ORIGINAL
The antiproliferative effects of ouabain and everolimus on adrenocortical tumor cells
Raffaele Pezzani1), Beatrice Rubin1), Marco Redaelli2), Claudia Radu3), Susi Barollo1), Maria Verena Cicala1), Monica Salvà1), Caterina Mian1), Carla Mucignat-Caretta2), Paolo Simioni3), Maurizio Iacobone4) and Franco Mantero1)
1) Endocrinology Unit, Department of Medicine, University of Padova, Padova 35128, Italy
2) Department of Molecular Medicine, University of Padova, Padova 35131, Italy
3) Department of Cardiologic, Thoracic, and Vascular Sciences, 2nd Chair of Internal Medicine, University of Padova Medical School, Padova 35128, Italy
4) Endocrine Surgery Unit, Department of Surgical and Gastroenterological Sciences, University of Padova, Padova 35128, Italy
Abstract. Ouabain is a cardiotonic steroid obtained from Strophanthus. Recently its role as antiproliferative agent has been investigated in tumor cells. Everolimus is a derivative of rapamycin and acts as a signal transduction inhibitor. Adrenocortical carcinoma is a rare cancer, with poor prognosis. This research focuses on antineoplastic properties of ouabain and its association with everolimus. We analyzed the effects of drugs on cells by MTT assay, by [3H] thymidine assay, by Wright’s staining, by homogeneous caspases assay, by flow cytometry analysis and by Western blot analysis on H295R and SW13 cells and on primary adrenocortical tumor cells. Ouabain induced cell viability reduction in SW13, H295R and 5 primary adrenocortical tumor cells. Combination of ouabain with everolimus produced a stronger cytotoxic effect on cell proliferation and viability. Marked morphological changes were observed in both SW13 and H295R cell lines after ouabain treatment, with an increase in necrosis. Cell cycle distribution was altered by ouabain in SW13. Analysis of apoptosis demonstrated an increase in caspase activity, clearly evident for SW13 at 72h. FACS analysis by Annexin V-FITC kit and propidium iodide confirmed an increased level of necrosis at higher concentrations. Western blot analysis showed that PI3k/Akt signaling pathway was modified after ouabain treatments in SW13. Ouabain exerts antiproliferative effects on SW13 and H295R cell lines and on primary adrenocortical tumor cells. These data suggest that ouabain or ouabain derivatives may be potential anticancer agents.
Key words: Ouabain, Antiproliferative, Adrenocortical tumor, Primary cell cultures, Everolimus
OUABAIN is a cardiotonic steroid (CS) obtained from Strophanthus. Its pharmacological action reflects its Na+/K+-ATPase binding activity, producing an increase in intracellular [Na+] at clinical doses. As a conse- quence, the influx of extracellular Ca2+ creates a sig- nificant increase in cardiac contraction. This positive inotropic action is considered the rationale for using ouabain (and other CS) in congestive heart failure [1]. In addition to its cardiotonic function, CS have been shown to influence many processes, such as cell prolif- eration, apoptosis, survival, etc. in different cell lines,
not only inhibiting these processes [2], but sometimes activating them [3].
Formerly the clinical observations of Stenkvist and collaborators have shown the beneficial effects of digi- talis treatments in women with breast cancer: very few patients treated with CS for chronic heart problems died of cancer [4, 5]. For this reason various laborato- ries have explored the role of ouabain and similar com- pounds as antiproliferative agents. For example CS induced apoptosis in human prostate cancer cell lines [6]. Other investigators found that ouabain and digoxin selectively induced apoptosis in a human acute T-cell lymphoblastic leukemia cell line [2]. In the estrogen dependent/independent breast cancer cell lines, CS similarly exerted antiproliferative effects, suggesting their potential use as anticancer agents [7].
Everolimus (RAD001) is a derivative of rapamycin acting on the mammalian target of rapamycin (mTOR), a key component of PI3K/Akt pathway, frequently acti- vated in human cancers. It is a well-known compound currently used for the treatment of renal cell cancer and other tumors [8]. Recently we described its antiprolif- erative effect in adrenocortical tumor models in asso- ciation with sorafenib [9].
Adrenocortical tumors can be benign or malig- nant. Benign tumors are frequent in the general popu- lation, being commonly discovered incidentally (inci- dentaloma) and may be functional or nonfunctional. Malignant adrenocortical carcinoma (ACC) has a poor prognosis with an estimated survival rate of 35% at 5 years, and to date pharmacologic therapy is principally based on the adrenotoxic drug, mitotane (o,p’-DDD), despite its low response rate and significant toxicity [10]. As this regimen is insufficient, it is necessary to develop new treatment modalities.
To determine whether ouabain alone, or in combina- tion with everolimus, exerts antineoplastic activity, we explored its effects on cell viability, cell proliferation, cell morphology, cell cycle and apoptosis in SW13 and H295R cells (the 2 most frequently used adrenocortical cell models), and in primary adrenocortical tumor cells.
Materials and Methods
Materials
Ouabain, everolimus, fetal bovine serum (FBS), 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT), trichloracetic acid (TCA) and pro- pidium iodide were purchased from Sigma Aldrich (Sigma Aldrich Corporation, St. Louis, MO, USA). DMEM-F12, 0.05% trypsin-EDTA, insulin, transfer- rin, selenium and antibiotics were from Invitrogen (Invitrogen S.R.L. Milano, Italy). Primary antibodies were: anti-Akt (cod. 9272), anti-phospho-Akt (Ser473)
(cod. 9271), anti-P70S6K (cod. 2708), anti-phospho- P70S6K (Thr389) (cod. 9234) all from Cell Signaling Technology and mouse monoclonal anti-ß-Actin anti- body (cod. A5441) from Sigma-Aldrich. Secondary antibodies were: horseradish peroxidase-labeled goat anti-mouse or anti-rabbit (Jackson ImmunoResearch laboratories Inc., West Grove, PA, USA).
Patients
Adrenocortical tumor tissues were obtained from 1 patient with ACC and 4 patients with aldosterone-pro- ducing adenomas (APA). The etiology of each adre- nocortical mass, clinical diagnose and staging were established as described [9]. All clinical data are sum- marized in Table 1. All patients gave written informed consent to the collection and use of adrenal tissue for research purposes, and the study was approved by the local ethics committee (conforming to the Declaration of Helsinki, revised in Tokyo, 2004).
Cell cultures
H295R and SW13 adrenocortical tumor cell lines were obtained from the American Type Culture Collection (ATCC, Rockwille, MD). H295R cells were established from a female patient diagnosed with an adrenocortical carcinoma. This strain secretes min- eralcorticoids, glucocorticoids and adrenal androgens. SW13 cells were derived from a small cell carcinoma in the adrenal cortex of a 55-year-old female, probably a metastasis to the adrenal cortex. This strain produces no steroid [11]. Five primary cell cultures were also studied, 1 ACC and 4 APA. All experiment involving cell manipulation were conducted incubating cell lines with 0.1% FBS as previously described [12].
MTT assay
Cells were plated in 96-well plates at a density of 5x103 cells/well in supplemented medium with or with-
| Patient | Sex | Age at diagnosis (years) | Secretion | Weiss score | Tumor size | Tumor location | Performed experiments | Metastasis at diagnosis |
|---|---|---|---|---|---|---|---|---|
| ACC | F | 63 | C | 6 | 8×9×4cm | Left | Exp1, Exp3 | No |
| APA-1 | F | 42 | A | NA | 2×6×1cm | Left | Exp1, Exp2, Exp3 | No |
| APA-2 | F | 40 | A | NA | 1×2×1cm | Left | Exp1 | No |
| APA-3 | M | 57 | A | NA | 1×4×1cm | Right | Exp1 | No |
| APA-4 | F | 59 | A | NA | 1×3×1cm | Right | Exp1 | No |
M = Male; F = Female; A = aldosterone; C = cortisol; Exp1=MTT test, Exp2=Wright’s staining, Exp3=flow cytometry analysis, NA=Not applicable.
out ouabain. Drug incubation was carried out at differ- ent concentrations 1, 10, 50, 100, 200, 500, 1000nM (and 5000, 10000nM, data not shown). Everolimus was used alone or in combination with ouabain at 5 uM. Methods and percentage of cell survival were as previ- ously described [12]. Each analysis was performed on 4 replicates and repeated in 2 independent experiments.
MTT assay was also performed on 5 primary adre- nocortical tumor cells, 1 ACC and 4 APA. The con- centrations of ouabain used were 1, 10, 100, 1000 nM for 24h or 72h (and 120h for APA-1 primary tumor cell culture).
[3H] thymidine assay
H295R and SW13 cells were seeded in 24-well plates at a density of 2x105 cells/well and incubated with varying concentrations of ouabain (1, 10, 100, 1000nM) for 24h and 72h. Six hours before ending the experiment, 1 µCi/well of [3H] thymidine (PerkinElmer, Monza, Italy) was added to the medium. Then cells were washed once with 10% TCA and twice with 5% TCA. Subsequently cells were lysed in 1 mL of 0.5 M NaOH. Scintillation liquid (Ultima Gold, PerkinElmer, Monza, Italy) was added and radioactivity incorporated into DNA measured in a scintillation counter (Wallac, PerkinElmer, Monza, Italy). The results are expressed as a percentage of control (100%). Each experiment was performed in quadruplicate and repeated 3 times.
Assessment of apoptosis and necrosis by Wright’s staining
H295R and SW13 cells were cultured on coverslips for 48h, incubated overnight in 0.1% FBS, and then incubated with drug for 24h at different concentrations (1, 10, 100, 1000nM) for both cell lines. The cells were then washed in PBS, fixed in methanol for 5 min and Wright’s stained for 5 min. Cell morphology was eval- uated by light microscopy at x400 magnification. At least 600 cells were counted for every experiment in 4 different fields and each experiment was repeated twice. One primary tumor cell culture (APA-1) was evaluated with ouabain (10, 100, 1000nM) for 120h.
Cell cycle distribution analysis
H295R and SW13 cells were plated into 25 cm2 flasks at a density of 1x10° cells/well and were treated with ouabain alone (1, 10, 100, 1000nM) for 24 and 72h or in combination with everolimus (5uM), trypsinized and harvested by centrifugation. Cells were resuspended
in ice-cold PBS and fixed in 70% ice-cold ethanol, fol- lowed by an overnight incubation at -20℃. After wash- ing, cells were stained with PI solution (50µg/mL PI, 10µg/mL RNaseA) and incubated for 1h at 37℃ in the dark. Cell cycle analysis was performed in triplicates and data were assessed by CXP analysis software.
Homogeneous caspases assay
H295R and SW13 cells were plated into 96-well plates at a density of 2x104 cells/well and were treated with ouabain (1, 10, 100, 1000nM) for 24h and 72h. Total caspase activity was evaluated with a Homogeneous Caspases Assay Kit (Roche Diagnostics GmbH, Mannheim, Germany) following the manufac- turer’s instructions. In brief, substrate solution was added to cell lysate for 2h at 37℃ and relative fluo- rescence was measured with excitation and emission filters (2ex=499nm and em=520nm). Data were cor- rected by subtracting the blank absorbance and concen- trations estimated against a calibration curve obtained with R110 standard solutions. Each experiment was performed in triplicate and repeated twice.
Apoptosis flow cytometry analysis
Apoptosis was determined by Annexin V-FITC kit (BD Biosciences, San Diego, CA, USA). In brief, H295R and SW13 cells were plated into 25 cm2 flasks at a density of 1x10° cells/well for two days, then main- tained overnight at 0.1% FBS. The day after the cells were treated with ouabain alone (1, 10, 100, 1000nM) for 24 and 72 h or in combination with everolimus 5uM, trypsinized and harvested by centrifugation. Cells were then stained with Annexin V-FITC and propidium iodide according to the manufacturer’s instruction.
Flow cytometry analysis was also performed on 2 primary adrenocortical tumor cells: APA-1 and ACC. The concentrations of ouabain used were 1, 10, 100, 1000 nM and the time of treatment was 24h. Analysis was performed in duplicate.
Western blot analysis
Cells were treated with different concentrations of ouabain 1, 10, 100 nM at 24h for SW13 cells and 1, 10, 100, 1000nM at 24h for H295R cells as previ- ously described [9]. In addition everolimus (5 uM) was used in combination with ouabain. Briefly proteins were extracted with lysis buffer containing a protease inhibi- tor cocktail (Sigma-Aldrich) for 1h at 4℃, loaded onto SDS/PAGE, electroblotted onto nitrocellulose mem-
branes, blocked for 1h with 5% nonfat dry milk and incu- bated overnight with different primary antibodies. After secondary antibody, immunoreactivity was detected with LiteAblot Extend Long Lasting Chemiluminescent Substrate (EuroClone). Films were scanned and band intensity quantified with ImageJ software 1.44p. Each experiment was performed in triplicate.
Statistical analysis
Statistical analysis were performed using GraphPad Prism 4 software (GraphPad Software, Inc., San Diego, CA) and Microsoft Excel software. Data comparisons were performed using the two-tailed Student’s t test and Kruskal-Wallis analysis followed by Dunn’s post- hoc test. P<0.05 was considered statistically signifi- cant. Data are presented as mean ± standard error of the mean (SEM).
Results
Ouabain and everolimus affect cell viability in SW13, H295R and in primary adrenocortical tumor cells
To examine cell viability in SW13, H295R and pri- mary adrenocortical tumor cells, MTT assay was per- formed. As shown in Fig. 1A-B, ouabain induced a concentration and time-dependent decrease in cell via- bility in SW13 cells with an IC50 of 48±1nM at 24h, and of 6.6±1nM at 72h; at 72h an extrapolated IC50 of 120±17nM was obtained for H295R cells. Combination strategy led to similar results to those with ouabain alone, but 72h cell viability was considerably more reduced at lower ouabain doses in H295R cells (oua- bain 10/100nM + everolimus 5uM) (Fig. 1C-D).
The data in Fig. 1E-F point out that in primary adre- nocortical tumor cells (ACC and APA-1) the effect of ouabain was more evident when cells were treated at 1000 nM at 72h. In the 3 other APA samples no similar result was found (data not shown).
Ouabain affects cell proliferation rate (DNA synthe- sis) of SW13 and H295R cells
To assess the proliferation rate of SW13 and H295R cells, [3H] thymidine incorporation assays were per- formed. At higher doses (100 and 1000nM), ouabain maximally inhibited [3H] thymidine uptake in both SW13 and H295R cells at 24h and 72h (p<0.001). At 10nM, ouabain produced a slight decrease in [3H] thy- midine uptake, and at the lower dose (1nM) cells are indistinguishable from control. No substantial differ-
ences were seen between 24h and 72h treatments in both cell lines (Fig. 2).
Ouabain alters cell morphology in SW13, H295R and primary adrenocortical tumor cells
To evaluate cell morphology, Wright’s staining method was used. As shown in Fig. 3, apoptosis, deter- mined by blebbing, cell shrinkage, nuclear fragmen- tation, condensation of chromatin and necrosis, was determined by cell disruption and membrane break- down, both of which were clearly detected. Marked morphological changes were observed both for SW13 and H295R cells (Fig. 3A-D). In H295R, SW13 and in primary adrenocortical tumor cells (APA-1), necrotic processes were similarly predominant over apoptotic processes (data not shown).
Ouabain induces cell cycle deregulation in SW13 cells
To study the cell cycle distribution, PI assays were performed. In H295R cells, ouabain treatment alone or in combination produced no effect at 24h, but at 72h a slight G0/G1 phase increase after ouabain treat- ment alone (1000nM) or combination regimen (oua- bain 1000nM + everolimus 5uM) was detected (data not shown). In SW13 cells, ouabain treatment pro- moted a G0/G1 phase increase both at 24h and 72h, with a concomitant S and G2/M phases reduction, par- ticularly evident at 100 and 1000nM at 72h (Fig. 4A). Combination treatment produced no change in G0/ G1 phase, a slight increment in S phase and a slight decrease in G2/M phase, especially evident when oua- bain 1000nM + everolimus 5uM was used (Fig. 4B).
Ouabain affects caspase activity in SW13 and H295R cells
To assess the first steps of apoptotic process, the acti- vation of a particular group of proteases, the caspases, was analyzed by a specific assay in both cell lines. The activity of caspases evaluated at 24h are not dissimi- lar to basal for both cell lines, despite a statistical dif- ference (Fig. 5A-B). The effect of ouabain (10, 100 and 1000nM) on caspase activity proved to be more marked at 72h for both cell lines; in SW13 cells, cas- pase activity clearly increased at 100 and 1000nM oua- bain at 72h (Fig. 5B).
Ouabain does not induce apoptosis in SW13, H295R and in primary adrenocortical tumor cells
To discriminate between apoptosis and necro-
*
1000
A
H295R
B
SW13
120
120
*
100
*
*
100
2
*
*
24h
24h
% cell viability
80
*
72h
*
*
% cell viability
80
72h
*
*
*
*
Z
1
60
I
60
*
*
40
X
*
*
40
*
*
20
*
*HH
20
*
*
*
*
0
0
T
T
T
0
1
10
50
100
200
500
1000
0
1
10
50
100
200
500
Ouabain (nM)
Ouabain (nM)
C
120
H295R
*
24h
D
SW13
100
24h
T
72h
*
*
72h
100
*
T
*
*
*
T
T
80
T
T
T
T
% cell viability
80
% cell viability
*
T
*
*
60
60
T
*
I
*
40
40
T
T
20
*
20
T
*
0
T
0
E: 5uM
E: 5uM+
E: 5uM+
E: 5uM+
E: 5uM
E: 5uM+
E: 5uM+
E: 5uM+
0: 1nM
O: 10nM
O: 100nM
0: 1nM
O: 10nM
O: 100nM
E
ACC
F
APA-1
120
*
120
*
*
100
2
*
100
*
*
*
% cell viability
80
*
*
% cell viability
80
*
*
*
*
60
60
*
*
40
24h
40
24h
72h
72h
20
20
120h
Y
*
0
0
0
1
10
100
1000
0
1
10
100
1000
Ouabain (nM)
Ouabain (nM)
[3H] Thymidine incorporation (% control)
SW13 24h
[3H] Thymidine incorporation (% control)
SW13 72h
100
*
100
*
50
50
0
0
control
1
10
100
1000
control
1
10
100
1000
Ouabain (nM)
Ouabain (nM)
[3H] Thymidine incorporation (% control)
H295R 24h
[3H] Thymidine incorporation (% control)
H295R 72h
100
100
*
50
50
0
control
1
10
100
1000
0
control
1
10
100
1000
Ouabain (nM)
Ouabain (nM)
A
B
C
D
A
Control 72h
1nM
10nM
100nM
1000nM
16±3%
15±3%
9±4%
3±1%
3±3%
7±2%
5±4%
4±1%
8±2%
10±3%
76±4%
75±4%
84±2%
94±3%
91±4%
GO/G1
S
☐ G2/M
B
Control 72h
E: 5uM
E: 5uM+O: 10nM
E: 5uM+O: 1000nM
21±4%
20±4%
18±2%
13±4%
11±1%
15±2%
8±2%
10±1%
71±4%
70±2%
71±2%
72±4%
GO/G1
S
☐ G2/M
A
24h
H295R
72h
B
24h
SW13
500
1
4
1
I
72h
1
2000
400
1500
Free R110 (% to control)
(caspase activity)
Free R110 (% to control)
300
(caspase activity)
1000
300
200
*
200
*
100
100
0
0
control
1
10 100 1000
control
1
10 100 1000
control
1
10 100 1000
control
1
10 100 1000
Ouabain (nM)
Ouabain (nM)
A
[Ungated] FL1 / FL3
B
[Ungated] FL1 / FL3
103
C1
C2
103
C1
C2
102
102
FL3
101
FL3
101
C3
C4
C3
C4
100
100
10º
101
102
103
10º
101
102
103
FL1
FL1
C
[Ungated] FL1 / FL3
D
[Ungated] FL1 / FL3
103
C1
C2
103
C1
C2
102
102
FL3
101
FL3
101
100
C3
C4
100
C3
C4
10º
101
102
103
10º
101
102
103
FL1
FL1
sis, Annexin V-FITC/propidium iodide staining and flow cytometry were performed in both cell models and 2 primary adrenocortical tumor cells (1 ACC and APA-1). In SW13 cells, we observed a clear increase in necrotic processes at 1000nM after 24h (data not shown). At 72h we detected a very toxic effect of oua- bain especially at 100 and 1000nM with high levels of necrosis (Fig. 6A-B), very similar to combination treat- ment (Fig. 6C-D). No appreciable effect was evident for apoptosis or necrosis in H295R cells at 24h and 72h (both with ouabain alone or combination treatment), and in ACC primary tumor cells, ouabain exerted a slight necrotic effect at 1000nM, with no effect evident in APA primary cells (data not shown).
Ouabain and everolimus suppress PI3k/Akt signaling pathways in SW13 cells
To evaluate the PI3k/Akt signaling pathway involved in apoptosis, survival and proliferation, Western blot analysis was done on both cell lines. No substantial change was observed for H295R cells, except for an absence of reactivity for phosphorylated p70S6k, both in ouabain (Fig. 7A) and combination treatments (Fig. 7C). An ouabain concentration dependent effect was particularly evident at 1000nM drug treatment. The results were different for SW13 cells, which showed a decrease in Akt, p-Akt, p70S6k and p-p70S6k reactiv- ity after ouabain administration (Fig. 7B) or after com- bination regimen (Fig. 7D). Band quantification statis- tical analysis are in Supplemental Tables 1, 2.
A
C
1
10
100
1000
B
C
1
10
100
ß-Actin
Akt
p-Akt
p70S6
p-p70S6
H295R
SW13
C
C
E
El
E2
D
C
E
E1
E3
ß-Actin
Akt
p-Akt
p70S6
p-p70S6
H295R
SW13
Discussion
To our knowledge, this is the first time that antipro- liferative effects of ouabain have been investigated in an established adrenocortical cell line (H295R), in a cell line from a suspected adrenal metastasis (SW13) and in 5 primary adrenocortical tumor cells. We used H295R cell line as an effective model of adrenocortical tumor: this line requires a longer population doubling time >72-96h and produced adrenal steroids, whereas SW13 cells show a rapid duplication time of 36-48h [11]. In the literature no data are available about the therapeutic range of ouabain in serum, even though Clark and collaborators found a range of 0.6-3.2nM for digoxin, which is a CS similar to ouabain in structure and properties [13].
Based on this information and on the fact that to block Na+/K+-ATPase lethal concentrations of ouabain (>10°M) are needed, we sought to treat cells from 1nM
to 1000nM. In combination strategy we used everoli- mus at 5uM, a concentration already tested in adreno- cortical tumors [9].
The present study demonstrated that ouabain alone or in combination with everolimus induced a decrease in cell viability, as shown by MTT assay. The activity of ouabain was more pronounced in SW13 cells and seemed to be time and concentration dependent (Fig. 1A-E). In H295R cells the efficacy of ouabain seemed to be related to time of treatment, as only at 72h was it possible to reach the IC50 (Fig. 1A). The difference between the 2 cell lines might mainly reflect the lon- ger doubling time of H295R than SW13, even if other cell-intrinsic causes (biochemical/cytological) are also involved; in fact H295R cells carry mutations in differ- ent genes, such as CTNNB1, TP53, etc. [15].
These data are in line with previous findings, where cytotoxic effect of ouabain in a panel of cancer cell lines were seen at an IC50 ranging from 50-100nM [16],
[17]. It is noteworthy that combination therapy (oua- bain 10-100nM + everolimus5uM) proved to be more successful in term of cell viability reduction in H295R cells at 72h compared with treatment alone (Fig. 1C).
Primary adrenocortical tumor cells showed no change in cell viability at 24h, but at 120h for APA-1 sample, and in some measure at 72h for another APA sample, we noted a sharp decline of cell viability (70-80%) at 1000nM, probably indicating that ouabain has enough time to act on growing cells (Fig. 1F) and underlining the importance of protracted treatment.
The [3H] thymidine incorporation assay confirmed that ouabain interferes with cell proliferation, thus caus- ing growth arrest (Fig. 2). A stochastic fall is clearly vis- ible between 10nM and 100nM ouabain, suggesting that drug treatments are not faithfully following a sigmoidal curve. Ouabain treatment at 100nM and 1000nM may imply an interference with DNA synthesis, but not with cell viability, as at 100nM (especially in H295R cell) about 50% of cells were still viable by MTT test.
Taken together, all the above results suggest that ouabain may exert antiproliferative effects on tumor cell lines, especially on SW13 cells, even if not at lower concentrations. One previous study reported that ouabain at very high concentration (1 and 10 uM) showed antiproliferative results on 3 different prostate cancer cell lines [14]. A second study described the effects of ouabain (from 0.01 to 100uM) on viability on a small panel of cancer cell lines with essentially the same results [18]. Furthermore combination strategy seemed to be more successful in H295R cells, at least in cell viability and proliferation assay.
Examination of cells by microscopy allows accu- rate distinction between necrosis and apoptosis [19]. Wright’s staining results clearly indicated that necrotic processes predominated over apoptosis (Fig. 3). The DNA cell cycle distribution was altered by ouabain only in SW13 cells, while combination treatment revealed no substantial change in both cell lines. In agreement with our data on ouabain, a previous study analyzed 2 different human hepatocellular cell lines for cell cycle distribution at 24h, after treatment with ouabain at 100 and 500 nM [20].
To discriminate between the two different cellular fates, apoptosis and necrosis, and to confirm micros- copy data, we performed the homogeneous caspase assay and FACS analysis by Annexin V and propidium iodide. The first method analyzes the activation of dif- ferent caspases: 2, 3, 6, 7, 8, 9 and 10, one of the initial
intracellular biochemical events during apoptosis [21]. It is noteworthy that the usual subdivision of necrosis versus apoptosis is a simplification of a more intricate picture. In fact cell death represents contributions by different cell processes, in which apoptosis and necrosis are overlapping phenomena; such overlapping cell death pathways are common behavior rather than an anom- aly [22]. We found activation of caspases (indicative of apoptosis), especially at 72h in SW13 cells, but a major increase in necrosis was found, clearly revealed by cell morphological evaluation (Fig. 3) and FACS analysis with Annexin V-FITC and propidium iodide (Fig. 6). These results are not in contrast, because: a) the activa- tion of caspases is an early event in cell apoptosis, but after ouabain triggers apoptosis, cells may not be able to complete the process and subsequently undergo death via necrosis [23]; b) some caspases (1 and 8) have been observed as driving events toward necrosis and conse- quently ouabain may activate this process via caspase [23]; c) as previously mentioned, the two different pro- cesses may be simultaneously present in a dying cell; d) the translocation of phosphatidylserine to the outer leaflet of cell membrane (evaluated by FACS analysis) may be uncoupled from caspase activation and conse- quently the 2 processes may occur at different times [24]. Moreover a combination regimen seemed not dif- ferently affect cell fate, since the results are similar to those obtained with the ouabain alone. Further studies are needed to discriminate between different modalities of cell death in adrenocortical cell lines.
We analyzed the PI3k/Akt signaling pathways. Constitutive activation of this pathway is found in many malignancies and also in H295R and SW13 cells, as demonstrated in our previous work [9]. In H295R cells, ouabain induced a reduction in p-p70S6k. On the one hand this effect may be associated with decreased proliferation and survival, as underlined by [3H] thymi- dine assay, FACS analysis and homogeneous caspase assay. On the other, it may suggest that a specific biomo- lecular mechanism is modulated by ouabain between Akt and p70S6k, although more research is needed to confirm this hypothesis. As expected, everolimus inhibited p-p70S6k both alone or in combination with ouabain. In SW13 cells, we observed a more general effect, probably indicating that ouabain exerts its con- sequences on the entire pathway. Blocking PI3k/Akt signaling pathways by ouabain may be therapeutically attractive, as it was for everolimus, but this speculation needs to be adequately corroborated.
Our study has shown for the first time antiprolif- erative effects of ouabain on SW13 and H295R cell models and on primary adrenocortical tumor cells. The results obtained are promising, but more data are needed to fully explore the role of ouabain as a poten- tial anticancer drug.
Acknowledgements
We thank Dr. Silvia Vettor for her help with flow cytometry analysis. We also thank Professor John Funder for his critical reading of the manuscript.
Sources of Financial Support
This work was partially supported by ENS@T- CANCER (European Network for the Study of Adrenal Tumors - grant agreement no. 259735) and by AIROB Association (Associazione Italiana per la Ricerca Oncologica di Base, Padova, Italy).
Contribution Statement
We declare that all the listed authors have partici- pated actively in the study and all meet the require- ments of the authorship. R. Pezzani, B. Rubin, M. Redaelli, C. Radu, S. Barollo did the lab work. MV. Cicala and M. Salvà contributed to the literature search. R. Pezzani, M. Redaelli and F. Mantero designed the study. M. Redaelli, R. Pezzani and C. Mucignat-Caretta undertook the statistical analysis. P. Simioni, C. Mian, C. Mucignat-Caretta, M. Iacobone and F. Mantero corrected the draft and directed the work. R. Pezzani wrote the draft of the manuscript and completed the study.
Disclosure
None of the authors have any potential conflicts on interest associated with this research.
| Akt | p-Akt | p70S6k | p-p70S6k | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P< 0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P <0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary |
| C vs 1 | No | ns | C vs 1 | No | ns | C vs 1 | No | ns | C vs 1 | Yes | * |
| C vs 10 | No | ns | C vs 10 | No | ns | C vs 10 | No | ns | C vs 10 | Yes | * |
| C vs 100 | No | ns | C vs 100 | No | ns | C vs 100 | No | ns | C vs 100 | Yes | * |
| C vs 1000 | No | ns | C vs 1000 | Yes | * | C vs 1000 | Yes | * | C vs 1000 | Yes | * |
| 1 vs 10 | No | ns | 1 vs 10 | No | ns | 1 vs 10 | No | ns | 1 vs 10 | No | ns |
| 1 vs 100 | No | ns | 1 vs 100 | No | ns | 1 vs 100 | No | ns | 1 vs 100 | No | ns |
| 1 vs 1000 | No | ns | 1 vs 1000 | No | ns | 1 vs 1000 | No | ns | 1 vs 1000 | No | ns |
| 10 vs 100 | No | ns | 10 vs 100 | No | ns | 10 vs 100 | No | ns | 10 vs 100 | No | ns |
| 10 vs 1000 | No | ns | 10 vs 1000 | No | ns | 10 vs 1000 | No | ns | 10 vs 1000 | No | ns |
| 100 vs 1000 | No | ns | 100 vs 1000 | No | ns | 100 vs 1000 | No | ns | 100 vs 1000 | No | ns |
| Akt | p-Akt | p70S6k | p-p70S6k | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary |
| C vs E | No | ns | C vs E | No | ns | C vs E | No | ns | C vs E | Yes | * |
| C vs E1 | No | ns | C vs E1 | No | ns | C vs E1 | No | ns | C vs E1 | Yes | * |
| C vs E2 | No | ns | C vs E2 | No | ns | C vs E2 | Yes | * | C vs E2 | Yes | * |
| E vs E1 | No | ns | E vs E1 | No | ns | E vs E1 | No | ns | E vs E1 | No | ns |
| E vs E2 | No | ns | E vs E2 | No | ns | E vs E2 | Yes | * | E vs E2 | No | ns |
| E1 vs E2 | No | ns | E1 vs E2 | No | ns | E1 vs E2 | No | ns | E1 vs E2 | No | ns |
| Akt | p-Akt | p70S6k | p-p70S6k | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P< 0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary |
| C vs 1 | No | ns | C vs 1 | Yes | * | C vs 1 | Yes | * | C vs 1 | No | ns |
| C vs 10 | No | ns | C vs 10 | No | ns | C vs 10 | Yes | * | C vs 10 | Yes | * |
| C vs 100 | No | ns | C vs 100 | No | ns | C vs 100 | Yes | * | C vs 100 | Yes | * |
| 1 vs 10 | No | ns | 1 vs 10 | No | ns | 1 vs 10 | No | ns | 1 vs 10 | No | ns |
| 1 vs 100 | No | ns | 1 vs 100 | No | ns | 1 vs 100 | No | ns | 1 vs 100 | No | ns |
| 10 vs 100 | No | ns | 10 vs 100 | No | ns | 10 vs 100 | No | ns | 10 vs 100 | No | ns |
| Akt | p-Akt | p70S6k | p-p70S6k | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P<0.05? | Summary | Dunn's Multiple Comparison Test | Significant? P < 0.05? | Summary |
| C vs E | No | ns | C vs E | No | ns | C vs E | No | ns | C vs E | Yes | * |
| C vs E1 | No | ns | C vs E1 | Yes | * | C vs E1 | Yes | * | C vs E1 | Yes | * |
| C vs E3 | No | ns | C vs E3 | Yes | * | C vs E3 | Yes | * | C vs E3 | Yes | * |
| E vs E1 | No | ns | E vs E1 | Yes | * | E vs E1 | Yes | * | E vs E1 | No | ns |
| E vs E3 | No | ns | E vs E3 | Yes | * | E vs E3 | Yes | * | E vs E3 | No | ns |
| E1 vs E3 | No | ns | E1 vs E3 | No | ns | E1 vs E3 | No | ns | E1 vs E3 | No | ns |
References
1. Zhang L, Zhang Z, Guo H, Wang Y (2008) Na+/ K+-ATPase-mediated signal transduction and Na+/ K+-ATPase regulation. Fundam Clin Pharmacol 22: 615-621.
2. Ihenetu K, Qazzaz HM, Crespo F, Fernandez-Botran R, Valdes R Jr. (2007) Digoxin-like immunoreactive fac- tors induce apoptosis in human acute T-cell lympho- blastic leukemia. Clin Chem 53: 1315-1322.
3. Qiu J, Gao HQ, Zhou RH, Liang Y, Zhang XH, et al. (2007) Proteomics analysis of the proliferative effect of low-dose ouabain on human endothelial cells. Biol Pharm Bull 30: 247-253.
4. Stenkvist B (1999) Is digitalis a therapy for breast carci- noma? Oncol Rep 6: 493-496.
5. Stenkvist B, Bengtsson E, Dahlqvist B, Eriksson O, Jarkrans T, et al. (1982) Cardiac glycosides and breast cancer, revisited. N Engl J Med 306: 484.
6. McConkey DJ, Lin Y, Nutt LK, Ozel HZ, Newman RA (2000) Cardiac glycosides stimulate Ca2+ increases and apoptosis in androgen-independent, metastatic human prostate adenocarcinoma cells. Cancer Res 60: 3807- 3812.
7. Winnicka K, Bielawski K, Bielawska A, Surazyński A (2008) Antiproliferative activity of derivatives of oua- bain, digoxin and proscillaridin A in human MCF-7 and
MDA-MB-231 breast cancer cells. Biol Pharm Bull 31: 1131-1140.
8. Houghton PJ (2010) Everolimus. Clin Cancer Res 16: 1368-1372.
9. Mariniello B, Rosato A, Zuccolotto G, Rubin B, Cicala MV, et al. (2012) Combination of sorafenib and everoli- mus impacts therapeutically on adrenocortical tumor models. Endocr Relat Cancer 19: 527-539.
10. Lacroix A (2010) Approach to the patient with adreno- cortical carcinoma. J Clin Endocrinol Metab 95: 4812- 4822.
11. Wang T, Rainey WE (2012) Human adrenocortical car- cinoma cell lines. Mol Cell Endocrinol 351: 58-65.
12. Mariniello B, Finco I, Sartorato P, Patalano A, Iacobone M, et al. (2011) Somatostatin receptor expression in adrenocortical tumors and effect of a new somatostatin analog SOM230 on hormone secretion in vitro and in ex vivo adrenal cells. J Endocrinol Invest 34: e131-138.
13. Clark WG, Brater DC, Johnson AR, eds. Goth’s Medical Pharmacology. 13th ed. St Louis, Mo: Mosby-Year Book Inc; 1992.
14. Yeh JY, Huang WJ, Kan SF, Wang PS (2001) Inhibitory effects of digitalis on the proliferation of androgen dependent and independent prostate cancer cells. J Urol 166: 1937-1942.
15. Cerquetti L, Sampaoli C, Amendola D, Bucci B, Misiti S, et al. (2010) Mitotane sensitizes adrenocortical can- cer cells to ionizing radiations by involvement of the cyclin B1/CDK complex in G2 arrest and mismatch repair enzymes modulation. Int J Oncol 37: 493-501.
16. Wang Z, Zheng M, Li Z, Li R, Jia L, et al. (2009) Cardiac glycosides inhibit p53 synthesis by a mecha- nism relieved by Src or MAPK inhibition. Cancer Res 69: 6556-6564.
17. Nilubol N, Zhang L, Shen M, Zhang YQ, He M, et al. (2012) Four clinically utilized drugs were identified and validated for treatment of adrenocortical cancer using quantitative high-throughput screening. J Transl Med 10: 198.
18. Simpson CD, Mawji IA, Anyiwe K, Williams MA, Wang X, et al. (2009) Inhibition of the sodium potas- sium adenosine triphosphatase pump sensitizes cancer cells to anoikis and prevents distant tumor formation. Cancer Res 69: 2739-2747.
19. Henery S, George T, Hall B, Basiji D, Ortyn W, et al. (2008) Quantitative image based apoptotic index mea- surement using multispectral imaging flow cytome- try: a comparison with standard photometric methods.
Apoptosis 13: 1054-1063.
20. Xu ZW, Wang FM, Gao MJ, Chen XY, Shan NN, et al. (2011) Cardiotonic steroids attenuate ERK phospho- rylation and generate cell cycle arrest to block human hepatoma cell growth. J Steroid Biochem Mol Biol 125: 181-191.
21. Joza N, Susin SA, Daugas E, Stanford WL, Cho SK, et al. (2001) Essential role of the mitochondrial apoptosis- inducing factor in programmed cell death. Nature 410: 549-554.
22. Overbeeke R, Steffens-Nakken H, Vermes I, Reutelingsperger C, Haanen C (1998) Early features of apoptosis detected by four different flow cytometry assays. Apoptosis 3: 115-121.
23. Leist M, Single B, Naumann H, Fava E, Simon B, et al. (1999) Inhibition of mitochondrial ATP generation by nitric oxide switches apoptosis to necrosis. Exp Cell Res 249: 396-403.
24. Foghsgaard L, Wissing D, Mauch D, Lademann U, Bastholm L, et al. (2001) Cathepsin B acts as a domi- nant execution protease in tumor cell apoptosis induced by tumor necrosis factor. J Cell Biol 153: 999-1010.