Antitumor evaluation of two selected Pakistani plant extracts on human bone and breast cancer cell lines

Four plants were employed in the present study. Vincetoxicum arnottianum and Caccinia mancranthera were collected from Baluchistan (Pakistan) and Berberis orthobotrys and Onosma hispida were collected from Gilgit-Baltistan (Pakistan) in 2014 (Table 1). The plants were identified by Dr. Sher Wali Khan and reference specimens were deposited at the Department of Biological Sciences, Karakoram International University, Pakistan.

Each plant sample including the aerial part of V. arnottianum (VSM), root (BORM) and fruit (BOFM) parts of B. orthobotrys, root (OHRM) and aerial (OHAM) parts of O. hispida, and the aerial part of C. macranthera (CMM) were air dried in shade and mechanically ground to fine powder. The finely-powdered material of each plant was soaked in methanol for several days and extracted. The dried methanolic extracts were obtained by removing the methanol by evaporation under reduced pressure. Furthermore, the fruit extract (BOFM) of B. orthobotrys was fractionated using solvent-solvent extraction and yielding n-hexane soluble oily substance (BO-23) and ethylacetate soluble oily substance (BO-5). Finally, eight samples i.e. VSM, BORM, BOFM, BO-5, BO-23, OHRM, OHAM and CMM were obtained and used for further study. Then, 50 mg of each dry sample was dissolved in 1 ml DMSO, EtOH or MeOH for the antitumor activity tests.

For soaking and extraction purposes, the commercial grade solvents were used. For preparation of the samples for the antitumor activity, absolute ethanol, DMSO, and absolute methanol from Sigma Aldrich were employed.

Human osteosarcoma cell lines MG-63 (CRL-1427), Saos-2 (HTB-85) and human breast adenocarcinoma cell lines MCF-7 (ATCC: HTB-22), BT-20 (HTB-19) as well as non-tumorigenic human epithelial breast cell line MCF-12A (CRL-10782) were purchased from ATCC (http://​www.​lgcstandards-atcc.​org/​) under the given numbers. The human non-tumorigenic, primary osteoblast cells (POB) were chosen as control cells. Briefly, cells were isolated from the spongiosa of the femoral heads of patients undergoing primary total hip replacement. The samples were collected with patient agreement and approval by the Local Ethical Committee (registration number: A 2010-10). Human primary osteoblasts were already used and isolation procedure was already described [46]. Except for MCF-12A, all other cell lines and the primary POB cells were cultivated in Dulbecco’s modified Eagle’s medium (Invitrogen, Germany) with 10 % fetal bovine serum (PAN Biotech GmbH, Germany) and 1 % gentamycin (Ratiopharm, Germany). MCF-12A was grown in Dulbecco’s modified Eagle’s medium Ham’s F12 without phenol red (Invitrogen, Germany) containing 10 % horse serum (PAA Laboratories GmbH, Germany), the Mammary Epithelial Cell Growth Medium SupplementPack (PromoCell, Germany) including Bovine Pituitary Extract 0.004 nl/ml, Epidermal Growth Factor (recombinant human) 10 ng/ml, Insulin (recombinant human) 5 g/ml, Hydrocortisone 0.5 g/ml and 1 % gentamycin (Ratiopharm, Germany).

Prior treatment with the plant extract cells were adapted to phenol-red-free Dulbecco’s modified Eagle’s medium (PAA Laboratories GmbH, Germany) with 10 % charcoal stripped fetal bovine serum (PAN Biotech GmbH, Germany) for 48 h to avoid unspecific stimulation of endogenous hormones in the serum (assay medium). Treatment with plant extracts (final concentration 1, 10, 25, 50, and 100 μg/ml) was carried out for 48 h in assay medium. As negative control substance the vehicle DMSO, ethanol or methanol (0.1 %) was used in the same manner.

MTS (3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay to determine cell viability was performed according to manufactures protocol (CellTiter 96® AQueous One Solution Cell Proliferation Assay; Promega Corp., Madison, WI, USA). Briefly, cells were seeded in 96-well plates at a density of 2000 cells/well in 100 μl medium and left to attach for 24 h. Treatment with plant extracts at final concentrations of 1, 10, 25, 50 and 100 μg/ml was carried out as described previously [47]. In parallel, control approaches were carried out with medium only and 0.1 % of the solvent DMSO, EtOH or MeOH to calculate background absorbance. No background absorbance was obtained for the extracts and MTS in the absence of cells, as some extracts are capable of reducing the MTS. After an initial incubation for 24 h cells were assayed with MTS. Colorimetric changes were measured at 490 nm and raw data was transferred to Microsoft Excel and analyzed. At least 8 replicates corrected with the background absorbance were performed. Reduction of cell viability at each concentration was plotted as a dose response curve. The IC₅₀ values were calculated using nonlinear regression to fit data to the dose–response.

Proliferation alterations as well as apoptosis induction under the exposure of the plant extracts were estimated by cell cycle analysis via flow cytometry (FACS Calibur, BD Biosciences) after propidium iodide (Roche Diagnostics, IN, USA) staining (50 mg/ml) of the cells as already described [47, 48]. For data acquisition and histogram preparation, the software FlowJo version 7.6.5 (Tree Star; www.​flowjo.​com) was used. A minimum of 15,000 ungated events were recorded. Doublets and clumps were excluded by gating on the DNA pulse width versus pulse area displays. For statistical evaluation, the sum of cells in S- and G2/M-phase was defined as proliferative events and the sub-G1-peak of the histogram as apoptotic ones.

Annexin-V detects the translocation of phosphatidylserine from the inner leaflets to the outer leaflets of the plasma membrane, which is a key feature of apoptotic cells, whereas PI detects necrotic cells with permeabilized plasma membrane. Labeling of early apoptotic and dead cells was performed according to the manufacturers’ instructions from the Alexa Fluor488 Annexin V/Dead Cell Apoptosis Kit (Thermo Fisher Scientific Inc., Germany). Cells were treated with 100 μg/ml plant extract for 48 h. After treatment detached as well as adherent cells were washed twice with cold PBS. The cell pellet was resuspended in 100 μl of annexin binding buffer at a density of 1 × 10⁶ cells per ml and incubated with 5 μl of Alexa488-conjugated Annexin-V and 5 μl of PI for 15 min at room temperature in the dark. 400 μl of 1× binding buffer was added to each sample tube, and the samples were immediately analyzed by flow cytometry. Histograms and statistics were designed with the software FlowJo Version 7.6.5.

For bright field as well as fluorescence microscopic imaging, cells were seeded on glass cover slips and cultured for 24 h. After treatment with plant extracts bright field images were obtained using Axio Scope A1 microscope and the software AxioVision Imaging Software Release 4.8.2. (Carl Zeiss, Germany). For fluorescence imaging cell were fixed with 4 % paraformaldehyde for 15 min, followed by three washings with PBS and then permeabilized with 0.1 % Triton X-100 for 15 min. After carefully washing, cells were incubated with 100 μl 6.6 μM Alexa Fluor594 phalloidin (Invitrogen, Germany) for 60 min in the dark at room temperature, washed again, counterstained with DAPI (Roche Diagnostics GmbH, Germany) for 15 min. Finally, cell were washed four times with PBS and embedded in mounting medium. Lysosomes were labeled with LysoTracker® Green DND-26 (Molecular Probes, Carlsbad, CA, USA) following the protocol supplied. The other cell compartments: mitochondria (MitoTracker® Mitochondrion-Selective Probes Green FM), Golgi complex (BODIPY® FL C5-ceramide complexed to BSA), endoplasmic reticulum (ER-Tracker™ Green BODIPY® FL glibenclamide), neutral lipids (4,4-difluoro-1,3,5,7,8-pentamethyl-4-bora-3a,4a-diaza-s-indacene BODIPY® 493/503), all from Molecular Probes, Germany were labeled following the manufactures’ instructions. All fluorescence signals were investigated with an inverted confocal laser scanning microscope (LSM780, Carl Zeiss, Germany) equipped with a helium/neon-ion laser and a ZEISS 63 × oil immersion objectives. The confocal images (1024 × 1024 pixel) were optimized using the ZEN software (Carl Zeiss, Germany).

For scanning electron microscopy (SEM) cells grown on glass cover slips were fixed with 2 % glutaraldehyde and 1 % PFA in 0.1 M phosphate buffer pH 7.3. After washes in 0.1 M phosphate buffer the cells were dehydrated with a graded series of ethanol and were processed for critical point drying using CO₂ as intermedium (Emitech K850 critical point dryer, Emitech Ltd. Ashford, UK). The cover slips were mounted on SEM stubs with adhesive carbon tape (Plano, Wetzlar, Germany) and sputter-coated with a gold layer (approximately 15–20 nm thickness) using a Bal-Tec SCD004 sputter coater (Balzers Union Ltd., Balzers, Liechtenstein). Specimens were viewed in a field-emission SEM operated at 5 kV (Merlin VP compact, Carl Zeiss Microscopy, Jena, Germany) and images with a size of 1024 x 768 pixels were recorded. Morphometric measurements of cell body axis length and width were taken with the free line measurement tool on calibrated pictures imported into iTEM imaging software (Olympus Soft Imaging Solutions, Münster, Germany).

Mitochondrial O₂-consumption as a measure for respiratory activity was determined by the Bionas® 2500 analyzing system combined with the metabolic chip Bionas DisocveryTM SC1000 equipped with Clark-type oxygen sensors. Prior experiments, chips were cleaned with 70 % ethanol for 10 min, washed with PBS and were adapted to the measurement medium for 5 min. Measurement medium was composed of DMEM without NaHCO3 (Invitrogen, Germany), 0.1 % charcoal stripped fetal bovine serum (PAN Biotech GmbH, Germany) and 1 % gentamycin (Ratiopharm, Germany), pH value 7.4 and sterile filtered. On each chip 2x10⁶ cells were seeded and let them adhere over night at 37 °C and in 5 % CO₂ so that 80 % sub-confluence on the sensor chips was reached. Bionas measurements were carried out with a pump rate of 56 ml/min [49]. After an adaption phase of 2 h to the new culture conditions, extracellular oxygen consumption of MG-63 cells after application of 25 μg/ml BORM or VSM was measured continuously for 20 h. Thereafter the recovery status (measurement medium without plant extracts) of the cells was monitored for additional 24 h. Data sets were evaluated and normalized with the software Bionas15002 Data analyzerV1.07.

Influence on migration was conducted on MG-63 cells, pre-incubated in assay medium for 48 h adaption in 6-well plates (Greiner, Germany). A scratch wound was made by Ibidi culture inserts (μ-Dish 35 mm; Ibidi GmbH, Martinsried, Germany) following the instructors recommendations. When cell layers reached confluence, the culture insert was removed and cells were treated with VSM (25–50 μg/ml) extract or control (vehicle, DMSO). Gap closure was analyzed as described previously [50]. Cell invasion assay was performed with the CytoSelect™ 24 –Well Cell Invasion Assay (Basement Membrane, Fluorometric Format) from Cell Biolabs, Inc., CA, USA. Briefly, 1x10⁶ MG-63 cells with the plant extracts were seeded in the membrane insert for 48 h. Fluorescence of invaded cells was counted with a plate reader at 480/520 nm.

The general steps of the Western blot procedure have been described previously [49]. Briefly, after treatment with the plant extracts VSM and BORM for at least 48 h the cells were trypsinized, washed with PBS and lysed in ice-cold lysis buffer (Bio-Plex Cell Lysis Kit, Bio-Rad, USA). After SDS-PAGE, protein content per lane as well separation quality was controlled with the Criterion Stain FreeTM gel imaging system (Bio-Rad, Germany). For protein detection primary antibodies (PCNA: sc-56, from Santa Cruz, USA; BCL-2: B3170, from Sigma) were incubated overnight at 4 °C followed by labeling with a horseradish peroxidase (HPR)-conjugated secondary antibody (Dako, Glostrup, Denmark) for 1 h at room temperature. Protein signals were visualized by using SuperSignal West Femto Chemiluminescent Substrate (Pierce Biotechnology, Rockford, USA). Band intensity was analyzed densitometrically with the Molecular Imager ChemiDoc XRS and Image Lab 3.0.1 software (Bio-Rad, USA). Protein detection was repeated at least three times with individually prepared cell lysates from independently passaged cells.

Every experiment was replicated three times with individually passaged cells and data sets were expressed as means ± standard deviations (SD). Statistical significance was determined by the unpaired one-way ANOVA or t-test (***P < 0.001, **P < 0.005, *P < 0.05).

To evaluate the anticancer properties of the Pakistani plant extracts two bone (MG-63, Saos-2) and two breast (BT-20, MCF-7) cancer cell lines in comparison with primary osteoblasts (POB-110) and non-tumorigenic mammary epithelial cells (MCF-12A) were selected. The osteosarcoma cell line MG-63 represents an early osteoblastic type while Saos-2 cells exhibited the most mature osteoblastic phenotype [51]. The breast cancer cell line MCF-7 represents the luminal, estrogen and progesterone receptor-positive subtype whereas BT-20 cells are invasive, triple-negative breast cancer cells [52]. For the initial screening all cells were treated with 50 μg/ml of each plant extract for 48 h and cell viability was measured right after (Fig. 1). The extracts VSM and BORM caused the greatest significant reduction (40–60 %) of cell viability in the osteosarcoma cell lines MG-63 and Saos-2. On primary osteoblast cells (POB-110) VSM induced only a slight decrease in cell viability while BORM lowered the viability up to 80 %. Beside VSM and BORM, only the treatment with BO-5 on MG-63 cells as well as OHRM on Saos-2 cell revealed a significant viability decrease of 10–15 %. These results illustrate that the VSM extract has anticancer potential on bone cancer cells, since it selectively reduced the vitality of osteosarcoma cells and only exerts a minimal effect on the primary osteoblasts. Similar results were achieved for the treatment of the breast cancer cell lines. On BT-20, hormone-independent and invasive carcinoma cells, only the BORM extract caused a slight viability reduction of approximately 20 %. On MCF-7 cells, VSM as well as BORM induced decreased vitality rate in a range of 10–20 %. This vitality reduction was also measured on the non-tumorigenic control cell line MCF-12A indicating that the extracts VSM and BORM displayed strong cytotoxic effects which will be analyzed in the next sections.

Besides the vitality measurements, the influence on cell growth and the induction of apoptosis are important parameters to evaluate the respective anticancer properties of the plant extracts. Therefore, cell cycle analyses were performed to determine the influence on the proliferation behavior (G2/M + S phase) and apoptosis initiation by DNA strand breaks (sub G1 phase), simultaneously (Fig. 2). Exemplarily, for all cell lines used, Fig. 2 demonstrates the DNA histogram (Fig. 2a), proliferation alterations (Fig. 2b) and the number of apoptotic cells (Fig. 2c) after 48 h treatment with 50 μg/ml plant extract on MG-63 osteosarcoma cell line.

Both, in the histogram and in the proliferation diagram is clearly evident that the extracts OHRM, VSM, BO-5 and BO-23 accelerate the cell number in the proliferative phases G2 and M. Only BORM caused a slight significant reduction in the proliferative phases (20 %). The number of apoptotic cells increased after treatment with 50 μg/ml of BOFM, VSM, CMM and BO-23, significantly. In summary, some of the plant extracts display an effect on the proliferative phases G2/M and S, but do not affect the sub-G1 phase. The DNA-histograms of the control treatments with medium, DMSO, EtOH and MeOH (final concentration: 0.1 μg/ml) are given in Additional file 1: Figure S1, showing no alterations in the cell cycle phases.

As BORM and VSM caused the most significant effects on all bone and breast cancer cell lines, both extracts were examined in concentration series ranging from 0.1 to 100 μg/ml to evaluate the concentration dependent effects (Figs. 3 and 4) on cell viability and to calculate the IC₅₀ values (Table 2). Therefore, all the cell lines were used in non-confluent cell cultures (confluence at treatment beginning: 60–80 %).

BORM induced linear concentration dependent effects on the osteosarcoma cell lines MG-63 and Saos-2 as well as on the breast cancer cell lines BT-20 and MCF-7 (Fig. 3). The highest concentration of 100 μg/ml reduced the cell viability by 60–80 % in all tumorigenic and control cells. However, in comparison with cancer cells, the control cells POB-110 and MCF-12A were not that strongly affected by BORM in concentrations below 50 μg/ml. This is confirmed by the calculation of the IC₅₀ values: BORM exhibited the lowest values for BT-20 and Saos-2 (~60 and 34 μg/ml), indicating a pronounced cytotoxicity (Table 2). In contrast, VSM caused only a linear concentration dependent effect on the bone cancer cell lines MG-63 and Saos-2 leading to IC₅₀ values of 62.53 and 57.09 μg/ml, respectively. The IC₅₀ value for the control primary osteoblasts POB was significantly higher (515.76 μg/ml) indicating that VSM mediates a stronger cytotoxic impact on the bone cancer cell lines. The vitality of the non-tumorigenic mammary epithelial cell line MCF-12A was only minimally affected by VSM leading to a no calculable IC₅₀ value. On the other hand VSM displayed moderate IC₅₀ values for BT-20 and MCF-7 (132.67, 116.82). These results illustrate that the plant extract VSM has anti-tumor potential, primarily because the viability of the cancer cells is reduced and the influence on the non-tumorigenic control cells is low.

These dose dependent effects were verified by cell cycle measurements (Fig. 5a, b) and apoptosis detection (Fig. 5c) on the osteosarcoma cell line MG-63, exemplarily. In comparison with the control treatments BORM induced a slight increase of the proliferative phase G2/M, starting at a concentration of 10 μg/ml. In contrast, the VSM extract caused a linear, concentration-dependent increase in the proliferative phase G2/M and S, indicating for a G2-arrest. At a concentration of 100 μg/ml VSM more than half of the analyzed cells were detected in the G2/M phase. To verify the apoptosis induction an Annexin V/PI staining was performed (Fig. 5c, Additional file 2: Figure S2). 100 μg/ml BORM induced an increase in early and late apoptotic events up to 50 %. By contrast, VSM caused a slight shift in early apoptotic events (~ 20 %). All together, these results suggest that BORM and VSM exhibit anti-tumorigenic potential. The precise mode of action is to be analyzed in the next chapters.

To characterize morphological and cytoskeletal changes (F-actin and ß-tubulin), MG-63 cells were cultured in the presence of BORM and VSM (0.1–100 μg/ml) or vehicle control (0.1 % DMSO) for 48 h, and monitored by bright field, scanning electron and laser scanning microscopy (Fig. 6; Additional file 3: Figure S3, Additional file 4: Figure S4). Under control conditions MG-63 cells form a confluent monolayer with the typical fibroblast-like cell structure (Fig. 6a). Confocal imaging revealed that untreated MG-63 cells possess cortical actin, some well-defined stress fibers, and cell polarity as shown by the presence of lamellipodia (Fig. 6c). VSM caused a remarkable change in cell shape: cells are wider and have a larger cell surface (Fig. 6b). Furthermore, a reduced formation of the cortical cytoskeleton and a solid reinforcement of actin stress fibers through the entire cell area are visible. The stress fibers are much longer, thicker and stabilize the entire cell, so that the cell contacts are partially broken. Exposure to BORM resulted in a strong increase of vesicles in the cell nucleus environment, observed both in the bright field image as well as in the F-actin staining (Fig. 6a-c). The cells are much more stretched and spindle-shaped, resulting in a smaller cell area (Fig. 6b). The formation of the actin fibers as well as the distribution of cortical actin did not change, substantially. The formation of tubulins is neither changed after treatment with BORM nor VSM. The increased formation of vesicles after exposure to BORM as well as the strengthening of the actin skeleton after treatment with VSM can be due to metabolic alteration or cell compartment disorders. Therefore, primary cell compartment alterations were monitored by live cell imaging (Fig. 7). The number and distribution of mitochondria as well as the appearance of the endoplasmic reticulum remained almost unchanged. But in contrast, the exposure to BORM revealed an accumulation and augmentation of lysosomes while treatment with VSM reduced the amount of lysosomes, significantly. Already at a starting concentration of 1 μg/ml the number of lysosomes increased, and reached the highest number at a concentration of 25 μg/ml BORM. Higher concentrations of BORN did not further elevate the amount of lysosomes but caused a merger of lysosomes so that the size increased up to the 3–5 fold (Additional file 5: Figure S5). Similarly, treatment with BORM caused a change of the Golgi apparatus: a stronger granularisation and formation of Golgi vesicles can be observed. The staining of neutral lipids, which was only very slightly visible in the untreated cells, was strongly upregulated after VSM exposure. Many small dots of lipids could be verified around the nucleus and in the cytoplasmic area. In contrast, treatment with BORM resulted in a diffuse distribution of neutral lipids in the cell without any specific dot distribution. Up to this state, it can be concluded that BORM and VSM mediate cytotoxic effect by affecting different metabolic pathways. To discuss this more profound, various metabolic and motility-specific investigations were carried out.

To examine the cell specific mode of action of the two plant extracts, the influence on cell metabolism was investigated, firstly. Therefore, live cell monitoring of three metabolic parameters (cell impedance, O₂-consumption and extracellular acidification) was performed (Fig. 8a, Additional file 6: Figure S6). Both, the treatment with VSM and BORM for 24 h resulted in a strong reduction (VSM: 75 %, BORM: 99 %) of the mitochondrial O₂-consumption in MG-63 cells (Fig. 8a). This decrease in respiratory capacity cannot be reverted after the discontinuation of the plant extracts. In contrast, the effect on primary osteoblasts (POB) was different: VSM did not alter the respiration capacity; BORM induced a slight decrease in O₂-consumption (~ 20 %) which could not only be reverted but enhanced up to 100 % after discontinuation of the plant extract. Beside this strong effect on cellular energy metabolism, the effect on cell motility was investigated. Because VSM induced an increased formation of stress fibers (Fig. 6), the influence on cell migration and invasion was determined (Fig. 8b, c). Concentrations of 25–50 μg/ml VSM decreased the migratory activity (90 %) and the invasiveness (35 %) of MG-63 cells. BORM did not alter the cell motility, significantly but induced apoptotic signals by enhanced BCL-2 expression and proliferation reduction by PCNA repression (Fig. 8d).