retinoic acid/alpha-interferon combination inhibits growth and … · therapeutics, targets, and...

Therapeutics, Targets, and Chemical Biology

Retinoic Acid/Alpha-Interferon Combination Inhibits Growthand Promotes Apoptosis in Mantle Cell Lymphoma throughAkt-Dependent Modulation of Critical Targets

Jessica Dal Col1, Katy Mastorci1, Damiana Antonia Fa�e1, Elena Muraro1, Debora Martorelli1,Giorgio Inghirami2, and Riccardo Dolcetti1

AbstractMantle cell lymphoma (MCL) is characterized by a profound deregulation of the mechanisms controlling

cell-cycle progression and survival. We herein show that the combination of 9-cis-retinoic acid (RA) and IFN-ainduces marked antiproliferative and proapoptotic effects in MCL cells through the modulation of criticaltargets. Particularly, IFN-a enhances RA-mediated G0–G1 cell accumulation by downregulating cyclin D1 andincreasing p27Kip1 and p21WAF1/Cip1 protein levels. Furthermore, RA/IFN-a combination also induces apoptosis bytriggering both caspases-8 and -9 resulting in Bax and Bak activation. In particular, RA/IFN-a treatmentdownregulates the antiapoptotic Bcl-xL and Bfl-1 proteins and upregulates the proapoptotic BH3-only Noxaprotein. Sequestration of Mcl-1 and Bfl-1 by upregulated Noxa results in the activation of Bid, and the consequentinduction of apoptosis is inhibited byNoxa silencing. Noxa upregulation is associatedwith nuclear translocation ofthe FOXO3a transcription factor as consequence of RA/IFN-a–induced Akt inhibition. Pharmacologic suppressionof Akt, but not of TORC1, increases Noxa protein levels and downregulates Bfl-1 protein supporting the conclusionthat the inhibition of the Akt pathway, the resulting FOXO3a activation and Noxa upregulation are criticalmolecular mechanisms underlying RA/IFN-a—dependent MCL cell apoptosis. These results support the potentialtherapeutic value of RA/IFN-a combination in MCL management. Cancer Res; 72(7); 1825–35. �2012 AACR.

IntroductionMantle cell lymphoma (MCL) is a distinct CD5þ B-cell non–

Hodgkin lymphoma (NHL) characterized by the t(11;14)(q13;q32) translocation that leads to overexpression of cyclin D1and subsequent cyclin D/Rb pathway deregulation (1, 2).Nevertheless, other genetic changes, such as c-myc overexpres-sion, loss of the ATM gene, low levels of the cyclin-dependentkinase (CDK) inhibitor p27Kip1, and p53 deregulation (3–5), areprobably required for MCL development. Gene expressionprofiling of MCL cells has recently shown the dysregulationof several genes/proteins involved in NF-kB, PI3K/Akt, andmTOR signaling pathways, which are all constitutively acti-vated in MCL cells (6–9). These findings suggest that aprofound alteration of pathways regulating cell survival islikely responsible for the aggressive clinical behavior of MCL,

which usually show poor response to conventional therapeu-tic regimens and a very unfavorable prognosis, even when thedisease is treated with high-dose therapy and hematopoieticstem cell transplantation (1). Therefore, new therapeuticoptions need to be explored in patients affected by theselymphomas.

Wehave recently shown that 9-cis-retinoic acid (RA) inducesmarked antiproliferative responses in MCL cells by interferingwith G1–S transition through the inhibition of ubiquitinationand proteasome-dependent degradation of p27Kip1. Retinoicacid–upregulated p27Kip1 binds to cyclin D1/CDK4 complexes,resulting in decreased CDK4 kinase activity and pRb hypopho-sphorylation. Notably, retinoic acid also inhibited the growth-promoting effect induced in primary MCL cells by microenvi-ronmental factors such as CD40 triggering and interleukin(IL)-4 (10). These findings make these compounds highlyattractive in terms of potential clinical efficacy in this setting.Because retinoic acid alone does not induce relevant apoptoticresponses in MCLs, we investigated the ability of retinoic acidto cooperate with IFNs in inducing apoptosis in MCL cells.Indeed, RA/IFN combination was previously shown to exertsynergistic antiproliferative and proapoptotic effects in differ-ent cancer cell systems (11, 12). With particular regard to theproapoptotic activity, these compounds can variably promoteboth the downregulation of antiapoptotic proteins such as Bcl-2, Bcl-xL, and Mcl-1 (12, 13), often overexpressed in B-NHLs,including MCL (14–18) and/or the induction of proapoptoticproteins, including Bax, Noxa, and XAF1 (12, 19–23).

Authors' Affiliations: 1Cancer Bio-Immunotherapy Unit, Department ofMedical Oncology, Centro di Riferimento Oncologico, IRCCS - NationalCancer Institute, Aviano (PN), Italy; and 2Department of Pathology andCeRMS, University of Torino, Torino, Italy

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Riccardo Dolcetti, Cancer Bio-ImmunotherapyUnit, C.R.O. - IRCCS, National Cancer Institute, Via F. Gallini 2, Aviano(PN) 33081, Italy. Phone: 39-0434-659660; Fax: 39-0434-659196;E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-11-2505

�2012 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 1825

on April 16, 2021. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 6, 2012; DOI: 10.1158/0008-5472.CAN-11-2505

http://cancerres.aacrjournals.org/

Intriguingly, recent studies showed a critical role of Noxa/Mcl-1 interaction in regulating MCL cell survival. In fact, bothNoxa-upregulating and BH3-mimetic drugs were shown toinduce significant apoptotic responses in MCL (24–26). Noxabelongs to the BH3-only subfamily of proteins with proa-poptotic activity (27), which may directly or indirectlypromote Bax and Bak activation. A crucial event in thisprocess is the selective interaction with and the consequentinactivation of the prosurvival members of the Bcl-2 family,which, specifically for Noxa, are represented by Mcl-1 andA1/Bfl-1 (28).

On these grounds, we have investigated the ability of RA/IFN-a treatment to induce apoptotic responses in MCL cellsand, particularly, to shift the critical balance between anti-and proapoptotic regulators in favor of apoptotic machineryactivation. Moreover, taking into account our previous find-ings indicating that the PI3K/Akt pathway is critical for MCLcell survival and that Akt, but not mTOR, inhibition inducesapoptotic responses in MCL (9), we also investigated theeffects of RA/IFN-a treatment on the inherent PI3K/Aktactivation.

Materials and MethodsPatient samples

Four patients with MCL were identified on the basis ofmorphologic, immunophenotypic, and molecular criteriaaccording to World Health Organization (WHO) lymphomaclassification (Table 1). The study was conducted in accor-dance with protocols approved by the local InstitutionalReview Board, and all patients gave their informed consent.Mononuclear cells were isolated from unicellular suspensionobtained from mechanically minced lymph nodes or spleen.Enriched MCL samples (>70% MCL) were cryopreserved in10% dimethyl sulfoxide until further study. Before use, sampleswere resuspended in RPMI-1640 medium (Lonza) containing10% fetal calf serum and antibiotics.

Cell linesMino, SP53, and Jeko-1 cell lines were generously contrib-

uted by Dr. Raymond Lai, University of Alberta and CrossCancer Institute (Edmonton, Alberta, Canada). Granta-519waspurchased from DSMZ (29). Cell lines were authenticated byfingerprinting (Power Plex 1.2, Promega) in January 2011. Cellswere cultured as previously described (10).

Proliferation assay, apoptosis, and caspase activitydetection

Cell proliferation was evaluated by 3H-thymidine uptake(10) and apoptosis by Annexin-V/propidium iodide (PI)stains and/or by active/cleaved caspase-3. Activation ofcaspases-8, -9, and -3 was evaluated using a fluorimetriccommercial kit (Immunochemistry Technologies). All flowcytometric analyses were conducted on a FC500 flowcytometer (Beckman Coulter).

Antibodies and reagentsBcl-2 antibody was from Calbiochem; Bcl-xL, A1/Bfl-1, Mcl-

1, Bax, Bid (rabbit), Puma, phospho-Akt (Ser473), phospho-S6rp (Ser235/236), phospho-GSK-3a/b (S21/9), phospho-FOXO3a (S318/321), Akt, and active caspase-3 (D175) antibo-dies were from Cell Signaling Technology; Noxa, Mcl-1 (Y37),Bcl-2A1 (EP517Y) antibodies were from Abcam; p27Kip1 andp21Waf1 from BD Transduction Laboratories; PARP (F2), cyclinD1 (DCS-6), p57Kip2, p45Skp2, and Cks1 from Santa Cruz Bio-technology; Bid (mouse) antibody from SouthernBiotech; andBak (NT) and FOXO3a antibodies fromMillipore. Vital nucleardye DRAQ5, SH5 (Akt inhibitor), and LY294002were purchasedfrom Alexis Biochemicals; rapamycin, G418, 9-cis-RA (Sigma);and IFN-a (IntronA, SP Europe) were used at 1 mmol/L and1,000 U/mL, respectively. RO 40-6055 (RARa agonist) waskindly provided by Dr. W. Bollag (Hoffmann-LaRoche, Inc.)and SR11237 (RXR agonist) was a kind gift of Dr. H. Grone-meyer (Institut de G�en�etique et de Biologie Mol�eculaire etCellulaire, Strasbourg, France).

Extracts preparation, immunoprecipitation,and immunoblotting

Whole-cell lysate extracts and immunoprecipitated sampleswere prepared as previously described (10). Proteins werefractionated using SDS-PAGE and transferred onto nitrocel-lulose membranes. Immunoblotting was conducted using theEnhanced Chemiluminescence plus Detection System(PerkinElmer).

Intracellular flow cytometryCells (106) were fixed with 2% of paraformaldehyde,

permeabilized with 500 mL of cold methanol, and incubatedwith the primary antibodies for 1 hour at room temperature.After 2 washes with PBS/0.5% bovine serum albumin, cellswere incubated for 30 minutes in ice with phycoerythrin

Table 1. MCL cases

Case no. Sex/age, y Malignant cells (%) Type Cyclin D1 p27Kip1 Sample analyzed

MCL4 F/72 95 Classical þ NA Lymph node biopsyMCL5 M/50 86 Classical þ Low Lymph node biopsyMCL6 M/64 95 Classical þ Low Lymph node biopsyMCL7 M/72 96 Classical þ Low Spleen biopsy

NOTE: The expression of cyclin D1 and p27Kip1 proteins was detected by immunohistochemistry.Abbreviation: NA, not available.

Dal Col et al.

Cancer Res; 72(7) April 1, 2012 Cancer Research1826




(PE)-anti-rabbit secondary antibody and analyzed by flowcytometry.

Bid–Mcl-1 and Bid–A1/Bfl-1 colocalization and FOXO3anuclear internalizationCells (106 per sample) were fixed, permeabilized, and labeled

as described above. Samples were acquired with the Image-Stream X (Amnis) using the INSPIRE software. For colocaliza-tion experiments, samples were labeled with Mcl-1 or A1/Bfl-1antibodies and Bid antibody and DRAQ5. Then, cells double-positive for Mcl-1 and Bid or A1/Bfl-1 and Bid were selectedand compared using an algorithm of the IDEAS software whichcalculates the specificity and the degree of the fluorescencesignals colocalization through the Similarity Bright DetailsScore (SBDS; see Supplementary Methods; ref. 30). For theanalysis of FOXO3a nuclear internalization, samples werelabeled with an antibody against FOXO3a (1:100) at 4�Covernight. Then, cells were stained with the PE-anti-rabbitsecondary antibody and DRAQ5. Using an algorithm of theIDEAS software, the similarity score (SS; ref. 31) betweenFOXO3a and DRAQ5 staining was calculated for each sample.

Noxa silencingTwo different short hairpin RNAs (shRNA) PMAIP1 (phor-

bol-12-myristate-13-acetate–induced protein 1) constructswere obtained by subcloning the double-stranded 64-meroligonucleotide containing the PMAIP1 target sequences (A:

50-AAACTGAACTTCCGGCAG-30; B: 50-TCTGATATCCAAAC-TCT-30) into the pSUPER.retro-GFP/neo vector (pSUPER;OligoEngine). Infectious supernatant from pSUPER, andpSUPER.retro-shPMAIP1 retrovirally transfected Phoenix cellswere collected after 48 hours and used for 3 cycles of infections(32). Upon infection, cells were selected with G418 (1 mg/mL)and the infection efficiency was checked by flow cytometry(97% GFP-positive cells). Different clones of infected cells werethen obtained after seeding cells in 96-well plate at an initialdensity of 25 cells per well in 200 mL of medium supplementedwith G418.

ResultsIFN-a significantly enhances the antiproliferativeactivity exerted by retinoic acid in MCL cells

The 9-cis-RA is the isomer with the strongest antiprolifera-tive activity against MCL cells (10). Treatment of SP53 cellswith 9-cis-RA (1 mmol/L) in combination with IFN-a (1,000 U/mL) for 2, 4, and 7 days resulted in a more pronouncedinhibition of MCL cell proliferation as compared with cellstreatedwith retinoic acid alone, showing an additive effect (Fig.1A). Consistently, 9-cis-RA/IFN-a combination increased thenumber of MCL cells in G0–G1 (not shown), suggesting a likelyinvolvement of regulators of G1 to S-phase transition. Immu-noblotting analysis showed that IFN-a enhances the upregula-tion of the p27Kip1 protein induced by 9-cis-RA as a result of amore pronounced inhibition of p45Skp2 and Cks1, 2 SCFSkp2

Figure 1. A, IFN-a enhances theantiproliferative activity exerted by 9-cis-RA in MCL cells. DNA synthesiswas assessed in SP53 cells by3H-thymidine incorporation after 6hours. Points, mean from triplicatewells; bars, SD. The results arerepresentative of 1 of 3 experiments.B, RA/IFN-a combinationupregulates p27Kip1 andp21WAF1/Cip1 in SP53 cells (3 days).C, RA/IFN-a–induced p21WAF1/Cip1

upregulation is not a p53-only–dependent event, being detected inMCL cell lines with wild-type (SP53,Granta-519) and mutated p53 (Mino,Jeko-1; 3 days). Mt, mutated; wt,wild-type. D, RA/IFN-a combinationdownregulates cyclin D1 expressionin SP53 and Mino cells (5 days).

1

10

100

1,000

10,000

100,000

1,000,000

7 d4 d2 d

Th

ym

idin

e i

nc

orp

ora

tio

n (

cp

m)

Untreated

IFN-α9-cis-RA

IFN-α + 9-cis-RA

IFN-αp21Cip1

- + - +- - + +

SP53Jeko-1- - + +- + - +

mtp53 wtp53

- - + +- + - +

Mino

mtp53

- - + +- + - +

wtp53

Granta-519

Tubulin

p27Kip1

(short exposure)

p21Cip1

p27Kip1

- + - + IFN-α- - + +

SP53

p45Skp2

Cks1

p57Kip2

Tubulin

A B

C

SP53

DMino

- - + + - - + +SP53

- + - + - + - +9-cis-RA

9-cis-RA

9-cis-RA

IFN-α

Tubulin

Cyclin D1

Effects of RA/a-IFN in Mantle Cell Lymphoma

www.aacrjournals.org Cancer Res; 72(7) April 1, 2012 1827




complex components that are required for proteasome-depen-dent p27Kip1 degradation (Fig. 1B). Furthermore, 9-cis-RA/IFN-a combination also induces p21WAF1/Cip1 upregulation, where-as the levels of p57Kip2 were unaffected (Fig. 1B). Notably,p21WAF1/Cip1 upregulation was detected in MCL cell lines witheitherwild-type p53 (Granta-519, SP53) ormutated p53 (Jeko-1,Mino), excluding a p53-only–dependent effect (Fig. 1C). Quan-titative real-time PCR experiments showed no or low effects onthemRNA levels of the factors studied (Supplementary Fig. S1),consistent with a mainly posttranslational modulationinduced by the treatment. More interestingly, after 5 days of9-cis-RA/IFN-a treatment, SP53 and Mino cells showed amarked downregulation of cyclin D1 protein levels, an effectthat was not observed in cells exposed to each drug alone (Fig.1D). These findings indicate that IFN-a enhances the anti-proliferative activity exerted by retinoic acid in MCL cells bydecreasing the protein levels of cyclin D1 and further upregu-lating p27Kip1 and p21WAF1/Cip1.

9-cis-RA sensitizes MCL cells to the proapoptotic effectof IFN-a through both RARa and RXRs

We previously showed that retinoic acid alone does notinduce relevant apoptotic responses in MCL (10). Given theability of IFN-a to cooperate with retinoic acid in inhibitingMCL cell growth, we also explored the possible induction ofproapoptotic effects. To this end, SP53 and Mino cells weretreatedwith each drug alone or in combination for 3 and 5days,and apoptosis was evaluated using Annexin-V/PI staining. RA/IFN-a combination induced more pronounced apoptoticeffects in both MCL cell lines than single treatments (Fig.2A). In particular, sequential treatment experiments indicatedthat a 24-hour pretreatment with 9-cis-RA sensitizes MCL cellsto the proapoptotic effect of IFN-a, whereas the reverse

induced only modest effects (Fig. 2B). Considering that thepleiotropic effects of retinoids aremainlymediated by 2 classesof nuclear receptors, the RARs andRXRs (33, 34), and that 9-cis-RA is a pan-RAR and -RXR agonist, the relative contribution ofdistinct retinoic receptors was assessed using synthetic selec-tive agonists for RARa (RO 40-6055) and RXR (SR11237). Asshown in Fig. 2C, both RARa and RXR agonists sensitize MCLcells to IFN-a—induced apoptosis, although the relative con-tribution of RARa is markedly higher. These findings areclinically relevant, as RARa agonists are associated with lesspronounced side effects than the pan-RAR/RXR agonist 9-cis-RA when used in vivo.

9-cis-RA/IFN-a–dependent MCL cell apoptosis involvesmultiple caspase activation andmodulation of anti- andproapoptotic proteins

The involvement of initiator and effectors caspases in9-cis-RA/IFN-a–induced apoptosis was investigated in SP53and Mino cells using specific fluorimetric caspase assays.Time course experiments showed that both caspases-8 and-9 are activated, almost simultaneously, after 36 hours of9-cis-RA/IFN-a treatment; nevertheless, at later time points,their activity differs (Fig. 3A). In fact, caspase-8 activationincreases progressively, whereas the caspase-9 activity, afterthe initial triggering, increases more slowly, showing adownmodulation at 60 hours and a subsequent increaseat 72 hours (Fig. 3A). Given the involvement of the mito-chondrial pathway, the role of Bak and Bax proteins wasanalyzed by flow cytometry using antibodies specific for theN-terminal domains of these proteins that are exposed onlyupon Bak and Bax activation. As shown in Fig. 3B, theconformational changes of Bak and Bax were detectableonly in cells exposed to 9-cis-RA/IFN-a (3 and 5 days)

An

nexin

-V

10.4%6.9% 4.5% 6.1%8.4% 15.0%12.0% 6.4%

5.5% 2.5% 8.2% 4.8% 13.5% 6.4% 26.3% 13.8%

SP53

Untreated + IFN-αα + IFN-αIFN-α

– IFN-α + IFN-α

IFN-α IFN-α α (24 h)

+RA

RA + IFN-α RA(24 h)

+ IFN-αRA

45

40

35

30

25

20

15

10

5

0

35

30

25

20

15

10

5

0

IFN-α9-cis-RA9-cis-RA

A

3.2% 3.3% 6.6% 6.7% 6.0% 3.6% 18.0% 23.3%

4.0% 3.7% 6.2% 6.2% 6.3% 6.8% 11.9% 10.2%

Mino

Untreated 9-cis-RA

9-cis-RA

9-cis-RA RO 40-6055 RO 40-6055

+ SR11237

SR11237

5 d

3 d

Propidium iodide

SP53

B

Inc

rem

en

to

fap

op

tos

is(%

)

SP53

C

Inc

rem

en

to

fap

op

tos

is(%

)

Figure 2. A, proapoptotic effectsinduced by RA/IFN-a in SP53 andMino cells. Data are representativeof 1 of 3 independent experiments.B, 9-cis-RA sensitizesMCL cells tothe apoptosis triggered by IFN-a.SP53 cells were sequentiallytreated with 9-cis-RA (or IFN-a) for24 hours and then IFN-a (or 9-cis-RA) was added. Apoptosis wasevaluated after 3 days. C, RA/IFN-a–dependent apoptosis involvesboth RARa and RXR. SP53 cellswere cultured with 1 mmol/L 9-cis-RA, 1 mmol/L RO 40-6055 (RARaagonist), or 1 mmol/L SR11237(RXR agonist) in the absence orpresence of IFN-a for 3 days.Results in B and C are reported aspercentage of increment relative tothe control (untreated sample).Bars, mean from 3 independentexperiments; error bars, SD.

Dal Col et al.





concomitantly with the presence of cleaved caspase-3, asan indicator of ongoing apoptosis.Given that the interactions between pro- and antiapoptotic

members of the Bcl-2 superfamily are essential for mitochon-drial integrity, the expression levels of several proteins of theBcl-2 and BH3-only families were analyzed by immunoblottingin SP53 cells treatedwith 9-cis-RA, IFN-a, or their combination.While no change in Bcl-2 and Mcl-1 expression levels wasobserved in the different conditions, a marked downregulationof the Bcl-xL and A1/Bfl-1 antiapoptotic proteins was inducedby 9-cis-RA/IFN-a (Fig. 3C). Although this treatment inducedconformational changes indicating Bax and Bak activation, theexpression levels of these proteins in untreated and treatedcells were comparable. Analysis of BH3-only proteins disclosedamarked increase ofNoxa protein levels, especially in 9-cis-RA/IFN-a–treated cells, whereas the expression of Puma did notappreciably change. The levels of the full-length Bid proteindecreased significantly as likely consequence of its activationby caspase-dependent cleavage (Fig. 3C). Notably, 9-cis-RA/

IFN-a treatment was also able to significantly upregulate Noxaconcomitantly with enhanced caspase-3 activation in all 4primary MCL cultures investigated (Fig. 3D). This effect wasapparently specific for lymphoma cells, as 9-cis-RA/IFN-a didnot upregulate Noxa, nor exerted any proapoptotic activity innormal B lymphocytes obtained from 2 different donors (notshown). Taken together, these results indicate that 9-cis-RA/IFN-a combination triggers both mitochondrial/intrinsic anddeath receptor/extrinsic apoptotic pathways andpromotes theshift of the critical balance between anti- and proapoptoticproteins in favor of apoptotic machinery activation.

Noxa is a critical mediator of 9-cis-RA/IFN-a–inducedapoptosis

The BH3-only protein Noxa was shown to be one of thecritical players of the apoptotic responses induced inMCL cellsby different drugs, such as bortezomib and theMDM2 inhibitorNutlin-3 (24, 25). Interestingly, immunoblotting experimentsshowed that Noxa upregulation was induced by 9-cis-RA and at

Figure 3. A, RA/IFN-a–inducedapoptosis is caspase-dependent.Mino cells were cultured in theabsence or presence of 9-cis-RA/IFN-a, and the activation ofcaspases-8, -9, and -3 wasevaluated at the indicated timepoints. B, RA/IFN-a treatmentinduces Bak and Bax activation.Mino and SP53 cells were treated ornot with 9-cis-RA/IFN-a for 3 and5 days and stained with antibodiesspecific for the N-terminus domain(NT) of Bak or Bax and for activecaspase-3. Data reported inA and B are representative of1 of 3 independent experiments.C, RA/IFN-a treatment modulatesBcl-2 family members and BH3-onlyprotein expression. Total cell lysates(100mg)were obtained after 3 days oftreatment. D, RA/IFN-a combinationinduces Noxa upregulationassociated with caspase-3activation in primary MCL cultures.Purifiedprimary lymphomacells from4 different patients with MCL weretreated with 9-cis-RA/IFN-a for 48hours. Total cell lysates (30 mg) wereanalyzed by immunoblotting forNoxa and cleaved caspase-3. Theextent of RA/IFN-a–induced Noxaupregulation is indicated in arbitraryunits, assigning to each untreatedsample the value of 1.00.

A

0.3% 2.5% 5.9% 32.4% 35.1% 78.4%

0.5% 2.3% 5.6% 13.6% 6.2% 55.9%

43.6%19.6%30.9%7.9%5.9%1.2%

12 h 24 h 36 h 48 h 60 h 72 h

Acti

ve

casp

ase-8

Acti

ve

casp

ase-9

Ac

tive

casp

ase-3

B

Ba

k-N

TB

ax

-NT

Cle

aved

casp

ase-3

5 d3 d3 d 5 d

Mino SP53

Noxa

Cleaved caspase-3

Tubulin

D

- + - +

MCL6 MCL7

1.00 1.37 1.00 1.45 1.00 4.22 1.00 2.79

- + - +

MCL5MCL4

Bid

Bcl-xL

Bax

A1/Bfl-1

Puma

Mcl-1

Bcl-2

Bak

Noxa

- + - + RA- - + +IFN-αα

RA+IFN-α

SP53C

Tubulin






higher levels by 9-cis-RA/IFN-a cotreatment. These effects areprobably mediated by transcriptional regulation, as shown bymicroarray-based expression profiling carried out in SP53 cells(not shown). To assess the role of Noxa in 9-cis-RA/IFN-a–dependent apoptosis, we knocked down Noxa expressionby retrovirally infecting Mino cells with a pSUPER.retro-GFP/neo vector (pSUPER) carrying 2 different shRNAs, correspond-ing to nucleotides 106 to 124 and 132 to 148 of PMAIP1

consensus coding sequence and GFP (pSUPER.retro-shPMAIP1). Silencing of Noxa prevents the 9-cis-RA/IFN-a–induced upregulation of the protein and consequentlyreduces the extent of treatment-dependent apoptosis in Minocells assessed by cleaved caspase-3 and PARP (Fig. 4A). Con-sidering the ability of Noxa to specifically bind and conse-quently inactivate the antiapoptotic Mcl-1 and A1/Bfl-1 pro-teins, the interactions between Noxa and these 2 Bcl-2 family

Bid

IgG

Noxa

Mcl-1

C

- - + + - - + + - + - + - + - +

RAIFN-αα

RA+IFN-α

Noxa

PARP

Cleaved caspase-3

Tubulin

pSUPERshPMAIP1

CL.a

shPMAIP1

CL.bpSUPER

shPMAIP1

CL.c

- + + - + + - + +- + - - + - - + -- - + - - + - - +

- + - - + -- - + - - +

- + + - + + IFN-α 1,000 U/mL

9-cis-RA 1 μμmol/L

9-cis-RA 0.1 μμmol/L

A

A1/Bfl-1

IgG

Noxa

Bound UnboundBound Unbound

IP: A1/Bfl-1

- + - + IgG

Mcl-1

Noxa

IP: Mcl-1

- + - +

BBound Unbound

IP: Mcl-1

3838

235

2614

711

SBDS2.49 ± 0.42

High Colocalizatiom

31 420

3

1

2

4

0

72.0%

Untreated Untreated

No

rma

lize

dfr

eq

ue

nc

y

Bid–Mcl-1 Bright Detail Similarity Score

Bid–Mcl-1 Bright Detail Similarity Score

High Colocalization

1 2 30 4

0

1

5

4

2

3

19.2%

9-cis-RA + IFN-α 9-cis-RA + IFN-α

No

rma

lize

dfr

eq

ue

nc

y

D

SBDS1.92 ± 0.37

High Colocalization

1 3 42

0

3

1

2

66.4%

High Colocalization

0 1 432

2

0

1

3

4

17.2%

No

rma

lize

dfr

eq

ue

nc

y

Bid–Bfl-1 Bright Detail Similarity Score

Bid–Bfl-1 Bright Detail Similarity Score

No

rma

lize

dfr

eq

ue

nc

y

E

4911

1562

1305

647

SBDS2.13 ± 0.42

SBDS2.45 ± 0.48

504

1586

2493

9736

8434

8872

11526

14834

Figure 4. A,Noxa knockdown reducesRA/IFN-a–dependent apoptosis.Mino cells infectedwith empty vector (pSUPER) and 2different clones (CL.a andCL.b)of cells infectedwith vector containing shPMAIP1 sequenceAand1clone (CL.c) with the sequenceB (seeMaterials andMethods)were treatedwith 9-cis-RA/IFN-a for 3 days. Total cell lysates (50 mg) were analyzed by immunoblotting for Noxa, cleaved caspase-3, and PARP. B, RA/IFN-a combination promotes theformation of Noxa–Mcl-1 and Noxa–A1/Bfl-1 complexes. SP53 cells were cultured in absence or presence of 9-cis-RA/IFN-a for 3 days. Mcl-1 (left) and A1/Bfl-1 (right) were immunoprecipitated from 500 mg of total proteins. Immunoprecipitated (IP; bound) and nonimmunoprecipitated (unbound) fractions wereanalyzed by immunoblotting for Noxa, Mcl-1, and A1/Bfl-1 proteins. C, RA/IFN-a–inducedNoxa favors Bid displacement fromBid–Mcl-1 complexes in SP53cells (3 days). Mcl-1was immunoprecipitated from 500 mg of total proteins followed by immunoblottingwith antibodies against Noxa, full-length Bid, andMcl-1. Data depicted inBandCare representative of 3 independent experiments. D,RA/IFN-a treatment inhibitsBid–Mcl-1 andBid–Bfl-1 interactions in vivo.Minocells were cultured in the absence (untreated) or presence of 9-cis-RA/IFN-a for 3 days, then 106 cells per samplewere labeledwith primary antibodies againstBid (1:50) and Mcl-1 (1:100) or (E) A1/Bfl-1 (1:75). The vital nuclear dye DRAQ5 was added to each sample. A total of 20 � 103 cells were acquired with theImageStream X and analyzed with a specific algorithm for the SBDS calculation. Data are representative of 1 of 3 independent experiments.

Dal Col et al.





members were investigated. Most of 9-cis-RA/IFN-a–inducedNoxa co-immunoprecipitated with Mcl-1, the remainingamount being associated to A1/Bfl-1 (Fig. 4B). Furthermore,the sequestration of Mcl-1 by upregulated Noxa results in thedisplacement of the full-length Bid protein from Bid–Mcl-1complexes (Fig. 4C), allowing thus the consequent Bid activa-tion through enzyme cleavage. The resulting truncated Bidmay thus directly contribute to the activation of the Bak andBax apoptotic effectors. Taking advantage from multispectralimaging flow cytometry, we analyzed the colocalizationbetween Bid and Mcl-1 and A1/Bfl-1 also in vivo. To this end,we set up a protocol in which the cells were stained withspecific antibodies to Mcl-1 or A1/Bfl-1 and Bid proteins andthen the Bid–Mcl-1 and Bid–A1/Bfl-1 colocalization was ana-lyzed only in double-positive live cells. As shown in Fig. 4D andE, the SBDS detected in untreated samples was 2.48 � 0.42 forBid-Mcl-1 and 2.45 � 0.49 for Bid-A1/Bfl-1, and in both cases,the score significantly decreased when the cells were treatedfor 3 days with RA/IFN-a. Moreover, in treated samples, thepercentage of cells showing a significant colocalization of the 2proteins (with SBDS� 2.25) was reduced from 72% to 19.2% forBid–Mcl-1 and from 66.4% to 17.2% for Bid–A1/Bfl-1. Theseresults indicate that the treatment induces the displacement ofBid from Bid–Mcl-1 and Bid–A1/Bfl-1 complexes throughNoxa upregulation and this event precedes and promotes theapoptotic process.

9-cis-RA/IFN-a–dependent MCL cell apoptosis ismediated by inhibition of the PI3K/Akt pathwayWeandothers recently showed that the PI3K/Akt pathway is

constitutively activated in MCL, resulting in a strong deregu-lation of both cell proliferation and survival (8, 9). Particularly,pharmacologic inhibition of phosphoinositide 3-kinase (PI3K)or Akt induced significant levels of apoptosis in both estab-lished cell lines and primaryMCL cultures, whereas the TORC1inhibitor rapamycin induced no or only modest effects on cellsurvival (9). On these grounds, we investigated the ability of 9-cis-RA/IFN-a to interfere with the constitutive activation ofthese kinases. In particular, SP53 cells were treated for 3 dayswith 9-cis-RA, IFN-a, or their combination and analyzed for thepresence of the phosphorylated form of Akt and of its sub-strates GSK-3b and FOXO3a. In addition, TORC1 activationwas investigated by analyzing the phosphorylation status ofone of its main substrates, the S6 ribosomal protein. Interest-ingly, 9-cis-RA inhibited Akt andmTOR activation, as shown bythe downregulation of phospho-(S473)-Akt, of its substratesphospho-(Ser21)-GSK-3b and phospho-(Ser318/321)-FOXO3a,and of phospho-(S235/236)-S6RP, and more importantly, IFN-a significantly enhanced the inhibitory effects exerted byretinoic acid on these kinases (Fig. 5A). In particular, 9-cis-RA/IFN-a–induced Akt inhibition was associated with Noxaupregulation and the decrease of A1/Bfl-1 protein levels (Fig.5B). Furthermore, the PI3k/Akt inhibitor LY294002 induced amarked upregulation of Noxa and a complete depletion of A1/Bfl-1, whereas rapamycin did not affect the levels of theseproteins (Fig. 5C), consistent with our previous demonstrationthat rapamycin is unable to induce apoptosis in MCL cells (9).Notably, Noxa upregulation induced by the treatment was

observed in MCL cell lines carrying either wild-type (SP53) ormutant p53 (Jeko-1; Fig. 5B), supporting the hypothesis that adifferent transcription factor is involved in this phenomenon.Interestingly, FOXO3a activates the transcription of severalgenes, including PMAIP1, which encodes for the Noxa protein(35). FOXO3a transcriptional activity is regulated by the con-trol of its intracellular localization through the phosphoryla-tion/dephosphorylation of different serine/threonine residues.In particular, Akt-dependent phosphorylation on Thr32, Ser318/321, and Ser253 abolishes its nuclear translocation. Giventhe ability of 9-cis-RA/IFN-a to inhibit Akt-dependent FOXO3aphosphorylation, using multispectral imaging flow cytometry,we evaluated FOXO3a intracellular localization after 48 hoursof exposure to 9-cis-RA/IFN-a, SH5 (10 mmol/L), or rapamycin(0.1 mmol/L). As shown in Fig. 5D, FOXO3a protein is clearlyretained in the cytoplasm of untreated cells, whereas in 9-cis-RA/IFN-a- and SH5-treated cells, the protein is also detectablewithin the nucleus in 58.7% and 64.1% of cells, respectively. Incontrast, rapamycin did not affect FOXO3a intracellular local-ization, confirming that this transcription factor is Akt- but notTORC1-dependent. Notably, the analyses were conductedexcluding apoptotic cells, given that FOXO3a nuclear internal-ization and consequent Noxa upregulation are 2 events occur-ring in the first steps of the apoptotic process. These results areconsistent with a role of FOXO3a as a molecular mediator ofthe 9-cis-RA/IFN-a–induced Noxa upregulation. Moreover,these findings indicate that the 9-cis-RA/IFN-a combinationinduces MCL cell apoptosis through inhibition of the inherentAkt activation, and the consequent increment of FOXO3aactivity and, in turn, of Noxa expression (Fig. 6A and B).

DiscussionAs our understanding of the biology of MCL advances, novel

agents rationally designed to target the key pathogenicmechanisms of MCL, such as cyclin D1, cyclin/CDK inhibitors,proapoptotic proteins, mTOR, and proteasome, continue toemerge. Previously, we showed that the constitutive PI3K/Akt/mTOR activation contributes to the stability of cyclin D1 andp27Kip1 in MCL cells (9, 36), suggesting that this signalingpathway may be a crucial therapeutic target. While the ther-apeutic potential of TORC1 inhibitors is being extensivelystudied in patients with relapsed or refractory MCL (37),specific inhibitors of the upstream kinase Akt are still underevaluation in phase I clinical trials (38). Nevertheless, the Aktkinase may constitute a more effective target in MCL thanTORC1, as Akt inhibition not only reduces proliferation butalso induces significant apoptotic responses (9).

Herein, we show that RA/IFN-a cotreatment has significanteffects on both proliferation and cell survival by affecting keymolecular targets of MCL, such as cyclin D1, p27Kip1, Akt, andNoxa (Fig. 6A and B). In particular, IFN-a enhances the anti-proliferative activity exerted by 9-cis-RA by inducing a down-regulation of cyclin D1, which is strongly overexpressed inmostMCLs. Nevertheless, the observation that cyclin D1 overexpres-sion alone is not sufficient for MCL development and that itsdownregulation has only limited effects on MCL cell prolifer-ation and survival (39) indicates that additional targets should






be affected to obtain clinically relevant therapeutic efficacy.Intriguingly, the antiproliferative activity of RA/IFN-a involvesalso the increased expression of the p27Kip1 and p21WAF1/Cip1

cell-cycle inhibitors as a consequence of enhanced proteinstability. This is particularly relevant for the p27Kip1 protein,which shows an abnormally short half-life in most of MCLs(5). Furthermore, p21WAF1/Cip1 upregulation is induced irre-spective of the p53 mutational status of the cells, thus exclud-ing a p53-only–dependent effect and suggesting that this drugcombination could be efficient also in cases showing deregu-lations in this critical pathway. This is particularly intriguingin the light of the observation that about 25% of MCLs showsa deregulated p53 (40, 41), a characteristic that could promotethe resistance to novel drugs targeting the HDM2/p53pathway, such as the MDM2 antagonist Nutlin-3 (25) andMI-63 (42).

More relevant in a therapeutic perspective is the demon-stration that, unlike retinoic acid alone (10), the RA/IFN-acombination induces significant levels of apoptosis in bothestablished cell lines and primary MCL cultures. The exposureofMCL cells to 9-cis-RA for 24 hours and the following additionof IFN-a indicate this sequential treatment as the most effec-tive combination, providing thus the rationale for the design ofappropriate treatment schedules. Moreover, we also show that9-cis-RA ability to sensitize MCL cells to IFN-a–dependentproapoptotic effect involves both RARa and RXRs. Thesefindings are of potential clinical relevance, as RARa agonistsare associated with less pronounced side effects than the pan-RAR/RXR agonist 9-cis-RA when used in vivo. RA/IFN-a com-bination triggers both the death receptor/extrinsic and themitochondrial/intrinsic apoptotic pathways and promotes theactivation of the proapoptotic effectors Bak and Bax.

C LY Rapa

SP53

p-Akt

Noxa

A1/Bfl-1

C

Tubulin

Akt

B

Noxa

A1/Bfl-1

- + - + - - + +

SP53 Jeko-1

p-Akt

Akt

Tubulin

p-FOXO3a

- + - + - - + +

p-Akt

Akt

p-GSK-3

p-S6rp

RAIFN-αα

RAIFN-α

S6rp

GSK-3

A

Tubulin

p-FOXO3a

- - + +- + - +

SP53

5424

1462

6463

SS –0.58 ± 0.69

DRAQ5

FOXO3a

Merge

DRAQ5

FOXO3a

Merge

DRAQ5

FOXO3a

Merge

DRAQ5

FOXO3a

Merge

High similarity

-1 0 2 4-3 -2 310

2

4

3

1

17.5%

Untreated

FOXO3a/DRAQ5_similarity score

No

rmalized

fre

qu

en

cy

981

1251

946

SS +0.16 ± 0.82

35

18

53

SS +0.27 ± 0.30

2570

791

751

SS –0.78 ± 0.37

High similarity

-1 0 2 4-3 -2 310

1

2

3

58.7% 64.1% 10.8%

9-cis-RA + IFN-α

High similarity

-1 0 2 4-3 -2 31

3

0

2

1 High similarity

-1 0 2 4-3 -2 310

1

2

4

3

SH5 (10 μμmol/L) Rapamycin (0.1 μμmol/L)

FOXO3a/DRAQ5_similarity score

No

rmalized

fre

qu

en

cy

D

Figure 5. A and B, RA/IFN-acombination inhibits the inherentPI3K/Akt pathway activation inSP53 and Jeko-1 cells (72 hours).Total cell lysates (100 mg) weresubjected to immunoblotting usingphosphospecific antibodies. C,inhibition of Akt, but not of TORC1,is associated with Noxaupregulation and A1/Bfl-1depletion. SP53 cells wereuntreated (C) or treated with50 mmol/L LY294002 (LY) or0.1 mmol/L rapamycin (Rapa) for48 hours and total cell lysates wereanalyzed for the expression ofphospho-Akt, Noxa, and A1/Bfl-1proteins. D, RA/IFN-a treatmentpromotes FOXO3a nuclearlocalization. Mino cells wereuntreated or treated with 9-cis-RA/IFN-a, SH5 (10 mmol/L), orrapamycin (0.1mmol/L) for 48 hoursand labeled with antibody againstFOXO3a (1:100) and the vitalnuclear dye DRAQ5. Cells (20 �103) were acquired withImageStream X and analyzed withthe IDEAS software. FOXO3anuclear localization was calculatedas similarity score betweenFOXO3a and DRAQ5 intensities.Data are representative of 1 of 3independent experiments.

Dal Col et al.





Moreover, RA/IFN-a treatment induced upregulation of Noxaand the concomitant Mcl-1 and A1/Bfl-1 inactivation as aresult of protein–protein interactions. (24–26). In contrast toother compounds inducing Noxa-dependent MCL cell apopto-sis, RA/IFN-a combination does not increase Mcl-1 proteinlevels and even downregulates A1/Bfl-1. Intriguingly, ourresults show that, under normal conditions, both Mcl-1 andA1/Bfl-1 can be bound to the full-length form of Bid, thuspreventing its cleavage and repressing its activation. MCL cellexposure to RA/IFN-a combination relieves this repressionthrough a competitive inhibition exerted byNoxa, which favorsBid displacement from pro/antiapoptotic protein complexesand its subsequent activation.Notably, we took advantage of multispectral imaging flow

cytometry to selectively analyze Bid–Mcl-1 and Bid–A1/Bfl-1colocalization in cells with morphometric features of early

apoptosis, a distinction that is not usually feasible in co-immu-noprecipitation experiments. This methodologic approach isparticularly relevant if we consider that the binding of Mcl-1 orA1/Bfl-1 to Bid abolishes its proapoptotic activity and that RA/IFN-a–induced Bid displacement from these complexes is anearly event in the activation of apoptotic machinery.

Noxa was initially identified as transcriptional target of p53(43), but now it is well known that it could be induced also byother factors, as it occurs in neuronal cells where it is inducedby the activation of the FOXO3a transcription factor, a down-stream target of the Akt kinase (35).We previously showed thatthe PI3K/Akt pathway is crucial for MCL cell survival and thatthe inhibition of Akt, but not of TORC1, induced apoptosis inMCL cells, suggesting that an Akt substrate different fromTORC1 is likely involved in mediating these effects. The resultspresented herein support the conclusion that the inhibition of

Figure 6. Antiproliferative andproapoptotic effects induced by RA/IFN-a in MCL cells. Untreated MCLcell (A) and modulation of criticalregulators by RA/IFN-a (B). tBid,truncated Bid.

Untreated MCL cell

Cell cycle

Cell cycle

Apoptosis

Apoptosis

RA/IFN-α–treated MCL cell

RA/IFN-α

RA/IFN-α

Proapoptotic

Prosurvival

Cell-cycle promoter

Cell-cycle inhibitor

Tentative stimulatory/inhibitory modification

Noxa

Cyclin D1

Cyclin D1

p27Kip1

p27Kip1

GSK-3

GSK-3

FOXO3a

FOXO3a

Cyto c

Cyto c

BaxBax

Bak

Bak

Bak

Bid

BidBid

Bid

Bid

Bid

Bak

p21Cip1

p21Cip1

Nucleus

Noxa

FOXO3a

FOXO3a

Noxa

Noxa

Noxa

Noxa

Noxa

Noxa

Noxa

Noxa Mcl-1

Mcl-1

Mcl-1 Mcl-1

Mcl-1

Mcl-1Mcl-1

Mcl-1

Mcl-1

Mcl-1

tBid

tBid

tBidtBid

tBid

Bfl-1

Bfl-1

Bfl-1

Bfl-1

Bfl-1

Bfl-1

Noxa

Noxa

Bcl-xL

Bcl-xL

Bcl-xL

BaxAkt

Akt

Bax

Nucleus

Stimulatory/inhibitory modification

B

A






the Akt pathway by RA/IFN-a, resulting in FOXO3a dephos-phorylation/activation and its subsequent nuclear internali-zation followed by Noxa upregulation, is one of the mainmolecularmechanismsunderlying RA/IFN-a–dependentMCLcell death.

Our findings are also consistent with a relevant role of RA/IFN-a–induced Akt inhibition in mediating MCL cell growthinhibition (9, 36). In particular, this drug combination mimicsthe effects induced inMCL cells by pharmacologic inhibition ofAkt, which results in cyclin D1 downregulation and p27Kip1

overexpression.Overall, the results of the present study show that how

the combination 9-cis-RA/IFN-a affects critical moleculartargets ofMCL, thus providing a strong rationale for the clinicalapplication of known and relatively cheap drugs in the man-agement of this aggressive and poorly responsive lymphoma.

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

Authors' ContributionsConception and design: J. Dal Col, R. Dolcetti.Development of methodology: J. Dal Col, K. Mastorci.Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J. Dal Col, E. Muraro, D. Martorelli, G. Inghirami.Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Dal Col, K. Mastorci, D.A. Fa�e, R. Dolcetti.Writing, review, and/or revision of the manuscript: J. Dal Col, R. DolcettiAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): K. Mastorci.Study supervision: R. Dolcetti.

Grant SupportThe study was supported by a grant from the Associazione Italiana per la

Ricerca sul Cancro (contract 10301 to R. Dolcetti and to Special ProgramMolecular Clinical Oncology 5-per-mille to G. Inghirami) and FondazioneFederica per la Vita.

The costs of publication of this article were defrayed in part 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 25, 2011; revised December 20, 2011; accepted January 23, 2012;published OnlineFirst February 6, 2012.

References1. Bertoni F, ZuccaE,Cavalli F.Mantle cell lymphoma.CurrOpinHematol

2004;11:411–8.2. Bosch F, Jares P, Campo E, Lopez-Guillermo A, Piris MA, Villamor N,

et al. PRAD-1/cyclin D1 gene overexpression in chronic lymphopro-liferative disorders: a highly specific marker of mantle cell lymphoma.Blood 1994;84:2726–32.

3. Bodrug SE,Warner BJ, BathML, LindemanGJ, Harris AW, Adams JM.Cyclin D1 transgene impedes lymphocytematuration andcollaboratesin lymphomagenesis with the myc gene. EMBO J 1994;13:2124–30.

4. Stilgenbauer S,Winkler D,Ott G, Schaffner C, Leupolt E, BentzM, et al.Molecular characterization of 11q deletions points to a pathogenic roleof the ATM gene in mantle cell lymphoma. Blood 1999;94:3262–4.

5. Chiarle R, Budel LM, Skolnik J, Frizzera G, Chilosi M, Corato A, et al.Increased proteasome degradation of cyclin-dependent kinase inhib-itor p27 is associated with a decreased overall survival in mantle celllymphoma. Blood 2000;95:619–26.

6. Martinez N, Camacho FI, Algara P, Rodriguez A, Dopazo A, Ruiz-Ballesteros E, et al. The molecular signature of mantle cell lymphomareveals multiple signals favoring cell survival. Cancer Res2003;63:8226–32.

7. Rosenwald A,Wright G,Wiestner A, ChanWC,Connors JM,Campo E,et al. The proliferation gene expression signature is a quantitativeintegrator of oncogenic events that predicts survival in mantle celllymphoma. Cancer Cell 2003;3:185–97.

8. Rizzatti EG, Falcao RP, Panepucci RA, Proto-Siqueira R, Anselmo-Lima WT, Okamoto OK, et al. Gene expression profiling of mantle celllymphoma cells reveals aberrant expression of genes from the PI3K-AKT, WNT and TGFbeta signalling pathways. Br J Haematol2005;130:516–26.

9. Dal Col J, Zancai P, Terrin L, Guidoboni M, Ponzoni M, Pavan A, et al.Distinct functional significance of Akt and mTOR constitutive activa-tion in mantle cell lymphoma. Blood 2008;111:5142–51.

10. Guidoboni M, Zancai P, Cariati R, Rizzo S, Dal Col J, Pavan A, et al.Retinoic acid inhibits the proliferative response induced by CD40activation and interleukin-4 in mantle cell lymphoma. Cancer Res2005;65:587–95.

11. Altucci L, Gronemeyer H. The promise of retinoids to fight againstcancer. Nat Rev Cancer 2001;1:181–93.

12. Chawla-Sarkar M, Lindner DJ, Liu YF,WilliamsBR, Sen GC, SilvermanRH, et al. Apoptosis and interferons: role of interferon-stimulatedgenes as mediators of apoptosis. Apoptosis 2003;8:237–49.

13. Vuletic A, Konjevic G, Milanovic D, Ruzdijic S, Jurisic V. Antiprolifera-tive effect of 13-cis-retinoic acid is associated with granulocyte dif-

ferentiation anddecrease in cyclin B1andBcl-2 protein levels inG0/G1arrested HL-60 cells. Pathol Oncol Res 2010;16:393–401.

14. Mrass P, Rendl M, Mildner M, Gruber F, Lengauer B, Ballaun C, et al.Retinoic acid increases the expression of p53 and proapoptoticcaspases and sensitizes keratinocytes to apoptosis: a possible expla-nation for tumor preventive action of retinoids. Cancer Res2004;64:6542–8.

15. Sun Y, Leaman DW. Involvement of noxa in cellular apoptoticresponses to interferon, double-stranded RNA, and virus infection.J Biol Chem 2005;280:15561–8.

16. Zhang R, Banik NL, Ray SK. Combination of all-trans retinoic acid andinterferon-gamma suppressed PI3K/Akt survival pathway in glioblas-toma T98G cells whereas NF-kappaB survival signaling in glioblasto-ma U87MG cells for induction of apoptosis. Neurochem Res2007;32:2194–202.

17. Bai Y, Ahmad U, Wang Y, Li HJ, Choy JC, Kim RW, et al. Interferon-yinduces x-linked inhibitor of apoptosis-associated factor-1 and noxaexpression and potentiates human vascular smooth muscle cell apo-ptosis by STAT3 activation. J Biol Chem 2008;283:6832–42.

18. Karabulut B, KaracaB, AtmacaH, Kisim A, Uzunoglu S, Sezgin C, et al.Regulation of apoptosis-relatedmolecules by synergistic combinationof all-trans retinoic acid and zoledronic acid in hormone-refractoryprostate cancer cell lines. Mol Biol Rep 2011;38:249–59.

19. Reed JC. Bcl-2 family proteins: regulators of apoptosis and chemore-sistance in hematologic malignancies. Semin Hematol 1997;34:9–19.

20. Nagy B, Lundan T, Larramendy ML, Aalto Y, Zhu Y, Niini T, et al.Abnormal expression of apoptosis-related genes in haematologicalmalignancies: overexpression of MYC is poor prognostic sign inmantle cell lymphoma. Br J Haematol 2003;120:434–41.

21. Rummel MJ, de VS, Hoelzer D, Koeffler HP, Hofmann WK. Alteredapoptosis pathways in mantle cell lymphoma. Leuk Lymphoma2004;45:49–54.

22. Cho-VegaJH,RassidakisGZ, Admirand JH,OyarzoM,RamalingamP,Paraguya A, et al. MCL-1 expression in B-cell non-Hodgkin's lympho-mas. Hum Pathol 2004;35:1095–100.

23. Reed JC. Bcl-2-family proteins and hematologic malignancies: historyand future prospects. Blood 2008;111:3322–30.

24. Perez-Galan P, Roue G, Villamor N, Montserrat E, Campo E, ColomerD. The proteasome inhibitor bortezomib induces apoptosis in mantle-cell lymphoma through generation of ROS and Noxa activation inde-pendent of p53 status. Blood 2006;107:257–64.

25. Tabe Y, Sebasigari D, Jin L, Rudelius M, Davies-Hill T, Miyake K,et al. MDM2 antagonist nutlin-3 displays antiproliferative and

Dal Col et al.





proapoptotic activity in mantle cell lymphoma. Clin Cancer Res2009;15:933–42.

26. Perez-Galan P, Roue G, Villamor N, Campo E, Colomer D. The BH3-mimetic GX15-070 synergizes with bortezomib in mantle cell lympho-ma by enhancing Noxa-mediated activation of Bak. Blood2007;109:4441–9.

27. Willis SN, Adams JM. Life in the balance: how BH3-only proteinsinduce apoptosis. Curr Opin Cell Biol 2005;17:617–25.

28. Adams JM, Cory S. The Bcl-2 apoptotic switch in cancer developmentand therapy. Oncogene 2007;26:1324–37.

29. Amin HM, McDonnell TJ, Medeiros LJ, Rassidakis GZ, Leventaki V,O'Connor SL, et al. Characterization of 4 mantle cell lymphoma celllines. Arch Pathol Lab Med 2003;127:424–31.

30. Beum PV, Lindorfer MA, Hall BE, George TC, Frost K, Morrissey PJ,et al. Quantitative analysis of protein co-localization on B cells opso-nized with rituximab and complement using the ImageStream multi-spectral imaging flow cytometer. J Immunol Methods 2006;317:90–9.

31. George TC, Fanning SL, Fitzgerald-Bocarsly P, Medeiros RB, HighfillS, Shimizu Y, et al. Quantitative measurement of nuclear translocationevents using similarity analysis of multispectral cellular imagesobtained in flow. J Immunol Methods 2006;311:117–29.

32. Becknell B, Trotta R, Yu J, Ding W, Mao HC, Hughes T, et al. Efficientinfection of human natural killer cells with an EBV/retroviral hybridvector. J Immunol Methods 2005;296:115–23.

33. Chambon P. A decade of molecular biology of retinoic acid receptors.FASEB J 1996;10:940–54.

34. Evans TR, Kaye SB. Retinoids: present role and future potential. Br JCancer 1999;80:1–8.

35. Obexer P, Geiger K, Ambros PF, Meister B, Ausserlechner MJ.FKHRL1-mediated expression of Noxa and Bim induces apoptosis

via the mitochondria in neuroblastoma cells. Cell Death Differ2007;14:534–47.

36. Dal Col J, Dolcetti R. GSK-3beta inhibition: at the crossroad betweenAkt and mTOR constitutive activation to enhance cyclin D1 proteinstability in mantle cell lymphoma. Cell Cycle 2008;7:2813–6.

37. Coiffier B, Ribrag V. Exploringmammalian target of rapamycin (mTOR)inhibition for treatment ofmantle cell lymphomaandother hematologicmalignancies. Leuk Lymphoma 2009;50:1916–30.

38. Perez-GalanP,DreylingM,Wiestner A.Mantle cell lymphoma: biology,pathogenesis, and themolecular basis of treatment in the genomic era.Blood 2011;117:26–38.

39. Klier M, Anastasov N, Hermann A, Meindl T, Angermeier D, Raffeld M,et al. Specific lentiviral shRNA-mediated knockdown of cyclin D1 inmantle cell lymphoma hasminimal effects on cell survival and reveals aregulatory circuit with cyclin D2. Leukemia 2008;22:2097–105.

40. Greiner TC, Dasgupta C, Ho VV, Weisenburger DD, Smith LM, LynchJC, et al. Mutation and genomic deletion status of ataxia telangiectasiamutated (ATM) and p53 confer specific gene expression profiles inmantle cell lymphoma. Proc Natl Acad Sci U S A 2006;103:2352–7.

41. Stefancikova L, Moulis M, Fabian P, Ravcukova B, Vasova I, Muzik J,et al. Loss of the p53 tumor suppressor activity is associated withnegative prognosis ofmantle cell lymphoma. Int J Oncol 2010;36:699–706.

42. JonesRJ,ChenQ,VoorheesPM,YoungKH,Bruey-SedanoN,YangD,et al. Inhibition of the p53E3 ligaseHDM-2 induces apoptosis andDNAdamage–independent p53 phosphorylation in mantle cell lymphoma.Clin Cancer Res 2008;14:5416–25.

43. Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T, et al.Noxa, a BH3-only member of the Bcl-2 family and candidate mediatorof p53-induced apoptosis. Science 2000;288:1053–8.






2012;72:1825-1835. Published OnlineFirst February 6, 2012.Cancer Res Jessica Dal Col, Katy Mastorci, Damiana Antonia Faè, et al. Akt-Dependent Modulation of Critical TargetsPromotes Apoptosis in Mantle Cell Lymphoma through Retinoic Acid/Alpha-Interferon Combination Inhibits Growth and

Updated version

10.1158/0008-5472.CAN-11-2505doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2012/02/06/0008-5472.CAN-11-2505.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/72/7/1825.full#ref-list-1

This article cites 43 articles, 17 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/72/7/1825.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://cancerres.aacrjournals.org/content/72/7/1825To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-11-2505

http://cancerres.aacrjournals.org/content/suppl/2012/02/06/0008-5472.CAN-11-2505.DC1

http://cancerres.aacrjournals.org/content/72/7/1825.full#ref-list-1

http://cancerres.aacrjournals.org/content/72/7/1825.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/72/7/1825


retinoic acid/alpha-interferon combination inhibits growth and … · therapeutics, targets, and...

Documents