Potent antitumor property of Allium bakhtiaricum extracts

A.bakhtiaricum was collected from Shiraz, Iran, in the spring, authenticated by Dr.Shahin Zarre and deposited at the Herbarium of Faculty of Sciences, Tehran University, Tehran, Iran (Voucher No:45496). The aerial parts were air dried prior to being grinded into powder. 50 g of dried powder was mixed with ethanol: water (80:20) at room temperature in order to obtain total extract. In addition, 100 g of plant powder was extracted sequentially by solvents with a wide range of polarities including n-hexane, chloroform, ethyl acetate and methanol using a maceration procedure. The process was repeated 3 times with the same plant material but using fresh solvents [36‐38]. The extracts were then filtered and evaporated to dryness on a rotary evaporator under reduced pressure below 40 °C. All the extracts were stored at 4 °C until used for experiments. The yield of extraction for total extract, n-hexane, chloroform, ethyl acetate and methanol fractions were as follows: 32.57, 1.63, 1.08, 0.4 and 15.25%, respectively.

MDA-MB-231 (human breast adenocarcinoma, C578), MCF-7 (human breast adenocarcinoma, C135), HT-29 (human colorectal adenocarcinoma, C466), HepG2 (liver hepatocellular carcinoma, C158), 4 T1(mouse mammary tumor, C604) and NIH3T3 (mouse embryonic fibroblasts, C156) cell lines were purchased from the cell bank of Pasture Institute of Iran (NCBI). Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) medium containing 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin (GibcoBRL, Rockville, IN, USA) at 37 °C with 5% CO2 in a humidified atmosphere inside a CO2 incubator. All solvents used were of analytical grade and purchased from Merck (Darmstadt, Germany), and the other chemicals were obtained from Sigma-Aldrich (St Louis, MO, USA).

The cytotoxic effects of total extract and fractions were assessed toward cancerous and non-cancerous cell lines by applying the MTT assay. Following seeding of the cells (MDA-MB-231, 3–7 × 10³; MCF-7, 4–8 × 10³; HT-29, 4–8 × 10³; HepG2, 5–9 × 10³; NIH3T3, 1 × 10³) in 96-well plates for different time exposures, samples were added at concentrations ranging from 0.002 to 0.25 mg/ml to each well and then incubated for 24, 48 and 72 h. The cultivation media without extract was used as a negative control, while DMSO (dimethyl sulfoxide) (0.5%) served as the solvent control. Afterwards, cells were subjected to MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) (0.5 mg/ml in phosphate buffered saline) for 4 h at 37 °C. Following the solubilization of the formed crystal formazan in DMSO, the absorbance was measured at 545 nm. The IC₅₀ values were calculated from dose-response curves at 24, 48 and 72 h exposure times.

Cell cycle phase distribution was determined by flowcytometry. Following exposure of MDA-MB-231 cells to the ¼ IC_50-72h, ½ IC_50-72h and IC_50-72h concentrations of the chloroform and ethyl acetate fractions for 48 h, cells were collected and stained with Propidium Iodide (PI) reagent at 37 °C for 15 min in the dark. PARTEC flowcytometer (Partec GmbH, Munster, Germany) using Flowjo Software was applied for determining the DNA content [38].

In order to detect apoptosis, Annexin-V/PI assay was carried out on the cells treated with ¼ IC_50-72h, ½ IC_50-72h and IC_50-72h concentrations of the chloroform and ethyl acetate fractions at different times. Afterwards, 100 μl Annexin-V-FLUOS (Roche Applied Science, Indianapolis, IN, USA) labeling solution was added to the suspended MDA-MB-231 cells and incubated at 37 °C. The tubes were then diluted with buffer prior to being subjected to flowcytometry [39].

MDA-MB-231 cells treated with the chloroform extract at ¼ IC_50-72h, ½ IC_50-72h and IC_50-72h concentrations for 48 h were lysed in lysis buffer (Tris 62.5 mM(pH 6.8), DTT 50 mM, SDS 10%, glycerol), and the extracted proteins were separated by 15% SDS-PAGE, electroblotted to polyvinylidene fluoride membrane (GE Health Care Life Sciences, Buckinghamshire, UK) and probed with primary antibody (LC3, 1:1000) (Cell Signaling Technology,Beverly, MA), followed by anti-rabbit IgG horseradish peroxidase (HRP) secondary antibody (1:8000) (Cell Signaling Technology, Beverly, MA). Protein bands were then detected by ECL (Enhanced chemiluminescence) advanced Western blotting detection kit (General Electric Health Care Life Sciences, Buckinghamshire, UK). Image J was used for data analysis to determine integrated density of bands. Protein concentration was measured using the Bradford assay [40].

MDA-MB-231 cells were treated for 2 h with high concentrations of the ethyl acetate (0.3, 0.06 mg/ml) and chloroform(0.2, 0.05 mg/ml) fractions and then the culture media were renewed. Besides, MDA-MB-231 cells were also treated with 0.002 and 0.001 mg/ml of the chloroform as well as 0.003 and 0.001 mg/ml of the ethyl acetate fractions. After washing the cells in PBS and fixing them with 2% formaldehyde, cells were incubated at 37 °C (no CO₂) with fresh senescence associated β-Gal (SA-β-Gal) (Abcam, Cambridge, MA) stain solution [41]. Stained cells were visualized maximal in 12–16 h. β-galactosidase activity was tracked under a Nikon eclipse TS100 inverted microscope for incidence of senescence.

MDA-MB-231 cells were seeded on 8-well glass slides(SPL Life Sciences, Korea) and treated with ¼ IC_50-72h concentration of the chloroform extract for 48 h. Having washed with phosphate-buffered saline, cells were fixed in glutaraldehyde (1% in PBS) at room temperature for 10 min. Afterwards, the fixed cells were washed again by PBS and permeabilized using washing buffer (0.1% Triton X-100, 1% bovine serum albumin in TBS) for 10 min. Cells were then incubated for 30 min with mouse anti-α-tubulin monoclonal antibody (1:100) (Sigma-Aldrich, USA) at room temperature followed by incubation with FITC conjugated anti-mouse IgG antibody (1:500) (Bioscience, CA) for 30 min. The nuclei were stained with propidium iodide (10 mg/ml) (Sigma–Aldrich, USA) [42, 43]. Microtubule networks were detected under a Nikon H600L fluorescence microscope.

Female BALB/c mice (6–8 week old) were purchased from the National Animal Center (Pasteur Institute of Karaj) and maintained under standard conditions of 12/12-h light–dark cycle, with food and water provided ad libitum. Treatment of animals was performed in accordance with the guidelines approved by the animal ethics committee of Pasteur Institute of Iran. Mice were inoculated with 10⁶cells/50 μl of exponentially 4 T1 cells into the mammary fat pad. Following daily observation, once the tumor masses were developed, mice bearing tumor were randomly distributed into eight groups (n = 8). To detect the tumor suppressive role of the chloroformic and ethyl acetate fractions, mice were daily administered by intra-peritoneal injection of vehicle alone (DMSO) and three doses of the chloroformic and ethyl acetate fractions (1, 10 and 20 mg/kg) 5 days a week for 28 days. Mice received no treatment in the negative control group. During the experiments, the tumor growth was tracked every other day and the tumor volume was determined in two dimensions thrice a week using a digital caliper. The tumor volume (mm³) was calculated according to the formula: (length × width²)/2. Treated mice were daily monitored for toxicity including weight loss, discomfort and death. Following anesthesia with 60 mg/kg of sodium pentobarbital, the mouse chest was surgically opened and concurrently perfused by 0.9% saline and then tissue samples including the tumor and lungs were dissected from the animals. Samples were weighed, measured and then placed in 10% formalin for fixation and histopathological analysis.

The fixed tissues were embedded into paraffin blocks, and 5 μm sections were prepared. The tissue sections were picked up onto a glass slide and deparaffinized, rehydrated, and subjected to Hematoxylin and Eosin (H&E) (Merck, Darmstadt, Germany) staining. A Carl Zeiss AxioImager microscope and Image M1 Software (Carl Zeiss, Jena, Germany) were used to provide images of randomly chosen fields at 400x magnification.

IC₅₀ values of total extract and different fractions of A.bakhtiaricum upon 24, 48 and 72 h of treatment toward all the tested cell lines are presented in Table 1. All the extracts caused loss of cell viability in a concentration and time dependent manner on the cell lines. Among different incubation times, a strong cytotoxic activity was revealed following 72 h of incubation of the cancerous cell lines with A.bakhtiaricum extracts. As shown in Table 1, the lowest IC₅₀ values for the extracts were recorded as follows: total < chloroform ≤ ethyl acetate fractions at 72 h. Remarkably, of all the cell lines, MDA-MB-231 cell line was the most sensitive to the chloroformic and ethyl acetate fractions with IC₅₀ values of 0.005 and 0.006 mg/ml, respectively.

In order to explore cell viability of the most effective fractions in a normal cell line, the MTT assay was also performed on NIH3T3 (mouse embryo fibroblast) after 72 h of treatment. Our results indicated that the IC₅₀ values of the chloroformic and ethyl acetate fractions (0.05 and 0.08 mg/ml) were higher against NIH3T3 cells than those obtained from these fractions on cancer cell lines.

Since progression of the cells through the various phases of the cell cycle results in cell proliferation [44], we next evaluated the effects of both chloroform and ethyl acetate fractions, as the most cytotoxic extracts, on cell cycle distribution of MDA-MB-231 cell line using flowcytometry. To do this, cultured MDA-MB-231 cells were exposed to the IC_50-72h, ½ IC_50-72h and ¼IC_50-72h concentrations of the chloroform and ethyl acetate fractions for 48 h, stained with PI and analyzed by flowcytometry. Our findings revealed that MDA-MB-231cells treated with ¼ and ½IC_50-72h concentrations of the chloroformic and ethyl acetate fractions caused a statistically significant rise in the percentage of cell population in G2 phase after 48 h (Table 2). Thus, it appears that these fractions are able to cause cell cytotoxicity by modulating the cell cycle profile and inducing G2/M phase arrest.

To ascertain whether cytotoxic effect of the chloroform or ethyl acetate fraction was due to apoptosis induction, MDA-MB-231 cells were labeled using annexinV/PI and analyzed by flowcytometry. Following treatment with IC_50-72h, ½ and ¼ IC_50-72h concentrations of chloroformic and ethyl acetate fractions, the percentage of viable cells were significantly declined after 72, 48, 24 and 12 h of incubation; however no significant increase was observed in the number of apoptotic cells. Of note, an upward trend was observed in the population of necrotic cells with increasing concentration of the fractions from ¼ IC_50-72h to IC_50-72h values. Also, apoptotic change was not detected after 6 h of exposure of the cells to the fractions (Additional file 1). As a whole, our results implied that ethyl acetate and chloroform extracts possess a non-apoptotic anti-cancer activity.

To determine whether cell death was resulted from autophagy activation, LC3-II accumulation in MDA-MB-231 cells, treated for 48 h with different concentrations of the chloroformic extract, was assessed by Western blotting. A hallmark of autophagy activation is the change of the cytosolic form of LC3 (LC3-I) to its lipidized form (LC3-II) [45]. Hence, we compared the LC3 expression in the chloroformic fraction-treated and untreated MDA-MB-231 cells. As our findings show, upon treatment of MDA-MB-231 cells with the chloroformic fraction, no significant change is observed in the LC3-II expression (Fig. 1.). Regarding Western blotting, the incidence of controlled autophagy in the untreated cell was shown by the presence of LC3-II.

It has been proposed that plant derived compounds may exert their anticancer activity via senescence mechanism due to the presence of alkaloids [46]. In addition, cell growth arrest is likely through the senescence induction [47]. Considering our findings, which indicated a cell cycle arrest following the treatment with the extracts and knowing that Allium possess a wide variety of natural compounds such as alkaloids [48, 49], we sought to explore whether a low dose and chronic treatment of the chloroform and ethyl acetate fractions could prevent the growth of breast cancer cells by inducing the premature senescence. Finding β-galactosidase positive cells implies an increased lysosomal mass and it is commonly considered as a well-established senescence marker [41]. We noticed that MDA-MB-231 cells treated with the chloroformic and ethyl acetate fractions showed no detectable SA-β-gal activity as illustrated in Fig. 2a, b. In contrast, the SA-β-gal positive cells, treated with 1 μM of Adriamycin, exhibited a blue color with a flattened and enlarged morphology, which are indicatives of senescence features (Fig. 2c) [50].

Because microtubule network perturbation can account for the subsequent mitotic arrest [43], herein, we examined whether the chloroform fraction treatment affected the cellular microtubule network. To do this, MDA-MB-231 cells were treated with 4 μM paclitaxel and ¼ IC_50-72h concentration of the chloroform extract, separately. Following 48 h of incubation, the microtubule network was observed by immunocytochemistry (Fig. 3). A normal arrangement and organization of microtubule network was visualized in the control cells (Fig. 3a), whereas paclitaxel enhanced microtubule density and caused long thick microtubule bundles to appear around the nucleus (Fig. 3b). Treating cells with the chloroformic fraction resulted in similar changes as those of paclitaxel-treated cells, demonstrating microtubule network as a possible intracellular target for components of the chloroform fraction. In order to quantify these findings, we calculated the percentage of cells bearing microtubule disarrangement and compared it to the paclitaxel-treated cells. The results were as follows: 96.95 ± 3.04% for paclitaxel- and 83.33 ± 4.16% for the chloroformic fraction-treated cells. Our outputs indicated a significant change in the microtubule network of the cells treated with the chloroformic extract.

To verify the antitumor activity of the chloroform and ethyl acetate fractions in vivo, BALB/c mice were subcutaneously (s.c.) injected with 4 T1 murine tumor cells. Following tumor development on day 7, the mice were distributed in 8 groups and treated with either the vehicle or the chloroformic and ethyl acetate fractions (1, 10, 20 mg/kg/day) for 21 days. As illustrated in Fig. 4a.,i.p. administration of 1 mg/kg/day of the chloroform extract led to a significant tumor growth suppression from the second week (p < 0.01) to the last week (p < 0.0001) of the treatment compared to the vehicle-control group. However, tumor volume reduction was observed after 4 weeks of treatment with 10 and 20 mg/kg/day of the chloroform extract. On the other hand, i.p. administration of 1 mg/kg/day of the ethyl acetate fraction was unable to significantly inhibit tumor growth within the 3 weeks of injection, and a remarkable potency was only observed following 4 weeks of treatment (Fig. 4b). With an increase in the dose of ethyl acetate fraction from 1 to 20 mg/kg/day, tumor growth began to regress significantly compared to the control animals from week 2 to 4 of the treatment. The maximum effect in terms of tumor size was seen for the chloroform fraction following 4 weeks of treatment at 1 mg/kg/day. Consistent with these data, tumor weight was also different among the control and the treated groups. The tumor weight was reduced from 2.02 ± 0.27 g in the control group to 1.01 ± 0.34 g and 0.98 ± 0.11 g in the chloroform and ethyl acetate treated groups, respectively. Also, according to Fig. 4c, the size of the tumor in the control group was larger than that of the treated mice. Remarkably, no weight loss was detected either in the control-vehicle or the treated groups after 28 days.

In the mice bearing 4 T1, the tumor tissues displayed malignant cells with the features including loss of polarity and a number of typical mitotic figures (Fig. 5a). These types of cells were nominated as an invasive lineage representing a suitable model to evaluate the potency of anticancer drugs due to its similarities with metastatic human breast cancer [51]. In the groups treated with 1 mg/kg/day of the chloroform or ethyl acetate fractions, a number of tumor cells in the process of necrosis were observed (Fig. 5b, c); however, tumors dissected from mice treated with the chloroform fraction presented a reduced number of proliferative cells than those from the ethyl acetate fraction at the same dose. These findings are in line with our tumor volume determinations, further corroborating a greater efficiency for the chloroformic fraction in suppressing tumor growth.

The findings of histopathology also exhibited less foci of metastasis in the lungs of the animals treated with chloroform and ethyl acetate fractions compared to the vehicle-treated animals, indicating an anti-invasive property of the fractions (Fig. 5d-f). Besides, a difference was observed between the metastatic lung foci of the chloroform and ethyl acetate treated groups, with the chloroform treated lung tissues harboring less nodules than those of the ethyl acetate-treated group.