Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: J Cell Biochem. 2021 Sep 22;122(12):1903–1914. doi: 10.1002/jcb.30150

Cytotoxic effect of carbohydrate derivatives of digitoxigenin involves modulation of plasma membrane Ca2+-ATPase

Jéssica Martins de Moura Valadares 1, Sumit O Bajaj 2, Hongyan Li 2, Hua-Yu L Wang 2, Simone Cavalcante Silva 2, Israel José Pereira Garcia 1, Duane Gischewiski Pereira 1, Pedro Azalim 3, Luis Eduardo Menezes Quintas 3, François Noël 3, Vanessa Faria Cortes 1, George Augustine O'Doherty 2, Leandro Augusto Barbosa 1
PMCID: PMC8671332  NIHMSID: NIHMS1740916  PMID: 34553411

Abstract

Cardiac glycosides, such as digoxin and digitoxin, are compounds that interact with Na+/K+-ATPase to induce anti-neoplastic effects; however, these cardiac glycosides have narrow therapeutic index. Thus, semi-synthetic analogs of digitoxin with modifications in the sugar moiety has been shown to be an interesting approach to obtain more selective and more effective analogs than the parent natural product. Therefore, the aim of this study was to assess the cytotoxic potential of novel digitoxigenin derivatives, digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2), in cervical carcinoma cells (HeLa) and human diploid lung fibroblasts (Wi-26-VA4). In addition, we studied the anticancer mechanisms of action of these compounds by comparing its cytotoxic effects with the potential to modulate the activity of three P-type ATPases; Na+/K+-ATPase, sarco/endoplasmic reticulum Ca2+-ATPase (SERCA), and plasma membrane Ca2+-ATPase (PMCA). Briefly, the results showed that compounds 1 and 2 were more cytotoxic and selectivity for HeLa tumor cells than the non-tumor cells Wi-26-VA4. While the anticancer cytotoxicity in HeLa cells involves the modulation of Na+/K+-ATPase, PMCA and SERCA, the modulation of these P-type ATPases was completely absent in Wi-26-VA4 cells, which suggest the importance of their role in the cytotoxic effect of compounds 1 and 2 in HeLa cells. Furthermore, the compound 2 inhibited directly erythrocyte ghosts PMCA and both compounds were more cytotoxic than digitoxin in HeLa cells. These results provide a better understanding of the mode of action of the synthetic cardiac glycosides and highlights 1 and 2 as potential anticancer agents.

Keywords: Cardiac Glycosides, Digitoxin, Digoxin, HeLa Cells, P-type ATPases, Sodium-Potassium-Exchanging ATPase

Graphical Abstract

graphic file with name nihms-1740916-f0001.jpg

Our data Showed that compounds 1 and 2 were more cytotoxic and selectivity for HeLa tumor cells than the non-tumor cells Wi-26-VA4. While the cytotoxicity in HeLa cells involves the modulation of Na+/K+-ATPase, PMCA and SERCA, the modulation of these P-type ATPases was completely absent in Wi-26-VA4 cells. Furthermore, the compound 2 inhibited directly erythrocyte ghosts PMCA and both compounds were more cytotoxic than digitoxin in HeLa cells.

Introduction

Cardiac glycosides (CGs) are compounds known for their cardiotonic and antineoplastic effects on various cell lines. CGs are able to interact with the Na+/K+-ATPase and use this enzyme as the receptor to elicit downstream effects that lead to programmed cell death of cancer cells (López-Lázaro, 2007; Newman, Yang, Pawlus, & Block, 2008). The structure of CGs consists of a steroid group, a sugar moiety attached to the steroid nucleus at C-3 and a lactone ring at C-17. The five or six membered lactone ring classifies these compounds into cardenolides or bufadienolides respectively(Cornelius, Kanai, & Toyoshima, 2013).

Many CGs have already been synthesized with changes in the lactone ring, in the steroidal nucleus and in the sugar moiety (Kulkarni et al., 2016; Rocha et al., 2014; Souza Gonçalves et al., 2019; Wang, Rojanasakul, & O’Doherty, 2011). It has been reported that the number and positions of the hydroxyl groups present in the steroid nucleus affect the interaction with Na+/K+-ATPase (Cornelius et al., 2013). Kanai et al., 2012 raise the possibility of conferring α2 specificity by steroid core modifications that favor the formation of hydrogen bonds with Ser119 in α2 (Kanai et al., 2021). Digoxin derivatives with substitutions in the lactone ring have also been shown to improve the selectivity to Na+/K+-ATPase isoforms (Pessôa et al., 2018).

The number and type of sugar residues can contribute positively through hydrogen bonds with the enzyme and increase the affinity in the equilibrium bond, but they can also be a limitation if they become obstacles to reaching the end of the binding cavity (Kanai et al., 2021). In addition, the inhibitory and cytotoxic potential of the compounds depends on the type, stereochemistry and size of their sugar portion, with the monosaccharide steroid being more potent than di- and trisaccharides, respectively (Azalim et al., 2020; Iyer et al., 2010; H.-Y. L. Wang et al., 2011; Hua Yu Leo Wang et al., 2011; H.-Y. L. Wang et al., 2011).

Digoxin and digitoxin are used to treat congestive heart failure. While both compounds share the same sugar residue (tris-β-D-digitoxose), digoxin presents an unique C-12 hydroxyl group, which moderately differentiates its pharmacological profile from digitoxin (Haux, 1999). The modification of digitoxin sugar moiety has proved to be a successful approach to developing more selective and effective anticancer compounds by induction of apoptosis and cell cycle arrest (Elbaz, Stueckle, Tse, Rojanasakul, & Dinu, 2012; Zhou & O'Doherty, 2008).

In the context of cancerous cells, P-type ATPases such as Na+/K+-ATPase, plasma membrane Ca2+-ATPase (PMCA) and sarco/endoplasmic reticulum Ca2+-ATPase (SERCA) have been studied for their differences in both genetic and protein expression level that are closely associated with the functional ablation in behavior and survival of cancer cells, which makes these enzymes desirable therapeutic targets for cancer (Kurnellas, Nicot, Shull, & Elkabes, 2005; Papp et al., 2012; Wu et al., 2013). Here, we report the anticancer activity of new digitoxigenin derivatives, digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2), in HeLa cells, and relate their cytotoxic effects to the modulation inhibitory expression and activity of P-ATPases as part of our effort to elucidate the mechanisms of anticancer action of these CGs.

Materials and methods

Compounds

Two digitoxigenin derivatives (digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2) were used in this work. The compounds were synthesized according to the method of Zhou and O'Doherty who have used a de novo approach to carbohydrates to build mono-, di- and trisaccharide libraries of cardiac glycoside analogues (Zhou & O'Doherty, 2008). The purity of the compounds was determined by high performance liquid chromatography (HPLC) (Figure S2).

Cell culture

Cervical carcinoma cells - HeLa [ATCC®CCL-2TM] and human diploid lung fibroblast - WI26 VA4 (ATCC® CCL95.1™) were grown in sterile growth medium DMEM-HG (Sigma-Aldrich, MO, USA) supplemented with 10% fetal bovine serum (GIBCO), and 0.1% gentamicin. Cells were seeded and incubated in incubator atmosphere at 37°C and 5% CO2.

Cell viability tests with MTT assay

Cells were cultured in 96 wells plate (1x105 cells/ml) for 24 hours until complete adherence, and then incubated with increasing concentrations of 1, 2, 3 for 24 h. After incubation 100 μL of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Sigma, St. Louis, MO, USA) 0.5 mg/ml was added in each well. The precipitated formazan was solubilized in 50 μL of DMSO in an ELISA plate reader (BIOTEK) at 550 nm. Cytotoxicity was scored as the percentage of reduction of absorbance, relative to untreated control cultures (Hilário et al., 2011).

Cell membrane preparation

2.25x107 cells were inoculated into each culture flask and treated with 2 (100 nM and 200 nM), 1 (65 nM and 100 nM) and 3 (100 nM, 9 μM, 20 μM). After 24 h of treatment, cells were removed from culture flasks with a membrane preparation buffer (6 mM Tris (pH 6.8), 20 mM imidazole, 0.25 M sucrose, 0.01% SDS, 3 mM EDTA and 2 mM phenylmethylsulfonyl fluoride (PMSF)). The cells were macerated in a potter-Elvehjem homogenizer. After centrifugation at 10,000 g at 4°C for 20 min, the supernatant was centrifuge at 70,000 g for 1 h and the pellet was resuspended in 300 μL buffer.

Na+/K+-ATPase pig kidney preparation

Kidneys from pigs were excised and used as previously described (Noël et al., 2018). A purified preparation was obtained, with a specific activity of 210 μmol Pi. mg−1 protein. h−1. The protein concentration was determined and adjusted for 3.7 mg/ml.

Erythrocyte plasma membrane preparation (ghosts)

Red blood cell ghosts were prepared according to Rega et al. (Rega, Garrahan, Barrabin, Horenstein, & Rossi, 1979). The blood concentrates were spun at 6,500 g for 10 min at 4 °C. The precipitates were dissolved in a solution of 20 mM Tris–HCl (pH 7.4), 130 mM KCl, and 0.6 mg/ml PMSF. This suspension was spun at 6,500 g for 10 min at 4°C. The cells were submitted to lysis by freezing in liquid nitrogen followed by thawing at room temperature (25°C). The lysates were then mixed with 5 mM HEPES (pH 7.4), 1 mM EDTA and 0.6 mg/ml PMSF and spun again at 9,000 g for 10 min at 4 °C. This centrifugation step was repeated four times, and the resulting pellet was resuspended in 10 mM HEPES (pH 7.4), 130 mM KCl, 0.5 mM MgCl2, and 0.05 mM CaCl2, and finally spun at 9,000 g for 10 min, mixed with a small volume of this last buffer, and stored in −80 °C until use (Maia et al., 2014; Sousa et al., 2015). It is worth mentioning that a free and informed consent form was applied to all volunteers involved in the research (Approval by the Ethics Committee 07004713.9.0000.5545).

ATPase activity determinations

ATPase activity was determined by measuring inorganic phosphate (Pi) that was released by ATP hydrolysis. The measurement of Pi was done through colorimetric method of Fiske (Fiske & Subbarow, 1925). Reactions were initiated by the addition of 3-4 mM ATP in a final volume of 50 μL and stopped after 60 min by addition of 1% SDS.

Maximal specific Na+/K+-ATPase activity

Reaction medium used to measure the maximal Na/K-ATPase activity was composed of 40 mM Tris (pH 7.4), 20 mM KCl, 4 mM MgCl2 and 120 mM NaCl, 3 mM ATP and 5 μg protein. The reaction was performed in the presence and absence of 1 mM ouabain (Maia et al., 2014).

Ghost PMCA activity

Reaction medium consisted of 0.2 mM CaCl2, 80 mM KCl, 0.2 mM EGTA, 20 mM Tris-HCl pH 7.4, 0.5 mM MgCl2, 3 mM ATP, 20 μg protein in each well and 1 mM ouabain to suppress Na/K-ATPase activity. The activity was measured in the presence of 2 μg/mL calmodulin (Cortes, Ribeiro, Barrabin, Alves-Ferreira, & Fontes, 2011). The enzyme was incubated for 60 min with each compound at different concentrations.

Culture cell PMCA activity

Reaction medium was composed of 50 mM Tris (pH 7.4), 8 mM MgCl2, 120 mM KCl, 1 mM EGTA, 1.008 mM CaCl2 to yield the required 17.5 μM free Ca2+ concentration calcium, 5 μg protein in each well, 1 mM ouabain and 50 nM thapsigargin to suppress Na/K-ATPase activity and SERCA, respectively. The activity were initiated by the addition of 3 mM de ATP and it was measured in the presence and absence of Ca2+ to subtract from Mg2+-ATPase activity and in the presence and absence of 2 μg/ml calmodulin (Kosk-Kosicka, 1999).

Culture cell SERCA activity

Reaction medium consisted of 100 mM Tris-HCl (pH 7.5), 100 mM KCl, 5 mM MgCl2, 4 mM ATP, 0.11 mM EGTA, 114 μM CaCl2 (pH 7.0), 5 μM free Ca2+ concentration (Ulianich et al., 2006) and 5 μg protein in each well. The reaction was performed in the presence and absence of 50 nM thapsigargin. The free Ca2+ concentration was calculated using the CHELATOR program (Schoenmakers, Visser, Flik, & Theuvenet, 1992).

Na+/K+-ATPase inhibition

In a final volume of 500 μL, the reaction was started by adding 3.7 μg pig kidney preparation to a medium containing 94 mM NaCl, 3 mM MgCl2, 3 mM KCl, 3 mM ATP-Na2, 1 mM EGTA, 20 mM maleate-Tris buffer (pH 7.4 at 37°C) and increasing concentrations of digitoxigenin derivatives. In this condition, different from the one used to estimate the maximal activity, the Na+/K+-ATPase activity was 87 μmol Pi. mg−1 protein. h−1. After 2 h, the reaction was stopped by addition of 1 mL of ice-cold Fiske solution and the absorbance determined by spectrophotometry (650 nm) 20 min later. The specific Na+/K+-ATPase activity was obtained by subtracting the basal activity measured in the presence of 1 mM ouabain. Curve fitting and determination of parameters were determined by non-linear regression analysis of the data (GraphPad Prism software) assuming the classical concentration–response curve model in order to estimate the concentration inhibiting 50% of maximal activity (IC50).

Immunoblotting

The cell extract samples were made according to the methodology described (Shoshani, 2004) and were prepared with the same amount of protein for SDS-PAGE electrophoresis on a Bio-Rad miniProtean III apparatus (Bio-Rad, Hercules, CA, USA) and then the proteins that ran in the gel were transferred to nitrocellulose membrane (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The Ponceau red method was used to ensure equal loading protein for all samples. The membrane was then blocked for 1 h, washed with Tris-buffered saline plus 0.1% Tween-20 (T-TBS) and incubated overnight with monoclonal antibodies specific for α1-Na/K-ATPase (sc-21712, Santa Cruz Biotechnology, Santa Cruz, CA, USA), PMCA4 antibody (JA9 - MA1-914, Thermo Fisher), β-actina (sc81178, Santa Cruz Biotechnology). The next day the membrane was washed with T-TBS and incubated for 2 h with peroxidase enzyme conjugated secondary antibody (sc-516102, Santa Cruz Biotechnology). The membrane was developed using an equimolar mixture of two solutions (Solution 1: Tris 100 mM pH 8.5, 2.5 mM Luminol and 0.396 mM p-coumaric acid; Solution 2: 100 mM Tris pH 8.5 and 0.06% H2O2). Immunoblots were quantified by densitometry using the ImageJ software (http://rsb.info.nih.gov/ij). Several exposure times were analyzed to ensure the linearity of the band intensities.

Statistical analyses

The statistical analyses were performed with GraphPad Prism software, version 5. One-way ANOVA test was used to compare more than two means, followed by Dunnet's post hoc test. Two-way ANOVA was used to compare more two means when there were two or more variables, and was followed by Bonferroni post-hoc test. P<0.05 was considered statistically significant.

Results

Direct effect on Na+,K+-ATPase activity

The compounds digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2) were synthesized according to the method of Zhou and O'Doherty (Zhou & O'Doherty, 2008). The structure of the compounds are shown in Figure 1A. For some analyzes we used digitoxin (3) as a standard. To evaluate the direct effect of such compounds on the Na+/K+-ATPase activity we directly incubated the compounds with purified Na+/K+-ATPase. The results showed that the compounds fully inhibited Na+/K+-ATPase activity with a low IC50 of 12 ± 1 nM (1) and 41 ± 3 nM (2) (Figure 1B).

Figure 1: Structure and direct effect of digitoxigenin derivatives on pig kidney Na+/K+-ATPase and PMCA erythrocytes activity.

Figure 1:

Open in a new tab

(A) Structure of digitoxigenin derivatives rhamno-digitoxigenin (1), amiceto-digitoxigenin (2) and digitoxin (3). (B) Increasing concentrations of (1) (2) and were incubated for 2 h with a purified Na+/K+-ATPase from pig kidney (only contains α1β1 protomer). Each point represents the mean ± SEM of three independent experiments performed in triplicate. (C) PMCA from human erythrocytes was incubated for 1 h with (2) at different concentrations and ATPase activity was determined. *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p <0.05) n=3.

Direct effect on Ca-ATPase activity

In order to assess whether 1 and 2 are specific modulators of Na+/K+-ATPase, we decided to first test the effect on PMCA activity. Thus, membrane preparations from erythrocytes were treated with these compounds at different concentrations, and PMCA activity was evaluated. Interestingly, 1 did not significantly alter the PMCA activity (Figure S1A), but the compound 2 was able to significantly inhibit the activity by around 25% at 100 nM (Figure 1C).

Cytotoxic effect in cervical carcinoma cells – HeLa

Based on the effect of the analogs on the activity of Na+/K+-ATPase and the ability of 2 to inhibit PMCA activity, we evaluated the effect of these compounds on the viability of HeLa cells by colorimetric MTT assay. The results showed IC50 = 2.340 ± 0.003 μM for digitoxin and that 1 and 2 are potent cytotoxic drugs on HeLa cells, with IC50 = 35.2 ± 1.6 nM and IC50 = 38.7 ± 1.3 nM, respectively (Figure 2A).

Figure 2: Effect of compounds on viability and Na+/K+-ATPase activity and expression in HeLa cells.

Figure 2:

Open in a new tab

(A) Cell viability (%) in HeLa cell line after treatment for 24 h with different concentrations of digitoxigenin-α-L-rhamno-pyranoside (1), digitoxigenin-α-L-amiceto-pyranoside (2), Digitoxin (3). *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p<0.05) n=3. Na+/K+-ATPase activity after cell membrane preparation of HeLa treated for 24 h with (B) digitoxigenin-α-L-rhamno-pyranoside (1) and (C) digitoxigenin-α-L-amiceto-pyranoside (2). (D) Digitoxin (3) (100 nM, 9 μM, 20 μM) n=3. (E) Na+/K+-ATPase α1 expression in HeLa cells treated for 24 h with 1 (65 nM and 100 nM) and 2 n=3. β-actin was used as a standard house-keeping protein. *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p<0.05).

We assess whether the cytotoxicity of these semi-synthetic compounds in HeLa cells could be attributed to the modulation of P-ATPases. For this, HeLa cells were treated for 24 h and membrane preparations were obtained to perform the measurement of Na+/K+-ATPase, PMCA and SERCA activity. Both 1 and 2 inhibited Na+/K+-ATPase activity in HeLa cells at nanomolar concentration (Figure 2B and C), consistent with direct inhibition in pig kidney Na+/K+-ATPase (Figure 1). Comparatively, digitoxin did not inhibit the activity of Na+/K+-ATPase at 100 nM as the synthesized compounds (Figure 3D). We also found that 1 and 2 decreased the expression of the α1 subunit of Na+/K+-ATPase (Figure 3E).

Figure 3: Effect of compounds on PMCA activity and expression in HeLa cells.

Figure 3:

Open in a new tab

PMCA activity of membrane preparation from HeLa cells treated for 24 h with (A) digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2) (B) digitoxin (3) *Significantly different from control (with calmodulin, CaM). Two-way ANOVA followed by Bonferroni post-hoc test (P <0.05) n = 3-4. PMCA4 expression levels. Treatment for 24 h with (C) (1) and (2) (D) (3). β-actin was used as a standard house-keeping protein. *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p<0.05) n=3-4.

Subsequently, we evaluated whether the drugs modulated PMCA and SERCA activity, since the proper functioning of these enzymes is essential for Ca2+ homeostasis and Ca2+ overload by their impairment may induce apoptosis (Di Leva, Domi, Fedrizzi, Lim, & Carafoli, 2008; Papp et al., 2012). Our results showed that 1, 2 and 3 decreased PMCA4 expression and inhibited total PMCA activity in intact HeLa cells, after 24 h of treatment (Figure 3). However, semi-synthetic compounds caused this inhibition at nanomolar concentrations, which was different than 3 that inhibited PMCA activity and expression at micromolar concentrations.

We also assessed whether both analogs directly affect the activity of SERCA. To this end, the SERCA membrane preparation was treated with the compounds at different concentrations. It was found that none of the analogs acted directly on the enzyme activity (Figure S1B and C). Interestingly, the treatment of HeLa cells with 1 at 100 nM decreased while 2 increased SERCA activity (Figure 4A and B, respectively). No effect was seen on nonspecific Mg2+-ATPase activity (Figure S1D).

Figure 4: Effect of compounds on SERCA activity in HeLa cells.

Figure 4:

Open in a new tab

Sarco(endo)plasmic reticulum Ca2+ATPase (SERCA) activity of membrane preparation from HeLa cells treated for 24 h with (A) digitoxigenin-α-L-rhamno-pyranoside (1) and (B) digitoxigenin-α-L-amiceto-pyranoside (2). *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p<0.05) n=5.

Human diploid lung fibroblasts (Wi-26-VA4)

Human diploid lung fibroblasts (Wi-26-VA4) were used to evaluate whether these compounds have cytotoxic selectivity for cancer cells. Both analogs had a higher cytotoxic effect in HeLa cells (Figure 2) than in non-tumor Wi-26-VA4 cells (IC50 ≥ 10 μM; Figure 5A). Consistent with the observed lower cytotoxicity in Wi-26-VA4 non-tumor cells, 1 and 2 had no significant modulatory effect on Na+/K+-ATPase and PMCA activity after 24 h (Figure 5B and C). This reinforces the involvement of these enzymes in the cytotoxic effect of both compounds as well as their selectivity for cancer cells.

Figure 5: Effect of compounds on cell viability, Na+/K+-ATPase and PMCA activity in Wi-26-VA4 cells.

Figure 5:

Open in a new tab

The cells were treated for 24 h with different concentrations of (A) digitoxigenin-α-L-rhamno-pyranoside (1) and digitoxigenin-α-L-amiceto-pyranoside (2). (B) Na+/K+-ATPase and (C) PMCA activity after membrane preparation of Wi-26-VA4 cells treated for 24 h (1) and (2). *Significantly different from control. One-way ANOVA followed by Dunnett post hoc test (p<0.05) n=3-5.

Discussion

The compounds 1 and 2 have shown high cytotoxic potency in non-small lung cancer cells - NCI-H460 (IC50 = 12-46 nM and IC50 =55 nM, respectively) (Elbaz, Stueckle, Wang, et al., 2012; Iyer et al., 2010; H. Y. L. Wang et al., 2011) which was significantly improved over that found for digitoxin (IC50 = 49-357 nM) (H.-Y. L. Wang et al., 2011). CGs are classical ligands of Na+/K+-ATPase, which is an important regulator of ionic homeostasis, cell growth and gene transcription in cancer cells.(Bagrov, Shapiro, & Fedorova, 2009). In this way, we have found that the compounds are potent inhibitors of Na, K-ATPase activity with IC50 in the nanomolar range. At concentrations >100 nM, CGs are classic Na,KATPase inhibitors, leading to their positive inotropic effect by increasing intracellular Ca2+ levels through indirect modulation of the Na+/Ca2+ exchanger (Dostanic, Schultz, Lorenz, & Lingrel, 2004). Here we demonstrate that these semi-synthetic compounds continue to show the inhibitory effect on Na,K-ATPase, but at very low concentrations.

Over 20 years ago, other P-ATPase, the colonic H+-K+ATPase was reported to be sensitive to ouabain (Cougnon et al., 1996). Recent work demonstrated a cardiotonic steroid, more specifically a bufadienolide extracted from Rhinella marina, directly inhibiting PMCA human erythrocytes and modulating the expression of this enzyme in human breast adenocarcinoma cell line, MDA-MB231(Garcia et al., 2019). Here we demonstrated that a cardiac glycoside, compound 2 also directly inhibited the activity of human PMCA (Figure 1C). Furthermore, it is possible that the higher lipophilicity of this compound, when compared to the more polar 1, it is essential for the interaction with PMCA.

PMCA is encoded by four genes (PMCA1–4) and their transcripts may be alternatively spliced and generate more variations of this enzyme. The isoforms PMCA1 and 4 are expressed ubiquitously, and are the only isoforms expressed in erythrocytes, but PMCA4 expression is more abundant than PMCA1 (Krebs, 2014; Stauffer, Guerini, & Carafoli, 1995). By the present experiments we cannot assure which PMCA isoform is inhibited by compound 2.

Our data showed that both analogs were more cytotoxic than digitoxin in HeLa cells. This effect is in agreement with the greater inhibition of Na,K-ATPase activity with treatment 1 and 2 than digitoxin. Comparatively, Hosseini et al. (2019) showed, using MTT in the same cells, that digitoxin had an IC50 = 7 μM (Hosseini, Taherkhani, & Ghorbani Nohooji, 2019). Yang et al. (2013) evidenced that the cytotoxic effect of digitoxin was only significant at 300 nM (Yang et al., 2013).

Furthermore, we demonstrate that the compounds inhibit both purified Na,K-ATPase (pig kidney membrane) and Na,K-ATPase activity of HeLa cells. This show that those compound can modulate the Na,K-ATPase activity directly and decreases its expression in the cell. Nanomolar concentrations of CGs can alter the expression levels of the Na+/K+-ATPase subunits and this regulation is cell type dependent (Tian et al., 2009). Comparative studies between the digitoxin and 1 (aka D6-MA) showed that 1 induces intrinsic pathway caspase 9-dependent apoptosis in non-small lung cancer cells (Iyer et al., 2010). D6-MA also caused cell cycle arrest in G2/M by sub-regulatory cyclin B1, cdc2, and survivin ChK 1/2, proteins that are essential for cell cycle progression.(Elbaz, Stueckle, Wang, et al., 2012).

The inhibition of Na+/K+-ATPase expression and activity, can lead to increased intracellular Ca2+ levels (Blaustein, 1993; Curry, Luk, Kenny, Roberts-Thomson, & Monteith, 2012). The compounds also decreased PMCA expression and activity more potently than digitoxin. These results indicate that by immediate and prolonged converging mechanisms, chronic increase in intracellular Ca2+ concentration may occur. In addition, the compounds modulated SERCA activity of HeLa cells. We hypothesize that for 1 the reduced SERCA activity would contribute to the increase of intracellular Ca2+ levels and for 2 the increment of SERCA activity would be a response to the increased intracellular Ca2+ levels additionally caused by direct inhibition of PMCA, besides the effect on Na+/K+-ATPase. These reinforce the fact that the two analogs have different mechanisms of action and might explain the higher cytotoxicity of 1.

It has been shown that PMCA inhibition in neurons leads to increased intracellular Ca2+ level, activation of caspase 3 and apoptosis (Kurnellas et al., 2005). The PMCA1 knockdown increased necrosis induced by the Ca2+ ionophore ionomycin and PMCA4 knockdown increased apoptosis mediated by ABT263 (Bcl-2 inhibitor) and by inhibition of NFκB nuclear translocation in MDA-MB231 breast cancer cells. In a previous study we showed a bufadienolide that modulated Na+/K+-ATPase and it also led to reduced PMCA4 expression and SERCA caspase 3 and 9 activation in MDA-MB231 cells, demonstrating a possible modulation of Calcium homeostasis in these cells (Garcia et al., 2019). This suggests that PMCA4 inhibitors could act as potential cancer cell sensitizers to apoptotic stimuli (Curry et al., 2012). In this way, the concentration range in which the compounds have cytotoxic effects corresponds to the impaired activity and expression of Na+/K+-ATPase and PMCA, and strongly supports the mechanistic involvement of these P-ATPases.

Na,K-ATPase has its expression altered in some types of cancers when compared to their respective normal tissues. Which could explain the difference in cytotoxicity of CGs in these cells (Mijatovic, Dufrasne, & Kiss, 2012). The compounds were less cytotoxic in non-tumor Wi-26-VA4 cells and did not modulate PMCA and Na,K-ATPase activity in these cells. Other studies have also shown this selectivity using different cardiotonic steroids (Kutluay, Makino, Inoue, & Saracoglu, 2019). However, bufadienolides have already been shown to be cytotoxic in Wi-26-VA4 cells with IC50 values in the micromolar range (Garcia et al., 2019). The lower cytotoxicity of digitoxin and 1 in non-tumorigenic immortalized bronchial and small airway epithelial cells compared to NCI-H460 has been demonstrated before (Elbaz, Stueckle, Wang, et al., 2012). Thus, our data reinforce this previous study and show that the compounds are also less cytotoxic in Wi-26-VA4. These findings reveal that the present cardiac glycosides are potential anticancer agents with high cytotoxicity and apparent selectivity for cancer cells, and further encourage efforts to fully elucidate their mechanism of action.

Conclusion

Our study shows that a semi-synthetic CGs, digitoxigenin-α-L-amycet-pyranoside (2), directly inhibits erythrocyte PMCA activity. These compounds also affected Na, K-ATPase, PMCA and SERCA activity in intact HeLa cells. In summary, we found that 1 and 2 at nanomolar concentrations inhibit the activity and expression of Na, K-ATPase, and PMCA 4 which could possibly contribute to ionic imbalance and increased intracellular Ca2+. Digitoxin (3) also modulated Na,K-ATPase activity and PMCA expression and activity, but this happened at micromolar concentrations and explains the greater cytotoxicity of 1 and 2. In addition, 1 and 2 also modulate SERCA activity of intact HeLa cells, which is likely related to intracellular Ca2+ levels rather than direct modulation of this enzyme. Both semisynthetic compounds also showed a selective cytotoxic effect on HeLa cells, but not on non-tumor Wi-26-VA4 cells, strongly suggesting that this was due to the lack of Na,K -ATPase and PMCA modulation in Wi-26 Cells- A4. In addition to the known mechanism of action of these CGs in Na,K-ATPase inhibition, we found a new mechanism of action through PMCA modulation, which expands the mechanistic understanding of these CGs analogues for further development as anticancer agents.

Supplementary Material

supinfo

Acknowledgment

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, Brazil) – Finance Code 001, FAPEMIG (Fundação de Amparo a Pesquisa do Estado de Minas Gerais) APQ-00290-16, APQ-00855-19, PPM-00307-18 and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) 401914/2016-0, 305173/2018-9. The National Science Foundation (CHE-1565788), and the National Institutes of Health (AI146485, AI144196 and AI142040) are also acknowledged for their support of the synthesis of analogues 1 and 2.

References

  1. Azalim P, do Monte FM, Rendeiro MM, Liu X, O'Doherty GA, Fontes CF, … Noël F (2020). Conformational states of the pig kidney Na+/K+-ATPase differently affect bufadienolides and cardenolides: A directed structure-activity and structure-kinetics study. Biochemical Pharmacology, 171, 113679. doi: 10.1016/j.bcp.2019.113679 [DOI] [PubMed] [Google Scholar]
  2. Bagrov AY, Shapiro JI, & Fedorova OV (2009). Endogenous cardiotonic steroids: physiology, pharmacology, and novel therapeutic targets. Pharmacological reviews, 61(1), 9–38. doi: 10.1124/pr.108.000711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Blaustein MP (1993). Physiological effects of endogenous ouabain: control of intracellular Ca2+ stores and cell responsiveness. The American journal of physiology, 264(6 Pt 1), C1367–1387. [DOI] [PubMed] [Google Scholar]
  4. Cornelius F, Kanai R, & Toyoshima C (2013). A structural view on the functional importance of the sugar moiety and steroid hydroxyls of cardiotonic steroids in binding to Na,K-ATPase. Journal of Biological Chemistry, 288(9), 6602–6616. doi: 10.1074/jbc.M112.442137 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cortes VF, Ribeiro IM, Barrabin H, Alves-Ferreira M, & Fontes CFL (2011). Regulatory phosphorylation of FXYD2 by PKC and cross interactions between FXYD2, plasmalemmal Ca-ATPase and Na,K-ATPase. ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS, 505(1), 75–82. doi: 10.1016/j.abb.2010.09.017 [DOI] [PubMed] [Google Scholar]
  6. Cougnon M, Planelles G, Crowson MS, Shull GE, Rossier BC, & Jaisser F (1996). The rat distal colon P-ATPase alpha subunit encodes a ouabain-sensitive H+, K+-ATPase. J Biol Chem, 271(13), 7277–7280. doi: 10.1074/jbc.271.13.7277 [DOI] [PubMed] [Google Scholar]
  7. Curry MC, Luk NA, Kenny PA, Roberts-Thomson SJ, & Monteith GR (2012). Distinct regulation of cytoplasmic calcium signals and cell death pathways by different plasma membrane calcium ATPase isoforms in MDA-MB-231 breast cancer cells. Journal of Biological Chemistry, 287(34), 28598–28608. doi: 10.1074/jbc.M112.364737 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Di Leva F, Domi T, Fedrizzi L, Lim D, & Carafoli E (2008). The plasma membrane Ca2+ ATPase of animal cells: Structure, function and regulation (Vol. 476, pp. 65–74). [DOI] [PubMed] [Google Scholar]
  9. Dostanic I, Schultz JEJ, Lorenz JN, & Lingrel JB (2004). The α1 Isoform of Na,K-ATPase Regulates Cardiac Contractility and Functionally Interacts and Co-localizes with the Na/Ca Exchanger in Heart*. Journal of Biological Chemistry, 279(52), 54053–54061. doi: 10.1074/jbc.M410737200 [DOI] [PubMed] [Google Scholar]
  10. Elbaz HA, Stueckle TA, Tse W, Rojanasakul Y, & Dinu CZ (2012). Digitoxin and its analogs as novel cancer therapeutics. Experimental hematology & oncology, 1(1), 1–4. doi: 10.1186/2162-3619-1-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Elbaz HA, Stueckle TA, Wang HYL, O'Doherty GA, Lowry DT, Sargent LM, … Rojanasakul Y (2012). Digitoxin and a synthetic monosaccharide analog inhibit cell viability in lung cancer cells. Toxicology and Applied Pharmacology, 258(1), 51–60. doi: 10.1016/j.taap.2011.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fiske CH, & Subbarow Y (1925). The colorimetric determination of phosphorus. Journal of Biological Chemistry, LXVI(2), 375–400. [Google Scholar]
  13. Garcia IJP, de Oliveira GC, de Moura Valadares JM, Banfi FF, Andrade SN, Freitas TR, … Barbosa LA (2019). New bufadienolides extracted from Rhinella marina inhibit Na,K-ATPase and induce apoptosis by activating caspases 3 and 9 in human breast and ovarian cancer cells. Steroids, 152, 108490. doi: 10.1016/j.steroids.2019.108490 [DOI] [PubMed] [Google Scholar]
  14. Haux J (1999). Digitoxin is a potential anticancer agent for several types of cancer. Medical hypotheses, 53(6), 543–548. doi: 10.1054/mehy.1999.0985 [DOI] [PubMed] [Google Scholar]
  15. Hilário FF, de Paula RC, Silveira MLT, Viana GHR, Alves RB, Pereira JRCS, … de Pilla Varotti F (2011). Synthesis and Evaluation of Antimalarial Activity of Oxygenated 3-alkylpyridine Marine Alkaloid Analogues (Vol. 78, pp. 477–482). [DOI] [PubMed] [Google Scholar]
  16. Hosseini M, Taherkhani M, & Ghorbani Nohooji M (2019). Introduction of Adonis aestivalis as a new source of effective cytotoxic cardiac glycoside. Nat Prod Res, 33(6), 915–920. doi: 10.1080/14786419.2017.1413573 [DOI] [PubMed] [Google Scholar]
  17. Iyer AKV, Zhou M, Azad N, Elbaz H, Wang L, Rogalsky DK, … Langenhan JM (2010). A Direct Comparison of the Anticancer Activities of Digitoxin MeON-Neoglycosides and O-Glycosides: Oligosaccharide Chain Length-Dependent Induction of Caspase-9-Mediated Apoptosis. ACS Medicinal Chemistry Letters, 1(7), 326–330. doi: 10.1021/ml1000933 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kanai R, Cornelius F, Ogawa H, Motoyama K, Vilsen B, & Toyoshima C (2021). Binding of cardiotonic steroids to Na(+),K(+)-ATPase in the E2P state. Proceedings of the National Academy of Sciences of the United States of America, 118(1), e2020438118. doi: 10.1073/pnas.2020438118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kosk-Kosicka D (1999). Measurement of Ca2+-ATPase Activity (in PMCA and SERCA1) (pp. 114–114). [Google Scholar]
  20. Krebs J (2014). The plethora of PMCA isoforms: Alternative splicing and differential expression. Biochimica et Biophysica Acta - Molecular Cell Research, 1853(9), 2018–2024. doi: 10.1016/j.bbamcr.2014.12.020 [DOI] [PubMed] [Google Scholar]
  21. Kulkarni YM, Kaushik V, Azad N, Wright C, Rojanasakul Y, O'Doherty G, & Iyer AKV (2016). Autophagy-Induced Apoptosis in Lung Cancer Cells by a Novel Digitoxin Analog. Journal of Cellular Physiology, 231(4), 817–828. doi: 10.1002/jcp.25129 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kurnellas MP, Nicot A, Shull GE, & Elkabes S (2005). Plasma membrane calcium ATPase deficiency causes neuronal pathology in the spinal cord: a potential mechanism for neurodegeneration in multiple sclerosis and spinal cord injury. The FASEB Journal, 19, 298–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kutluay VM, Makino T, Inoue M, & Saracoglu I (2019). New knowledge about old drugs; a cardenolide type glycoside with cytotoxic effect and unusual secondary metabolites from Digitalis grandiflora Miller. Fitoterapia, 134, 73–80. doi: 10.1016/j.fitote.2019.02.001 [DOI] [PubMed] [Google Scholar]
  24. López-Lázaro M (2007). Digitoxin as an anticancer agent with selectivity for cancer cells: possible mechanisms involved. Expert opinion on therapeutic targets, 11, 1043–1053. doi: 10.1517/14728222.11.8.1043 [DOI] [PubMed] [Google Scholar]
  25. Maia GAS, Cortes VF, Ribeiro RIMDA, Mignaco JA, de Lima Santos H, Fontes CFL, & Barbosa LA (2014). Could Na,K-ATPase play a role in potassium leakage from irradiated erythrocytes? Clinica Chimica Acta, 433, 58–61. doi: 10.1016/j.cca.2014.02.025 [DOI] [PubMed] [Google Scholar]
  26. Mijatovic T, Dufrasne F, & Kiss R (2012). Na+/K+-ATPase and cancer. Pharm Pat Anal, 1(1), 91–106. doi: 10.4155/ppa.12.3 [DOI] [PubMed] [Google Scholar]
  27. Newman R. a., Yang P, Pawlus AD, & Block KI (2008). Cardiac glycosides as novel cancer therapeutic agents. Mol.Interv, 8, 36–49. doi: 10.1124/mi.8.1.8 [DOI] [PubMed] [Google Scholar]
  28. Noël F, Azalim P, do Monte FM, Quintas LEM, Katz A, & Karlish SJD (2018). Revisiting the binding kinetics and inhibitory potency of cardiac glycosides on Na + ,K + -ATPase (α1β1): Methodological considerations. Journal of Pharmacological and Toxicological Methods, 94, 64–72. doi: 10.1016/j.vascn.2018.09.001 [DOI] [PubMed] [Google Scholar]
  29. Papp B, Brouland J-P, Arbabian A, Gélébart P, Kovács T, Bobe R, … Apáti A (2012). Endoplasmic reticulum calcium pumps and cancer cell differentiation. Biomolecules, 2(1), 165–186. doi: 10.3390/biom2010165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pessôa MTC, Alves SLG, Taranto AG, Villar J, Blanco G, & Barbosa LA (2018). Selectivity analyses of γ-benzylidene digoxin derivatives to different Na,K-ATPase α isoforms: a molecular docking approach. 33(1), 85–97. doi: 10.1080/14756366.2017.1380637 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rega AF, Garrahan PJ, Barrabin H, Horenstein A, & Rossi JP (1979). Cation flux across biomembranes. Academic, New York, Academic, New York-Academic, New York. [Google Scholar]
  32. Rocha SC, Pessoa MT, Neves LD, Alves SL, Silva LM, Santos HL, … Barbosa LA (2014). 21-Benzylidene digoxin: a proapoptotic cardenolide of cancer cells that up-regulates Na,K-ATPase and epithelial tight junctions. PLoS ONE, 9(10), e108776. doi: 10.1371/journal.pone.0108776 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Schoenmakers TJM, Visser GJ, Flik G, & Theuvenet APR (1992). Chelator - An Improved Method for Computing Metal-Ion Concentrations in Physiological Solutions (Vol. 12, pp. 870–874). [PubMed] [Google Scholar]
  34. Shoshani L (2004). The Polarized Expression of Na+,K+-ATPase in Epithelia Depends on the Association between -Subunits Located in Neighboring Cells. Molecular Biology of the Cell, 16, 1071–1081. doi: 10.1091/mbc.e04-03-0267 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sousa L, Garcia IJP, Costa TGF, Silva LND, Renó CO, Oliveira ES, … Barbosa LA (2015). Effects of Iron Overload on the Activity of Na,K-ATPase and Lipid Profile of the Human Erythrocyte Membrane. PLoS ONE, 10(7), e0132852. doi: 10.1371/journal.pone.0132852 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Souza Gonçalves B, Moura Valadares JM, Alves SLG, Silva SC, Rangel LP, Cortes VF, … Lima Santos H (2019). Evaluation of neuroprotective activity of digoxin and semisynthetic derivatives against partial chemical ischemia. Journal of Cellular Biochemistry, 20(June 2018), 17108–17122. doi: 10.1002/jcb.28971 [DOI] [PubMed] [Google Scholar]
  37. Stauffer TP, Guerini D, & Carafoli E (1995). Tissue distribution of the four gene products of the plasma membrane Ca2+ pump. A study using specific antibodies. Journal of Biological Chemistry, 270, 12184–12190. [DOI] [PubMed] [Google Scholar]
  38. Tian J, Li X, Liang M, Liu L, Xie JX, Ye Q, … Xie Z (2009). Changes in sodium pump expression dictate the effects of ouabain on cell growth. Journal of Biological Chemistry, 284(22), 14921–14929. doi: 10.1074/jbc.M808355200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ulianich L, Elia MG, Treglia AS, Muscella A, Di Jeso B, Storelli C, & Marsigliante S (2006). The sarcoplasmic-endoplasmic reticulum Ca2+ ATPase 2b regulates the Ca2+ transients elicited by P2Y2 activation in PC Cl3 thyroid cells. Journal of Endocrinology, 190(3), 641–649. doi: 10.1677/joe.1.06455 [DOI] [PubMed] [Google Scholar]
  40. Wang H-YL, Rojanasakul Y, & O’Doherty GA (2011). Synthesis and Evaluation of the α-d-/α-l-Rhamnosyl and Amicetosyl Digitoxigenin Oligomers as Antitumor Agents. ACS Med Chem, 2(4), 264–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wang HYL, Wu B, Zhang Q, Kang SW, Rojanasakul Y, apos, & Doherty GA (2011). C5'-alkyl substitution effects on digitoxigenin a-l-glycoside cancer cytotoxicity. ACS Medicinal Chemistry Letters, 2(4), 259–263. doi: 10.1021/ml100291n [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wang HYL, Xin W, Zhou M, Stueckle TA, Rojanasakul Y, & O'Doherty GA (2011). Stereochemical survey of digitoxin monosaccharides. ACS Medicinal Chemistry Letters, 2(1), 73–78. doi: 10.1021/ml100219d [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wu J, Akkuratov EE, Bai Y, Gaskill CM, Askari A, & Liu L (2013). Cell signaling associated with Na+/K+-ATPase: Activation of phosphatidylinositide 3-kinase IA/Akt by ouabain is independent of Src. Biochemistry, 52(50), 9059–9067. doi: 10.1021/bi4011804 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yang QF, Dalgard CL, Eidelman O, Jozwik C, Pollard BS, Srivastava M, & Pollard HB (2013). Digitoxin induces apoptosis in cancer cells by inhibiting nuclear factor of activated T-cells-driven c-MYC expression. Journal of carcinogenesis, 12(8). doi: 10.4103/1477-3163.112268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Zhou M, & O'Doherty G (2008). The De Novo synthesis of oligosaccharides: application to the medicinal chemistry SAR-study of digitoxin. Curr. Top. Med. Chem. (Sharjah, United Arab Emirates), 8(2), 114–125. doi: 10.2174/156802608783378828 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supinfo
NIHMS1740916-supplement-supinfo.docx (421KB, docx)

RESOURCES