Upregulated JAG1 Enhances Cell Proliferation in ......Upregulated JAG1 Enhances Cell Proliferation...

Human Cancer Biology

Upregulated JAG1 Enhances Cell Proliferation inAdrenocortical Carcinoma

Derek P. Simon1, Thomas J. Giordano2, and Gary D. Hammer1,3

AbstractPurpose: The purpose of this study was to examine the expression andmolecular significance of JAG1, a

ligand for the Notch developmental signaling pathway, in adrenocortical carcinoma (ACC).

Experimental Design:Humanmicroarray data were analyzed for genes expressing ligands for theNotch

pathway and validated with quantitative real-time PCR (QPCR) and immunoblots of RNA and protein,

respectively. ACC cells lines were assessed for Notch pathway member expression by immunoblot, QPCR,

and immunofluorescence. Notch pathway activity was also determined using a reporter gene (luciferase)

activation. Proliferation experiments using a Jag1 knockdown strategy (Jag1KD) and an inhibitor of Notch-

dependent transcription (DNMaml) used a coculture systemwith fluorescence-activated cell-sorting (FACS)

analysis. Tumor stage and mitotic rate of human ACC samples were correlated to JAG1 expression.

Results: The Notch ligand JAG1 mRNA and protein are upregulated in ACCs. JAG1 upregulation can be

modeled in the Y1 mouse ACC cell line that expresses Jag1, Notch receptors, downstream signaling

molecules, and exhibits density-dependent Notch activation. Jag1 enhances cell proliferation through

activation of canonical Notch signaling as shown through Jag1KD and coculture experiments. Inhibition of

Notch signaling at the level of postreceptor signaling (DNMaml), results in similar inhibition of cell

proliferation. Analysis of clinical data indicates that Jag1 expression correlates with both grade and stage of

ACCs, supporting a role of JAG1-dependent Notch activation in late-stage ACCs.

Conclusions: JAG1 is theprimary upregulatedNotch ligand inACCs and enhancesACCcell proliferation

and tumor aggressiveness in a non–cell-autonomous manner through activation of Notch signaling in

adjacent cells. Clin Cancer Res; 18(9); 2452–64. �2012 AACR.

IntroductionAdrenocortical tumors (ACT) are extremely common

neoplasms, the vast majority being benign adrenocorticaladenomas (ACA) that occur in as many as 4% to 7% of thepopulation, whereas adrenocortical carcinomas (ACC) areextremely rare (�0.5–2 cases/million) accounting for 0.2%of cancer deaths annually (1, 2). ACC is typically an aggres-sive neoplasm with many patients presenting with metas-tases upon diagnosis (1). Because of difficulty of earlydetection and lackof effective treatments for advanced-stageACCs, the average survival for surgically unresectabletumors is 12 months and the overall 5-year survival rate ishistorically less than 10% (3, 4). The molecular pathogen-esis of ACC has remained elusive until recently. Dysregula-

tion of developmental signal transduction pathways isfound in an increasing number of cancers including ACCs.Specifically, the Wnt signaling pathway, a critical mediatorof adrenal development, plays an important role in theetiology of ACCs, where constitutively active, nuclearb-catenin is frequently observed (5–9). The developmentof visible adrenal tumors in mice engineered to expressconstitutively active b-catenin in the mouse adrenal cortexsupports the hypothesis that dysregulation of Wnt/b-cate-nin signaling is a vital step in adrenocortical tumorigenesis(10).

Similar to the Wnt pathway, Notch signaling is involvedin a wide range of cell fate decisions during development.Although its dysregulation is a commonmolecular event ina variety of cancers, its role in adrenal developmentand ACC is unknown (11, 12). Notch signaling involvesinteraction between a transmembrane ligand, of eitherthe Jagged (JAG1/2) or Delta-like (DLL1/3/4) family, anda transmembrane receptor (Notch1/2/3/4) generallyexpressed onadjacent cells (13, 14).UponbindingofNotchligand to receptor, the g-secretase complex cleaves theNotch receptor in 2 locations releasing the active signalingmolecule NICD (cleaved Notch intracellular domain).NICD interacts with constitutively DNA-bound CSL[CBF-1/RBPjk/Su(H)/Lag-1], recruits the essential tran-scriptional coactivator MAML (Mastermind-like), and

Authors' Affiliations: 1Cellular and Molecular Biology Training Program,2Department of Pathology, and 3Endocrine Oncology Program, Universityof Michigan Comprehensive Cancer Center, Ann Arbor, Michigan

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Gary D. Hammer, University of Michigan, 1502BSRB, 109 Zina Pitcher Pl, 1528 BSRB, Ann Arbor, MI 48109. Phone: 734-615-2421; Fax: 734-647-7950; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-11-2371

�2012 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 18(9) May 1, 20122452

on July 5, 2020. © 2012 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 16, 2012; DOI: 10.1158/1078-0432.CCR-11-2371

http://clincancerres.aacrjournals.org/

initiates transcription of Notch-dependent genes such asthe HES (hairy enhancer of split) family of transcriptionfactors.The upregulation of the Notch ligand, Jagged1 (JAG1),

in a variety of cancers implies a ligand-dependent acti-vation of the Notch signaling pathway (15–17). Indeed,the upregulation of JAG1 in breast and prostate cancerhas been implicated in metastatic disease and correlatedwith poor prognosis (18–21). Mechanistically, JAG1 isthought to enhance the metastatic potential of breastcancer through a Notch-dependent induction of epithe-lial-to-mesenchymal transition of mammary epithelialcells (22). Like all Notch ligands, JAG1 classically inter-acts with receptors on adjacent cells (non–cell-autono-mous) rather than with receptors on the cells in whichthey are expressed (cell-autonomous). However, the abil-ity of JAG1 to induce malignant transformation of RKEcells despite the absence of Notch receptors raises thepossibility that noncanonical actions of JAG1 mediatesome of its oncogenic manifestations (23). In this study,we report for the first time that JAG1 is the primaryupregulated Notch ligand in ACCs and enhances ACCcell proliferation and tumor aggressiveness in a non–cell-autonomous manner through activation of Notch signal-ing in adjacent cells.

Materials and MethodsMicroarray analysisDNA microarray analyses were carried out with Affy-

metrix U133A 2.0 Plus oligonucleotide arrays and havebeen published previously (24, 25). Probe sets for JAG1,JAG2, DLL1, DLL3, and DLL4 were presented in a heat-map with clustering delineated by tumor type; individualsamples were ordered on the basis of JAG1 expression asdetermined by probe sets 216268_s_at and 209099_x_at.Data presented as dot plots used the following probe sets:JAG1 Set #1 231183_s_at, JAG1 Set #2 209099_x_at, JAG2

Set #1 209784_s_at, and JAG2 Set #2 32137_at. Correla-tions were carried out using JAG1 probe set 216268_s_atwith KI67 212022_s_at and with TOP2A 201292_at.Similar correlations were obtained with other JAG1 probesets.

Human samplesProtein and RNA were extracted using routine protocols

from frozen adrenocortical tissues obtained via the Univer-sity of Michigan (Ann Arbor, MI) Comprehensive CancerCenter Tissue Procurement Service with InstitutionalReview Board approval. Samples for protein and RNAanalysis were randomly selected [normal adrenal (NL): n¼ 5; ACA: n¼ 5; ACC: n¼ 10]. Because of tissue availability,different pools of samples were analyzed for message andprotein.

Plasmids, short hairpin RNA, and transfectionNotch reporter (pJH23A: 4xwtCBF1Luc) and control

reporter (pJH25A: 4xmtCBF1Luc) expression vectors werea generous gift from Dr. S. Dianne Hayward (John Hop-kins University Medical School, Baltimore, MD; ref. 26).The Notch reporter contains 4 consensus CSL-bindingsites driving expression of firefly luciferase whereas thesesites are mutated in the control reporter. pGIPZ vectors(Open Biosystems) expressing short hairpin RNA(shRNA) against JAG1 and a nonspecific scrambled con-trol shRNA (scramble) were obtained from the Universityof Michigan shRNA core (http://www.med.umich.edu/vcore/vector-shRNA.htm). In addition to the shRNA,pGIPZ vectors contain a puromycin selection cassette andan IRES GFP sequence. Sequences for JAG1 shRNA are #1:50-gtcagaattgtgacataaa-30 and #2: 50-gggatttggttaatggtta-30.pdsREDII expresses dsREDII under control of the cyto-megalovirus (CMV) promoter and was obtained fromDr. Claudius Vincenz (University of Michigan). Control(MigR1) plasmid, which expresses GFP, and DNMamlplasmid, which expresses a fusion protein of GFP andamino acids 13 to 74 of Maml1 and acts as a dominant-negative, were a generous gift of Dr. Ivan Maillard (Univer-sity of Michigan; refs. 27, 28). Both control (MigR1) andDNMaml plasmids contain flanking long terminal repeat(LTR) sequences and expression is driven by an MSCVpromoter. Retroviral packagingprotein expressionplasmidspGag/Pol and pVSV were kindly provided by Dr. MichaelMalim (King’s College, London, United Kingdom).

Cell culture and generation of stable cell linesCulture of the mouse ACC cell line Y1 (29) and the

human ACC cell lines NCI-H295A (30) and RL251 (31)has been described previously (25, 32). All standard cellculture reagents were obtained from Invitrogen Life Tech-nologies. Virus-competent 293T cells, a gift from Dr. Ben-jamin Margolis (University of Michigan), were maintainedin Dulbecco’s Modified Eagle’s Media (DMEM) with 10%cosmic calf serum (CCS; Hyclone) and penicillin/strepto-mycin. In some experiments, Y1 cells were treated for 6hours with 5 mmol/L EDTA prepared in PBS. Transient

Translational RelevanceAdrenocortical carcinoma (ACC) is a rare, highly

aggressive endocrine neoplasm. Because of difficulty ofdetection and lack of effective treatments, ACC frequent-ly presents with an extremely poor prognosis. Despiterecent technological advancements in genetic profilingof ACCs, the molecular pathogenesis of ACC hasremained unclear, particularly pertaining to factorsinvolved in late-stage disease. We report here for thefirst time that JAG1, a ligand for the Notch signalingpathway, is upregulated in ACCs, enhances ACC cellproliferation in a non–cell-autonomous manner, andis positively correlated with late-stage, aggressiveACCs. JAG1 and the Notch signaling pathway may benovel targets for therapeutic intervention in late-stageACCs.

Upregulated JAG1 in Adrenocortical Carcinoma

www.aacrjournals.org Clin Cancer Res; 18(9) May 1, 2012 2453




transfections were carried out using Fugene (Roche) accord-ing to the manufacturer’s instructions and optimized at a4:1 ratio (4 mL Fugene/1 mg DNA) for Y1 cells and 2:1 ratiofor 293T cells.

For generation of scramble (GFPþ) and Jag1KD (GFPþ)stable cell lines, Y1 cells were transfected with 2 mg of pGIPZvectors expressing shRNA directed against Jag1 or a control(scramble) as described earlier, followed by 4 weeks ofpuromycin selection (2 mg/mL; Roche). Cells were thenenriched for GFP expression within the 104 to 105 rangeusing fluorescence-activated cell sorting (FACS) asdescribed later. Sorted cells were replated in 10-cm dishesand allowed to expand.

To obtain the dsREDII [wildtype Y1 (Redþ)] cell line, 10-cm dishes of Y1 cells were transiently transfected with 2 mgof pdsREDII as described. Because pdsREDII lacks a mam-malian selection cassette, cells were passaged after 2 daysand were transiently transfected an additional time. After 2days, cells were enriched for dsREDII expression within the104 to 105 range by FACS. Sorted cells were replated in 10-cm dishes and allowed to expand.

To generate the control (GFPþ) and DNMaml (GFPþ)stable cell lines, viral supernatant was generated by co-transfection of 293T cells with 2 mg each of pGag/Pol,pVSV, and either control (MigR1) or DNMaml constructs.After 2 days, medium was collected and centrifuged at5,000 � g followed by filtration through a 0.22-mmsyringe nylon filter (Fisher Scientific). Viral supernatantwas adjusted to a final volume of 10 mL with DMEM andpolybrene was added (10 U/mL; Sigma). Y1 cells werethen transducted with viral supernatant for 24 hours.Cells were passaged and transducted an additional timeunder identical conditions. Cells were then enriched forGFP expression within 104 to 105 range using FACS andwere replated into 10-cm dishes and allowed to expand.Because the GFP and dsREDII expression evidently dimin-ished overtime, cells were resorted under identical para-meters every 3 months.

FACS and analysisTrypsinized cells were pelleted at 1,000� g for 5 minutes

and resuspended in 1� PBS containing 10% CCS at aconcentration of 1 to 2million cells/mL. FACS experimentswere done by the University of Michigan Flow CytometryCore (http://www.med.umich.edu/flowcytometry/) witheither BD Biosciences FACSDiVa High-Speed Cell Sorter(3-laser: 488, 350, and633nm)orBDBiosciences FACSAriaHigh-Speed Cell Sorter (3-laser: 488, 407, and 633 nm).

Quantitative real-time PCR analysisRNA was isolated with TRIzol (Invitrogen) according the

manufacturer’s instructions, and cDNAwas generated usingiScript cDNA Synthesis kit (Bio-Rad Laboratories). Quan-titative real-time PCR (QPCR) experiments were carried outas previously described (33, 34). A comprehensive list ofhuman and mouse QPCR primers is found in Supplemen-tary Table S1. Analysis was conducted with either theefficiency-corrected DCT method or the DDCT method as

indicated (35). Expression of mRNA was normalized tob-actin.

Immunocytochemistry and immunoblotsFor a comprehensive list of primary and secondary

antibodies used for immunocytochemistry and immuno-blots, see Supplementary Table S2. For immunocyto-chemical localization: Y1 cells were plated on glass slidescoated with fibronectin (10 mg/mL; Sigma). Slides werewashed in 1� PBS, fixed in 4% paraformaldehyde (Fish-er) for 15 minutes at 4�C, and permeabilized with 0.02%Igepal CA-630 (Sigma). Slides were blocked with 2%milkin 1� PBS, and primary/secondary antibodies (Supple-mentary Table S2) were diluted in 0.2% milk in 1� PBS.For detection of native fluorescence, slides were not fixedto preserve the activity of GFP and dsRedII. Coverslipswere applied and images obtained as previously described(33, 34).

Immunoblot analysis of protein lysates from cell cultureswere conducted as previously described (33). Analysis ofsome protein lysate was conducted as described but block-ing, primary, and secondary dilutionswere done inOdysseyBlocking buffer (LICOR), secondary antibodies used wereOdyssey IRdyes (Supplementary Table S2). Immunoblotsusing protein lysate from human adrenal tumor sampleswere quantified using ImageJ software (NIH, Bethesda,MD). The btan20 (Notch1) and C651.6DbHN (Notch2)monoclonal antibodies were developed by Spyros Artava-nis-Tsakonas and were obtained from the DevelopmentalStudies Hybridoma Bank under the auspices of the NICHDand maintained by the Department of Biological Sciences(University of Iowa, Iowa City, IA).

MTS proliferation assayThe MTS cell viability assay translates cell number into a

colorimetric readout (absorbance) via metabolic break-down of tetrazolium salts (Promega Corp.). Cells wereplated in 96-well plates and assay conducted according tothe manufacturer’s instructions. Absorbance values wereobtained using a SpectraMAX190 plate reader (MolecularDevices).

Coculture experimentsStable cell lines of Y1 cells expressing dsRedII [wildtype

Y1 (Redþ)] were cocultured with either scramble (GFPþ) orJag1KD (GFPþ) cells in triplicate wells of a 6-well plate in 2ratios: 90% Redþ/10% GFPþ and 10% Redþ/90% GFPþ.The combined initial concentration for each ratio at eachtime point was 150,000 cells per well. Cells analyzed at theday 4 time point were plated 4 days in advance of analysis,day 3 time point were plated 3 days in advance of analysis,and so on. At the end of the 4-day time course, harvestedcells were analyzed by FACS as described earlier. Ten thou-sand cells were analyzed for each sample and the cellnumber for each color (Redþ or GFPþ) was determinedand data are presented as a percentage change from day 1(for schematic, see Fig. 4A). Identical conditions and anal-ysis were carried out for coculture of wild-type Y1 (Redþ)

Simon et al.

Clin Cancer Res; 18(9) May 1, 2012 Clinical Cancer Research2454




with either control (GFPþ) or DNMaml (GFPþ) except thata 50%/50% ratio was used.

Luciferase assaysY1 cells were plated in 24-well plates andwere transiently

transfected with 50 ng of pRL-TK Renilla Luciferase (Pro-mega Corp.) and 0.5 mg of either control reporter (pJH25A)or Notch reporter (pJH23A) firefly luciferase constructsdescribed earlier. Assays were conducted 24 hours aftertransfection using the Dual-Luciferase Reporter Assay (Pro-mega) according to the manufacturer’s instructions andoptimized for Y1 cells. Cells were lysed in 1� passive lysisbuffer and lysates analyzed on the Glomax Multi-detectionSystem (Promega). Expression was normalized to pRL-TKRenilla Luciferase.

StatisticsAll comparisons made used the Student t test. Statistical

analysis of microarrays has been described elsewhere(24, 36).

ResultsJAG1 is upregulated in human ACCIn an effort to better understand the molecular character-

istics of human ACCs, our group has previously conductedDNA microarray analyses using frozen human tissues—most recently with a total of 33 ACC, 22 ACA, and 10normal adrenals (24, 36). Analysis of differentiallyexpressed probe sets revealed an upregulation of the Notchligand JAG1 in ACC samples compared with normal andadenomatous tissue (Fig. 1A). The 5 JAG1 probes setsdepicted are within the top 0.8% of all rank-ordered upre-gulated probe sets represented in the microarray. The other4Notch ligands (JAG2, DLL1/3/4) were upregulated in onlya few ACCs.Quantitative analysis of 2 independent probe sets for

each JAG ligand (JAG1 and JAG2) confirmed that JAG1expression is significantly higher in ACC samples than innormal adrenals and ACA (Fig. 1B, top). JAG2 exhibits astatistically significant, albeit less dramatic, difference inexpression among samples (Fig. 1B, bottom). Furthermore,interrogating 2 other adrenal tumor microarray data setsrevealed a similar upregulation of JAG1 in ACCs (36, 37).Microarray analyses were validated with QPCR of mRNA

from human adrenal tumor samples (Fig. 1C). Correlationof QPCR and microarray data for each samples is shownin Fig. 1D. Although both JAG1 and JAG2were significantlydifferent in ACC versus ACA/normal adrenals, JAG1 isexpressed at a higher level and a greater increase in ACCthan JAG2. In addition, JAG1 QPCR expression was moretightly correlated to the microarray data (JAG1: r ¼ 0.874,JAG2: r ¼ 0.545). These data support the validity andbiologic relevance of the microarray results.Furthermore, immunoblot analysis of human adrenal

tumor samples revealed a higher expression of JAG1 proteinin the majority of ACC than in normal adrenals and ACA(Fig. 1E, top). Quantification of band intensity of 2 immu-

noblots using the same set of human samples identifiesrobust protein levels of JAG1 in most ACCs and barelydetectable quantities in ACAs and normal adrenals (Fig. 1E,bottom). Together, these data suggest that JAG1mRNA andprotein is upregulated in a majority of ACC samples and isconsistent with biologic relevance of JAG1-activated Notchsignaling contributing to adrenocortical carcinogenesis.Although JAG2 is also upregulated, it exhibits a lower levelof expression and a poorer correlation of QPCR and arraydata. Therefore, we decided to focus exclusively on JAG1,the significance of its upregulation in ACC, and its role inadrenocortical carcinogenesis.

The Y1 mouse ACC cell line exhibits active Notchsignaling

Predicated on the canonical role of JAG1 as an obligateligand for Notch activation, normal mouse adrenal (adre-nal), the mouse ACC cell line (Y1), and the 2 cell linesderived from human ACC (H295A and RL251) were sur-veyed for concurrent Jag1 abundance and expression of theNotch signaling components. Jag1 protein is highlyexpressed in both Y1 and H295A lines, recapitulating inan in vitro context the upregulation of JAG1 observed in thehuman ACC samples (Fig. 2A), with Jag1mRNA showing a43-fold increase over Jag2 andDll1/3/4 ligands being barelydetectable (Fig. 2B). This latter comparison shows that theY1 cell recapitulates the Notch ligand expression profileobserved in the human ACC microarray where the expres-sion of the other 4 Notch ligands is only modestly elevatedand is consistent with Jag1 functioning as the biologicallyrelevant Notch ligand in Y1 cells. Y1 cells also express boththe Notch1 and Notch2 receptors together with the activesignaling molecule NICD. Immunocytochemistry revealsubiquitous expression of core Notch pathway components(ligand: Jagged1, receptors: Notch1/2, target gene: Hes1) inY1 cells (Fig. 2C), suggesting that juxtaposed Y1 cells arecapable of activating canonical Notch signaling in adjacentcells and/or are capable of self-activation (i.e., ligand-medi-ated receptor activation).

The active engagement of Notch signaling in Y1 cellsprovides an appropriate model system to examine Jag1-dependent Notch activation in ACCs. Because Mg2þ isrequired for Notch receptor stability, Mg2þ depletion canbe used to induce Notch receptor cleavage and biochemi-cally release the active NICD peptide to test induction ofNotch-dependent transcription (as opposed to ligand-inde-pendent constitutively active Notch-mediated transcrip-tion) in Jag1-expressing tumor cells (38, 39). Treatment ofY1 cells for 6 hours with the chelator EDTA to deplete Mg2þ

resulted in an increase in NICD protein in EDTA-treatedcells compared with vehicle-treated cells (Fig. 2D, top).H295A human ACC cells also exhibited a similar responseto EDTA treatment (Supplementary Fig. S1). To confirmthat NICD cleavage results in productive Notch-dependenttransduction, Y1 cells were transiently transfected with aspecific Notch luciferase reporter containing 4 NICD con-sensus-binding sites or an otherwise identical reporter inwhich theNICD sites aremutated. After EDTA treatment, an






increase in the expression of the Notch luciferase reporterwas observed (EDTA-Notch vs. vehicle-Notch ¼ 2.3-foldincrease; Fig. 2D, bottom). These data indicate that canon-ical Notch signaling can be activated in Y1 cells, presumablybecause of the presence of Jag1.

Because Notch signaling is generally dependent upon thejuxtaposition of 2 adjacent cells expressing membrane-bound ligand and receptor, respectively, we hypothesizedthat activation of Notch signaling in Y1 cells is density-

dependent. When Y1 cells were plated at increasing density(10%, 25%, 50%, and 90% confluence), a density-depen-dent increase in NICD protein was observed (Fig. 2E, top).When the transcriptional activity of the Notch reporter wasevaluated by luciferase assay, an elevated activity wasobserved in the highest (90%) when compared with thelowest (10%) density (high-density Notch versus low-den-sity Notch¼ 2.3-fold; Fig. 2E, bottom). In addition, immu-noblot analysis did not detect Jag1 in conditioned medium

-2 0 2 4 6 8

2.2

2.4

2.6

2.8

3.0 NL

ACA

ACC

Base-2 log human JAG2mRNA expression

JA

G2 (

pro

be s

et

#1)

Base-1

0 lo

g e

xp

ressio

n

JAG1

NL ACA ACC

0

50

100

150

200

250

300

350

Re

lati

ve

mR

NA

ex

pre

ss

ion

(no

rma

lize

d t

o β

-ac

tin

)

Probe set #1

NL ACA ACC2.0

2.5

3.0

3.5

Base-1

0 l

og

exp

ressio

n

Probe set #1

NL ACA ACC1.5

2.0

2.5

3.0

3.5

4.0

Base-1

0 lo

g e

xp

ressio

n

Probe set #2

NL ACA ACC2.5

3.0

3.5

4.0

4.5

Base-1

0 lo

g e

xp

ressio

n

Probe set #2

NL ACA ACC2.0

2.5

3.0

3.5

4.0

Ba

se

-10

lo

g e

xp

res

sio

n

JAG1

β-Actin

JAG1

β-Actin

NL

JAG1

β-Actin

ACC

ACA

ACC

JAG1

β-Actin

190 kDa

42 kDa

190 kDa

42 kDa

190 kDa

42 kDa

190 kDa

42 kDa

Gene symbol

JAG1JAG1JAG1JAG1JAG2JAG2DLL1DLL1DLL3DLL3DLL4

Ratio to the median

1/4 1/2 1 2 4

Normal Adenoma Carcinoma

JAG2

NL ACA ACC

0

50

100

150

200

250

300

350

Re

lati

ve

mR

NA

ex

pre

ss

ion

(no

rma

lize

d t

o β

-ac

tin

)

0 2 4 6 8 102.5

3.0

3.5

4.0

4.5 NL

ACA

ACC

Base-2 log human JAG1mRNA expression

JA

G1 (

pro

be s

et

#2)

Base-1

0 lo

g e

xp

ressio

n

NL ACA ACC

0

5

10

15

20

Re

lati

ve

JA

G1

pro

tein

(no

rmalized

to

β-a

cti

n)

A

B

D

C

E

* #

###***

** ##

****####

JAG1

JAG2

r = 0.545r = 0.874

Figure 1. JAG1 is upregulated in humanACC. A, heatmap of Affymetrix U133A 2.0 Plus oligonucleotide array representingNotch ligand genes. Normal adrenal(NL), n¼ 10; ACA, n¼ 22; ACC, n¼ 33. Scale is indicated. B, dot plot of 2 JAG1 and JAG2 probe sets. Each dot indicates one tissue sample. Lines indicatemean expression levels. JAG1 #1: ACC versus NL (�, P ¼ 4 � 10�6, ACC vs. ACA; #, P ¼ 7 � 10�12), JAG1 #2: ACC versus NL (��, P ¼ 1 � 10�6,ACAvs.NL; ##,P¼2 � 10�11), JAG2#1: ACCversusNL (��,P¼6 � 10�4, ACCvs. ACA; ###,P¼4 � 10�4), JAG2#2:ACCversusNL (��,P¼5.3 � 10�3,ACC vs. ACA; ####, P ¼ 3 � 10�4). C, QPCR analysis of mRNA from randomly selected human samples (NL: n ¼ 5, ACA: n ¼ 5, ACC: n ¼ 10) forJAG1 and JAG2. Each data point represents an average of triplicate determinations. D, correlation of log-transformed JAG1 QPCR expression data(from C) with the corresponding JAG1 (probe set #2) microarray data (from B; r ¼ 0.874, P ¼ 5 � 10�7) and JAG2 QPCR data with JAG2 (probe set #1)microarray data (r ¼ 0.545, P ¼ 0.013). E, top, immunoblot analysis of 5 mg of protein lysates from randomly selected human adrenal samples (NL, n ¼ 5;ACA, n ¼ 5; ACC, n ¼ 10). Blots were probed for Jagged1 and b-actin used as a loading control. Molecular weights are indicated. Bottom, quantification ofimmunoblots. Jag1 protein was normalized to b-actin and then to NL sample #06. The average of 2 experiments is presented. Lines represent mean.

Simon et al.





from Y1 cells, which eliminates the possibility that Jag1 isacting as a secreted factor (data not shown). Taken together,these data confirm that active Notch signaling is occurringin Y1 cells.To investigate the dependence of Notch activation on

Jag1 in ACC cells, an shRNA knockdown strategy was used.Two shRNAs with 100% homology to Jag1 mRNA werefound to be sufficient to knockdown Jag1 in Y1 cells whenused in combination. Two stable cell lines were generated, aJag1-knockdown line expressing both Jag1 shRNAs [Jag1KD(GFPþ)] and a control line expressing a nonspecific shRNA[scramble (GFPþ)]. GFP is expressed concurrently with theshRNA and scramble vectors; thus, both cell lines are GFPþ.Jag1 protein was significantly decreased in the Jag1KD(GFPþ) cell line, an effect that was stable for more than 3weeks as determined by immunoblot analysis (Fig. 3A).Jag1mRNA expression was also reduced by 63%, althoughthe related ligand Jag2 showed no statistically significant

change (Fig. 3B). The concurrent suppression of the Notchtarget gene Hes1 is consistent with a Jag1-dependent acti-vation of Notch signaling in ACC cells.

Jag1 has a non–cell-autonomous effect on ACC cellproliferation

To analyze the effect of Jag1 knockdown on proliferationin Y1 cells, the MTS viability assay was used. Although nodifference in proliferation was observed between scramble(GFPþ) and Jag1KD (GFPþ) cells when plated at low-den-sity (10%confluence; Fig. 3C), Jag1KD(GFPþ) cells showeda 32% reduction in proliferation when plated at a higherconcentration (40% confluence; Fig. 3D). These data con-firm that Jag1 exerts an effect on ACC cell proliferation in adensity-dependent manner. Furthermore, immunoblotanalysis revealed a reduction in proliferating cell nuclearantigen (PCNA), a marker of proliferation, in the Jag1KD(GFPþ) cell lysates compared with scramble (GFPþ),

Jag1 Jag2 Dll1 Dll3 Dll40.000

0.0010.05

0.102.0

2.5

3.0

3.5

4.0

Rela

tive q

uan

tity

Adre

nal

Y1 H29

5A

RL25

1

Jag1

β-Actin

NICD

Notch2

Notch1

Cell density

NICD

β-Actin

NICD

β-Actin

EDTA

Vehic

le

Control Notch0

1

2

3

4

5 Vehicle

EDTA

Re

lati

ve

ex

pre

ss

ion

(fir

efl

y/R

enill

a)

Control Notch0

5

10

15

20 Low density

High density

Re

lati

ve

ex

pre

ss

ion

(fir

efl

y/R

enill

a)

Merge

Jag1

Jag1

Jag1

Merge

Merge

Hes1

Notch2

Notch1

* #

* #

*

A

C E

B D

Figure 2. The Y1 mouse ACC cell line exhibits active Notch signaling. A, immunoblot analysis of 10 mg of protein lysate from wild-type (WT) mouse adrenal(adrenal), mouse ACC cell line (Y1), and human ACC (H295, RL251) cell lines for Jag1, Notch1, Notch2, and NICD. b-Actin is used as loading control. B,efficiency-correctedDCTQPCRmethodof triplicate samples used toquantify the expression of the 5Notch ligands inY1cells (�, JAG1vs. JAG2,P¼0.00001).Representative experiment of 3 repetitions. C, immunofluorescent colocalization of Jag1 with Notch1, Notch2, and Hes1 in Y1 cells. D, top, immunoblotanalysis of 10 mg of protein lysate from Y1 cells treated with 5 mmol/L EDTA or vehicle (PBS) for 6 hours. Blots were probe for NICD and b-actin was usedas loading control. Bottom, luciferase assay of triplicate samples of Notch (4xwtCBF1Luc; pJH23A) and control (4xmtCBF1Luc; pJH25A) reporter inY1cells treatedwith 5mmol/LEDTAor vehicle for 6 hours. Luciferaseexpression is normalized topRL-TKRenillaexpression (EDTA,Notch: � vs. EDTA, controlP¼0.004, # vs. vehicle, NotchP¼0.0039). Representative experiment of 3 repetitions. E, top, immunoblot analysis ofNICD levels of 10mgof protein lysates ofY1 cells grown at 10%, 25%, 50%, and 90% confluence. b-Actin is used as loading control. Bottom, luciferase assay of triplicate samples of Notchand control reporter expression (normalized to Renilla) in Y1 cells grown at 10% (low density) and 90% (high density) confluence (high density, Notch:�, vs. high density, control P ¼ 0.00003; #, vs. low density, Notch P ¼ 0.0006). Representative experiment of 3 repetitions.






whereas cleaved caspase-3 (Clv-Csp-3), a marker of apo-ptosis, was not detected in either cell line (SupplementaryFig. S2A and data not shown). These data confirm that a lossof Jag1 protein inhibits the proliferation of Y1 cells. Pred-icated on the assumption that Jag1 acts through Notchreceptors on adjacent cells, itwould be expected to influenceproliferation in a non–cell-autonomous manner consistentwith the density dependence observed.

In light of recent data that suggest additional Notchreceptor–independent biologic functions of Jag1 and therelated Notch ligands DLL1 and DLL3 within the cells inwhich theyareexpressed(cell-autonomous; refs.23,40,41),a coculture system was designed using FACS analysis tofurther interrogate the hypothesis that Jag1 acts in a non–cell-autonomous manner in ACC cells. Jag1KD (GFPþ) orscramble (GFPþ) Y1 cell lines were grown in combinationwith wild-type Y1 cells expressing dsRedII [wild-type Y1(Redþ); Fig. 4A]. Wild-type Y1 (Redþ) cells were culturedwith scramble (GFPþ) or Jag1KD (GFPþ) in 2 differentratios: 90% Redþ/10%GFPþ or 10% Redþ/90%GFPþ (Fig.4B). The former condition (90%Redþ/10%GFPþ) assessedthe ability of Jag1KD [GFPþ; relative to the scramble(GFPþ)] cells to proliferate upon receiving Jag1 inputs fromwild-type Y1 (Redþ), whereas the latter condition (10%Redþ/90% GFPþ) interrogates the effect of decreased Jag1input to wild-type Y1 (Redþ) cells. To guarantee sufficientcellular interactions, cells were initially plated at high den-sity (70% confluence by day 2; Fig. 4B). To assure that allcells could be analyzed at the same time at the end of the 4-day time course, the same initial concentration of cells was

plated 4days before analysis for day 4, 3days before analysisfor day 3, and so on (Fig. 4A). Cell number for each FACS-sorted Redþ and GFPþ populations was determined at eachtime point, and relative proliferation of the populations ispresented as a percentage change in these cell numbers fromday 1 (Fig. 4A).

To examine the hypothesis that Jag1 functions in a non–cell-autonomousmanner, the followingmixing experimentwas conducted. In the 90% Redþ/10% GFPþ ratio, Jag1KD(GFPþ) cells were cultured with an abundance of wild-typeY1 (Redþ) cells expressing high amounts of Jag1, thusJag1KD(GFPþ) cells shouldbe able to receive an abundanceof Jag1 signaling inputs from neighboring wild-type Y1(Redþ) cells (Fig. 4B, left). Under these conditions, wild-type Y1 (Redþ) cells show no relative change in prolifera-tion whether cultured with scramble (GFPþ) and Jag1KD(GFPþ) cells. This is expected because wild-type Y1 (Redþ)cells are most likely receiving Jag1 inputs predominantlyfrom other wild-type Y1 (Redþ) cells (Fig. 4C, top; Supple-mentary Fig. S3). Importantly, no difference in the prolif-eration of scramble (GFPþ) and Jag1KD (GFPþ) cells areobserved under these coculture conditions (Fig. 4C, bot-tom; Supplementary Fig. S3). These data indicate that thescramble (GFPþ) and Jag1KD (GFPþ) cells are capable ofreceiving Jag1 inputs from wild-type Y1 (Redþ) cells and,hence, proliferate, normally supporting the hypothesis thatJag1 has a non–cell-autonomous effect (SupplementaryFig. S3).

To further test the hypothesis that Jag1 acts in a non–cell-autonomous manner on adjacent cells, wild-type Y1

0 1 2 3 4 50.0

0.5

1.0

1.5

2.0

2.5 Scramble (GFP+)

Jag1KD (GFP+)

Time (Days)

Rela

tive p

rolife

rati

on

(ab

so

rban

ce)

Jag1

β-Actin

Jag1KD

(GFP+)Scramble

(GFP+)

Jag1 Jag2 Hes1 Sf10.0

0.5

1.0

1.5

Scramble (GFP+)

Jag1KD (GFP+)

Rela

tive m

RN

A e

xp

ressio

n

(no

rmalized

to

β-a

cti

n)

* #

*

0 1 2 3 4 50.0

0.2

0.4

0.6

0.8

1.0 Scramble (GFP+ )Jag1KD (GFP+)

Time (Days)

Re

lati

ve

pro

life

rati

on

(ab

so

rba

nc

e)

10% Confluence 40% Confluence

A B

C D

Figure 3. Jag1 knockdown in Y1cells inhibits proliferation in adensity-dependent manner. A,immunoblot analysis of proteinlysates from stable cell linesexpressing shRNAs for eitherscramble or Jag1 [scramble (GFPþ)and Jag1KD (GFPþ), respectively].Blots were probed for Jag1 andb-actin was used as loadingcontrol. B,QPCRanalysis ofmRNAfrom scramble (GFPþ) and Jag1KD(GFPþ) stable cell lines analyzed bythe DDCT method and normalizedto b-actin [scramble (GFPþ) vs.Jag1KD (GFPþ): �, Jag1 P ¼0.0002; #, Hes1 P ¼ 0.001].Representative experiment of 5repetitions. Absorbance valuesobtained from MTS viability assayon cell lines [scramble (GFPþ) vs.Jag1KD (GFPþ)] plated at (C) 10%confluence at day 1 and growth to35% confluence by day 4 and (D)40% confluence at day 1 andgrowth to 85% by day 4, scramble(GFPþ) vs. Jag1KD (GFPþ) at day 4.�, P ¼ 0.0045. Each data pointrepresents an average � SD of 6determinations. Representativeexperiment of 4 repetitions.

Simon et al.





(Redþ) cells were cultured with Jag1KD (GFPþ) or scram-ble (GFPþ) cells at the ratio of 10% Redþ/90% GFPþ (Fig.4B, right). Under these conditions, wild-type Y1 (Redþ)cells receive a majority of signaling input from GFPþ

(scramble or Jag1KD) cells. Specifically, the wild-typeY1 (Redþ) cells receive numerous Jag1 inputs from scram-ble (GFPþ) cells and reduced Jag1 inputs from the Jag1KD(GFPþ) cells (right, Fig. 4B; Supplementary Fig. S3).Under these conditions, wild-type Y1 (Redþ) cells showa 23% reduction in proliferation at day 2 and a 27%reduction in proliferation at day 3 [when cocultured withJag1KD (GFPþ) cells (Fig. 4D, top]. The Jag1KD (GFPþ)cells also exhibit a maximal 35% reduction at day 4 (Fig.4D, bottom) predicated on the assumption that they are

receiving the majority of signaling inputs from neighbor-ing Jag1KD (GFPþ) cells.

In summary, wild-type Y1 (Redþ) cells proliferate lesswell when cocultured with 90% Jag1KD (GFPþ) cells,suggesting that a decrease of Jag1 inputs results in retard-ed Y1 cell growth. Jag1KD (GFPþ) cells remain competentto receive Jag1 inputs from wild-type Y1 (Redþ) cells asreflected in the increased proliferation of the Jag1KD(GFPþ) cells grown in the presence of 90% wild-typeY1 (Redþ) cells. Jag1KD (GFPþ) cells proliferated lesswell in the coculture containing 90% Jag1KD (GFPþ)cells suggesting that the decrease in Jag1 inputs resultsin retarded Jag1KD (GFPþ) cell growth. Together, thecoculture studies indicate that Jag1 enhances ACC cell

Figure 4. Jag1 enhances ACCproliferation in a non–cell-autonomous manner. A, schematicindicating experimental design.dsRedIIþ (normal Y1) cells werecocultured with GFPþ (scramble orJag1KD) cells in ratios 90% Redþ/10% GFPþ or 10% Redþ/90%GFPþ. Initial combined cell number(Redþ plus GFPþ) was 150,000 cellsand triplicate wells were plated. Thesame initial platingwas used for eachtime point and cells were plated 4days from harvest for the day 4 timepoint, 3 days from the harvest for theday 3 time point, and so on.Harvested cells were analyzed byFACS. Ten thousand sorted cellswere counted for each time point andthe number of Redþ and GFPþ weredetermined for each count. Thepercentage change in cell numberfrom day 1 was determined by theformulas indicated and based on the10,000 cells counted for each timepoint. B, immunocytochemicalimages of the 2 different cocultureconditions at day 2. C, top, thepercentage change of Redþ cellsfrom day 1 for each time point in the90% Redþ/10% GFPþ condition.Bottom, the percentage change ofGFPþ cells from day 1 for each timepoint in the 90% Redþ/10% GFPþ

condition. D, top, the percentagechange of Redþ cells from day 1 foreach time point in the 10% Redþ/90% GFPþ condition (�, P < 0.03);bottom, the percentage change ofGFPþ cells from day 1 for each timepoint in the 10% Redþ/90% GFPþ

condition. Scramble (GFPþ) versusJag1KD (GFPþ) at days 2 to 4.�,P <0.0001). Eachbar represents anaverage � SD of 3 determinations.Representative experiment of 3repetitions.

1 2 3 40

50

100

150

200 Scramble (GFP+) cultured w/ Wt Y1 (Red+)

Jag1KD (GFP+) cultured w/ Wt Y1 (Red+)

Time (Days)

: (Y

n/Y

1)

100

1 2 3 40

50

100

150

200 Wt Y1 (Red+) cultured w/

scramble (GFP+)

Wt Y1 (Red+) cultured w/

Jag1KD (GFP+)

Time (Days)

: (X

n/X

1)

100

GFP

dsR

ed

GFP+

(scramble vs. Jag1KD)

Plating Data collectection and analysis

GFP

dsR

ed

(X1)

(Y1)

Day 1

(Xn)

(Yn)

Data analyzed by the formulas (Xn/X1) ×100 and

(Yn /Y1) × 100 for each day n where X = Red+

population and Y = GFP+ population

90% Red / 10% GFP (Day 2) 10% Red / 90% GFP (Day 2)

2 conditions:90% Red/ 10% GFP10% Red/ 90% GFP

***

*

1 2 3 40

50

100

150

200 Wt Y1 (Red

+) cultured w/

scramble (GFP+)

Wt Y1 (Red+) cultured w/

Jag1KD (GFP+)

Time (Days)

% o

f D

ay 1

% o

f D

ay

1

: (X

n/X 1)

10

0

1 2 3 40

50

100

150

200

Scramble (GFP+) cultured w/ Wt Y1 (Red+)

Jag1KD (GFP+) cultured w/ Wt Y1 (Red+)

Time (Days)

% o

f D

ay 1

% o

f D

ay

1: (Y

n/Y

1)

100

*

Day n

Red+

(Wt Y1 cells)

Identical plating for 4 day time course

Red

+

(Wt Y

1 c

ells)

GF

P+

(exp

eri

men

tal cells)

A

B

C D






proliferation in a non–cell-autonomous manner. Seeexperimental model in Supplementary Fig. S3.

Inhibition of Notch-dependent transcription reducesACC cell proliferation

The non–cell-autonomous enhancement of ACC cellproliferation by Jag1 is consistent with a Notch receptor–dependent process. As such, an inhibition of Notch-depen-dent transcription should phenocopy the Jag1 knockdownin a cell-autonomous manner. Notch-dependent transcrip-tion is initiatedby a ternary complexof thebasally repressiveCSL, active signaling molecule NICD, and transcriptionalcoactivator MAML1–4 (13, 14). An engineered peptidesequence derived from Maml1, which has a dominant-negative effect on all Notch-dependent transcription bycompeting for the endogenous Maml proteins and prevent-ing their binding to NICD and CSL (27, 28) was used[DNMaml (GFPþ): expresses GFP fusion protein of aminoacids 13–74 of Maml1, control (GFPþ): expresses GFP].Stable cell lines expressing either DNMaml or the controlconstruct were generated and RNA was isolated and ana-lyzed by QPCR. In DNMaml (GFPþ) cells, the canonicalNotch target gene Hes1 and the putative target Cdkn1a arereduced by 64% and 43%, respectively (Fig. 5A). Twounrelated but highly expressed genes in Y1 cells, Ctnnb1(b-catenin) and Sf1 (steroidogenic factor 1), were unaffect-ed, consistentwitha specific inhibitionofNotch target genesin the DNMaml (GFPþ) cell line.

Because DNMaml inhibits Notch-dependent transcrip-tion, we hypothesized that DNMaml (GFPþ) cells wouldhave a reduced ability to proliferate when compared withcontrol (GFPþ) cells. Using the MTS viability assay to assessproliferation, DNMaml (GFPþ) cells plated at 40% con-fluency showed a 37% reduction in proliferation whencomparedwith control (GFPþ) cells (Fig. 5B). Furthermore,immunoblot analysis revealed a reduction in PCNA proteinlevel inDNMaml (GFPþ) cell lysates comparedwith control(GFPþ), whereas Clv-Csp-3 was undetectable and the pro-tein level of cleaved caspase-6 (Clv-Csp-6), another markerof apoptosis, was unchanged (Supplementary Fig. S2B).

Although Jag1 functions non–cell-autonomously toinfluence ACC cell proliferation, DNMaml targets down-streamNotch signaling and should have a cell-autonomouseffect on proliferation. To directly address this supposition,a similar coculture study was carried out using a 50% wild-type Y1 (Redþ)/50% control (GFPþ) or DNMaml (GFPþ)ratio. Wild-type Y1 (Redþ) cells cocultured with eithercontrol (GFPþ) or DNMaml (GFPþ) cells maintain robustproliferation (Fig. 5C, left). No statistically significant dif-ference in Hes1 expression was observed on day 4 in wild-type Y1 (Redþ) cells cultured with GFPþ (control orDNMaml) cells, indicating that DNMaml is not affectingNotch signaling in adjacent wild-type Y1 (Redþ) cells (Fig.5C, right). Conversely, DNMaml (GFPþ) cells culturedwithwild-type Y1 (Redþ) cells exhibit a 34.3% reduction inproliferation when compared with control (GFPþ) cellscultured with wild-type Y1 (Redþ) cells (left, Fig. 5D).Hes1mRNA was reduced 71.34% in DNMaml (GFPþ) versus

control (GFPþ) cells at day 4 (Fig. 5D, right). These dataindicate DNMaml is acting specifically in DNMaml (GFPþ)cells. Together with the Jag1 coculture studies, these datasupport a Jag1-dependent activation of Notch signaling inACC that can be targeted at the level of ligand (presentingcell) or receptor (receiving cell) to inhibit ACC cellproliferation.

JAG1 expression is correlated with increasedaggressiveness of ACC

JAG1 is upregulated in ACCs and acts through canonicalNotch signaling to enhance density-dependent ACC cellproliferation. To determine whether elevated JAG1 mRNAexpression levels in human ACCs correspond to an increasein cancer aggressiveness, tumor stage and grade (as assessedbymitotic rate) were examined in the 33 ACC samples usedin the microarray analysis. JAG1 mRNA expression levelscorrelatedwith advanced stage (r¼ 0.35;P¼ 0.04) andwithmitotic rate (r¼ 0.40; P¼ 0.02; Fig. 6A and 6B). Specifically,JAG1 expression was increased 1.67-fold (P¼ 0.05) in late-stage ACCs (stages III and IV) compared with early-stageACCs (stages I and II; Fig. 6A). Our previous microarray hasshown strong correlations between KI67 and topoisomer-ase 2A (TOP2A) expression, 2 markers of proliferation thatare highly upregulated inACCs, and immunohistochemicalstaining for Ki67 and Top2a protein (36). We identified apositive correlation of JAG1 expression with KI67 expres-sion (overall correlation r ¼ 0.62, P < 0.0001) and TOP2A(overall correlation r ¼ 0.69, P < 0.0001). These data areconsistent with the significant role of JAG1 in ACC cellproliferation and advanced stage of disease.

DiscussionThe Notch ligand JAG1 mRNA and protein are upregu-

lated in ACCs. JAG1 upregulation can bemodeled in the Y1mouse ACC cell line that expresses Jag1, Notch receptors,and downstream signaling molecules. Y1 cells exhibit den-sity-dependent Notch activation. Jag1 enhances cell prolif-eration through activation of canonical Notch signaling asshown through knockdown and coculture experiments.Inhibition ofNotch signaling at the level of ligand (Jag1KD)or postreceptor signaling (DNMaml) results in similarinhibition of cell proliferation. Analysis of clinical dataindicates that Jag1 expression correlates with both gradeand stage of ACCs, supporting a role of JAG1-dependentNotch activation in ACCs.

JAG1 upregulation has been observed in several cancerssuch as breast and prostate cancer where it facilitates pro-liferation and metastasis (22, 40). In breast cancer, JAG1 iscorrelated with poor prognosis and lower survival rates inwomen with late-stage, aggressive cancer (18–20). Mecha-nistically, JAG1 has been shown to induce expression ofcyclin D1 in prostate cancer (21), enhance the number ofcancer cells in S-phase (41), and facilitate proliferation inWnt1-transformed breast epithelial cells (42, 43).

Although the canonical mechanism by which Jag1mediates cellular effects in numerous systems is through

Simon et al.





1 2 3 40

50

100

150

200

250

Wt Y1 (Red+) cultured w/control (GFP+)

Wt Y1 (Red+) cultured w/DNMaml (GFP+)

Time (Days)

%o

fD

ay

1%

of

Da

y1

:(X

n/X

1)10

0

0.0

0.5

1.0

1.5

2.0

Day 4

Rela

tive m

RN

A e

xp

ressio

n(n

orm

alized

to

β-a

cti

n)

Control (GFP+) w/ DNMaml (GFP+) w/Wt Y1 (Red+)

Hes1

Wt Y1 (Red+)

*

Hes1 Cdkn1a Ctnnb1 Sf10.0

0.5

1.0

1.5Control (GFP+)

DNMaml (GFP+)

Re

lati

ve

mR

NA

ex

pre

ss

ion

(no

rma

lize

d t

o β

-ac

tin

)

* #

0 1 2 3 4 50.0

0.5

1.0

1.5

2.0

2.5 Control (GFP+)

DNMaml (GFP+)

Time (Days)

Re

lati

ve

pro

life

rati

on

(ab

so

rba

nc

e)

*

*

1 2 3 40

50

100

150

200

250 Control (GFP+) cultured w/Wt Y1 (Red+)

DNMaml (GFP+) cultured w/Wt Y1 (Red+)

Time (Days)

:(Y

n/Y

1)10

0

* *#

0.0

0.5

1.0

1.5

2.0

Wt Y1 (Red+) w/control (GFP+)

Wt Y1 (Red+) w/DNMaml (GFP+)

Hes1

Day 4

Re

lati

ve

mR

NA

ex

pre

ss

ion

(no

rma

lize

d t

o β

-ac

tin

)

Re

d+

(Wt Y

1 c

ell

s)

GF

P+

(ex

pe

rim

en

tal

ce

lls

)

50% Wild-type Y1 (Red+) / 50% control (GFP+) or DNMaml (GFP+)

A

C

D

B

Figure 5. DNMaml suppression ofNotch-dependent transcription reducesY1cell proliferation to a similar degree as Jag1Knockdown. A,QPCRofmRNA fromstable cells linesexpressingeither control (GFPþ) orDNMaml (GFPþ) constructsanalyzedusing theDDCTmethodandnormalized tob-actin [control (GFPþ) vs.DNMaml (GFPþ): Hes1 �, P ¼ 0.0001; Cdkn1a #, P ¼ 0.02]. B, absorbance values obtained from MTS viability assay on cell lines [control (GFPþ) vs.DNMaml (GFPþ) �,P<0.0001]. Eachdata point represents an average�SDof 6 determinations. Representative experiment of 4 repetitions.Coculture of 50%normal Y1 cells (Redþ) and either 50% control (GFPþ) or 50% DNMaml cells (GFPþ). Initial combined cell number (Redþ plus GFPþ) was 150,000cells and triplicate wells were plated. The same initial plating was used for each time point and cells were plated 4 days from harvest for the day 4 time point, 3days from the harvest for the day 3 time point, and so on. Harvested cells were analyzed by FACS. Ten thousand cells were counted for each time pointand the number of Redþ and GFPþ were determined for each count. The percentage change in cell number from day 1 was determined by the formulasindicated (y-axis) and based on the 10,000 cells counted for each time point. C, left, the percentage change of Redþ cells from day 1 for each timepoint; right, mRNA was harvested from Redþ cells at the day 4 time point for the control and DNMaml coculture. Hes1 expression was determined bythe DDCT method and normalized to b-actin. D, left, the percentage change of GFPþ cells from day 1 for each time point [control (GFPþ) vs. DNMaml (GFPþ)�, P < 0.0001; #, P ¼ 0.06); right, mRNA was harvested from GFPþ cells at the day 4 time point for the control and DNMaml coculture. Hes1 expression wasdetermined by the DDCT method and normalized to b-actin (�, P ¼ 0.0066). Each bar represents an average � SD of 3 determinations. Representativeexperiment of 3 repetitions.






its binding to the Notch receptors and activation ofdownstream signaling (11–14), Jag1 and the other Notchligands may also have receptor-independent roles(23, 44, 45). Overexpression of Jag1 has been shown tocell autonomously induce transformation of RKE cellsindependent of Notch receptors but dependent on intra-cellular interaction between the cytoplasmic tail of Jag1and Affadin, a cell adherens junction protein (23). Fur-thermore, Jag1 and DLL1 are able to be processed by theg-secretase complex to release intracellular signaling frag-ments (44). In this report, knockdown of Jag1 in mouseadrenocortical cancer cells using specific shRNAs resultedin a density-dependent reduction in proliferation. Cocul-ture experiments of normal Y1 cells with either Jag1KD orscramble cell lines tested whether Jag1 has a cell-auton-omous or non–cell-autonomous effect. Jag1KD cells werecompetent to proliferate provided they received sufficientJag1 signaling inputs from adjacent cells. Cells receivingdiminished Jag1 inputs from Jag1KD cells did not pro-liferate as well as cells receiving inputs from control(scramble) cells. These data indicate that Jag1 does nothave a cell-autonomous effect but instead mediates adre-nal cancer cell proliferation by binding to and activatingNotch receptors on adjacent cells. The similar cell-auton-omous reduction of growth after inhibition of Notch-dependent transcription using a dominant-negative ver-sion of the transcriptional coactivator Maml1 supports

the conclusion that Jag1 effects ACC cell proliferation in anon–cell-autonomous manner.

Of obvious interest is the molecular mechanism of JAG1upregulation inhumanACCs. It is informative thatWnt andNotch are known to synergize in a variety of developmentalsystems such as the ear where Jag1 acts to mediate some ofthe effects of downstream Wnt/b-catenin signaling on theformation of the otic placode (46). Moreover, JAG1 hasbeen shown to be a direct target of b-catenin in the epider-mis where Notch signaling is required for b-catenin–medi-ated melanoma formation (47, 48). A synergistic effectbetween Notch and Wnt on tumorigenesis is also seen inbreast and colon carcinoma where JAG1 is upregulated inboth of these cancers (15, 18, 42, 49).

Whether theNotch andWnt pathways interact in ACCs isunknown. Although repression of Notch-dependent tran-scription had no effect on b-catenin (Ctnnb1) expression inthe DNMaml experiments (Fig. 5A), it remains unknownwhether Wnt activation synergizes or activates variouscomponents of the Notch pathway. Nuclear b-catenin hasbeen observed in both benign ACAs and malignant ACCs(5–9) as well as the known ACC cell lines H295A and Y1(Kim and Hammer, unpublished observations). WhetherJAG1 is a downstream target of Wnt signaling in ACCs iscurrently unknown. In addition, a mouse model of ACC inwhich b-catenin is constitutively active has been recentlyreported (10). It would be informative to examine whether

4 8 12 168

10

12

14

16 NL

ACA

ACC

TOP2A expression (Base-2 log-transformed)

JA

G1

ex

pre

ss

ion

(B

as

e-2

lo

g-t

ran

sfo

rme

d)

6 8 10 12 148

10

12

14

16 NL

ACA

ACC

KI67 expression (Base-2 log-transformed)

JA

G1

ex

pre

ss

ion

(B

as

e-2

lo

g-t

ran

sfo

rme

d)

0 2 4 6 88

10

12

14

16

Mitotic rate (Base-2 log-transformed)

JA

G1 e

xp

ressio

n

(Base-2

lo

g-t

ran

sfo

rmed

)

r = 0.40

0 1 2 3 48

10

12

14

16

Stage

JA

G1 e

xp

ressio

n

(Base-2

lo

g-t

ran

sfo

rmed

)

r = 0.35

r = 0.62 r = 0.69

A B

C D

Figure 6. JAG1 expression ishighest in aggressive, highlyproliferating ACC. A, correlation ofJAG1 expression (base-2 log-transformed) for stage in ACCs(n ¼ 33). Nineteen stage I þ IIversus 14 stage IIIþ IV P¼ 0.0551,overall correlation r ¼ 0.35, P ¼0.04. B, correlation of JAG1expression (base-2 log-transformed) with mitotic rate(base-2 log-transformed). Overallcorrelation r ¼ 0.40, P ¼ 0.02.C, correlation of JAG1 expression(base-2 log-transformed) with KI67expression (base-2 log-transformed) across all humanadrenal samples used in themicroarray data set. Overallcorrelation r ¼ 0.62, P < 0.0001.D, correlation of JAG1 expression(base-2 log-transformed) withTOP2A expression (base-2 logtransformed) across all humanadrenal samples used in themicroarray data set. Overallcorrelation r ¼ 0.69, P < 0.0001.NL, normal adrenals.

Simon et al.





b-catenin activation has an effect on Jag1 and other Notchfactor expression in this model. Furthermore, conditionalknockout of Jag1 in Wnt/b-catenin–induced colorectaltumors results in a reduction in tumor size when comparedwith tumors in which Jag1 expression is not geneticallyaltered (50). Understanding the mechanism of JAG1 upre-gulation in ACC will be an important area of investigation.Moreover, the correlation of high JAG1 levels with high-grade and late-stage ACC in this study is provocative andsuggests a potential novel target for therapy.

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

Authors' ContributionsConception and design: D.P. Simon, G.D. HammerDevelopment of methodology: D.P. SimonAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): D.P. Simon, T.J. Giordano, G.D. HammerAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis):D.P. Simon, T.J. Giordano,G.D.Hammer

Writing, review, and/or revision of the manuscript: D.P. Simon, T.J.Giordano, G.D. HammerAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): D.P. Simon, G.D. HammerStudy supervision: G.D. Hammer

AcknowledgmentsThe authors thank Michelle Vinco for providing human ACC samples for

protein and RNA analysis, Aaron Robida and Michael Pihalja from theUniversity of Michigan Flow Cytometry Core for conducting FACS experi-ments, and Joanne Heaton andMichelle Wood for their help in preparationof the manuscript.

Grant SupportThis work was supported by NIH RO1 grant CA-134606 (to G.D.

Hammer).The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received September 13, 2011; revised February 16, 2012; accepted March9, 2012; published OnlineFirst March 16, 2012.

References1. Wajchenberg BL, Albergaria Pereira MA, Medonca BB, Latronico AC,

Campos Carneiro P, Alves VA, et al. Adrenocortical carcinoma: clinicaland laboratory observations. Cancer 2000;88:711–36.

2. Guerrieri M, De Sanctis A, Crosta F, Arnaldi G, BoscaroM, Lezoche G,et al. Adrenal incidentaloma: surgical update. J Endocrinol Invest2007;30:200–4.

3. Icard P, Goudet P, Charpenay C, Andreassian B, Carnaille B, ChapuisY, et al. Adrenocortical carcinomas: surgical trends and results of a253-patient series from the FrenchAssociation of Endocrine Surgeonsstudy group. World J Surg 2001;25:891–7.

4. Vassilopoulou-Sellin R, Schultz PN. Adrenocortical carcinoma.Clinical outcome at the end of the 20th century. Cancer 2001;92:1113–21.

5. Tissier F, Cavard C, Groussin L, Perlemoine K, Fumey G, HagnereAM, et al. Mutations of beta-catenin in adrenocortical tumors:activation of the Wnt signaling pathway is a frequent event in bothbenign and malignant adrenocortical tumors. Cancer Res 2005;65:7622–7.

6. Gaujoux S, Tissier F, Groussin L, Libe R, Ragazzon B, Launay P, et al.Wnt/beta-catenin and 30,50-cyclic adenosine 50-monophosphate/pro-tein kinase A signaling pathways alterations and somatic beta-cateningene mutations in the progression of adrenocortical tumors. J ClinEndocrinol Metab 2008;93:4135–40.

7. Gaujoux S, Grabar S, Fassnacht M, Ragazzon B, Launay P, Libe R,et al. Beta-catenin activation is associated with specific clinical andpathologic characteristics and a poor outcome in adrenocorticalcarcinoma. Clin Cancer Res 2011;17:328–36.

8. El Wakil A, Lalli E. The Wnt/beta-catenin pathway in adrenocorticaldevelopment and cancer. Mol Cell Endocrinol 2011;332:32–7.

9. Bonnet S, Gaujoux S, Launay P, Baudry C, Chokri I, Ragazzon B, et al.Wnt/beta-catenin pathway activation in adrenocortical adenomas isfrequently due to somatic CTNNB1-activating mutations, which areassociated with larger and nonsecreting tumors: a study in cortisol-secreting and -nonsecreting tumors. J Clin EndocrinolMetab 2010;96:E419–26.

10. Berthon A, Sahut-Barnola I, Lambert-Langlais S, de Joussineau C,Damon-Soubeyrand C, Louiset E, et al. Constitutive beta-cateninactivation induces adrenal hyperplasia and promotes adrenal cancerdevelopment. Hum Mol Genet 2010;19:1561–76.

11. Bolos V, Grego-Bessa J, de la Pompa JL. Notch signaling in devel-opment and cancer. Endocr Rev 2007;28:339–63.

12. Wilson A, Radtke F. Multiple functions of Notch signaling in self-renewing organs and cancer. FEBS Lett 2006;580:2860–8.

13. Bray SJ. Notch signalling: a simple pathway becomes complex. NatRev Mol Cell Biol 2006;7:678–89.

14. KopanR, IlaganMX. ThecanonicalNotch signalingpathway: unfoldingthe activation mechanism. Cell 2009;137:216–33.

15. Guilmeau S, Flandez M, Mariadason JM, Augenlicht LH. Heteroge-neity of Jagged1 expression in human and mouse intestinal tumors:implications for targeting Notch signaling. Oncogene 2009;29:992–1002.

16. Gao J, Chen C, Hong L, Wang J, Du Y, Song J, et al. Expression ofJagged1 and its association with hepatitis B virus X protein in hepa-tocellular carcinoma. Biochem Biophys Res Commun 2007;356:341–7.

17. Yeh TS,WuCW,HsuKW, LiaoWJ, YangMC, Li AF, et al. The activatedNotch1 signal pathway is associated with gastric cancer progressionthrough cyclooxygenase-2. Cancer Res 2009;69:5039–48.

18. Reedijk M, Odorcic S, Chang L, Zhang H,Miller N, McCready DR, et al.High-level coexpression of JAG1 and NOTCH1 is observed in humanbreast cancer and is associated with poor overall survival. Cancer Res2005;65:8530–7.

19. Dickson BC, Mulligan AM, Zhang H, Lockwood G, O'Malley FP, EganSE, et al. High-level JAG1 mRNA and protein predict poor outcome inbreast cancer. Mod Pathol 2007;20:685–93.

20. Reedijk M, Pinnaduwage D, Dickson BC, Mulligan AM, Zhang H, BullSB, et al. JAG1 expression is associated with a basal phenotype andrecurrence in lymph node-negative breast cancer. Breast Cancer ResTreat 2008;111:439–48.

21. Cohen B, ShimizuM, Izrailit J, NgNF, BuchmanY, Pan JG, et al. CyclinD1 is a direct target of JAG1-mediated Notch signaling in breastcancer. Breast Cancer Res Treat 2010;123:113–24.

22. LeongKG,NiessenK, Kulic I, Raouf A, EavesC, Pollet I, et al. Jagged1-mediated Notch activation induces epithelial-to-mesenchymal transi-tion through Slug-induced repression of E-cadherin. J Exp Med2007;204:2935–48.

23. Ascano JM, Beverly LJ, Capobianco AJ. The C-terminal PDZ-ligand ofJAGGED1 is essential for cellular transformation. J Biol Chem2003;278:8771–9.

24. Giordano TJ, Kuick R, Else T, Gauger PG, Vinco M, Bauersfeld J, et al.Molecular classification and prognostication of adrenocortical tumorsby transcriptome profiling. Clin Cancer Res 2009;15:668–76.

25. Barlaskar FM, Spalding AC, Heaton JH, Kuick R, Kim AC, Thomas DG,et al. Preclinical targeting of the type I insulin-like growth factorreceptor in adrenocortical carcinoma. J Clin Endocrinol Metab2009;94:204–12.






26. Hsieh JJ, Henkel T, Salmon P, Robey E, Peterson MG, Hayward SD.Truncated mammalian Notch1 activates CBF1/RBPJk-repressedgenes by a mechanism resembling that of Epstein-Barr virus EBNA2.Mol Cell Biol 1996;16:952–9.

27. Maillard I, Weng AP, Carpenter AC, Rodriguez CG, Sai H, Xu L, et al.Mastermind critically regulates Notch-mediated lymphoid cell fatedecisions. Blood 2004;104:1696–702.

28. Maillard I, Tu L, Sambandam A, Yashiro-Ohtani Y, Millholland J,Keeshan K, et al. The requirement for Notch signaling at the beta-selection checkpoint in vivo is absolute and independent of the pre-Tcell receptor. J Exp Med 2006;203:2239–45.

29. Yasumura Y, Buonassisi V, Sato G. Clonal analysis of differentiatedfunction in animal cell cultures. I. Possible correlated maintenance ofdifferentiated function and the diploid karyotype. Cancer Res 1966;26:529–35.

30. Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE,et al. Establishment and characterization of a human adrenocorticalcarcinoma cell line that expresses multiple pathways of steroid bio-synthesis. Cancer Res 1990;50:5488–96.

31. Schteingart DE, Giordano TJ, Benitez RS, Burdick MD, Starkman MN,Arenberg DA, et al. Overexpression of CXC chemokines by an adre-nocortical carcinoma: a novel clinical syndrome. J Clin EndocrinolMetab 2001;86:3968–74.

32. Yang WH, Heaton JH, Brevig H, Mukherjee S, Iniguez-Lluhi JA,Hammer GD. SUMOylation inhibits SF-1 activity by reducing CDK7-mediated serine 203 phosphorylation. Mol Cell Biol 2009;29:613–25.

33. Looyenga BD, Hammer GD. Genetic removal of Smad3 from inhibin-null mice attenuates tumor progression by uncoupling extracellularmitogenic signals from the cell cycle machinery. Mol Endocrinol2007;21:2440–57.

34. Kim AC, Reuter AL, Zubair M, Else T, Serecky K, Bingham NC, et al.Targeted disruption of beta-catenin in Sf1-expressing cells impairsdevelopment and maintenance of the adrenal cortex. Development2008;135:2593–602.

35. Bookout AL, Cummins CL, Mangelsdorf DJ, Pesola JM, Kramer MF.High-throughput real-timequantitative reverse transcriptionPCR.CurrProtoc Mol Biol 2006;Chapter 15:Unit 15.8.

36. Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL,et al. Distinct transcriptional profiles of adrenocortical tumors uncov-ered by DNA microarray analysis. Am J Pathol 2003;162:521–31.

37. West AN, Neale GA, Pounds S, Figueredo BC, Rodriguez Galindo C,Pianovski MA, et al. Gene expression profiling of childhood adreno-cortical tumors. Cancer Res 2007;67:600–8.

38. Rand MD, Grimm LM, Artavanis-Tsakonas S, Patriub V, Blacklow SC,Sklar J, et al. Calcium depletion dissociates and activates heterodi-meric notch receptors. Mol Cell Biol 2000;20:1825–35.

39. Bozkulak EC,WeinmasterG.Selective use ofADAM10andADAM17 inactivation of Notch1 signaling. Mol Cell Biol 2009;29:5679–95.

40. Santagata S, Demichelis F, Riva A, Varambally S, Hofer MD, Kutok JL,et al. JAGGED1 expression is associated with prostate cancer metas-tasis and recurrence. Cancer Res 2004;64:6854–7.

41. Zhang Y, Wang Z, Ahmed F, Banerjee S, Li Y, Sarkar FH. Down-regulation of Jagged-1 induces cell growth inhibition and S phasearrest in prostate cancer cells. Int J Cancer 2006;119:2071–7.

42. Ayyanan A, Civenni G, Ciarloni L, Morel C, Mueller N, Lefort K, et al.Increased Wnt signaling triggers oncogenic conversion of humanbreast epithelial cells by a Notch-dependent mechanism. Proc NatlAcad Sci U S A 2006;103:3799–804.

43. Shimizu M, Cohen B, Goldvasser P, Berman H, Virtanen C, Reedijk M.Plasminogen activator uPA is a direct transcriptional target of theJAG1-Notch receptor signaling pathway in breast cancer. Cancer Res2011;71:277–86.

44. LaVoie MJ, Selkoe DJ. The Notch ligands, Jagged and Delta, aresequentially processed by alpha-secretase and presenilin/gamma-secretase and release signaling fragments. J Biol Chem 2003;278:34427–37.

45. Ladi E, Nichols JT, Ge W, Miyamoto A, Yao C, Yang LT, et al. Thedivergent DSL ligand Dll3 does not activate Notch signaling but cellautonomously attenuates signaling induced by other DSL ligands. JCell Biol 2005;170:983–92.

46. Jayasena CS, Ohyama T, Segil N, Groves AK. Notch signaling aug-ments the canonical Wnt pathway to specify the size of the oticplacode. Development 2008;135:2251–61.

47. Estrach S, Ambler CA, Lo Celso C, Hozumi K, Watt FM. Jagged 1 is abeta-catenin target gene required for ectopic hair follicle formation inadult epidermis. Development 2006;133:4427–38.

48. Ambler CA, Watt FM. Expression of Notch pathway genes in mam-malian epidermis and modulation by beta-catenin. Dev Dyn 2007;236:1595–601.

49. Alves-Guerra MC, Ronchini C, Capobianco AJ. Mastermind-like 1 Is aspecific coactivator of beta-catenin transcription activation and isessential for colon carcinoma cell survival. Cancer Res 2007;67:8690–8.

50. Rodilla V, Villanueva A, Obrador-Hevia A, Robert-Moreno A, Fernan-dez-Majada V, Grilli A, et al. Jagged1 is the pathological link betweenWnt andNotchpathways in colorectal cancer. ProcNatl AcadSciUSA2009;106:6315–20.

Simon et al.





2012;18:2452-2464. Published OnlineFirst March 16, 2012.Clin Cancer Res Derek P. Simon, Thomas J. Giordano and Gary D. Hammer CarcinomaUpregulated JAG1 Enhances Cell Proliferation in Adrenocortical

Updated version

10.1158/1078-0432.CCR-11-2371doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2012/03/14/1078-0432.CCR-11-2371.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/18/9/2452.full#ref-list-1

This article cites 49 articles, 26 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/18/9/2452.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://clincancerres.aacrjournals.org/content/18/9/2452To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-11-2371

http://clincancerres.aacrjournals.org/content/suppl/2012/03/14/1078-0432.CCR-11-2371.DC1

http://clincancerres.aacrjournals.org/content/18/9/2452.full#ref-list-1

http://clincancerres.aacrjournals.org/content/18/9/2452.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/18/9/2452


Upregulated JAG1 Enhances Cell Proliferation in ......Upregulated JAG1 Enhances Cell Proliferation...

Documents

Transcript of Upregulated JAG1 Enhances Cell Proliferation in ......Upregulated JAG1 Enhances Cell Proliferation...