1. Introduction
Several natural compounds from plants receive great attention for their potential application to prevent and contrast cancer development or to support conventional therapies thanks to their high efficacy and often low toxicity to normal cells [
1,
2]. Interacting with several cellular targets, natural compounds may activate signalling pathways, finally leading to the apoptosis of cancer cells, sometimes engaging in autophagic or endoplasmic reticulum (ER)-stress responses [
2,
3,
4,
5].
Among emerging bioactive compounds explored for their anticancer properties, there are cardiac glycosides, a group of molecules displaying the ability to bind the main ion transport protein pump, the Na
+/K
+-ATPase [
6]. They have long been and continue to be employed as positive ionotropic agents in congestive heart failure [
7], but, more recently, the potential anticancer activities of some of them, for example, digitoxin, digoxin or ouabain, have been demonstrated in vitro and in vivo [
8,
9]. From the chemical point of view, cardiac glycosides present a steroid nucleus with a lactone moiety at position 17 and a sugar moiety at position 3. The type of lactone moiety allows defining two subgroups, cardenolides and bufadienolides, whereas further subfamilies are defined on the bases of the sugar moiety [
8].
The Asclepiadaceae family represents a rich source of cardenolides with peculiar features [
10]. The A/B rings of the steroid nucleus are transfused, giving more flat structures and a more potent binding ability to the Na
+/K
+-ATPase pump [
11]. Moreover, the sugar moiety is linked at position 3 by a double link, thus generating a dioxanoid structure [
8].
Several cardenolides have been isolated and characterised from roots and leaves of
Pergularia tomentosa [
11,
12,
13,
14], a tropical herb species of the Asclepiadaceae family, native to the Middle East, Egypt and South Africa. Studies on cytotoxic, antiproliferative, and pro-apoptotic properties of cardenolides from
P. tomentosa, as well as on their strong inhibitory activity of the Na
+/K
+-ATPase pump have been reported, highlighting their prospective usefulness in cancer management [
11,
13,
15,
16,
17].
The present work aimed to investigate, in a deeper way, the potential anticancer properties of some cardenolides isolated from the aerial parts of
P. tomentosa [
13]. In particular, we focused attention on five selected compounds, three of which were identified in the roots and the aerial parts of
P. tomentosa, i.e., 16α-hydroxycalotropin (
2), calotropin (
4), and calactin (
5), while 12β,6′-dihydroxycalotropin (
1), and 12β-hydroxycalactin (
3) were only found in the aerial parts of
P. tomentosa (
Figure 1). In particular, compounds
1–3 were chosen for their high antiproliferative activity in PC3 cells [
13]. Here, we investigated the cytotoxicity of these selected compounds in cancer cell lines not previously employed and explored their antiproliferative ability in several cell cultures and their pro-apoptotic and pro-autophagic activities in a hepatic cancer cell line.
3. Discussion
In the present work, we investigated the mechanism at the bases of cytotoxicity of some cardenolides obtained from leaves of
P. tomentosa. Cardenolides, a subgroup of cardiac glycosides, are abundantly present in the roots and aerial parts of several plants [
14]. The anticancer properties of cardenolides (and, more in general, of cardiac glycosides) have been studied in different cell models, and there is a lot of evidence of their efficacy [
8,
18,
19]. The mechanisms of action of these phytochemicals may include the inhibition of proliferation, the induction of apoptosis or autophagy and the sensitisation to chemotherapy [
9,
10,
20]. Previous studies performed with cardenolide-rich extracts or pure cardenolides from
P. tomentosa demonstrated that they not only exerted a cardiotonic activity [
21] but also reduced the cell viability of cancer cells, had a pro-apoptotic effect on Kaposi’s sarcoma cells and exerted an antiangiogenic effect both in in vitro and in vivo models [
11,
12,
13,
17,
22]. The inhibition of Na
+/K
+-ATPase activity has been considered the possible main mechanism responsible for the anticancer properties of these bioactive compounds [
12].
In our work, we focused on five doubly linked cardenolides, a subclass of cardenolides found in
Pergularia genus [
23]. These compounds were selected from a bigger group already tested in a previous work by Hosseini et al. [
13] demonstrating that known cardenolides, together with new isolated ones, were able to decrease the cell viability of five different human cancer cell lines. Interestingly, IC
50 for each compound was very variable from one cell line to another, indicating that different cancer cells presented their sensitivity towards these compounds [
13].
Generally, in cardiac glycosides isolated from plants belonging to the genus
Digitalis (family Scrophulariaceae) and
Strophanthus (Apocynaceae), rings A/B and C/D are
cis fused, while rings B/C are
trans fused [
24,
25]. Such ring fusion gives the aglycon nucleus of these cardiac glycosides a characteristic “U” shape [
11]. On the other hand, in cardiac glycosides produced by plants from the family Asclepiadaceae, such as
Pergularia, the A/B rings are
trans fused, resulting, thus, in rather flat structures. It is worth noting that, whereas cardiac glycosides from
Digitalis and
Strophanthus species contain sugar units linked through the 3β-OH of the steroid aglycon (single link), compounds produced by plants from the Asclepiadaceae are characterised by a sugar moiety linked to the 2α- and 3β-positions of the aglycon by hemiacetal and acetal functions, respectively, generating a “dioxanoid” structure [
11,
13]. The sugar unit can be represented by a 4,6-dideoxyhexosulose (
1) and a modified form of the latter, 4-deoxyhexosulose (
2–5). Moreover, compounds
1–5 can be considered as belonging to two groups: the calotropin derivatives (
1,
2 and
4) and the calactin derivatives (
3 and
5), differing in the configuration at C-3′. These structural features are rarely found in cardenolides from families different from Asclepiadaceae.
An MTT assay analysed the cytotoxic activity of these compounds in two cancer cell lines not previously tested, i.e., HepG2 and Caco-2. Moreover, an immortalised nontumoral cell line (MRC5) and a culture of normal cells (HUVEC) were used. Our findings showed that compounds were cytotoxic towards all cells tested, compound 3 being the less active. Normal/noncancer cells and cancer cells were similarly affected by compounds 1, 2 and 3, whereas compounds 4 and 5 were more cytotoxic for normal/noncancer cells than for cancer cells. These findings indicate a non-specific action of the tested compounds toward cancer cells. However, intestinal- and hepatic-cancer-cell lines appeared very sensitive, particularly to the activity of calactin and calotropin. Calculated IC50 confirmed that each cell type displayed a different sensitivity and also revealed that HepG2 cells were more sensitive to the effect of calactin (5) but less sensitive toward the calactin derivative (3) than Caco-2 cells. Focusing the attention on HepG2 cells, all molecules used at 1 μM reduced the entry into the S-phase. The effect was more marked for compound 5, in line with the result of MTT assays. The ability to reduce cell migration also highlighted the potential to interfere with signalling pathways in cancer cells.
Furthermore, also apoptosis could contribute to reducing cell viability. We detected an increase in the appearance of cleaved caspase 3 as a consequence of the treatment with each compound, with a more marked signal after treatments with compounds 4 and 5. The caspase 3 activity assay confirmed this trend, even if compound 3, at the condition tested, did not seem to affect enzyme activity. Microscopic visualisation of the presence of apoptotic nuclei by the TUNEL technique was in line with biochemical assays. The Western blot analysis on HepG2 samples showed an increase in the pro-apoptotic protein p53 for all compounds, except for calactin (5), which reduced the p53 level. This finding highlighted that calactin had a mechanism of action that, in part, was different from the other tested compounds.
Several natural products, including cardenolides, can induce an ER-stress response, that can cause a mechanism of programmed cell death [
2,
5,
26]. However, on the contrary, natural products may also inhibit proteins related to the unfolded protein response, thus reducing the adaptative response of the cell [
5]. Herein, two early markers of ER stress, XBP1 and GRP78, were investigated, but the activation of an unfolded protein response was not observed. Conversely, we observed a slight decrease in GRP78 level, which is compatible with this protein’s pro-survival role; indeed, in several conditions, a reduction in GRP78 has been associated with increased apoptosis [
27,
28].
A desirable feature of an anticancer drug may be the ability to modulate autophagy [
1,
4]. Autophagy is a process that, at the basal level, promotes cellular homeostasis, but in stressful conditions, it may act as a protective response or a pro-death pathway, depending on the cell type, the environmental context and the nature of the stressor [
29]. The pro-death role of autophagy can occur by inducing the apoptotic mechanism or without the involvement of the apoptotic machinery [
2,
29,
30]. Several cardiac glycosides have displayed a pro-autophagic activity as part of their anticancer effect [
31,
32,
33]. We found that
P. tomentosa compounds induced autophagy, as highlighted by the reduction in p62 level and the appearance of the LC3-II form. The phenomenon was more evident for calactin than for other compounds. This could be the reason of the reduction in p53 level by calactin, in line with the reported suppression of p53 by autophagy [
30,
34]. A reduction in p53 protein synthesis has been already reported as a mechanism to mediate the anticancer effects of other cardiac glycosides [
35]. Thus, we attest to the induction of apoptosis and autophagy in the same cell culture, and autophagy could precede apoptosis [
30].
5. Materials and Methods
5.1. Extraction and Isolation Procedures of Compounds 1–5 from the Aerial Parts of P. tomentosa
Aerial parts of
P. tomentosa L. were collected near Nehbandan, Iran, in April 2018. The aerial parts of
P. tomentosa (70 g) were dried and extracted at room temperature using solvents of increasing polarity including hexane (0.5 L for 3 days, two times), CHCl
3 (0.5 L for 3 days, two times), and MeOH (0.5 L for 3 days, three times). After filtration and evaporation of the solvent to dryness in vacuo, 3 g of a crude MeOH extract was obtained. The extract was fractionated using a Sephadex LH-20 (Pharmacia) column (100 × 5 cm), with MeOH as mobile phase, affording 53 fractions (8 mL), as monitored by TLC. Fractions were analysed by an RP-HPLC-UV system. The elution gradient was obtained using water with 0.1% formic acid as eluent A and acetonitrile with 0.1% formic acid as B at a flow rate of 2.0 mL/min. For fractions 13−15 (950 mg), the HPLC gradient started at 10% B, and after 10 min % B was at 30%; after 16 min, it was at 54%, after 7 min, it was at 63%; and after 11 min, it was at 100%, holding it for 10 min. In this way, the compounds were as follows:
1 (3.4 mg, t
R = 14.25 min),
2 (4.2 mg, t
R = 16.20 min),
3 (4.5 mg, t
R = 18.36 min),
4 (4.5 mg, t
R = 22.90 min), and
5 (3.0 mg, t
R = 24.82 min). Compounds
1−5 were identified by comparison of their
1H and
13C NMR data with those reported previously in the literature [
12]. NMR spectroscopic data were acquired in MeOH-
d4 (99.95%, Sigma-Aldrich, St. Louis, MO, USA) on a Bruker DRX-600 spectrometer (Bruker BioSpin GmBH, Rheinstetten, Germany) equipped with a Bruker 5 mm TCI CryoProbe at 300 K. Data processing was carried out with Topspin 3.2 software. The purity of these compounds (>99%) was determined by HPLC analysis. Stock solutions of
P. tomentosa compounds were prepared in dimethyl sulfoxide (DMSO) and stored at −20° C in small aliquots.
5.2. Cell Cultures
The human liver hepatocellular carcinoma cell line, HepG2, and the human adenocarcinoma of colon cell line, Caco-2, were cultured in Eagle’s Minimum Essential medium supplemented with 1% (v/v) non-essential amino acids, 0.2 mM L-glutamine, 50 units/mL penicillin, 50 μg/mL streptomycin and 10% (v/v) or 20% (v/v) fetal bovine serum (Sigma Aldrich, Milan, Italy) for HepG2 or Caco-2, respectively. The human embryonal lung fibroblast cell line, MRC5, was cultured in Dulbecco’s Modified Eagle medium (Sigma Aldrich, Milan, Italy) supplemented with 10% (v/v) fetal bovine serum, 0.2 mM L-glutamine, 50 units/mL penicillin and 50 μg/mL streptomycin. HUVEC cells were cultured in endothelial growth medium-2, supplemented with fetal bovine serum (2%), VEGF (0.1%), rH FGF-B (0.4%), rH EGF (0.1%), GA-1000 (0.1%), hydrocortisone (0.04%), R3-IGF-1 (0.1%), heparin (0.1%) and ascorbic acid (0.1%). Cells were maintained at 37 °C in a 5% CO2, 95% air-humidified atmosphere and passaged twice a week. Cell lines were obtained from Interlab Cell Line Collection, Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy. HUVEC were obtained from Lonza (Milan, Italy).
5.3. Cell Viability Assay and IC50 Calculation
An MTT assay was used to analyse cell viability. Cells were seeded in 96 wells at the density of 2 × 10
4/cm
2 and, after 24 h, they were treated with
P. tomentosa compounds or with the vehicle (DMSO). After a further 24 h of treatment, 0.5 mg/mL of MTT was added to the cell medium and incubated for 1.5 h at 37 °C. The resulting formazan crystals were dissolved in 100 μL of DMSO and absorbances were measured at 595 nm and 655 nm. The background signals at 655 nm were subtracted from 595 nm signals, and data were expressed as a percent of cell viability. The concentration of each compound that is required for 50% inhibition in vitro (IC
50) was calculated as described in Husseini et al. [
12]. IC
50 values were obtained for Caco-2, HepG2 and MRC5 cells by performing MTT assays with concentrations ranging from 0.01 μg/mL to 5 μg/mL.
5.4. Proliferation Assay
Cell proliferation was analysed by a BrdU (Roche Diagnostics SpA, Monza, Italy) incorporation assay, as previously described [
25]. Briefly, HepG2 cells, seeded at adensity of 3 × 10
3/cm
2 on round glass cover slips, were treated for 24 h with
P. tomentosa compounds at 1 μM. Then, BrdU was added to cell medium (final concentration, 100 μM) for 1 h. Cells were fixed (in 3% paraformaldehyde, 10 min), permeabilised (in 0.2% Triton X-100, 5 min) and treated with HCl 1.5 N (5 min). Finally, cover slips were incubated with a monoclonal antibody against BrdU (Sigma-Aldrich, Milan, Italy), 1:100, and with a TRITC-conjugated secondary antibody, 1:100. After staining with Hoechst, cover slips were mounted with Mowiol (Sigma, Milan, Italy). Stained cells were observed with an AxioSkop 40 fluorescent microscope (Carl Zeiss MicroImaging, Inc., Jena, Germany). Images were acquired with Axiocam MRc5 and processed with ImageJ Launcher. The number of cells into S-phase was expressed as the ratio between the number of cells incorporating BrdU and the total number of cells.
5.5. Scratch-Wound-Healing Assay
Cells were seeded at a density of 5 × 104 cells/well in 24-well plates and allowed to grow for 72 h, reaching confluence. Mitomycin C (10 μg/mL) was added to inhibit proliferation, and then a sterile 2–10 μL pipette tip was used to perform a vertical scratch in the monolayer of each well. Detached cells were removed by washing with phosphate-buffered saline. Then, 300 μL of fresh medium containing P. tomentosa compounds (1 μM) or vehicle was added to each well and incubated for 72 h. Images of migrating cells were taken at 0, 24, 48 and 72 h, by using an Olympus CKX41 fluorescent microscope (Olympus Italia srl, Segrate, Italy) and elaborated by the ImageJ software. For each scratch, wound width was calculated as the average distance between the edges in four different sites. Then, reductions in wound wights at 24, 48 and 72 h were expressed as % of the width at time zero.
5.6. Transwell Migration Assay
Transwell filters (coated with collagen) in serum-free media containing 0.1% of BSA, with a PET membrane with 8 μm pores (BD Biosciences, Milan, Italy), were rehydrated for 2 h at 37 °C. HepG2 were seeded in the upper chambers at a density of 75 × 103 cells in MEM with 0.1% BSA and treated with or without 1 µM of Pergularia compounds. MEM−0.1%BSA, supplemented with 10% FBS, was added to the lower chamber as chemoattractant. Control wells without 10% FBS were included to assess random migration. After incubation at 37 °C in 5% CO2 for 18 h, the migrated cells were fixed with 4% paraformaldehyde for 5 min, permeabilised with methanol for 20 min and stained with crystal violet 0.5% for 15 min at room temperature. The cells on the lower side of the filter were allowed to dry before counting. Eight separate bright-field images were randomly acquired of each filter using an Olympus CKX41 Image Analyzer. The cells in each image were counted and analysed compared to control-transfected cells.
5.7. Caspase-3 Assay
To evaluate apoptosis, a caspase-3 colorimetric assay was performed. Briefly, HepG2 cell lines were seeded in plates of 60 mm diameter at a density of 3.5 × 104/cm2 and, after 24 h, treated with P. tomentosa compounds (1 μM each) for 24 h. Then, cells were harvested in phosphate-buffered saline, collected by centrifugation and lysed in a buffer containing 50 mM HEPES, 0.1% CHAPS, 10 mM dithiothreitol, 100 mM NaCl, 1 mM EDTA and 10% sucrose. Proteins (30 μg) were incubated for 2 h at 37 °C in a reaction mixture containing the caspase-3 substrate (acetyl-Asp-Glu-Val-Asp, labelled with p-nitroanilide) (Sigma-aldrich, Milan, Italy) at a concentration of 0.2 mM. Finally, the release of free p-nitroanilide was monitored spectrophotometrically at 405 nm. Enzyme activity was expressed as percentage of absorption of treated cells versus vehicle-treated cells.
5.8. Western Blot Analyses
To analyse the protein levels of p53, caspase-3, GRP78, LC3 and p62, HepG2 cells were treated with P. tomentosa compounds (1 or 2 μM) or with DMSO for the indicated times; then, Western blot analyses were performed. Briefly, cells were lysed in RIPA buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 0.1% sodium dodecyl sulphate, 1% triton X-100, 1 mM orthovanadate, and a cocktail of inhibitors (Sigma-Aldrich, Milan, Italy)). After separation on a sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transfer to a poly-vinylidene fluoride membrane (Euroclone, Milan, Italy), the following primary antibodies were used, at a dilution of 1:1000 in tris-buffered saline containing 1% non-fat dry milk, overnight at 4 °C: mouse anti-p53 antibody, mouse anti-MAP LC3 β antibody, mouse anti-p62 antibody (DBA, Milan, Italy), mouse anti-GRP78 antibody, and mouse anti-caspase 3 antibody (Thermo Fisher Scientific, Milan, Italy). For normalisation, a mouse anti-GAPDH antibody (Microtech, Naples, Italy) was used. A horseradish-peroxidase-conjugated anti-mouse secondary antibody (Bio-Rad laboratories S.r.l, Milan, Italy) was used for 1 h; finally, immunocomplexes were revealed using a chemiluminescence detection kit (Microtech, Naples, Italy) according to the manufacturer’s instructions.
5.9. TUNEL Assay
Microscopic evaluation of apoptosis was performed by a TUNEL assay, using a commercial kit (Fragment End Labelling (FragEL™), Jena, Germany) DNA Fragmentation Detection Kit, Sigma-Aldrich, Milan, Italy) according to the manufacturer-provided protocol, with some modifications. Briefly, cells were seeded on glass cover slips placed in 24-well plates at a density of 3.5 × 103/cm2 and, after 48 h, they were treated with 1 μΜ of each compound for a further 24 h. Cells were fixed in 3% paraformaldehyde for 10 min, permeabilised in 0.2% Triton X-100 for 5 min and treated with HCl 1.5 N for 8 min. Finally, cover slips were equilibrated in equilibration buffer (10 min) and incubated with label mix containing the TdT enzyme for 1 h at 37 °C, then mounted with mounting media. Microscope observations were performed by the AxioSkop 40 fluorescent microscope. Images were acquired with Axiocam MRc5 (Carl Zeiss MicroImaging, Inc., Jena, Germany).
5.10. XBP1 Splicing Detection
To detect the unspliced and the spliced form of XBP1, we followed the protocol described by Martucciello et al. [
25]. After treatments for 4 h with 1 μM of each compound, or DMSO, or thapsigargin (1 μM), RNA was isolated and retro-transcripted. PCR analyses were performed with the following primers: 5′- CCTGGTTGCTGAAGAGGAGG-3′; 5′-CCATGGGGAGATGTTCTGGAG-3′. After 35 cycles (heating at 94 °C for 30 s, annealing at 58 °C for 45 s, and polymerisation at 72 °C for 60 s), amplified cDNA was visualised on 2.5% agarose gel, stained with an ethidium bromide solution.
5.11. Statistics
All data were expressed as means ± standard error (SE) of at least 3 independent experiments conducted in triplicates. Statistical analysis was performed using the Student’s t-test. In all experiments, differences were considered to be statistically significant at p < 0.01 or at p < 0.05.