAC inhibited G1 decrement and G2/M increment of METS-treated CAL 27 cells. For normal S-G cells, NAC enhanced G1 decrement and G2/M increment at 6 µg/mL of 24 h METS treatment.

METS Causes More Annexin V-Based Apoptosis to Oral Cancer Cells than Normal Cells
SubG1 increment is an apoptosis-like change, and it needs further validation for apoptosis. The annexin V (+) events of histograms were used to measure apoptosis. Annexin V (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) ( Figure 4). However, it indicated lower annexin (+) events in normal S-G cells than in oral cancer cells by METS treatment. Accordingly, METS exerts a greater extent of apoptosis induction in oral cancer cells than normal cells.
Additionally, NAC was used to test the contribution of ROS in annexin V increment caused by METS. NAC inhibited annexin V increment of METS-treated oral cancer cells ( Figure 4B), particularly for 24 h. It reveals that oxidative stress modulated METStriggered apoptosis. Antioxidants 2022, 11,1982 6 of 21

METS Causes More Annexin V-Based Apoptosis to Oral Cancer Cells Than Normal Cells
SubG1 increment is an apoptosis-like change, and it needs further validation for apoptosis. The annexin V (+) events of histograms were used to measure apoptosis. Annexin V (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) ( Figure 4). However, it indicated lower annexin (+) events in normal S-G cells than in oral cancer cells by METS treatment. Accordingly, METS exerts a greater extent of apoptosis induction in oral cancer cells than normal cells. Additionally, NAC was used to test the contribution of ROS in annexin V increment caused by METS. NAC inhibited annexin V increment of METS-treated oral cancer cells ( Figure 4B), particularly for 24 h. It reveals that oxidative stress modulated METS-triggered apoptosis.

METS Causes More Caspase 3 and 3/7 Activations to Oral Cancer Cells Than Normal Cells
Caspase 3 activation was monitored by flow cytometry and luminescence detection. For flow cytometry, the caspase 3 (+) events of histograms were used to measure   (Figure 5A,C). However, it showed lower caspase 3 (+) events in normal S-G cells than oral cancer cells by METS treatment. For the luminescent assay, caspase 3/7 activities were dose-responsively increased in oral cancer cells but not in normal cells ( Figure 5C). Accordingly, METS exert more caspase 3 and 3/7 activations in oral cancer cells than in normal cells. Additionally, NAC was used to test the contribution of ROS in caspase 3 and 3/7 activation caused by METS. NAC moderately inhibited caspase 3 activations of Ca9-22 cells at 24 h METS treatment ( Figure 5B). NAC dramatically inhibited caspase 3 activations of CAL 27 cells at 12 and 24 h METS treatment. NAC also moderately inhibited 3/7 activations of Ca9-22 and CAL 27 cells at 6 μg/mL of 24 h METS treatment ( Figure 5B,C). In contrast, normal S-G cells showed low changes in caspase 3 and 3/7 activities. These results reveal that oxidative stress modulated METS-triggered caspase 3 and 3/7 activations.

METS Causes More Caspases 8 and 9 Activations to Oral Cancer Cells Than Normal Cells
Caspase 8 and 9 activations were monitored by flow cytometry. The caspase 8 and 9 (+) events of histograms were used to measure the apoptosis. Caspase 8 and 9 (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) ( Figure 6A,C). However, it showed lower caspase 8 and 9 (+) events in normal S-G cells than oral cancer cells by METS treatment. Accordingly, METS exert more caspase 8 and 9 activations in oral cancer cells than in normal cells. Additionally, NAC inhibited caspase 8 and 9 activations of METS-treated oral cancer cells ( Figure 6B,D), revealing that oxidative stress modulated METS-triggered apoptotic signaling.

METS Causes More Caspases 8 and 9 Activations to Oral Cancer Cells Than Normal Cells
Caspase 8 and 9 activations were monitored by flow cytometry. The caspase 8 and 9 (+) events of histograms were used to measure the apoptosis. Caspase 8 and 9 (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) ( Figure 6A,C). However, it showed lower caspase 8 and 9 (+) events in normal S-G cells than oral cancer cells by METS treatment. Accordingly, METS exert more caspase 8 and 9 activations in oral cancer cells than in normal cells. Additionally, NAC inhibited caspase 8 and 9 activations of METS-treated oral cancer cells ( Figure 6B,D), revealing that oxidative stress modulated METS-triggered apoptotic signaling.

METS Causes More ROS and MitoSOX but Less GSH Generations to Oral Cancer Cells Than Normal Cells
The ROS and MitoSOX (+) events of histograms were used to measure oxidative stress. In general, ROS and MitoSOX (+) events were dose-and time-

METS Causes More DNA Damage to Oral Cancer Cells Than Normal Cells
The γH2AX and 8-OHdG (+) events of histograms were used to measure DNA damage. γH2AX and 8-OHdG (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) (Figures 9A and 10A). However, it showed lower γH2AX and 8-OHdG (+) events in normal S-G cells than in oral cancer cells by METS treatment. Accordingly, METS exerts more γH2AX and 8-OHdG in oral cancer cells than in normal cells. Additionally, NAC inhibited γH2AX and 8-OHdG increment of METS-

METS Causes More DNA Damage to Oral Cancer Cells than Normal Cells
The γH2AX and 8-OHdG (+) events of histograms were used to measure DNA damage. γH2AX and 8-OHdG (+) events were dose-and time-dependently increased by METS in oral cancer cells (Ca9-22 and CAL 27) (Figures 9A and 10A). However, it showed lower γH2AX and 8-OHdG (+) events in normal S-G cells than in oral cancer cells by METS treatment. Accordingly, METS exerts more γH2AX and 8-OHdG in oral cancer cells than in normal cells. Additionally, NAC inhibited γH2AX and 8-OHdG increment of METS-treated oral cancer cells ( Figures 9B and 10B), particularly for 24 h. These results reveal that oxidative stress was involved in METS-triggered γH2AX and 8-OHdG generation.

Discussion
Several kinds of marine sponge extracts have demonstrated anticancer [43] effects. However, the impacts of antiproliferation of the marine sponge Theonella extract (METS) on oral cancer cells were rarely investigated. Moreover, most marine sponge extract studies reported only on the cytotoxicity of some cancer cells but did not consider the drug response to normal cells. These studies for marine sponge extracts lack a detailed investigation of their acting mechanism. The present study demonstrated the preferential

Discussion
Several kinds of marine sponge extracts have demonstrated anticancer [43] effects. However, the impacts of antiproliferation of the marine sponge Theonella extract (METS) on oral cancer cells were rarely investigated. Moreover, most marine sponge extract studies reported only on the cytotoxicity of some cancer cells but did not consider the drug response to normal cells. These studies for marine sponge extracts lack a detailed investigation of their acting mechanism. The present study demonstrated the preferential antiproliferation to oral cancer cells but only revealed the low cytotoxic effect on normal cells. The exact impacts of METS treatments on oral cancer cells were discussed.
In the present assessment, METS shows IC 50 values of 4.5 and 5 µg/mL in oral cancer cells (Ca9-22 and CAL 27) at a 24 h MTS assay. In contrast, normal cells (S-G) only show no change in viability. Moreover, this antiproliferation was alleviated by ROS inhibitor (NAC) pretreatment. Accordingly, these results suggest that METS possesses ROS-dependent preferential antiproliferation to oral cancer cells but causes little cell death in normal cells. Since METS is the crude extract from the marine sponge, it shows a high drug sensitivity (IC 50 4.5 µg/mL) for oral cancer treatment; however, the in vivo antioral cancer effects of METS were not investigated. The in vivo effect of METS warrants a detailed evaluation that may improve the drug progression of its clinical treatment to oral cancer. Moreover, combining natural products with clinical drugs can improve oral cancer therapy [46,47]. Combined treatment may sensitize cancer cells to clinical medications and reduce their potential adverse effects. It warrants a detailed assessment of future combined treatment, including METS and other clinical drugs.

METS Exhibits Preferential Generation of Oxidative Stress to Oral Cancer Cells
Cancer cells exhibit a higher extent of basal oxidative stress than normal cells [42,[48][49][50][51]. Elevated oxidative stress may overload cancer cells and cause cell death but tolerate normal cells to maintain cell survival. Similarly, METS also demonstrated oxidative stress inductions, such as upregulation of ROS and MitoSOX in oral cancer cells more than in normal cells (Figure 7). These oxidative stress inductions were suppressed by NAC, suggesting that MET is an oxidative modulating agent. Accordingly, METS showed preferential induction of oxidative stress in oral cancer but not in normal cells.
Moreover, redox homeostasis is balanced by cellular antioxidants and oxidative stress [52,53]. GSH can alleviate cellular oxidative stress [54]. Some drugs suppress endogenous antioxidants and induce oxidative stress. For instance, amygdalin decreases the GSH level and is associated with oxidative stress induction of breast cancer cells [55]. Similarly, METS downregulate GSH levels in oral cancer cells (Figure 8), accompanied by oxidative stress generation (Figure 7). After METS treatment, the GSH level of oral cancer cells is lower than normal cells. Therefore, METS-induced oxidative stress may partly be attributed to GSH depletion.

METS Preferentially Provokes Apoptosis in Oral Cancer Cells
Oxidative stress elevation is an anticancer strategy promoting apoptosis [49,56]. Marine sponge extracts promote apoptosis in cancer cells [57]. For example, petroleum ether extract of Negombata magnifica induces apoptosis in terms of DNA ladder assay in liver cancer cells [57]. Ethyl acetate extract of Hyalella cribriformis activates caspase 3 in rhabdomyosarcoma cells [58].
Similarly, METS causes a dramatic elevation of subG1 (Figure 3), and it also was associated with apoptosis inductions as detected by annexin V staining ( Figure 4). Moreover, apoptosis-related caspase signaling, such as caspases 8, 9, and 3, were activated by METS of oral cancer cells but displayed a small activation in normal cells. These results indicate that METS preferentially causes apoptosis in oral cancer cells compared to normal cells. Moreover, it also suggests that METS activates extrinsic and intrinsic apoptosis in oral cancer cells since caspases 8 and 9 are activated. Furthermore, these apoptosis inductions were suppressed by NAC ( Figure 6). Hence, METS promotes apoptosis of oral cancer cells relying on oxidative stress.

METS Preferentially Provokes DNA Damage to Oral Cancer Cells
Several natural products use the anticancer strategy by targeting DNA damage responses, such as γH2AX expression [59]. 8-OHdG is a typical oxidative-stress-dependent DNA damage [60]. Since METS induces oxidative stress, the DNA damage status warrants a detailed assessment. Using flow cytometry, the γH2AX and 8-OHdG DNA damage were caused after METS treatment for oral cancer, but they showed lower levels in normal cells (Figures 9 and 10). Moreover, these DNA damage inductions were suppressed by NAC. Accordingly, METS preferentially provokes ROS-dependent DNA damage to oral cancer cells, solidly impacting preferential antiproliferation.

METS Preferentially Arrests the Cell Cycle in Oral Cancer Cells
Marine sponge extracts disturb cell cycle progression in cancer cells. For example, petroleum ether extract of Negombata magnifica induces G1 arrest in liver cancer cells [57]. Methanol extract of Crambe crambe causes G2/M arrest in prostate cancer cells [61]. In addition to subG1 accumulation, METS caused G1 blocking of oral cancer Ca9-22 cells and G2/M blocking of oral cancer CAL 27 and normal S-G cells. Consequently, METS induced different responses to cell cycle disturbance in different oral cancer cell lines. Therefore, different marine sponge extracts may block the cell cycle progression at different phases in different cancer cells.

Conclusions
This study confirms that the antiproliferation effects of METS are effective in oral cancer cells but do not affect normal cells. This METS-induced preferential antiproliferation effects is associated with higher expressions of cellular and mitochondrial oxidative stress (ROS and mitoSOX), apoptosis (subG1 accumulation, annexin V enhancement, and upregulation of the extrinsic and intrinsic caspase signaling), and DNA damage (DNA double-strand breaks and oxidative damage) in breast cancer cells than in normal cells. Utilizing NAC pretreatment demonstrates that preferential antiproliferation function and mechanism of METS provides an oxidative-stress-mediated regulation. In conclusion, METS is a promising marine-sponge-derived natural product for antioral cancer treatment.