In vitro and in silico investigations of the pro-apoptotic activity of Opuntia ficus-indica cladode extracts against K562 cells

Preparation of cladode extracts

Spiny and thornless cladodes were collected Trikomo regions in Northern Cyprus in July. Collected cladode were washed with water, air-dried, and then they were peeled by hand. Extracts of thorny and thornless cladodes were obtained following the previously mentioned procedure [15]. The cladodes’ thorns were removed, allowed to dry outside at 25 °C, and then sliced and crushed by a blender. Using the Soxhlet apparatus, 10 g of cladodes powdered flour was extracted with EtOH over the course of 3 h. EtOH was evaporated until dry using a rotary evaporator. The cladode extracts were freeze-dried for 36 h to get pure-dried extracts. Extract powders were dissolved in dimethyl sulfoxide (DMSO; 10 mg/mL) prior to use, and then they were promptly placed into the culture medium at the prescribed concentrations (using <0.05 % DMSO final concentration).

Cell culture, treatment and cytotoxicity assay

K-562 cell line (Cat No: CCL-243, ATCC) was cultured in RPMI-1640 (Cat No: FG1215, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10 % FBS (FBS; Cat No: FBS-11B Capricorn Scientific), 1 % l-glutamine (Cat No: K 0283, Merck millipore) and 1 % penicillin–streptomycin (Cat no: A 2212, Pan Biotech), and maintained in a humidified atmosphere of 5 % carbon dioxide at 37 °C. By using the MTT assay, the inhibitory effect of OFI extracts on cell proliferation was ascertained. In basic terms, the cells were propagated in 96-well plates. After 24 h of culture, cells were subjected to various doses of OFI (spiny and thornless extracts: 0, 10, 25, 50, 75, and 100 g/mL, respectively) for 24 and 48 h. Following treatment, each well received 10 μL of MTT (Cat No: 11465007001, Merck), and the plate was incubated for a further 4 h at 37 °C. The absorbance of DMSO-solubilized violet formazan crystals, which was directly proportional to cell viability, was measured at 570 nm with a hybrid multi-mode microplate reader (Synergy H1; Bio Tek Instruments, Inc., Winooski, VT, USA). A plot of percent cell viability vs. OFI extracts (spiny and thornless) concentration was graphed (n=3), and the concentration that gave half-maximal inhibition of the cell growth (IC₅₀) was determined graphically.

Determination of apoptosis

Apoptosis of K-562 cells upon OFI extracts treatment was determined by using both Tali apoptosis assay and MMP assay. In Tali apoptosis assay, cells were initially stained with 5 μL of annexin V Alexa Fluor 488 (Tali™ Apoptosis Kit - Annexin V Alexa Fluor™ 488 & Propidium Iodide Cat. No: A10788, Invitrogen) in darkness for 20 min. Following centrifugation, they were then stained with 1 μL of propidium iodide (PI) in darkness for 1 min. The fluorescence of both dyes was measured simultaneously by using a benchtop cell analyzer (Tali Image-Based Cytometer; Invitrogen, Carlsbad, CA, USA). Green fluorescence and red fluorescence (with or without green fluorescence) were used to differentiate apoptotic cells (annexin V-positive/PI negative) from dead cells (annexin V-positive/PI positive or annexin V-negative/PI positive), respectively. In MMP assay, cells were stained with the JC-1 probe according to the instructions provided with the test kit. The alternating fluorescence emission color of the dye was quantitated with a hybrid multi-mode microplate reader endowed with appropriate filters (Synergy H1; Bio Tek Instruments, Inc., Winooski, VT, USA). Green fluorescence and red fluorescence were used to differentiate apoptotic cells with poor mitochondrial integrity (low ΔΨm) from healthy cells with good mitochondrial integrity (high ΔΨm), respectively.

Western blotting

The effect of OFI extract on the expression of apoptotic proteins [Bax (1:200; Cat no: sc-20067, Santa Cruz Biotechnology), Bcl-2 (1:200; Cat no: sc-7382, Santa Cruz Biotechnology), and β-actin (1:200; Cat no: sc-130657, Santa Cruz Biotechnology) in K562 cells was assessed using the western blot method. K562 cells were subjected to varying concentrations of OFI extracts on a medium with 1 % serum. To prepared the whole cell extract, cells (5.0 × 10⁵ cells/mL) were lysed with 1× radio-immune-precipitation assay (RIPA) buffer containing protease inhibitor cocktail (Cat No: 11873580001; Sigma Aldrich, St. Louis, MO, USA), The amount of protein in the obtained K562 cells lysates was determined using the Lowry method and 50 g of protein were loaded into 12 % SDS-PAGE (for 90 min) and then blotted onto the membranes (Schleicher and Schuell, 0.45 m, German). All membrane was blocked with 1 % BSA including 0.05 % Tween 20 (for 1 h, at +22 °C), before incubating primary antibody for 14 h. Densitometric studies were performed using the Bio-Rad Molecular Analyst software after a 1 h incubation period with the secondary antibody on membranes treated with the primary antibody (free edition, www.totallab.com).

Molecular docking

The solution NMR structures of (i) human Bcl-2 in complex with an acylsulfonamide-based inhibitor (PDB ID: 2O21) [16], (ii) human Bcl-x_L in complex with the same dual Bcl-2/Bcl-x_L inhibitor (PDB ID: 2O1Y) [16] and (iii) human Mcl-1 in complex with the selective inhibitor topotoclax (PDB ID: 6OVC) [17] were retrieved from RCSB Protein Data Bank [18]. The 3D conformers of narcissin (compound ID: 5481663) and nicotiflorin (compound ID: 5318767) were downloaded from PubChem [19]. The ligands were docked in moderate-precision mode onto the proteins, which were prepared via the Protoss optimization routine [20, 21], using the JAMDA fully automated protein–ligand docking tool [22], [23], [24] available at https://proteins.plus/. The sphere within a radius of 6.5 Å from the native (co-crystallised) inhibitor was used to define the conformational search space. Based on the computed JAMDA score and additional structural parameters listed for known flavonoids, the optimal docking solution was chosen [25]. Favorable non-covalent interactions between the anti-apoptotic proteins and the flavonol glycosides were estimated by using Discovery Studio v16.1.0 (Dassault Systèmes BIOVIA Corp., USA).

Statistical analysis

GraphPad Prism 3.0 (GraphPad Software, San Diego, CA, USA) was used for statistical analysis. All data were reported as mean±SD One-way analysis of variance (ANOVA) was employed to compare multiple groups followed by Bonferroni’s multiple comparison test to discover the differences between multiple groups. Significant values of p<0.05 were considered.

Effects of OFI extracts on K562 cells viability

To evaluate the anti-cancer activity of spiny and thornless ethanol extracts of the OFI cladode with various concentrations (10–100 μg/mL) were screened on K562 myeloid cancer cell lines for 24 and 48 h using the MTT assay. The spiny OFI cladode extract exhibited potent antiproliferative activities with IC₅₀ values of 44.67 μg/mL for 48 h on K562 cells (Figure 1A and B). Furthermore, the IC₅₀ value of thornless ethanol extracts of the OFI cladode after 48 h was 44.11 μg/mL in the K562 cells (Figure 1A and B).

Figure 1:

Effect of cell growth inhibition by different (A) spiny and (B) thornless OFI extracts. K562 cells were treated with each OFI extract (10, 25, 50, 75, and 100 μg/mL) for 24 and 48 h. The cell proliferation inhibition was measured using the MTT assay. Values are expressed as mean±SD of independent experiments.

Apoptosis is characterized by a reduction in the ΔΨm level of the mitochondrial membrane and malfunction [26]. K562 cells were exposed to varying concentrations of thornless and spiny ethanol extracts of the OFI cladode (10–100 μg/mL) for 48 h before being incubated with JC-1 dye to ascertain the role played by the mitochondria in the induction of apoptosis by OFI. Based on our results, as indicated in Figure 2A, both spiny and thornless ethanol extracts of the OFI cladode caused loss of ΔΨm level in a dose-dependent manner in K562 cells.

Figure 2:

(A) Mitochondrial membrane potential was detected by using JC-1 dye, and (B) apoptotic cell ratios were evaluated using the Tali apoptosis kit in K562 cells (n=3). ***The data was significant when compared with control group (p=0.0008). ****The data was significant when compared with control group (p<0.0001).

Furthermore, it was determined that cell proliferation is suppressed and viability is reduced in the treatment of K562 cells with OFI cladode extract obtained spiny and thornless for 48 h. Inhibition of K562 cell proliferation was determined with a decrease of about 22 and 11 %, especially after the addition of 100 μg/mL spiny and thornless OFI cladode extracts, due to increased extracts concentrations according to the cell count results using TALI image-based cytometer after trypan blue staining.

After 48 h of treatment with four different spiny OFI extracts, the highest significant variation in inhibition was 45 % at 50 g/mL. Consistent with ∼30 % reduction in cell proliferation, when annexin V/PI labelling was performed, it was discovered that 40 % of the total cell population was apoptotic at the IC₅₀ concentration (44.67 g/mL) and 45 % at 50 g/mL. The results of the JC-1 assay revealed that viability decreased by 35 % upon the addition of spiny OFI cladode extract at 50 μg/mL concentrations. The cytotoxic effect of thornless OFI cladode extract showed the inhibition of 20 % in the K562 cells after 48 h (Figure 2B). Overall, it may be possible to say that spiny OFI cladode extract is more sensitive to the cytotoxic effect of K562 cells than thornless OFI cladode extract.

Western blotting results

We investigated whether anti-apoptotic Bcl-2 and proapoptotic Bax antibodies were associated with the anti-proliferative activity of spiny and thornless OFI cladode extract using western blot. Our cell results showed that the spiny and thornless OFI cladode extract significantly inhibits the growth of K562 cells. We found that spiny OFI extract (30 and 50 μg/mL) in K562 cells exhibited increased efficacy on the Bax/Bcl-2 ratio due to decreasing effect on the level of Bcl-2 expression and increasing effect on the level of Bax expression at 48 h (p<0.001) when compared control group (Figure 3A–C). The Bax/Bcl-2 ratio increased 1.23-fold (p<0.001) at 30 μg/mL. Also, we found that thornless OFI extract (30 and 50 μg/mL) was not found to have any effect on the Bcl-2/Bax ratio at 48 h in K562 cells (Figure 3B and C). It was determined that thornless OFI extract in K562 cells exhibited at 48 h increased efficacy on the level of Bcl-2 expression (p<0.05), but not a significant increasing effect on the level of Bax expression (Figure 3C).

Figure 3:

The expression levels Bax and Bcl-2 after spiny (A) and thornless (B) OFI extract application in K562 cells. (C) The expression levels of Bax/Bcl-2 ratio after spiny and thornless OFI extract application in K562 cells. Western blots were determined using the antibodies directed against Bax and Bcl-2. The immunoblots indicated the Bcl-2 (≅ 26 kDa), and Bax (≅ 23 kDa) levels (duplicated). β-actin (≅ 43 kDa) was used to normalize the amount of protein loaded in each lane. ***The data was significant when compared with control group (p<0.001).

Protein–ligand docking and interaction profiling results

The results of our computational studies revealed that both the isorhamnetin glycoside narcissin (i.e. isorhamnetin 3-O-rutinoside) (Figure 4A) and the kaempferol glycoside nicotiflorin (i.e. kaempferol 3-O-rutinoside) (Figure 4B) could fill in the vacant space in the inhibitor-binding groove of human Bcl-2. The flavonol groups of the ligands appeared to form conventional hydrogen-bonding interactions with Gly145 and Arg146 as well as a pi–pi stacking interaction with Phe104. Arg146 was in sufficiently close proximity to the A-rings of the ligands to also form a cation–pi interaction with them. Asp111, Glu136 and Ala149 could establish conventional hydrogen-bonding interactions with the glucose and rhamnose sugar groups of the ligands. With its additional methoxyl group at the C-3′ position, narcissin was also able to interact with Tyr108 through a weaker carbon hydrogen bond. A somewhat similar bonding pattern was observed for the predicted Bcl-x_L–narcissin (Figure 4C) and Bcl-x_L–nicotiflorin (Figure 4D) complexes. Ala93, Arg100 and Gly138 appeared to form conventional hydrogen bonds with the flavonol groups of the ligands, while Glu92 and Tyr195 appeared to form conventional hydrogen bonds with the disaccharide moieties of the molecules. Phe97, which corresponds to Phe104 in Bcl-2, could be involved in a pi–pi stacking interaction with the C-rings of the ligands. Considering flavonol glycosides in complex with Mcl-1 (Figure 4D and E), Met231 and Met250 were estimated to play a key role in stabilising the bound (docked) conformations of the molecules by establishing sulphur–pi interactions with the flavonol groups of them. In addition, Val253 and Arg263 were able to establish conventional hydrogen-bonding interactions with the terminal sugar groups of the ligands.

Figure 4:

Favorable non-covalent interactions occurring between the predicted (A) Bcl-2–narcissin, (B) Bcl-2–nicotiflorin, (C) Bcl-xL–narcissin, (D) Bcl-xL–nicotiflorin, (E) Mcl-1–narcissin, and (F) Mcl-1–nicotiflorin complexes. Magenta dashed lines correspond to pi–pi stacking interactions, pink dashed lines correspond to hydrophobic interactions, lime green dashed lines correspond to conventional hydrogen bonds, light green dashed lines correspond to weak hydrogen bonds, bright orange dashed lines correspond to electrostatic interactions, and light orange dashed lines correspond to sulfur–pi interactions. The images (A)–(F) were rendered using Discovery Studio Visualizer, v16.1.0 (Dassault Systèmes BIOVIA Corp., San Diego, CA, USA).