Thymoquinone synergizes gemcitabine anti-breast cancer activity via modulating its apoptotic and autophagic activities

The use of anti-cancer adjuvant therapy is rationalized by potentiating the efficacy, and/or protecting from major side effects of chemotherapeutics. Thymoquinone (TQ) is a naturally occurring compound with cumulative evidence of anti-cancer properties. In this study, we assessed the chemomodulatory potential of TQ to gemcitabine (GCB) against human breast adenocarcinoma (MCF-7), and ductal carcinoma (T47D) cells. TQ showed cytotoxic effects against MCF-7 and T47D with IC50’s of 64.9 ± 14 µM and 165 ± 2 µM, respectively. The IC50’s of GCB against MCF-7 and T47D were 0.9 ± 0.18 µM and 14.3 ± 2.8 µM and were significantly reduced after combination with TQ to 0.058 ± 12 µM and 2.3 ± 0.2 µM, respectively. The CI- values were indicative of synergism in MCF-7 and T47D cells (0.15 and 0.30, respectively). Further investigation showed that GCB caused significant anti-proliferative effect reflected by increasing cell population in S-phase in both cell lines. TQ potentiated GCB-induced anti-proliferative activity in both cell lines. GCB induced considerable apoptosis in T47D cell line, and TQ significantly increased GCB-induced apoptotic effects by 1.5 to 3.6 folds. Interestingly, GCB, TQ and their combination induced significant autophagic cell death in the apoptosis defected MCF-7 cells. In addition, TQ, GCB and their combination depleted breast cancer associated stem cell (CD44(+)/CD24(−)/(low)) clone within MCF-7 and T47D cells by 3.8% to 27.5%. In conclusion, TQ showed promising chemomodulatory effects to GCB against breast cancer cells via inducing apoptosis, necrosis and autophagy, in addition to depleting tumor associated resistant stem cell fraction.

Natural compounds and even crude medicinal plants are believed to be promising source of alternative anti-cancer remedy. They are well-known to suppress or block the carcinogenic processes 10 . Amongst, Nigella sativa is extensively studied for potential anticancer properties. It was even described as a miracle herb since many studies revealed its outstanding pharmacological potential 11 . Thymoquinone (TQ) is one of the major bioactive compounds isolated from Nigella sativa, it possesses anti-inflammation, anti-hypertensive, anti-oxidant, and anti-cancer effects 12,13 . Combination of natural compounds with conventional cancer chemotherapy showed promising outcomes and gained the attention of scientists worldwide due to enhancing the anti-cancer efficacy without increasing normal tissue toxicity 14 In our previous work, combination of TQ with cisplatin diminished the resistance fraction to cisplatin and improved its anticancer activity against head and neck cancer cells 15 . Yet, TQ might overcome resistance to GCB and would be a potential successful combination therapy for breast cancer. Therefore, the aim of the current work is to investigate the potential chemomodulatory effects of TQ to the cytotoxic profile of GCB against human breast cancer cell lines.

TQ improves the cytotoxicity of GCB in breast cancer cell lines.
To study the influence of TQ on the cytotoxic profile of GCB against breast cancer cells, dose response curves of GCB, TQ and their combination were conducted against MCF-7 and T47D cells (Table 1).
In T47D cell line, GCB exerted gradient cytotoxic activity with increasing concentration at all exposure time points; viability started to drop at concentrations equal to or higher than 1 μM with an IC 50 's of 7.6 ± 0.7, 6.4 ± 1.4 and 14.3 ± 2.8 μM after 24 h, 48 h and 72 h, respectively. TQ treatment showed sudden cytotoxic effect at concentrations higher than 100 μM; IC 50 of TQ was found to be higher than 100 μM at all exposure time points (Fig. 1B,D and F). Equitoxic combination of TQ with GCB significantly improved the cytotoxic profile of GCB, inducing remarkable decrease in its IC 50 by 6.2 to 25 folds (0.3 ± 0.02, 0.27 ± 0.07 and 2.3 ± 0.2 µM). Combination analysis revealed a strong synergism between GCB and TQ with CI-values ranging from 0.1 to 0.34 ( Table 1).

Effect of GCB, TQ and their combination on cell cycle distribution of breast cancer cells.
To assess the influence of GCB, TQ and their combination on the cell cycle distribution of breast cancer cells, MCF-7 and T47D cells were treated with the pre-determined IC 50 's of treatments under investigation for 24 and 48 h, and assessed for DNA content using flow cytometry. In MCF-7, TQ alone did not induce any significant change in all cell cycle phases. However, GCB caused significant anti-proliferative effect manifested by increasing the cell population at G 0 /G 1 phase after 24 h from 38.1 ± 3% to 67.7 ± 1.2% (Supp. Table 1). Further exposure (48 h) of MCF-7 to GCB induced significant S-phase arrest increasing its cell population from 38.0 ± 1.9% to 50.4 ± 3.2%. Combining TQ with GCB did not further increase GCB-induced antiproliferative effects either in G 0 /G 1 -phase or S-phase ( Fig. 2A,B,D,E). After 24 h, GCB alone or GCB in combination with TQ induced significant increase in pre-G cell population from 4.7 ± 1.5% to 8.8 ± 0.6% and 7.2 ± 1.2%, respectively. Longer exposure (48 h) of MCF-7 cells to GCB significantly induced more cell death manifested by 23.1 ± 3.4% of cells in the pre-G phase  Table 2). In T47D, TQ alone did not cause any significant change in cell cycle distribution after 24 h. However, it induced cell cycle arrest at S-phase after 48 h; cell population within S-phase was increased from 29.1 ± 1.7% to 38.4 ± 0.2%. In addition, GCB induced significant anti-proliferative effect manifested by increasing cells at G 0 /G 1 phase after 24 h from 53.2 ± 0.6% to 60.5 ± 1.8%. After 48 h, GCB induced significant S-phase arrest increasing its cell population from 29.1 ± 2.7% to 36.5 ± 3.4%. The combination of TQ with GCB for 48 h increased GCB-induced antiproliferative effects manifested by increasing cells in S-phase from 44.3 ± 2.5% to 49.9 ± 1.3% (Fig. 3A,B,D,E) and (Supp. Table 3). After 24 h of treatment, TQ induced significant increase in the pre-G cell population from 8.5 ± 0.3% to 10.8 ± 0.2%. Further exposure (48 h) of T47D to TQ significantly induced more cell death manifested by increased pre-G phase cell population from 12.6 ± 1.4% to 67.6 ± 5.2%. Moreover, GCB alone or GCB in combination with TQ induced significant increase in pre-G cell population    Cell cycle distribution was determined using DNA cytometry analysis and different cell phases were plotted (B,E) as percentage of total events. Sub-G cell population was plotted as percent of total events (C,F). Data is presented as mean ± SD; n = 3. Data is presented as mean ± SD; n = 3. *Significantly different from control group. **Significantly different from GCB treatment.  4B). In addition, treatment with GCB or GCB combination with TQ for 24 h induced significant necrotic cell death compared to control cell (1.3 ± 0.1%, 4.9 ± 0.2% and 0.84 ± 0.1%, respectively) (Fig. 4A). Prolonged exposure of cells to single treatment of GCB or TQ induced cell necrosis 22.64 ± 0.09% and 4.01 ± 0.3%, respectively. However, TQ in combination with GCB induced necrosis in only 6.9 ± 1.1% of cells (Fig. 4B).

Assessment of autophagy.
Other than apoptosis, programmed cell death via autophagy represents a big research controversy. Herein, we further investigated the effect of GCB, TQ and their combination on autophagy process within MCF-7 and T47D cells using Cyto-ID autophagy detection dye coupled with flowcytometry. In MCF-7, treatment with GCB or TQ alone increased autophagic cell death by 39.4% and 60.6%, respectively. Combination of TQ with GCB significantly increased autophagic cell death by 29.1% compared to control untreated cells (Fig. 5A). Unlike MCF-7 cells, TQ alone did not induce any significant change in autophagic cell death in T47D cells. Furthermore, GCB and combination of GCB with TQ significantly decreased autophagic cell death by 54.1% and 52.2%, respectively (Fig. 5B).

Discussion
Breast cancer remains a significant health problem worldwide and is the most common cancer diagnosed in females as it increases morbidity and mortality rates 16 . Treatment options include surgery, hormonal therapy, radiotherapy, and chemotherapy 17 . GCB remains as one of the most common chemotherapeutic agents for the treatment of breast cancer 18 . Although GCB treatment cause initial responses, its long-term success may be discontinued because of the development of drug resistance and toxicities 9,19 . Recently, the combination of natural products with chemotherapies have attracted researchers' attention as it was found to augment the effect of standard cancer chemotherapeutic drugs and/or protect from its side effects [20][21][22] . TQ is among the active components of Nigella sativa which is commonly used for several medicinal purposes 11,23 .
Herein, we showed a strong synergism between GCB and TQ against breast adenocarcinoma (MCF-7), as well as breast ductal carcinoma (T47D) cells. It is also worth mentioning the weaker cytotoxic effect of GCB against breast cancer cells by longer exposure (72 h) might be attributed to its stability issues. GCB is unstable in serum condition and this is due to protein binding and enzyme dependent and independent degradation 24,25 . Moreover, GCB suffers from several physico-chemical stability issues in solutions 26 . Accordingly, further detailed assessment for GCB-induced influences to cell cycle, apoptosis and autophagy were carried out after treatment for 24 and 48 h. According to our observation, TQ alone showed significant but weak anti-proliferative effects in comparison to GCB. However, TQ enhanced the cytotoxic profile of GCB by 9-15 folds and 6-25 folds against MCF-7 and T47D, respectively. Several publications reported the significance of TQ alone as an anti-cancer agent in different types of cancer [27][28][29] . In addition, several studies including ours showed promising chemomodulatory effects of TQ to several chemotherapeutic agents against different types of cancer 15,30 . Earlier in 2014, Pandita and colleagues reported a synergistic interaction between TQ and GCB against pancreatic cancer cells. TQ down regulate Pyruvate kinase which is involved in a wide range of cancer cell metabolism 22 . Later on, Zhang and colleagues showed a chemosensetizing effect of TQ to cisplatin against colorectal cancer cells via inhibiting NF-κB signaling 31 .
In the current work, we tried to further explain the synergistic interaction between GCB and TQ in breast cancer cells from the aspect of cell cycle interference. GCB slowed down the cell cycle progression in G 0 /G 1 and S-phases in both cell lines which was also reported by previous studies 32 . The anti-proliferative effect of GCB alone or in combination with TQ was found to be stressful enough to induce cell death observed by increased Pre-G cell population. TQ alone did not induce any significant cell cycle interference; except delayed S-phase arrest after 48 h of exposure. However, TQ was found to potentiate the killing effect of GCB increasing the Pre-G cell population in both cell lines under investigation compared to GCB treatment alone. Previous studies for the influence of TQ to cell cycle progression denoted interference with different cell cycle phases such as G 0 /G 1 and S-phases 33,34 . TQ-induced cell cycle arrest in S-phase was also stressful to T47D cells and induced elevated Pre-G cell population. The increased Pre-G cell population might not be specific enough to determine the exact cell death mechanism. Some studies referred TQ-induced anticancer effect to its ability to induce apoptosis via TGF-family, p53, p21, c-FLIP, Bax and Bcl-2 interference [35][36][37] . Furthermore, we examined apoptotic, necrotic and autophagic cell death induced by GCB, TQ and their combination. According to our observation, both TQ and GCB induced significant apoptosis in T47D by more than 4 folds after 24 h. Yet, the synergistic interaction between TQ and GCB against T47D cells could be clearly explained by the excessive increase of apoptotic cell fraction compared to single GCB or TQ treatments (2.5 fold higher apoptosis). Altered apoptosis is one of the important underlying reasons for GCB resistance among cancer cells 38,39 . Many altered pathways were suggested to be responsible for GCB-induced apoptosis resistance, such as Bcl-2 40 . It is worth mentioning that combination treatment induced significantly higher necrosis compared to GCB or TQ treatments alone which in turn might bypass apoptosis pathways altogether. In MCF-7 cells, apoptosis is not detectable due to lack of caspase-3 expression 41 . In such cases, alternative cell death pathways are mandatory. Autophagy is another suggested cell death pathway; however it possesses complicated roles and controversy in cancer cell death 42,43 . In MCF-7 cells, TQ induced significant autophagic cell death and this might be a pro-death mechanism due to defective apoptosis in this cell line 41,43 . It was reported for MCF-7 cells to undergo programmed cell death dominantly via autophagy 43 . In contrast to MCF-7, TQ did not exert any autophagic response in T47D. Furthermore, GCB and combination of GCB with TQ significantly decreased autophagic cell death. Yet, this might be explained by forced apoptosis induction in T47D cell after these treatments. In other words, autophagy is considered herein as an apoptosis escape shelter 43,44 .
Besides the synergistic interaction between GCB and TQ against breast cancer cells, we studied the influence of these treatments against breast cancer associated stem cells (CD44 + /CD24 − ) 45 . To the best of our knowledge, this is the first study demonstrating the effect of GCB and TQ against breast cancer stem cells. According to our observations, GCB alone minimally affected CD44 + /CD24 − cell clone only in MCF-7 cells. Yet, it was found that GCB treatment activates a group of developmental pathways known to be responsible for chemotherapeutic treatment resistance 9 . Interestingly, TQ significantly decreased CD44 + /CD24 − cell clone in both cell lines under investigation; and TQ combination with GCB further suppressed this stem cell clone in MCF-7 cells. In conclusion, TQ proved and is still proving to possess strong chemomodulatory potential to many chemotherapeutic agents such as GCB, against breast cancer cell lines. TQ induces cell death via apoptosis, necrosis and autophagy. In addition, TQ decreases tumor associated resistant stem cell fraction. Cell culture. Human breast cancer cell lines, MCF-7 and T47D, were obtained from the Vaccera (Giza, Egypt). Cells were maintained in DMEM media supplemented with streptomycin (100 μg/mL); penicillin (100 units/mL) and heat-inactivated fetal bovine serum (10% v/v) in a humidified, 5% (v/v) CO 2 atmosphere at 37 °C.

Materials and Methods
Cytotoxicity assays. The cytotoxicity of TQ, GCB, and their combination were tested against MCF-7 and T47D cells by sulforhodamine B (SRB) assay. Exponentially growing cells were collected using 0.25% Trypsin-EDTA and seeded in 96-well plates at 1000-2000 cells/well. Cells were treated with serial concentration (0.01 to 300 µM) of TQ, GCB and their combination for 24, 48 and 72 h and subsequently fixed with trichloroacetic acid (TCA) (10% w/v) for 1 h at 4 °C. After several washings with double distilled water, cells were stained with SRB solution 0.4% (w/v) for 10 min in a dark place at room temperature and finally washed with 1% (v/v) acetic acid. After the plates became dry by overnight incubation, Tris-HCl (50 mM, pH 7.4) was used to dissolve the SRB-stained cells and color intensity was measured at 540 nm with ELISA microplate reader and calculated as percent viability of control cells (cells exposed to drug free media).

Data analysis.
The dose response curves of drugs under investigation were analyzed using E max model in the following formula where "R" is the residual unaffected fraction (the resistance fraction); "[D]" is the drug concentration used; "K d " is the drug concentration that produces 50% reduction of the maximum inhibition rate and m is a Hill-type coefficient. "IC 50 " is defined as the drug concentration required to reduce absorbance to 50% of the control (i.e., K d = IC 50 when R = 0 and E max = 100 − R). Combination index (CI) was calculated from the formula:

Analysis of Cell Cycle Distribution.
To assess the effect of the TQ, GCB and their combination on cell cycle distribution, MCF-7 and T47D cells were subjected to the pre-determined IC 50 's of test drugs or drug free media for 24 and 48 h. After treatment, cells were collected by trypsinization and washed twice with ice-cold PBS and re-suspended in 0.5 mL of PBS. Two milliliters of 60% ice-cold ethanol were added gently while vortexing and cells were incubated at 4 °C for 1 h for fixation. Upon analysis, fixed cells were washed and re-suspended in 1 mL of PBS containing 50 µg/mL RNAase A and 10 µg/mL propidium iodide (PI). After 20 min of incubation in dark at 37 °C, cells were analyzed for DNA contents using flow cytometry analysis FL2 (λ ex/em 535/617 nm) signal detector (ACEA Novocyte ™ flowcytometer, ACEA Biosciences Inc., San Diego, CA, USA). For each sample, 12,000 events were acquired. Cell cycle distribution was calculated using ACEA NovoExpress ™ software (ACEA Biosciences Inc., San Diego, CA, USA).

Apoptosis assay.
To elucidate the method of cell death by which breast cancer cells are killed in response to treatment with GCB, TQ and their combination, apoptosis and necrosis cell populations were determined using Annexin V-FITC apoptosis detection kit (Abcam Inc., Cambridge Science Park, Cambridge, UK). Briefly, the cells were exposed to the predetermined IC 50 's of test drugs (single or combined treatments) or drug free media (control group) for 24 h and 48 h. Cells were harvested and washed twice with PBS, and incubated in dark with 0.5 ml of Annexin V-FITC/PI solution for 30 min in a dark place at room temperature according to manufacturer protocol. After staining,cells were injected via ACEA Novocyte ™ flowcytometer (ACEA Biosciences Inc., San Diego, CA, USA) and analyzed for FITC and PI fluorescent signals using FL1 and FL2 signal detector, respectively (λ ex/em 488/530 nm for FITC and λ ex/em 535/617 nm for PI). For each sample, 12,000 events were acquired and positive FITC and/or PI cells were quantified by quadrant analysis and calculated using ACEA NovoExpress ™ software (ACEA Biosciences Inc., San Diego, CA, USA).

Autophagy assay.
To further elucidate the method of cell death by which breast cancer cells are killed in response to treatment with GCB, TQ and their combination, Autophagic cell death was quantitatively assessed using Cyto-ID Autophagy Detection Kit (Abcam Inc., Cambridge Science Park, Cambridge, UK). In brief, cells were exposed to the predetermined IC 50 's of test compounds (single or combined treatments) for 24 h. Simultaneously, cells were exposed to 10 µM chloroquine (CQ) as a positive control (autophagy inducing agent), and drug free media (control group) for 24 h. After treatment, cells were collected and washed twice with PBS.
Stem cell detection. The effects of TQ, GCB, and their combination against breast cancer associated stem cell clone (CD44 + /CD24 − ) were assessed using flow cytometry coupled with FITC labeled anti-CD44 and APC/ Cy7 labeled anti-CD24 antibodies (Abcam Inc., Cambridge Science Park, Cambridge, UK). Briefly, cells were treated for 24 h with the predetermined IC 50 's of test compounds (single or combined treatments), and drug free media (control group). After treatment, cells were collected and washed with 10% FBS in ice cold PBS. Cells were incubated with the conjugated anti-CD44 and anti-CD24 antibodies in a dark place at room temperature. After staining, cells were washed three times with 10% FBS in ice cold PBS. Finally, cells were injected via ACEA Novocyte ™ flowcytometer (ACEA Biosciences Inc., San Diego, CA, USA) and analyzed for FITC and APC/CY7 fluorescent signals using FL1 and FL2 signal detector, respectively (λ ex/em 488/530 nm for FITC and λ ex/em 535/617 nm for APC/CY7). For each sample, 12,000 events were acquired and positive FITC and/or APC/CY7 cells were quantified by quadrant analysis and calculated using ACEA NovoExpress ™ software (ACEA Biosciences Inc., San Diego, CA, USA).
Statistical analysis. Data are presented as mean ± SD using Prism ® for Windows, ver. 5.00 (GraphPad Software Inc., La Jolla, CA, USA). Analysis of variance (ANOVA) with LSD post hoc test was used for testing the significance using SPSS ® for windows, version 17.0.0. p < 0.05 was taken as a cut off value for significance.