Digitalis-like Compounds Facilitate Non- Medullary Thyroid ...mct.aacrjournals.org/content/molcanther/16/1/169.full.pdf · Cancer Biology and Signal Transduction Digitalis-like Compounds

Cancer Biology and Signal Transduction

Digitalis-like Compounds Facilitate Non-Medullary Thyroid Cancer Redifferentiationthrough Intracellular Ca2þ, FOS, and Autophagy-Dependent PathwaysMarika H. Tesselaar1, Thomas Crezee1, Herman G. Swarts2, Danny Gerrits3,Otto C. Boerman3, Jan B. Koenderink2, Hendrik G. Stunnenberg4, Mihai G. Netea5,Johannes W.A. Smit5, Romana T. Netea-Maier5, and Theo S. Plantinga1

Abstract

Up to 20%–30% of patients with metastatic non-medullarythyroid cancer have persistent or recurrent disease resultingfrom tumor dedifferentiation. Tumor redifferentiation torestore sensitivity to radioactive iodide (RAI) therapy is con-sidered a promising strategy to overcome RAI resistance.Autophagy has emerged as an important mechanism in cancerdedifferentiation. Here, we demonstrate the therapeuticpotential of autophagy activators for redifferentiation of thy-roid cancer cell lines. Five autophagy-activating compounds,all known as digitalis-like compounds, restored hNIS expres-

sion and iodide uptake in thyroid cancer cell lines. Upregula-tion of hNIS was mediated by intracellular Ca2þ and FOSactivation. Cell proliferation was inhibited by downregulatingAKT1 and by induction of autophagy and p21-dependent cell-cycle arrest. Digitalis-like compounds, also designated ascardiac glycosides for their well-characterized beneficial effectsin the treatment of heart disease, could therefore represent apromising repositioned treatment modality for patients withRAI-refractory thyroid carcinoma. Mol Cancer Ther; 16(1); 169–81.�2016 AACR.

IntroductionPatients diagnosed with non-medullary thyroid cancer gener-

ally have a good prognosis as a result of implementing treatmentregimens consisting of thyroidectomy and 131I radioactive iodide(RAI) ablation. In contrast, treatment success and overall survivalof patients with RAI-refractory thyroid cancer is poor as no otherefficient treatments are available, leading to recurrences, persistentdisease, and thyroid cancer–related mortality. Mechanistically, itis well established that RAI resistance is accompanied by severetumor dedifferentiation (1, 2). Strategies for restoring thyroid-specific gene expression, designated as redifferentiation, are there-

fore considered as a promising treatment modality to improveclinical responses to RAI treatment.

Autophagy is a ubiquitously expressed degradation machineryfor recycling of intracellular content such as cytoplasmic orga-nelles, proteins, andmacromolecules (3). In addition, autophagyplays important roles in cancer initiation and progression byinfluencing proliferation, differentiation, and anticancer therapyresistance (4, 5). Interestingly, loss of autophagy activity wasshown to be associated with thyroid cancer dedifferentiationand reduced clinical response to RAI therapy (6). On the basisof these observations, we hypothesized that pharmacologic acti-vation of autophagy could facilitate restoration of hNIS expres-sion and, hence, could establish reactivation of iodide uptake inRAI-refractory thyroid cancer. To explore this hypothesis, weinitiated a screening of pharmacologic compounds for theircapacity to induce autophagy and redifferentiation. Selection ofautophagy-activating compounds was based on systematic high-throughput screenings that have identified FDA-approved autop-hagy-activating small molecules, which was demonstratedby a number of standardized readout assays (7, 8). For the currentstudy, we selected 15 small molecules for their effect on thyroidcancer redifferentiation. Furthermore, we have investigatedthe underlying mechanism of thyroid cancer redifferentiationinduced by autophagy activators through complementaryapproaches at both the cellular and biochemical level.

Materials and MethodsCell culture and reagents

Thyroid cancer cell lines BC-PAP (papillary, BRAFV600Emuta-tion), FTC133 (follicular, PTEN deficient), and TPC-1 (papillary,

1Department of Pathology, Radboud University Medical Center and RadboudInstitute for Molecular Life Sciences, Nijmegen, the Netherlands. 2Department ofPharmacology & Toxicology, Radboud University Medical Center and RadboudInstitute for Molecular Life Sciences, Nijmegen, the Netherlands. 3Department ofRadiology &NuclearMedicine, RadboudUniversityMedical Center andRadboudInstitute for Molecular Life Sciences, Nijmegen, the Netherlands. 4Department ofMolecular Biology, Radboud University Medical Center and Radboud Institutefor Molecular Life Sciences, Nijmegen, the Netherlands. 5Department of InternalMedicine, Radboud University Medical Center and Radboud Institute for Molec-ular Life Sciences, Nijmegen, the Netherlands.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Theo S. Plantinga, Department of Pathology, RadboudUniversity Medical Center, Geert Grooteplein Zuid 10, 6500 HB Nijmegen, theNetherlands. Phone: 312-4361-4382; Fax: 312-4366-8750; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-16-0460

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 169

on August 25, 2019. © 2017 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst November 11, 2016; DOI: 10.1158/1535-7163.MCT-16-0460

http://mct.aacrjournals.org/

RET/PTC rearrangement) were obtained from the sources previ-ously described and were authenticated by short tandem repeatprofiling (9). Cell lines FTC133 and TPC-1 were cultured inDMEM (Invitrogen) supplemented with gentamicin (10 mg/mL),pyruvate (10 mmol/L), and 10% FCS (Invitrogen). Cell line BC-PAP was cultured in RPMI medium (Invitrogen) supplementedwith gentamicin (10 mg/mL), pyruvate (10 mmol/L), and 10%FCS (Invitrogen). Cells were incubated with DMSO (vehiclecontrol), the mTOR inhibitor rapamycin (Biovision), or withthe selected autophagy-activating compounds (SupplementaryTable S1) for the indicated time points and concentrations. For allcompounds, stock concentrations ranged from50 to100mmol/L.Selection of the compounds and applied concentrations arederived from earlier studies (7, 8). Distribution coefficients(cLogD) at pH ¼ 7.0 for estimating lipophilicity and polarityof DLCs was determined in silico by ChemBioOffice softwarepackage. For mRNA expression knockdown of ATF3 and FOS,SMARTpool siRNA oligonucleotides targeting scramble, ATF3, orFOS (Thermo Fisher Scientific) were transfected with FuGene HD(Promega) according to the manufacturer's instructions.

Real-time quantitative PCRThyroid cancer cell lines were treated with TRIzol reagent

(Invitrogen), and total RNApurificationwas performed accordingto the manufacturer's instructions. Isolated RNAwas subsequent-ly transcribed into cDNA using iScript cDNA synthesis kit (Bio-Rad Laboratories) followed by quantitative PCR using the SYBRGreen method (Life Technologies). The following primers wereused: hNIS (SLC5A5) forward 50-TCC-TGT-CCA-CCG-GAA-TTA-TCT-30 and reverse 50-ACG-ACC-TGG-AAC-ACA-TCA-GTC-30;ATF3 forward 50-CCT-CTG-CGC-TGG-AAT-CAG-TC-30 andreverse 50-TTC-TTT-CTC-GTC-GCC-TCT-TTT-T-30; FOS forward50-GGG-GCA-AGG-TGG-AAC-AGT-TAT-30 and reverse 50-CCG-CTT-GGA-GTG-TAT-CAG-TCA-30; JUN forward 50-TCC-AAG-TGC-CGA-AAA-AGG-AAG-30 and reverse 50-CGA-GTT-CTG-AGC-TTT-CAA-GGT-30; TTF1 forward 50-AGC-ACA-CGA-CTC-CGT-TCT-C-30 and reverse 50-GCC-CAC-TTT-CTT-GTA-GCT-TTC-C-30, TTF2 forward 50-CAC-GGT-GGA-CTT-CTA-CGG-G-30

and reverse 50-GGA-CAC-GAA-CCG-ATC-TAT-CCC-30; and PAX8forward 50-AGT-CAC-CCC-AGT-CGG-ATT-C-30 and reverse 50-CTG-CTC-TGT-GAG-TCA-ATG-CTT-A-30. Data were corrected forexpression of the housekeeping gene b2microglobulin, for whichthe primers forward, 50-ATG-AGT-ATG-CCT-GCC-GTG-TG-30,and reverse, 50-CCA-AAT-GCG-GCA-TCT-TCA-AAC-30, were used.

Western blottingForWestern blotting of hNIS protein, 5� 106 cells were lysed in

40 mL of lysis buffer [50mmol/L Tris (pH 7.4), 150mmol/LNaCl,2mmol/L EDTA, 2mmol/L EGTA, 10%glycerol, 1%TritonX-100,40 mmol/L b-glycerophosphate, 50 mmol/L sodium fluoride,200 mmol/L sodium vanadate, 10 mg/mL leupeptin, 10 mg/mLaprotinin, 1 mmol/L pepstatin A, and 1 mmol/L phenylmethyl-sulfonyl fluoride]. The homogenate was frozen and then thawedand centrifuged at 4�C for 3 minutes at 14,000 � g, and thesupernatant was mixed with a loading buffer containing dithio-threitol, incubated at 37�C for 30minutes (no boiling), and takenfor Western blot analysis. Equal amounts of protein were sub-jected to SDS-PAGE using 10% polyacrylamide gels. After SDS-PAGE, proteins were transferred to nitrocellulose membrane(0.2 mm). The membrane was blocked with 5% (w/v) milk

powder in TBS/Tween 20 for 1 hour at room temperature, fol-lowed by incubation overnight at 4�C with an hNIS antibody(1:500, 250552; Abbiotec) in 5% milk powder in TBS/Tween 20or with an actin antibody (loading control, 1:1,000, A2066;Sigma) in 5% milk powder in TBS/Tween 20. After overnightincubation, the blots were washed three times with TBS/Tween 20and then incubated with horseradish peroxidase–conjugatedswine anti-rabbit antibody at a dilution of 1:5,000 in 5% (w/v)milk powder in TBS/Tween 20 for 1 hour at room temperature.After being washed three times with TBS/Tween 20, the blots weredeveloped with enhanced chemiluminescence (GE Healthcare)according to the manufacturer's instructions.

Cell proliferation and viability assaysFor measurement of proliferation, MTT assays have been per-

formed according to the manufacturer's instructions (Sigma-Aldrich). In brief, cells were plated in 96-well plates with a celldensity of 1,000 cells per well. After the indicated time points andtreatments, MTT substrate was added and the amount of con-verted substrate was measured by a plate reader at 570 nm. Inaddition, at the same time points, activity of lactate dehydroge-nase (LDH) has been measured in cell culture supernatants forassessment of cell death in accordance with the manufacturer'sprotocol (CytoTox 96 kit, Promega).

In vitro iodide uptakeIn vitro [125I]-iodide uptake was determined as described pre-

viously (10). During 72 hours before the assessment of iodideuptake, cells were cultured with rapamycin (50 mg/mL), digoxin,strophanthin K, lanatoside C, digoxigenin, or proscillaridin A, allin 50 mmol/L concentration. Cells were incubated for 30 minuteswith 1 kBq Na125I (Perkin-Elmer) and 20 mmol/L nonradioactiveNaI as a carrier, with or without 80 mmol/L of sodium perchlorateto control for hNIS-specific uptake. The radioactive medium wasaspirated and the cells were washedwith ice-cold PBS and lysed in0.1 mol/L NaOH buffer at room temperature. Radioactivity wasmeasured in the cell lysates in an automatic gamma counter(Wizard2, Perkin-Elmer). In parallel experiments, DNA was iso-lated from the cells by standard procedures (Puregene kit, GentraSystems) and quantified by Nanodrop measurements (ThermoFisher Scientific). Accumulated radioactivitywas expressed as cpmper nanogram of DNA.

RNA sequencingFor RNA sequencing analysis, 5 � 106 cells were cultured for

either 24, 48, or 72 hours with 50 mmol/L hNIS-inducing DLCs.Thyroid cancer cell lines were treated with TRIzol reagent (Invi-trogen) and total RNA purification was performed according tothe manufacturer's instructions. Successively, total RNA was sub-jected to (i) ribosomal RNA depletion by using riboZero (Illu-mina), (ii) DNAse treatment, (iii) RNA fragmentation, (iv) first-and second-strand cDNA synthesis with random hexamers, (v)library preparation by applying Kapa Hyper Prep (Kapa Biosys-tems), and (vi) RNA sequencing by Illumina HiSeq 2000. RawunmappedRNA sequencingdataweremappedon theUCSChg19human genome assembly and were analyzed by Tuxedo toolsTopHat and CuffDiff in the RNA sequencing module on theGenePattern server of Broad Institute (www.genepattern.broad-institute.org) to analyze differential expression. Processed datawere visualized by the Rstudio CummeRbund tool (11). Venn

Tesselaar et al.

Mol Cancer Ther; 16(1) January 2017 Molecular Cancer Therapeutics170




diagramswere constructed byusingVenny v2.1 (http://bioinfogp.cnb.csic.es/tools/venny/). Protein networks were built by apply-ing STRING v10 (http://string-db.org/; ref. 12). Raw RNAsequencing data are deposited in the GEO database under acces-sion number GSE79400.

Intracellular calcium assaysFor assessment of intracellular Ca2þ accumulation evoked by

DLCs, cells were loaded with Fura-2 (Thermo Fisher Scientific,dilution 1:500 in culture medium without phenol red) for 30minutes, washed, and administered to 50 mmol/L or 5 mmol/Lconcentrations ofDLCs. Immediately after addition ofDLCs, cellswere consecutively imaged at 340 and 380 nm by a BD Pathway855 Robotic high content microscope every 30minutes for a totalduration of 7.5 hours. An automatic threshold algorithm wasapplied, background subtractions were performed, and ratios of340 nm:380 nm were calculated by FIJI software.

Cellular [3H]ouabain binding assaysAffinity of thyroid cancer cell lines for Naþ/Kþ ATPase–

dependent DLC binding was measured by competition assayswith [3H]-ouabain (Perkin-Elmer). Cells (5 � 105/well) wereseeded in 24-well plates and were simultaneously incubated for90 minutes at 37�C with 10 nmol/L [3H]-ouabain and differentconcentrations of DLCs in RM-medium (20 mmol/L Hepes-NaOH, 130 mmol/L NaCl, 0.5 mmol/L MgCl2, 0.2 mmol/LNa2HPO4, 0.4 mmol/L NaH2PO4, pH 7.4) supplemented with1.5 mmol/L sodium vanadate. To control for specific binding toNaþ/Kþ ATPases, 5 mmol/L KCl was added. Afterwards, cellswere washed with cold RM-medium and lysed in 50% meth-anol. The amount of [3H]-ouabain radioactivity was measuredin cell lysates by a liquid scintillation counter after addition of4 mL OptiFluor (Canberra Packard). IC50 was determined bycalculating the DLC concentration corresponding to 50% ofmaximal 3H-ouabain–binding capacity that was measured inthe absence of unlabeled DLCs.

Statistical analysisStatistical significance was tested with Student t test, Mann–

Whitney U test, Spearman rank correlation coefficient or byWilcoxon matched-pairs signed rank test, when appropriate.P values below 0.05 were considered statistically significant.For RNA sequencing data, a false discovery rate of 0.05 wasincorporated.

ResultsAutophagy-activating compounds restore hNIS expression andfunctional iodide uptake in thyroid cancer cell lines

To assess whether autophagy-activating compounds are capableof reactivating hNIS expression in dedifferentiated thyroid cancercell lines BC-PAP, FTC133, and TPC-1, cells were incubated for72 hours with DMSO vehicle, 15 different autophagy-activatingcompounds that were previously identified as such (Supplemen-tary Table S1; ref. 7), or with the mTOR inhibitor rapamycin aspositive control for activating hNIS expression (Fig. 1A). Five of15 compounds were shown to upregulate hNIS in at least one ofthe cell lines, some only at 50 mmol/L concentrations and somealso at 5 mmol/L, all of which are digitalis-like compounds (DLC);four compounds (digoxin, digoxigenin, proscillaridinA, strophan-tin K) in FTC133, lanatoside C in TPC-1, and proscillaridin A in

BC-PAP. Gene expression data were confirmed by Western blot-ting for detection of hNIS protein, demonstrating similar orhigher expression of hNIS after 72 hours, both the inactive 56kDa as the glycosylated 87 kDa active form, as compared with thepositive control rapamycin (Fig. 1B). Functional hNIS expressionwas assessed by measurement of iodide uptake capacity. After 72hours of culturing in the presence of DLCs, [125I]-iodide uptakewas increased comparable with rapamycin and was efficientlyinhibited by the hNIS high-affinity substrate sodium perchlorate(Fig. 1C).

DLCs induce cell death and cell-cycle arrest in thyroid cancercell lines

Additional sets of experiments were conducted to investigatethe effects of DLCs on cell survival, metabolic activity, andproliferation. First, MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assays were performed to mea-sure metabolic activity of BC-PAP, FTC133, and TPC-1 afteradministration of DLCs. The results indicate that metabolicactivity is completely inhibited upon DLC treatment after 48–72 hours of culture, for most DLCs at and above concentrationsof 1 mmol/L (Fig. 2A). Second, to determine the contribution ofeither cell death or cell-cycle arrest pathways in blockingmetabolic activity, lactate dehydrogenase (LDH) activity wasmeasured in the respective cell culture supernatants. AlthoughLDH release was induced to some extent by DLCs in all celllines, 30%–90% of cells survived after 72 hours of treatment(Fig. 2B). As these surviving cells exhibit inactive metabolism,cell-cycle arrest has likely occurred.

Genome-wide transcriptomics reveals cellular pathwaysdriving hNIS reactivation and cell-cycle arrest by DLCs inthyroid cancer cell lines

To provide mechanistic insights into the molecular pathwaysdriving hNIS reactivation by DLCs, RNA sequencing of thyroidcancer cell lines was performed after 24, 48, and 72 hours ofDLC administration (proscillaridin A for BC-PAP; digoxin, digox-igenin, proscillaridin A, and strophantin K for FTC133; lanatosideC for TPC-1) and was compared with vehicle-treated cells.High quality RNA sequencing data, on average 3 � 107 reads persample, was obtained as determined by the construction ofdispersion, density, and volcano plots. Furthermore, greatly dis-tinct gene expression patternswere observed between vehicle- andDLC-treated cells by performing principal component analysis(Supplementary Figs. 1–6). Total number of identified transcriptsand total number of differentially expressed genes reaching sta-tistical significance [based on uncorrected P values �0.05 andwith expression difference of at least 2 fold (2log � 1 or 2log ��1)] was on average 20,406 and 5,168 for BC-PAP, 19,777and 5,946 for FTC133, 19,880 and 7,609 for TPC-1, respectively.For subsequent analyses, thementioned gene sets only containingdifferentially expressed genes reaching statistical significancewere selected. Significantly, differently expressed genes were fur-ther analyzed in Venn diagrams to depict overlapping genesbetween 24-, 48-, and 72-hour time points for over- and under-expressed genes separately and per single cell line–DLC combi-nation (Supplementary Figs. 1–6). Furthermore, heatmaps wereconstructed to depict the 25 most pronounced up- and down-regulated genes per cell line–DLC combination (Fig. 3A–C).To identify common players involved in hNIS reactivation in

DLC-induced Thyroid Cancer Redifferentiation

www.aacrjournals.org Mol Cancer Ther; 16(1) January 2017 171




0

0 0 0

100

200

300

20

40

60

80

200

[12

5 I]

up

take

(c

pm

/ng

DN

A)

[12

5 I]

up

take

(c

pm

/ng

DN

A)

[12

5 I]

up

take

(c

pm

/ng

DN

A)

400

600

ND

Vehicle

Vehicle

Vehicle Lanatoside CRapamycin

Rapamycin Digoxin Strophantin K Digoxigenin Proscillaridin A

Proscillaridin A

hNIS 87 kDa

hNIS 56 kDa

b-Actin

hNIS 87 kDa

hNIS 87 kDa

hNIS 56 kDa

hNIS 56 kDa

b-Actin

b-Actin

ND ND ND

Vehicl

eVeh

icle

Vehicl

e

Rapam

ycin

Vehicl

e

Vehicl

e

Vehicl

e

Lanat

oside C

Lanat

oside C

+ N

a+ CIO

4–

Digox

in

Digox

igen

in

Digox

igen

in +

Na+ CIO

4–

Prosc

illarid

in A

Prosc

illarid

in A

+ N

a+ CIO

4–

Strophan

tin K

Strophan

tin K

+ N

a+ CIO

4–

Digox

in +

Na+ CIO

4–

Rapam

ycin

Rapam

ycin

Prosc

illarid

in A

Prosc

illarid

in A

+ N

a+ CIO

4–

Rapam

ycin

+ N

a+ CIO

4–

Rapam

ycin

+ N

a+ CIO

4–

Na+ CIO

4–

Na+ CIO

4–

Na+ CIO

4–

Rapam

ycin

Rapam

ycin

Lanat

oside C

50 mm

ol/L

Prosc

illarid

in A

50 mm

ol/L

Prosc

illarid

in A

5 mm

ol/L

Digox

in 50

mmol/L

Digox

igen

in 50

mmol/L

Prosc

illarid

in A

50 mm

ol/L

Prosc

illarid

in A

5 mm

ol/L

Digox

in 5

mmol/L

Strophan

tin K

50 mm

ol/L

Strophan

tin K

5 mm

ol/L

50

0 0

500

1,000

1,500

TPC-1

200400600800

1,000

2,0003,0004,0005,0006,000

100

150

200R

elat

ive

hN

IS e

xpre

ssio

n(f

old

incr

ease

)

Rel

ativ

e h

NIS

exp

ress

ion

(fo

ld in

crea

se)

Rel

ativ

e h

NIS

exp

ress

ion

(fo

ld in

crea

se)

250

A

B

C

BC-PAP

BC-PAP

FTC133

TPC-1

FTC133

TPC-1BC-PAP FTC133

Figure 1.

hNIS mRNA and protein expression and iodide uptake induced by digitalis-like compounds. A, hNIS mRNA expression in thyroid cancer (TC) cell lines BC-PAP,FTC133, and TPC-1 after 72 hours of treatment with the indicated digitalis-like compounds or 50 mg/mL rapamycin. Data are obtained from three independentexperiments. Data are means � SD. ND, not detectable. B, hNIS protein expression in thyroid cancer cell lines BC-PAP, FTC133 and TPC-1 after 72 hours oftreatment with 50 mmol/L of the indicated digitalis-like compounds or 50 mg/mL rapamycin. Data are representative of three independent experiments.Images are cropped; uncropped images are provided in Supplementary Fig. S12. C, Radioactive iodide [125I] uptake by thyroid cancer cell lines BC-PAP, FTC133,and TPC-1 after 72 hours of treatment with 50 mmol/L of the indicated digitalis-like compounds or 50 mg/mL rapamycin and with or without sodium perchlorate.Data are obtained from three independent experiments. Data are means � SD. � , P < 0.05 (Mann–Whitney U test).

Tesselaar et al.





thyroid cancer cell lines by DLC treatment, genes that wereidentified as consistently up- or downregulated at all three timepoints, represented as overlapping centres of Venn diagrams inSupplementary Figures 1–6 comprising on average 508 over- and900 underexpressed genes per thyroid cancer cell line–DLC com-bination, were selected for comparison between thyroid cancercell line–DLC combinations. Before composing overall Venndiagrams comprising all three cell lines, other Venn diagramswere drawn for comparison of FTC133 datasets separately basedon treatment with different DLCs (digoxin, digoxigenin, proscil-laridinA, and strophantinK). Between these FTC133datasets, 391and 496 genes were commonly up- or downregulated, respec-tively (Fig. 3D). Ultimately, based on total 27 independenttranscriptomics datasets, Venn diagrams were built to identifyoverlapping genes that are differentially expressed in all thyroidcancer cell line–DLC combinations at all 24, 48, and 72-hour timepoints (Fig. 3D) and resulted in concise gene lists of 128 over- and125 underexpressed genes that are listed in Supplementary TableS2 and Supplementary Table S3, respectively. To visualize gene

networks and to perform Gene Ontology (GO) pathway analysisbased on the lists of commonly regulated genes (128 up and 125down), functional protein–protein interaction networks weregenerated by applying STRINGv10 (http://string-db.org/; ref. 12).These analyses revealed broad protein interaction networks ofboth up- and downregulated genes (Supplementary Fig. S7) andidentified upregulated FOS, ATF3, EGR1, CDKN1A, and autop-hagy genes and downregulated AKT1 as central molecular playerswithin these networks. Furthermore, GO Biological Processespathway analysis demonstrated statistically significant enrich-ment scores (FDR ¼ 0.05) for numerous upregulated pathways,especially those involving autophagy (Supplementary Fig. S8). Incontrast, no significantly downregulatedGOpathways were iden-tified (data not shown).

Reactivation of hNIS by DLCs is FOS, but not ATF3, dependentFor functional validation of RNA sequencing data, additional

experiments were performed into the role of the early-responsegenes FOS and ATF3, as both are consistently upregulated by

Figure 2.

Effects of digitalis-like compounds on proliferation and survival of thyroid cancer (TC) cell lines. A, Metabolic activity of thyroid cancer cell lines treated with eitherDMSO vehicle (Veh.) or indicated concentrations of digitalis-like compounds for up to 72 hours. Data are means � SD (N ¼ 4). B, Induction of thyroid cancercell death by either DMSO vehicle or different concentrations of digitalis-like compounds for up to 72 hours. Data are expressed as percentage of lactatedehydrogenase release, mean � SD (N ¼ 4).






DLCs in all cell lines and both have previously been recognized aspotential drivers of hNIS expression through hNIS upstreamenhancer (hNUE) binding in cooperation with the transcriptionfactor PAX8 (13, 14). First, modulation of ATF3 and FOS geneexpression has been determined in BC-PAP, FTC133, and TPC-1

by treatment with DLCs for 48 and 72 hours (Fig. 4). As antic-ipated, in all cell lines, both ATF3 and FOS expression wasincreased by a number of DLCs. Interestingly, except for theBC-PAP cell line, DLCs capable of upregulating hNIS also mostpotently induced ATF3 and FOS expression, that is, digoxin,

Figure 3.

Transcriptomics, systembiology, and pathway analysis on thyroid cancer cell lines treatedwith digitalis-like compounds.A–C,RNA sequencing heatmaps of BC-PAP(A), FTC133 (B), and TPC-1 (C) after treatment with vehicle or 50 mmol/L of the indicated digitalis-like compounds (DLC) at 24, 48, and 72 hours. The 25 mostupregulated and the 25 most downregulated genes are depicted for all three time points combined. D, Venn diagrams of differentially expressed genesof (1, left panels) digitalis-like compound treated FTC133 (all four compounds) and (2, right panels) of proscillaridin A–treated BC-PAP, commonly up- ordownregulated genes in FTC133 and lanatoside C–treated TPC-1.

Tesselaar et al.





digoxigenin, strophantin K, and proscillaridin A for FTC133 andlanatoside C for TPC-1, also reflected by Spearman rho tests andcorresponding P values. Follow-up experiments of ATF3 and FOSgene silencing by siRNA revealed that FOS, but not ATF3, repre-sents the transcription factor driving hNIS expression in all celllines tested (Fig. 5).On the basis of RNA sequencing data, anotherhNIS transcription factor, JUN, was the highest upregulated genein hNIS-positive BC-PAP cells and could therefore be of specificimportance as cofactor of FOS for proscillaridin A–induced hNIS

expression inBC-PAP. Indeed, JUNexpressionwas increasedmostpotently in BC-PAP treated with proscillaridin A as comparedwith the other treatment conditions, again corroborated bySpearman rho statistics. In contrast, for FTC133 and TPC-1, nosignificant correlations were observed between expression of JUNand hNIS (Supplementary Fig. S9). Importantly, expression ofthyroid transcription factors TTF1, TTF2, and PAX8 was notmodulated by DLC treatment in any of the cell lines (Supple-mentary Fig. S10).

Figure 4.

Functional validation of transcriptomics data. Expression of FOS and ATF3 after treatment for 48 and 72 hours of BC-PAP, FTC133, and TPC-1 with theindicated digitalis-like compounds. Data are obtained from three independent experiments. Data are means � SD. Square boxes represent statistical outputof Spearman rho tests for degree of correlation between expression of FOS/ATF3 with hNIS expression at 48- and 72-hour time points. ND, not detectable.






Increase of intracellular Ca2þ by DLCsPharmacologically, DLCs selectively bind to membranous

Naþ/Kþ ATPases, thereby inhibiting Naþ and Kþfluxes across

the plasma membrane resulting in intracellular Ca2þ accumu-lation facilitated by the Naþ/Ca2þ exchanger (NCX). To inves-tigate the mechanism underlying differential responses of thy-roid cancer cell lines BC-PAP, FTC133, and TPC-1 to thedifferent DLCs tested, intracellular Ca2þ concentrations weremeasured by labeling of cells with the ratiometric fluorescentdye Fura-2 which binds to free intracellular Ca2þ. After loadingcells with Fura-2, cells were exposed to DLCs in 50 mmol/L or 5mmol/L concentrations for up to 7.5 hours and Fura-2 ratioswere determined by measuring fluorescence at 340 and 380 nmevery 30 minutes. Significant changes in intracellular Ca2þ

concentrations were observed by comparing the effects ofdifferent DLCs. In FTC133 and TPC-1, DLCs that are capableof upregulating FOS and hNIS induced the highest elevation ofintracellular Ca2þ, whereas this was less increased by the hNIS-negative compounds in the respective cell lines. In BC-PAP,however, intracellular Ca2þ was increased at early time pointsby proscillaridin A and remained stable afterwards, whereashNIS-negative DLCs evoked increased intracellular Ca2þ only atlater time points ultimately resulting in higher concentrationsof intracellular Ca2þ (Fig. 6). These data suggest that in a celltype–specific manner the magnitude of intracellular Ca2þ accu-mulation induced by DLCs determines whether reactivation ofhNIS expression occurs.

DLC binding affinity and pharmacokineticsTo further explore the pharmacologic processes responsible for

DLC-specific responses in thyroid cancer cell lines, cellular bind-ing affinity of DLCs was assessed by [3H]-ouabain competitionassays andwas analyzed in relation toNaþ/KþATPase expression.Proscillaridin A, digoxigenin, and strophantin K were shown tobind with highest affinity to all thyroid cancer cell lines, whereasfor digoxin and lanatoside C, binding affinity was about 10-foldlower (Fig. 7A). Expression of Naþ/Kþ ATPase isoforms andsubunits, of which only ATP1A1, ATP1B1, ATP1B3, and FXYD5could be detected, revealedmarginal differences in both na€�ve andDLC-treated cells, although DLC treatment mostly upregulatedNaþ/Kþ ATPase expression (Fig. 7B). Other players influencingDLC pharmacokinetics are membrane transporters facilitatinginflux, that is, organic anion transporters such as SLCO4C1, andefflux by ABC transporters including multidrug resistance gene(MDR1/ABCB1) of most, but not all, DLCs. Heatmaps based onRNA sequencing data were constructed to compare expression oforganic anion transporters as potential active influx mechanismand ABC transporters representing putative efflux channels.Whereas in FTC133 several ABC transporters are highly expressedand SLC transporters relatively low, in BC-PAP and TPC-1, highexpression of SLCO4A1 is observed in vehicle-treated cells (Sup-plementary Fig. S11). In addition to cellular features that deter-mine DLC efficacy, also physicochemical properties of DLCscould modulate its bioavailability for Naþ/Kþ ATPase inhibition.Importantly, between tested DLCs, large differences in

Figure 5.

Functional validation of transcriptomics data. Expression of ATF3, FOS, and hNIS after 72 hours of treatment with 50 mmol/L of the indicated digitalis-likecompounds in the presence or absence of siRNA oligonucleotides targeting either scramble, ATF3, or FOS. Data are obtained from three independent experiments.Data are means � SD. ND, not detectable.

Tesselaar et al.





lipophilicity and polarity were predicted, with proscillaridin A asmost lipophilic and lanatoside C as least lipophilic (Supplemen-tary Table S4).

DiscussionMechanisms responsible for dedifferentiation and concom-

itant development of RAI refractoriness in thyroid cancerpatients are incompletely understood, although recent findingsindicate that interactions between oncogenic and autophagypathways are highly relevant (1, 15, 16). Furthermore, molec-ular targeting of MAPK and mTOR kinases, with autophagy aspotential effect modifier (6), have emerged as promising treat-ment modalities leading to, although partially, restored sensi-tivity to RAI therapy by inducing redifferentiation (17–19). Wetherefore hypothesized that autophagy could represent a targetfor adjunctive therapy to restore RAI-sensitivity in dedifferen-tiated thyroid cancer. Accordingly, the current study demon-strates that DLCs, a specific class of autophagy-activatingagents, induce robust redifferentiation of thyroid cancer cell

lines irrespective of the tumor genetic background being eithermutated BRAF, PTEN loss, or RET/PTC rearrangement. In alltested thyroid cancer cell lines, at least one of five DLCs restoredfunctional hNIS expression and increased capacity for iodideuptake up to 400-fold. Genome wide transcriptomic profilingby RNA sequencing and functional assays revealed intracellularCa2þ-dependent expression of the transcription factor FOS,with in specific conditions JUN as potential cofactor, as pre-requisite for functional hNIS expression. In addition, DLCswere confirmed to activate autophagy, were shown to evokeCDKN1A/p21 expression, and to repress AKT1, culminating incell-cycle arrest and, to some extent, cell death.

DLCs, also termed cardiac glycosides, are well characterizedNaþ/Kþ ATPase inhibitors that block Naþ efflux and, conse-quently, accelerate passive transport through the Naþ/Ca2þ

exchanger (NCX) and promote accumulation of cytoplasmicCa2þ (20). Because of this, DLCs are effective drugs for treat-ment of cardiac failure or arrhythmia by increasing cardiomyo-cyte contractility, for which digoxin is still in clinical use (21,22). Furthermore, regulation of cytoplasmic Ca2þ, at the level

Figure 6.

Intracellular calcium accumulation in thyroid cancer cell lines BC-PAP, FTC133, and TPC-1 treated with five different digitalis-like compounds (DLC) stratifiedfor their capacity to reactivate hNIS expression (hNIS-positive vs. hNIS-negative). Cells were loaded with Fura-2 and fluorescent intensity was measured at340and380nmevery 30minuteswith a total duration of 7.5 hours. Ratios of 340:380were calculated byFIJI software. Data are representative for three independentexperiments (N ¼ 15). Data are means � SEM. � , P < 0.05 (Mann–Whitney U test).






of the plasma membrane as well as intracellular storage com-partments, has been previously demonstrated to influence abroad spectrum of cellular processes implicated in both phys-iologic and pathologic conditions. Pathways of differentiationare among the most prominently influenced mechanisms inthis respect, which has been shown for an extensive collectionof cell types (23–27). Changes in intracellular Ca2þ have alsobeen linked to carcinogenesis, proliferation, and dedifferenti-ation grade of malignant cells, with autophagy as importantpathway involved in processes of UPR and ER stress and in

regulation of Ca2þ transport between different intracellularCa2þ storage compartments (28–32). DLCs have been exten-sively investigated as repositioned anticancer treatment basedon their antiproliferative and apoptotic effects, selectivelyinduced in malignant cells only, and was revealed to resultfrom perturbations of multiple pathways (reviewed in ref. 33).In addition, systematic high-throughput screening studies havedemonstrated DLCs as potentially effective treatment modal-ities for myelodysplastic syndrome (34) and prostate cancer(35).

Figure 7.

Pharmacokinetics and pharmacodynamics of effects of digitalis-like compounds on thyroid cancer (TC) cell lines. A,DLC-binding affinity for thyroid cancer cell linesBC-PAP, FTC133, and TPC-1 determined by 3H-ouabain competition assays. Data are means � SD (N ¼ 6). IC50, 50% inhibitory concentration. B, Geneexpression of Naþ/Kþ ATPase subunits ATP1A1, ATP1B1, ATP1B3, and FXYD5 in untreated and digitalis-like compound treated BC-PAP, FTC133, and TPC-1. Data aremeans � SD (N ¼ 6).

Tesselaar et al.





In the current study, not all tested DLCs reactivated hNISexpression in all thyroid cancer cell lines, which was demonstrat-ed to be associated with their capacity to sufficiently elevateintracellular Ca2þ concentrations and, for BC-PAP specifically,the capacity to enable long-term Ca2þ accumulation within acertain range. Pharmacologically, this could potentially be due todifferential binding affinityofDLCs toBC-PAP, FTC133, andTPC-1 based on expression profiles of Naþ/Kþ ATPase genes or todifferences inDLCpharmacokinetics. Althoughminor differencesin ATPase gene expression were present between cell lines, bind-ing affinities for DLCs were highly similar. Pharmacokinetics ofDLCs is complex and at least involves a combination of active andpassive transport across the cell membrane, with Naþ/Kþ ATPaseinternalization (36), SLCO4C1-mediated influx (37), andABCB1/MDR1-dependent efflux (38, 39) as active transportmechanisms. Whereas ABCB1/MDR1 and SLCO4C1 are mini-mally expressed by thyroid cancer cell lines, other ABC transporterhomologs (ABCB2/TAP1, ABCB3/TAP2, ABCB7, ABCB8) areespecially expressed by FTC133 and the organic anion transporterSLCO4A1 by specifically BC-PAP and TPC-1. These differencescould potentially confer differential responses of thyroid cancercell lines to DLCs bymodulating their extracellular concentrationin time, although it remains to be proven whether these trans-porters are involved in DLC transport. In addition, molecularlipophilicity andpolarity properties ofDLCs are important factorsproportionally related to both passive and active transport acrossthe cell membrane, thereby influencing its bioavailability forNaþ/Kþ ATPase inhibition (40). By in silico molecular modeling,proscillaridin Awas predicted asmost lipophilic and lanatoside Cas least lipophilic of all tested DLCs. On the basis of theseobservations, one could envision that variable configurations ofDLC lipophilicity combined with relative contributions of activeand passive DLC transport, overall determining in vitro DLCbioavailability for Naþ/Kþ ATPase inhibition, are responsible forthe differential responses of thyroid cancer cell lines to DLCtreatment. Differences in in vitro DLC bioavailability also hasimportant implications for the interpretation of DLC dosagesapplied here, as the biologically available dose forNaþ/KþATPaseinhibition is probably variable and could differ per DLC and perthyroid cancer cell line.

The proto-oncogene FOS is an important player in the phys-iologic response of thyroid follicular cells to thyroid-stimulatinghormone (TSH), mediating signals of proliferation and thyroid-specific gene expression downstream of the second messengercyclic adenosine monophosphate (cAMP). For hNIS expressionspecifically, FOS has been identified as transcription factor bind-ing to the hNIS upstream enhancer (hNUE) in conjunction withPAX8 (14). Accordingly, FOS expression has been associated withbenign thyroid neoplasms and differentiated thyroid cancer,suggesting FOS as a marker of the thyroid differentiation state(41, 42). In both physiologic and malignant cell types, it is wellestablished that FOS expression is regulated by its transcriptionfactor cAMP response element-binding protein (CREB) that isphosphorylated under the influence of intracellular Ca2þ withcalmodulin, cAMP, RAS, and ERK as intermediate factors (illus-trated in Fig. 8) (43–45). As opposed to FOS-induced hNISexpression, proliferative effects of FOS are strongly inhibited byDLC treatment. One major explanation for the observed inhibi-tion of proliferation is concomitant activation of autophagy, thatis likely to be elicited by DLCs through both overlapping anddistinct mechanisms regulated by intracellular Ca2þ involving

ERK signaling, mTOR inhibition by AMPK and TFEB signaling(illustrated in Fig. 8) enabled by activation of calcineurin, that isalso known to inhibit BRAF (46–49). Furthermore, DLCs induceCDKN1A/p21-dependent cell-cycle arrest, confirming previousfindings (50–54), and to some extent cell death occurs that couldbe attributed to either autophagic cell death or apoptosis (55, 56).

Of note, as opposed to the mechanism of hNIS induction bymTOR inhibition (19), TTF1 mRNA expression remained unaf-fected after DLC treatment, confirming a differential underlyingmechanism for restoration of hNIS expression by DLCs throughelevation of intracellular Ca2þ and FOS expression. This is rein-forced by the observation that lanatoside C restores functionalhNIS expression in TPC-1 despite its TTF1 deficiency, whereas thiscell line remains negative for TTF1-dependent hNIS expressionupon mTOR inhibition (19). As opposed to DLCs, the otherselected autophagy-activating compounds listed in Supplemen-tary Table S1 were not able to reactivate hNIS expression, indi-cating that autophagy activation could be necessary but is notsufficient for thyroid cancer redifferentiation. This notion isreinforced by the crucial role of a Ca2þ-driven second pathwayculminating in FOS-dependent thyroid cancer redifferentiation aswe have demonstrated for DLCs (Fig. 8), which are the onlycompounds in Supplementary Table S1 known to directly influ-ence intracellular Ca2þ concentrations.

This is the first report to reveal the ability of DLCs to facilitatetumor redifferentiation and, hence, uncovers additional thera-peutic benefits of DLCs as anticancer treatment besides their well-established effects on proliferation and cell survival. These find-ings need confirmation in subsequent studies involving mousemodels and human cohorts. Importantly, because of the narrowtherapeutic index of DLCs, the dosage and chemical structure ofDLCs required for thyroid cancer redifferentiation in vivo inrelation to its toxicity remains an important aspect to beaddressed. In conclusion, DLCs exhibit multiple beneficial effectsas potential treatment of RAI-refractory thyroid cancer by restor-ing hNIS expression and iodide uptake capacity and by inducingautophagy and cell-cycle arrest, all elicited at least in part by

Figure 8.

Model of digitalis-like compound-dependent activation of autophagy andreactivation of hNIS expression through intracellular Ca2þ and FOS in thyroidcarcinoma. NCX, sodium-calciumexchanger; ERK, extracellular signal-regulatedkinase; TFEB, transcription factor EB; AMPK, AMP-activated protein kinase;cAMP, cyclic adenosine monophosphate; CREB, cAMP response element-binding protein; PAX8, paired box gene 8; hNIS, human sodium-iodidesymporter; hNUE, hNIS upstream enhancer.






intracellular Ca2þ accumulation. Therefore, DLC treatmentemerges as a promising adjunctive therapy for patients withRAI-refractory thyroid cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: M.G. Netea, J.W.A. Smit, R.T. Netea-Maier, T.S.PlantingaDevelopment of methodology: M.H. Tesselaar, H.G. Swarts, O.C. Boerman,J.W.A. Smit, R.T. Netea-Maier, T.S. PlantingaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M.H. Tesselaar, T. Crezee, D. Gerrits, O.C. Boerman,H.G. Stunnenberg, R.T. Netea-Maier, T.S. PlantingaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):M.H. Tesselaar, T. Crezee,H.G. Swarts,D.Gerrits, J.W.A. Smit, R.T. Netea-Maier, T.S. Plantinga

Writing, review, and/or revision of the manuscript: M.H. Tesselaar, O.C.Boerman, M.G. Netea, J.W.A. Smit, R.T. Netea-Maier, T.S. PlantingaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): T.S. PlantingaStudy supervision: J.W.A. Smit, R.T. Netea-Maier, T.S. PlantingaOther (development and interpretation of Na,K-ATPase/digitalis-like com-pound interactions assay): J.B. Koenderink

Grant SupportT.S. Plantingawas supported by aVeni grant of theNetherlandsOrganization

for Scientific Research (NWO) and by the Alpe d'HuZes fund of the DutchCancer Society (KUN2014-6728).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received July 13, 2016; revised September 26, 2016; accepted October 17,2016; published OnlineFirst November 11, 2016.

References1. Xing M.Molecular pathogenesis and mechanisms of thyroid cancer. Nat

Rev Cancer 2013;13:184–99.2. Papp S, Asa SL. When thyroid carcinoma goes bad: a morphological and

molecular analysis. Head Neck Pathol 2015;9:16–23.3. Parzych KR, Klionsky DJ. An overview of autophagy: morphology, mech-

anism, and regulation. Antioxid Redox Signal 2014;20:460–73.4. Hombach-Klonisch S, Natarajan S, Thanasupawat T, Medapati M, Pathak

A, Ghavami S, et al. Mechanisms of therapeutic resistance in cancer (stem)cells with emphasis on thyroid cancer cells. Front Endocrinol 2014;5:37.

5. Choi AMK, Ryter SW, Levine B. Autophagy in human health and disease. NEngl J Med 2013;368:1845–6.

6. Plantinga TS, Tesselaar MH, Morreau H, Corssmit EP, Willemsen BK,Kusters B, et al. Autophagy activity is associated withmembranous sodiumiodide symporter expression and clinical response to radioiodine therapyin non-medullary thyroid cancer. Autophagy 2016;12:1195–205.

7. Shaw SY, Tran K, Castoreno AB, Peloquin JM, Lassen KG, Khor B, et al.Selectivemodulation of autophagy, innate immunity, and adaptive immu-nity by small molecules. ACS Chem Biol 2013;8:2724–33.

8. Hundeshagen P, Hamacher-Brady A, Eils R, Brady NR. Concurrent detec-tion of autolysosome formation and lysosomal degradation by flowcytometry in a high-content screen for inducers of autophagy. BMC Biol2011;9:38.

9. Schweppe RE, Klopper JP, Korch C, Pugazhenthi U, Benezra M, Knauf JA,et al. Deoxyribonucleic acid profiling analysis of 40 human thyroid cancercell lines reveals cross-contamination resulting in cell line redundancy andmisidentification. J Clin Endocrinol Metab 2008;93:4331–41.

10. Hou P, Bojdani E, Xing M. Induction of thyroid gene expression andradioiodine uptake in thyroid cancer cells by targeting major signalingpathways. J Clin Endocrinol Metab 2010;95:820–8.

11. Trapnell C, Roberts A, Goff L, Pertea G, KimD, Kelley DR, et al. Differentialgene and transcript expression analysis of RNA-seq experiments withTopHat and Cufflinks. Nat Protoc 2012;7:562–78.

12. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-CepasJ, et al. STRING v10: protein-protein interaction networks, integrated overthe tree of life. Nucleic Acids Res 2015;43:D447–52.

13. Ohno M, Zannini M, Levy O, Carrasco N, di Lauro R. The paired-domaintranscription factor Pax8binds to the upstream enhancer of the rat sodium/iodide symporter gene and participates in both thyroid-specific and cyclic-AMP-dependent transcription. Mol Cell Biol 1999;19:2051–60.

14. Chun JT, Di Dato V, D'Andrea B, Zannini M, Di Lauro R. The CRE-likeelement inside the 5'-upstream region of the rat sodium/iodide symportergene interacts with diverse classes of b-Zip molecules that regulate tran-scriptional activities through strong synergy with Pax-8. Mol Endocrinol2004;18:2817–29.

15. Netea-Maier RT, Kluck V, Plantinga TS, Smit JWA. Autophagy in thyroidcancer: present knowledge and future perspectives. Front Endocrinol2015;6:22.

16. Petrulea MS, Plantinga TS, Smit JW, Georgescu CE, Netea-Maier RT. PI3K/Akt/mTOR: A promising therapeutic target for non-medullary thyroidcarcinoma. Cancer Treat Rev 2015;41:707–13.

17. Ho AL, Grewal RK, Leboeuf R, Sherman EJ, Pfister DG, Deandreis D, et al.Selumetinib-enhanced radioiodine uptake in advanced thyroid cancer. NEngl J Med 2013;368:623–32.

18. Rothenberg SM, McFadden DG, Palmer EL, Daniels GH, Wirth LJ. Redif-ferentiation of iodine-refractory BRAF V600E-mutant metastatic papillarythyroid cancer with dabrafenib. Clin Cancer Res 2015;21:1028–35.

19. Plantinga TS, Heinhuis B, Gerrits D, Netea MG, Joosten LAB, HermusARMM, et al. mTOR inhibition promotes TTF1-dependent redifferentia-tion and restores iodine uptake in thyroid carcinoma cell lines. J ClinEndocrinol Metab 2014;99:E1368–75.

20. Katz AM, Lorell BH. Regulation of cardiac contraction and relaxation.Circulation 2000;102:IV69–74.

21. Hauptman PJ, Garg R, Kelly RA. Cardiac glycosides in the nextmillennium.Prog Cardiovasc Dis 1999;41:247–54.

22. Rahimtoola SH.Digitalis therapy for patients in clinical heart failure.Circulation 2004;109:2942–6.

23. Popp T, Steinritz D, Breit A, Deppe J, Egea V, Schmidt A, et al. Wnt5a/beta-catenin signalingdrives calcium-induceddifferentiationof humanprimarykeratinocytes. J Invest Dermatol 2014;134:2183–91.

24. Collado B, SanchezMG, Diaz-Laviada I, Prieto JC, CarmenaMJ. Vasoactiveintestinal peptide (VIP) induces c-fos expression in LNCaP prostate cancercells through a mechanism that involves Ca2þ signalling. Implications inangiogenesis and neuroendocrine differentiation. Biochim Biophys Acta2005;1744:224–33.

25. Holliday J, AdamsRJ, Sejnowski TJ, SpitzerNC. Calcium-induced release ofcalcium regulates differentiation of cultured spinal neurons. Neuron1991;7:787–96.

26. Shi H, Halvorsen YD, Ellis PN, Wilkison WO, Zemel MB. Role of intra-cellular calcium in human adipocyte differentiation. Physiol Genomics2000;3:75–82.

27. Kim H, Kim T, Jeong B-C, Cho I-T, Han D, Takegahara N, et al. Tmem64modulates calcium signaling during RANKL-mediated osteoclast differen-tiation. Cell Metab 2013;17:249–60.

28. Satheesh NJ, Busselberg D. The role of intracellular calcium for thedevelopment and treatment of neuroblastoma. Cancers 2015;7:823–48.

29. Okazaki T, Mochizuki T, Tashima M, Sawada H, Uchino H. Role ofintracellular calcium ion in human promyelocytic leukemia HL-60 celldifferentiation. Cancer Res 1986;46:6059–63.

30. Papp B, Brouland J-P, Arbabian A, Gelebart P, Kovacs T, Bobe R, et al.Endoplasmic reticulum calcium pumps and cancer cell differentiation.Biomolecules 2012;2:165–86.

31. Kondratskyi A, Yassine M, Kondratska K, Skryma R, Slomianny C, Pre-varskaya N. Calcium-permeable ion channels in control of autophagy andcancer. Front Physiol 2013;4:272.


Tesselaar et al.




32. Kania E, Pajak B, Orzechowski A. Calcium homeostasis and ER stress incontrol of autophagy in cancer cells. Biomed Res Int 2015;2015:352794.

33. Prassas I, Diamandis EP. Novel therapeutic applications of cardiac glyco-sides. Nat Rev Drug Discov 2008;7:926–35.

34. CaoH, Yin Z, Chen S, Liu T, Liu J, Hu Y, et al. Systematic drug repositioningby integrating transcriptome and historical clinical data, identification ofdigoxin as a novel drug reposition candidate for high-risk myelodysplasticsyndromes [abstract]. In: Proceedings of the 57th ASHAnnualMeeting andExposition; 2015 Dec 5–8; Orlando, FL. Washington, DC: ASH; 2015.Abstract nr 4118.

35. Platz EA, Yegnasubramanian S, Liu JO, Chong CR, Shim JS, Kenfield SA,et al. A novel two-stage, transdisciplinary study identifies digoxin as apossible drug for prostate cancer treatment. Cancer Discov 2011;1:68–77.

36. Cherniavsky-Lev M, Golani O, Karlish SJD, Garty H. Ouabain-inducedinternalization and lysosomal degradation of the Naþ/Kþ-ATPase. J BiolChem 2014;289:1049–59.

37. Mikkaichi T, Suzuki T, Onogawa T, TanemotoM,Mizutamari H, OkadaM,et al. Isolation and characterization of a digoxin transporter and its rathomologue expressed in the kidney. Proc Natl Acad Sci U S A 2004;101:3569–74.

38. Gozalpour E, Wittgen HGM, van den Heuvel JJMW, Greupink R, RusselFGM, Koenderink JB. Interaction of digitalis-like compounds with p-glycoprotein. Toxicol Sci 2013;131:502–11.

39. de Lannoy IA, Silverman M. The MDR1 gene product, P-glycoprotein,mediates the transport of the cardiac glycoside, digoxin. Biochem BiophysRes Commun 1992;189:551–7.

40. Weigand KM, Laursen M, Swarts HGP, Engwerda AHJ, Prufert C, SandrockJ, et al. Na(þ),K(þ)-ATPase isoform selectivity for digitalis-like com-pounds is determined by two amino acids in the first extracellular loop.Chem Res Toxicol 2014;27:2082–92.

41. Terrier P, Sheng ZM, Schlumberger M, Tubiana M, Caillou B, Travagli JP,et al. Structure and expression of c-myc and c-fos proto-oncogenes inthyroid carcinomas. Br J Cancer 1988;57:43–7.

42. Kataki A, Sotirianakos S, Memos N, Karayiannis M,Messaris E, Leandros E,et al. P53 and C-FOS overexpression in patients with thyroid cancer: animmunohistochemical study. Neoplasma 2003;50:26–30.

43. Carrasco MA, Riveros N, Rios J, Muller M, Torres F, Pineda J, et al.Depolarization-induced slow calcium transients activate early genes inskeletal muscle cells. Am J Physiol Cell Physiol 2003;284:C1438–47.

44. Sheng M, McFadden G, Greenberg ME. Membrane depolarization andcalcium induce c-fos transcription via phosphorylation of transcriptionfactor CREB. Neuron 1990;4:571–82.

45. Sakamoto KM, Frank DA. CREB in the pathophysiology of cancer: implica-tions for targeting transcription factors for cancer therapy. Clin Cancer Res2009;15:2583–7.

46. Medina DL, Di Paola S, Peluso I, Armani A, De Stefani D, Venditti R, et al.Lysosomal calcium signalling regulates autophagy through calcineurin andTFEB. Nat Cell Biol 2015;17:288–99.

47. Hoyer-Hansen M, Bastholm L, Szyniarowski P, Campanella M, SzabadkaiG, Farkas T, et al. Control of macroautophagy by calcium, calmodulin-dependent kinase kinase-beta, and Bcl-2. Mol Cell 2007;25:193–205.

48. Wang Y, Qiu Q, Shen J-J, Li D-D, Jiang X-J, Si S-Y, et al. Cardiac glycosidesinduce autophagy in human non-small cell lung cancer cells throughregulation of dual signaling pathways. Int J Biochem Cell Biol 2012;44:1813–24.

49. Duan L, Cobb MH. Calcineurin increases glucose activation of ERK1/2by reversing negative feedback. Proc Natl Acad Sci U S A 2010;107:22314–9.

50. Kometiani P, Liu L, Askari A. Digitalis-induced signaling by Naþ/Kþ-ATPase in human breast cancer cells. Mol Pharmacol 2005;67:929–36.

51. Jiang Y, Zhang Y, Luan J, Duan H, Zhang F, Yagasaki K, et al. Effects ofbufalin on the proliferation of human lung cancer cells and its molecularmechanisms of action. Cytotechnology 2010;62:573–83.

52. Elbaz HA, Stueckle TA, Wang H-YL, O'Doherty GA, Lowry DT, Sargent LM,et al. Digitoxin and a syntheticmonosaccharide analog inhibit cell viabilityin lung cancer cells. Toxicol Appl Pharmacol 2012;258:51–60.

53. Haas M, Wang H, Tian J, Xie Z. Src-mediated inter-receptor cross-talkbetween the Naþ/Kþ-ATPase and the epidermal growth factor receptorrelays the signal from ouabain to mitogen-activated protein kinases. J BiolChem 2002;277:18694–702.

54. KawazoeN,WatabeM,MasudaY,Nakajo S,NakayaK. Tiam1 is involved inthe regulation of bufalin-induced apoptosis in human leukemia cells.Oncogene 1999;18:2413–21.

55. Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: thecalcium-apoptosis link. Nat Rev Mol Cell Biol 2003;4:552–65.

56. Pinton P,Giorgi C, Siviero R, Zecchini E, RizzutoR. Calciumand apoptosis:ER-mitochondria Ca2þ transfer in the control of apoptosis. Oncogene2008;27:6407–18.






2017;16:169-181. Published OnlineFirst November 11, 2016.Mol Cancer Ther Marika H. Tesselaar, Thomas Crezee, Herman G. Swarts, et al. Autophagy-Dependent Pathways

, FOS, and2+Redifferentiation through Intracellular CaDigitalis-like Compounds Facilitate Non-Medullary Thyroid Cancer

Updated version

10.1158/1535-7163.MCT-16-0460doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/11/11/1535-7163.MCT-16-0460.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/16/1/169.full#ref-list-1

This article cites 55 articles, 11 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/16/1/169.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/16/1/169To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-16-0460

http://mct.aacrjournals.org/content/suppl/2016/11/11/1535-7163.MCT-16-0460.DC1

http://mct.aacrjournals.org/content/16/1/169.full#ref-list-1

http://mct.aacrjournals.org/content/16/1/169.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/16/1/169


Documents

Digitalis-like Compounds Facilitate Non- Medullary Thyroid ...mct.aacrjournals.org/content/molcanther/16/1/169.full.pdf · Cancer Biology and Signal Transduction Digitalis-like Compounds